Merge branch 'master' of /usr/src/ntfs-2.6/

320 files changed, 5784 insertions, 4256 deletions

diff --git a/Documentation/kernel-parameters.txt b/Documentation/kernel-parameters.txt
index fc99075e0af..7b7382d0f75 100644
--- a/Documentation/kernel-parameters.txt
+++ b/Documentation/kernel-parameters.txt
@@ -1008,7 +1008,9 @@ running once the system is up.
 			noexec=on: enable non-executable mappings (default)
 			noexec=off: disable nn-executable mappings
 
-	nofxsr		[BUGS=IA-32]
+	nofxsr		[BUGS=IA-32] Disables x86 floating point extended
+			register save and restore. The kernel will only save
+			legacy floating-point registers on task switch.
 
 	nohlt		[BUGS=ARM]
 
@@ -1053,6 +1055,8 @@ running once the system is up.
 
 	nosbagart	[IA-64]
 
+	nosep		[BUGS=IA-32] Disables x86 SYSENTER/SYSEXIT support.
+
 	nosmp		[SMP] Tells an SMP kernel to act as a UP kernel.
 
 	nosync		[HW,M68K] Disables sync negotiation for all devices.
@@ -1122,6 +1126,11 @@ running once the system is up.
 	pas16=		[HW,SCSI]
 			See header of drivers/scsi/pas16.c.
 
+	pause_on_oops=
+			Halt all CPUs after the first oops has been printed for
+			the specified number of seconds.  This is to be used if
+			your oopses keep scrolling off the screen.
+
 	pcbit=		[HW,ISDN]
 
 	pcd.		[PARIDE]
diff --git a/Documentation/power/swsusp.txt b/Documentation/power/swsusp.txt
index b28b7f04abb..d7814a113ee 100644
--- a/Documentation/power/swsusp.txt
+++ b/Documentation/power/swsusp.txt
@@ -17,6 +17,11 @@ Some warnings, first.
  * but it will probably only crash.
  *
  * (*) suspend/resume support is needed to make it safe.
+ *
+ * If you have any filesystems on USB devices mounted before suspend,
+ * they won't be accessible after resume and you may lose data, as though
+ * you have unplugged the USB devices with mounted filesystems on them
+ * (see the FAQ below for details).
 
 You need to append resume=/dev/your_swap_partition to kernel command
 line. Then you suspend by
@@ -27,19 +32,18 @@ echo shutdown > /sys/power/disk; echo disk > /sys/power/state
 
 echo platform > /sys/power/disk; echo disk > /sys/power/state
 
+. If you have SATA disks, you'll need recent kernels with SATA suspend
+support. For suspend and resume to work, make sure your disk drivers
+are built into kernel -- not modules. [There's way to make
+suspend/resume with modular disk drivers, see FAQ, but you probably
+should not do that.]
+
 If you want to limit the suspend image size to N bytes, do
 
 echo N > /sys/power/image_size
 
 before suspend (it is limited to 500 MB by default).
 
-Encrypted suspend image:
-------------------------
-If you want to store your suspend image encrypted with a temporary
-key to prevent data gathering after resume you must compile
-crypto and the aes algorithm into the kernel - modules won't work
-as they cannot be loaded at resume time.
-
 
 Article about goals and implementation of Software Suspend for Linux
 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
@@ -333,4 +337,37 @@ init=/bin/bash, then swapon and starting suspend sequence manually
 usually does the trick. Then it is good idea to try with latest
 vanilla kernel.
 
+Q: How can distributions ship a swsusp-supporting kernel with modular
+disk drivers (especially SATA)?
+
+A: Well, it can be done, load the drivers, then do echo into
+/sys/power/disk/resume file from initrd. Be sure not to mount
+anything, not even read-only mount, or you are going to lose your
+data.
+
+Q: How do I make suspend more verbose?
+
+A: If you want to see any non-error kernel messages on the virtual
+terminal the kernel switches to during suspend, you have to set the
+kernel console loglevel to at least 5, for example by doing
+
+	echo 5 > /proc/sys/kernel/printk
+
+Q: Is this true that if I have a mounted filesystem on a USB device and
+I suspend to disk, I can lose data unless the filesystem has been mounted
+with "sync"?
+
+A: That's right.  It depends on your hardware, and it could be true even for
+suspend-to-RAM.  In fact, even with "-o sync" you can lose data if your
+programs have information in buffers they haven't written out to disk.
+
+If you're lucky, your hardware will support low-power modes for USB
+controllers while the system is asleep.  Lots of hardware doesn't,
+however.  Shutting off the power to a USB controller is equivalent to
+unplugging all the attached devices.
+
+Remember that it's always a bad idea to unplug a disk drive containing a
+mounted filesystem.  With USB that's true even when your system is asleep!
+The safest thing is to unmount all USB-based filesystems before suspending
+and remount them after resuming.
 
diff --git a/Documentation/power/userland-swsusp.txt b/Documentation/power/userland-swsusp.txt
new file mode 100644
index 00000000000..94058220aaf
--- /dev/null
+++ b/Documentation/power/userland-swsusp.txt
@@ -0,0 +1,149 @@
+Documentation for userland software suspend interface
+	(C) 2006 Rafael J. Wysocki <rjw@sisk.pl>
+
+First, the warnings at the beginning of swsusp.txt still apply.
+
+Second, you should read the FAQ in swsusp.txt _now_ if you have not
+done it already.
+
+Now, to use the userland interface for software suspend you need special
+utilities that will read/write the system memory snapshot from/to the
+kernel.  Such utilities are available, for example, from
+<http://www.sisk.pl/kernel/utilities/suspend>.  You may want to have
+a look at them if you are going to develop your own suspend/resume
+utilities.
+
+The interface consists of a character device providing the open(),
+release(), read(), and write() operations as well as several ioctl()
+commands defined in kernel/power/power.h.  The major and minor
+numbers of the device are, respectively, 10 and 231, and they can
+be read from /sys/class/misc/snapshot/dev.
+
+The device can be open either for reading or for writing.  If open for
+reading, it is considered to be in the suspend mode.  Otherwise it is
+assumed to be in the resume mode.  The device cannot be open for reading
+and writing.  It is also impossible to have the device open more than once
+at a time.
+
+The ioctl() commands recognized by the device are:
+
+SNAPSHOT_FREEZE - freeze user space processes (the current process is
+	not frozen); this is required for SNAPSHOT_ATOMIC_SNAPSHOT
+	and SNAPSHOT_ATOMIC_RESTORE to succeed
+
+SNAPSHOT_UNFREEZE - thaw user space processes frozen by SNAPSHOT_FREEZE
+
+SNAPSHOT_ATOMIC_SNAPSHOT - create a snapshot of the system memory; the
+	last argument of ioctl() should be a pointer to an int variable,
+	the value of which will indicate whether the call returned after
+	creating the snapshot (1) or after restoring the system memory state
+	from it (0) (after resume the system finds itself finishing the
+	SNAPSHOT_ATOMIC_SNAPSHOT ioctl() again); after the snapshot
+	has been created the read() operation can be used to transfer
+	it out of the kernel
+
+SNAPSHOT_ATOMIC_RESTORE - restore the system memory state from the
+	uploaded snapshot image; before calling it you should transfer
+	the system memory snapshot back to the kernel using the write()
+	operation; this call will not succeed if the snapshot
+	image is not available to the kernel
+
+SNAPSHOT_FREE - free memory allocated for the snapshot image
+
+SNAPSHOT_SET_IMAGE_SIZE - set the preferred maximum size of the image
+	(the kernel will do its best to ensure the image size will not exceed
+	this number, but if it turns out to be impossible, the kernel will
+	create the smallest image possible)
+
+SNAPSHOT_AVAIL_SWAP - return the amount of available swap in bytes (the last
+	argument should be a pointer to an unsigned int variable that will
+	contain the result if the call is successful).
+
+SNAPSHOT_GET_SWAP_PAGE - allocate a swap page from the resume partition
+	(the last argument should be a pointer to a loff_t variable that
+	will contain the swap page offset if the call is successful)
+
+SNAPSHOT_FREE_SWAP_PAGES - free all swap pages allocated with
+	SNAPSHOT_GET_SWAP_PAGE
+
+SNAPSHOT_SET_SWAP_FILE - set the resume partition (the last ioctl() argument
+	should specify the device's major and minor numbers in the old
+	two-byte format, as returned by the stat() function in the .st_rdev
+	member of the stat structure); it is recommended to always use this
+	call, because the code to set the resume partition could be removed from
+	future kernels
+
+The device's read() operation can be used to transfer the snapshot image from
+the kernel.  It has the following limitations:
+- you cannot read() more than one virtual memory page at a time
+- read()s accross page boundaries are impossible (ie. if ypu read() 1/2 of
+	a page in the previous call, you will only be able to read()
+	_at_ _most_ 1/2 of the page in the next call)
+
+The device's write() operation is used for uploading the system memory snapshot
+into the kernel.  It has the same limitations as the read() operation.
+
+The release() operation frees all memory allocated for the snapshot image
+and all swap pages allocated with SNAPSHOT_GET_SWAP_PAGE (if any).
+Thus it is not necessary to use either SNAPSHOT_FREE or
+SNAPSHOT_FREE_SWAP_PAGES before closing the device (in fact it will also
+unfreeze user space processes frozen by SNAPSHOT_UNFREEZE if they are
+still frozen when the device is being closed).
+
+Currently it is assumed that the userland utilities reading/writing the
+snapshot image from/to the kernel will use a swap parition, called the resume
+partition, as storage space.  However, this is not really required, as they
+can use, for example, a special (blank) suspend partition or a file on a partition
+that is unmounted before SNAPSHOT_ATOMIC_SNAPSHOT and mounted afterwards.
+
+These utilities SHOULD NOT make any assumptions regarding the ordering of
+data within the snapshot image, except for the image header that MAY be
+assumed to start with an swsusp_info structure, as specified in
+kernel/power/power.h.  This structure MAY be used by the userland utilities
+to obtain some information about the snapshot image, such as the size
+of the snapshot image, including the metadata and the header itself,
+contained in the .size member of swsusp_info.
+
+The snapshot image MUST be written to the kernel unaltered (ie. all of the image
+data, metadata and header MUST be written in _exactly_ the same amount, form
+and order in which they have been read).  Otherwise, the behavior of the
+resumed system may be totally unpredictable.
+
+While executing SNAPSHOT_ATOMIC_RESTORE the kernel checks if the
+structure of the snapshot image is consistent with the information stored
+in the image header.  If any inconsistencies are detected,
+SNAPSHOT_ATOMIC_RESTORE will not succeed.  Still, this is not a fool-proof
+mechanism and the userland utilities using the interface SHOULD use additional
+means, such as checksums, to ensure the integrity of the snapshot image.
+
+The suspending and resuming utilities MUST lock themselves in memory,
+preferrably using mlockall(), before calling SNAPSHOT_FREEZE.
+
+The suspending utility MUST check the value stored by SNAPSHOT_ATOMIC_SNAPSHOT
+in the memory location pointed to by the last argument of ioctl() and proceed
+in accordance with it:
+1. 	If the value is 1 (ie. the system memory snapshot has just been
+	created and the system is ready for saving it):
+	(a)	The suspending utility MUST NOT close the snapshot device
+		_unless_ the whole suspend procedure is to be cancelled, in
+		which case, if the snapshot image has already been saved, the
+		suspending utility SHOULD destroy it, preferrably by zapping
+		its header.  If the suspend is not to be cancelled, the
+		system MUST be powered off or rebooted after the snapshot
+		image has been saved.
+	(b)	The suspending utility SHOULD NOT attempt to perform any
+		file system operations (including reads) on the file systems
+		that were mounted before SNAPSHOT_ATOMIC_SNAPSHOT has been
+		called.  However, it MAY mount a file system that was not
+		mounted at that time and perform some operations on it (eg.
+		use it for saving the image).
+2.	If the value is 0 (ie. the system state has just been restored from
+	the snapshot image), the suspending utility MUST close the snapshot
+	device.  Afterwards it will be treated as a regular userland process,
+	so it need not exit.
+
+The resuming utility SHOULD NOT attempt to mount any file systems that could
+be mounted before suspend and SHOULD NOT attempt to perform any operations
+involving such file systems.
+
+For details, please refer to the source code.
diff --git a/Documentation/power/video.txt b/Documentation/power/video.txt
index 912bed87c75..d18a57d1a53 100644
--- a/Documentation/power/video.txt
+++ b/Documentation/power/video.txt
@@ -1,7 +1,7 @@
 
 		Video issues with S3 resume
 		~~~~~~~~~~~~~~~~~~~~~~~~~~~
-		  2003-2005, Pavel Machek
+		  2003-2006, Pavel Machek
 
 During S3 resume, hardware needs to be reinitialized. For most
 devices, this is easy, and kernel driver knows how to do
@@ -15,6 +15,27 @@ run normally so video card is normally initialized. It should not be
 problem for S1 standby, because hardware should retain its state over
 that.
 
+We either have to run video BIOS during early resume, or interpret it
+using vbetool later, or maybe nothing is neccessary on particular
+system because video state is preserved. Unfortunately different
+methods work on different systems, and no known method suits all of
+them.
+
+Userland application called s2ram has been developed; it contains long
+whitelist of systems, and automatically selects working method for a
+given system. It can be downloaded from CVS at
+www.sf.net/projects/suspend . If you get a system that is not in the
+whitelist, please try to find a working solution, and submit whitelist
+entry so that work does not need to be repeated.
+
+Currently, VBE_SAVE method (6 below) works on most
+systems. Unfortunately, vbetool only runs after userland is resumed,
+so it makes debugging of early resume problems
+hard/impossible. Methods that do not rely on userland are preferable.
+
+Details
+~~~~~~~
+
 There are a few types of systems where video works after S3 resume:
 
 (1) systems where video state is preserved over S3.
@@ -104,6 +125,7 @@ HP NX7000			??? (*)
 HP Pavilion ZD7000		vbetool post needed, need open-source nv driver for X
 HP Omnibook XE3	athlon version	none (1)
 HP Omnibook XE3GC		none (1), video is S3 Savage/IX-MV
+HP Omnibook 5150		none (1), (S1 also works OK)
 IBM TP T20, model 2647-44G	none (1), video is S3 Inc. 86C270-294 Savage/IX-MV, vesafb gets "interesting" but X work.
 IBM TP A31 / Type 2652-M5G      s3_mode (3) [works ok with BIOS 1.04 2002-08-23, but not at all with BIOS 1.11 2004-11-05 :-(]
 IBM TP R32 / Type 2658-MMG      none (1)
@@ -120,18 +142,24 @@ IBM ThinkPad T42p (2373-GTG)	s3_bios (2)
 IBM TP X20			??? (*)
 IBM TP X30			s3_bios (2)
 IBM TP X31 / Type 2672-XXH      none (1), use radeontool (http://fdd.com/software/radeon/) to turn off backlight.
-IBM TP X32			none (1), but backlight is on and video is trashed after long suspend
+IBM TP X32			none (1), but backlight is on and video is trashed after long suspend. s3_bios,s3_mode (4) works too. Perhaps that gets better results?
 IBM Thinkpad X40 Type 2371-7JG  s3_bios,s3_mode (4)
+IBM TP 600e			none(1), but a switch to console and back to X is needed
 Medion MD4220			??? (*)
 Samsung P35			vbetool needed (6)
-Sharp PC-AR10 (ATI rage)	none (1)
+Sharp PC-AR10 (ATI rage)	none (1), backlight does not switch off
 Sony Vaio PCG-C1VRX/K		s3_bios (2)
 Sony Vaio PCG-F403		??? (*)
+Sony Vaio PCG-GRT995MP		none (1), works with 'nv' X driver
+Sony Vaio PCG-GR7/K		none (1), but needs radeonfb, use radeontool (http://fdd.com/software/radeon/) to turn off backlight.
 Sony Vaio PCG-N505SN		??? (*)
 Sony Vaio vgn-s260		X or boot-radeon can init it (5)
+Sony Vaio vgn-S580BH		vga=normal, but suspend from X. Console will be blank unless you return to X.
+Sony Vaio vgn-FS115B		s3_bios (2),s3_mode (4)
 Toshiba Libretto L5		none (1)
-Toshiba Satellite 4030CDT	s3_mode (3)
-Toshiba Satellite 4080XCDT      s3_mode (3)
+Toshiba Portege 3020CT		s3_mode (3)
+Toshiba Satellite 4030CDT	s3_mode (3) (S1 also works OK)
+Toshiba Satellite 4080XCDT      s3_mode (3) (S1 also works OK)
 Toshiba Satellite 4090XCDT      ??? (*)
 Toshiba Satellite P10-554       s3_bios,s3_mode (4)(****)
 Toshiba M30                     (2) xor X with nvidia driver using internal AGP
@@ -151,39 +179,3 @@ Asus A7V8X	    nVidia RIVA TNT2 model 64	  s3_bios,s3_mode (4)
 (***) To be tested with a newer kernel.
 
 (****) Not with SMP kernel, UP only.
-
-VBEtool details
-~~~~~~~~~~~~~~~
-(with thanks to Carl-Daniel Hailfinger)
-
-First, boot into X and run the following script ONCE:
-#!/bin/bash
-statedir=/root/s3/state
-mkdir -p $statedir
-chvt 2
-sleep 1
-vbetool vbestate save >$statedir/vbe
-
-
-To suspend and resume properly, call the following script as root:
-#!/bin/bash
-statedir=/root/s3/state
-curcons=`fgconsole`
-fuser /dev/tty$curcons 2>/dev/null|xargs ps -o comm= -p|grep -q X && chvt 2
-cat /dev/vcsa >$statedir/vcsa
-sync
-echo 3 >/proc/acpi/sleep
-sync
-vbetool post
-vbetool vbestate restore <$statedir/vbe
-cat $statedir/vcsa >/dev/vcsa
-rckbd restart
-chvt $[curcons%6+1]
-chvt $curcons
-
-
-Unless you change your graphics card or other hardware configuration,
-the state once saved will be OK for every resume afterwards.
-NOTE: The "rckbd restart" command may be different for your
-distribution. Simply replace it with the command you would use to
-set the fonts on screen.
diff --git a/arch/cris/kernel/irq.c b/arch/cris/kernel/irq.c
index 30deaf1b728..b504def3e34 100644
--- a/arch/cris/kernel/irq.c
+++ b/arch/cris/kernel/irq.c
@@ -52,9 +52,8 @@ int show_interrupts(struct seq_file *p, void *v)
 
 	if (i == 0) {
 		seq_printf(p, "           ");
-		for (j=0; j<NR_CPUS; j++)
-			if (cpu_online(j))
-				seq_printf(p, "CPU%d       ",j);
+		for_each_online_cpu(j)
+			seq_printf(p, "CPU%d       ",j);
 		seq_putc(p, '\n');
 	}
 
@@ -67,9 +66,8 @@ int show_interrupts(struct seq_file *p, void *v)
 #ifndef CONFIG_SMP
 		seq_printf(p, "%10u ", kstat_irqs(i));
 #else
-		for (j = 0; j < NR_CPUS; j++)
-			if (cpu_online(j))
-				seq_printf(p, "%10u ", kstat_cpu(j).irqs[i]);
+		for_each_online_cpu(j)
+			seq_printf(p, "%10u ", kstat_cpu(j).irqs[i]);
 #endif
 		seq_printf(p, " %14s", irq_desc[i].handler->typename);
 		seq_printf(p, "  %s", action->name);
diff --git a/arch/frv/kernel/irq.c b/arch/frv/kernel/irq.c
index 27ab4c30aac..11fa326a8f6 100644
--- a/arch/frv/kernel/irq.c
+++ b/arch/frv/kernel/irq.c
@@ -75,9 +75,8 @@ int show_interrupts(struct seq_file *p, void *v)
 	switch (i) {
 	case 0:
 		seq_printf(p, "           ");
-		for (j = 0; j < NR_CPUS; j++)
-			if (cpu_online(j))
-				seq_printf(p, "CPU%d       ",j);
+		for_each_online_cpu(j)
+			seq_printf(p, "CPU%d       ",j);
 
 		seq_putc(p, '\n');
 		break;
@@ -100,9 +99,8 @@ int show_interrupts(struct seq_file *p, void *v)
 #ifndef CONFIG_SMP
 		seq_printf(p, "%10u ", kstat_irqs(i));
 #else
-		for (j = 0; j < NR_CPUS; j++)
-			if (cpu_online(j))
-				seq_printf(p, "%10u ", kstat_cpu(j).irqs[i - 1]);
+		for_each_online_cpu(j)
+			seq_printf(p, "%10u ", kstat_cpu(j).irqs[i - 1]);
 #endif
 
 		level = group->sources[ix]->level - frv_irq_levels;
diff --git a/arch/i386/Kconfig b/arch/i386/Kconfig
index 5b1a7d46d1d..bfea1bedcbf 100644
--- a/arch/i386/Kconfig
+++ b/arch/i386/Kconfig
@@ -80,6 +80,7 @@ config X86_VOYAGER
 
 config X86_NUMAQ
 	bool "NUMAQ (IBM/Sequent)"
+	select SMP
 	select NUMA
 	help
 	  This option is used for getting Linux to run on a (IBM/Sequent) NUMA
@@ -400,6 +401,7 @@ choice
 
 config NOHIGHMEM
 	bool "off"
+	depends on !X86_NUMAQ
 	---help---
 	  Linux can use up to 64 Gigabytes of physical memory on x86 systems.
 	  However, the address space of 32-bit x86 processors is only 4
@@ -436,6 +438,7 @@ config NOHIGHMEM
 
 config HIGHMEM4G
 	bool "4GB"
+	depends on !X86_NUMAQ
 	help
 	  Select this if you have a 32-bit processor and between 1 and 4
 	  gigabytes of physical RAM.
@@ -503,10 +506,6 @@ config NUMA
 	default n if X86_PC
 	default y if (X86_NUMAQ || X86_SUMMIT)
 
-# Need comments to help the hapless user trying to turn on NUMA support
-comment "NUMA (NUMA-Q) requires SMP, 64GB highmem support"
-	depends on X86_NUMAQ && (!HIGHMEM64G || !SMP)
-
 comment "NUMA (Summit) requires SMP, 64GB highmem support, ACPI"
 	depends on X86_SUMMIT && (!HIGHMEM64G || !ACPI)
 
@@ -660,13 +659,18 @@ config BOOT_IOREMAP
 	default y
 
 config REGPARM
-	bool "Use register arguments (EXPERIMENTAL)"
-	depends on EXPERIMENTAL
-	default n
+	bool "Use register arguments"
+	default y
 	help
-	Compile the kernel with -mregparm=3. This uses a different ABI
-	and passes the first three arguments of a function call in registers.
-	This will probably break binary only modules.
+	Compile the kernel with -mregparm=3. This instructs gcc to use
+	a more efficient function call ABI which passes the first three
+	arguments of a function call via registers, which results in denser
+	and faster code.
+
+	If this option is disabled, then the default ABI of passing
+	arguments via the stack is used.
+
+	If unsure, say Y.
 
 config SECCOMP
 	bool "Enable seccomp to safely compute untrusted bytecode"
diff --git a/arch/i386/Kconfig.debug b/arch/i386/Kconfig.debug
index bf32ecc9ad0..00108ba9a78 100644
--- a/arch/i386/Kconfig.debug
+++ b/arch/i386/Kconfig.debug
@@ -31,6 +31,15 @@ config DEBUG_STACK_USAGE
 
 	  This option will slow down process creation somewhat.
 
+config STACK_BACKTRACE_COLS
+	int "Stack backtraces per line" if DEBUG_KERNEL
+	range 1 3
+	default 2
+	help
+	  Selects how many stack backtrace entries per line to display.
+
+	  This can save screen space when displaying traces.
+
 comment "Page alloc debug is incompatible with Software Suspend on i386"
 	depends on DEBUG_KERNEL && SOFTWARE_SUSPEND
 
diff --git a/arch/i386/kernel/Makefile b/arch/i386/kernel/Makefile
index 65656c033d7..5b9ed21216c 100644
--- a/arch/i386/kernel/Makefile
+++ b/arch/i386/kernel/Makefile
@@ -7,7 +7,7 @@ extra-y := head.o init_task.o vmlinux.lds
 obj-y	:= process.o semaphore.o signal.o entry.o traps.o irq.o \
 		ptrace.o time.o ioport.o ldt.o setup.o i8259.o sys_i386.o \
 		pci-dma.o i386_ksyms.o i387.o dmi_scan.o bootflag.o \
-		quirks.o i8237.o topology.o
+		quirks.o i8237.o topology.o alternative.o
 
 obj-y				+= cpu/
 obj-y				+= timers/
diff --git a/arch/i386/kernel/alternative.c b/arch/i386/kernel/alternative.c
new file mode 100644
index 00000000000..5cbd6f99fb2
--- /dev/null
+++ b/arch/i386/kernel/alternative.c
@@ -0,0 +1,321 @@
+#include <linux/module.h>
+#include <linux/spinlock.h>
+#include <linux/list.h>
+#include <asm/alternative.h>
+#include <asm/sections.h>
+
+#define DEBUG 0
+#if DEBUG
+# define DPRINTK(fmt, args...) printk(fmt, args)
+#else
+# define DPRINTK(fmt, args...)
+#endif
+
+/* Use inline assembly to define this because the nops are defined
+   as inline assembly strings in the include files and we cannot
+   get them easily into strings. */
+asm("\t.data\nintelnops: "
+	GENERIC_NOP1 GENERIC_NOP2 GENERIC_NOP3 GENERIC_NOP4 GENERIC_NOP5 GENERIC_NOP6
+	GENERIC_NOP7 GENERIC_NOP8);
+asm("\t.data\nk8nops: "
+	K8_NOP1 K8_NOP2 K8_NOP3 K8_NOP4 K8_NOP5 K8_NOP6
+	K8_NOP7 K8_NOP8);
+asm("\t.data\nk7nops: "
+	K7_NOP1 K7_NOP2 K7_NOP3 K7_NOP4 K7_NOP5 K7_NOP6
+	K7_NOP7 K7_NOP8);
+
+extern unsigned char intelnops[], k8nops[], k7nops[];
+static unsigned char *intel_nops[ASM_NOP_MAX+1] = {
+	NULL,
+	intelnops,
+	intelnops + 1,
+	intelnops + 1 + 2,
+	intelnops + 1 + 2 + 3,
+	intelnops + 1 + 2 + 3 + 4,
+	intelnops + 1 + 2 + 3 + 4 + 5,
+	intelnops + 1 + 2 + 3 + 4 + 5 + 6,
+	intelnops + 1 + 2 + 3 + 4 + 5 + 6 + 7,
+};
+static unsigned char *k8_nops[ASM_NOP_MAX+1] = {
+	NULL,
+	k8nops,
+	k8nops + 1,
+	k8nops + 1 + 2,
+	k8nops + 1 + 2 + 3,
+	k8nops + 1 + 2 + 3 + 4,
+	k8nops + 1 + 2 + 3 + 4 + 5,
+	k8nops + 1 + 2 + 3 + 4 + 5 + 6,
+	k8nops + 1 + 2 + 3 + 4 + 5 + 6 + 7,
+};
+static unsigned char *k7_nops[ASM_NOP_MAX+1] = {
+	NULL,
+	k7nops,
+	k7nops + 1,
+	k7nops + 1 + 2,
+	k7nops + 1 + 2 + 3,
+	k7nops + 1 + 2 + 3 + 4,
+	k7nops + 1 + 2 + 3 + 4 + 5,
+	k7nops + 1 + 2 + 3 + 4 + 5 + 6,
+	k7nops + 1 + 2 + 3 + 4 + 5 + 6 + 7,
+};
+static struct nop {
+	int cpuid;
+	unsigned char **noptable;
+} noptypes[] = {
+	{ X86_FEATURE_K8, k8_nops },
+	{ X86_FEATURE_K7, k7_nops },
+	{ -1, NULL }
+};
+
+
+extern struct alt_instr __alt_instructions[], __alt_instructions_end[];
+extern struct alt_instr __smp_alt_instructions[], __smp_alt_instructions_end[];
+extern u8 *__smp_locks[], *__smp_locks_end[];
+
+extern u8 __smp_alt_begin[], __smp_alt_end[];
+
+
+static unsigned char** find_nop_table(void)
+{
+	unsigned char **noptable = intel_nops;
+	int i;
+
+	for (i = 0; noptypes[i].cpuid >= 0; i++) {
+		if (boot_cpu_has(noptypes[i].cpuid)) {
+			noptable = noptypes[i].noptable;
+			break;
+		}
+	}
+	return noptable;
+}
+
+/* Replace instructions with better alternatives for this CPU type.
+   This runs before SMP is initialized to avoid SMP problems with
+   self modifying code. This implies that assymetric systems where
+   APs have less capabilities than the boot processor are not handled.
+   Tough. Make sure you disable such features by hand. */
+
+void apply_alternatives(struct alt_instr *start, struct alt_instr *end)
+{
+	unsigned char **noptable = find_nop_table();
+	struct alt_instr *a;
+	int diff, i, k;
+
+	DPRINTK("%s: alt table %p -> %p\n", __FUNCTION__, start, end);
+	for (a = start; a < end; a++) {
+		BUG_ON(a->replacementlen > a->instrlen);
+		if (!boot_cpu_has(a->cpuid))
+			continue;
+		memcpy(a->instr, a->replacement, a->replacementlen);
+		diff = a->instrlen - a->replacementlen;
+		/* Pad the rest with nops */
+		for (i = a->replacementlen; diff > 0; diff -= k, i += k) {
+			k = diff;
+			if (k > ASM_NOP_MAX)
+				k = ASM_NOP_MAX;
+			memcpy(a->instr + i, noptable[k], k);
+		}
+	}
+}
+
+static void alternatives_smp_save(struct alt_instr *start, struct alt_instr *end)
+{
+	struct alt_instr *a;
+
+	DPRINTK("%s: alt table %p-%p\n", __FUNCTION__, start, end);
+	for (a = start; a < end; a++) {
+		memcpy(a->replacement + a->replacementlen,
+		       a->instr,
+		       a->instrlen);
+	}
+}
+
+static void alternatives_smp_apply(struct alt_instr *start, struct alt_instr *end)
+{
+	struct alt_instr *a;
+
+	for (a = start; a < end; a++) {
+		memcpy(a->instr,
+		       a->replacement + a->replacementlen,
+		       a->instrlen);
+	}
+}
+
+static void alternatives_smp_lock(u8 **start, u8 **end, u8 *text, u8 *text_end)
+{
+	u8 **ptr;
+
+	for (ptr = start; ptr < end; ptr++) {
+		if (*ptr < text)
+			continue;
+		if (*ptr > text_end)
+			continue;
+		**ptr = 0xf0; /* lock prefix */
+	};
+}
+
+static void alternatives_smp_unlock(u8 **start, u8 **end, u8 *text, u8 *text_end)
+{
+	unsigned char **noptable = find_nop_table();
+	u8 **ptr;
+
+	for (ptr = start; ptr < end; ptr++) {
+		if (*ptr < text)
+			continue;
+		if (*ptr > text_end)
+			continue;
+		**ptr = noptable[1][0];
+	};
+}
+
+struct smp_alt_module {
+	/* what is this ??? */
+	struct module	*mod;
+	char		*name;
+
+	/* ptrs to lock prefixes */
+	u8		**locks;
+	u8		**locks_end;
+
+	/* .text segment, needed to avoid patching init code ;) */
+	u8		*text;
+	u8		*text_end;
+
+	struct list_head next;
+};
+static LIST_HEAD(smp_alt_modules);
+static DEFINE_SPINLOCK(smp_alt);
+
+static int smp_alt_once = 0;
+static int __init bootonly(char *str)
+{
+	smp_alt_once = 1;
+	return 1;
+}
+__setup("smp-alt-boot", bootonly);
+
+void alternatives_smp_module_add(struct module *mod, char *name,
+				 void *locks, void *locks_end,
+				 void *text,  void *text_end)
+{
+	struct smp_alt_module *smp;
+	unsigned long flags;
+
+	if (smp_alt_once) {
+		if (boot_cpu_has(X86_FEATURE_UP))
+			alternatives_smp_unlock(locks, locks_end,
+						text, text_end);
+		return;
+	}
+
+	smp = kzalloc(sizeof(*smp), GFP_KERNEL);
+	if (NULL == smp)
+		return; /* we'll run the (safe but slow) SMP code then ... */
+
+	smp->mod	= mod;
+	smp->name	= name;
+	smp->locks	= locks;
+	smp->locks_end	= locks_end;
+	smp->text	= text;
+	smp->text_end	= text_end;
+	DPRINTK("%s: locks %p -> %p, text %p -> %p, name %s\n",
+		__FUNCTION__, smp->locks, smp->locks_end,
+		smp->text, smp->text_end, smp->name);
+
+	spin_lock_irqsave(&smp_alt, flags);
+	list_add_tail(&smp->next, &smp_alt_modules);
+	if (boot_cpu_has(X86_FEATURE_UP))
+		alternatives_smp_unlock(smp->locks, smp->locks_end,
+					smp->text, smp->text_end);
+	spin_unlock_irqrestore(&smp_alt, flags);
+}
+
+void alternatives_smp_module_del(struct module *mod)
+{
+	struct smp_alt_module *item;
+	unsigned long flags;
+
+	if (smp_alt_once)
+		return;
+
+	spin_lock_irqsave(&smp_alt, flags);
+	list_for_each_entry(item, &smp_alt_modules, next) {
+		if (mod != item->mod)
+			continue;
+		list_del(&item->next);
+		spin_unlock_irqrestore(&smp_alt, flags);
+		DPRINTK("%s: %s\n", __FUNCTION__, item->name);
+		kfree(item);
+		return;
+	}
+	spin_unlock_irqrestore(&smp_alt, flags);
+}
+
+void alternatives_smp_switch(int smp)
+{
+	struct smp_alt_module *mod;
+	unsigned long flags;
+
+	if (smp_alt_once)
+		return;
+	BUG_ON(!smp && (num_online_cpus() > 1));
+
+	spin_lock_irqsave(&smp_alt, flags);
+	if (smp) {
+		printk(KERN_INFO "SMP alternatives: switching to SMP code\n");
+		clear_bit(X86_FEATURE_UP, boot_cpu_data.x86_capability);
+		clear_bit(X86_FEATURE_UP, cpu_data[0].x86_capability);
+		alternatives_smp_apply(__smp_alt_instructions,
+				       __smp_alt_instructions_end);
+		list_for_each_entry(mod, &smp_alt_modules, next)
+			alternatives_smp_lock(mod->locks, mod->locks_end,
+					      mod->text, mod->text_end);
+	} else {
+		printk(KERN_INFO "SMP alternatives: switching to UP code\n");
+		set_bit(X86_FEATURE_UP, boot_cpu_data.x86_capability);
+		set_bit(X86_FEATURE_UP, cpu_data[0].x86_capability);
+		apply_alternatives(__smp_alt_instructions,
+				   __smp_alt_instructions_end);
+		list_for_each_entry(mod, &smp_alt_modules, next)
+			alternatives_smp_unlock(mod->locks, mod->locks_end,
+						mod->text, mod->text_end);
+	}
+	spin_unlock_irqrestore(&smp_alt, flags);
+}
+
+void __init alternative_instructions(void)
+{
+	apply_alternatives(__alt_instructions, __alt_instructions_end);
+
+	/* switch to patch-once-at-boottime-only mode and free the
+	 * tables in case we know the number of CPUs will never ever
+	 * change */
+#ifdef CONFIG_HOTPLUG_CPU
+	if (num_possible_cpus() < 2)
+		smp_alt_once = 1;
+#else
+	smp_alt_once = 1;
+#endif
+
+	if (smp_alt_once) {
+		if (1 == num_possible_cpus()) {
+			printk(KERN_INFO "SMP alternatives: switching to UP code\n");
+			set_bit(X86_FEATURE_UP, boot_cpu_data.x86_capability);
+			set_bit(X86_FEATURE_UP, cpu_data[0].x86_capability);
+			apply_alternatives(__smp_alt_instructions,
+					   __smp_alt_instructions_end);
+			alternatives_smp_unlock(__smp_locks, __smp_locks_end,
+						_text, _etext);
+		}
+		free_init_pages("SMP alternatives",
+				(unsigned long)__smp_alt_begin,
+				(unsigned long)__smp_alt_end);
+	} else {
+		alternatives_smp_save(__smp_alt_instructions,
+				      __smp_alt_instructions_end);
+		alternatives_smp_module_add(NULL, "core kernel",
+					    __smp_locks, __smp_locks_end,
+					    _text, _etext);
+		alternatives_smp_switch(0);
+	}
+}
diff --git a/arch/i386/kernel/apic.c b/arch/i386/kernel/apic.c
index 776c90989e0..eb5279d23b7 100644
--- a/arch/i386/kernel/apic.c
+++ b/arch/i386/kernel/apic.c
@@ -38,6 +38,7 @@
 #include <asm/i8253.h>
 
 #include <mach_apic.h>
+#include <mach_apicdef.h>
 #include <mach_ipi.h>
 
 #include "io_ports.h"
diff --git a/arch/i386/kernel/cpu/centaur.c b/arch/i386/kernel/cpu/centaur.c
index f52669ecb93..bd75629dd26 100644
--- a/arch/i386/kernel/cpu/centaur.c
+++ b/arch/i386/kernel/cpu/centaur.c
@@ -4,6 +4,7 @@
 #include <asm/processor.h>
 #include <asm/msr.h>
 #include <asm/e820.h>
+#include <asm/mtrr.h>
 #include "cpu.h"
 
 #ifdef CONFIG_X86_OOSTORE
diff --git a/arch/i386/kernel/cpu/common.c b/arch/i386/kernel/cpu/common.c
index e6bd095ae10..7e3d6b6a4e9 100644
--- a/arch/i386/kernel/cpu/common.c
+++ b/arch/i386/kernel/cpu/common.c
@@ -25,9 +25,10 @@ EXPORT_PER_CPU_SYMBOL(cpu_gdt_descr);
 DEFINE_PER_CPU(unsigned char, cpu_16bit_stack[CPU_16BIT_STACK_SIZE]);
 EXPORT_PER_CPU_SYMBOL(cpu_16bit_stack);
 
-static int cachesize_override __devinitdata = -1;
-static int disable_x86_fxsr __devinitdata = 0;
-static int disable_x86_serial_nr __devinitdata = 1;
+static int cachesize_override __cpuinitdata = -1;
+static int disable_x86_fxsr __cpuinitdata;
+static int disable_x86_serial_nr __cpuinitdata = 1;
+static int disable_x86_sep __cpuinitdata;
 
 struct cpu_dev * cpu_devs[X86_VENDOR_NUM] = {};
 
@@ -59,7 +60,7 @@ static int __init cachesize_setup(char *str)
 }
 __setup("cachesize=", cachesize_setup);
 
-int __devinit get_model_name(struct cpuinfo_x86 *c)
+int __cpuinit get_model_name(struct cpuinfo_x86 *c)
 {
 	unsigned int *v;
 	char *p, *q;
@@ -89,7 +90,7 @@ int __devinit get_model_name(struct cpuinfo_x86 *c)
 }
 
 
-void __devinit display_cacheinfo(struct cpuinfo_x86 *c)
+void __cpuinit display_cacheinfo(struct cpuinfo_x86 *c)
 {
 	unsigned int n, dummy, ecx, edx, l2size;
 
@@ -130,7 +131,7 @@ void __devinit display_cacheinfo(struct cpuinfo_x86 *c)
 /* in particular, if CPUID levels 0x80000002..4 are supported, this isn't used */
 
 /* Look up CPU names by table lookup. */
-static char __devinit *table_lookup_model(struct cpuinfo_x86 *c)
+static char __cpuinit *table_lookup_model(struct cpuinfo_x86 *c)
 {
 	struct cpu_model_info *info;
 
@@ -151,7 +152,7 @@ static char __devinit *table_lookup_model(struct cpuinfo_x86 *c)
 }
 
 
-static void __devinit get_cpu_vendor(struct cpuinfo_x86 *c, int early)
+static void __cpuinit get_cpu_vendor(struct cpuinfo_x86 *c, int early)
 {
 	char *v = c->x86_vendor_id;
 	int i;
@@ -187,6 +188,14 @@ static int __init x86_fxsr_setup(char * s)
 __setup("nofxsr", x86_fxsr_setup);
 
 
+static int __init x86_sep_setup(char * s)
+{
+	disable_x86_sep = 1;
+	return 1;
+}
+__setup("nosep", x86_sep_setup);
+
+
 /* Standard macro to see if a specific flag is changeable */
 static inline int flag_is_changeable_p(u32 flag)
 {
@@ -210,7 +219,7 @@ static inline int flag_is_changeable_p(u32 flag)
 
 
 /* Probe for the CPUID instruction */
-static int __devinit have_cpuid_p(void)
+static int __cpuinit have_cpuid_p(void)
 {
 	return flag_is_changeable_p(X86_EFLAGS_ID);
 }
@@ -254,7 +263,7 @@ static void __init early_cpu_detect(void)
 	}
 }
 
-void __devinit generic_identify(struct cpuinfo_x86 * c)
+void __cpuinit generic_identify(struct cpuinfo_x86 * c)
 {
 	u32 tfms, xlvl;
 	int junk;
@@ -307,7 +316,7 @@ void __devinit generic_identify(struct cpuinfo_x86 * c)
 #endif
 }
 
-static void __devinit squash_the_stupid_serial_number(struct cpuinfo_x86 *c)
+static void __cpuinit squash_the_stupid_serial_number(struct cpuinfo_x86 *c)
 {
 	if (cpu_has(c, X86_FEATURE_PN) && disable_x86_serial_nr ) {
 		/* Disable processor serial number */
@@ -335,7 +344,7 @@ __setup("serialnumber", x86_serial_nr_setup);
 /*
  * This does the hard work of actually picking apart the CPU stuff...
  */
-void __devinit identify_cpu(struct cpuinfo_x86 *c)
+void __cpuinit identify_cpu(struct cpuinfo_x86 *c)
 {
 	int i;
 
@@ -405,6 +414,10 @@ void __devinit identify_cpu(struct cpuinfo_x86 *c)
 		clear_bit(X86_FEATURE_XMM, c->x86_capability);
 	}
 
+	/* SEP disabled? */
+	if (disable_x86_sep)
+		clear_bit(X86_FEATURE_SEP, c->x86_capability);
+
 	if (disable_pse)
 		clear_bit(X86_FEATURE_PSE, c->x86_capability);
 
@@ -417,7 +430,7 @@ void __devinit identify_cpu(struct cpuinfo_x86 *c)
 		else
 			/* Last resort... */
 			sprintf(c->x86_model_id, "%02x/%02x",
-				c->x86_vendor, c->x86_model);
+				c->x86, c->x86_model);
 	}
 
 	/* Now the feature flags better reflect actual CPU features! */
@@ -453,7 +466,7 @@ void __devinit identify_cpu(struct cpuinfo_x86 *c)
 }
 
 #ifdef CONFIG_X86_HT
-void __devinit detect_ht(struct cpuinfo_x86 *c)
+void __cpuinit detect_ht(struct cpuinfo_x86 *c)
 {
 	u32 	eax, ebx, ecx, edx;
 	int 	index_msb, core_bits;
@@ -500,7 +513,7 @@ void __devinit detect_ht(struct cpuinfo_x86 *c)
 }
 #endif
 
-void __devinit print_cpu_info(struct cpuinfo_x86 *c)
+void __cpuinit print_cpu_info(struct cpuinfo_x86 *c)
 {
 	char *vendor = NULL;
 
@@ -523,7 +536,7 @@ void __devinit print_cpu_info(struct cpuinfo_x86 *c)
 		printk("\n");
 }
 
-cpumask_t cpu_initialized __devinitdata = CPU_MASK_NONE;
+cpumask_t cpu_initialized __cpuinitdata = CPU_MASK_NONE;
 
 /* This is hacky. :)
  * We're emulating future behavior.
@@ -570,7 +583,7 @@ void __init early_cpu_init(void)
  * and IDT. We reload them nevertheless, this function acts as a
  * 'CPU state barrier', nothing should get across.
  */
-void __devinit cpu_init(void)
+void __cpuinit cpu_init(void)
 {
 	int cpu = smp_processor_id();
 	struct tss_struct * t = &per_cpu(init_tss, cpu);
@@ -670,7 +683,7 @@ void __devinit cpu_init(void)
 }
 
 #ifdef CONFIG_HOTPLUG_CPU
-void __devinit cpu_uninit(void)
+void __cpuinit cpu_uninit(void)
 {
 	int cpu = raw_smp_processor_id();
 	cpu_clear(cpu, cpu_initialized);
diff --git a/arch/i386/kernel/cpu/cpufreq/powernow-k8.c b/arch/i386/kernel/cpu/cpufreq/powernow-k8.c
index e11a09207ec..3d5110b65cc 100644
--- a/arch/i386/kernel/cpu/cpufreq/powernow-k8.c
+++ b/arch/i386/kernel/cpu/cpufreq/powernow-k8.c
@@ -1145,9 +1145,7 @@ static int __cpuinit powernowk8_init(void)
 {
 	unsigned int i, supported_cpus = 0;
 
-	for (i=0; i<NR_CPUS; i++) {
-		if (!cpu_online(i))
-			continue;
+	for_each_cpu(i) {
 		if (check_supported_cpu(i))
 			supported_cpus++;
 	}
diff --git a/arch/i386/kernel/cpu/intel.c b/arch/i386/kernel/cpu/intel.c
index 8c0120186b9..5386b29bb5a 100644
--- a/arch/i386/kernel/cpu/intel.c
+++ b/arch/i386/kernel/cpu/intel.c
@@ -29,7 +29,7 @@ extern int trap_init_f00f_bug(void);
 struct movsl_mask movsl_mask __read_mostly;
 #endif
 
-void __devinit early_intel_workaround(struct cpuinfo_x86 *c)
+void __cpuinit early_intel_workaround(struct cpuinfo_x86 *c)
 {
 	if (c->x86_vendor != X86_VENDOR_INTEL)
 		return;
@@ -44,7 +44,7 @@ void __devinit early_intel_workaround(struct cpuinfo_x86 *c)
  *	This is called before we do cpu ident work
  */
  
-int __devinit ppro_with_ram_bug(void)
+int __cpuinit ppro_with_ram_bug(void)
 {
 	/* Uses data from early_cpu_detect now */
 	if (boot_cpu_data.x86_vendor == X86_VENDOR_INTEL &&
@@ -62,7 +62,7 @@ int __devinit ppro_with_ram_bug(void)
  * P4 Xeon errata 037 workaround.
  * Hardware prefetcher may cause stale data to be loaded into the cache.
  */
-static void __devinit Intel_errata_workarounds(struct cpuinfo_x86 *c)
+static void __cpuinit Intel_errata_workarounds(struct cpuinfo_x86 *c)
 {
 	unsigned long lo, hi;
 
@@ -81,7 +81,7 @@ static void __devinit Intel_errata_workarounds(struct cpuinfo_x86 *c)
 /*
  * find out the number of processor cores on the die
  */
-static int __devinit num_cpu_cores(struct cpuinfo_x86 *c)
+static int __cpuinit num_cpu_cores(struct cpuinfo_x86 *c)
 {
 	unsigned int eax, ebx, ecx, edx;
 
@@ -96,7 +96,7 @@ static int __devinit num_cpu_cores(struct cpuinfo_x86 *c)
 		return 1;
 }
 
-static void __devinit init_intel(struct cpuinfo_x86 *c)
+static void __cpuinit init_intel(struct cpuinfo_x86 *c)
 {
 	unsigned int l2 = 0;
 	char *p = NULL;
@@ -205,7 +205,7 @@ static unsigned int intel_size_cache(struct cpuinfo_x86 * c, unsigned int size)
 	return size;
 }
 
-static struct cpu_dev intel_cpu_dev __devinitdata = {
+static struct cpu_dev intel_cpu_dev __cpuinitdata = {
 	.c_vendor	= "Intel",
 	.c_ident 	= { "GenuineIntel" },
 	.c_models = {
diff --git a/arch/i386/kernel/cpu/intel_cacheinfo.c b/arch/i386/kernel/cpu/intel_cacheinfo.c
index ffe58cee0c4..ce61921369e 100644
--- a/arch/i386/kernel/cpu/intel_cacheinfo.c
+++ b/arch/i386/kernel/cpu/intel_cacheinfo.c
@@ -174,7 +174,7 @@ unsigned int __cpuinit init_intel_cacheinfo(struct cpuinfo_x86 *c)
 	unsigned int new_l1d = 0, new_l1i = 0; /* Cache sizes from cpuid(4) */
 	unsigned int new_l2 = 0, new_l3 = 0, i; /* Cache sizes from cpuid(4) */
 
-	if (c->cpuid_level > 4) {
+	if (c->cpuid_level > 3) {
 		static int is_initialized;
 
 		if (is_initialized == 0) {
@@ -330,7 +330,7 @@ static void __cpuinit cache_shared_cpu_map_setup(unsigned int cpu, int index)
 		}
 	}
 }
-static void __devinit cache_remove_shared_cpu_map(unsigned int cpu, int index)
+static void __cpuinit cache_remove_shared_cpu_map(unsigned int cpu, int index)
 {
 	struct _cpuid4_info	*this_leaf, *sibling_leaf;
 	int sibling;
diff --git a/arch/i386/kernel/cpu/proc.c b/arch/i386/kernel/cpu/proc.c
index 89a85af33d2..5cfbd801169 100644
--- a/arch/i386/kernel/cpu/proc.c
+++ b/arch/i386/kernel/cpu/proc.c
@@ -40,7 +40,7 @@ static int show_cpuinfo(struct seq_file *m, void *v)
 		/* Other (Linux-defined) */
 		"cxmmx", "k6_mtrr", "cyrix_arr", "centaur_mcr",
 		NULL, NULL, NULL, NULL,
-		"constant_tsc", NULL, NULL, NULL, NULL, NULL, NULL, NULL,
+		"constant_tsc", "up", NULL, NULL, NULL, NULL, NULL, NULL,
 		NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL,
 		NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL,
 
diff --git a/arch/i386/kernel/crash.c b/arch/i386/kernel/crash.c
index d49dbe8dc96..e3c5fca0aa8 100644
--- a/arch/i386/kernel/crash.c
+++ b/arch/i386/kernel/crash.c
@@ -105,7 +105,7 @@ static int crash_nmi_callback(struct pt_regs *regs, int cpu)
 		return 1;
 	local_irq_disable();
 
-	if (!user_mode(regs)) {
+	if (!user_mode_vm(regs)) {
 		crash_fixup_ss_esp(&fixed_regs, regs);
 		regs = &fixed_regs;
 	}
diff --git a/arch/i386/kernel/entry.S b/arch/i386/kernel/entry.S
index 4d704724b2f..cfc683f153b 100644
--- a/arch/i386/kernel/entry.S
+++ b/arch/i386/kernel/entry.S
@@ -226,6 +226,10 @@ ENTRY(system_call)
 	pushl %eax			# save orig_eax
 	SAVE_ALL
 	GET_THREAD_INFO(%ebp)
+	testl $TF_MASK,EFLAGS(%esp)
+	jz no_singlestep
+	orl $_TIF_SINGLESTEP,TI_flags(%ebp)
+no_singlestep:
 					# system call tracing in operation / emulation
 	/* Note, _TIF_SECCOMP is bit number 8, and so it needs testw and not testb */
 	testw $(_TIF_SYSCALL_EMU|_TIF_SYSCALL_TRACE|_TIF_SECCOMP|_TIF_SYSCALL_AUDIT),TI_flags(%ebp)
diff --git a/arch/i386/kernel/head.S b/arch/i386/kernel/head.S
index e0b7c632efb..3debc2e2654 100644
--- a/arch/i386/kernel/head.S
+++ b/arch/i386/kernel/head.S
@@ -450,7 +450,6 @@ int_msg:
 
 .globl boot_gdt_descr
 .globl idt_descr
-.globl cpu_gdt_descr
 
 	ALIGN
 # early boot GDT descriptor (must use 1:1 address mapping)
@@ -470,8 +469,6 @@ cpu_gdt_descr:
 	.word GDT_ENTRIES*8-1
 	.long cpu_gdt_table
 
-	.fill NR_CPUS-1,8,0		# space for the other GDT descriptors
-
 /*
  * The boot_gdt_table must mirror the equivalent in setup.S and is
  * used only for booting.
@@ -485,7 +482,7 @@ ENTRY(boot_gdt_table)
 /*
  * The Global Descriptor Table contains 28 quadwords, per-CPU.
  */
-	.align PAGE_SIZE_asm
+	.align L1_CACHE_BYTES
 ENTRY(cpu_gdt_table)
 	.quad 0x0000000000000000	/* NULL descriptor */
 	.quad 0x0000000000000000	/* 0x0b reserved */
diff --git a/arch/i386/kernel/io_apic.c b/arch/i386/kernel/io_apic.c
index 39d9a5fa907..311b4e7266f 100644
--- a/arch/i386/kernel/io_apic.c
+++ b/arch/i386/kernel/io_apic.c
@@ -351,8 +351,8 @@ static inline void rotate_irqs_among_cpus(unsigned long useful_load_threshold)
 {
 	int i, j;
 	Dprintk("Rotating IRQs among CPUs.\n");
-	for (i = 0; i < NR_CPUS; i++) {
-		for (j = 0; cpu_online(i) && (j < NR_IRQS); j++) {
+	for_each_online_cpu(i) {
+		for (j = 0; j < NR_IRQS; j++) {
 			if (!irq_desc[j].action)
 				continue;
 			/* Is it a significant load ?  */
@@ -381,7 +381,7 @@ static void do_irq_balance(void)
 	unsigned long imbalance = 0;
 	cpumask_t allowed_mask, target_cpu_mask, tmp;
 
-	for (i = 0; i < NR_CPUS; i++) {
+	for_each_cpu(i) {
 		int package_index;
 		CPU_IRQ(i) = 0;
 		if (!cpu_online(i))
@@ -422,9 +422,7 @@ static void do_irq_balance(void)
 		}
 	}
 	/* Find the least loaded processor package */
-	for (i = 0; i < NR_CPUS; i++) {
-		if (!cpu_online(i))
-			continue;
+	for_each_online_cpu(i) {
 		if (i != CPU_TO_PACKAGEINDEX(i))
 			continue;
 		if (min_cpu_irq > CPU_IRQ(i)) {
@@ -441,9 +439,7 @@ tryanothercpu:
 	 */
 	tmp_cpu_irq = 0;
 	tmp_loaded = -1;
-	for (i = 0; i < NR_CPUS; i++) {
-		if (!cpu_online(i))
-			continue;
+	for_each_online_cpu(i) {
 		if (i != CPU_TO_PACKAGEINDEX(i))
 			continue;
 		if (max_cpu_irq <= CPU_IRQ(i)) 
@@ -619,9 +615,7 @@ static int __init balanced_irq_init(void)
 	if (smp_num_siblings > 1 && !cpus_empty(tmp))
 		physical_balance = 1;
 
-	for (i = 0; i < NR_CPUS; i++) {
-		if (!cpu_online(i))
-			continue;
+	for_each_online_cpu(i) {
 		irq_cpu_data[i].irq_delta = kmalloc(sizeof(unsigned long) * NR_IRQS, GFP_KERNEL);
 		irq_cpu_data[i].last_irq = kmalloc(sizeof(unsigned long) * NR_IRQS, GFP_KERNEL);
 		if (irq_cpu_data[i].irq_delta == NULL || irq_cpu_data[i].last_irq == NULL) {
@@ -638,9 +632,11 @@ static int __init balanced_irq_init(void)
 	else 
 		printk(KERN_ERR "balanced_irq_init: failed to spawn balanced_irq");
 failed:
-	for (i = 0; i < NR_CPUS; i++) {
+	for_each_cpu(i) {
 		kfree(irq_cpu_data[i].irq_delta);
+		irq_cpu_data[i].irq_delta = NULL;
 		kfree(irq_cpu_data[i].last_irq);
+		irq_cpu_data[i].last_irq = NULL;
 	}
 	return 0;
 }
@@ -1761,7 +1757,8 @@ static void __init setup_ioapic_ids_from_mpc(void)
 	 * Don't check I/O APIC IDs for xAPIC systems.  They have
 	 * no meaning without the serial APIC bus.
 	 */
-	if (!(boot_cpu_data.x86_vendor == X86_VENDOR_INTEL && boot_cpu_data.x86 < 15))
+	if (!(boot_cpu_data.x86_vendor == X86_VENDOR_INTEL)
+		|| APIC_XAPIC(apic_version[boot_cpu_physical_apicid]))
 		return;
 	/*
 	 * This is broken; anything with a real cpu count has to
diff --git a/arch/i386/kernel/kprobes.c b/arch/i386/kernel/kprobes.c
index 694a1399763..7a59050242a 100644
--- a/arch/i386/kernel/kprobes.c
+++ b/arch/i386/kernel/kprobes.c
@@ -84,9 +84,9 @@ void __kprobes arch_disarm_kprobe(struct kprobe *p)
 
 void __kprobes arch_remove_kprobe(struct kprobe *p)
 {
-	down(&kprobe_mutex);
+	mutex_lock(&kprobe_mutex);
 	free_insn_slot(p->ainsn.insn);
-	up(&kprobe_mutex);
+	mutex_unlock(&kprobe_mutex);
 }
 
 static inline void save_previous_kprobe(struct kprobe_ctlblk *kcb)
diff --git a/arch/i386/kernel/module.c b/arch/i386/kernel/module.c
index 5149c8a621f..470cf97e7cd 100644
--- a/arch/i386/kernel/module.c
+++ b/arch/i386/kernel/module.c
@@ -104,26 +104,38 @@ int apply_relocate_add(Elf32_Shdr *sechdrs,
 	return -ENOEXEC;
 }
 
-extern void apply_alternatives(void *start, void *end); 
-
 int module_finalize(const Elf_Ehdr *hdr,
 		    const Elf_Shdr *sechdrs,
 		    struct module *me)
 {
-	const Elf_Shdr *s;
+	const Elf_Shdr *s, *text = NULL, *alt = NULL, *locks = NULL;
 	char *secstrings = (void *)hdr + sechdrs[hdr->e_shstrndx].sh_offset;
 
-	/* look for .altinstructions to patch */ 
 	for (s = sechdrs; s < sechdrs + hdr->e_shnum; s++) { 
-		void *seg; 		
-		if (strcmp(".altinstructions", secstrings + s->sh_name))
-			continue;
-		seg = (void *)s->sh_addr; 
-		apply_alternatives(seg, seg + s->sh_size); 
-	} 	
+		if (!strcmp(".text", secstrings + s->sh_name))
+			text = s;
+		if (!strcmp(".altinstructions", secstrings + s->sh_name))
+			alt = s;
+		if (!strcmp(".smp_locks", secstrings + s->sh_name))
+			locks= s;
+	}
+
+	if (alt) {
+		/* patch .altinstructions */
+		void *aseg = (void *)alt->sh_addr;
+		apply_alternatives(aseg, aseg + alt->sh_size);
+	}
+	if (locks && text) {
+		void *lseg = (void *)locks->sh_addr;
+		void *tseg = (void *)text->sh_addr;
+		alternatives_smp_module_add(me, me->name,
+					    lseg, lseg + locks->sh_size,
+					    tseg, tseg + text->sh_size);
+	}
 	return 0;
 }
 
 void module_arch_cleanup(struct module *mod)
 {
+	alternatives_smp_module_del(mod);
 }
diff --git a/arch/i386/kernel/mpparse.c b/arch/i386/kernel/mpparse.c
index e6e2f43db85..8d8aa9d1796 100644
--- a/arch/i386/kernel/mpparse.c
+++ b/arch/i386/kernel/mpparse.c
@@ -828,6 +828,8 @@ void __init find_smp_config (void)
 		smp_scan_config(address, 0x400);
 }
 
+int es7000_plat;
+
 /* --------------------------------------------------------------------------
                             ACPI-based MP Configuration
    -------------------------------------------------------------------------- */
@@ -935,7 +937,8 @@ void __init mp_register_ioapic (
 	mp_ioapics[idx].mpc_apicaddr = address;
 
 	set_fixmap_nocache(FIX_IO_APIC_BASE_0 + idx, address);
-	if ((boot_cpu_data.x86_vendor == X86_VENDOR_INTEL) && (boot_cpu_data.x86 < 15))
+	if ((boot_cpu_data.x86_vendor == X86_VENDOR_INTEL)
+		&& !APIC_XAPIC(apic_version[boot_cpu_physical_apicid]))
 		tmpid = io_apic_get_unique_id(idx, id);
 	else
 		tmpid = id;
@@ -1011,8 +1014,6 @@ void __init mp_override_legacy_irq (
 	return;
 }
 
-int es7000_plat;
-
 void __init mp_config_acpi_legacy_irqs (void)
 {
 	struct mpc_config_intsrc intsrc;
diff --git a/arch/i386/kernel/nmi.c b/arch/i386/kernel/nmi.c
index be87c5e2ee9..9074818b947 100644
--- a/arch/i386/kernel/nmi.c
+++ b/arch/i386/kernel/nmi.c
@@ -143,7 +143,7 @@ static int __init check_nmi_watchdog(void)
 	local_irq_enable();
 	mdelay((10*1000)/nmi_hz); // wait 10 ticks
 
-	for (cpu = 0; cpu < NR_CPUS; cpu++) {
+	for_each_cpu(cpu) {
 #ifdef CONFIG_SMP
 		/* Check cpu_callin_map here because that is set
 		   after the timer is started. */
@@ -510,7 +510,7 @@ void touch_nmi_watchdog (void)
 	 * Just reset the alert counters, (other CPUs might be
 	 * spinning on locks we hold):
 	 */
-	for (i = 0; i < NR_CPUS; i++)
+	for_each_cpu(i)
 		alert_counter[i] = 0;
 
 	/*
@@ -543,7 +543,7 @@ void nmi_watchdog_tick (struct pt_regs * regs)
 			/*
 			 * die_nmi will return ONLY if NOTIFY_STOP happens..
 			 */
-			die_nmi(regs, "NMI Watchdog detected LOCKUP");
+			die_nmi(regs, "BUG: NMI Watchdog detected LOCKUP");
 	} else {
 		last_irq_sums[cpu] = sum;
 		alert_counter[cpu] = 0;
diff --git a/arch/i386/kernel/process.c b/arch/i386/kernel/process.c
index 0480454ebff..299e6167408 100644
--- a/arch/i386/kernel/process.c
+++ b/arch/i386/kernel/process.c
@@ -295,7 +295,7 @@ void show_regs(struct pt_regs * regs)
 	printk("EIP: %04x:[<%08lx>] CPU: %d\n",0xffff & regs->xcs,regs->eip, smp_processor_id());
 	print_symbol("EIP is at %s\n", regs->eip);
 
-	if (user_mode(regs))
+	if (user_mode_vm(regs))
 		printk(" ESP: %04x:%08lx",0xffff & regs->xss,regs->esp);
 	printk(" EFLAGS: %08lx    %s  (%s %.*s)\n",
 	       regs->eflags, print_tainted(), system_utsname.release,
diff --git a/arch/i386/kernel/ptrace.c b/arch/i386/kernel/ptrace.c
index 5c1fb6aada5..506462ef36a 100644
--- a/arch/i386/kernel/ptrace.c
+++ b/arch/i386/kernel/ptrace.c
@@ -34,10 +34,10 @@
 
 /*
  * Determines which flags the user has access to [1 = access, 0 = no access].
- * Prohibits changing ID(21), VIP(20), VIF(19), VM(17), IOPL(12-13), IF(9).
+ * Prohibits changing ID(21), VIP(20), VIF(19), VM(17), NT(14), IOPL(12-13), IF(9).
  * Also masks reserved bits (31-22, 15, 5, 3, 1).
  */
-#define FLAG_MASK 0x00054dd5
+#define FLAG_MASK 0x00050dd5
 
 /* set's the trap flag. */
 #define TRAP_FLAG 0x100
diff --git a/arch/i386/kernel/semaphore.c b/arch/i386/kernel/semaphore.c
index 7455ab64394..967dc74df9e 100644
--- a/arch/i386/kernel/semaphore.c
+++ b/arch/i386/kernel/semaphore.c
@@ -110,11 +110,11 @@ asm(
 ".align	4\n"
 ".globl	__write_lock_failed\n"
 "__write_lock_failed:\n\t"
-	LOCK "addl	$" RW_LOCK_BIAS_STR ",(%eax)\n"
+	LOCK_PREFIX "addl	$" RW_LOCK_BIAS_STR ",(%eax)\n"
 "1:	rep; nop\n\t"
 	"cmpl	$" RW_LOCK_BIAS_STR ",(%eax)\n\t"
 	"jne	1b\n\t"
-	LOCK "subl	$" RW_LOCK_BIAS_STR ",(%eax)\n\t"
+	LOCK_PREFIX "subl	$" RW_LOCK_BIAS_STR ",(%eax)\n\t"
 	"jnz	__write_lock_failed\n\t"
 	"ret"
 );
@@ -124,11 +124,11 @@ asm(
 ".align	4\n"
 ".globl	__read_lock_failed\n"
 "__read_lock_failed:\n\t"
-	LOCK "incl	(%eax)\n"
+	LOCK_PREFIX "incl	(%eax)\n"
 "1:	rep; nop\n\t"
 	"cmpl	$1,(%eax)\n\t"
 	"js	1b\n\t"
-	LOCK "decl	(%eax)\n\t"
+	LOCK_PREFIX "decl	(%eax)\n\t"
 	"js	__read_lock_failed\n\t"
 	"ret"
 );
diff --git a/arch/i386/kernel/setup.c b/arch/i386/kernel/setup.c
index ab62a9f4701..2d8782960f4 100644
--- a/arch/i386/kernel/setup.c
+++ b/arch/i386/kernel/setup.c
@@ -1377,101 +1377,6 @@ static void __init register_memory(void)
 		pci_mem_start, gapstart, gapsize);
 }
 
-/* Use inline assembly to define this because the nops are defined 
-   as inline assembly strings in the include files and we cannot 
-   get them easily into strings. */
-asm("\t.data\nintelnops: " 
-    GENERIC_NOP1 GENERIC_NOP2 GENERIC_NOP3 GENERIC_NOP4 GENERIC_NOP5 GENERIC_NOP6
-    GENERIC_NOP7 GENERIC_NOP8); 
-asm("\t.data\nk8nops: " 
-    K8_NOP1 K8_NOP2 K8_NOP3 K8_NOP4 K8_NOP5 K8_NOP6
-    K8_NOP7 K8_NOP8); 
-asm("\t.data\nk7nops: " 
-    K7_NOP1 K7_NOP2 K7_NOP3 K7_NOP4 K7_NOP5 K7_NOP6
-    K7_NOP7 K7_NOP8); 
-    
-extern unsigned char intelnops[], k8nops[], k7nops[];
-static unsigned char *intel_nops[ASM_NOP_MAX+1] = { 
-     NULL,
-     intelnops,
-     intelnops + 1,
-     intelnops + 1 + 2,
-     intelnops + 1 + 2 + 3,
-     intelnops + 1 + 2 + 3 + 4,
-     intelnops + 1 + 2 + 3 + 4 + 5,
-     intelnops + 1 + 2 + 3 + 4 + 5 + 6,
-     intelnops + 1 + 2 + 3 + 4 + 5 + 6 + 7,
-}; 
-static unsigned char *k8_nops[ASM_NOP_MAX+1] = { 
-     NULL,
-     k8nops,
-     k8nops + 1,
-     k8nops + 1 + 2,
-     k8nops + 1 + 2 + 3,
-     k8nops + 1 + 2 + 3 + 4,
-     k8nops + 1 + 2 + 3 + 4 + 5,
-     k8nops + 1 + 2 + 3 + 4 + 5 + 6,
-     k8nops + 1 + 2 + 3 + 4 + 5 + 6 + 7,
-}; 
-static unsigned char *k7_nops[ASM_NOP_MAX+1] = { 
-     NULL,
-     k7nops,
-     k7nops + 1,
-     k7nops + 1 + 2,
-     k7nops + 1 + 2 + 3,
-     k7nops + 1 + 2 + 3 + 4,
-     k7nops + 1 + 2 + 3 + 4 + 5,
-     k7nops + 1 + 2 + 3 + 4 + 5 + 6,
-     k7nops + 1 + 2 + 3 + 4 + 5 + 6 + 7,
-}; 
-static struct nop { 
-     int cpuid; 
-     unsigned char **noptable; 
-} noptypes[] = { 
-     { X86_FEATURE_K8, k8_nops }, 
-     { X86_FEATURE_K7, k7_nops }, 
-     { -1, NULL }
-}; 
-
-/* Replace instructions with better alternatives for this CPU type.
-
-   This runs before SMP is initialized to avoid SMP problems with
-   self modifying code. This implies that assymetric systems where
-   APs have less capabilities than the boot processor are not handled. 
-   Tough. Make sure you disable such features by hand. */ 
-void apply_alternatives(void *start, void *end) 
-{ 
-	struct alt_instr *a; 
-	int diff, i, k;
-        unsigned char **noptable = intel_nops; 
-	for (i = 0; noptypes[i].cpuid >= 0; i++) { 
-		if (boot_cpu_has(noptypes[i].cpuid)) { 
-			noptable = noptypes[i].noptable;
-			break;
-		}
-	} 
-	for (a = start; (void *)a < end; a++) { 
-		if (!boot_cpu_has(a->cpuid))
-			continue;
-		BUG_ON(a->replacementlen > a->instrlen); 
-		memcpy(a->instr, a->replacement, a->replacementlen); 
-		diff = a->instrlen - a->replacementlen; 
-		/* Pad the rest with nops */
-		for (i = a->replacementlen; diff > 0; diff -= k, i += k) {
-			k = diff;
-			if (k > ASM_NOP_MAX)
-				k = ASM_NOP_MAX;
-			memcpy(a->instr + i, noptable[k], k); 
-		} 
-	}
-} 
-
-void __init alternative_instructions(void)
-{
-	extern struct alt_instr __alt_instructions[], __alt_instructions_end[];
-	apply_alternatives(__alt_instructions, __alt_instructions_end);
-}
-
 static char * __init machine_specific_memory_setup(void);
 
 #ifdef CONFIG_MCA
@@ -1554,6 +1459,16 @@ void __init setup_arch(char **cmdline_p)
 
 	parse_cmdline_early(cmdline_p);
 
+#ifdef CONFIG_EARLY_PRINTK
+	{
+		char *s = strstr(*cmdline_p, "earlyprintk=");
+		if (s) {
+			setup_early_printk(strchr(s, '=') + 1);
+			printk("early console enabled\n");
+		}
+	}
+#endif
+
 	max_low_pfn = setup_memory();
 
 	/*
@@ -1578,19 +1493,6 @@ void __init setup_arch(char **cmdline_p)
 	 * NOTE: at this point the bootmem allocator is fully available.
 	 */
 
-#ifdef CONFIG_EARLY_PRINTK
-	{
-		char *s = strstr(*cmdline_p, "earlyprintk=");
-		if (s) {
-			extern void setup_early_printk(char *);
-
-			setup_early_printk(strchr(s, '=') + 1);
-			printk("early console enabled\n");
-		}
-	}
-#endif
-
-
 	dmi_scan_machine();
 
 #ifdef CONFIG_X86_GENERICARCH
diff --git a/arch/i386/kernel/signal.c b/arch/i386/kernel/signal.c
index 963616d364e..5c352c3a9e7 100644
--- a/arch/i386/kernel/signal.c
+++ b/arch/i386/kernel/signal.c
@@ -123,7 +123,8 @@ restore_sigcontext(struct pt_regs *regs, struct sigcontext __user *sc, int *peax
 	  err |= __get_user(tmp, &sc->seg);				\
 	  loadsegment(seg,tmp); }
 
-#define	FIX_EFLAGS	(X86_EFLAGS_AC | X86_EFLAGS_OF | X86_EFLAGS_DF | \
+#define	FIX_EFLAGS	(X86_EFLAGS_AC | X86_EFLAGS_RF |		 \
+			 X86_EFLAGS_OF | X86_EFLAGS_DF |		 \
 			 X86_EFLAGS_TF | X86_EFLAGS_SF | X86_EFLAGS_ZF | \
 			 X86_EFLAGS_AF | X86_EFLAGS_PF | X86_EFLAGS_CF)
 
@@ -582,9 +583,6 @@ static void fastcall do_signal(struct pt_regs *regs)
 	if (!user_mode(regs))
 		return;
 
-	if (try_to_freeze())
-		goto no_signal;
-
 	if (test_thread_flag(TIF_RESTORE_SIGMASK))
 		oldset = &current->saved_sigmask;
 	else
@@ -613,7 +611,6 @@ static void fastcall do_signal(struct pt_regs *regs)
 		return;
 	}
 
-no_signal:
 	/* Did we come from a system call? */
 	if (regs->orig_eax >= 0) {
 		/* Restart the system call - no handlers present */
diff --git a/arch/i386/kernel/smpboot.c b/arch/i386/kernel/smpboot.c
index 7007e178379..4c470e99a74 100644
--- a/arch/i386/kernel/smpboot.c
+++ b/arch/i386/kernel/smpboot.c
@@ -899,6 +899,7 @@ static int __devinit do_boot_cpu(int apicid, int cpu)
 	unsigned short nmi_high = 0, nmi_low = 0;
 
 	++cpucount;
+	alternatives_smp_switch(1);
 
 	/*
 	 * We can't use kernel_thread since we must avoid to
@@ -1368,6 +1369,8 @@ void __cpu_die(unsigned int cpu)
 		/* They ack this in play_dead by setting CPU_DEAD */
 		if (per_cpu(cpu_state, cpu) == CPU_DEAD) {
 			printk ("CPU %d is now offline\n", cpu);
+			if (1 == num_online_cpus())
+				alternatives_smp_switch(0);
 			return;
 		}
 		msleep(100);
diff --git a/arch/i386/kernel/topology.c b/arch/i386/kernel/topology.c
index 67a0e1baa28..296355292c7 100644
--- a/arch/i386/kernel/topology.c
+++ b/arch/i386/kernel/topology.c
@@ -41,6 +41,15 @@ int arch_register_cpu(int num){
 		parent = &node_devices[node].node;
 #endif /* CONFIG_NUMA */
 
+	/*
+	 * CPU0 cannot be offlined due to several
+	 * restrictions and assumptions in kernel. This basically
+	 * doesnt add a control file, one cannot attempt to offline
+	 * BSP.
+	 */
+	if (!num)
+		cpu_devices[num].cpu.no_control = 1;
+
 	return register_cpu(&cpu_devices[num].cpu, num, parent);
 }
 
diff --git a/arch/i386/kernel/traps.c b/arch/i386/kernel/traps.c
index b814dbdcc91..de5386b01d3 100644
--- a/arch/i386/kernel/traps.c
+++ b/arch/i386/kernel/traps.c
@@ -99,6 +99,8 @@ int register_die_notifier(struct notifier_block *nb)
 {
 	int err = 0;
 	unsigned long flags;
+
+	vmalloc_sync_all();
 	spin_lock_irqsave(&die_notifier_lock, flags);
 	err = notifier_chain_register(&i386die_chain, nb);
 	spin_unlock_irqrestore(&die_notifier_lock, flags);
@@ -112,12 +114,30 @@ static inline int valid_stack_ptr(struct thread_info *tinfo, void *p)
 		p < (void *)tinfo + THREAD_SIZE - 3;
 }
 
-static void print_addr_and_symbol(unsigned long addr, char *log_lvl)
+/*
+ * Print CONFIG_STACK_BACKTRACE_COLS address/symbol entries per line.
+ */
+static inline int print_addr_and_symbol(unsigned long addr, char *log_lvl,
+					int printed)
 {
-	printk(log_lvl);
+	if (!printed)
+		printk(log_lvl);
+
+#if CONFIG_STACK_BACKTRACE_COLS == 1
 	printk(" [<%08lx>] ", addr);
+#else
+	printk(" <%08lx> ", addr);
+#endif
 	print_symbol("%s", addr);
-	printk("\n");
+
+	printed = (printed + 1) % CONFIG_STACK_BACKTRACE_COLS;
+
+	if (printed)
+		printk("  ");
+	else
+		printk("\n");
+
+	return printed;
 }
 
 static inline unsigned long print_context_stack(struct thread_info *tinfo,
@@ -125,20 +145,24 @@ static inline unsigned long print_context_stack(struct thread_info *tinfo,
 				char *log_lvl)
 {
 	unsigned long addr;
+	int printed = 0; /* nr of entries already printed on current line */
 
 #ifdef	CONFIG_FRAME_POINTER
 	while (valid_stack_ptr(tinfo, (void *)ebp)) {
 		addr = *(unsigned long *)(ebp + 4);
-		print_addr_and_symbol(addr, log_lvl);
+		printed = print_addr_and_symbol(addr, log_lvl, printed);
 		ebp = *(unsigned long *)ebp;
 	}
 #else
 	while (valid_stack_ptr(tinfo, stack)) {
 		addr = *stack++;
 		if (__kernel_text_address(addr))
-			print_addr_and_symbol(addr, log_lvl);
+			printed = print_addr_and_symbol(addr, log_lvl, printed);
 	}
 #endif
+	if (printed)
+		printk("\n");
+
 	return ebp;
 }
 
@@ -166,8 +190,7 @@ static void show_trace_log_lvl(struct task_struct *task,
 		stack = (unsigned long*)context->previous_esp;
 		if (!stack)
 			break;
-		printk(log_lvl);
-		printk(" =======================\n");
+		printk("%s =======================\n", log_lvl);
 	}
 }
 
@@ -194,21 +217,17 @@ static void show_stack_log_lvl(struct task_struct *task, unsigned long *esp,
 	for(i = 0; i < kstack_depth_to_print; i++) {
 		if (kstack_end(stack))
 			break;
-		if (i && ((i % 8) == 0)) {
-			printk("\n");
-			printk(log_lvl);
-			printk("       ");
-		}
+		if (i && ((i % 8) == 0))
+			printk("\n%s       ", log_lvl);
 		printk("%08lx ", *stack++);
 	}
-	printk("\n");
-	printk(log_lvl);
-	printk("Call Trace:\n");
+	printk("\n%sCall Trace:\n", log_lvl);
 	show_trace_log_lvl(task, esp, log_lvl);
 }
 
 void show_stack(struct task_struct *task, unsigned long *esp)
 {
+	printk("       ");
 	show_stack_log_lvl(task, esp, "");
 }
 
@@ -233,7 +252,7 @@ void show_registers(struct pt_regs *regs)
 
 	esp = (unsigned long) (&regs->esp);
 	savesegment(ss, ss);
-	if (user_mode(regs)) {
+	if (user_mode_vm(regs)) {
 		in_kernel = 0;
 		esp = regs->esp;
 		ss = regs->xss & 0xffff;
@@ -333,6 +352,8 @@ void die(const char * str, struct pt_regs * regs, long err)
 	static int die_counter;
 	unsigned long flags;
 
+	oops_enter();
+
 	if (die.lock_owner != raw_smp_processor_id()) {
 		console_verbose();
 		spin_lock_irqsave(&die.lock, flags);
@@ -385,6 +406,7 @@ void die(const char * str, struct pt_regs * regs, long err)
 		ssleep(5);
 		panic("Fatal exception");
 	}
+	oops_exit();
 	do_exit(SIGSEGV);
 }
 
@@ -623,7 +645,7 @@ void die_nmi (struct pt_regs *regs, const char *msg)
 	/* If we are in kernel we are probably nested up pretty bad
 	 * and might aswell get out now while we still can.
 	*/
-	if (!user_mode(regs)) {
+	if (!user_mode_vm(regs)) {
 		current->thread.trap_no = 2;
 		crash_kexec(regs);
 	}
@@ -694,6 +716,7 @@ fastcall void do_nmi(struct pt_regs * regs, long error_code)
 
 void set_nmi_callback(nmi_callback_t callback)
 {
+	vmalloc_sync_all();
 	rcu_assign_pointer(nmi_callback, callback);
 }
 EXPORT_SYMBOL_GPL(set_nmi_callback);
diff --git a/arch/i386/kernel/vmlinux.lds.S b/arch/i386/kernel/vmlinux.lds.S
index 4710195b6b7..3f21c6f6466 100644
--- a/arch/i386/kernel/vmlinux.lds.S
+++ b/arch/i386/kernel/vmlinux.lds.S
@@ -68,6 +68,26 @@ SECTIONS
 	*(.data.init_task)
   }
 
+  /* might get freed after init */
+  . = ALIGN(4096);
+  __smp_alt_begin = .;
+  __smp_alt_instructions = .;
+  .smp_altinstructions : AT(ADDR(.smp_altinstructions) - LOAD_OFFSET) {
+	*(.smp_altinstructions)
+  }
+  __smp_alt_instructions_end = .;
+  . = ALIGN(4);
+  __smp_locks = .;
+  .smp_locks : AT(ADDR(.smp_locks) - LOAD_OFFSET) {
+	*(.smp_locks)
+  }
+  __smp_locks_end = .;
+  .smp_altinstr_replacement : AT(ADDR(.smp_altinstr_replacement) - LOAD_OFFSET) {
+	*(.smp_altinstr_replacement)
+  }
+  . = ALIGN(4096);
+  __smp_alt_end = .;
+
   /* will be freed after init */
   . = ALIGN(4096);		/* Init code and data */
   __init_begin = .;
diff --git a/arch/i386/kernel/vsyscall-sysenter.S b/arch/i386/kernel/vsyscall-sysenter.S
index 76b72815940..3b62baa6a37 100644
--- a/arch/i386/kernel/vsyscall-sysenter.S
+++ b/arch/i386/kernel/vsyscall-sysenter.S
@@ -21,6 +21,9 @@
  * instruction clobbers %esp, the user's %esp won't even survive entry
  * into the kernel. We store %esp in %ebp. Code in entry.S must fetch
  * arg6 from the stack.
+ *
+ * You can not use this vsyscall for the clone() syscall because the
+ * three dwords on the parent stack do not get copied to the child.
  */
 	.text
 	.globl __kernel_vsyscall
diff --git a/arch/i386/mach-es7000/es7000.h b/arch/i386/mach-es7000/es7000.h
index f1e3204f5de..80566ca4a80 100644
--- a/arch/i386/mach-es7000/es7000.h
+++ b/arch/i386/mach-es7000/es7000.h
@@ -83,6 +83,7 @@ struct es7000_oem_table {
 	struct psai psai;
 };
 
+#ifdef CONFIG_ACPI
 struct acpi_table_sdt {
 	unsigned long pa;
 	unsigned long count;
@@ -99,6 +100,9 @@ struct oem_table {
 	u32 OEMTableSize;
 };
 
+extern int find_unisys_acpi_oem_table(unsigned long *oem_addr);
+#endif
+
 struct mip_reg {
 	unsigned long long off_0;
 	unsigned long long off_8;
@@ -114,7 +118,6 @@ struct mip_reg {
 #define	MIP_FUNC(VALUE) 	(VALUE & 0xff)
 
 extern int parse_unisys_oem (char *oemptr);
-extern int find_unisys_acpi_oem_table(unsigned long *oem_addr);
 extern void setup_unisys(void);
 extern int es7000_start_cpu(int cpu, unsigned long eip);
 extern void es7000_sw_apic(void);
diff --git a/arch/i386/mach-es7000/es7000plat.c b/arch/i386/mach-es7000/es7000plat.c
index a9ab0644f40..3d0fc853516 100644
--- a/arch/i386/mach-es7000/es7000plat.c
+++ b/arch/i386/mach-es7000/es7000plat.c
@@ -51,8 +51,6 @@ struct mip_reg		*host_reg;
 int 			mip_port;
 unsigned long		mip_addr, host_addr;
 
-#if defined(CONFIG_X86_IO_APIC) && defined(CONFIG_ACPI)
-
 /*
  * GSI override for ES7000 platforms.
  */
@@ -76,8 +74,6 @@ es7000_rename_gsi(int ioapic, int gsi)
 	return gsi;
 }
 
-#endif	/* (CONFIG_X86_IO_APIC) && (CONFIG_ACPI) */
-
 void __init
 setup_unisys(void)
 {
@@ -160,6 +156,7 @@ parse_unisys_oem (char *oemptr)
 	return es7000_plat;
 }
 
+#ifdef CONFIG_ACPI
 int __init
 find_unisys_acpi_oem_table(unsigned long *oem_addr)
 {
@@ -212,6 +209,7 @@ find_unisys_acpi_oem_table(unsigned long *oem_addr)
 	}
 	return -1;
 }
+#endif
 
 static void
 es7000_spin(int n)
diff --git a/arch/i386/mm/fault.c b/arch/i386/mm/fault.c
index cf572d9a3b6..7f0fcf219a2 100644
--- a/arch/i386/mm/fault.c
+++ b/arch/i386/mm/fault.c
@@ -214,6 +214,68 @@ static noinline void force_sig_info_fault(int si_signo, int si_code,
 
 fastcall void do_invalid_op(struct pt_regs *, unsigned long);
 
+static inline pmd_t *vmalloc_sync_one(pgd_t *pgd, unsigned long address)
+{
+	unsigned index = pgd_index(address);
+	pgd_t *pgd_k;
+	pud_t *pud, *pud_k;
+	pmd_t *pmd, *pmd_k;
+
+	pgd += index;
+	pgd_k = init_mm.pgd + index;
+
+	if (!pgd_present(*pgd_k))
+		return NULL;
+
+	/*
+	 * set_pgd(pgd, *pgd_k); here would be useless on PAE
+	 * and redundant with the set_pmd() on non-PAE. As would
+	 * set_pud.
+	 */
+
+	pud = pud_offset(pgd, address);
+	pud_k = pud_offset(pgd_k, address);
+	if (!pud_present(*pud_k))
+		return NULL;
+
+	pmd = pmd_offset(pud, address);
+	pmd_k = pmd_offset(pud_k, address);
+	if (!pmd_present(*pmd_k))
+		return NULL;
+	if (!pmd_present(*pmd))
+		set_pmd(pmd, *pmd_k);
+	else
+		BUG_ON(pmd_page(*pmd) != pmd_page(*pmd_k));
+	return pmd_k;
+}
+
+/*
+ * Handle a fault on the vmalloc or module mapping area
+ *
+ * This assumes no large pages in there.
+ */
+static inline int vmalloc_fault(unsigned long address)
+{
+	unsigned long pgd_paddr;
+	pmd_t *pmd_k;
+	pte_t *pte_k;
+	/*
+	 * Synchronize this task's top level page-table
+	 * with the 'reference' page table.
+	 *
+	 * Do _not_ use "current" here. We might be inside
+	 * an interrupt in the middle of a task switch..
+	 */
+	pgd_paddr = read_cr3();
+	pmd_k = vmalloc_sync_one(__va(pgd_paddr), address);
+	if (!pmd_k)
+		return -1;
+	pte_k = pte_offset_kernel(pmd_k, address);
+	if (!pte_present(*pte_k))
+		return -1;
+	return 0;
+}
+
 /*
  * This routine handles page faults.  It determines the address,
  * and the problem, and then passes it off to one of the appropriate
@@ -223,6 +285,8 @@ fastcall void do_invalid_op(struct pt_regs *, unsigned long);
  *	bit 0 == 0 means no page found, 1 means protection fault
  *	bit 1 == 0 means read, 1 means write
  *	bit 2 == 0 means kernel, 1 means user-mode
+ *	bit 3 == 1 means use of reserved bit detected
+ *	bit 4 == 1 means fault was an instruction fetch
  */
 fastcall void __kprobes do_page_fault(struct pt_regs *regs,
 				      unsigned long error_code)
@@ -237,13 +301,6 @@ fastcall void __kprobes do_page_fault(struct pt_regs *regs,
 	/* get the address */
         address = read_cr2();
 
-	if (notify_die(DIE_PAGE_FAULT, "page fault", regs, error_code, 14,
-					SIGSEGV) == NOTIFY_STOP)
-		return;
-	/* It's safe to allow irq's after cr2 has been saved */
-	if (regs->eflags & (X86_EFLAGS_IF|VM_MASK))
-		local_irq_enable();
-
 	tsk = current;
 
 	si_code = SEGV_MAPERR;
@@ -259,17 +316,29 @@ fastcall void __kprobes do_page_fault(struct pt_regs *regs,
 	 *
 	 * This verifies that the fault happens in kernel space
 	 * (error_code & 4) == 0, and that the fault was not a
-	 * protection error (error_code & 1) == 0.
+	 * protection error (error_code & 9) == 0.
 	 */
-	if (unlikely(address >= TASK_SIZE)) { 
-		if (!(error_code & 5))
-			goto vmalloc_fault;
-		/* 
+	if (unlikely(address >= TASK_SIZE)) {
+		if (!(error_code & 0x0000000d) && vmalloc_fault(address) >= 0)
+			return;
+		if (notify_die(DIE_PAGE_FAULT, "page fault", regs, error_code, 14,
+						SIGSEGV) == NOTIFY_STOP)
+			return;
+		/*
 		 * Don't take the mm semaphore here. If we fixup a prefetch
 		 * fault we could otherwise deadlock.
 		 */
 		goto bad_area_nosemaphore;
-	} 
+	}
+
+	if (notify_die(DIE_PAGE_FAULT, "page fault", regs, error_code, 14,
+					SIGSEGV) == NOTIFY_STOP)
+		return;
+
+	/* It's safe to allow irq's after cr2 has been saved and the vmalloc
+	   fault has been handled. */
+	if (regs->eflags & (X86_EFLAGS_IF|VM_MASK))
+		local_irq_enable();
 
 	mm = tsk->mm;
 
@@ -440,24 +509,31 @@ no_context:
 
 	bust_spinlocks(1);
 
-#ifdef CONFIG_X86_PAE
-	if (error_code & 16) {
-		pte_t *pte = lookup_address(address);
+	if (oops_may_print()) {
+	#ifdef CONFIG_X86_PAE
+		if (error_code & 16) {
+			pte_t *pte = lookup_address(address);
 
-		if (pte && pte_present(*pte) && !pte_exec_kernel(*pte))
-			printk(KERN_CRIT "kernel tried to execute NX-protected page - exploit attempt? (uid: %d)\n", current->uid);
+			if (pte && pte_present(*pte) && !pte_exec_kernel(*pte))
+				printk(KERN_CRIT "kernel tried to execute "
+					"NX-protected page - exploit attempt? "
+					"(uid: %d)\n", current->uid);
+		}
+	#endif
+		if (address < PAGE_SIZE)
+			printk(KERN_ALERT "BUG: unable to handle kernel NULL "
+					"pointer dereference");
+		else
+			printk(KERN_ALERT "BUG: unable to handle kernel paging"
+					" request");
+		printk(" at virtual address %08lx\n",address);
+		printk(KERN_ALERT " printing eip:\n");
+		printk("%08lx\n", regs->eip);
 	}
-#endif
-	if (address < PAGE_SIZE)
-		printk(KERN_ALERT "Unable to handle kernel NULL pointer dereference");
-	else
-		printk(KERN_ALERT "Unable to handle kernel paging request");
-	printk(" at virtual address %08lx\n",address);
-	printk(KERN_ALERT " printing eip:\n");
-	printk("%08lx\n", regs->eip);
 	page = read_cr3();
 	page = ((unsigned long *) __va(page))[address >> 22];
-	printk(KERN_ALERT "*pde = %08lx\n", page);
+	if (oops_may_print())
+		printk(KERN_ALERT "*pde = %08lx\n", page);
 	/*
 	 * We must not directly access the pte in the highpte
 	 * case, the page table might be allocated in highmem.
@@ -465,7 +541,7 @@ no_context:
 	 * it's allocated already.
 	 */
 #ifndef CONFIG_HIGHPTE
-	if (page & 1) {
+	if ((page & 1) && oops_may_print()) {
 		page &= PAGE_MASK;
 		address &= 0x003ff000;
 		page = ((unsigned long *) __va(page))[address >> PAGE_SHIFT];
@@ -510,51 +586,41 @@ do_sigbus:
 	tsk->thread.error_code = error_code;
 	tsk->thread.trap_no = 14;
 	force_sig_info_fault(SIGBUS, BUS_ADRERR, address, tsk);
-	return;
-
-vmalloc_fault:
-	{
-		/*
-		 * Synchronize this task's top level page-table
-		 * with the 'reference' page table.
-		 *
-		 * Do _not_ use "tsk" here. We might be inside
-		 * an interrupt in the middle of a task switch..
-		 */
-		int index = pgd_index(address);
-		unsigned long pgd_paddr;
-		pgd_t *pgd, *pgd_k;
-		pud_t *pud, *pud_k;
-		pmd_t *pmd, *pmd_k;
-		pte_t *pte_k;
-
-		pgd_paddr = read_cr3();
-		pgd = index + (pgd_t *)__va(pgd_paddr);
-		pgd_k = init_mm.pgd + index;
-
-		if (!pgd_present(*pgd_k))
-			goto no_context;
-
-		/*
-		 * set_pgd(pgd, *pgd_k); here would be useless on PAE
-		 * and redundant with the set_pmd() on non-PAE. As would
-		 * set_pud.
-		 */
+}
 
-		pud = pud_offset(pgd, address);
-		pud_k = pud_offset(pgd_k, address);
-		if (!pud_present(*pud_k))
-			goto no_context;
-		
-		pmd = pmd_offset(pud, address);
-		pmd_k = pmd_offset(pud_k, address);
-		if (!pmd_present(*pmd_k))
-			goto no_context;
-		set_pmd(pmd, *pmd_k);
+#ifndef CONFIG_X86_PAE
+void vmalloc_sync_all(void)
+{
+	/*
+	 * Note that races in the updates of insync and start aren't
+	 * problematic: insync can only get set bits added, and updates to
+	 * start are only improving performance (without affecting correctness
+	 * if undone).
+	 */
+	static DECLARE_BITMAP(insync, PTRS_PER_PGD);
+	static unsigned long start = TASK_SIZE;
+	unsigned long address;
 
-		pte_k = pte_offset_kernel(pmd_k, address);
-		if (!pte_present(*pte_k))
-			goto no_context;
-		return;
+	BUILD_BUG_ON(TASK_SIZE & ~PGDIR_MASK);
+	for (address = start; address >= TASK_SIZE; address += PGDIR_SIZE) {
+		if (!test_bit(pgd_index(address), insync)) {
+			unsigned long flags;
+			struct page *page;
+
+			spin_lock_irqsave(&pgd_lock, flags);
+			for (page = pgd_list; page; page =
+					(struct page *)page->index)
+				if (!vmalloc_sync_one(page_address(page),
+								address)) {
+					BUG_ON(page != pgd_list);
+					break;
+				}
+			spin_unlock_irqrestore(&pgd_lock, flags);
+			if (!page)
+				set_bit(pgd_index(address), insync);
+		}
+		if (address == start && test_bit(pgd_index(address), insync))
+			start = address + PGDIR_SIZE;
 	}
 }
+#endif
diff --git a/arch/i386/mm/init.c b/arch/i386/mm/init.c
index 7ba55a6e2db..9f66ac582a8 100644
--- a/arch/i386/mm/init.c
+++ b/arch/i386/mm/init.c
@@ -720,21 +720,6 @@ static int noinline do_test_wp_bit(void)
 	return flag;
 }
 
-void free_initmem(void)
-{
-	unsigned long addr;
-
-	addr = (unsigned long)(&__init_begin);
-	for (; addr < (unsigned long)(&__init_end); addr += PAGE_SIZE) {
-		ClearPageReserved(virt_to_page(addr));
-		init_page_count(virt_to_page(addr));
-		memset((void *)addr, 0xcc, PAGE_SIZE);
-		free_page(addr);
-		totalram_pages++;
-	}
-	printk (KERN_INFO "Freeing unused kernel memory: %dk freed\n", (__init_end - __init_begin) >> 10);
-}
-
 #ifdef CONFIG_DEBUG_RODATA
 
 extern char __start_rodata, __end_rodata;
@@ -758,17 +743,31 @@ void mark_rodata_ro(void)
 }
 #endif
 
+void free_init_pages(char *what, unsigned long begin, unsigned long end)
+{
+	unsigned long addr;
+
+	for (addr = begin; addr < end; addr += PAGE_SIZE) {
+		ClearPageReserved(virt_to_page(addr));
+		init_page_count(virt_to_page(addr));
+		memset((void *)addr, 0xcc, PAGE_SIZE);
+		free_page(addr);
+		totalram_pages++;
+	}
+	printk(KERN_INFO "Freeing %s: %ldk freed\n", what, (end - begin) >> 10);
+}
+
+void free_initmem(void)
+{
+	free_init_pages("unused kernel memory",
+			(unsigned long)(&__init_begin),
+			(unsigned long)(&__init_end));
+}
 
 #ifdef CONFIG_BLK_DEV_INITRD
 void free_initrd_mem(unsigned long start, unsigned long end)
 {
-	if (start < end)
-		printk (KERN_INFO "Freeing initrd memory: %ldk freed\n", (end - start) >> 10);
-	for (; start < end; start += PAGE_SIZE) {
-		ClearPageReserved(virt_to_page(start));
-		init_page_count(virt_to_page(start));
-		free_page(start);
-		totalram_pages++;
-	}
+	free_init_pages("initrd memory", start, end);
 }
 #endif
+
diff --git a/arch/i386/oprofile/nmi_int.c b/arch/i386/oprofile/nmi_int.c
index 0493e8b8ec4..1accce50c2c 100644
--- a/arch/i386/oprofile/nmi_int.c
+++ b/arch/i386/oprofile/nmi_int.c
@@ -122,7 +122,7 @@ static void nmi_save_registers(void * dummy)
 static void free_msrs(void)
 {
 	int i;
-	for (i = 0; i < NR_CPUS; ++i) {
+	for_each_cpu(i) {
 		kfree(cpu_msrs[i].counters);
 		cpu_msrs[i].counters = NULL;
 		kfree(cpu_msrs[i].controls);
@@ -138,10 +138,7 @@ static int allocate_msrs(void)
 	size_t counters_size = sizeof(struct op_msr) * model->num_counters;
 
 	int i;
-	for (i = 0; i < NR_CPUS; ++i) {
-		if (!cpu_online(i))
-			continue;
-
+	for_each_online_cpu(i) {
 		cpu_msrs[i].counters = kmalloc(counters_size, GFP_KERNEL);
 		if (!cpu_msrs[i].counters) {
 			success = 0;
diff --git a/arch/ia64/hp/sim/simserial.c b/arch/ia64/hp/sim/simserial.c
index 626cdc83668..0e5c6ae5022 100644
--- a/arch/ia64/hp/sim/simserial.c
+++ b/arch/ia64/hp/sim/simserial.c
@@ -46,11 +46,6 @@
 #define KEYBOARD_INTR	3	/* must match with simulator! */
 
 #define NR_PORTS	1	/* only one port for now */
-#define SERIAL_INLINE	1
-
-#ifdef SERIAL_INLINE
-#define _INLINE_ inline
-#endif
 
 #define IRQ_T(info) ((info->flags & ASYNC_SHARE_IRQ) ? SA_SHIRQ : SA_INTERRUPT)
 
@@ -237,7 +232,7 @@ static void rs_put_char(struct tty_struct *tty, unsigned char ch)
 	local_irq_restore(flags);
 }
 
-static _INLINE_ void transmit_chars(struct async_struct *info, int *intr_done)
+static void transmit_chars(struct async_struct *info, int *intr_done)
 {
 	int count;
 	unsigned long flags;
diff --git a/arch/m32r/kernel/irq.c b/arch/m32r/kernel/irq.c
index 1ce63926a3c..a4634b06f67 100644
--- a/arch/m32r/kernel/irq.c
+++ b/arch/m32r/kernel/irq.c
@@ -37,9 +37,8 @@ int show_interrupts(struct seq_file *p, void *v)
 
 	if (i == 0) {
 		seq_printf(p, "           ");
-		for (j=0; j<NR_CPUS; j++)
-			if (cpu_online(j))
-				seq_printf(p, "CPU%d       ",j);
+		for_each_online_cpu(j)
+			seq_printf(p, "CPU%d       ",j);
 		seq_putc(p, '\n');
 	}
 
@@ -52,9 +51,8 @@ int show_interrupts(struct seq_file *p, void *v)
 #ifndef CONFIG_SMP
 		seq_printf(p, "%10u ", kstat_irqs(i));
 #else
-		for (j = 0; j < NR_CPUS; j++)
-			if (cpu_online(j))
-				seq_printf(p, "%10u ", kstat_cpu(j).irqs[i]);
+		for_each_online_cpu(j)
+			seq_printf(p, "%10u ", kstat_cpu(j).irqs[i]);
 #endif
 		seq_printf(p, " %14s", irq_desc[i].handler->typename);
 		seq_printf(p, "  %s", action->name);
diff --git a/arch/m68k/bvme6000/rtc.c b/arch/m68k/bvme6000/rtc.c
index 703cbc6dc9c..15c16b62dff 100644
--- a/arch/m68k/bvme6000/rtc.c
+++ b/arch/m68k/bvme6000/rtc.c
@@ -18,6 +18,7 @@
 #include <linux/module.h>
 #include <linux/mc146818rtc.h>	/* For struct rtc_time and ioctls, etc */
 #include <linux/smp_lock.h>
+#include <linux/bcd.h>
 #include <asm/bvme6000hw.h>
 
 #include <asm/io.h>
@@ -32,9 +33,6 @@
  *	ioctls.
  */
 
-#define BCD2BIN(val) (((val)&15) + ((val)>>4)*10)
-#define BIN2BCD(val) ((((val)/10)<<4) + (val)%10)
-
 static unsigned char days_in_mo[] =
 {0, 31, 28, 31, 30, 31, 30, 31, 31, 30, 31, 30, 31};
 
diff --git a/arch/mips/kernel/irq.c b/arch/mips/kernel/irq.c
index 7d93992e462..3dd76b3d296 100644
--- a/arch/mips/kernel/irq.c
+++ b/arch/mips/kernel/irq.c
@@ -68,9 +68,8 @@ int show_interrupts(struct seq_file *p, void *v)
 
 	if (i == 0) {
 		seq_printf(p, "           ");
-		for (j=0; j<NR_CPUS; j++)
-			if (cpu_online(j))
-				seq_printf(p, "CPU%d       ",j);
+		for_each_online_cpu(j)
+			seq_printf(p, "CPU%d       ",j);
 		seq_putc(p, '\n');
 	}
 
@@ -83,9 +82,8 @@ int show_interrupts(struct seq_file *p, void *v)
 #ifndef CONFIG_SMP
 		seq_printf(p, "%10u ", kstat_irqs(i));
 #else
-		for (j = 0; j < NR_CPUS; j++)
-			if (cpu_online(j))
-				seq_printf(p, "%10u ", kstat_cpu(j).irqs[i]);
+		for_each_online_cpu(j)
+			seq_printf(p, "%10u ", kstat_cpu(j).irqs[i]);
 #endif
 		seq_printf(p, " %14s", irq_desc[i].handler->typename);
 		seq_printf(p, "  %s", action->name);
diff --git a/arch/mips/kernel/smp.c b/arch/mips/kernel/smp.c
index 06ed9075242..78d171bfa33 100644
--- a/arch/mips/kernel/smp.c
+++ b/arch/mips/kernel/smp.c
@@ -167,8 +167,8 @@ int smp_call_function (void (*func) (void *info), void *info, int retry,
 	mb();
 
 	/* Send a message to all other CPUs and wait for them to respond */
-	for (i = 0; i < NR_CPUS; i++)
-		if (cpu_online(i) && i != cpu)
+	for_each_online_cpu(i)
+		if (i != cpu)
 			core_send_ipi(i, SMP_CALL_FUNCTION);
 
 	/* Wait for response */
diff --git a/arch/mips/sgi-ip27/ip27-irq.c b/arch/mips/sgi-ip27/ip27-irq.c
index 73e5e52781d..2854ac4c9be 100644
--- a/arch/mips/sgi-ip27/ip27-irq.c
+++ b/arch/mips/sgi-ip27/ip27-irq.c
@@ -88,12 +88,9 @@ static inline int find_level(cpuid_t *cpunum, int irq)
 {
 	int cpu, i;
 
-	for (cpu = 0; cpu <= NR_CPUS; cpu++) {
+	for_each_online_cpu(cpu) {
 		struct slice_data *si = cpu_data[cpu].data;
 
-		if (!cpu_online(cpu))
-			continue;
-
 		for (i = BASE_PCI_IRQ; i < LEVELS_PER_SLICE; i++)
 			if (si->level_to_irq[i] == irq) {
 				*cpunum = cpu;
diff --git a/arch/parisc/kernel/smp.c b/arch/parisc/kernel/smp.c
index 25564b7ca6b..d6ac1c60a47 100644
--- a/arch/parisc/kernel/smp.c
+++ b/arch/parisc/kernel/smp.c
@@ -298,8 +298,8 @@ send_IPI_allbutself(enum ipi_message_type op)
 {
 	int i;
 	
-	for (i = 0; i < NR_CPUS; i++) {
-		if (cpu_online(i) && i != smp_processor_id())
+	for_each_online_cpu(i) {
+		if (i != smp_processor_id())
 			send_IPI_single(i, op);
 	}
 }
@@ -643,14 +643,13 @@ int sys_cpus(int argc, char **argv)
 	if ( argc == 1 ){
 	
 #ifdef DUMP_MORE_STATE
-		for(i=0; i<NR_CPUS; i++) {
+		for_each_online_cpu(i) {
 			int cpus_per_line = 4;
-			if(cpu_online(i)) {
-				if (j++ % cpus_per_line)
-					printk(" %3d",i);
-				else
-					printk("\n %3d",i);
-			}
+
+			if (j++ % cpus_per_line)
+				printk(" %3d",i);
+			else
+				printk("\n %3d",i);
 		}
 		printk("\n"); 
 #else
@@ -659,9 +658,7 @@ int sys_cpus(int argc, char **argv)
 	} else if((argc==2) && !(strcmp(argv[1],"-l"))) {
 		printk("\nCPUSTATE  TASK CPUNUM CPUID HARDCPU(HPA)\n");
 #ifdef DUMP_MORE_STATE
-		for(i=0;i<NR_CPUS;i++) {
-			if (!cpu_online(i))
-				continue;
+		for_each_online_cpu(i) {
 			if (cpu_data[i].cpuid != NO_PROC_ID) {
 				switch(cpu_data[i].state) {
 					case STATE_RENDEZVOUS:
@@ -695,9 +692,7 @@ int sys_cpus(int argc, char **argv)
 	} else if ((argc==2) && !(strcmp(argv[1],"-s"))) { 
 #ifdef DUMP_MORE_STATE
      		printk("\nCPUSTATE   CPUID\n");
-		for (i=0;i<NR_CPUS;i++) {
-			if (!cpu_online(i))
-				continue;
+		for_each_online_cpu(i) {
 			if (cpu_data[i].cpuid != NO_PROC_ID) {
 				switch(cpu_data[i].state) {
 					case STATE_RENDEZVOUS:
diff --git a/arch/powerpc/kernel/irq.c b/arch/powerpc/kernel/irq.c
index 24dc8117b82..771a59cbd21 100644
--- a/arch/powerpc/kernel/irq.c
+++ b/arch/powerpc/kernel/irq.c
@@ -135,9 +135,8 @@ skip:
 #ifdef CONFIG_TAU_INT
 		if (tau_initialized){
 			seq_puts(p, "TAU: ");
-			for (j = 0; j < NR_CPUS; j++)
-				if (cpu_online(j))
-					seq_printf(p, "%10u ", tau_interrupts(j));
+			for_each_online_cpu(j)
+				seq_printf(p, "%10u ", tau_interrupts(j));
 			seq_puts(p, "  PowerPC             Thermal Assist (cpu temp)\n");
 		}
 #endif
diff --git a/arch/powerpc/kernel/kprobes.c b/arch/powerpc/kernel/kprobes.c
index 258039fb301..cb1fe5878e8 100644
--- a/arch/powerpc/kernel/kprobes.c
+++ b/arch/powerpc/kernel/kprobes.c
@@ -81,9 +81,9 @@ void __kprobes arch_disarm_kprobe(struct kprobe *p)
 
 void __kprobes arch_remove_kprobe(struct kprobe *p)
 {
-	down(&kprobe_mutex);
+	mutex_lock(&kprobe_mutex);
 	free_insn_slot(p->ainsn.insn);
-	up(&kprobe_mutex);
+	mutex_unlock(&kprobe_mutex);
 }
 
 static inline void prepare_singlestep(struct kprobe *p, struct pt_regs *regs)
diff --git a/arch/powerpc/kernel/setup-common.c b/arch/powerpc/kernel/setup-common.c
index be12041c0fc..c1d62bf11f2 100644
--- a/arch/powerpc/kernel/setup-common.c
+++ b/arch/powerpc/kernel/setup-common.c
@@ -162,9 +162,8 @@ static int show_cpuinfo(struct seq_file *m, void *v)
 #if defined(CONFIG_SMP) && defined(CONFIG_PPC32)
 		unsigned long bogosum = 0;
 		int i;
-		for (i = 0; i < NR_CPUS; ++i)
-			if (cpu_online(i))
-				bogosum += loops_per_jiffy;
+		for_each_online_cpu(i)
+			bogosum += loops_per_jiffy;
 		seq_printf(m, "total bogomips\t: %lu.%02lu\n",
 			   bogosum/(500000/HZ), bogosum/(5000/HZ) % 100);
 #endif /* CONFIG_SMP && CONFIG_PPC32 */
diff --git a/arch/powerpc/kernel/setup_32.c b/arch/powerpc/kernel/setup_32.c
index db72a92943b..dc2770df25b 100644
--- a/arch/powerpc/kernel/setup_32.c
+++ b/arch/powerpc/kernel/setup_32.c
@@ -272,9 +272,8 @@ int __init ppc_init(void)
 	if ( ppc_md.progress ) ppc_md.progress("             ", 0xffff);
 
 	/* register CPU devices */
-	for (i = 0; i < NR_CPUS; i++)
-		if (cpu_possible(i))
-			register_cpu(&cpu_devices[i], i, NULL);
+	for_each_cpu(i)
+		register_cpu(&cpu_devices[i], i, NULL);
 
 	/* call platform init */
 	if (ppc_md.init != NULL) {
diff --git a/arch/powerpc/platforms/powermac/smp.c b/arch/powerpc/platforms/powermac/smp.c
index 6d64a9bf347..1065d87fc27 100644
--- a/arch/powerpc/platforms/powermac/smp.c
+++ b/arch/powerpc/platforms/powermac/smp.c
@@ -191,9 +191,7 @@ static void smp_psurge_message_pass(int target, int msg)
 	if (num_online_cpus() < 2)
 		return;
 
-	for (i = 0; i < NR_CPUS; i++) {
-		if (!cpu_online(i))
-			continue;
+	for_each_online_cpu(i) {
 		if (target == MSG_ALL
 		    || (target == MSG_ALL_BUT_SELF && i != smp_processor_id())
 		    || target == i) {
diff --git a/arch/ppc/kernel/setup.c b/arch/ppc/kernel/setup.c
index c08ab432e95..53e9deacee8 100644
--- a/arch/ppc/kernel/setup.c
+++ b/arch/ppc/kernel/setup.c
@@ -168,9 +168,8 @@ int show_cpuinfo(struct seq_file *m, void *v)
 		/* Show summary information */
 #ifdef CONFIG_SMP
 		unsigned long bogosum = 0;
-		for (i = 0; i < NR_CPUS; ++i)
-			if (cpu_online(i))
-				bogosum += cpu_data[i].loops_per_jiffy;
+		for_each_online_cpu(i)
+			bogosum += cpu_data[i].loops_per_jiffy;
 		seq_printf(m, "total bogomips\t: %lu.%02lu\n",
 			   bogosum/(500000/HZ), bogosum/(5000/HZ) % 100);
 #endif /* CONFIG_SMP */
@@ -712,9 +711,8 @@ int __init ppc_init(void)
 	if ( ppc_md.progress ) ppc_md.progress("             ", 0xffff);
 
 	/* register CPU devices */
-	for (i = 0; i < NR_CPUS; i++)
-		if (cpu_possible(i))
-			register_cpu(&cpu_devices[i], i, NULL);
+	for_each_cpu(i)
+		register_cpu(&cpu_devices[i], i, NULL);
 
 	/* call platform init */
 	if (ppc_md.init != NULL) {
diff --git a/arch/s390/kernel/smp.c b/arch/s390/kernel/smp.c
index 7dbe00c76c6..d52d6d211d9 100644
--- a/arch/s390/kernel/smp.c
+++ b/arch/s390/kernel/smp.c
@@ -799,9 +799,7 @@ void __init smp_prepare_cpus(unsigned int max_cpus)
          */
 	print_cpu_info(&S390_lowcore.cpu_data);
 
-        for(i = 0; i < NR_CPUS; i++) {
-		if (!cpu_possible(i))
-			continue;
+        for_each_cpu(i) {
 		lowcore_ptr[i] = (struct _lowcore *)
 			__get_free_pages(GFP_KERNEL|GFP_DMA, 
 					sizeof(void*) == 8 ? 1 : 0);
diff --git a/arch/sh/kernel/irq.c b/arch/sh/kernel/irq.c
index 6883c00728c..b56e79632f2 100644
--- a/arch/sh/kernel/irq.c
+++ b/arch/sh/kernel/irq.c
@@ -35,9 +35,8 @@ int show_interrupts(struct seq_file *p, void *v)
 
 	if (i == 0) {
 		seq_puts(p, "           ");
-		for (j=0; j<NR_CPUS; j++)
-			if (cpu_online(j))
-				seq_printf(p, "CPU%d       ",j);
+		for_each_online_cpu(j)
+			seq_printf(p, "CPU%d       ",j);
 		seq_putc(p, '\n');
 	}
 
diff --git a/arch/sh/kernel/setup.c b/arch/sh/kernel/setup.c
index a067a34e0b6..c0e79843f58 100644
--- a/arch/sh/kernel/setup.c
+++ b/arch/sh/kernel/setup.c
@@ -404,9 +404,8 @@ static int __init topology_init(void)
 {
 	int cpu_id;
 
-	for (cpu_id = 0; cpu_id < NR_CPUS; cpu_id++)
-		if (cpu_possible(cpu_id))
-			register_cpu(&cpu[cpu_id], cpu_id, NULL);
+	for_each_cpu(cpu_id)
+		register_cpu(&cpu[cpu_id], cpu_id, NULL);
 
 	return 0;
 }
diff --git a/arch/sh64/kernel/irq.c b/arch/sh64/kernel/irq.c
index 9fc2b71dbd8..d69879c0e06 100644
--- a/arch/sh64/kernel/irq.c
+++ b/arch/sh64/kernel/irq.c
@@ -53,9 +53,8 @@ int show_interrupts(struct seq_file *p, void *v)
 
 	if (i == 0) {
 		seq_puts(p, "           ");
-		for (j=0; j<NR_CPUS; j++)
-			if (cpu_online(j))
-				seq_printf(p, "CPU%d       ",j);
+		for_each_online_cpu(j)
+			seq_printf(p, "CPU%d       ",j);
 		seq_putc(p, '\n');
 	}
 
diff --git a/arch/sparc/kernel/irq.c b/arch/sparc/kernel/irq.c
index 410b9a72aba..4c60a6ef54a 100644
--- a/arch/sparc/kernel/irq.c
+++ b/arch/sparc/kernel/irq.c
@@ -184,9 +184,8 @@ int show_interrupts(struct seq_file *p, void *v)
 #ifndef CONFIG_SMP
 		seq_printf(p, "%10u ", kstat_irqs(i));
 #else
-		for (j = 0; j < NR_CPUS; j++) {
-			if (cpu_online(j))
-				seq_printf(p, "%10u ",
+		for_each_online_cpu(j) {
+			seq_printf(p, "%10u ",
 				    kstat_cpu(cpu_logical_map(j)).irqs[i]);
 		}
 #endif
diff --git a/arch/sparc/kernel/smp.c b/arch/sparc/kernel/smp.c
index c6e721d8f47..ea5682ce703 100644
--- a/arch/sparc/kernel/smp.c
+++ b/arch/sparc/kernel/smp.c
@@ -243,9 +243,8 @@ int setup_profiling_timer(unsigned int multiplier)
 		return -EINVAL;
 
 	spin_lock_irqsave(&prof_setup_lock, flags);
-	for(i = 0; i < NR_CPUS; i++) {
-		if (cpu_possible(i))
-			load_profile_irq(i, lvl14_resolution / multiplier);
+	for_each_cpu(i) {
+		load_profile_irq(i, lvl14_resolution / multiplier);
 		prof_multiplier(i) = multiplier;
 	}
 	spin_unlock_irqrestore(&prof_setup_lock, flags);
@@ -273,13 +272,12 @@ void smp_bogo(struct seq_file *m)
 {
 	int i;
 	
-	for (i = 0; i < NR_CPUS; i++) {
-		if (cpu_online(i))
-			seq_printf(m,
-				   "Cpu%dBogo\t: %lu.%02lu\n", 
-				   i,
-				   cpu_data(i).udelay_val/(500000/HZ),
-				   (cpu_data(i).udelay_val/(5000/HZ))%100);
+	for_each_online_cpu(i) {
+		seq_printf(m,
+			   "Cpu%dBogo\t: %lu.%02lu\n",
+			   i,
+			   cpu_data(i).udelay_val/(500000/HZ),
+			   (cpu_data(i).udelay_val/(5000/HZ))%100);
 	}
 }
 
@@ -288,8 +286,6 @@ void smp_info(struct seq_file *m)
 	int i;
 
 	seq_printf(m, "State:\n");
-	for (i = 0; i < NR_CPUS; i++) {
-		if (cpu_online(i))
-			seq_printf(m, "CPU%d\t\t: online\n", i);
-	}
+	for_each_online_cpu(i)
+		seq_printf(m, "CPU%d\t\t: online\n", i);
 }
diff --git a/arch/sparc/kernel/sun4d_irq.c b/arch/sparc/kernel/sun4d_irq.c
index 52621348a56..cea7fc6fc6e 100644
--- a/arch/sparc/kernel/sun4d_irq.c
+++ b/arch/sparc/kernel/sun4d_irq.c
@@ -103,11 +103,9 @@ found_it:	seq_printf(p, "%3d: ", i);
 #ifndef CONFIG_SMP
 		seq_printf(p, "%10u ", kstat_irqs(i));
 #else
-		for (x = 0; x < NR_CPUS; x++) {
-			if (cpu_online(x))
-				seq_printf(p, "%10u ",
-				       kstat_cpu(cpu_logical_map(x)).irqs[i]);
-		}
+		for_each_online_cpu(x)
+			seq_printf(p, "%10u ",
+			       kstat_cpu(cpu_logical_map(x)).irqs[i]);
 #endif
 		seq_printf(p, "%c %s",
 			(action->flags & SA_INTERRUPT) ? '+' : ' ',
diff --git a/arch/sparc/kernel/sun4d_smp.c b/arch/sparc/kernel/sun4d_smp.c
index 4219dd2ce3a..41bb9596be4 100644
--- a/arch/sparc/kernel/sun4d_smp.c
+++ b/arch/sparc/kernel/sun4d_smp.c
@@ -249,11 +249,9 @@ void __init smp4d_boot_cpus(void)
 	} else {
 		unsigned long bogosum = 0;
 		
-		for(i = 0; i < NR_CPUS; i++) {
-			if (cpu_isset(i, cpu_present_map)) {
-				bogosum += cpu_data(i).udelay_val;
-				smp_highest_cpu = i;
-			}
+		for_each_present_cpu(i) {
+			bogosum += cpu_data(i).udelay_val;
+			smp_highest_cpu = i;
 		}
 		SMP_PRINTK(("Total of %d Processors activated (%lu.%02lu BogoMIPS).\n", cpucount + 1, bogosum/(500000/HZ), (bogosum/(5000/HZ))%100));
 		printk("Total of %d Processors activated (%lu.%02lu BogoMIPS).\n",
diff --git a/arch/sparc/kernel/sun4m_smp.c b/arch/sparc/kernel/sun4m_smp.c
index fbbd8a474c4..1dde312eebd 100644
--- a/arch/sparc/kernel/sun4m_smp.c
+++ b/arch/sparc/kernel/sun4m_smp.c
@@ -218,10 +218,8 @@ void __init smp4m_boot_cpus(void)
 		cpu_present_map = cpumask_of_cpu(smp_processor_id());
 	} else {
 		unsigned long bogosum = 0;
-		for(i = 0; i < NR_CPUS; i++) {
-			if (cpu_isset(i, cpu_present_map))
-				bogosum += cpu_data(i).udelay_val;
-		}
+		for_each_present_cpu(i)
+			bogosum += cpu_data(i).udelay_val;
 		printk("Total of %d Processors activated (%lu.%02lu BogoMIPS).\n",
 		       cpucount + 1,
 		       bogosum/(500000/HZ),
diff --git a/arch/sparc64/kernel/irq.c b/arch/sparc64/kernel/irq.c
index 8c93ba655b3..e505a4125e3 100644
--- a/arch/sparc64/kernel/irq.c
+++ b/arch/sparc64/kernel/irq.c
@@ -117,9 +117,7 @@ int show_interrupts(struct seq_file *p, void *v)
 #ifndef CONFIG_SMP
 		seq_printf(p, "%10u ", kstat_irqs(i));
 #else
-		for (j = 0; j < NR_CPUS; j++) {
-			if (!cpu_online(j))
-				continue;
+		for_each_online_cpu(j) {
 			seq_printf(p, "%10u ",
 				   kstat_cpu(j).irqs[i]);
 		}
diff --git a/arch/sparc64/kernel/smp.c b/arch/sparc64/kernel/smp.c
index 373a701c90a..1b6e2ade100 100644
--- a/arch/sparc64/kernel/smp.c
+++ b/arch/sparc64/kernel/smp.c
@@ -57,25 +57,21 @@ void smp_info(struct seq_file *m)
 	int i;
 	
 	seq_printf(m, "State:\n");
-	for (i = 0; i < NR_CPUS; i++) {
-		if (cpu_online(i))
-			seq_printf(m,
-				   "CPU%d:\t\tonline\n", i);
-	}
+	for_each_online_cpu(i)
+		seq_printf(m, "CPU%d:\t\tonline\n", i);
 }
 
 void smp_bogo(struct seq_file *m)
 {
 	int i;
 	
-	for (i = 0; i < NR_CPUS; i++)
-		if (cpu_online(i))
-			seq_printf(m,
-				   "Cpu%dBogo\t: %lu.%02lu\n"
-				   "Cpu%dClkTck\t: %016lx\n",
-				   i, cpu_data(i).udelay_val / (500000/HZ),
-				   (cpu_data(i).udelay_val / (5000/HZ)) % 100,
-				   i, cpu_data(i).clock_tick);
+	for_each_online_cpu(i)
+		seq_printf(m,
+			   "Cpu%dBogo\t: %lu.%02lu\n"
+			   "Cpu%dClkTck\t: %016lx\n",
+			   i, cpu_data(i).udelay_val / (500000/HZ),
+			   (cpu_data(i).udelay_val / (5000/HZ)) % 100,
+			   i, cpu_data(i).clock_tick);
 }
 
 void __init smp_store_cpu_info(int id)
@@ -1282,7 +1278,7 @@ int setup_profiling_timer(unsigned int multiplier)
 		return -EINVAL;
 
 	spin_lock_irqsave(&prof_setup_lock, flags);
-	for (i = 0; i < NR_CPUS; i++)
+	for_each_cpu(i)
 		prof_multiplier(i) = multiplier;
 	current_tick_offset = (timer_tick_offset / multiplier);
 	spin_unlock_irqrestore(&prof_setup_lock, flags);
@@ -1384,10 +1380,8 @@ void __init smp_cpus_done(unsigned int max_cpus)
 	unsigned long bogosum = 0;
 	int i;
 
-	for (i = 0; i < NR_CPUS; i++) {
-		if (cpu_online(i))
-			bogosum += cpu_data(i).udelay_val;
-	}
+	for_each_online_cpu(i)
+		bogosum += cpu_data(i).udelay_val;
 	printk("Total of %ld processors activated "
 	       "(%lu.%02lu BogoMIPS).\n",
 	       (long) num_online_cpus(),
diff --git a/arch/sparc64/mm/init.c b/arch/sparc64/mm/init.c
index ded63ee9c4f..1539a8362b6 100644
--- a/arch/sparc64/mm/init.c
+++ b/arch/sparc64/mm/init.c
@@ -1828,8 +1828,8 @@ void __flush_tlb_all(void)
 void online_page(struct page *page)
 {
 	ClearPageReserved(page);
-	set_page_count(page, 0);
-	free_cold_page(page);
+	init_page_count(page);
+	__free_page(page);
 	totalram_pages++;
 	num_physpages++;
 }
diff --git a/arch/um/kernel/um_arch.c b/arch/um/kernel/um_arch.c
index 27cdf916442..80c9c18aae9 100644
--- a/arch/um/kernel/um_arch.c
+++ b/arch/um/kernel/um_arch.c
@@ -491,6 +491,16 @@ void __init check_bugs(void)
 	check_devanon();
 }
 
-void apply_alternatives(void *start, void *end)
+void apply_alternatives(struct alt_instr *start, struct alt_instr *end)
+{
+}
+
+void alternatives_smp_module_add(struct module *mod, char *name,
+				 void *locks, void *locks_end,
+				 void *text,  void *text_end)
+{
+}
+
+void alternatives_smp_module_del(struct module *mod)
 {
 }
diff --git a/arch/x86_64/kernel/early_printk.c b/arch/x86_64/kernel/early_printk.c
index 6dffb498ccd..a8a6aa70d69 100644
--- a/arch/x86_64/kernel/early_printk.c
+++ b/arch/x86_64/kernel/early_printk.c
@@ -17,11 +17,8 @@
 #define VGABASE		((void __iomem *)0xffffffff800b8000UL)
 #endif
 
-#define MAX_YPOS	max_ypos
-#define MAX_XPOS	max_xpos
-
 static int max_ypos = 25, max_xpos = 80;
-static int current_ypos = 1, current_xpos = 0; 
+static int current_ypos = 25, current_xpos = 0;
 
 static void early_vga_write(struct console *con, const char *str, unsigned n)
 {
@@ -29,26 +26,26 @@ static void early_vga_write(struct console *con, const char *str, unsigned n)
 	int  i, k, j;
 
 	while ((c = *str++) != '\0' && n-- > 0) {
-		if (current_ypos >= MAX_YPOS) {
+		if (current_ypos >= max_ypos) {
 			/* scroll 1 line up */
-			for (k = 1, j = 0; k < MAX_YPOS; k++, j++) {
-				for (i = 0; i < MAX_XPOS; i++) {
-					writew(readw(VGABASE + 2*(MAX_XPOS*k + i)),
-					       VGABASE + 2*(MAX_XPOS*j + i));
+			for (k = 1, j = 0; k < max_ypos; k++, j++) {
+				for (i = 0; i < max_xpos; i++) {
+					writew(readw(VGABASE+2*(max_xpos*k+i)),
+					       VGABASE + 2*(max_xpos*j + i));
 				}
 			}
-			for (i = 0; i < MAX_XPOS; i++)
-				writew(0x720, VGABASE + 2*(MAX_XPOS*j + i));
-			current_ypos = MAX_YPOS-1;
+			for (i = 0; i < max_xpos; i++)
+				writew(0x720, VGABASE + 2*(max_xpos*j + i));
+			current_ypos = max_ypos-1;
 		}
 		if (c == '\n') {
 			current_xpos = 0;
 			current_ypos++;
 		} else if (c != '\r')  {
 			writew(((0x7 << 8) | (unsigned short) c),
-			       VGABASE + 2*(MAX_XPOS*current_ypos +
+			       VGABASE + 2*(max_xpos*current_ypos +
 						current_xpos++));
-			if (current_xpos >= MAX_XPOS) {
+			if (current_xpos >= max_xpos) {
 				current_xpos = 0;
 				current_ypos++;
 			}
@@ -244,6 +241,7 @@ int __init setup_early_printk(char *opt)
 	           && SCREEN_INFO.orig_video_isVGA == 1) {
 		max_xpos = SCREEN_INFO.orig_video_cols;
 		max_ypos = SCREEN_INFO.orig_video_lines;
+		current_ypos = SCREEN_INFO.orig_y;
 		early_console = &early_vga_console; 
  	} else if (!strncmp(buf, "simnow", 6)) {
  		simnow_init(buf + 6);
diff --git a/arch/x86_64/kernel/irq.c b/arch/x86_64/kernel/irq.c
index 30d2a1e545f..d8bd0b345b1 100644
--- a/arch/x86_64/kernel/irq.c
+++ b/arch/x86_64/kernel/irq.c
@@ -38,9 +38,8 @@ int show_interrupts(struct seq_file *p, void *v)
 
 	if (i == 0) {
 		seq_printf(p, "           ");
-		for (j=0; j<NR_CPUS; j++)
-			if (cpu_online(j))
-				seq_printf(p, "CPU%d       ",j);
+		for_each_online_cpu(j)
+			seq_printf(p, "CPU%d       ",j);
 		seq_putc(p, '\n');
 	}
 
@@ -53,10 +52,8 @@ int show_interrupts(struct seq_file *p, void *v)
 #ifndef CONFIG_SMP
 		seq_printf(p, "%10u ", kstat_irqs(i));
 #else
-		for (j=0; j<NR_CPUS; j++)
-			if (cpu_online(j))
-			seq_printf(p, "%10u ",
-				kstat_cpu(j).irqs[i]);
+		for_each_online_cpu(j)
+			seq_printf(p, "%10u ", kstat_cpu(j).irqs[i]);
 #endif
 		seq_printf(p, " %14s", irq_desc[i].handler->typename);
 
@@ -68,15 +65,13 @@ skip:
 		spin_unlock_irqrestore(&irq_desc[i].lock, flags);
 	} else if (i == NR_IRQS) {
 		seq_printf(p, "NMI: ");
-		for (j = 0; j < NR_CPUS; j++)
-			if (cpu_online(j))
-				seq_printf(p, "%10u ", cpu_pda(j)->__nmi_count);
+		for_each_online_cpu(j)
+			seq_printf(p, "%10u ", cpu_pda(j)->__nmi_count);
 		seq_putc(p, '\n');
 #ifdef CONFIG_X86_LOCAL_APIC
 		seq_printf(p, "LOC: ");
-		for (j = 0; j < NR_CPUS; j++)
-			if (cpu_online(j))
-				seq_printf(p, "%10u ", cpu_pda(j)->apic_timer_irqs);
+		for_each_online_cpu(j)
+			seq_printf(p, "%10u ", cpu_pda(j)->apic_timer_irqs);
 		seq_putc(p, '\n');
 #endif
 		seq_printf(p, "ERR: %10u\n", atomic_read(&irq_err_count));
diff --git a/arch/x86_64/kernel/kprobes.c b/arch/x86_64/kernel/kprobes.c
index 8b866a8572c..14f0ced613b 100644
--- a/arch/x86_64/kernel/kprobes.c
+++ b/arch/x86_64/kernel/kprobes.c
@@ -222,9 +222,9 @@ void __kprobes arch_disarm_kprobe(struct kprobe *p)
 
 void __kprobes arch_remove_kprobe(struct kprobe *p)
 {
-	down(&kprobe_mutex);
+	mutex_lock(&kprobe_mutex);
 	free_insn_slot(p->ainsn.insn);
-	up(&kprobe_mutex);
+	mutex_unlock(&kprobe_mutex);
 }
 
 static inline void save_previous_kprobe(struct kprobe_ctlblk *kcb)
diff --git a/arch/x86_64/kernel/nmi.c b/arch/x86_64/kernel/nmi.c
index 5bf17e41cd2..66c009e10ba 100644
--- a/arch/x86_64/kernel/nmi.c
+++ b/arch/x86_64/kernel/nmi.c
@@ -162,9 +162,7 @@ int __init check_nmi_watchdog (void)
 	local_irq_enable();
 	mdelay((10*1000)/nmi_hz); // wait 10 ticks
 
-	for (cpu = 0; cpu < NR_CPUS; cpu++) {
-		if (!cpu_online(cpu))
-			continue;
+	for_each_online_cpu(cpu) {
 		if (cpu_pda(cpu)->__nmi_count - counts[cpu] <= 5) {
 			endflag = 1;
 			printk("CPU#%d: NMI appears to be stuck (%d->%d)!\n",
diff --git a/arch/x86_64/kernel/signal.c b/arch/x86_64/kernel/signal.c
index 5876df116c9..e5f5ce7909a 100644
--- a/arch/x86_64/kernel/signal.c
+++ b/arch/x86_64/kernel/signal.c
@@ -443,9 +443,6 @@ int do_signal(struct pt_regs *regs, sigset_t *oldset)
 	if (!user_mode(regs))
 		return 1;
 
-	if (try_to_freeze())
-		goto no_signal;
-
 	if (!oldset)
 		oldset = &current->blocked;
 
@@ -463,7 +460,6 @@ int do_signal(struct pt_regs *regs, sigset_t *oldset)
 		return handle_signal(signr, &info, &ka, oldset, regs);
 	}
 
- no_signal:
 	/* Did we come from a system call? */
 	if ((long)regs->orig_rax >= 0) {
 		/* Restart the system call - no handlers present */
diff --git a/arch/xtensa/kernel/irq.c b/arch/xtensa/kernel/irq.c
index 4cbf6d91571..51f9bed455f 100644
--- a/arch/xtensa/kernel/irq.c
+++ b/arch/xtensa/kernel/irq.c
@@ -83,9 +83,8 @@ int show_interrupts(struct seq_file *p, void *v)
 
 	if (i == 0) {
 		seq_printf(p, "           ");
-		for (j=0; j<NR_CPUS; j++)
-			if (cpu_online(j))
-				seq_printf(p, "CPU%d       ",j);
+		for_each_online_cpu(j)
+			seq_printf(p, "CPU%d       ",j);
 		seq_putc(p, '\n');
 	}
 
@@ -98,9 +97,8 @@ int show_interrupts(struct seq_file *p, void *v)
 #ifndef CONFIG_SMP
 		seq_printf(p, "%10u ", kstat_irqs(i));
 #else
-		for (j = 0; j < NR_CPUS; j++)
-			if (cpu_online(j))
-				seq_printf(p, "%10u ", kstat_cpu(j).irqs[i]);
+		for_each_online_cpu(j)
+			seq_printf(p, "%10u ", kstat_cpu(j).irqs[i]);
 #endif
 		seq_printf(p, " %14s", irq_desc[i].handler->typename);
 		seq_printf(p, "  %s", action->name);
@@ -113,9 +111,8 @@ skip:
 		spin_unlock_irqrestore(&irq_desc[i].lock, flags);
 	} else if (i == NR_IRQS) {
 		seq_printf(p, "NMI: ");
-		for (j = 0; j < NR_CPUS; j++)
-			if (cpu_online(j))
-				seq_printf(p, "%10u ", nmi_count(j));
+		for_each_online_cpu(j)
+			seq_printf(p, "%10u ", nmi_count(j));
 		seq_putc(p, '\n');
 		seq_printf(p, "ERR: %10u\n", atomic_read(&irq_err_count));
 	}
diff --git a/arch/xtensa/platform-iss/console.c b/arch/xtensa/platform-iss/console.c
index 94fdfe474ac..2a580efb58e 100644
--- a/arch/xtensa/platform-iss/console.c
+++ b/arch/xtensa/platform-iss/console.c
@@ -31,10 +31,6 @@
 #include <linux/tty.h>
 #include <linux/tty_flip.h>
 
-#ifdef SERIAL_INLINE
-#define _INLINE_ inline
-#endif
-
 #define SERIAL_MAX_NUM_LINES 1
 #define SERIAL_TIMER_VALUE (20 * HZ)
 
diff --git a/block/ioctl.c b/block/ioctl.c
index e1109491c23..35fdb7dc651 100644
--- a/block/ioctl.c
+++ b/block/ioctl.c
@@ -42,9 +42,9 @@ static int blkpg_ioctl(struct block_device *bdev, struct blkpg_ioctl_arg __user
 					return -EINVAL;
 			}
 			/* partition number in use? */
-			down(&bdev->bd_sem);
+			mutex_lock(&bdev->bd_mutex);
 			if (disk->part[part - 1]) {
-				up(&bdev->bd_sem);
+				mutex_unlock(&bdev->bd_mutex);
 				return -EBUSY;
 			}
 			/* overlap? */
@@ -55,13 +55,13 @@ static int blkpg_ioctl(struct block_device *bdev, struct blkpg_ioctl_arg __user
 					continue;
 				if (!(start+length <= s->start_sect ||
 				      start >= s->start_sect + s->nr_sects)) {
-					up(&bdev->bd_sem);
+					mutex_unlock(&bdev->bd_mutex);
 					return -EBUSY;
 				}
 			}
 			/* all seems OK */
 			add_partition(disk, part, start, length);
-			up(&bdev->bd_sem);
+			mutex_unlock(&bdev->bd_mutex);
 			return 0;
 		case BLKPG_DEL_PARTITION:
 			if (!disk->part[part-1])
@@ -71,9 +71,9 @@ static int blkpg_ioctl(struct block_device *bdev, struct blkpg_ioctl_arg __user
 			bdevp = bdget_disk(disk, part);
 			if (!bdevp)
 				return -ENOMEM;
-			down(&bdevp->bd_sem);
+			mutex_lock(&bdevp->bd_mutex);
 			if (bdevp->bd_openers) {
-				up(&bdevp->bd_sem);
+				mutex_unlock(&bdevp->bd_mutex);
 				bdput(bdevp);
 				return -EBUSY;
 			}
@@ -81,10 +81,10 @@ static int blkpg_ioctl(struct block_device *bdev, struct blkpg_ioctl_arg __user
 			fsync_bdev(bdevp);
 			invalidate_bdev(bdevp, 0);
 
-			down(&bdev->bd_sem);
+			mutex_lock(&bdev->bd_mutex);
 			delete_partition(disk, part);
-			up(&bdev->bd_sem);
-			up(&bdevp->bd_sem);
+			mutex_unlock(&bdev->bd_mutex);
+			mutex_unlock(&bdevp->bd_mutex);
 			bdput(bdevp);
 
 			return 0;
@@ -102,10 +102,10 @@ static int blkdev_reread_part(struct block_device *bdev)
 		return -EINVAL;
 	if (!capable(CAP_SYS_ADMIN))
 		return -EACCES;
-	if (down_trylock(&bdev->bd_sem))
+	if (!mutex_trylock(&bdev->bd_mutex))
 		return -EBUSY;
 	res = rescan_partitions(disk, bdev);
-	up(&bdev->bd_sem);
+	mutex_unlock(&bdev->bd_mutex);
 	return res;
 }
 
diff --git a/drivers/base/power/suspend.c b/drivers/base/power/suspend.c
index 8660779fb28..bdb60663f2e 100644
--- a/drivers/base/power/suspend.c
+++ b/drivers/base/power/suspend.c
@@ -8,6 +8,7 @@
  *
  */
 
+#include <linux/vt_kern.h>
 #include <linux/device.h>
 #include "../base.h"
 #include "power.h"
@@ -62,7 +63,6 @@ int suspend_device(struct device * dev, pm_message_t state)
 	return error;
 }
 
-
 /**
  *	device_suspend - Save state and stop all devices in system.
  *	@state:		Power state to put each device in.
@@ -82,6 +82,9 @@ int device_suspend(pm_message_t state)
 {
 	int error = 0;
 
+	if (!is_console_suspend_safe())
+		return -EINVAL;
+
 	down(&dpm_sem);
 	down(&dpm_list_sem);
 	while (!list_empty(&dpm_active) && error == 0) {
diff --git a/drivers/block/cciss.c b/drivers/block/cciss.c
index cf39cf9aac2..e29b8926f80 100644
--- a/drivers/block/cciss.c
+++ b/drivers/block/cciss.c
@@ -3268,8 +3268,8 @@ clean2:
 	unregister_blkdev(hba[i]->major, hba[i]->devname);
 clean1:
 	release_io_mem(hba[i]);
-	free_hba(i);
 	hba[i]->busy_initializing = 0;
+	free_hba(i);
 	return(-1);
 }
 
diff --git a/drivers/block/floppy.c b/drivers/block/floppy.c
index d23b54332d7..fb2d0be7cde 100644
--- a/drivers/block/floppy.c
+++ b/drivers/block/floppy.c
@@ -179,6 +179,7 @@ static int print_unex = 1;
 #include <linux/devfs_fs_kernel.h>
 #include <linux/platform_device.h>
 #include <linux/buffer_head.h>	/* for invalidate_buffers() */
+#include <linux/mutex.h>
 
 /*
  * PS/2 floppies have much slower step rates than regular floppies.
@@ -413,7 +414,7 @@ static struct floppy_write_errors write_errors[N_DRIVE];
 static struct timer_list motor_off_timer[N_DRIVE];
 static struct gendisk *disks[N_DRIVE];
 static struct block_device *opened_bdev[N_DRIVE];
-static DECLARE_MUTEX(open_lock);
+static DEFINE_MUTEX(open_lock);
 static struct floppy_raw_cmd *raw_cmd, default_raw_cmd;
 
 /*
@@ -3333,7 +3334,7 @@ static inline int set_geometry(unsigned int cmd, struct floppy_struct *g,
 	if (type) {
 		if (!capable(CAP_SYS_ADMIN))
 			return -EPERM;
-		down(&open_lock);
+		mutex_lock(&open_lock);
 		LOCK_FDC(drive, 1);
 		floppy_type[type] = *g;
 		floppy_type[type].name = "user format";
@@ -3347,7 +3348,7 @@ static inline int set_geometry(unsigned int cmd, struct floppy_struct *g,
 				continue;
 			__invalidate_device(bdev);
 		}
-		up(&open_lock);
+		mutex_unlock(&open_lock);
 	} else {
 		int oldStretch;
 		LOCK_FDC(drive, 1);
@@ -3674,7 +3675,7 @@ static int floppy_release(struct inode *inode, struct file *filp)
 {
 	int drive = (long)inode->i_bdev->bd_disk->private_data;
 
-	down(&open_lock);
+	mutex_lock(&open_lock);
 	if (UDRS->fd_ref < 0)
 		UDRS->fd_ref = 0;
 	else if (!UDRS->fd_ref--) {
@@ -3684,7 +3685,7 @@ static int floppy_release(struct inode *inode, struct file *filp)
 	if (!UDRS->fd_ref)
 		opened_bdev[drive] = NULL;
 	floppy_release_irq_and_dma();
-	up(&open_lock);
+	mutex_unlock(&open_lock);
 	return 0;
 }
 
@@ -3702,7 +3703,7 @@ static int floppy_open(struct inode *inode, struct file *filp)
 	char *tmp;
 
 	filp->private_data = (void *)0;
-	down(&open_lock);
+	mutex_lock(&open_lock);
 	old_dev = UDRS->fd_device;
 	if (opened_bdev[drive] && opened_bdev[drive] != inode->i_bdev)
 		goto out2;
@@ -3785,7 +3786,7 @@ static int floppy_open(struct inode *inode, struct file *filp)
 		if ((filp->f_mode & 2) && !(UTESTF(FD_DISK_WRITABLE)))
 			goto out;
 	}
-	up(&open_lock);
+	mutex_unlock(&open_lock);
 	return 0;
 out:
 	if (UDRS->fd_ref < 0)
@@ -3796,7 +3797,7 @@ out:
 		opened_bdev[drive] = NULL;
 	floppy_release_irq_and_dma();
 out2:
-	up(&open_lock);
+	mutex_unlock(&open_lock);
 	return res;
 }
 
diff --git a/drivers/block/loop.c b/drivers/block/loop.c
index 0010704739e..74bf0255e98 100644
--- a/drivers/block/loop.c
+++ b/drivers/block/loop.c
@@ -1144,7 +1144,7 @@ static int lo_ioctl(struct inode * inode, struct file * file,
 	struct loop_device *lo = inode->i_bdev->bd_disk->private_data;
 	int err;
 
-	down(&lo->lo_ctl_mutex);
+	mutex_lock(&lo->lo_ctl_mutex);
 	switch (cmd) {
 	case LOOP_SET_FD:
 		err = loop_set_fd(lo, file, inode->i_bdev, arg);
@@ -1170,7 +1170,7 @@ static int lo_ioctl(struct inode * inode, struct file * file,
 	default:
 		err = lo->ioctl ? lo->ioctl(lo, cmd, arg) : -EINVAL;
 	}
-	up(&lo->lo_ctl_mutex);
+	mutex_unlock(&lo->lo_ctl_mutex);
 	return err;
 }
 
@@ -1178,9 +1178,9 @@ static int lo_open(struct inode *inode, struct file *file)
 {
 	struct loop_device *lo = inode->i_bdev->bd_disk->private_data;
 
-	down(&lo->lo_ctl_mutex);
+	mutex_lock(&lo->lo_ctl_mutex);
 	lo->lo_refcnt++;
-	up(&lo->lo_ctl_mutex);
+	mutex_unlock(&lo->lo_ctl_mutex);
 
 	return 0;
 }
@@ -1189,9 +1189,9 @@ static int lo_release(struct inode *inode, struct file *file)
 {
 	struct loop_device *lo = inode->i_bdev->bd_disk->private_data;
 
-	down(&lo->lo_ctl_mutex);
+	mutex_lock(&lo->lo_ctl_mutex);
 	--lo->lo_refcnt;
-	up(&lo->lo_ctl_mutex);
+	mutex_unlock(&lo->lo_ctl_mutex);
 
 	return 0;
 }
@@ -1233,12 +1233,12 @@ int loop_unregister_transfer(int number)
 	xfer_funcs[n] = NULL;
 
 	for (lo = &loop_dev[0]; lo < &loop_dev[max_loop]; lo++) {
-		down(&lo->lo_ctl_mutex);
+		mutex_lock(&lo->lo_ctl_mutex);
 
 		if (lo->lo_encryption == xfer)
 			loop_release_xfer(lo);
 
-		up(&lo->lo_ctl_mutex);
+		mutex_unlock(&lo->lo_ctl_mutex);
 	}
 
 	return 0;
@@ -1285,7 +1285,7 @@ static int __init loop_init(void)
 		lo->lo_queue = blk_alloc_queue(GFP_KERNEL);
 		if (!lo->lo_queue)
 			goto out_mem4;
-		init_MUTEX(&lo->lo_ctl_mutex);
+		mutex_init(&lo->lo_ctl_mutex);
 		init_completion(&lo->lo_done);
 		init_completion(&lo->lo_bh_done);
 		lo->lo_number = i;
diff --git a/drivers/block/nbd.c b/drivers/block/nbd.c
index 6997d8e6bfb..a9bde30dada 100644
--- a/drivers/block/nbd.c
+++ b/drivers/block/nbd.c
@@ -459,9 +459,9 @@ static void do_nbd_request(request_queue_t * q)
 		req->errors = 0;
 		spin_unlock_irq(q->queue_lock);
 
-		down(&lo->tx_lock);
+		mutex_lock(&lo->tx_lock);
 		if (unlikely(!lo->sock)) {
-			up(&lo->tx_lock);
+			mutex_unlock(&lo->tx_lock);
 			printk(KERN_ERR "%s: Attempted send on closed socket\n",
 			       lo->disk->disk_name);
 			req->errors++;
@@ -484,7 +484,7 @@ static void do_nbd_request(request_queue_t * q)
 		}
 
 		lo->active_req = NULL;
-		up(&lo->tx_lock);
+		mutex_unlock(&lo->tx_lock);
 		wake_up_all(&lo->active_wq);
 
 		spin_lock_irq(q->queue_lock);
@@ -534,9 +534,9 @@ static int nbd_ioctl(struct inode *inode, struct file *file,
  
 	case NBD_CLEAR_SOCK:
 		error = 0;
-		down(&lo->tx_lock);
+		mutex_lock(&lo->tx_lock);
 		lo->sock = NULL;
-		up(&lo->tx_lock);
+		mutex_unlock(&lo->tx_lock);
 		file = lo->file;
 		lo->file = NULL;
 		nbd_clear_que(lo);
@@ -590,7 +590,7 @@ static int nbd_ioctl(struct inode *inode, struct file *file,
 		 * FIXME: This code is duplicated from sys_shutdown, but
 		 * there should be a more generic interface rather than
 		 * calling socket ops directly here */
-		down(&lo->tx_lock);
+		mutex_lock(&lo->tx_lock);
 		if (lo->sock) {
 			printk(KERN_WARNING "%s: shutting down socket\n",
 				lo->disk->disk_name);
@@ -598,7 +598,7 @@ static int nbd_ioctl(struct inode *inode, struct file *file,
 				SEND_SHUTDOWN|RCV_SHUTDOWN);
 			lo->sock = NULL;
 		}
-		up(&lo->tx_lock);
+		mutex_unlock(&lo->tx_lock);
 		file = lo->file;
 		lo->file = NULL;
 		nbd_clear_que(lo);
@@ -683,7 +683,7 @@ static int __init nbd_init(void)
 		nbd_dev[i].flags = 0;
 		spin_lock_init(&nbd_dev[i].queue_lock);
 		INIT_LIST_HEAD(&nbd_dev[i].queue_head);
-		init_MUTEX(&nbd_dev[i].tx_lock);
+		mutex_init(&nbd_dev[i].tx_lock);
 		init_waitqueue_head(&nbd_dev[i].active_wq);
 		nbd_dev[i].blksize = 1024;
 		nbd_dev[i].bytesize = 0x7ffffc00ULL << 10; /* 2TB */
diff --git a/drivers/block/pktcdvd.c b/drivers/block/pktcdvd.c
index 476a5b553f3..1d261f985f3 100644
--- a/drivers/block/pktcdvd.c
+++ b/drivers/block/pktcdvd.c
@@ -56,6 +56,7 @@
 #include <linux/seq_file.h>
 #include <linux/miscdevice.h>
 #include <linux/suspend.h>
+#include <linux/mutex.h>
 #include <scsi/scsi_cmnd.h>
 #include <scsi/scsi_ioctl.h>
 #include <scsi/scsi.h>
@@ -81,7 +82,7 @@
 static struct pktcdvd_device *pkt_devs[MAX_WRITERS];
 static struct proc_dir_entry *pkt_proc;
 static int pkt_major;
-static struct semaphore ctl_mutex;	/* Serialize open/close/setup/teardown */
+static struct mutex ctl_mutex;	/* Serialize open/close/setup/teardown */
 static mempool_t *psd_pool;
 
 
@@ -2018,7 +2019,7 @@ static int pkt_open(struct inode *inode, struct file *file)
 
 	VPRINTK("pktcdvd: entering open\n");
 
-	down(&ctl_mutex);
+	mutex_lock(&ctl_mutex);
 	pd = pkt_find_dev_from_minor(iminor(inode));
 	if (!pd) {
 		ret = -ENODEV;
@@ -2044,14 +2045,14 @@ static int pkt_open(struct inode *inode, struct file *file)
 		set_blocksize(inode->i_bdev, CD_FRAMESIZE);
 	}
 
-	up(&ctl_mutex);
+	mutex_unlock(&ctl_mutex);
 	return 0;
 
 out_dec:
 	pd->refcnt--;
 out:
 	VPRINTK("pktcdvd: failed open (%d)\n", ret);
-	up(&ctl_mutex);
+	mutex_unlock(&ctl_mutex);
 	return ret;
 }
 
@@ -2060,14 +2061,14 @@ static int pkt_close(struct inode *inode, struct file *file)
 	struct pktcdvd_device *pd = inode->i_bdev->bd_disk->private_data;
 	int ret = 0;
 
-	down(&ctl_mutex);
+	mutex_lock(&ctl_mutex);
 	pd->refcnt--;
 	BUG_ON(pd->refcnt < 0);
 	if (pd->refcnt == 0) {
 		int flush = test_bit(PACKET_WRITABLE, &pd->flags);
 		pkt_release_dev(pd, flush);
 	}
-	up(&ctl_mutex);
+	mutex_unlock(&ctl_mutex);
 	return ret;
 }
 
@@ -2596,21 +2597,21 @@ static int pkt_ctl_ioctl(struct inode *inode, struct file *file, unsigned int cm
 	case PKT_CTRL_CMD_SETUP:
 		if (!capable(CAP_SYS_ADMIN))
 			return -EPERM;
-		down(&ctl_mutex);
+		mutex_lock(&ctl_mutex);
 		ret = pkt_setup_dev(&ctrl_cmd);
-		up(&ctl_mutex);
+		mutex_unlock(&ctl_mutex);
 		break;
 	case PKT_CTRL_CMD_TEARDOWN:
 		if (!capable(CAP_SYS_ADMIN))
 			return -EPERM;
-		down(&ctl_mutex);
+		mutex_lock(&ctl_mutex);
 		ret = pkt_remove_dev(&ctrl_cmd);
-		up(&ctl_mutex);
+		mutex_unlock(&ctl_mutex);
 		break;
 	case PKT_CTRL_CMD_STATUS:
-		down(&ctl_mutex);
+		mutex_lock(&ctl_mutex);
 		pkt_get_status(&ctrl_cmd);
-		up(&ctl_mutex);
+		mutex_unlock(&ctl_mutex);
 		break;
 	default:
 		return -ENOTTY;
@@ -2656,7 +2657,7 @@ static int __init pkt_init(void)
 		goto out;
 	}
 
-	init_MUTEX(&ctl_mutex);
+	mutex_init(&ctl_mutex);
 
 	pkt_proc = proc_mkdir("pktcdvd", proc_root_driver);
 
diff --git a/drivers/block/rd.c b/drivers/block/rd.c
index ffd6abd6d5a..1c54f46d3f7 100644
--- a/drivers/block/rd.c
+++ b/drivers/block/rd.c
@@ -310,12 +310,12 @@ static int rd_ioctl(struct inode *inode, struct file *file,
 	 * cache
 	 */
 	error = -EBUSY;
-	down(&bdev->bd_sem);
+	mutex_lock(&bdev->bd_mutex);
 	if (bdev->bd_openers <= 2) {
 		truncate_inode_pages(bdev->bd_inode->i_mapping, 0);
 		error = 0;
 	}
-	up(&bdev->bd_sem);
+	mutex_unlock(&bdev->bd_mutex);
 	return error;
 }
 
diff --git a/drivers/cdrom/cdrom.c b/drivers/cdrom/cdrom.c
index 879bbc26ce9..a59876a0bfa 100644
--- a/drivers/cdrom/cdrom.c
+++ b/drivers/cdrom/cdrom.c
@@ -407,7 +407,6 @@ int register_cdrom(struct cdrom_device_info *cdi)
 	ENSURE(get_mcn, CDC_MCN);
 	ENSURE(reset, CDC_RESET);
 	ENSURE(audio_ioctl, CDC_PLAY_AUDIO);
-	ENSURE(dev_ioctl, CDC_IOCTLS);
 	ENSURE(generic_packet, CDC_GENERIC_PACKET);
 	cdi->mc_flags = 0;
 	cdo->n_minors = 0;
@@ -2196,395 +2195,586 @@ retry:
 	return cdrom_read_cdda_old(cdi, ubuf, lba, nframes);	
 }
 
-/* Just about every imaginable ioctl is supported in the Uniform layer
- * these days. ATAPI / SCSI specific code now mainly resides in
- * mmc_ioct().
- */
-int cdrom_ioctl(struct file * file, struct cdrom_device_info *cdi,
-		struct inode *ip, unsigned int cmd, unsigned long arg)
+static int cdrom_ioctl_multisession(struct cdrom_device_info *cdi,
+		void __user *argp)
 {
-	struct cdrom_device_ops *cdo = cdi->ops;
+	struct cdrom_multisession ms_info;
+	u8 requested_format;
 	int ret;
 
-	/* Try the generic SCSI command ioctl's first.. */
-	ret = scsi_cmd_ioctl(file, ip->i_bdev->bd_disk, cmd, (void __user *)arg);
-	if (ret != -ENOTTY)
+	cdinfo(CD_DO_IOCTL, "entering CDROMMULTISESSION\n");
+
+	if (!(cdi->ops->capability & CDC_MULTI_SESSION))
+		return -ENOSYS;
+
+	if (copy_from_user(&ms_info, argp, sizeof(ms_info)))
+		return -EFAULT;
+
+	requested_format = ms_info.addr_format;
+	if (requested_format != CDROM_MSF && requested_format != CDROM_LBA)
+		return -EINVAL;
+	ms_info.addr_format = CDROM_LBA;
+
+	ret = cdi->ops->get_last_session(cdi, &ms_info);
+	if (ret)
 		return ret;
 
-	/* the first few commands do not deal with audio drive_info, but
-	   only with routines in cdrom device operations. */
-	switch (cmd) {
-	case CDROMMULTISESSION: {
-		struct cdrom_multisession ms_info;
-		u_char requested_format;
-		cdinfo(CD_DO_IOCTL, "entering CDROMMULTISESSION\n"); 
-                if (!(cdo->capability & CDC_MULTI_SESSION))
-                        return -ENOSYS;
-		IOCTL_IN(arg, struct cdrom_multisession, ms_info);
-		requested_format = ms_info.addr_format;
-		if (!((requested_format == CDROM_MSF) ||
-			(requested_format == CDROM_LBA)))
-				return -EINVAL;
-		ms_info.addr_format = CDROM_LBA;
-		if ((ret=cdo->get_last_session(cdi, &ms_info)))
+	sanitize_format(&ms_info.addr, &ms_info.addr_format, requested_format);
+
+	if (copy_to_user(argp, &ms_info, sizeof(ms_info)))
+		return -EFAULT;
+
+	cdinfo(CD_DO_IOCTL, "CDROMMULTISESSION successful\n");
+	return 0;
+}
+
+static int cdrom_ioctl_eject(struct cdrom_device_info *cdi)
+{
+	cdinfo(CD_DO_IOCTL, "entering CDROMEJECT\n");
+
+	if (!CDROM_CAN(CDC_OPEN_TRAY))
+		return -ENOSYS;
+	if (cdi->use_count != 1 || keeplocked)
+		return -EBUSY;
+	if (CDROM_CAN(CDC_LOCK)) {
+		int ret = cdi->ops->lock_door(cdi, 0);
+		if (ret)
 			return ret;
-		sanitize_format(&ms_info.addr, &ms_info.addr_format,
-				requested_format);
-		IOCTL_OUT(arg, struct cdrom_multisession, ms_info);
-		cdinfo(CD_DO_IOCTL, "CDROMMULTISESSION successful\n"); 
-		return 0;
-		}
+	}
 
-	case CDROMEJECT: {
-		cdinfo(CD_DO_IOCTL, "entering CDROMEJECT\n"); 
-		if (!CDROM_CAN(CDC_OPEN_TRAY))
-			return -ENOSYS;
-		if (cdi->use_count != 1 || keeplocked)
-			return -EBUSY;
-		if (CDROM_CAN(CDC_LOCK))
-			if ((ret=cdo->lock_door(cdi, 0)))
-				return ret;
+	return cdi->ops->tray_move(cdi, 1);
+}
 
-		return cdo->tray_move(cdi, 1);
-		}
+static int cdrom_ioctl_closetray(struct cdrom_device_info *cdi)
+{
+	cdinfo(CD_DO_IOCTL, "entering CDROMCLOSETRAY\n");
 
-	case CDROMCLOSETRAY: {
-		cdinfo(CD_DO_IOCTL, "entering CDROMCLOSETRAY\n"); 
-		if (!CDROM_CAN(CDC_CLOSE_TRAY))
-			return -ENOSYS;
-		return cdo->tray_move(cdi, 0);
-		}
+	if (!CDROM_CAN(CDC_CLOSE_TRAY))
+		return -ENOSYS;
+	return cdi->ops->tray_move(cdi, 0);
+}
 
-	case CDROMEJECT_SW: {
-		cdinfo(CD_DO_IOCTL, "entering CDROMEJECT_SW\n"); 
-		if (!CDROM_CAN(CDC_OPEN_TRAY))
-			return -ENOSYS;
-		if (keeplocked)
-			return -EBUSY;
-		cdi->options &= ~(CDO_AUTO_CLOSE | CDO_AUTO_EJECT);
-		if (arg)
-			cdi->options |= CDO_AUTO_CLOSE | CDO_AUTO_EJECT;
-		return 0;
-		}
+static int cdrom_ioctl_eject_sw(struct cdrom_device_info *cdi,
+		unsigned long arg)
+{
+	cdinfo(CD_DO_IOCTL, "entering CDROMEJECT_SW\n");
 
-	case CDROM_MEDIA_CHANGED: {
-		struct cdrom_changer_info *info;
-		int changed;
+	if (!CDROM_CAN(CDC_OPEN_TRAY))
+		return -ENOSYS;
+	if (keeplocked)
+		return -EBUSY;
 
-		cdinfo(CD_DO_IOCTL, "entering CDROM_MEDIA_CHANGED\n"); 
-		if (!CDROM_CAN(CDC_MEDIA_CHANGED))
-			return -ENOSYS;
+	cdi->options &= ~(CDO_AUTO_CLOSE | CDO_AUTO_EJECT);
+	if (arg)
+		cdi->options |= CDO_AUTO_CLOSE | CDO_AUTO_EJECT;
+	return 0;
+}
 
-		/* cannot select disc or select current disc */
-		if (!CDROM_CAN(CDC_SELECT_DISC) || arg == CDSL_CURRENT)
-			return media_changed(cdi, 1);
+static int cdrom_ioctl_media_changed(struct cdrom_device_info *cdi,
+		unsigned long arg)
+{
+	struct cdrom_changer_info *info;
+	int ret;
 
-		if ((unsigned int)arg >= cdi->capacity)
-			return -EINVAL;
+	cdinfo(CD_DO_IOCTL, "entering CDROM_MEDIA_CHANGED\n");
 
-		info = kmalloc(sizeof(*info), GFP_KERNEL);
-		if (!info)
-			return -ENOMEM;
+	if (!CDROM_CAN(CDC_MEDIA_CHANGED))
+		return -ENOSYS;
 
-		if ((ret = cdrom_read_mech_status(cdi, info))) {
-			kfree(info);
-			return ret;
-		}
+	/* cannot select disc or select current disc */
+	if (!CDROM_CAN(CDC_SELECT_DISC) || arg == CDSL_CURRENT)
+		return media_changed(cdi, 1);
 
-		changed = info->slots[arg].change;
-		kfree(info);
-		return changed;
-		}
+	if ((unsigned int)arg >= cdi->capacity)
+		return -EINVAL;
 
-	case CDROM_SET_OPTIONS: {
-		cdinfo(CD_DO_IOCTL, "entering CDROM_SET_OPTIONS\n"); 
-		/* options need to be in sync with capability. too late for
-		   that, so we have to check each one separately... */
-		switch (arg) {
-		case CDO_USE_FFLAGS:
-		case CDO_CHECK_TYPE:
-			break;
-		case CDO_LOCK:
-			if (!CDROM_CAN(CDC_LOCK))
-				return -ENOSYS;
-			break;
-		case 0:
-			return cdi->options;
-		/* default is basically CDO_[AUTO_CLOSE|AUTO_EJECT] */
-		default:
-			if (!CDROM_CAN(arg))
-				return -ENOSYS;
-		}
-		cdi->options |= (int) arg;
-		return cdi->options;
-		}
+	info = kmalloc(sizeof(*info), GFP_KERNEL);
+	if (!info)
+		return -ENOMEM;
 
-	case CDROM_CLEAR_OPTIONS: {
-		cdinfo(CD_DO_IOCTL, "entering CDROM_CLEAR_OPTIONS\n"); 
-		cdi->options &= ~(int) arg;
-		return cdi->options;
-		}
+	ret = cdrom_read_mech_status(cdi, info);
+	if (!ret)
+		ret = info->slots[arg].change;
+	kfree(info);
+	return ret;
+}
 
-	case CDROM_SELECT_SPEED: {
-		cdinfo(CD_DO_IOCTL, "entering CDROM_SELECT_SPEED\n"); 
-		if (!CDROM_CAN(CDC_SELECT_SPEED))
-			return -ENOSYS;
-		return cdo->select_speed(cdi, arg);
-		}
+static int cdrom_ioctl_set_options(struct cdrom_device_info *cdi,
+		unsigned long arg)
+{
+	cdinfo(CD_DO_IOCTL, "entering CDROM_SET_OPTIONS\n");
 
-	case CDROM_SELECT_DISC: {
-		cdinfo(CD_DO_IOCTL, "entering CDROM_SELECT_DISC\n"); 
-		if (!CDROM_CAN(CDC_SELECT_DISC))
+	/*
+	 * Options need to be in sync with capability.
+	 * Too late for that, so we have to check each one separately.
+	 */
+	switch (arg) {
+	case CDO_USE_FFLAGS:
+	case CDO_CHECK_TYPE:
+		break;
+	case CDO_LOCK:
+		if (!CDROM_CAN(CDC_LOCK))
+			return -ENOSYS;
+		break;
+	case 0:
+		return cdi->options;
+	/* default is basically CDO_[AUTO_CLOSE|AUTO_EJECT] */
+	default:
+		if (!CDROM_CAN(arg))
 			return -ENOSYS;
+	}
+	cdi->options |= (int) arg;
+	return cdi->options;
+}
 
-                if ((arg != CDSL_CURRENT) && (arg != CDSL_NONE))
-			if ((int)arg >= cdi->capacity)
-				return -EINVAL;
-
-		/* cdo->select_disc is a hook to allow a driver-specific
-		 * way of seleting disc.  However, since there is no
-		 * equiv hook for cdrom_slot_status this may not 
-		 * actually be useful...
-		 */
-		if (cdo->select_disc != NULL)
-			return cdo->select_disc(cdi, arg);
-
-		/* no driver specific select_disc(), call our own */
-		cdinfo(CD_CHANGER, "Using generic cdrom_select_disc()\n"); 
-		return cdrom_select_disc(cdi, arg);
-		}
+static int cdrom_ioctl_clear_options(struct cdrom_device_info *cdi,
+		unsigned long arg)
+{
+	cdinfo(CD_DO_IOCTL, "entering CDROM_CLEAR_OPTIONS\n");
 
-	case CDROMRESET: {
-		if (!capable(CAP_SYS_ADMIN))
-			return -EACCES;
-		cdinfo(CD_DO_IOCTL, "entering CDROM_RESET\n");
-		if (!CDROM_CAN(CDC_RESET))
-			return -ENOSYS;
-		invalidate_bdev(ip->i_bdev, 0);
-		return cdo->reset(cdi);
-		}
+	cdi->options &= ~(int) arg;
+	return cdi->options;
+}
 
-	case CDROM_LOCKDOOR: {
-		cdinfo(CD_DO_IOCTL, "%socking door.\n", arg ? "L" : "Unl");
-		if (!CDROM_CAN(CDC_LOCK))
-			return -EDRIVE_CANT_DO_THIS;
-		keeplocked = arg ? 1 : 0;
-		/* don't unlock the door on multiple opens,but allow root
-		 * to do so */
-		if ((cdi->use_count != 1) && !arg && !capable(CAP_SYS_ADMIN))
-			return -EBUSY;
-		return cdo->lock_door(cdi, arg);
-		}
+static int cdrom_ioctl_select_speed(struct cdrom_device_info *cdi,
+		unsigned long arg)
+{
+	cdinfo(CD_DO_IOCTL, "entering CDROM_SELECT_SPEED\n");
 
-	case CDROM_DEBUG: {
-		if (!capable(CAP_SYS_ADMIN))
-			return -EACCES;
-		cdinfo(CD_DO_IOCTL, "%sabling debug.\n", arg ? "En" : "Dis");
-		debug = arg ? 1 : 0;
-		return debug;
-		}
+	if (!CDROM_CAN(CDC_SELECT_SPEED))
+		return -ENOSYS;
+	return cdi->ops->select_speed(cdi, arg);
+}
 
-	case CDROM_GET_CAPABILITY: {
-		cdinfo(CD_DO_IOCTL, "entering CDROM_GET_CAPABILITY\n");
-		return (cdo->capability & ~cdi->mask);
-		}
+static int cdrom_ioctl_select_disc(struct cdrom_device_info *cdi,
+		unsigned long arg)
+{
+	cdinfo(CD_DO_IOCTL, "entering CDROM_SELECT_DISC\n");
+
+	if (!CDROM_CAN(CDC_SELECT_DISC))
+		return -ENOSYS;
+
+	if (arg != CDSL_CURRENT && arg != CDSL_NONE) {
+		if ((int)arg >= cdi->capacity)
+			return -EINVAL;
+	}
+
+	/*
+	 * ->select_disc is a hook to allow a driver-specific way of
+	 * seleting disc.  However, since there is no equivalent hook for
+	 * cdrom_slot_status this may not actually be useful...
+	 */
+	if (cdi->ops->select_disc)
+		return cdi->ops->select_disc(cdi, arg);
+
+	cdinfo(CD_CHANGER, "Using generic cdrom_select_disc()\n");
+	return cdrom_select_disc(cdi, arg);
+}
+
+static int cdrom_ioctl_reset(struct cdrom_device_info *cdi,
+		struct block_device *bdev)
+{
+	cdinfo(CD_DO_IOCTL, "entering CDROM_RESET\n");
+
+	if (!capable(CAP_SYS_ADMIN))
+		return -EACCES;
+	if (!CDROM_CAN(CDC_RESET))
+		return -ENOSYS;
+	invalidate_bdev(bdev, 0);
+	return cdi->ops->reset(cdi);
+}
+
+static int cdrom_ioctl_lock_door(struct cdrom_device_info *cdi,
+		unsigned long arg)
+{
+	cdinfo(CD_DO_IOCTL, "%socking door.\n", arg ? "L" : "Unl");
+
+	if (!CDROM_CAN(CDC_LOCK))
+		return -EDRIVE_CANT_DO_THIS;
+
+	keeplocked = arg ? 1 : 0;
+
+	/*
+	 * Don't unlock the door on multiple opens by default, but allow
+	 * root to do so.
+	 */
+	if (cdi->use_count != 1 && !arg && !capable(CAP_SYS_ADMIN))
+		return -EBUSY;
+	return cdi->ops->lock_door(cdi, arg);
+}
+
+static int cdrom_ioctl_debug(struct cdrom_device_info *cdi,
+		unsigned long arg)
+{
+	cdinfo(CD_DO_IOCTL, "%sabling debug.\n", arg ? "En" : "Dis");
+
+	if (!capable(CAP_SYS_ADMIN))
+		return -EACCES;
+	debug = arg ? 1 : 0;
+	return debug;
+}
 
-/* The following function is implemented, although very few audio
+static int cdrom_ioctl_get_capability(struct cdrom_device_info *cdi)
+{
+	cdinfo(CD_DO_IOCTL, "entering CDROM_GET_CAPABILITY\n");
+	return (cdi->ops->capability & ~cdi->mask);
+}
+
+/*
+ * The following function is implemented, although very few audio
  * discs give Universal Product Code information, which should just be
  * the Medium Catalog Number on the box.  Note, that the way the code
  * is written on the CD is /not/ uniform across all discs!
  */
-	case CDROM_GET_MCN: {
-		struct cdrom_mcn mcn;
-		cdinfo(CD_DO_IOCTL, "entering CDROM_GET_MCN\n"); 
-		if (!(cdo->capability & CDC_MCN))
-			return -ENOSYS;
-		if ((ret=cdo->get_mcn(cdi, &mcn)))
-			return ret;
-		IOCTL_OUT(arg, struct cdrom_mcn, mcn);
-		cdinfo(CD_DO_IOCTL, "CDROM_GET_MCN successful\n"); 
-		return 0;
-		}
+static int cdrom_ioctl_get_mcn(struct cdrom_device_info *cdi,
+		void __user *argp)
+{
+	struct cdrom_mcn mcn;
+	int ret;
 
-	case CDROM_DRIVE_STATUS: {
-		cdinfo(CD_DO_IOCTL, "entering CDROM_DRIVE_STATUS\n"); 
-		if (!(cdo->capability & CDC_DRIVE_STATUS))
-			return -ENOSYS;
-		if (!CDROM_CAN(CDC_SELECT_DISC))
-			return cdo->drive_status(cdi, CDSL_CURRENT);
-                if ((arg == CDSL_CURRENT) || (arg == CDSL_NONE)) 
-			return cdo->drive_status(cdi, CDSL_CURRENT);
-		if (((int)arg >= cdi->capacity))
-			return -EINVAL;
-		return cdrom_slot_status(cdi, arg);
-		}
+	cdinfo(CD_DO_IOCTL, "entering CDROM_GET_MCN\n");
 
-	/* Ok, this is where problems start.  The current interface for the
-	   CDROM_DISC_STATUS ioctl is flawed.  It makes the false assumption
-	   that CDs are all CDS_DATA_1 or all CDS_AUDIO, etc.  Unfortunatly,
-	   while this is often the case, it is also very common for CDs to
-	   have some tracks with data, and some tracks with audio.  Just 
-	   because I feel like it, I declare the following to be the best
-	   way to cope.  If the CD has ANY data tracks on it, it will be
-	   returned as a data CD.  If it has any XA tracks, I will return
-	   it as that.  Now I could simplify this interface by combining these 
-	   returns with the above, but this more clearly demonstrates
-	   the problem with the current interface.  Too bad this wasn't 
-	   designed to use bitmasks...         -Erik 
-
-	   Well, now we have the option CDS_MIXED: a mixed-type CD. 
-	   User level programmers might feel the ioctl is not very useful.
-	   					---david
-	*/
-	case CDROM_DISC_STATUS: {
-		tracktype tracks;
-		cdinfo(CD_DO_IOCTL, "entering CDROM_DISC_STATUS\n"); 
-		cdrom_count_tracks(cdi, &tracks);
-		if (tracks.error) 
-			return(tracks.error);
-
-		/* Policy mode on */
-		if (tracks.audio > 0) {
-			if (tracks.data==0 && tracks.cdi==0 && tracks.xa==0) 
-				return CDS_AUDIO;
-			else
-				return CDS_MIXED;
-		}
-		if (tracks.cdi > 0) return CDS_XA_2_2;
-		if (tracks.xa > 0) return CDS_XA_2_1;
-		if (tracks.data > 0) return CDS_DATA_1;
-		/* Policy mode off */
+	if (!(cdi->ops->capability & CDC_MCN))
+		return -ENOSYS;
+	ret = cdi->ops->get_mcn(cdi, &mcn);
+	if (ret)
+		return ret;
 
-		cdinfo(CD_WARNING,"This disc doesn't have any tracks I recognize!\n");
-		return CDS_NO_INFO;
-		}
+	if (copy_to_user(argp, &mcn, sizeof(mcn)))
+		return -EFAULT;
+	cdinfo(CD_DO_IOCTL, "CDROM_GET_MCN successful\n");
+	return 0;
+}
 
-	case CDROM_CHANGER_NSLOTS: {
-		cdinfo(CD_DO_IOCTL, "entering CDROM_CHANGER_NSLOTS\n"); 
-		return cdi->capacity;
-		}
+static int cdrom_ioctl_drive_status(struct cdrom_device_info *cdi,
+		unsigned long arg)
+{
+	cdinfo(CD_DO_IOCTL, "entering CDROM_DRIVE_STATUS\n");
+
+	if (!(cdi->ops->capability & CDC_DRIVE_STATUS))
+		return -ENOSYS;
+	if (!CDROM_CAN(CDC_SELECT_DISC) ||
+	    (arg == CDSL_CURRENT || arg == CDSL_NONE))
+		return cdi->ops->drive_status(cdi, CDSL_CURRENT);
+	if (((int)arg >= cdi->capacity))
+		return -EINVAL;
+	return cdrom_slot_status(cdi, arg);
+}
+
+/*
+ * Ok, this is where problems start.  The current interface for the
+ * CDROM_DISC_STATUS ioctl is flawed.  It makes the false assumption that
+ * CDs are all CDS_DATA_1 or all CDS_AUDIO, etc.  Unfortunatly, while this
+ * is often the case, it is also very common for CDs to have some tracks
+ * with data, and some tracks with audio.  Just because I feel like it,
+ * I declare the following to be the best way to cope.  If the CD has ANY
+ * data tracks on it, it will be returned as a data CD.  If it has any XA
+ * tracks, I will return it as that.  Now I could simplify this interface
+ * by combining these  returns with the above, but this more clearly
+ * demonstrates the problem with the current interface.  Too bad this
+ * wasn't designed to use bitmasks...         -Erik
+ *
+ * Well, now we have the option CDS_MIXED: a mixed-type CD.
+ * User level programmers might feel the ioctl is not very useful.
+ *					---david
+ */
+static int cdrom_ioctl_disc_status(struct cdrom_device_info *cdi)
+{
+	tracktype tracks;
+
+	cdinfo(CD_DO_IOCTL, "entering CDROM_DISC_STATUS\n");
+
+	cdrom_count_tracks(cdi, &tracks);
+	if (tracks.error)
+		return tracks.error;
+
+	/* Policy mode on */
+	if (tracks.audio > 0) {
+		if (!tracks.data && !tracks.cdi && !tracks.xa)
+			return CDS_AUDIO;
+		else
+			return CDS_MIXED;
 	}
 
-	/* use the ioctls that are implemented through the generic_packet()
-	   interface. this may look at bit funny, but if -ENOTTY is
-	   returned that particular ioctl is not implemented and we
-	   let it go through the device specific ones. */
+	if (tracks.cdi > 0)
+		return CDS_XA_2_2;
+	if (tracks.xa > 0)
+		return CDS_XA_2_1;
+	if (tracks.data > 0)
+		return CDS_DATA_1;
+	/* Policy mode off */
+
+	cdinfo(CD_WARNING,"This disc doesn't have any tracks I recognize!\n");
+	return CDS_NO_INFO;
+}
+
+static int cdrom_ioctl_changer_nslots(struct cdrom_device_info *cdi)
+{
+	cdinfo(CD_DO_IOCTL, "entering CDROM_CHANGER_NSLOTS\n");
+	return cdi->capacity;
+}
+
+static int cdrom_ioctl_get_subchnl(struct cdrom_device_info *cdi,
+		void __user *argp)
+{
+	struct cdrom_subchnl q;
+	u8 requested, back;
+	int ret;
+
+	/* cdinfo(CD_DO_IOCTL,"entering CDROMSUBCHNL\n");*/
+
+	if (!CDROM_CAN(CDC_PLAY_AUDIO))
+		return -ENOSYS;
+	if (copy_from_user(&q, argp, sizeof(q)))
+		return -EFAULT;
+
+	requested = q.cdsc_format;
+	if (requested != CDROM_MSF && requested != CDROM_LBA)
+		return -EINVAL;
+	q.cdsc_format = CDROM_MSF;
+
+	ret = cdi->ops->audio_ioctl(cdi, CDROMSUBCHNL, &q);
+	if (ret)
+		return ret;
+
+	back = q.cdsc_format; /* local copy */
+	sanitize_format(&q.cdsc_absaddr, &back, requested);
+	sanitize_format(&q.cdsc_reladdr, &q.cdsc_format, requested);
+
+	if (copy_to_user(argp, &q, sizeof(q)))
+		return -EFAULT;
+	/* cdinfo(CD_DO_IOCTL, "CDROMSUBCHNL successful\n"); */
+	return 0;
+}
+
+static int cdrom_ioctl_read_tochdr(struct cdrom_device_info *cdi,
+		void __user *argp)
+{
+	struct cdrom_tochdr header;
+	int ret;
+
+	/* cdinfo(CD_DO_IOCTL, "entering CDROMREADTOCHDR\n"); */
+
+	if (!CDROM_CAN(CDC_PLAY_AUDIO))
+		return -ENOSYS;
+	if (copy_from_user(&header, argp, sizeof(header)))
+		return -EFAULT;
+
+	ret = cdi->ops->audio_ioctl(cdi, CDROMREADTOCHDR, &header);
+	if (ret)
+		return ret;
+
+	if (copy_to_user(argp, &header, sizeof(header)))
+		return -EFAULT;
+	/* cdinfo(CD_DO_IOCTL, "CDROMREADTOCHDR successful\n"); */
+	return 0;
+}
+
+static int cdrom_ioctl_read_tocentry(struct cdrom_device_info *cdi,
+		void __user *argp)
+{
+	struct cdrom_tocentry entry;
+	u8 requested_format;
+	int ret;
+
+	/* cdinfo(CD_DO_IOCTL, "entering CDROMREADTOCENTRY\n"); */
+
+	if (!CDROM_CAN(CDC_PLAY_AUDIO))
+		return -ENOSYS;
+	if (copy_from_user(&entry, argp, sizeof(entry)))
+		return -EFAULT;
+
+	requested_format = entry.cdte_format;
+	if (requested_format != CDROM_MSF && requested_format != CDROM_LBA)
+		return -EINVAL;
+	/* make interface to low-level uniform */
+	entry.cdte_format = CDROM_MSF;
+	ret = cdi->ops->audio_ioctl(cdi, CDROMREADTOCENTRY, &entry);
+	if (ret)
+		return ret;
+	sanitize_format(&entry.cdte_addr, &entry.cdte_format, requested_format);
+
+	if (copy_to_user(argp, &entry, sizeof(entry)))
+		return -EFAULT;
+	/* cdinfo(CD_DO_IOCTL, "CDROMREADTOCENTRY successful\n"); */
+	return 0;
+}
+
+static int cdrom_ioctl_play_msf(struct cdrom_device_info *cdi,
+		void __user *argp)
+{
+	struct cdrom_msf msf;
+
+	cdinfo(CD_DO_IOCTL, "entering CDROMPLAYMSF\n");
+
+	if (!CDROM_CAN(CDC_PLAY_AUDIO))
+		return -ENOSYS;
+	if (copy_from_user(&msf, argp, sizeof(msf)))
+		return -EFAULT;
+	return cdi->ops->audio_ioctl(cdi, CDROMPLAYMSF, &msf);
+}
+
+static int cdrom_ioctl_play_trkind(struct cdrom_device_info *cdi,
+		void __user *argp)
+{
+	struct cdrom_ti ti;
+	int ret;
+
+	cdinfo(CD_DO_IOCTL, "entering CDROMPLAYTRKIND\n");
+
+	if (!CDROM_CAN(CDC_PLAY_AUDIO))
+		return -ENOSYS;
+	if (copy_from_user(&ti, argp, sizeof(ti)))
+		return -EFAULT;
+
+	ret = check_for_audio_disc(cdi, cdi->ops);
+	if (ret)
+		return ret;
+	return cdi->ops->audio_ioctl(cdi, CDROMPLAYTRKIND, &ti);
+}
+static int cdrom_ioctl_volctrl(struct cdrom_device_info *cdi,
+		void __user *argp)
+{
+	struct cdrom_volctrl volume;
+
+	cdinfo(CD_DO_IOCTL, "entering CDROMVOLCTRL\n");
+
+	if (!CDROM_CAN(CDC_PLAY_AUDIO))
+		return -ENOSYS;
+	if (copy_from_user(&volume, argp, sizeof(volume)))
+		return -EFAULT;
+	return cdi->ops->audio_ioctl(cdi, CDROMVOLCTRL, &volume);
+}
+
+static int cdrom_ioctl_volread(struct cdrom_device_info *cdi,
+		void __user *argp)
+{
+	struct cdrom_volctrl volume;
+	int ret;
+
+	cdinfo(CD_DO_IOCTL, "entering CDROMVOLREAD\n");
+
+	if (!CDROM_CAN(CDC_PLAY_AUDIO))
+		return -ENOSYS;
+
+	ret = cdi->ops->audio_ioctl(cdi, CDROMVOLREAD, &volume);
+	if (ret)
+		return ret;
+
+	if (copy_to_user(argp, &volume, sizeof(volume)))
+		return -EFAULT;
+	return 0;
+}
+
+static int cdrom_ioctl_audioctl(struct cdrom_device_info *cdi,
+		unsigned int cmd)
+{
+	int ret;
+
+	cdinfo(CD_DO_IOCTL, "doing audio ioctl (start/stop/pause/resume)\n");
+
+	if (!CDROM_CAN(CDC_PLAY_AUDIO))
+		return -ENOSYS;
+	ret = check_for_audio_disc(cdi, cdi->ops);
+	if (ret)
+		return ret;
+	return cdi->ops->audio_ioctl(cdi, cmd, NULL);
+}
+
+/*
+ * Just about every imaginable ioctl is supported in the Uniform layer
+ * these days.
+ * ATAPI / SCSI specific code now mainly resides in mmc_ioctl().
+ */
+int cdrom_ioctl(struct file * file, struct cdrom_device_info *cdi,
+		struct inode *ip, unsigned int cmd, unsigned long arg)
+{
+	void __user *argp = (void __user *)arg;
+	int ret;
+
+	/*
+	 * Try the generic SCSI command ioctl's first.
+	 */
+	ret = scsi_cmd_ioctl(file, ip->i_bdev->bd_disk, cmd, argp);
+	if (ret != -ENOTTY)
+		return ret;
+
+	switch (cmd) {
+	case CDROMMULTISESSION:
+		return cdrom_ioctl_multisession(cdi, argp);
+	case CDROMEJECT:
+		return cdrom_ioctl_eject(cdi);
+	case CDROMCLOSETRAY:
+		return cdrom_ioctl_closetray(cdi);
+	case CDROMEJECT_SW:
+		return cdrom_ioctl_eject_sw(cdi, arg);
+	case CDROM_MEDIA_CHANGED:
+		return cdrom_ioctl_media_changed(cdi, arg);
+	case CDROM_SET_OPTIONS:
+		return cdrom_ioctl_set_options(cdi, arg);
+	case CDROM_CLEAR_OPTIONS:
+		return cdrom_ioctl_clear_options(cdi, arg);
+	case CDROM_SELECT_SPEED:
+		return cdrom_ioctl_select_speed(cdi, arg);
+	case CDROM_SELECT_DISC:
+		return cdrom_ioctl_select_disc(cdi, arg);
+	case CDROMRESET:
+		return cdrom_ioctl_reset(cdi, ip->i_bdev);
+	case CDROM_LOCKDOOR:
+		return cdrom_ioctl_lock_door(cdi, arg);
+	case CDROM_DEBUG:
+		return cdrom_ioctl_debug(cdi, arg);
+	case CDROM_GET_CAPABILITY:
+		return cdrom_ioctl_get_capability(cdi);
+	case CDROM_GET_MCN:
+		return cdrom_ioctl_get_mcn(cdi, argp);
+	case CDROM_DRIVE_STATUS:
+		return cdrom_ioctl_drive_status(cdi, arg);
+	case CDROM_DISC_STATUS:
+		return cdrom_ioctl_disc_status(cdi);
+	case CDROM_CHANGER_NSLOTS:
+		return cdrom_ioctl_changer_nslots(cdi);
+	}
+
+	/*
+	 * Use the ioctls that are implemented through the generic_packet()
+	 * interface. this may look at bit funny, but if -ENOTTY is
+	 * returned that particular ioctl is not implemented and we
+	 * let it go through the device specific ones.
+	 */
 	if (CDROM_CAN(CDC_GENERIC_PACKET)) {
 		ret = mmc_ioctl(cdi, cmd, arg);
-		if (ret != -ENOTTY) {
+		if (ret != -ENOTTY)
 			return ret;
-		}
 	}
 
-	/* note: most of the cdinfo() calls are commented out here,
-	   because they fill up the sys log when CD players poll
-	   the drive. */
+	/*
+	 * Note: most of the cdinfo() calls are commented out here,
+	 * because they fill up the sys log when CD players poll
+	 * the drive.
+	 */
 	switch (cmd) {
-	case CDROMSUBCHNL: {
-		struct cdrom_subchnl q;
-		u_char requested, back;
-		if (!CDROM_CAN(CDC_PLAY_AUDIO))
-			return -ENOSYS;
-		/* cdinfo(CD_DO_IOCTL,"entering CDROMSUBCHNL\n");*/ 
-		IOCTL_IN(arg, struct cdrom_subchnl, q);
-		requested = q.cdsc_format;
-		if (!((requested == CDROM_MSF) ||
-		      (requested == CDROM_LBA)))
-			return -EINVAL;
-		q.cdsc_format = CDROM_MSF;
-		if ((ret=cdo->audio_ioctl(cdi, cmd, &q)))
-			return ret;
-		back = q.cdsc_format; /* local copy */
-		sanitize_format(&q.cdsc_absaddr, &back, requested);
-		sanitize_format(&q.cdsc_reladdr, &q.cdsc_format, requested);
-		IOCTL_OUT(arg, struct cdrom_subchnl, q);
-		/* cdinfo(CD_DO_IOCTL, "CDROMSUBCHNL successful\n"); */ 
-		return 0;
-		}
-	case CDROMREADTOCHDR: {
-		struct cdrom_tochdr header;
-		if (!CDROM_CAN(CDC_PLAY_AUDIO))
-			return -ENOSYS;
-		/* cdinfo(CD_DO_IOCTL, "entering CDROMREADTOCHDR\n"); */ 
-		IOCTL_IN(arg, struct cdrom_tochdr, header);
-		if ((ret=cdo->audio_ioctl(cdi, cmd, &header)))
-			return ret;
-		IOCTL_OUT(arg, struct cdrom_tochdr, header);
-		/* cdinfo(CD_DO_IOCTL, "CDROMREADTOCHDR successful\n"); */ 
-		return 0;
-		}
-	case CDROMREADTOCENTRY: {
-		struct cdrom_tocentry entry;
-		u_char requested_format;
-		if (!CDROM_CAN(CDC_PLAY_AUDIO))
-			return -ENOSYS;
-		/* cdinfo(CD_DO_IOCTL, "entering CDROMREADTOCENTRY\n"); */ 
-		IOCTL_IN(arg, struct cdrom_tocentry, entry);
-		requested_format = entry.cdte_format;
-		if (!((requested_format == CDROM_MSF) || 
-			(requested_format == CDROM_LBA)))
-				return -EINVAL;
-		/* make interface to low-level uniform */
-		entry.cdte_format = CDROM_MSF;
-		if ((ret=cdo->audio_ioctl(cdi, cmd, &entry)))
-			return ret;
-		sanitize_format(&entry.cdte_addr,
-		&entry.cdte_format, requested_format);
-		IOCTL_OUT(arg, struct cdrom_tocentry, entry);
-		/* cdinfo(CD_DO_IOCTL, "CDROMREADTOCENTRY successful\n"); */ 
-		return 0;
-		}
-	case CDROMPLAYMSF: {
-		struct cdrom_msf msf;
-		if (!CDROM_CAN(CDC_PLAY_AUDIO))
-			return -ENOSYS;
-		cdinfo(CD_DO_IOCTL, "entering CDROMPLAYMSF\n"); 
-		IOCTL_IN(arg, struct cdrom_msf, msf);
-		return cdo->audio_ioctl(cdi, cmd, &msf);
-		}
-	case CDROMPLAYTRKIND: {
-		struct cdrom_ti ti;
-		if (!CDROM_CAN(CDC_PLAY_AUDIO))
-			return -ENOSYS;
-		cdinfo(CD_DO_IOCTL, "entering CDROMPLAYTRKIND\n"); 
-		IOCTL_IN(arg, struct cdrom_ti, ti);
-		CHECKAUDIO;
-		return cdo->audio_ioctl(cdi, cmd, &ti);
-		}
-	case CDROMVOLCTRL: {
-		struct cdrom_volctrl volume;
-		if (!CDROM_CAN(CDC_PLAY_AUDIO))
-			return -ENOSYS;
-		cdinfo(CD_DO_IOCTL, "entering CDROMVOLCTRL\n"); 
-		IOCTL_IN(arg, struct cdrom_volctrl, volume);
-		return cdo->audio_ioctl(cdi, cmd, &volume);
-		}
-	case CDROMVOLREAD: {
-		struct cdrom_volctrl volume;
-		if (!CDROM_CAN(CDC_PLAY_AUDIO))
-			return -ENOSYS;
-		cdinfo(CD_DO_IOCTL, "entering CDROMVOLREAD\n"); 
-		if ((ret=cdo->audio_ioctl(cdi, cmd, &volume)))
-			return ret;
-		IOCTL_OUT(arg, struct cdrom_volctrl, volume);
-		return 0;
-		}
+	case CDROMSUBCHNL:
+		return cdrom_ioctl_get_subchnl(cdi, argp);
+	case CDROMREADTOCHDR:
+		return cdrom_ioctl_read_tochdr(cdi, argp);
+	case CDROMREADTOCENTRY:
+		return cdrom_ioctl_read_tocentry(cdi, argp);
+	case CDROMPLAYMSF:
+		return cdrom_ioctl_play_msf(cdi, argp);
+	case CDROMPLAYTRKIND:
+		return cdrom_ioctl_play_trkind(cdi, argp);
+	case CDROMVOLCTRL:
+		return cdrom_ioctl_volctrl(cdi, argp);
+	case CDROMVOLREAD:
+		return cdrom_ioctl_volread(cdi, argp);
 	case CDROMSTART:
 	case CDROMSTOP:
 	case CDROMPAUSE:
-	case CDROMRESUME: {
-		if (!CDROM_CAN(CDC_PLAY_AUDIO))
-			return -ENOSYS;
-		cdinfo(CD_DO_IOCTL, "doing audio ioctl (start/stop/pause/resume)\n"); 
-		CHECKAUDIO;
-		return cdo->audio_ioctl(cdi, cmd, NULL);
-		}
-	} /* switch */
+	case CDROMRESUME:
+		return cdrom_ioctl_audioctl(cdi, cmd);
+	}
 
-	/* do the device specific ioctls */
-	if (CDROM_CAN(CDC_IOCTLS))
-		return cdo->dev_ioctl(cdi, cmd, arg);
-	
 	return -ENOSYS;
 }
 
diff --git a/drivers/cdrom/cdu31a.c b/drivers/cdrom/cdu31a.c
index 378e88d2075..72ffd64e8b1 100644
--- a/drivers/cdrom/cdu31a.c
+++ b/drivers/cdrom/cdu31a.c
@@ -2668,7 +2668,7 @@ static int scd_audio_ioctl(struct cdrom_device_info *cdi,
 	return retval;
 }
 
-static int scd_dev_ioctl(struct cdrom_device_info *cdi,
+static int scd_read_audio(struct cdrom_device_info *cdi,
 			 unsigned int cmd, unsigned long arg)
 {
 	void __user *argp = (void __user *)arg;
@@ -2894,11 +2894,10 @@ static struct cdrom_device_ops scd_dops = {
 	.get_mcn		= scd_get_mcn,
 	.reset			= scd_reset,
 	.audio_ioctl		= scd_audio_ioctl,
-	.dev_ioctl		= scd_dev_ioctl,
 	.capability		= CDC_OPEN_TRAY | CDC_CLOSE_TRAY | CDC_LOCK |
 				  CDC_SELECT_SPEED | CDC_MULTI_SESSION |
 				  CDC_MCN | CDC_MEDIA_CHANGED | CDC_PLAY_AUDIO |
-				  CDC_RESET | CDC_IOCTLS | CDC_DRIVE_STATUS,
+				  CDC_RESET | CDC_DRIVE_STATUS,
 	.n_minors		= 1,
 };
 
@@ -2936,6 +2935,9 @@ static int scd_block_ioctl(struct inode *inode, struct file *file,
 		case CDROMCLOSETRAY:
 			retval = scd_tray_move(&scd_info, 0);
 			break;
+		case CDROMREADAUDIO:
+			retval = scd_read_audio(&scd_info, CDROMREADAUDIO, arg);
+			break;
 		default:
 			retval = cdrom_ioctl(file, &scd_info, inode, cmd, arg);
 	}
diff --git a/drivers/cdrom/cm206.c b/drivers/cdrom/cm206.c
index ce127f7ec0f..fad27a87ce3 100644
--- a/drivers/cdrom/cm206.c
+++ b/drivers/cdrom/cm206.c
@@ -1157,32 +1157,6 @@ static int cm206_audio_ioctl(struct cdrom_device_info *cdi, unsigned int cmd,
 	}
 }
 
-/* Ioctl. These ioctls are specific to the cm206 driver. I have made
-   some driver statistics accessible through ioctl calls.
- */
-
-static int cm206_ioctl(struct cdrom_device_info *cdi, unsigned int cmd,
-		       unsigned long arg)
-{
-	switch (cmd) {
-#ifdef STATISTICS
-	case CM206CTL_GET_STAT:
-		if (arg >= NR_STATS)
-			return -EINVAL;
-		else
-			return cd->stats[arg];
-	case CM206CTL_GET_LAST_STAT:
-		if (arg >= NR_STATS)
-			return -EINVAL;
-		else
-			return cd->last_stat[arg];
-#endif
-	default:
-		debug(("Unknown ioctl call 0x%x\n", cmd));
-		return -EINVAL;
-	}
-}
-
 static int cm206_media_changed(struct cdrom_device_info *cdi, int disc_nr)
 {
 	if (cd != NULL) {
@@ -1321,11 +1295,10 @@ static struct cdrom_device_ops cm206_dops = {
 	.get_mcn		= cm206_get_upc,
 	.reset			= cm206_reset,
 	.audio_ioctl		= cm206_audio_ioctl,
-	.dev_ioctl		= cm206_ioctl,
 	.capability		= CDC_CLOSE_TRAY | CDC_OPEN_TRAY | CDC_LOCK |
 				  CDC_MULTI_SESSION | CDC_MEDIA_CHANGED |
 				  CDC_MCN | CDC_PLAY_AUDIO | CDC_SELECT_SPEED |
-				  CDC_IOCTLS | CDC_DRIVE_STATUS,
+				  CDC_DRIVE_STATUS,
 	.n_minors		= 1,
 };
 
@@ -1350,6 +1323,21 @@ static int cm206_block_release(struct inode *inode, struct file *file)
 static int cm206_block_ioctl(struct inode *inode, struct file *file,
 				unsigned cmd, unsigned long arg)
 {
+	switch (cmd) {
+#ifdef STATISTICS
+	case CM206CTL_GET_STAT:
+		if (arg >= NR_STATS)
+			return -EINVAL;
+		return cd->stats[arg];
+	case CM206CTL_GET_LAST_STAT:
+		if (arg >= NR_STATS)
+			return -EINVAL;
+		return cd->last_stat[arg];
+#endif
+	default:
+		break;
+	}
+
 	return cdrom_ioctl(file, &cm206_info, inode, cmd, arg);
 }
 
diff --git a/drivers/cdrom/sbpcd.c b/drivers/cdrom/sbpcd.c
index 466e9c2974b..4760f515f59 100644
--- a/drivers/cdrom/sbpcd.c
+++ b/drivers/cdrom/sbpcd.c
@@ -4160,332 +4160,6 @@ static int sbpcd_get_last_session(struct cdrom_device_info *cdi, struct cdrom_mu
 	return  0;
 }
 
-/*==========================================================================*/
-/*==========================================================================*/
-/*
- * ioctl support
- */
-static int sbpcd_dev_ioctl(struct cdrom_device_info *cdi, u_int cmd,
-		      u_long arg)
-{
-	struct sbpcd_drive *p = cdi->handle;
-	int i;
-	
-	msg(DBG_IO2,"ioctl(%s, 0x%08lX, 0x%08lX)\n", cdi->name, cmd, arg);
-	if (p->drv_id==-1) {
-		msg(DBG_INF, "ioctl: bad device: %s\n", cdi->name);
-		return (-ENXIO);             /* no such drive */
-	}
-	down(&ioctl_read_sem);
-	if (p != current_drive)
-		switch_drive(p);
-	
-	msg(DBG_IO2,"ioctl: device %s, request %04X\n",cdi->name,cmd);
-	switch (cmd) 		/* Sun-compatible */
-	{
-	case DDIOCSDBG:		/* DDI Debug */
-		if (!capable(CAP_SYS_ADMIN)) RETURN_UP(-EPERM);
-		i=sbpcd_dbg_ioctl(arg,1);
-		RETURN_UP(i);
-	case CDROMRESET:      /* hard reset the drive */
-		msg(DBG_IOC,"ioctl: CDROMRESET entered.\n");
-		i=DriveReset();
-		current_drive->audio_state=0;
-		RETURN_UP(i);
-		
-	case CDROMREADMODE1:
-		msg(DBG_IOC,"ioctl: CDROMREADMODE1 requested.\n");
-#ifdef SAFE_MIXED
-		if (current_drive->has_data>1) RETURN_UP(-EBUSY);
-#endif /* SAFE_MIXED */
-		cc_ModeSelect(CD_FRAMESIZE);
-		cc_ModeSense();
-		current_drive->mode=READ_M1;
-		RETURN_UP(0);
-		
-	case CDROMREADMODE2: /* not usable at the moment */
-		msg(DBG_IOC,"ioctl: CDROMREADMODE2 requested.\n");
-#ifdef SAFE_MIXED
-		if (current_drive->has_data>1) RETURN_UP(-EBUSY);
-#endif /* SAFE_MIXED */
-		cc_ModeSelect(CD_FRAMESIZE_RAW1);
-		cc_ModeSense();
-		current_drive->mode=READ_M2;
-		RETURN_UP(0);
-		
-	case CDROMAUDIOBUFSIZ: /* configure the audio buffer size */
-		msg(DBG_IOC,"ioctl: CDROMAUDIOBUFSIZ entered.\n");
-		if (current_drive->sbp_audsiz>0)
-			vfree(current_drive->aud_buf);
-		current_drive->aud_buf=NULL;
-		current_drive->sbp_audsiz=arg;
-		
-		if (current_drive->sbp_audsiz>16)
-		{
-			current_drive->sbp_audsiz = 0;
-			RETURN_UP(current_drive->sbp_audsiz);
-		}
-	
-		if (current_drive->sbp_audsiz>0)
-		{
-			current_drive->aud_buf=(u_char *) vmalloc(current_drive->sbp_audsiz*CD_FRAMESIZE_RAW);
-			if (current_drive->aud_buf==NULL)
-			{
-				msg(DBG_INF,"audio buffer (%d frames) not available.\n",current_drive->sbp_audsiz);
-				current_drive->sbp_audsiz=0;
-			}
-			else msg(DBG_INF,"audio buffer size: %d frames.\n",current_drive->sbp_audsiz);
-		}
-		RETURN_UP(current_drive->sbp_audsiz);
-
-	case CDROMREADAUDIO:
-	{ /* start of CDROMREADAUDIO */
-		int i=0, j=0, frame, block=0;
-		u_int try=0;
-		u_long timeout;
-		u_char *p;
-		u_int data_tries = 0;
-		u_int data_waits = 0;
-		u_int data_retrying = 0;
-		int status_tries;
-		int error_flag;
-		
-		msg(DBG_IOC,"ioctl: CDROMREADAUDIO entered.\n");
-		if (fam0_drive) RETURN_UP(-EINVAL);
-		if (famL_drive) RETURN_UP(-EINVAL);
-		if (famV_drive) RETURN_UP(-EINVAL);
-		if (famT_drive) RETURN_UP(-EINVAL);
-#ifdef SAFE_MIXED
-		if (current_drive->has_data>1) RETURN_UP(-EBUSY);
-#endif /* SAFE_MIXED */ 
-		if (current_drive->aud_buf==NULL) RETURN_UP(-EINVAL);
-		if (copy_from_user(&read_audio, (void __user *)arg,
-				   sizeof(struct cdrom_read_audio)))
-			RETURN_UP(-EFAULT);
-		if (read_audio.nframes < 0 || read_audio.nframes>current_drive->sbp_audsiz) RETURN_UP(-EINVAL);
-		if (!access_ok(VERIFY_WRITE, read_audio.buf,
-			      read_audio.nframes*CD_FRAMESIZE_RAW))
-                	RETURN_UP(-EFAULT);
-		
-		if (read_audio.addr_format==CDROM_MSF) /* MSF-bin specification of where to start */
-			block=msf2lba(&read_audio.addr.msf.minute);
-		else if (read_audio.addr_format==CDROM_LBA) /* lba specification of where to start */
-			block=read_audio.addr.lba;
-		else RETURN_UP(-EINVAL);
-#if 000
-		i=cc_SetSpeed(speed_150,0,0);
-		if (i) msg(DBG_AUD,"read_audio: SetSpeed error %d\n", i);
-#endif
-		msg(DBG_AUD,"read_audio: lba: %d, msf: %06X\n",
-		    block, blk2msf(block));
-		msg(DBG_AUD,"read_audio: before cc_ReadStatus.\n");
-#if OLD_BUSY
-		while (busy_data) sbp_sleep(HZ/10); /* wait a bit */
-		busy_audio=1;
-#endif /* OLD_BUSY */ 
-		error_flag=0;
-		for (data_tries=5; data_tries>0; data_tries--)
-		{
-			msg(DBG_AUD,"data_tries=%d ...\n", data_tries);
-			current_drive->mode=READ_AU;
-			cc_ModeSelect(CD_FRAMESIZE_RAW);
-			cc_ModeSense();
-			for (status_tries=3; status_tries > 0; status_tries--)
-			{
-				flags_cmd_out |= f_respo3;
-				cc_ReadStatus();
-				if (sbp_status() != 0) break;
-				if (st_check) cc_ReadError();
-				sbp_sleep(1);    /* wait a bit, try again */
-			}
-			if (status_tries == 0)
-			{
-				msg(DBG_AUD,"read_audio: sbp_status: failed after 3 tries in line %d.\n", __LINE__);
-				continue;
-			}
-			msg(DBG_AUD,"read_audio: sbp_status: ok.\n");
-			
-			flags_cmd_out = f_putcmd | f_respo2 | f_ResponseStatus | f_obey_p_check;
-			if (fam0L_drive)
-			{
-				flags_cmd_out |= f_lopsta | f_getsta | f_bit1;
-				cmd_type=READ_M2;
-				drvcmd[0]=CMD0_READ_XA; /* "read XA frames", old drives */
-				drvcmd[1]=(block>>16)&0x000000ff;
-				drvcmd[2]=(block>>8)&0x000000ff;
-				drvcmd[3]=block&0x000000ff;
-				drvcmd[4]=0;
-				drvcmd[5]=read_audio.nframes; /* # of frames */
-				drvcmd[6]=0;
-			}
-			else if (fam1_drive)
-			{
-				drvcmd[0]=CMD1_READ; /* "read frames", new drives */
-				lba2msf(block,&drvcmd[1]); /* msf-bin format required */
-				drvcmd[4]=0;
-				drvcmd[5]=0;
-				drvcmd[6]=read_audio.nframes; /* # of frames */
-			}
-			else if (fam2_drive)
-			{
-				drvcmd[0]=CMD2_READ_XA2;
-				lba2msf(block,&drvcmd[1]); /* msf-bin format required */
-				drvcmd[4]=0;
-				drvcmd[5]=read_audio.nframes; /* # of frames */
-				drvcmd[6]=0x11; /* raw mode */
-			}
-			else if (famT_drive) /* CD-55A: not tested yet */
-			{
-			}
-			msg(DBG_AUD,"read_audio: before giving \"read\" command.\n");
-			flags_cmd_out=f_putcmd;
-			response_count=0;
-			i=cmd_out();
-			if (i<0) msg(DBG_INF,"error giving READ AUDIO command: %0d\n", i);
-			sbp_sleep(0);
-			msg(DBG_AUD,"read_audio: after giving \"read\" command.\n");
-			for (frame=1;frame<2 && !error_flag; frame++)
-			{
-				try=maxtim_data;
-				for (timeout=jiffies+9*HZ; ; )
-				{
-					for ( ; try!=0;try--)
-					{
-						j=inb(CDi_status);
-						if (!(j&s_not_data_ready)) break;
-						if (!(j&s_not_result_ready)) break;
-						if (fam0L_drive) if (j&s_attention) break;
-					}
-					if (try != 0 || time_after_eq(jiffies, timeout)) break;
-					if (data_retrying == 0) data_waits++;
-					data_retrying = 1;
-					sbp_sleep(1);
-					try = 1;
-				}
-				if (try==0)
-				{
-					msg(DBG_INF,"read_audio: sbp_data: CDi_status timeout.\n");
-					error_flag++;
-					break;
-				}
-				msg(DBG_AUD,"read_audio: sbp_data: CDi_status ok.\n");
-				if (j&s_not_data_ready)
-				{
-					msg(DBG_INF, "read_audio: sbp_data: DATA_READY timeout.\n");
-					error_flag++;
-					break;
-				}
-				msg(DBG_AUD,"read_audio: before reading data.\n");
-				error_flag=0;
-				p = current_drive->aud_buf;
-				if (sbpro_type==1) OUT(CDo_sel_i_d,1);
-				if (do_16bit)
-				{
-					u_short *p2 = (u_short *) p;
-
-					for (; (u_char *) p2 < current_drive->aud_buf + read_audio.nframes*CD_FRAMESIZE_RAW;)
-				  	{
-						if ((inb_p(CDi_status)&s_not_data_ready)) continue;
-
-						/* get one sample */
-						*p2++ = inw_p(CDi_data);
-						*p2++ = inw_p(CDi_data);
-					}
-				} else {
-					for (; p < current_drive->aud_buf + read_audio.nframes*CD_FRAMESIZE_RAW;)
-				  	{
-						if ((inb_p(CDi_status)&s_not_data_ready)) continue;
-
-						/* get one sample */
-						*p++ = inb_p(CDi_data);
-						*p++ = inb_p(CDi_data);
-						*p++ = inb_p(CDi_data);
-						*p++ = inb_p(CDi_data);
-					}
-				}
-				if (sbpro_type==1) OUT(CDo_sel_i_d,0);
-				data_retrying = 0;
-			}
-			msg(DBG_AUD,"read_audio: after reading data.\n");
-			if (error_flag)    /* must have been spurious D_RDY or (ATTN&&!D_RDY) */
-			{
-				msg(DBG_AUD,"read_audio: read aborted by drive\n");
-#if 0000
-				i=cc_DriveReset();                /* ugly fix to prevent a hang */
-#else
-				i=cc_ReadError();
-#endif
-				continue;
-			}
-			if (fam0L_drive)
-			{
-				i=maxtim_data;
-				for (timeout=jiffies+9*HZ; time_before(jiffies, timeout); timeout--)
-				{
-					for ( ;i!=0;i--)
-					{
-						j=inb(CDi_status);
-						if (!(j&s_not_data_ready)) break;
-						if (!(j&s_not_result_ready)) break;
-						if (j&s_attention) break;
-					}
-					if (i != 0 || time_after_eq(jiffies, timeout)) break;
-					sbp_sleep(0);
-					i = 1;
-				}
-				if (i==0) msg(DBG_AUD,"read_audio: STATUS TIMEOUT AFTER READ");
-				if (!(j&s_attention))
-				{
-					msg(DBG_AUD,"read_audio: sbp_data: timeout waiting DRV_ATTN - retrying\n");
-					i=cc_DriveReset();  /* ugly fix to prevent a hang */
-					continue;
-				}
-			}
-			do
-			{
-				if (fam0L_drive) cc_ReadStatus();
-				i=ResponseStatus();  /* builds status_bits, returns orig. status (old) or faked p_success (new) */
-				if (i<0) { msg(DBG_AUD,
-					       "read_audio: cc_ReadStatus error after read: %02X\n",
-					       current_drive->status_bits);
-					   continue; /* FIXME */
-				   }
-			}
-			while ((fam0L_drive)&&(!st_check)&&(!(i&p_success)));
-			if (st_check)
-			{
-				i=cc_ReadError();
-				msg(DBG_AUD,"read_audio: cc_ReadError was necessary after read: %02X\n",i);
-				continue;
-			}
-			if (copy_to_user(read_audio.buf,
-					 current_drive->aud_buf,
-					 read_audio.nframes * CD_FRAMESIZE_RAW))
-				RETURN_UP(-EFAULT);
-			msg(DBG_AUD,"read_audio: copy_to_user done.\n");
-			break;
-		}
-		cc_ModeSelect(CD_FRAMESIZE);
-		cc_ModeSense();
-		current_drive->mode=READ_M1;
-#if OLD_BUSY
-		busy_audio=0;
-#endif /* OLD_BUSY */ 
-		if (data_tries == 0)
-		{
-			msg(DBG_AUD,"read_audio: failed after 5 tries in line %d.\n", __LINE__);
-			RETURN_UP(-EIO);
-		}
-		msg(DBG_AUD,"read_audio: successful return.\n");
-		RETURN_UP(0);
-	} /* end of CDROMREADAUDIO */
-		
-	default:
-		msg(DBG_IOC,"ioctl: unknown function request %04X\n", cmd);
-		RETURN_UP(-EINVAL);
-	} /* end switch(cmd) */
-}
-
 static int sbpcd_audio_ioctl(struct cdrom_device_info *cdi, u_int cmd,
 		       void * arg)
 {
@@ -4530,7 +4204,7 @@ static int sbpcd_audio_ioctl(struct cdrom_device_info *cdi, u_int cmd,
 		default:
 			RETURN_UP(-EINVAL);
 		}
-		
+
 	case CDROMRESUME: /* resume paused audio play */
 		msg(DBG_IOC,"ioctl: CDROMRESUME entered.\n");
 		/* resume playing audio tracks when a previous PLAY AUDIO call has  */
@@ -4544,12 +4218,12 @@ static int sbpcd_audio_ioctl(struct cdrom_device_info *cdi, u_int cmd,
 		if (i<0) RETURN_UP(-EIO);
 		current_drive->audio_state=audio_playing;
 		RETURN_UP(0);
-		
+
 	case CDROMPLAYMSF:
 		msg(DBG_IOC,"ioctl: CDROMPLAYMSF entered.\n");
 #ifdef SAFE_MIXED
 		if (current_drive->has_data>1) RETURN_UP(-EBUSY);
-#endif /* SAFE_MIXED */ 
+#endif /* SAFE_MIXED */
 		if (current_drive->audio_state==audio_playing)
 		{
 			i=cc_Pause_Resume(1);
@@ -4584,7 +4258,7 @@ static int sbpcd_audio_ioctl(struct cdrom_device_info *cdi, u_int cmd,
 		msg(DBG_IOC,"ioctl: CDROMPLAYTRKIND entered.\n");
 #ifdef SAFE_MIXED
 		if (current_drive->has_data>1) RETURN_UP(-EBUSY);
-#endif /* SAFE_MIXED */ 
+#endif /* SAFE_MIXED */
 		if (current_drive->audio_state==audio_playing)
 		{
 			msg(DBG_IOX,"CDROMPLAYTRKIND: already audio_playing.\n");
@@ -4654,13 +4328,13 @@ static int sbpcd_audio_ioctl(struct cdrom_device_info *cdi, u_int cmd,
 		cc_DriveReset();
 #endif
 		RETURN_UP(i);
-		
+
 	case CDROMSTART:  /* Spin up the drive */
 		msg(DBG_IOC,"ioctl: CDROMSTART entered.\n");
 		cc_SpinUp();
 		current_drive->audio_state=0;
 		RETURN_UP(0);
-		
+
 	case CDROMVOLCTRL:   /* Volume control */
 		msg(DBG_IOC,"ioctl: CDROMVOLCTRL entered.\n");
 		memcpy(&volctrl,(char *) arg,sizeof(volctrl));
@@ -4670,7 +4344,7 @@ static int sbpcd_audio_ioctl(struct cdrom_device_info *cdi, u_int cmd,
 		current_drive->vol_ctrl1=volctrl.channel1;
 		i=cc_SetVolume();
 		RETURN_UP(0);
-		
+
 	case CDROMVOLREAD:   /* read Volume settings from drive */
 		msg(DBG_IOC,"ioctl: CDROMVOLREAD entered.\n");
 		st=cc_GetVolume();
@@ -4694,7 +4368,7 @@ static int sbpcd_audio_ioctl(struct cdrom_device_info *cdi, u_int cmd,
 		if (i<0) {
 			j=cc_ReadError(); /* clear out error status from drive */
 			current_drive->audio_state=CDROM_AUDIO_NO_STATUS;
-			/* get and set the disk state here, 
+			/* get and set the disk state here,
 			probably not the right place, but who cares!
 			It makes it work properly! --AJK */
 			if (current_drive->CD_changed==0xFF) {
@@ -4715,8 +4389,8 @@ static int sbpcd_audio_ioctl(struct cdrom_device_info *cdi, u_int cmd,
 			}
 		}
 		memcpy(&SC, (void *) arg, sizeof(struct cdrom_subchnl));
-		/* 
-			This virtual crap is very bogus! 
+		/*
+			This virtual crap is very bogus!
 			It doesn't detect when the cd is done playing audio!
 			Lets do this right with proper hardware register reading!
 		*/
@@ -4775,7 +4449,7 @@ static int sbpcd_audio_ioctl(struct cdrom_device_info *cdi, u_int cmd,
 		    SC.cdsc_trk,SC.cdsc_ind,
 		    SC.cdsc_absaddr,SC.cdsc_reladdr);
 		RETURN_UP(0);
-		
+
 	default:
 		msg(DBG_IOC,"ioctl: unknown function request %04X\n", cmd);
 		RETURN_UP(-EINVAL);
@@ -4788,7 +4462,7 @@ static int sbpcd_audio_ioctl(struct cdrom_device_info *cdi, u_int cmd,
 static void sbp_transfer(struct request *req)
 {
 	long offs;
-	
+
 	while ( (req->nr_sectors > 0) &&
 	       (req->sector/4 >= current_drive->sbp_first_frame) &&
 	       (req->sector/4 <= current_drive->sbp_last_frame) )
@@ -4807,11 +4481,11 @@ static void sbp_transfer(struct request *req)
  *
  *  This is a kludge so we don't need to modify end_request.
  *  We put the req we take out after INIT_REQUEST in the requests list,
- *  so that end_request will discard it. 
+ *  so that end_request will discard it.
  *
  *  The bug could be present in other block devices, perhaps we
  *  should modify INIT_REQUEST and end_request instead, and
- *  change every block device.. 
+ *  change every block device..
  *
  *  Could be a race here?? Could e.g. a timer interrupt schedule() us?
  *  If so, we should copy end_request here, and do it right.. (or
@@ -4883,19 +4557,19 @@ static void do_sbpcd_request(request_queue_t * q)
 	while (busy_audio) sbp_sleep(HZ); /* wait a bit */
 	busy_data=1;
 #endif /* OLD_BUSY */
-	
+
 	if (p->audio_state==audio_playing) goto err_done;
 	if (p != current_drive)
 		switch_drive(p);
 
 	block = req->sector; /* always numbered as 512-byte-pieces */
 	nsect = req->nr_sectors; /* always counted as 512-byte-pieces */
-	
+
 	msg(DBG_BSZ,"read sector %d (%d sectors)\n", block, nsect);
 #if 0
 	msg(DBG_MUL,"read LBA %d\n", block/4);
 #endif
-	
+
 	sbp_transfer(req);
 	/* if we satisfied the request from the buffer, we're done. */
 	if (req->nr_sectors == 0)
@@ -4914,10 +4588,10 @@ static void do_sbpcd_request(request_queue_t * q)
 	i=prepare(0,0); /* at moment not really a hassle check, but ... */
 	if (i!=0)
 		msg(DBG_INF,"\"prepare\" tells error %d -- ignored\n", i);
-#endif /* FUTURE */ 
-	
+#endif /* FUTURE */
+
 	if (!st_spinning) cc_SpinUp();
-	
+
 	for (data_tries=n_retries; data_tries > 0; data_tries--)
 	{
 		for (status_tries=3; status_tries > 0; status_tries--)
@@ -4940,7 +4614,7 @@ static void do_sbpcd_request(request_queue_t * q)
 		{
 #ifdef SAFE_MIXED
 			current_drive->has_data=2; /* is really a data disk */
-#endif /* SAFE_MIXED */ 
+#endif /* SAFE_MIXED */
 #ifdef DEBUG_GTL
 			printk(" do_sbpcd_request[%do](%p:%ld+%ld) end 3, Time:%li\n",
 				xnr, req, req->sector, req->nr_sectors, jiffies);
@@ -4951,7 +4625,7 @@ static void do_sbpcd_request(request_queue_t * q)
 			goto request_loop;
 		}
 	}
-	
+
  err_done:
 #if OLD_BUSY
 	busy_data=0;
@@ -4976,7 +4650,7 @@ static void sbp_read_cmd(struct request *req)
 
 	int i;
 	int block;
-	
+
 	current_drive->sbp_first_frame=current_drive->sbp_last_frame=-1;      /* purge buffer */
 	current_drive->sbp_current = 0;
 	block=req->sector/4;
@@ -4993,7 +4667,7 @@ static void sbp_read_cmd(struct request *req)
 			current_drive->sbp_read_frames=1;
 		}
 	}
-	
+
 	flags_cmd_out = f_putcmd | f_respo2 | f_ResponseStatus | f_obey_p_check;
 	clr_cmdbuf();
 	if (famV_drive)
@@ -5092,7 +4766,7 @@ static int sbp_data(struct request *req)
 	int success;
 	int wait;
 	int duration;
-	
+
 	error_flag=0;
 	success=0;
 #if LONG_TIMING
@@ -5105,12 +4779,12 @@ static int sbp_data(struct request *req)
 	for (frame=0;frame<current_drive->sbp_read_frames&&!error_flag; frame++)
 	{
 		SBPCD_CLI;
-		
+
 		del_timer(&data_timer);
 		data_timer.expires=jiffies+max_latency;
 		timed_out_data=0;
 		add_timer(&data_timer);
-		while (!timed_out_data) 
+		while (!timed_out_data)
 		{
 			if (current_drive->f_multisession) try=maxtim_data*4;
 			else try=maxtim_data;
@@ -5207,9 +4881,9 @@ static int sbp_data(struct request *req)
                         else
                           {
                             sbp_sleep(1);
-                            OUT(CDo_sel_i_d,0); 
+                            OUT(CDo_sel_i_d,0);
                             i=inb(CDi_status);
-                          } 
+                          }
 			if (!(i&s_not_data_ready))
 			{
 				OUT(CDo_sel_i_d,1);
@@ -5311,7 +4985,7 @@ static int sbp_data(struct request *req)
 		}
 		SBPCD_STI;
 	}
-	
+
 #if 0
 	if (!success)
 #endif
@@ -5370,7 +5044,326 @@ static int sbpcd_block_ioctl(struct inode *inode, struct file *file,
 				unsigned cmd, unsigned long arg)
 {
 	struct sbpcd_drive *p = inode->i_bdev->bd_disk->private_data;
-	return cdrom_ioctl(file, p->sbpcd_infop, inode, cmd, arg);
+	struct cdrom_device_info *cdi = p->sbpcd_infop;
+	int ret, i;
+
+	ret = cdrom_ioctl(file, p->sbpcd_infop, inode, cmd, arg);
+	if (ret != -ENOSYS)
+		return ret;
+
+	msg(DBG_IO2,"ioctl(%s, 0x%08lX, 0x%08lX)\n", cdi->name, cmd, arg);
+	if (p->drv_id==-1) {
+		msg(DBG_INF, "ioctl: bad device: %s\n", cdi->name);
+		return (-ENXIO);             /* no such drive */
+	}
+	down(&ioctl_read_sem);
+	if (p != current_drive)
+		switch_drive(p);
+	
+	msg(DBG_IO2,"ioctl: device %s, request %04X\n",cdi->name,cmd);
+	switch (cmd) 		/* Sun-compatible */
+	{
+	case DDIOCSDBG:		/* DDI Debug */
+		if (!capable(CAP_SYS_ADMIN)) RETURN_UP(-EPERM);
+		i=sbpcd_dbg_ioctl(arg,1);
+		RETURN_UP(i);
+	case CDROMRESET:      /* hard reset the drive */
+		msg(DBG_IOC,"ioctl: CDROMRESET entered.\n");
+		i=DriveReset();
+		current_drive->audio_state=0;
+		RETURN_UP(i);
+
+	case CDROMREADMODE1:
+		msg(DBG_IOC,"ioctl: CDROMREADMODE1 requested.\n");
+#ifdef SAFE_MIXED
+		if (current_drive->has_data>1) RETURN_UP(-EBUSY);
+#endif /* SAFE_MIXED */
+		cc_ModeSelect(CD_FRAMESIZE);
+		cc_ModeSense();
+		current_drive->mode=READ_M1;
+		RETURN_UP(0);
+
+	case CDROMREADMODE2: /* not usable at the moment */
+		msg(DBG_IOC,"ioctl: CDROMREADMODE2 requested.\n");
+#ifdef SAFE_MIXED
+		if (current_drive->has_data>1) RETURN_UP(-EBUSY);
+#endif /* SAFE_MIXED */
+		cc_ModeSelect(CD_FRAMESIZE_RAW1);
+		cc_ModeSense();
+		current_drive->mode=READ_M2;
+		RETURN_UP(0);
+
+	case CDROMAUDIOBUFSIZ: /* configure the audio buffer size */
+		msg(DBG_IOC,"ioctl: CDROMAUDIOBUFSIZ entered.\n");
+		if (current_drive->sbp_audsiz>0)
+			vfree(current_drive->aud_buf);
+		current_drive->aud_buf=NULL;
+		current_drive->sbp_audsiz=arg;
+
+		if (current_drive->sbp_audsiz>16)
+		{
+			current_drive->sbp_audsiz = 0;
+			RETURN_UP(current_drive->sbp_audsiz);
+		}
+	
+		if (current_drive->sbp_audsiz>0)
+		{
+			current_drive->aud_buf=(u_char *) vmalloc(current_drive->sbp_audsiz*CD_FRAMESIZE_RAW);
+			if (current_drive->aud_buf==NULL)
+			{
+				msg(DBG_INF,"audio buffer (%d frames) not available.\n",current_drive->sbp_audsiz);
+				current_drive->sbp_audsiz=0;
+			}
+			else msg(DBG_INF,"audio buffer size: %d frames.\n",current_drive->sbp_audsiz);
+		}
+		RETURN_UP(current_drive->sbp_audsiz);
+
+	case CDROMREADAUDIO:
+	{ /* start of CDROMREADAUDIO */
+		int i=0, j=0, frame, block=0;
+		u_int try=0;
+		u_long timeout;
+		u_char *p;
+		u_int data_tries = 0;
+		u_int data_waits = 0;
+		u_int data_retrying = 0;
+		int status_tries;
+		int error_flag;
+
+		msg(DBG_IOC,"ioctl: CDROMREADAUDIO entered.\n");
+		if (fam0_drive) RETURN_UP(-EINVAL);
+		if (famL_drive) RETURN_UP(-EINVAL);
+		if (famV_drive) RETURN_UP(-EINVAL);
+		if (famT_drive) RETURN_UP(-EINVAL);
+#ifdef SAFE_MIXED
+		if (current_drive->has_data>1) RETURN_UP(-EBUSY);
+#endif /* SAFE_MIXED */
+		if (current_drive->aud_buf==NULL) RETURN_UP(-EINVAL);
+		if (copy_from_user(&read_audio, (void __user *)arg,
+				   sizeof(struct cdrom_read_audio)))
+			RETURN_UP(-EFAULT);
+		if (read_audio.nframes < 0 || read_audio.nframes>current_drive->sbp_audsiz) RETURN_UP(-EINVAL);
+		if (!access_ok(VERIFY_WRITE, read_audio.buf,
+			      read_audio.nframes*CD_FRAMESIZE_RAW))
+                	RETURN_UP(-EFAULT);
+
+		if (read_audio.addr_format==CDROM_MSF) /* MSF-bin specification of where to start */
+			block=msf2lba(&read_audio.addr.msf.minute);
+		else if (read_audio.addr_format==CDROM_LBA) /* lba specification of where to start */
+			block=read_audio.addr.lba;
+		else RETURN_UP(-EINVAL);
+#if 000
+		i=cc_SetSpeed(speed_150,0,0);
+		if (i) msg(DBG_AUD,"read_audio: SetSpeed error %d\n", i);
+#endif
+		msg(DBG_AUD,"read_audio: lba: %d, msf: %06X\n",
+		    block, blk2msf(block));
+		msg(DBG_AUD,"read_audio: before cc_ReadStatus.\n");
+#if OLD_BUSY
+		while (busy_data) sbp_sleep(HZ/10); /* wait a bit */
+		busy_audio=1;
+#endif /* OLD_BUSY */
+		error_flag=0;
+		for (data_tries=5; data_tries>0; data_tries--)
+		{
+			msg(DBG_AUD,"data_tries=%d ...\n", data_tries);
+			current_drive->mode=READ_AU;
+			cc_ModeSelect(CD_FRAMESIZE_RAW);
+			cc_ModeSense();
+			for (status_tries=3; status_tries > 0; status_tries--)
+			{
+				flags_cmd_out |= f_respo3;
+				cc_ReadStatus();
+				if (sbp_status() != 0) break;
+				if (st_check) cc_ReadError();
+				sbp_sleep(1);    /* wait a bit, try again */
+			}
+			if (status_tries == 0)
+			{
+				msg(DBG_AUD,"read_audio: sbp_status: failed after 3 tries in line %d.\n", __LINE__);
+				continue;
+			}
+			msg(DBG_AUD,"read_audio: sbp_status: ok.\n");
+
+			flags_cmd_out = f_putcmd | f_respo2 | f_ResponseStatus | f_obey_p_check;
+			if (fam0L_drive)
+			{
+				flags_cmd_out |= f_lopsta | f_getsta | f_bit1;
+				cmd_type=READ_M2;
+				drvcmd[0]=CMD0_READ_XA; /* "read XA frames", old drives */
+				drvcmd[1]=(block>>16)&0x000000ff;
+				drvcmd[2]=(block>>8)&0x000000ff;
+				drvcmd[3]=block&0x000000ff;
+				drvcmd[4]=0;
+				drvcmd[5]=read_audio.nframes; /* # of frames */
+				drvcmd[6]=0;
+			}
+			else if (fam1_drive)
+			{
+				drvcmd[0]=CMD1_READ; /* "read frames", new drives */
+				lba2msf(block,&drvcmd[1]); /* msf-bin format required */
+				drvcmd[4]=0;
+				drvcmd[5]=0;
+				drvcmd[6]=read_audio.nframes; /* # of frames */
+			}
+			else if (fam2_drive)
+			{
+				drvcmd[0]=CMD2_READ_XA2;
+				lba2msf(block,&drvcmd[1]); /* msf-bin format required */
+				drvcmd[4]=0;
+				drvcmd[5]=read_audio.nframes; /* # of frames */
+				drvcmd[6]=0x11; /* raw mode */
+			}
+			else if (famT_drive) /* CD-55A: not tested yet */
+			{
+			}
+			msg(DBG_AUD,"read_audio: before giving \"read\" command.\n");
+			flags_cmd_out=f_putcmd;
+			response_count=0;
+			i=cmd_out();
+			if (i<0) msg(DBG_INF,"error giving READ AUDIO command: %0d\n", i);
+			sbp_sleep(0);
+			msg(DBG_AUD,"read_audio: after giving \"read\" command.\n");
+			for (frame=1;frame<2 && !error_flag; frame++)
+			{
+				try=maxtim_data;
+				for (timeout=jiffies+9*HZ; ; )
+				{
+					for ( ; try!=0;try--)
+					{
+						j=inb(CDi_status);
+						if (!(j&s_not_data_ready)) break;
+						if (!(j&s_not_result_ready)) break;
+						if (fam0L_drive) if (j&s_attention) break;
+					}
+					if (try != 0 || time_after_eq(jiffies, timeout)) break;
+					if (data_retrying == 0) data_waits++;
+					data_retrying = 1;
+					sbp_sleep(1);
+					try = 1;
+				}
+				if (try==0)
+				{
+					msg(DBG_INF,"read_audio: sbp_data: CDi_status timeout.\n");
+					error_flag++;
+					break;
+				}
+				msg(DBG_AUD,"read_audio: sbp_data: CDi_status ok.\n");
+				if (j&s_not_data_ready)
+				{
+					msg(DBG_INF, "read_audio: sbp_data: DATA_READY timeout.\n");
+					error_flag++;
+					break;
+				}
+				msg(DBG_AUD,"read_audio: before reading data.\n");
+				error_flag=0;
+				p = current_drive->aud_buf;
+				if (sbpro_type==1) OUT(CDo_sel_i_d,1);
+				if (do_16bit)
+				{
+					u_short *p2 = (u_short *) p;
+
+					for (; (u_char *) p2 < current_drive->aud_buf + read_audio.nframes*CD_FRAMESIZE_RAW;)
+				  	{
+						if ((inb_p(CDi_status)&s_not_data_ready)) continue;
+
+						/* get one sample */
+						*p2++ = inw_p(CDi_data);
+						*p2++ = inw_p(CDi_data);
+					}
+				} else {
+					for (; p < current_drive->aud_buf + read_audio.nframes*CD_FRAMESIZE_RAW;)
+				  	{
+						if ((inb_p(CDi_status)&s_not_data_ready)) continue;
+
+						/* get one sample */
+						*p++ = inb_p(CDi_data);
+						*p++ = inb_p(CDi_data);
+						*p++ = inb_p(CDi_data);
+						*p++ = inb_p(CDi_data);
+					}
+				}
+				if (sbpro_type==1) OUT(CDo_sel_i_d,0);
+				data_retrying = 0;
+			}
+			msg(DBG_AUD,"read_audio: after reading data.\n");
+			if (error_flag)    /* must have been spurious D_RDY or (ATTN&&!D_RDY) */
+			{
+				msg(DBG_AUD,"read_audio: read aborted by drive\n");
+#if 0000
+				i=cc_DriveReset();                /* ugly fix to prevent a hang */
+#else
+				i=cc_ReadError();
+#endif
+				continue;
+			}
+			if (fam0L_drive)
+			{
+				i=maxtim_data;
+				for (timeout=jiffies+9*HZ; time_before(jiffies, timeout); timeout--)
+				{
+					for ( ;i!=0;i--)
+					{
+						j=inb(CDi_status);
+						if (!(j&s_not_data_ready)) break;
+						if (!(j&s_not_result_ready)) break;
+						if (j&s_attention) break;
+					}
+					if (i != 0 || time_after_eq(jiffies, timeout)) break;
+					sbp_sleep(0);
+					i = 1;
+				}
+				if (i==0) msg(DBG_AUD,"read_audio: STATUS TIMEOUT AFTER READ");
+				if (!(j&s_attention))
+				{
+					msg(DBG_AUD,"read_audio: sbp_data: timeout waiting DRV_ATTN - retrying\n");
+					i=cc_DriveReset();  /* ugly fix to prevent a hang */
+					continue;
+				}
+			}
+			do
+			{
+				if (fam0L_drive) cc_ReadStatus();
+				i=ResponseStatus();  /* builds status_bits, returns orig. status (old) or faked p_success (new) */
+				if (i<0) { msg(DBG_AUD,
+					       "read_audio: cc_ReadStatus error after read: %02X\n",
+					       current_drive->status_bits);
+					   continue; /* FIXME */
+				   }
+			}
+			while ((fam0L_drive)&&(!st_check)&&(!(i&p_success)));
+			if (st_check)
+			{
+				i=cc_ReadError();
+				msg(DBG_AUD,"read_audio: cc_ReadError was necessary after read: %02X\n",i);
+				continue;
+			}
+			if (copy_to_user(read_audio.buf,
+					 current_drive->aud_buf,
+					 read_audio.nframes * CD_FRAMESIZE_RAW))
+				RETURN_UP(-EFAULT);
+			msg(DBG_AUD,"read_audio: copy_to_user done.\n");
+			break;
+		}
+		cc_ModeSelect(CD_FRAMESIZE);
+		cc_ModeSense();
+		current_drive->mode=READ_M1;
+#if OLD_BUSY
+		busy_audio=0;
+#endif /* OLD_BUSY */
+		if (data_tries == 0)
+		{
+			msg(DBG_AUD,"read_audio: failed after 5 tries in line %d.\n", __LINE__);
+			RETURN_UP(-EIO);
+		}
+		msg(DBG_AUD,"read_audio: successful return.\n");
+		RETURN_UP(0);
+	} /* end of CDROMREADAUDIO */
+
+	default:
+		msg(DBG_IOC,"ioctl: unknown function request %04X\n", cmd);
+		RETURN_UP(-EINVAL);
+	} /* end switch(cmd) */
 }
 
 static int sbpcd_block_media_changed(struct gendisk *disk)
@@ -5478,10 +5471,9 @@ static struct cdrom_device_ops sbpcd_dops = {
 	.get_mcn		= sbpcd_get_mcn,
 	.reset			= sbpcd_reset,
 	.audio_ioctl		= sbpcd_audio_ioctl,
-	.dev_ioctl		= sbpcd_dev_ioctl,
 	.capability		= CDC_CLOSE_TRAY | CDC_OPEN_TRAY | CDC_LOCK |
 				CDC_MULTI_SESSION | CDC_MEDIA_CHANGED |
-				CDC_MCN | CDC_PLAY_AUDIO | CDC_IOCTLS,
+				CDC_MCN | CDC_PLAY_AUDIO,
 	.n_minors		= 1,
 };
 
diff --git a/drivers/cdrom/viocd.c b/drivers/cdrom/viocd.c
index e2761725955..c0f817ba7ad 100644
--- a/drivers/cdrom/viocd.c
+++ b/drivers/cdrom/viocd.c
@@ -627,7 +627,7 @@ static struct cdrom_device_ops viocd_dops = {
 	.media_changed = viocd_media_changed,
 	.lock_door = viocd_lock_door,
 	.generic_packet = viocd_packet,
-	.capability = CDC_CLOSE_TRAY | CDC_OPEN_TRAY | CDC_LOCK | CDC_SELECT_SPEED | CDC_SELECT_DISC | CDC_MULTI_SESSION | CDC_MCN | CDC_MEDIA_CHANGED | CDC_PLAY_AUDIO | CDC_RESET | CDC_IOCTLS | CDC_DRIVE_STATUS | CDC_GENERIC_PACKET | CDC_CD_R | CDC_CD_RW | CDC_DVD | CDC_DVD_R | CDC_DVD_RAM | CDC_RAM
+	.capability = CDC_CLOSE_TRAY | CDC_OPEN_TRAY | CDC_LOCK | CDC_SELECT_SPEED | CDC_SELECT_DISC | CDC_MULTI_SESSION | CDC_MCN | CDC_MEDIA_CHANGED | CDC_PLAY_AUDIO | CDC_RESET | CDC_DRIVE_STATUS | CDC_GENERIC_PACKET | CDC_CD_R | CDC_CD_RW | CDC_DVD | CDC_DVD_R | CDC_DVD_RAM | CDC_RAM
 };
 
 static int __init find_capability(const char *type)
diff --git a/drivers/char/amiserial.c b/drivers/char/amiserial.c
index 7ac365b5d9e..6602b3156df 100644
--- a/drivers/char/amiserial.c
+++ b/drivers/char/amiserial.c
@@ -46,8 +46,6 @@
 
 /* Sanity checks */
 
-#define SERIAL_INLINE
-  
 #if defined(MODULE) && defined(SERIAL_DEBUG_MCOUNT)
 #define DBG_CNT(s) printk("(%s): [%x] refc=%d, serc=%d, ttyc=%d -> %s\n", \
  tty->name, (info->flags), serial_driver->refcount,info->count,tty->count,s)
@@ -95,10 +93,6 @@ static char *serial_version = "4.30";
 #include <asm/amigahw.h>
 #include <asm/amigaints.h>
 
-#ifdef SERIAL_INLINE
-#define _INLINE_ inline
-#endif
-
 #define custom amiga_custom
 static char *serial_name = "Amiga-builtin serial driver";
 
@@ -253,14 +247,14 @@ static void rs_start(struct tty_struct *tty)
  * This routine is used by the interrupt handler to schedule
  * processing in the software interrupt portion of the driver.
  */
-static _INLINE_ void rs_sched_event(struct async_struct *info,
-				  int event)
+static void rs_sched_event(struct async_struct *info,
+			   int event)
 {
 	info->event |= 1 << event;
 	tasklet_schedule(&info->tlet);
 }
 
-static _INLINE_ void receive_chars(struct async_struct *info)
+static void receive_chars(struct async_struct *info)
 {
         int status;
 	int serdatr;
@@ -349,7 +343,7 @@ out:
 	return;
 }
 
-static _INLINE_ void transmit_chars(struct async_struct *info)
+static void transmit_chars(struct async_struct *info)
 {
 	custom.intreq = IF_TBE;
 	mb();
@@ -389,7 +383,7 @@ static _INLINE_ void transmit_chars(struct async_struct *info)
 	}
 }
 
-static _INLINE_ void check_modem_status(struct async_struct *info)
+static void check_modem_status(struct async_struct *info)
 {
 	unsigned char status = ciab.pra & (SER_DCD | SER_CTS | SER_DSR);
 	unsigned char dstatus;
@@ -1959,7 +1953,7 @@ done:
  * number, and identifies which options were configured into this
  * driver.
  */
-static _INLINE_ void show_serial_version(void)
+static void show_serial_version(void)
 {
  	printk(KERN_INFO "%s version %s\n", serial_name, serial_version);
 }
diff --git a/drivers/char/generic_serial.c b/drivers/char/generic_serial.c
index e38a5f0e07b..5e59c0b4273 100644
--- a/drivers/char/generic_serial.c
+++ b/drivers/char/generic_serial.c
@@ -48,8 +48,8 @@ static int gs_debug;
 #define NEW_WRITE_LOCKING 1
 #if NEW_WRITE_LOCKING
 #define DECL      /* Nothing */
-#define LOCKIT    down (& port->port_write_sem);
-#define RELEASEIT up (&port->port_write_sem);
+#define LOCKIT    mutex_lock(& port->port_write_mutex);
+#define RELEASEIT mutex_unlock(&port->port_write_mutex);
 #else
 #define DECL      unsigned long flags;
 #define LOCKIT    save_flags (flags);cli ()
@@ -124,14 +124,14 @@ int gs_write(struct tty_struct * tty,
 	/* get exclusive "write" access to this port (problem 3) */
 	/* This is not a spinlock because we can have a disk access (page 
 		 fault) in copy_from_user */
-	down (& port->port_write_sem);
+	mutex_lock(& port->port_write_mutex);
 
 	while (1) {
 
 		c = count;
  
 		/* This is safe because we "OWN" the "head". Noone else can 
-		   change the "head": we own the port_write_sem. */
+		   change the "head": we own the port_write_mutex. */
 		/* Don't overrun the end of the buffer */
 		t = SERIAL_XMIT_SIZE - port->xmit_head;
 		if (t < c) c = t;
@@ -153,7 +153,7 @@ int gs_write(struct tty_struct * tty,
 		count -= c;
 		total += c;
 	}
-	up (& port->port_write_sem);
+	mutex_unlock(& port->port_write_mutex);
 
 	gs_dprintk (GS_DEBUG_WRITE, "write: interrupts are %s\n", 
 	            (port->flags & GS_TX_INTEN)?"enabled": "disabled"); 
@@ -214,7 +214,7 @@ int gs_write(struct tty_struct * tty,
 		c = count;
 
 		/* This is safe because we "OWN" the "head". Noone else can 
-		   change the "head": we own the port_write_sem. */
+		   change the "head": we own the port_write_mutex. */
 		/* Don't overrun the end of the buffer */
 		t = SERIAL_XMIT_SIZE - port->xmit_head;
 		if (t < c) c = t;
@@ -888,7 +888,7 @@ int gs_init_port(struct gs_port *port)
 	spin_lock_irqsave (&port->driver_lock, flags);
 	if (port->tty) 
 		clear_bit(TTY_IO_ERROR, &port->tty->flags);
-	init_MUTEX(&port->port_write_sem);
+	mutex_init(&port->port_write_mutex);
 	port->xmit_cnt = port->xmit_head = port->xmit_tail = 0;
 	spin_unlock_irqrestore(&port->driver_lock, flags);
 	gs_set_termios(port->tty, NULL);
diff --git a/drivers/char/istallion.c b/drivers/char/istallion.c
index 28c5a3193b8..ede128356af 100644
--- a/drivers/char/istallion.c
+++ b/drivers/char/istallion.c
@@ -181,7 +181,6 @@ static struct tty_driver	*stli_serial;
  *	is already swapping a shared buffer won't make things any worse.
  */
 static char			*stli_tmpwritebuf;
-static DECLARE_MUTEX(stli_tmpwritesem);
 
 #define	STLI_TXBUFSIZE		4096
 
diff --git a/drivers/char/n_tty.c b/drivers/char/n_tty.c
index ccad7ae9454..ede365d0538 100644
--- a/drivers/char/n_tty.c
+++ b/drivers/char/n_tty.c
@@ -132,7 +132,7 @@ static void put_tty_queue(unsigned char c, struct tty_struct *tty)
  *	We test the TTY_THROTTLED bit first so that it always
  *	indicates the current state. The decision about whether
  *	it is worth allowing more input has been taken by the caller.
- *	Can sleep, may be called under the atomic_read semaphore but
+ *	Can sleep, may be called under the atomic_read_lock mutex but
  *	this is not guaranteed.
  */
  
@@ -1132,7 +1132,7 @@ static inline int input_available_p(struct tty_struct *tty, int amt)
  *	buffer, and once to drain the space from the (physical) beginning of
  *	the buffer to head pointer.
  *
- *	Called under the tty->atomic_read sem and with TTY_DONT_FLIP set
+ *	Called under the tty->atomic_read_lock sem and with TTY_DONT_FLIP set
  *
  */
  
@@ -1262,11 +1262,11 @@ do_it_again:
 	 *	Internal serialization of reads.
 	 */
 	if (file->f_flags & O_NONBLOCK) {
-		if (down_trylock(&tty->atomic_read))
+		if (!mutex_trylock(&tty->atomic_read_lock))
 			return -EAGAIN;
 	}
 	else {
-		if (down_interruptible(&tty->atomic_read))
+		if (mutex_lock_interruptible(&tty->atomic_read_lock))
 			return -ERESTARTSYS;
 	}
 
@@ -1393,7 +1393,7 @@ do_it_again:
 			timeout = time;
 	}
 	clear_bit(TTY_DONT_FLIP, &tty->flags);
-	up(&tty->atomic_read);
+	mutex_unlock(&tty->atomic_read_lock);
 	remove_wait_queue(&tty->read_wait, &wait);
 
 	if (!waitqueue_active(&tty->read_wait))
diff --git a/drivers/char/nwflash.c b/drivers/char/nwflash.c
index ca41d62b1d9..8865387d344 100644
--- a/drivers/char/nwflash.c
+++ b/drivers/char/nwflash.c
@@ -27,6 +27,7 @@
 #include <linux/rwsem.h>
 #include <linux/init.h>
 #include <linux/smp_lock.h>
+#include <linux/mutex.h>
 
 #include <asm/hardware/dec21285.h>
 #include <asm/io.h>
@@ -56,7 +57,7 @@ static int gbWriteEnable;
 static int gbWriteBase64Enable;
 static volatile unsigned char *FLASH_BASE;
 static int gbFlashSize = KFLASH_SIZE;
-static DECLARE_MUTEX(nwflash_sem);
+static DEFINE_MUTEX(nwflash_mutex);
 
 extern spinlock_t gpio_lock;
 
@@ -140,7 +141,7 @@ static ssize_t flash_read(struct file *file, char __user *buf, size_t size,
 		/*
 		 * We now lock against reads and writes. --rmk
 		 */
-		if (down_interruptible(&nwflash_sem))
+		if (mutex_lock_interruptible(&nwflash_mutex))
 			return -ERESTARTSYS;
 
 		ret = copy_to_user(buf, (void *)(FLASH_BASE + p), count);
@@ -149,7 +150,7 @@ static ssize_t flash_read(struct file *file, char __user *buf, size_t size,
 			*ppos += count;
 		} else
 			ret = -EFAULT;
-		up(&nwflash_sem);
+		mutex_unlock(&nwflash_mutex);
 	}
 	return ret;
 }
@@ -188,7 +189,7 @@ static ssize_t flash_write(struct file *file, const char __user *buf,
 	/*
 	 * We now lock against reads and writes. --rmk
 	 */
-	if (down_interruptible(&nwflash_sem))
+	if (mutex_lock_interruptible(&nwflash_mutex))
 		return -ERESTARTSYS;
 
 	written = 0;
@@ -277,7 +278,7 @@ static ssize_t flash_write(struct file *file, const char __user *buf,
 	 */
 	leds_event(led_release);
 
-	up(&nwflash_sem);
+	mutex_unlock(&nwflash_mutex);
 
 	return written;
 }
diff --git a/drivers/char/raw.c b/drivers/char/raw.c
index 30e4cbe16bb..15a7b408652 100644
--- a/drivers/char/raw.c
+++ b/drivers/char/raw.c
@@ -19,6 +19,7 @@
 #include <linux/uio.h>
 #include <linux/cdev.h>
 #include <linux/device.h>
+#include <linux/mutex.h>
 
 #include <asm/uaccess.h>
 
@@ -29,7 +30,7 @@ struct raw_device_data {
 
 static struct class *raw_class;
 static struct raw_device_data raw_devices[MAX_RAW_MINORS];
-static DECLARE_MUTEX(raw_mutex);
+static DEFINE_MUTEX(raw_mutex);
 static struct file_operations raw_ctl_fops;	     /* forward declaration */
 
 /*
@@ -53,7 +54,7 @@ static int raw_open(struct inode *inode, struct file *filp)
 		return 0;
 	}
 
-	down(&raw_mutex);
+	mutex_lock(&raw_mutex);
 
 	/*
 	 * All we need to do on open is check that the device is bound.
@@ -78,7 +79,7 @@ static int raw_open(struct inode *inode, struct file *filp)
 		filp->f_dentry->d_inode->i_mapping =
 			bdev->bd_inode->i_mapping;
 	filp->private_data = bdev;
-	up(&raw_mutex);
+	mutex_unlock(&raw_mutex);
 	return 0;
 
 out2:
@@ -86,7 +87,7 @@ out2:
 out1:
 	blkdev_put(bdev);
 out:
-	up(&raw_mutex);
+	mutex_unlock(&raw_mutex);
 	return err;
 }
 
@@ -99,14 +100,14 @@ static int raw_release(struct inode *inode, struct file *filp)
 	const int minor= iminor(inode);
 	struct block_device *bdev;
 
-	down(&raw_mutex);
+	mutex_lock(&raw_mutex);
 	bdev = raw_devices[minor].binding;
 	if (--raw_devices[minor].inuse == 0) {
 		/* Here  inode->i_mapping == bdev->bd_inode->i_mapping  */
 		inode->i_mapping = &inode->i_data;
 		inode->i_mapping->backing_dev_info = &default_backing_dev_info;
 	}
-	up(&raw_mutex);
+	mutex_unlock(&raw_mutex);
 
 	bd_release(bdev);
 	blkdev_put(bdev);
@@ -187,9 +188,9 @@ static int raw_ctl_ioctl(struct inode *inode, struct file *filp,
 				goto out;
 			}
 
-			down(&raw_mutex);
+			mutex_lock(&raw_mutex);
 			if (rawdev->inuse) {
-				up(&raw_mutex);
+				mutex_unlock(&raw_mutex);
 				err = -EBUSY;
 				goto out;
 			}
@@ -211,11 +212,11 @@ static int raw_ctl_ioctl(struct inode *inode, struct file *filp,
 					bind_device(&rq);
 				}
 			}
-			up(&raw_mutex);
+			mutex_unlock(&raw_mutex);
 		} else {
 			struct block_device *bdev;
 
-			down(&raw_mutex);
+			mutex_lock(&raw_mutex);
 			bdev = rawdev->binding;
 			if (bdev) {
 				rq.block_major = MAJOR(bdev->bd_dev);
@@ -223,7 +224,7 @@ static int raw_ctl_ioctl(struct inode *inode, struct file *filp,
 			} else {
 				rq.block_major = rq.block_minor = 0;
 			}
-			up(&raw_mutex);
+			mutex_unlock(&raw_mutex);
 			if (copy_to_user((void __user *)arg, &rq, sizeof(rq))) {
 				err = -EFAULT;
 				goto out;
diff --git a/drivers/char/ser_a2232.c b/drivers/char/ser_a2232.c
index fee68cc895f..510bd3e0e88 100644
--- a/drivers/char/ser_a2232.c
+++ b/drivers/char/ser_a2232.c
@@ -97,7 +97,7 @@
 #include <asm/amigahw.h>
 #include <linux/zorro.h>
 #include <asm/irq.h>
-#include <asm/semaphore.h>
+#include <linux/mutex.h>
 
 #include <linux/delay.h>
 
@@ -654,7 +654,7 @@ static void a2232_init_portstructs(void)
 		port->gs.closing_wait = 30 * HZ;
 		port->gs.rd = &a2232_real_driver;
 #ifdef NEW_WRITE_LOCKING
-		init_MUTEX(&(port->gs.port_write_sem));
+		init_MUTEX(&(port->gs.port_write_mutex));
 #endif
 		init_waitqueue_head(&port->gs.open_wait);
 		init_waitqueue_head(&port->gs.close_wait);
diff --git a/drivers/char/snsc.c b/drivers/char/snsc.c
index 0e7d216e7eb..b543821d8cb 100644
--- a/drivers/char/snsc.c
+++ b/drivers/char/snsc.c
@@ -5,7 +5,7 @@
  * License.  See the file "COPYING" in the main directory of this archive
  * for more details.
  *
- * Copyright (C) 2004 Silicon Graphics, Inc. All rights reserved.
+ * Copyright (C) 2004, 2006 Silicon Graphics, Inc. All rights reserved.
  */
 
 /*
@@ -77,7 +77,7 @@ scdrv_open(struct inode *inode, struct file *file)
 	scd = container_of(inode->i_cdev, struct sysctl_data_s, scd_cdev);
 
 	/* allocate memory for subchannel data */
-	sd = kmalloc(sizeof (struct subch_data_s), GFP_KERNEL);
+	sd = kzalloc(sizeof (struct subch_data_s), GFP_KERNEL);
 	if (sd == NULL) {
 		printk("%s: couldn't allocate subchannel data\n",
 		       __FUNCTION__);
@@ -85,7 +85,6 @@ scdrv_open(struct inode *inode, struct file *file)
 	}
 
 	/* initialize subch_data_s fields */
-	memset(sd, 0, sizeof (struct subch_data_s));
 	sd->sd_nasid = scd->scd_nasid;
 	sd->sd_subch = ia64_sn_irtr_open(scd->scd_nasid);
 
@@ -394,7 +393,7 @@ scdrv_init(void)
 			sprintf(devnamep, "#%d", geo_slab(geoid));
 
 			/* allocate sysctl device data */
-			scd = kmalloc(sizeof (struct sysctl_data_s),
+			scd = kzalloc(sizeof (struct sysctl_data_s),
 				      GFP_KERNEL);
 			if (!scd) {
 				printk("%s: failed to allocate device info"
@@ -402,7 +401,6 @@ scdrv_init(void)
 				       SYSCTL_BASENAME, devname);
 				continue;
 			}
-			memset(scd, 0, sizeof (struct sysctl_data_s));
 
 			/* initialize sysctl device data fields */
 			scd->scd_nasid = cnodeid_to_nasid(cnode);
diff --git a/drivers/char/snsc_event.c b/drivers/char/snsc_event.c
index a4fa507eed9..e234d50e142 100644
--- a/drivers/char/snsc_event.c
+++ b/drivers/char/snsc_event.c
@@ -287,7 +287,7 @@ scdrv_event_init(struct sysctl_data_s *scd)
 {
 	int rv;
 
-	event_sd = kmalloc(sizeof (struct subch_data_s), GFP_KERNEL);
+	event_sd = kzalloc(sizeof (struct subch_data_s), GFP_KERNEL);
 	if (event_sd == NULL) {
 		printk(KERN_WARNING "%s: couldn't allocate subchannel info"
 		       " for event monitoring\n", __FUNCTION__);
@@ -295,7 +295,6 @@ scdrv_event_init(struct sysctl_data_s *scd)
 	}
 
 	/* initialize subch_data_s fields */
-	memset(event_sd, 0, sizeof (struct subch_data_s));
 	event_sd->sd_nasid = scd->scd_nasid;
 	spin_lock_init(&event_sd->sd_rlock);
 
@@ -321,5 +320,3 @@ scdrv_event_init(struct sysctl_data_s *scd)
 		return;
 	}
 }
-
-
diff --git a/drivers/char/stallion.c b/drivers/char/stallion.c
index bdaab699210..3f5d6077f39 100644
--- a/drivers/char/stallion.c
+++ b/drivers/char/stallion.c
@@ -148,7 +148,6 @@ static struct tty_driver	*stl_serial;
  *	is already swapping a shared buffer won't make things any worse.
  */
 static char			*stl_tmpwritebuf;
-static DECLARE_MUTEX(stl_tmpwritesem);
 
 /*
  *	Define a local default termios struct. All ports will be created
diff --git a/drivers/char/sx.c b/drivers/char/sx.c
index a6b4f02bdce..3b474723027 100644
--- a/drivers/char/sx.c
+++ b/drivers/char/sx.c
@@ -2318,7 +2318,7 @@ static int sx_init_portstructs (int nboards, int nports)
 			port->board = board;
 			port->gs.rd = &sx_real_driver;
 #ifdef NEW_WRITE_LOCKING
-			port->gs.port_write_sem = MUTEX;
+			port->gs.port_write_mutex = MUTEX;
 #endif
 			port->gs.driver_lock = SPIN_LOCK_UNLOCKED;
 			/*
diff --git a/drivers/char/tty_io.c b/drivers/char/tty_io.c
index 53d3d066554..76592ee1fb3 100644
--- a/drivers/char/tty_io.c
+++ b/drivers/char/tty_io.c
@@ -130,7 +130,7 @@ LIST_HEAD(tty_drivers);			/* linked list of tty drivers */
 
 /* Semaphore to protect creating and releasing a tty. This is shared with
    vt.c for deeply disgusting hack reasons */
-DECLARE_MUTEX(tty_sem);
+DEFINE_MUTEX(tty_mutex);
 
 #ifdef CONFIG_UNIX98_PTYS
 extern struct tty_driver *ptm_driver;	/* Unix98 pty masters; for /dev/ptmx */
@@ -1188,11 +1188,11 @@ void disassociate_ctty(int on_exit)
 
 	lock_kernel();
 
-	down(&tty_sem);
+	mutex_lock(&tty_mutex);
 	tty = current->signal->tty;
 	if (tty) {
 		tty_pgrp = tty->pgrp;
-		up(&tty_sem);
+		mutex_unlock(&tty_mutex);
 		if (on_exit && tty->driver->type != TTY_DRIVER_TYPE_PTY)
 			tty_vhangup(tty);
 	} else {
@@ -1200,7 +1200,7 @@ void disassociate_ctty(int on_exit)
 			kill_pg(current->signal->tty_old_pgrp, SIGHUP, on_exit);
 			kill_pg(current->signal->tty_old_pgrp, SIGCONT, on_exit);
 		}
-		up(&tty_sem);
+		mutex_unlock(&tty_mutex);
 		unlock_kernel();	
 		return;
 	}
@@ -1211,7 +1211,7 @@ void disassociate_ctty(int on_exit)
 	}
 
 	/* Must lock changes to tty_old_pgrp */
-	down(&tty_sem);
+	mutex_lock(&tty_mutex);
 	current->signal->tty_old_pgrp = 0;
 	tty->session = 0;
 	tty->pgrp = -1;
@@ -1222,7 +1222,7 @@ void disassociate_ctty(int on_exit)
 		p->signal->tty = NULL;
 	} while_each_task_pid(current->signal->session, PIDTYPE_SID, p);
 	read_unlock(&tasklist_lock);
-	up(&tty_sem);
+	mutex_unlock(&tty_mutex);
 	unlock_kernel();
 }
 
@@ -1306,7 +1306,7 @@ static inline ssize_t do_tty_write(
 	ssize_t ret = 0, written = 0;
 	unsigned int chunk;
 	
-	if (down_interruptible(&tty->atomic_write)) {
+	if (mutex_lock_interruptible(&tty->atomic_write_lock)) {
 		return -ERESTARTSYS;
 	}
 
@@ -1329,7 +1329,7 @@ static inline ssize_t do_tty_write(
 	if (count < chunk)
 		chunk = count;
 
-	/* write_buf/write_cnt is protected by the atomic_write semaphore */
+	/* write_buf/write_cnt is protected by the atomic_write_lock mutex */
 	if (tty->write_cnt < chunk) {
 		unsigned char *buf;
 
@@ -1338,7 +1338,7 @@ static inline ssize_t do_tty_write(
 
 		buf = kmalloc(chunk, GFP_KERNEL);
 		if (!buf) {
-			up(&tty->atomic_write);
+			mutex_unlock(&tty->atomic_write_lock);
 			return -ENOMEM;
 		}
 		kfree(tty->write_buf);
@@ -1374,7 +1374,7 @@ static inline ssize_t do_tty_write(
 		inode->i_mtime = current_fs_time(inode->i_sb);
 		ret = written;
 	}
-	up(&tty->atomic_write);
+	mutex_unlock(&tty->atomic_write_lock);
 	return ret;
 }
 
@@ -1442,8 +1442,8 @@ static inline void tty_line_name(struct tty_driver *driver, int index, char *p)
 
 /*
  * WSH 06/09/97: Rewritten to remove races and properly clean up after a
- * failed open.  The new code protects the open with a semaphore, so it's
- * really quite straightforward.  The semaphore locking can probably be
+ * failed open.  The new code protects the open with a mutex, so it's
+ * really quite straightforward.  The mutex locking can probably be
  * relaxed for the (most common) case of reopening a tty.
  */
 static int init_dev(struct tty_driver *driver, int idx,
@@ -1640,7 +1640,7 @@ fast_track:
 success:
 	*ret_tty = tty;
 	
-	/* All paths come through here to release the semaphore */
+	/* All paths come through here to release the mutex */
 end_init:
 	return retval;
 
@@ -1837,7 +1837,7 @@ static void release_dev(struct file * filp)
 		/* Guard against races with tty->count changes elsewhere and
 		   opens on /dev/tty */
 		   
-		down(&tty_sem);
+		mutex_lock(&tty_mutex);
 		tty_closing = tty->count <= 1;
 		o_tty_closing = o_tty &&
 			(o_tty->count <= (pty_master ? 1 : 0));
@@ -1868,7 +1868,7 @@ static void release_dev(struct file * filp)
 
 		printk(KERN_WARNING "release_dev: %s: read/write wait queue "
 				    "active!\n", tty_name(tty, buf));
-		up(&tty_sem);
+		mutex_unlock(&tty_mutex);
 		schedule();
 	}	
 
@@ -1934,7 +1934,7 @@ static void release_dev(struct file * filp)
 		read_unlock(&tasklist_lock);
 	}
 
-	up(&tty_sem);
+	mutex_unlock(&tty_mutex);
 
 	/* check whether both sides are closing ... */
 	if (!tty_closing || (o_tty && !o_tty_closing))
@@ -2040,11 +2040,11 @@ retry_open:
 	index  = -1;
 	retval = 0;
 	
-	down(&tty_sem);
+	mutex_lock(&tty_mutex);
 
 	if (device == MKDEV(TTYAUX_MAJOR,0)) {
 		if (!current->signal->tty) {
-			up(&tty_sem);
+			mutex_unlock(&tty_mutex);
 			return -ENXIO;
 		}
 		driver = current->signal->tty->driver;
@@ -2070,18 +2070,18 @@ retry_open:
 			noctty = 1;
 			goto got_driver;
 		}
-		up(&tty_sem);
+		mutex_unlock(&tty_mutex);
 		return -ENODEV;
 	}
 
 	driver = get_tty_driver(device, &index);
 	if (!driver) {
-		up(&tty_sem);
+		mutex_unlock(&tty_mutex);
 		return -ENODEV;
 	}
 got_driver:
 	retval = init_dev(driver, index, &tty);
-	up(&tty_sem);
+	mutex_unlock(&tty_mutex);
 	if (retval)
 		return retval;
 
@@ -2167,9 +2167,9 @@ static int ptmx_open(struct inode * inode, struct file * filp)
 	}
 	up(&allocated_ptys_lock);
 
-	down(&tty_sem);
+	mutex_lock(&tty_mutex);
 	retval = init_dev(ptm_driver, index, &tty);
-	up(&tty_sem);
+	mutex_unlock(&tty_mutex);
 	
 	if (retval)
 		goto out;
@@ -2915,8 +2915,8 @@ static void initialize_tty_struct(struct tty_struct *tty)
 	init_waitqueue_head(&tty->write_wait);
 	init_waitqueue_head(&tty->read_wait);
 	INIT_WORK(&tty->hangup_work, do_tty_hangup, tty);
-	sema_init(&tty->atomic_read, 1);
-	sema_init(&tty->atomic_write, 1);
+	mutex_init(&tty->atomic_read_lock);
+	mutex_init(&tty->atomic_write_lock);
 	spin_lock_init(&tty->read_lock);
 	INIT_LIST_HEAD(&tty->tty_files);
 	INIT_WORK(&tty->SAK_work, NULL, NULL);
diff --git a/drivers/char/vme_scc.c b/drivers/char/vme_scc.c
index d9325281e48..fd00822ac14 100644
--- a/drivers/char/vme_scc.c
+++ b/drivers/char/vme_scc.c
@@ -184,7 +184,7 @@ static void scc_init_portstructs(void)
 		port->gs.closing_wait = 30 * HZ;
 		port->gs.rd = &scc_real_driver;
 #ifdef NEW_WRITE_LOCKING
-		port->gs.port_write_sem = MUTEX;
+		port->gs.port_write_mutex = MUTEX;
 #endif
 		init_waitqueue_head(&port->gs.open_wait);
 		init_waitqueue_head(&port->gs.close_wait);
diff --git a/drivers/char/vt.c b/drivers/char/vt.c
index 0900d1dbee5..ca4844c527d 100644
--- a/drivers/char/vt.c
+++ b/drivers/char/vt.c
@@ -2489,7 +2489,7 @@ static int con_open(struct tty_struct *tty, struct file *filp)
 }
 
 /*
- * We take tty_sem in here to prevent another thread from coming in via init_dev
+ * We take tty_mutex in here to prevent another thread from coming in via init_dev
  * and taking a ref against the tty while we're in the process of forgetting
  * about it and cleaning things up.
  *
@@ -2497,7 +2497,7 @@ static int con_open(struct tty_struct *tty, struct file *filp)
  */
 static void con_close(struct tty_struct *tty, struct file *filp)
 {
-	down(&tty_sem);
+	mutex_lock(&tty_mutex);
 	acquire_console_sem();
 	if (tty && tty->count == 1) {
 		struct vc_data *vc = tty->driver_data;
@@ -2507,15 +2507,15 @@ static void con_close(struct tty_struct *tty, struct file *filp)
 		tty->driver_data = NULL;
 		release_console_sem();
 		vcs_remove_devfs(tty);
-		up(&tty_sem);
+		mutex_unlock(&tty_mutex);
 		/*
-		 * tty_sem is released, but we still hold BKL, so there is
+		 * tty_mutex is released, but we still hold BKL, so there is
 		 * still exclusion against init_dev()
 		 */
 		return;
 	}
 	release_console_sem();
-	up(&tty_sem);
+	mutex_unlock(&tty_mutex);
 }
 
 static void vc_init(struct vc_data *vc, unsigned int rows,
@@ -2869,9 +2869,9 @@ void unblank_screen(void)
 }
 
 /*
- * We defer the timer blanking to work queue so it can take the console semaphore
+ * We defer the timer blanking to work queue so it can take the console mutex
  * (console operations can still happen at irq time, but only from printk which
- * has the console semaphore. Not perfect yet, but better than no locking
+ * has the console mutex. Not perfect yet, but better than no locking
  */
 static void blank_screen_t(unsigned long dummy)
 {
@@ -3234,6 +3234,14 @@ void vcs_scr_writew(struct vc_data *vc, u16 val, u16 *org)
 	}
 }
 
+int is_console_suspend_safe(void)
+{
+	/* It is unsafe to suspend devices while X has control of the
+	 * hardware. Make sure we are running on a kernel-controlled console.
+	 */
+	return vc_cons[fg_console].d->vc_mode == KD_TEXT;
+}
+
 /*
  *	Visible symbols for modules
  */
diff --git a/drivers/char/watchdog/pcwd_usb.c b/drivers/char/watchdog/pcwd_usb.c
index 1533f56baa4..2700c5c45b8 100644
--- a/drivers/char/watchdog/pcwd_usb.c
+++ b/drivers/char/watchdog/pcwd_usb.c
@@ -42,6 +42,7 @@
 #include <linux/completion.h>
 #include <asm/uaccess.h>
 #include <linux/usb.h>
+#include <linux/mutex.h>
 
 
 #ifdef CONFIG_USB_DEBUG
@@ -143,7 +144,7 @@ struct usb_pcwd_private {
 static struct usb_pcwd_private *usb_pcwd_device;
 
 /* prevent races between open() and disconnect() */
-static DECLARE_MUTEX (disconnect_sem);
+static DEFINE_MUTEX(disconnect_mutex);
 
 /* local function prototypes */
 static int usb_pcwd_probe	(struct usb_interface *interface, const struct usb_device_id *id);
@@ -723,7 +724,7 @@ static void usb_pcwd_disconnect(struct usb_interface *interface)
 	struct usb_pcwd_private *usb_pcwd;
 
 	/* prevent races with open() */
-	down (&disconnect_sem);
+	mutex_lock(&disconnect_mutex);
 
 	usb_pcwd = usb_get_intfdata (interface);
 	usb_set_intfdata (interface, NULL);
@@ -749,7 +750,7 @@ static void usb_pcwd_disconnect(struct usb_interface *interface)
 
 	cards_found--;
 
-	up (&disconnect_sem);
+	mutex_unlock(&disconnect_mutex);
 
 	printk(KERN_INFO PFX "USB PC Watchdog disconnected\n");
 }
diff --git a/drivers/connector/connector.c b/drivers/connector/connector.c
index d7125f4d911..35897079a78 100644
--- a/drivers/connector/connector.c
+++ b/drivers/connector/connector.c
@@ -26,6 +26,7 @@
 #include <linux/netlink.h>
 #include <linux/moduleparam.h>
 #include <linux/connector.h>
+#include <linux/mutex.h>
 
 #include <net/sock.h>
 
@@ -41,7 +42,7 @@ module_param(cn_val, uint, 0);
 MODULE_PARM_DESC(cn_idx, "Connector's main device idx.");
 MODULE_PARM_DESC(cn_val, "Connector's main device val.");
 
-static DECLARE_MUTEX(notify_lock);
+static DEFINE_MUTEX(notify_lock);
 static LIST_HEAD(notify_list);
 
 static struct cn_dev cdev;
@@ -260,7 +261,7 @@ static void cn_notify(struct cb_id *id, u32 notify_event)
 {
 	struct cn_ctl_entry *ent;
 
-	down(&notify_lock);
+	mutex_lock(&notify_lock);
 	list_for_each_entry(ent, &notify_list, notify_entry) {
 		int i;
 		struct cn_notify_req *req;
@@ -293,7 +294,7 @@ static void cn_notify(struct cb_id *id, u32 notify_event)
 			cn_netlink_send(&m, ctl->group, GFP_KERNEL);
 		}
 	}
-	up(&notify_lock);
+	mutex_unlock(&notify_lock);
 }
 
 /*
@@ -407,14 +408,14 @@ static void cn_callback(void *data)
 	if (ctl->group == 0) {
 		struct cn_ctl_entry *n;
 
-		down(&notify_lock);
+		mutex_lock(&notify_lock);
 		list_for_each_entry_safe(ent, n, &notify_list, notify_entry) {
 			if (cn_ctl_msg_equals(ent->msg, ctl)) {
 				list_del(&ent->notify_entry);
 				kfree(ent);
 			}
 		}
-		up(&notify_lock);
+		mutex_unlock(&notify_lock);
 
 		return;
 	}
@@ -429,9 +430,9 @@ static void cn_callback(void *data)
 
 	memcpy(ent->msg, ctl, size - sizeof(*ent));
 
-	down(&notify_lock);
+	mutex_lock(&notify_lock);
 	list_add(&ent->notify_entry, &notify_list);
-	up(&notify_lock);
+	mutex_unlock(&notify_lock);
 }
 
 static int __init cn_init(void)
diff --git a/drivers/firmware/dcdbas.c b/drivers/firmware/dcdbas.c
index 3a4e5c5b4e1..d6543fc4a92 100644
--- a/drivers/firmware/dcdbas.c
+++ b/drivers/firmware/dcdbas.c
@@ -33,6 +33,7 @@
 #include <linux/spinlock.h>
 #include <linux/string.h>
 #include <linux/types.h>
+#include <linux/mutex.h>
 #include <asm/io.h>
 #include <asm/semaphore.h>
 
@@ -48,7 +49,7 @@ static u8 *smi_data_buf;
 static dma_addr_t smi_data_buf_handle;
 static unsigned long smi_data_buf_size;
 static u32 smi_data_buf_phys_addr;
-static DECLARE_MUTEX(smi_data_lock);
+static DEFINE_MUTEX(smi_data_lock);
 
 static unsigned int host_control_action;
 static unsigned int host_control_smi_type;
@@ -139,9 +140,9 @@ static ssize_t smi_data_buf_size_store(struct device *dev,
 	buf_size = simple_strtoul(buf, NULL, 10);
 
 	/* make sure SMI data buffer is at least buf_size */
-	down(&smi_data_lock);
+	mutex_lock(&smi_data_lock);
 	ret = smi_data_buf_realloc(buf_size);
-	up(&smi_data_lock);
+	mutex_unlock(&smi_data_lock);
 	if (ret)
 		return ret;
 
@@ -154,7 +155,7 @@ static ssize_t smi_data_read(struct kobject *kobj, char *buf, loff_t pos,
 	size_t max_read;
 	ssize_t ret;
 
-	down(&smi_data_lock);
+	mutex_lock(&smi_data_lock);
 
 	if (pos >= smi_data_buf_size) {
 		ret = 0;
@@ -165,7 +166,7 @@ static ssize_t smi_data_read(struct kobject *kobj, char *buf, loff_t pos,
 	ret = min(max_read, count);
 	memcpy(buf, smi_data_buf + pos, ret);
 out:
-	up(&smi_data_lock);
+	mutex_unlock(&smi_data_lock);
 	return ret;
 }
 
@@ -174,7 +175,7 @@ static ssize_t smi_data_write(struct kobject *kobj, char *buf, loff_t pos,
 {
 	ssize_t ret;
 
-	down(&smi_data_lock);
+	mutex_lock(&smi_data_lock);
 
 	ret = smi_data_buf_realloc(pos + count);
 	if (ret)
@@ -183,7 +184,7 @@ static ssize_t smi_data_write(struct kobject *kobj, char *buf, loff_t pos,
 	memcpy(smi_data_buf + pos, buf, count);
 	ret = count;
 out:
-	up(&smi_data_lock);
+	mutex_unlock(&smi_data_lock);
 	return ret;
 }
 
@@ -201,9 +202,9 @@ static ssize_t host_control_action_store(struct device *dev,
 	ssize_t ret;
 
 	/* make sure buffer is available for host control command */
-	down(&smi_data_lock);
+	mutex_lock(&smi_data_lock);
 	ret = smi_data_buf_realloc(sizeof(struct apm_cmd));
-	up(&smi_data_lock);
+	mutex_unlock(&smi_data_lock);
 	if (ret)
 		return ret;
 
@@ -302,7 +303,7 @@ static ssize_t smi_request_store(struct device *dev,
 	unsigned long val = simple_strtoul(buf, NULL, 10);
 	ssize_t ret;
 
-	down(&smi_data_lock);
+	mutex_lock(&smi_data_lock);
 
 	if (smi_data_buf_size < sizeof(struct smi_cmd)) {
 		ret = -ENODEV;
@@ -334,7 +335,7 @@ static ssize_t smi_request_store(struct device *dev,
 	}
 
 out:
-	up(&smi_data_lock);
+	mutex_unlock(&smi_data_lock);
 	return ret;
 }
 
diff --git a/drivers/ide/ide-cd.c b/drivers/ide/ide-cd.c
index 3325660f724..c7671e18801 100644
--- a/drivers/ide/ide-cd.c
+++ b/drivers/ide/ide-cd.c
@@ -313,6 +313,7 @@
 #include <linux/cdrom.h>
 #include <linux/ide.h>
 #include <linux/completion.h>
+#include <linux/mutex.h>
 
 #include <scsi/scsi.h>	/* For SCSI -> ATAPI command conversion */
 
@@ -324,7 +325,7 @@
 
 #include "ide-cd.h"
 
-static DECLARE_MUTEX(idecd_ref_sem);
+static DEFINE_MUTEX(idecd_ref_mutex);
 
 #define to_ide_cd(obj) container_of(obj, struct cdrom_info, kref) 
 
@@ -335,11 +336,11 @@ static struct cdrom_info *ide_cd_get(struct gendisk *disk)
 {
 	struct cdrom_info *cd = NULL;
 
-	down(&idecd_ref_sem);
+	mutex_lock(&idecd_ref_mutex);
 	cd = ide_cd_g(disk);
 	if (cd)
 		kref_get(&cd->kref);
-	up(&idecd_ref_sem);
+	mutex_unlock(&idecd_ref_mutex);
 	return cd;
 }
 
@@ -347,9 +348,9 @@ static void ide_cd_release(struct kref *);
 
 static void ide_cd_put(struct cdrom_info *cd)
 {
-	down(&idecd_ref_sem);
+	mutex_lock(&idecd_ref_mutex);
 	kref_put(&cd->kref, ide_cd_release);
-	up(&idecd_ref_sem);
+	mutex_unlock(&idecd_ref_mutex);
 }
 
 /****************************************************************************
@@ -2471,52 +2472,6 @@ static int ide_cdrom_packet(struct cdrom_device_info *cdi,
 }
 
 static
-int ide_cdrom_dev_ioctl (struct cdrom_device_info *cdi,
-			 unsigned int cmd, unsigned long arg)
-{
-	struct packet_command cgc;
-	char buffer[16];
-	int stat;
-
-	init_cdrom_command(&cgc, buffer, sizeof(buffer), CGC_DATA_UNKNOWN);
-
-	/* These will be moved into the Uniform layer shortly... */
-	switch (cmd) {
- 	case CDROMSETSPINDOWN: {
- 		char spindown;
- 
- 		if (copy_from_user(&spindown, (void __user *) arg, sizeof(char)))
-			return -EFAULT;
- 
-                if ((stat = cdrom_mode_sense(cdi, &cgc, GPMODE_CDROM_PAGE, 0)))
-			return stat;
-
- 		buffer[11] = (buffer[11] & 0xf0) | (spindown & 0x0f);
-
- 		return cdrom_mode_select(cdi, &cgc);
- 	} 
- 
- 	case CDROMGETSPINDOWN: {
- 		char spindown;
- 
-                if ((stat = cdrom_mode_sense(cdi, &cgc, GPMODE_CDROM_PAGE, 0)))
-			return stat;
- 
- 		spindown = buffer[11] & 0x0f;
- 
-		if (copy_to_user((void __user *) arg, &spindown, sizeof (char)))
-			return -EFAULT;
- 
- 		return 0;
- 	}
-  
-	default:
-		return -EINVAL;
-	}
-
-}
-
-static
 int ide_cdrom_audio_ioctl (struct cdrom_device_info *cdi,
 			   unsigned int cmd, void *arg)
 			   
@@ -2852,12 +2807,11 @@ static struct cdrom_device_ops ide_cdrom_dops = {
 	.get_mcn		= ide_cdrom_get_mcn,
 	.reset			= ide_cdrom_reset,
 	.audio_ioctl		= ide_cdrom_audio_ioctl,
-	.dev_ioctl		= ide_cdrom_dev_ioctl,
 	.capability		= CDC_CLOSE_TRAY | CDC_OPEN_TRAY | CDC_LOCK |
 				CDC_SELECT_SPEED | CDC_SELECT_DISC |
 				CDC_MULTI_SESSION | CDC_MCN |
 				CDC_MEDIA_CHANGED | CDC_PLAY_AUDIO | CDC_RESET |
-				CDC_IOCTLS | CDC_DRIVE_STATUS | CDC_CD_R |
+				CDC_DRIVE_STATUS | CDC_CD_R |
 				CDC_CD_RW | CDC_DVD | CDC_DVD_R| CDC_DVD_RAM |
 				CDC_GENERIC_PACKET | CDC_MO_DRIVE | CDC_MRW |
 				CDC_MRW_W | CDC_RAM,
@@ -3367,6 +3321,45 @@ static int idecd_release(struct inode * inode, struct file * file)
 	return 0;
 }
 
+static int idecd_set_spindown(struct cdrom_device_info *cdi, unsigned long arg)
+{
+	struct packet_command cgc;
+	char buffer[16];
+	int stat;
+	char spindown;
+
+	if (copy_from_user(&spindown, (void __user *)arg, sizeof(char)))
+		return -EFAULT;
+
+	init_cdrom_command(&cgc, buffer, sizeof(buffer), CGC_DATA_UNKNOWN);
+
+	stat = cdrom_mode_sense(cdi, &cgc, GPMODE_CDROM_PAGE, 0);
+	if (stat)
+		return stat;
+
+	buffer[11] = (buffer[11] & 0xf0) | (spindown & 0x0f);
+	return cdrom_mode_select(cdi, &cgc);
+}
+
+static int idecd_get_spindown(struct cdrom_device_info *cdi, unsigned long arg)
+{
+	struct packet_command cgc;
+	char buffer[16];
+	int stat;
+ 	char spindown;
+
+	init_cdrom_command(&cgc, buffer, sizeof(buffer), CGC_DATA_UNKNOWN);
+
+	stat = cdrom_mode_sense(cdi, &cgc, GPMODE_CDROM_PAGE, 0);
+	if (stat)
+		return stat;
+
+	spindown = buffer[11] & 0x0f;
+	if (copy_to_user((void __user *)arg, &spindown, sizeof (char)))
+		return -EFAULT;
+	return 0;
+}
+
 static int idecd_ioctl (struct inode *inode, struct file *file,
 			unsigned int cmd, unsigned long arg)
 {
@@ -3374,7 +3367,16 @@ static int idecd_ioctl (struct inode *inode, struct file *file,
 	struct cdrom_info *info = ide_cd_g(bdev->bd_disk);
 	int err;
 
-	err  = generic_ide_ioctl(info->drive, file, bdev, cmd, arg);
+	switch (cmd) {
+ 	case CDROMSETSPINDOWN:
+		return idecd_set_spindown(&info->devinfo, arg);
+ 	case CDROMGETSPINDOWN:
+		return idecd_get_spindown(&info->devinfo, arg);
+	default:
+		break;
+ 	}
+
+	err = generic_ide_ioctl(info->drive, file, bdev, cmd, arg);
 	if (err == -EINVAL)
 		err = cdrom_ioctl(file, &info->devinfo, inode, cmd, arg);
 
diff --git a/drivers/ide/ide-disk.c b/drivers/ide/ide-disk.c
index 09086b8b648..e238b7da824 100644
--- a/drivers/ide/ide-disk.c
+++ b/drivers/ide/ide-disk.c
@@ -60,6 +60,7 @@
 #include <linux/genhd.h>
 #include <linux/slab.h>
 #include <linux/delay.h>
+#include <linux/mutex.h>
 
 #define _IDE_DISK
 
@@ -78,7 +79,7 @@ struct ide_disk_obj {
 	struct kref	kref;
 };
 
-static DECLARE_MUTEX(idedisk_ref_sem);
+static DEFINE_MUTEX(idedisk_ref_mutex);
 
 #define to_ide_disk(obj) container_of(obj, struct ide_disk_obj, kref)
 
@@ -89,11 +90,11 @@ static struct ide_disk_obj *ide_disk_get(struct gendisk *disk)
 {
 	struct ide_disk_obj *idkp = NULL;
 
-	down(&idedisk_ref_sem);
+	mutex_lock(&idedisk_ref_mutex);
 	idkp = ide_disk_g(disk);
 	if (idkp)
 		kref_get(&idkp->kref);
-	up(&idedisk_ref_sem);
+	mutex_unlock(&idedisk_ref_mutex);
 	return idkp;
 }
 
@@ -101,9 +102,9 @@ static void ide_disk_release(struct kref *);
 
 static void ide_disk_put(struct ide_disk_obj *idkp)
 {
-	down(&idedisk_ref_sem);
+	mutex_lock(&idedisk_ref_mutex);
 	kref_put(&idkp->kref, ide_disk_release);
-	up(&idedisk_ref_sem);
+	mutex_unlock(&idedisk_ref_mutex);
 }
 
 /*
diff --git a/drivers/ide/ide-floppy.c b/drivers/ide/ide-floppy.c
index 1f8db9ac05d..a53e3ce4a14 100644
--- a/drivers/ide/ide-floppy.c
+++ b/drivers/ide/ide-floppy.c
@@ -98,6 +98,7 @@
 #include <linux/cdrom.h>
 #include <linux/ide.h>
 #include <linux/bitops.h>
+#include <linux/mutex.h>
 
 #include <asm/byteorder.h>
 #include <asm/irq.h>
@@ -517,7 +518,7 @@ typedef struct {
 	u8		reserved[4];
 } idefloppy_mode_parameter_header_t;
 
-static DECLARE_MUTEX(idefloppy_ref_sem);
+static DEFINE_MUTEX(idefloppy_ref_mutex);
 
 #define to_ide_floppy(obj) container_of(obj, struct ide_floppy_obj, kref)
 
@@ -528,11 +529,11 @@ static struct ide_floppy_obj *ide_floppy_get(struct gendisk *disk)
 {
 	struct ide_floppy_obj *floppy = NULL;
 
-	down(&idefloppy_ref_sem);
+	mutex_lock(&idefloppy_ref_mutex);
 	floppy = ide_floppy_g(disk);
 	if (floppy)
 		kref_get(&floppy->kref);
-	up(&idefloppy_ref_sem);
+	mutex_unlock(&idefloppy_ref_mutex);
 	return floppy;
 }
 
@@ -540,9 +541,9 @@ static void ide_floppy_release(struct kref *);
 
 static void ide_floppy_put(struct ide_floppy_obj *floppy)
 {
-	down(&idefloppy_ref_sem);
+	mutex_lock(&idefloppy_ref_mutex);
 	kref_put(&floppy->kref, ide_floppy_release);
-	up(&idefloppy_ref_sem);
+	mutex_unlock(&idefloppy_ref_mutex);
 }
 
 /*
diff --git a/drivers/ide/ide-tape.c b/drivers/ide/ide-tape.c
index 0101d0def7c..ebc59064b47 100644
--- a/drivers/ide/ide-tape.c
+++ b/drivers/ide/ide-tape.c
@@ -443,6 +443,7 @@
 #include <linux/smp_lock.h>
 #include <linux/completion.h>
 #include <linux/bitops.h>
+#include <linux/mutex.h>
 
 #include <asm/byteorder.h>
 #include <asm/irq.h>
@@ -1011,7 +1012,7 @@ typedef struct ide_tape_obj {
          int debug_level; 
 } idetape_tape_t;
 
-static DECLARE_MUTEX(idetape_ref_sem);
+static DEFINE_MUTEX(idetape_ref_mutex);
 
 static struct class *idetape_sysfs_class;
 
@@ -1024,11 +1025,11 @@ static struct ide_tape_obj *ide_tape_get(struct gendisk *disk)
 {
 	struct ide_tape_obj *tape = NULL;
 
-	down(&idetape_ref_sem);
+	mutex_lock(&idetape_ref_mutex);
 	tape = ide_tape_g(disk);
 	if (tape)
 		kref_get(&tape->kref);
-	up(&idetape_ref_sem);
+	mutex_unlock(&idetape_ref_mutex);
 	return tape;
 }
 
@@ -1036,9 +1037,9 @@ static void ide_tape_release(struct kref *);
 
 static void ide_tape_put(struct ide_tape_obj *tape)
 {
-	down(&idetape_ref_sem);
+	mutex_lock(&idetape_ref_mutex);
 	kref_put(&tape->kref, ide_tape_release);
-	up(&idetape_ref_sem);
+	mutex_unlock(&idetape_ref_mutex);
 }
 
 /*
@@ -1290,11 +1291,11 @@ static struct ide_tape_obj *ide_tape_chrdev_get(unsigned int i)
 {
 	struct ide_tape_obj *tape = NULL;
 
-	down(&idetape_ref_sem);
+	mutex_lock(&idetape_ref_mutex);
 	tape = idetape_devs[i];
 	if (tape)
 		kref_get(&tape->kref);
-	up(&idetape_ref_sem);
+	mutex_unlock(&idetape_ref_mutex);
 	return tape;
 }
 
@@ -4870,11 +4871,11 @@ static int ide_tape_probe(ide_drive_t *drive)
 
 	drive->driver_data = tape;
 
-	down(&idetape_ref_sem);
+	mutex_lock(&idetape_ref_mutex);
 	for (minor = 0; idetape_devs[minor]; minor++)
 		;
 	idetape_devs[minor] = tape;
-	up(&idetape_ref_sem);
+	mutex_unlock(&idetape_ref_mutex);
 
 	idetape_setup(drive, tape, minor);
 
diff --git a/drivers/isdn/capi/kcapi.c b/drivers/isdn/capi/kcapi.c
index feec40cf590..8c4fcb9027b 100644
--- a/drivers/isdn/capi/kcapi.c
+++ b/drivers/isdn/capi/kcapi.c
@@ -32,6 +32,7 @@
 #ifdef CONFIG_AVMB1_COMPAT
 #include <linux/b1lli.h>
 #endif
+#include <linux/mutex.h>
 
 static char *revision = "$Revision: 1.1.2.8 $";
 
@@ -66,7 +67,7 @@ LIST_HEAD(capi_drivers);
 DEFINE_RWLOCK(capi_drivers_list_lock);
 
 static DEFINE_RWLOCK(application_lock);
-static DECLARE_MUTEX(controller_sem);
+static DEFINE_MUTEX(controller_mutex);
 
 struct capi20_appl *capi_applications[CAPI_MAXAPPL];
 struct capi_ctr *capi_cards[CAPI_MAXCONTR];
@@ -395,20 +396,20 @@ attach_capi_ctr(struct capi_ctr *card)
 {
 	int i;
 
-	down(&controller_sem);
+	mutex_lock(&controller_mutex);
 
 	for (i = 0; i < CAPI_MAXCONTR; i++) {
 		if (capi_cards[i] == NULL)
 			break;
 	}
 	if (i == CAPI_MAXCONTR) {
-		up(&controller_sem);
+		mutex_unlock(&controller_mutex);
 		printk(KERN_ERR "kcapi: out of controller slots\n");
 	   	return -EBUSY;
 	}
 	capi_cards[i] = card;
 
-	up(&controller_sem);
+	mutex_unlock(&controller_mutex);
 
 	card->nrecvctlpkt = 0;
 	card->nrecvdatapkt = 0;
@@ -531,13 +532,13 @@ u16 capi20_register(struct capi20_appl *ap)
 
 	write_unlock_irqrestore(&application_lock, flags);
 	
-	down(&controller_sem);
+	mutex_lock(&controller_mutex);
 	for (i = 0; i < CAPI_MAXCONTR; i++) {
 		if (!capi_cards[i] || capi_cards[i]->cardstate != CARD_RUNNING)
 			continue;
 		register_appl(capi_cards[i], applid, &ap->rparam);
 	}
-	up(&controller_sem);
+	mutex_unlock(&controller_mutex);
 
 	if (showcapimsgs & 1) {
 		printk(KERN_DEBUG "kcapi: appl %d up\n", applid);
@@ -560,13 +561,13 @@ u16 capi20_release(struct capi20_appl *ap)
 	capi_applications[ap->applid - 1] = NULL;
 	write_unlock_irqrestore(&application_lock, flags);
 
-	down(&controller_sem);
+	mutex_lock(&controller_mutex);
 	for (i = 0; i < CAPI_MAXCONTR; i++) {
 		if (!capi_cards[i] || capi_cards[i]->cardstate != CARD_RUNNING)
 			continue;
 		release_appl(capi_cards[i], ap->applid);
 	}
-	up(&controller_sem);
+	mutex_unlock(&controller_mutex);
 
 	flush_scheduled_work();
 	skb_queue_purge(&ap->recv_queue);
diff --git a/drivers/isdn/hisax/config.c b/drivers/isdn/hisax/config.c
index df9d6520181..27332506f9f 100644
--- a/drivers/isdn/hisax/config.c
+++ b/drivers/isdn/hisax/config.c
@@ -25,7 +25,6 @@
 #include <linux/workqueue.h>
 #include <linux/interrupt.h>
 #define HISAX_STATUS_BUFSIZE 4096
-#define INCLUDE_INLINE_FUNCS
 
 /*
  * This structure array contains one entry per card. An entry looks
diff --git a/drivers/isdn/hisax/elsa.c b/drivers/isdn/hisax/elsa.c
index 110e9fd669c..f8ca4b32333 100644
--- a/drivers/isdn/hisax/elsa.c
+++ b/drivers/isdn/hisax/elsa.c
@@ -108,7 +108,6 @@ static const char *ITACVer[] =
 #define ELSA_ASSIGN      4
 
 #define RS_ISR_PASS_LIMIT 256
-#define _INLINE_ inline
 #define FLG_MODEM_ACTIVE 1
 /* IPAC AUX */
 #define ELSA_IPAC_LINE_LED	0x40	/* Bit 6 Gelbe LED */
diff --git a/drivers/net/loopback.c b/drivers/net/loopback.c
index 690a1aae0b3..0c13795dca3 100644
--- a/drivers/net/loopback.c
+++ b/drivers/net/loopback.c
@@ -172,11 +172,9 @@ static struct net_device_stats *get_stats(struct net_device *dev)
 
 	memset(stats, 0, sizeof(struct net_device_stats));
 
-	for (i=0; i < NR_CPUS; i++) {
+	for_each_cpu(i) {
 		struct net_device_stats *lb_stats;
 
-		if (!cpu_possible(i)) 
-			continue;
 		lb_stats = &per_cpu(loopback_stats, i);
 		stats->rx_bytes   += lb_stats->rx_bytes;
 		stats->tx_bytes   += lb_stats->tx_bytes;
diff --git a/drivers/net/ppp_generic.c b/drivers/net/ppp_generic.c
index f608c12e3e8..b2073fce821 100644
--- a/drivers/net/ppp_generic.c
+++ b/drivers/net/ppp_generic.c
@@ -46,6 +46,7 @@
 #include <linux/rwsem.h>
 #include <linux/stddef.h>
 #include <linux/device.h>
+#include <linux/mutex.h>
 #include <net/slhc_vj.h>
 #include <asm/atomic.h>
 
@@ -198,11 +199,11 @@ static unsigned int cardmap_find_first_free(struct cardmap *map);
 static void cardmap_destroy(struct cardmap **map);
 
 /*
- * all_ppp_sem protects the all_ppp_units mapping.
+ * all_ppp_mutex protects the all_ppp_units mapping.
  * It also ensures that finding a ppp unit in the all_ppp_units map
  * and updating its file.refcnt field is atomic.
  */
-static DECLARE_MUTEX(all_ppp_sem);
+static DEFINE_MUTEX(all_ppp_mutex);
 static struct cardmap *all_ppp_units;
 static atomic_t ppp_unit_count = ATOMIC_INIT(0);
 
@@ -804,7 +805,7 @@ static int ppp_unattached_ioctl(struct ppp_file *pf, struct file *file,
 		/* Attach to an existing ppp unit */
 		if (get_user(unit, p))
 			break;
-		down(&all_ppp_sem);
+		mutex_lock(&all_ppp_mutex);
 		err = -ENXIO;
 		ppp = ppp_find_unit(unit);
 		if (ppp != 0) {
@@ -812,7 +813,7 @@ static int ppp_unattached_ioctl(struct ppp_file *pf, struct file *file,
 			file->private_data = &ppp->file;
 			err = 0;
 		}
-		up(&all_ppp_sem);
+		mutex_unlock(&all_ppp_mutex);
 		break;
 
 	case PPPIOCATTCHAN:
@@ -2446,7 +2447,7 @@ ppp_create_interface(int unit, int *retp)
 	dev->do_ioctl = ppp_net_ioctl;
 
 	ret = -EEXIST;
-	down(&all_ppp_sem);
+	mutex_lock(&all_ppp_mutex);
 	if (unit < 0)
 		unit = cardmap_find_first_free(all_ppp_units);
 	else if (cardmap_get(all_ppp_units, unit) != NULL)
@@ -2465,12 +2466,12 @@ ppp_create_interface(int unit, int *retp)
 
 	atomic_inc(&ppp_unit_count);
 	cardmap_set(&all_ppp_units, unit, ppp);
-	up(&all_ppp_sem);
+	mutex_unlock(&all_ppp_mutex);
 	*retp = 0;
 	return ppp;
 
 out2:
-	up(&all_ppp_sem);
+	mutex_unlock(&all_ppp_mutex);
 	free_netdev(dev);
 out1:
 	kfree(ppp);
@@ -2500,7 +2501,7 @@ static void ppp_shutdown_interface(struct ppp *ppp)
 {
 	struct net_device *dev;
 
-	down(&all_ppp_sem);
+	mutex_lock(&all_ppp_mutex);
 	ppp_lock(ppp);
 	dev = ppp->dev;
 	ppp->dev = NULL;
@@ -2514,7 +2515,7 @@ static void ppp_shutdown_interface(struct ppp *ppp)
 	ppp->file.dead = 1;
 	ppp->owner = NULL;
 	wake_up_interruptible(&ppp->file.rwait);
-	up(&all_ppp_sem);
+	mutex_unlock(&all_ppp_mutex);
 }
 
 /*
@@ -2556,7 +2557,7 @@ static void ppp_destroy_interface(struct ppp *ppp)
 
 /*
  * Locate an existing ppp unit.
- * The caller should have locked the all_ppp_sem.
+ * The caller should have locked the all_ppp_mutex.
  */
 static struct ppp *
 ppp_find_unit(int unit)
@@ -2601,7 +2602,7 @@ ppp_connect_channel(struct channel *pch, int unit)
 	int ret = -ENXIO;
 	int hdrlen;
 
-	down(&all_ppp_sem);
+	mutex_lock(&all_ppp_mutex);
 	ppp = ppp_find_unit(unit);
 	if (ppp == 0)
 		goto out;
@@ -2626,7 +2627,7 @@ ppp_connect_channel(struct channel *pch, int unit)
  outl:
 	write_unlock_bh(&pch->upl);
  out:
-	up(&all_ppp_sem);
+	mutex_unlock(&all_ppp_mutex);
 	return ret;
 }
 
diff --git a/drivers/oprofile/cpu_buffer.c b/drivers/oprofile/cpu_buffer.c
index 78193e4bbdb..330d3869b41 100644
--- a/drivers/oprofile/cpu_buffer.c
+++ b/drivers/oprofile/cpu_buffer.c
@@ -38,9 +38,8 @@ void free_cpu_buffers(void)
 {
 	int i;
  
-	for_each_online_cpu(i) {
+	for_each_online_cpu(i)
 		vfree(cpu_buffer[i].buffer);
-	}
 }
 
 int alloc_cpu_buffers(void)
diff --git a/drivers/pnp/pnpbios/rsparser.c b/drivers/pnp/pnpbios/rsparser.c
index 5e38cd7335f..c89c98a2cca 100644
--- a/drivers/pnp/pnpbios/rsparser.c
+++ b/drivers/pnp/pnpbios/rsparser.c
@@ -448,11 +448,7 @@ pnpbios_parse_resource_option_data(unsigned char * p, unsigned char * end, struc
 			break;
 
 		case SMALL_TAG_END:
-			if (option_independent != option)
-				printk(KERN_WARNING "PnPBIOS: Missing SMALL_TAG_ENDDEP tag\n");
-			p = p + 2;
-        		return (unsigned char *)p;
-			break;
+        		return p + 2;
 
 		default: /* an unkown tag */
 			len_err:
diff --git a/drivers/s390/block/dasd_ioctl.c b/drivers/s390/block/dasd_ioctl.c
index fafeeae5267..f9930552ab5 100644
--- a/drivers/s390/block/dasd_ioctl.c
+++ b/drivers/s390/block/dasd_ioctl.c
@@ -151,9 +151,9 @@ dasd_ioctl_enable(struct block_device *bdev, int no, long args)
 		return -ENODEV;
 	dasd_enable_device(device);
 	/* Formatting the dasd device can change the capacity. */
-	down(&bdev->bd_sem);
+	mutex_lock(&bdev->bd_mutex);
 	i_size_write(bdev->bd_inode, (loff_t)get_capacity(device->gdp) << 9);
-	up(&bdev->bd_sem);
+	mutex_unlock(&bdev->bd_mutex);
 	return 0;
 }
 
@@ -184,9 +184,9 @@ dasd_ioctl_disable(struct block_device *bdev, int no, long args)
 	 * Set i_size to zero, since read, write, etc. check against this
 	 * value.
 	 */
-	down(&bdev->bd_sem);
+	mutex_lock(&bdev->bd_mutex);
 	i_size_write(bdev->bd_inode, 0);
-	up(&bdev->bd_sem);
+	mutex_unlock(&bdev->bd_mutex);
 	return 0;
 }
 
diff --git a/drivers/scsi/ide-scsi.c b/drivers/scsi/ide-scsi.c
index 0cf0e4c7ac0..39b760a2424 100644
--- a/drivers/scsi/ide-scsi.c
+++ b/drivers/scsi/ide-scsi.c
@@ -47,6 +47,7 @@
 #include <linux/ide.h>
 #include <linux/scatterlist.h>
 #include <linux/delay.h>
+#include <linux/mutex.h>
 
 #include <asm/io.h>
 #include <asm/bitops.h>
@@ -109,7 +110,7 @@ typedef struct ide_scsi_obj {
 	unsigned long log;			/* log flags */
 } idescsi_scsi_t;
 
-static DECLARE_MUTEX(idescsi_ref_sem);
+static DEFINE_MUTEX(idescsi_ref_mutex);
 
 #define ide_scsi_g(disk) \
 	container_of((disk)->private_data, struct ide_scsi_obj, driver)
@@ -118,19 +119,19 @@ static struct ide_scsi_obj *ide_scsi_get(struct gendisk *disk)
 {
 	struct ide_scsi_obj *scsi = NULL;
 
-	down(&idescsi_ref_sem);
+	mutex_lock(&idescsi_ref_mutex);
 	scsi = ide_scsi_g(disk);
 	if (scsi)
 		scsi_host_get(scsi->host);
-	up(&idescsi_ref_sem);
+	mutex_unlock(&idescsi_ref_mutex);
 	return scsi;
 }
 
 static void ide_scsi_put(struct ide_scsi_obj *scsi)
 {
-	down(&idescsi_ref_sem);
+	mutex_lock(&idescsi_ref_mutex);
 	scsi_host_put(scsi->host);
-	up(&idescsi_ref_sem);
+	mutex_unlock(&idescsi_ref_mutex);
 }
 
 static inline idescsi_scsi_t *scsihost_to_idescsi(struct Scsi_Host *host)
diff --git a/drivers/scsi/sr.c b/drivers/scsi/sr.c
index f9c1192dc15..7c80711e18e 100644
--- a/drivers/scsi/sr.c
+++ b/drivers/scsi/sr.c
@@ -71,7 +71,7 @@ MODULE_ALIAS_BLOCKDEV_MAJOR(SCSI_CDROM_MAJOR);
 #define SR_CAPABILITIES \
 	(CDC_CLOSE_TRAY|CDC_OPEN_TRAY|CDC_LOCK|CDC_SELECT_SPEED| \
 	 CDC_SELECT_DISC|CDC_MULTI_SESSION|CDC_MCN|CDC_MEDIA_CHANGED| \
-	 CDC_PLAY_AUDIO|CDC_RESET|CDC_IOCTLS|CDC_DRIVE_STATUS| \
+	 CDC_PLAY_AUDIO|CDC_RESET|CDC_DRIVE_STATUS| \
 	 CDC_CD_R|CDC_CD_RW|CDC_DVD|CDC_DVD_R|CDC_DVD_RAM|CDC_GENERIC_PACKET| \
 	 CDC_MRW|CDC_MRW_W|CDC_RAM)
 
@@ -118,7 +118,6 @@ static struct cdrom_device_ops sr_dops = {
 	.get_mcn		= sr_get_mcn,
 	.reset			= sr_reset,
 	.audio_ioctl		= sr_audio_ioctl,
-	.dev_ioctl		= sr_dev_ioctl,
 	.capability		= SR_CAPABILITIES,
 	.generic_packet		= sr_packet,
 };
@@ -456,17 +455,33 @@ static int sr_block_ioctl(struct inode *inode, struct file *file, unsigned cmd,
 {
 	struct scsi_cd *cd = scsi_cd(inode->i_bdev->bd_disk);
 	struct scsi_device *sdev = cd->device;
+	void __user *argp = (void __user *)arg;
+	int ret;
 
-        /*
-         * Send SCSI addressing ioctls directly to mid level, send other
-         * ioctls to cdrom/block level.
-         */
-        switch (cmd) {
-                case SCSI_IOCTL_GET_IDLUN:
-                case SCSI_IOCTL_GET_BUS_NUMBER:
-                        return scsi_ioctl(sdev, cmd, (void __user *)arg);
+	/*
+	 * Send SCSI addressing ioctls directly to mid level, send other
+	 * ioctls to cdrom/block level.
+	 */
+	switch (cmd) {
+	case SCSI_IOCTL_GET_IDLUN:
+	case SCSI_IOCTL_GET_BUS_NUMBER:
+		return scsi_ioctl(sdev, cmd, argp);
 	}
-	return cdrom_ioctl(file, &cd->cdi, inode, cmd, arg);
+
+	ret = cdrom_ioctl(file, &cd->cdi, inode, cmd, arg);
+	if (ret != ENOSYS)
+		return ret;
+
+	/*
+	 * ENODEV means that we didn't recognise the ioctl, or that we
+	 * cannot execute it in the current device state.  In either
+	 * case fall through to scsi_ioctl, which will return ENDOEV again
+	 * if it doesn't recognise the ioctl
+	 */
+	ret = scsi_nonblockable_ioctl(sdev, cmd, argp, NULL);
+	if (ret != -ENODEV)
+		return ret;
+	return scsi_ioctl(sdev, cmd, argp);
 }
 
 static int sr_block_media_changed(struct gendisk *disk)
diff --git a/drivers/scsi/sr.h b/drivers/scsi/sr.h
index d2bcd99c272..d65de9621b2 100644
--- a/drivers/scsi/sr.h
+++ b/drivers/scsi/sr.h
@@ -55,7 +55,6 @@ int sr_get_mcn(struct cdrom_device_info *, struct cdrom_mcn *);
 int sr_reset(struct cdrom_device_info *);
 int sr_select_speed(struct cdrom_device_info *cdi, int speed);
 int sr_audio_ioctl(struct cdrom_device_info *, unsigned int, void *);
-int sr_dev_ioctl(struct cdrom_device_info *, unsigned int, unsigned long);
 
 int sr_is_xa(Scsi_CD *);
 
diff --git a/drivers/scsi/sr_ioctl.c b/drivers/scsi/sr_ioctl.c
index b65462f7648..d1268cb4683 100644
--- a/drivers/scsi/sr_ioctl.c
+++ b/drivers/scsi/sr_ioctl.c
@@ -562,22 +562,3 @@ int sr_is_xa(Scsi_CD *cd)
 #endif
 	return is_xa;
 }
-
-int sr_dev_ioctl(struct cdrom_device_info *cdi,
-		 unsigned int cmd, unsigned long arg)
-{
-	Scsi_CD *cd = cdi->handle;
-	int ret;
-	
-	ret = scsi_nonblockable_ioctl(cd->device, cmd,
-				      (void __user *)arg, NULL);
-	/*
-	 * ENODEV means that we didn't recognise the ioctl, or that we
-	 * cannot execute it in the current device state.  In either
-	 * case fall through to scsi_ioctl, which will return ENDOEV again
-	 * if it doesn't recognise the ioctl
-	 */
-	if (ret != -ENODEV)
-		return ret;
-	return scsi_ioctl(cd->device, cmd, (void __user *)arg);
-}
diff --git a/drivers/serial/68328serial.c b/drivers/serial/68328serial.c
index 7f0f35a05dc..b88a7c1158a 100644
--- a/drivers/serial/68328serial.c
+++ b/drivers/serial/68328serial.c
@@ -101,8 +101,6 @@ struct tty_driver *serial_driver;
 
 #define RS_ISR_PASS_LIMIT 256
 
-#define _INLINE_ inline
-
 static void change_speed(struct m68k_serial *info);
 
 /*
@@ -262,7 +260,7 @@ static void batten_down_hatches(void)
 	/* Drop into the debugger */
 }
 
-static _INLINE_ void status_handle(struct m68k_serial *info, unsigned short status)
+static void status_handle(struct m68k_serial *info, unsigned short status)
 {
 #if 0
 	if(status & DCD) {
@@ -289,7 +287,8 @@ static _INLINE_ void status_handle(struct m68k_serial *info, unsigned short stat
 	return;
 }
 
-static _INLINE_ void receive_chars(struct m68k_serial *info, struct pt_regs *regs, unsigned short rx)
+static void receive_chars(struct m68k_serial *info, struct pt_regs *regs,
+			  unsigned short rx)
 {
 	struct tty_struct *tty = info->tty;
 	m68328_uart *uart = &uart_addr[info->line];
@@ -359,7 +358,7 @@ clear_and_exit:
 	return;
 }
 
-static _INLINE_ void transmit_chars(struct m68k_serial *info)
+static void transmit_chars(struct m68k_serial *info)
 {
 	m68328_uart *uart = &uart_addr[info->line];
 
diff --git a/drivers/serial/au1x00_uart.c b/drivers/serial/au1x00_uart.c
index 29f94bbb79b..948880ac587 100644
--- a/drivers/serial/au1x00_uart.c
+++ b/drivers/serial/au1x00_uart.c
@@ -133,13 +133,12 @@ static const struct serial_uart_config uart_config[PORT_MAX_8250+1] = {
 	{ "AU1X00_UART",16,	UART_CLEAR_FIFO | UART_USE_FIFO },
 };
 
-static _INLINE_ unsigned int serial_in(struct uart_8250_port *up, int offset)
+static unsigned int serial_in(struct uart_8250_port *up, int offset)
 {
 	return au_readl((unsigned long)up->port.membase + offset);
 }
 
-static _INLINE_ void
-serial_out(struct uart_8250_port *up, int offset, int value)
+static void serial_out(struct uart_8250_port *up, int offset, int value)
 {
 	au_writel(value, (unsigned long)up->port.membase + offset);
 }
@@ -237,7 +236,7 @@ static void serial8250_enable_ms(struct uart_port *port)
 	serial_out(up, UART_IER, up->ier);
 }
 
-static _INLINE_ void
+static void
 receive_chars(struct uart_8250_port *up, int *status, struct pt_regs *regs)
 {
 	struct tty_struct *tty = up->port.info->tty;
@@ -312,7 +311,7 @@ receive_chars(struct uart_8250_port *up, int *status, struct pt_regs *regs)
 	spin_lock(&up->port.lock);
 }
 
-static _INLINE_ void transmit_chars(struct uart_8250_port *up)
+static void transmit_chars(struct uart_8250_port *up)
 {
 	struct circ_buf *xmit = &up->port.info->xmit;
 	int count;
@@ -346,7 +345,7 @@ static _INLINE_ void transmit_chars(struct uart_8250_port *up)
 		serial8250_stop_tx(&up->port);
 }
 
-static _INLINE_ void check_modem_status(struct uart_8250_port *up)
+static void check_modem_status(struct uart_8250_port *up)
 {
 	int status;
 
diff --git a/drivers/serial/crisv10.c b/drivers/serial/crisv10.c
index be12623d854..89700141f87 100644
--- a/drivers/serial/crisv10.c
+++ b/drivers/serial/crisv10.c
@@ -481,8 +481,6 @@ static char *serial_version = "$Revision: 1.25 $";
 #include "serial_compat.h"
 #endif
 
-#define _INLINE_ inline
-
 struct tty_driver *serial_driver;
 
 /* serial subtype definitions */
@@ -591,8 +589,6 @@ static void rs_throttle(struct tty_struct * tty);
 static void rs_wait_until_sent(struct tty_struct *tty, int timeout);
 static int rs_write(struct tty_struct * tty, int from_user,
                     const unsigned char *buf, int count);
-extern _INLINE_ int rs_raw_write(struct tty_struct * tty, int from_user,
-                            const unsigned char *buf, int count);
 #ifdef CONFIG_ETRAX_RS485
 static int e100_write_rs485(struct tty_struct * tty, int from_user,
                             const unsigned char *buf, int count);
@@ -1538,8 +1534,7 @@ e100_enable_rxdma_irq(struct e100_serial *info)
 
 /* the tx DMA uses only dma_descr interrupt */
 
-static _INLINE_ void
-e100_disable_txdma_irq(struct e100_serial *info)
+static void e100_disable_txdma_irq(struct e100_serial *info)
 {
 #ifdef SERIAL_DEBUG_INTR
 	printk("txdma_irq(%d): 0\n",info->line);
@@ -1548,8 +1543,7 @@ e100_disable_txdma_irq(struct e100_serial *info)
 	*R_IRQ_MASK2_CLR = info->irq;
 }
 
-static _INLINE_ void
-e100_enable_txdma_irq(struct e100_serial *info)
+static void e100_enable_txdma_irq(struct e100_serial *info)
 {
 #ifdef SERIAL_DEBUG_INTR
 	printk("txdma_irq(%d): 1\n",info->line);
@@ -1558,8 +1552,7 @@ e100_enable_txdma_irq(struct e100_serial *info)
 	*R_IRQ_MASK2_SET = info->irq;
 }
 
-static _INLINE_ void
-e100_disable_txdma_channel(struct e100_serial *info)
+static void e100_disable_txdma_channel(struct e100_serial *info)
 {
 	unsigned long flags;
 
@@ -1599,8 +1592,7 @@ e100_disable_txdma_channel(struct e100_serial *info)
 }
 
 
-static _INLINE_ void
-e100_enable_txdma_channel(struct e100_serial *info)
+static void e100_enable_txdma_channel(struct e100_serial *info)
 {
 	unsigned long flags;
 
@@ -1625,8 +1617,7 @@ e100_enable_txdma_channel(struct e100_serial *info)
 	restore_flags(flags);
 }
 
-static _INLINE_ void
-e100_disable_rxdma_channel(struct e100_serial *info)
+static void e100_disable_rxdma_channel(struct e100_serial *info)
 {
 	unsigned long flags;
 
@@ -1665,8 +1656,7 @@ e100_disable_rxdma_channel(struct e100_serial *info)
 }
 
 
-static _INLINE_ void
-e100_enable_rxdma_channel(struct e100_serial *info)
+static void e100_enable_rxdma_channel(struct e100_serial *info)
 {
 	unsigned long flags;
 
@@ -1913,9 +1903,7 @@ rs_start(struct tty_struct *tty)
  * This routine is used by the interrupt handler to schedule
  * processing in the software interrupt portion of the driver.
  */
-static _INLINE_ void
-rs_sched_event(struct e100_serial *info,
-				    int event)
+static void rs_sched_event(struct e100_serial *info, int event)
 {
 	if (info->event & (1 << event))
 		return;
@@ -2155,8 +2143,9 @@ add_char_and_flag(struct e100_serial *info, unsigned char data, unsigned char fl
 	return 1;
 }
 
-extern _INLINE_ unsigned int
-handle_descr_data(struct e100_serial *info, struct etrax_dma_descr *descr, unsigned int recvl)
+static unsigned int handle_descr_data(struct e100_serial *info,
+				      struct etrax_dma_descr *descr,
+				      unsigned int recvl)
 {
 	struct etrax_recv_buffer *buffer = phys_to_virt(descr->buf) - sizeof *buffer;
 
@@ -2182,8 +2171,7 @@ handle_descr_data(struct e100_serial *info, struct etrax_dma_descr *descr, unsig
 	return recvl;
 }
 
-static _INLINE_ unsigned int
-handle_all_descr_data(struct e100_serial *info)
+static unsigned int handle_all_descr_data(struct e100_serial *info)
 {
 	struct etrax_dma_descr *descr;
 	unsigned int recvl;
@@ -2230,8 +2218,7 @@ handle_all_descr_data(struct e100_serial *info)
 	return ret;
 }
 
-static _INLINE_ void
-receive_chars_dma(struct e100_serial *info)
+static void receive_chars_dma(struct e100_serial *info)
 {
 	struct tty_struct *tty;
 	unsigned char rstat;
@@ -2292,8 +2279,7 @@ receive_chars_dma(struct e100_serial *info)
 	*info->icmdadr = IO_STATE(R_DMA_CH6_CMD, cmd, restart);
 }
 
-static _INLINE_ int
-start_recv_dma(struct e100_serial *info)
+static int start_recv_dma(struct e100_serial *info)
 {
 	struct etrax_dma_descr *descr = info->rec_descr;
 	struct etrax_recv_buffer *buffer;
@@ -2348,11 +2334,6 @@ start_receive(struct e100_serial *info)
 }
 
 
-static _INLINE_ void
-status_handle(struct e100_serial *info, unsigned short status)
-{
-}
-
 /* the bits in the MASK2 register are laid out like this:
    DMAI_EOP DMAI_DESCR DMAO_EOP DMAO_DESCR
    where I is the input channel and O is the output channel for the port.
@@ -2454,8 +2435,7 @@ rec_interrupt(int irq, void *dev_id, struct pt_regs * regs)
 	return IRQ_RETVAL(handled);
 } /* rec_interrupt */
 
-static _INLINE_ int
-force_eop_if_needed(struct e100_serial *info)
+static int force_eop_if_needed(struct e100_serial *info)
 {
 	/* We check data_avail bit to determine if data has
 	 * arrived since last time
@@ -2499,8 +2479,7 @@ force_eop_if_needed(struct e100_serial *info)
 	return 1;
 }
 
-extern _INLINE_ void
-flush_to_flip_buffer(struct e100_serial *info)
+static void flush_to_flip_buffer(struct e100_serial *info)
 {
 	struct tty_struct *tty;
 	struct etrax_recv_buffer *buffer;
@@ -2611,8 +2590,7 @@ flush_to_flip_buffer(struct e100_serial *info)
 	tty_flip_buffer_push(tty);
 }
 
-static _INLINE_ void
-check_flush_timeout(struct e100_serial *info)
+static void check_flush_timeout(struct e100_serial *info)
 {
 	/* Flip what we've got (if we can) */
 	flush_to_flip_buffer(info);
@@ -2741,7 +2719,7 @@ TODO: The break will be delayed until an F or V character is received.
 
 */
 
-extern _INLINE_
+static
 struct e100_serial * handle_ser_rx_interrupt_no_dma(struct e100_serial *info)
 {
 	unsigned long data_read;
@@ -2875,8 +2853,7 @@ more_data:
 	return info;
 }
 
-extern _INLINE_
-struct e100_serial* handle_ser_rx_interrupt(struct e100_serial *info)
+static struct e100_serial* handle_ser_rx_interrupt(struct e100_serial *info)
 {
 	unsigned char rstat;
 
@@ -2995,7 +2972,7 @@ struct e100_serial* handle_ser_rx_interrupt(struct e100_serial *info)
 	return info;
 } /* handle_ser_rx_interrupt */
 
-extern _INLINE_ void handle_ser_tx_interrupt(struct e100_serial *info)
+static void handle_ser_tx_interrupt(struct e100_serial *info)
 {
 	unsigned long flags;
 
@@ -3621,9 +3598,8 @@ rs_flush_chars(struct tty_struct *tty)
 	restore_flags(flags);
 }
 
-extern _INLINE_ int
-rs_raw_write(struct tty_struct * tty, int from_user,
-	  const unsigned char *buf, int count)
+static int rs_raw_write(struct tty_struct * tty, int from_user,
+			const unsigned char *buf, int count)
 {
 	int	c, ret = 0;
 	struct e100_serial *info = (struct e100_serial *)tty->driver_data;
@@ -4710,7 +4686,7 @@ rs_open(struct tty_struct *tty, struct file * filp)
  * /proc fs routines....
  */
 
-extern _INLINE_ int line_info(char *buf, struct e100_serial *info)
+static int line_info(char *buf, struct e100_serial *info)
 {
 	char	stat_buf[30];
 	int	ret;
diff --git a/drivers/serial/m32r_sio.c b/drivers/serial/m32r_sio.c
index 876bc5e027b..e9c10c0a30f 100644
--- a/drivers/serial/m32r_sio.c
+++ b/drivers/serial/m32r_sio.c
@@ -248,17 +248,17 @@ static void sio_error(int *status)
 
 #endif /* CONFIG_SERIAL_M32R_PLDSIO */
 
-static _INLINE_ unsigned int sio_in(struct uart_sio_port *up, int offset)
+static unsigned int sio_in(struct uart_sio_port *up, int offset)
 {
 	return __sio_in(up->port.iobase + offset);
 }
 
-static _INLINE_ void sio_out(struct uart_sio_port *up, int offset, int value)
+static void sio_out(struct uart_sio_port *up, int offset, int value)
 {
 	__sio_out(value, up->port.iobase + offset);
 }
 
-static _INLINE_ unsigned int serial_in(struct uart_sio_port *up, int offset)
+static unsigned int serial_in(struct uart_sio_port *up, int offset)
 {
 	if (!offset)
 		return 0;
@@ -266,8 +266,7 @@ static _INLINE_ unsigned int serial_in(struct uart_sio_port *up, int offset)
 	return __sio_in(offset);
 }
 
-static _INLINE_ void
-serial_out(struct uart_sio_port *up, int offset, int value)
+static void serial_out(struct uart_sio_port *up, int offset, int value)
 {
 	if (!offset)
 		return;
@@ -326,8 +325,8 @@ static void m32r_sio_enable_ms(struct uart_port *port)
 	serial_out(up, UART_IER, up->ier);
 }
 
-static _INLINE_ void receive_chars(struct uart_sio_port *up, int *status,
-	struct pt_regs *regs)
+static void receive_chars(struct uart_sio_port *up, int *status,
+			  struct pt_regs *regs)
 {
 	struct tty_struct *tty = up->port.info->tty;
 	unsigned char ch;
@@ -400,7 +399,7 @@ static _INLINE_ void receive_chars(struct uart_sio_port *up, int *status,
 	tty_flip_buffer_push(tty);
 }
 
-static _INLINE_ void transmit_chars(struct uart_sio_port *up)
+static void transmit_chars(struct uart_sio_port *up)
 {
 	struct circ_buf *xmit = &up->port.info->xmit;
 	int count;
diff --git a/drivers/serial/sunsu.c b/drivers/serial/sunsu.c
index 7fc3d3b41d1..9fe2283d91e 100644
--- a/drivers/serial/sunsu.c
+++ b/drivers/serial/sunsu.c
@@ -102,9 +102,7 @@ struct uart_sunsu_port {
 #endif
 };
 
-#define _INLINE_
-
-static _INLINE_ unsigned int serial_in(struct uart_sunsu_port *up, int offset)
+static unsigned int serial_in(struct uart_sunsu_port *up, int offset)
 {
 	offset <<= up->port.regshift;
 
@@ -121,8 +119,7 @@ static _INLINE_ unsigned int serial_in(struct uart_sunsu_port *up, int offset)
 	}
 }
 
-static _INLINE_ void
-serial_out(struct uart_sunsu_port *up, int offset, int value)
+static void serial_out(struct uart_sunsu_port *up, int offset, int value)
 {
 #ifndef CONFIG_SPARC64
 	/*
@@ -316,7 +313,7 @@ static void sunsu_enable_ms(struct uart_port *port)
 	spin_unlock_irqrestore(&up->port.lock, flags);
 }
 
-static _INLINE_ struct tty_struct *
+static struct tty_struct *
 receive_chars(struct uart_sunsu_port *up, unsigned char *status, struct pt_regs *regs)
 {
 	struct tty_struct *tty = up->port.info->tty;
@@ -395,7 +392,7 @@ receive_chars(struct uart_sunsu_port *up, unsigned char *status, struct pt_regs
 	return tty;
 }
 
-static _INLINE_ void transmit_chars(struct uart_sunsu_port *up)
+static void transmit_chars(struct uart_sunsu_port *up)
 {
 	struct circ_buf *xmit = &up->port.info->xmit;
 	int count;
@@ -431,7 +428,7 @@ static _INLINE_ void transmit_chars(struct uart_sunsu_port *up)
 		__stop_tx(up);
 }
 
-static _INLINE_ void check_modem_status(struct uart_sunsu_port *up)
+static void check_modem_status(struct uart_sunsu_port *up)
 {
 	int status;
 
diff --git a/drivers/tc/zs.c b/drivers/tc/zs.c
index 6756d0fab6f..2dffa8e303b 100644
--- a/drivers/tc/zs.c
+++ b/drivers/tc/zs.c
@@ -186,8 +186,6 @@ static struct tty_driver *serial_driver;
 #define RS_STROBE_TIME 10
 #define RS_ISR_PASS_LIMIT 256
 
-#define _INLINE_ inline
-
 static void probe_sccs(void);
 static void change_speed(struct dec_serial *info);
 static void rs_wait_until_sent(struct tty_struct *tty, int timeout);
@@ -344,14 +342,13 @@ static inline void rs_recv_clear(struct dec_zschannel *zsc)
  * This routine is used by the interrupt handler to schedule
  * processing in the software interrupt portion of the driver.
  */
-static _INLINE_ void rs_sched_event(struct dec_serial *info, int event)
+static void rs_sched_event(struct dec_serial *info, int event)
 {
 	info->event |= 1 << event;
 	tasklet_schedule(&info->tlet);
 }
 
-static _INLINE_ void receive_chars(struct dec_serial *info,
-				   struct pt_regs *regs)
+static void receive_chars(struct dec_serial *info, struct pt_regs *regs)
 {
 	struct tty_struct *tty = info->tty;
 	unsigned char ch, stat, flag;
@@ -441,7 +438,7 @@ static void transmit_chars(struct dec_serial *info)
 		rs_sched_event(info, RS_EVENT_WRITE_WAKEUP);
 }
 
-static _INLINE_ void status_handle(struct dec_serial *info)
+static void status_handle(struct dec_serial *info)
 {
 	unsigned char stat;
 
diff --git a/fs/9p/mux.c b/fs/9p/mux.c
index ea1134eb47c..8e8356c1c22 100644
--- a/fs/9p/mux.c
+++ b/fs/9p/mux.c
@@ -31,6 +31,7 @@
 #include <linux/poll.h>
 #include <linux/kthread.h>
 #include <linux/idr.h>
+#include <linux/mutex.h>
 
 #include "debug.h"
 #include "v9fs.h"
@@ -110,7 +111,7 @@ static void v9fs_pollwait(struct file *filp, wait_queue_head_t * wait_address,
 static u16 v9fs_mux_get_tag(struct v9fs_mux_data *);
 static void v9fs_mux_put_tag(struct v9fs_mux_data *, u16);
 
-static DECLARE_MUTEX(v9fs_mux_task_lock);
+static DEFINE_MUTEX(v9fs_mux_task_lock);
 static struct workqueue_struct *v9fs_mux_wq;
 
 static int v9fs_mux_num;
@@ -166,7 +167,7 @@ static int v9fs_mux_poll_start(struct v9fs_mux_data *m)
 
 	dprintk(DEBUG_MUX, "mux %p muxnum %d procnum %d\n", m, v9fs_mux_num,
 		v9fs_mux_poll_task_num);
-	up(&v9fs_mux_task_lock);
+	mutex_lock(&v9fs_mux_task_lock);
 
 	n = v9fs_mux_calc_poll_procs(v9fs_mux_num + 1);
 	if (n > v9fs_mux_poll_task_num) {
@@ -225,7 +226,7 @@ static int v9fs_mux_poll_start(struct v9fs_mux_data *m)
 	}
 
 	v9fs_mux_num++;
-	down(&v9fs_mux_task_lock);
+	mutex_unlock(&v9fs_mux_task_lock);
 
 	return 0;
 }
@@ -235,7 +236,7 @@ static void v9fs_mux_poll_stop(struct v9fs_mux_data *m)
 	int i;
 	struct v9fs_mux_poll_task *vpt;
 
-	up(&v9fs_mux_task_lock);
+	mutex_lock(&v9fs_mux_task_lock);
 	vpt = m->poll_task;
 	list_del(&m->mux_list);
 	for(i = 0; i < ARRAY_SIZE(m->poll_waddr); i++) {
@@ -252,7 +253,7 @@ static void v9fs_mux_poll_stop(struct v9fs_mux_data *m)
 		v9fs_mux_poll_task_num--;
 	}
 	v9fs_mux_num--;
-	down(&v9fs_mux_task_lock);
+	mutex_unlock(&v9fs_mux_task_lock);
 }
 
 /**
diff --git a/fs/adfs/file.c b/fs/adfs/file.c
index afebbfde696..6af10885f9d 100644
--- a/fs/adfs/file.c
+++ b/fs/adfs/file.c
@@ -19,11 +19,7 @@
  *
  *  adfs regular file handling primitives           
  */
-#include <linux/errno.h>
 #include <linux/fs.h>
-#include <linux/fcntl.h>
-#include <linux/time.h>
-#include <linux/stat.h>
 #include <linux/buffer_head.h>			/* for file_fsync() */
 #include <linux/adfs_fs.h>
 
diff --git a/fs/autofs4/autofs_i.h b/fs/autofs4/autofs_i.h
index 385bed09b0d..f54c5b21f87 100644
--- a/fs/autofs4/autofs_i.h
+++ b/fs/autofs4/autofs_i.h
@@ -13,6 +13,7 @@
 /* Internal header file for autofs */
 
 #include <linux/auto_fs4.h>
+#include <linux/mutex.h>
 #include <linux/list.h>
 
 /* This is the range of ioctl() numbers we claim as ours */
@@ -102,7 +103,7 @@ struct autofs_sb_info {
 	int reghost_enabled;
 	int needs_reghost;
 	struct super_block *sb;
-	struct semaphore wq_sem;
+	struct mutex wq_mutex;
 	spinlock_t fs_lock;
 	struct autofs_wait_queue *queues; /* Wait queue pointer */
 };
diff --git a/fs/autofs4/inode.c b/fs/autofs4/inode.c
index 2d3082854a2..1ad98d48e55 100644
--- a/fs/autofs4/inode.c
+++ b/fs/autofs4/inode.c
@@ -269,7 +269,7 @@ int autofs4_fill_super(struct super_block *s, void *data, int silent)
 	sbi->sb = s;
 	sbi->version = 0;
 	sbi->sub_version = 0;
-	init_MUTEX(&sbi->wq_sem);
+	mutex_init(&sbi->wq_mutex);
 	spin_lock_init(&sbi->fs_lock);
 	sbi->queues = NULL;
 	s->s_blocksize = 1024;
diff --git a/fs/autofs4/waitq.c b/fs/autofs4/waitq.c
index 394ff36ef8f..be78e9378c0 100644
--- a/fs/autofs4/waitq.c
+++ b/fs/autofs4/waitq.c
@@ -178,7 +178,7 @@ int autofs4_wait(struct autofs_sb_info *sbi, struct dentry *dentry,
 		return -ENOENT;
 	}
 
-	if (down_interruptible(&sbi->wq_sem)) {
+	if (mutex_lock_interruptible(&sbi->wq_mutex)) {
 		kfree(name);
 		return -EINTR;
 	}
@@ -194,7 +194,7 @@ int autofs4_wait(struct autofs_sb_info *sbi, struct dentry *dentry,
 		/* Can't wait for an expire if there's no mount */
 		if (notify == NFY_NONE && !d_mountpoint(dentry)) {
 			kfree(name);
-			up(&sbi->wq_sem);
+			mutex_unlock(&sbi->wq_mutex);
 			return -ENOENT;
 		}
 
@@ -202,7 +202,7 @@ int autofs4_wait(struct autofs_sb_info *sbi, struct dentry *dentry,
 		wq = kmalloc(sizeof(struct autofs_wait_queue),GFP_KERNEL);
 		if ( !wq ) {
 			kfree(name);
-			up(&sbi->wq_sem);
+			mutex_unlock(&sbi->wq_mutex);
 			return -ENOMEM;
 		}
 
@@ -218,10 +218,10 @@ int autofs4_wait(struct autofs_sb_info *sbi, struct dentry *dentry,
 		wq->status = -EINTR; /* Status return if interrupted */
 		atomic_set(&wq->wait_ctr, 2);
 		atomic_set(&wq->notified, 1);
-		up(&sbi->wq_sem);
+		mutex_unlock(&sbi->wq_mutex);
 	} else {
 		atomic_inc(&wq->wait_ctr);
-		up(&sbi->wq_sem);
+		mutex_unlock(&sbi->wq_mutex);
 		kfree(name);
 		DPRINTK("existing wait id = 0x%08lx, name = %.*s, nfy=%d",
 			(unsigned long) wq->wait_queue_token, wq->len, wq->name, notify);
@@ -282,19 +282,19 @@ int autofs4_wait_release(struct autofs_sb_info *sbi, autofs_wqt_t wait_queue_tok
 {
 	struct autofs_wait_queue *wq, **wql;
 
-	down(&sbi->wq_sem);
+	mutex_lock(&sbi->wq_mutex);
 	for ( wql = &sbi->queues ; (wq = *wql) != 0 ; wql = &wq->next ) {
 		if ( wq->wait_queue_token == wait_queue_token )
 			break;
 	}
 
 	if ( !wq ) {
-		up(&sbi->wq_sem);
+		mutex_unlock(&sbi->wq_mutex);
 		return -EINVAL;
 	}
 
 	*wql = wq->next;	/* Unlink from chain */
-	up(&sbi->wq_sem);
+	mutex_unlock(&sbi->wq_mutex);
 	kfree(wq->name);
 	wq->name = NULL;	/* Do not wait on this queue */
 
diff --git a/fs/bio.c b/fs/bio.c
index 1f3bb501c26..8f1d2e815c9 100644
--- a/fs/bio.c
+++ b/fs/bio.c
@@ -1243,11 +1243,11 @@ static int __init init_bio(void)
 		scale = 4;
 
 	/*
-	 * scale number of entries
+	 * Limit number of entries reserved -- mempools are only used when
+	 * the system is completely unable to allocate memory, so we only
+	 * need enough to make progress.
 	 */
-	bvec_pool_entries = megabytes * 2;
-	if (bvec_pool_entries > 256)
-		bvec_pool_entries = 256;
+	bvec_pool_entries = 1 + scale;
 
 	fs_bio_set = bioset_create(BIO_POOL_SIZE, bvec_pool_entries, scale);
 	if (!fs_bio_set)
diff --git a/fs/block_dev.c b/fs/block_dev.c
index 6e50346fb1e..44d05e6e34d 100644
--- a/fs/block_dev.c
+++ b/fs/block_dev.c
@@ -265,8 +265,8 @@ static void init_once(void * foo, kmem_cache_t * cachep, unsigned long flags)
 	    SLAB_CTOR_CONSTRUCTOR)
 	{
 		memset(bdev, 0, sizeof(*bdev));
-		sema_init(&bdev->bd_sem, 1);
-		sema_init(&bdev->bd_mount_sem, 1);
+		mutex_init(&bdev->bd_mutex);
+		mutex_init(&bdev->bd_mount_mutex);
 		INIT_LIST_HEAD(&bdev->bd_inodes);
 		INIT_LIST_HEAD(&bdev->bd_list);
 		inode_init_once(&ei->vfs_inode);
@@ -574,7 +574,7 @@ static int do_open(struct block_device *bdev, struct file *file)
 	}
 	owner = disk->fops->owner;
 
-	down(&bdev->bd_sem);
+	mutex_lock(&bdev->bd_mutex);
 	if (!bdev->bd_openers) {
 		bdev->bd_disk = disk;
 		bdev->bd_contains = bdev;
@@ -605,21 +605,21 @@ static int do_open(struct block_device *bdev, struct file *file)
 			if (ret)
 				goto out_first;
 			bdev->bd_contains = whole;
-			down(&whole->bd_sem);
+			mutex_lock(&whole->bd_mutex);
 			whole->bd_part_count++;
 			p = disk->part[part - 1];
 			bdev->bd_inode->i_data.backing_dev_info =
 			   whole->bd_inode->i_data.backing_dev_info;
 			if (!(disk->flags & GENHD_FL_UP) || !p || !p->nr_sects) {
 				whole->bd_part_count--;
-				up(&whole->bd_sem);
+				mutex_unlock(&whole->bd_mutex);
 				ret = -ENXIO;
 				goto out_first;
 			}
 			kobject_get(&p->kobj);
 			bdev->bd_part = p;
 			bd_set_size(bdev, (loff_t) p->nr_sects << 9);
-			up(&whole->bd_sem);
+			mutex_unlock(&whole->bd_mutex);
 		}
 	} else {
 		put_disk(disk);
@@ -633,13 +633,13 @@ static int do_open(struct block_device *bdev, struct file *file)
 			if (bdev->bd_invalidated)
 				rescan_partitions(bdev->bd_disk, bdev);
 		} else {
-			down(&bdev->bd_contains->bd_sem);
+			mutex_lock(&bdev->bd_contains->bd_mutex);
 			bdev->bd_contains->bd_part_count++;
-			up(&bdev->bd_contains->bd_sem);
+			mutex_unlock(&bdev->bd_contains->bd_mutex);
 		}
 	}
 	bdev->bd_openers++;
-	up(&bdev->bd_sem);
+	mutex_unlock(&bdev->bd_mutex);
 	unlock_kernel();
 	return 0;
 
@@ -652,7 +652,7 @@ out_first:
 	put_disk(disk);
 	module_put(owner);
 out:
-	up(&bdev->bd_sem);
+	mutex_unlock(&bdev->bd_mutex);
 	unlock_kernel();
 	if (ret)
 		bdput(bdev);
@@ -714,7 +714,7 @@ int blkdev_put(struct block_device *bdev)
 	struct inode *bd_inode = bdev->bd_inode;
 	struct gendisk *disk = bdev->bd_disk;
 
-	down(&bdev->bd_sem);
+	mutex_lock(&bdev->bd_mutex);
 	lock_kernel();
 	if (!--bdev->bd_openers) {
 		sync_blockdev(bdev);
@@ -724,9 +724,9 @@ int blkdev_put(struct block_device *bdev)
 		if (disk->fops->release)
 			ret = disk->fops->release(bd_inode, NULL);
 	} else {
-		down(&bdev->bd_contains->bd_sem);
+		mutex_lock(&bdev->bd_contains->bd_mutex);
 		bdev->bd_contains->bd_part_count--;
-		up(&bdev->bd_contains->bd_sem);
+		mutex_unlock(&bdev->bd_contains->bd_mutex);
 	}
 	if (!bdev->bd_openers) {
 		struct module *owner = disk->fops->owner;
@@ -746,7 +746,7 @@ int blkdev_put(struct block_device *bdev)
 		bdev->bd_contains = NULL;
 	}
 	unlock_kernel();
-	up(&bdev->bd_sem);
+	mutex_unlock(&bdev->bd_mutex);
 	bdput(bdev);
 	return ret;
 }
diff --git a/fs/buffer.c b/fs/buffer.c
index 1d3683d496f..0d6ca7bac6c 100644
--- a/fs/buffer.c
+++ b/fs/buffer.c
@@ -201,7 +201,7 @@ int fsync_bdev(struct block_device *bdev)
  * freeze_bdev  --  lock a filesystem and force it into a consistent state
  * @bdev:	blockdevice to lock
  *
- * This takes the block device bd_mount_sem to make sure no new mounts
+ * This takes the block device bd_mount_mutex to make sure no new mounts
  * happen on bdev until thaw_bdev() is called.
  * If a superblock is found on this device, we take the s_umount semaphore
  * on it to make sure nobody unmounts until the snapshot creation is done.
@@ -210,7 +210,7 @@ struct super_block *freeze_bdev(struct block_device *bdev)
 {
 	struct super_block *sb;
 
-	down(&bdev->bd_mount_sem);
+	mutex_lock(&bdev->bd_mount_mutex);
 	sb = get_super(bdev);
 	if (sb && !(sb->s_flags & MS_RDONLY)) {
 		sb->s_frozen = SB_FREEZE_WRITE;
@@ -264,7 +264,7 @@ void thaw_bdev(struct block_device *bdev, struct super_block *sb)
 		drop_super(sb);
 	}
 
-	up(&bdev->bd_mount_sem);
+	mutex_unlock(&bdev->bd_mount_mutex);
 }
 EXPORT_SYMBOL(thaw_bdev);
 
diff --git a/fs/cifs/dir.c b/fs/cifs/dir.c
index fed55e3c53d..632561dd9c5 100644
--- a/fs/cifs/dir.c
+++ b/fs/cifs/dir.c
@@ -138,9 +138,9 @@ cifs_create(struct inode *inode, struct dentry *direntry, int mode,
 	cifs_sb = CIFS_SB(inode->i_sb);
 	pTcon = cifs_sb->tcon;
 
-	down(&direntry->d_sb->s_vfs_rename_sem);
+	mutex_lock(&direntry->d_sb->s_vfs_rename_mutex);
 	full_path = build_path_from_dentry(direntry);
-	up(&direntry->d_sb->s_vfs_rename_sem);
+	mutex_unlock(&direntry->d_sb->s_vfs_rename_mutex);
 	if(full_path == NULL) {
 		FreeXid(xid);
 		return -ENOMEM;
@@ -317,9 +317,9 @@ int cifs_mknod(struct inode *inode, struct dentry *direntry, int mode,
 	cifs_sb = CIFS_SB(inode->i_sb);
 	pTcon = cifs_sb->tcon;
 
-	down(&direntry->d_sb->s_vfs_rename_sem);
+	mutex_lock(&direntry->d_sb->s_vfs_rename_mutex);
 	full_path = build_path_from_dentry(direntry);
-	up(&direntry->d_sb->s_vfs_rename_sem);
+	mutex_unlock(&direntry->d_sb->s_vfs_rename_mutex);
 	if(full_path == NULL)
 		rc = -ENOMEM;
 	else if (pTcon->ses->capabilities & CAP_UNIX) {
diff --git a/fs/cifs/fcntl.c b/fs/cifs/fcntl.c
index a7a47bb36bf..ec4dfe9bf5e 100644
--- a/fs/cifs/fcntl.c
+++ b/fs/cifs/fcntl.c
@@ -86,9 +86,9 @@ int cifs_dir_notify(struct file * file, unsigned long arg)
 	cifs_sb = CIFS_SB(file->f_dentry->d_sb);
 	pTcon = cifs_sb->tcon;
 
-	down(&file->f_dentry->d_sb->s_vfs_rename_sem);
+	mutex_lock(&file->f_dentry->d_sb->s_vfs_rename_mutex);
 	full_path = build_path_from_dentry(file->f_dentry);
-	up(&file->f_dentry->d_sb->s_vfs_rename_sem);
+	mutex_unlock(&file->f_dentry->d_sb->s_vfs_rename_mutex);
 
 	if(full_path == NULL) {
 		rc = -ENOMEM;
diff --git a/fs/cifs/file.c b/fs/cifs/file.c
index 675bd256829..165d6742638 100644
--- a/fs/cifs/file.c
+++ b/fs/cifs/file.c
@@ -203,9 +203,9 @@ int cifs_open(struct inode *inode, struct file *file)
 		}
 	}
 
-	down(&inode->i_sb->s_vfs_rename_sem);
+	mutex_lock(&inode->i_sb->s_vfs_rename_mutex);
 	full_path = build_path_from_dentry(file->f_dentry);
-	up(&inode->i_sb->s_vfs_rename_sem);
+	mutex_unlock(&inode->i_sb->s_vfs_rename_mutex);
 	if (full_path == NULL) {
 		FreeXid(xid);
 		return -ENOMEM;
diff --git a/fs/cifs/inode.c b/fs/cifs/inode.c
index 59359911f48..ff93a9f81d1 100644
--- a/fs/cifs/inode.c
+++ b/fs/cifs/inode.c
@@ -574,9 +574,9 @@ int cifs_unlink(struct inode *inode, struct dentry *direntry)
 
 	/* Unlink can be called from rename so we can not grab the sem here
 	   since we deadlock otherwise */
-/*	down(&direntry->d_sb->s_vfs_rename_sem);*/
+/*	mutex_lock(&direntry->d_sb->s_vfs_rename_mutex);*/
 	full_path = build_path_from_dentry(direntry);
-/*	up(&direntry->d_sb->s_vfs_rename_sem);*/
+/*	mutex_unlock(&direntry->d_sb->s_vfs_rename_mutex);*/
 	if (full_path == NULL) {
 		FreeXid(xid);
 		return -ENOMEM;
@@ -718,9 +718,9 @@ int cifs_mkdir(struct inode *inode, struct dentry *direntry, int mode)
 	cifs_sb = CIFS_SB(inode->i_sb);
 	pTcon = cifs_sb->tcon;
 
-	down(&inode->i_sb->s_vfs_rename_sem);
+	mutex_lock(&inode->i_sb->s_vfs_rename_mutex);
 	full_path = build_path_from_dentry(direntry);
-	up(&inode->i_sb->s_vfs_rename_sem);
+	mutex_unlock(&inode->i_sb->s_vfs_rename_mutex);
 	if (full_path == NULL) {
 		FreeXid(xid);
 		return -ENOMEM;
@@ -803,9 +803,9 @@ int cifs_rmdir(struct inode *inode, struct dentry *direntry)
 	cifs_sb = CIFS_SB(inode->i_sb);
 	pTcon = cifs_sb->tcon;
 
-	down(&inode->i_sb->s_vfs_rename_sem);
+	mutex_lock(&inode->i_sb->s_vfs_rename_mutex);
 	full_path = build_path_from_dentry(direntry);
-	up(&inode->i_sb->s_vfs_rename_sem);
+	mutex_unlock(&inode->i_sb->s_vfs_rename_mutex);
 	if (full_path == NULL) {
 		FreeXid(xid);
 		return -ENOMEM;
@@ -1137,9 +1137,9 @@ int cifs_setattr(struct dentry *direntry, struct iattr *attrs)
 			rc = 0;
 	}
 		
-	down(&direntry->d_sb->s_vfs_rename_sem);
+	mutex_lock(&direntry->d_sb->s_vfs_rename_mutex);
 	full_path = build_path_from_dentry(direntry);
-	up(&direntry->d_sb->s_vfs_rename_sem);
+	mutex_unlock(&direntry->d_sb->s_vfs_rename_mutex);
 	if (full_path == NULL) {
 		FreeXid(xid);
 		return -ENOMEM;
diff --git a/fs/cifs/link.c b/fs/cifs/link.c
index 0f99aae3316..8d0da7c87c7 100644
--- a/fs/cifs/link.c
+++ b/fs/cifs/link.c
@@ -48,10 +48,10 @@ cifs_hardlink(struct dentry *old_file, struct inode *inode,
 /* No need to check for cross device links since server will do that
    BB note DFS case in future though (when we may have to check) */
 
-	down(&inode->i_sb->s_vfs_rename_sem);
+	mutex_lock(&inode->i_sb->s_vfs_rename_mutex);
 	fromName = build_path_from_dentry(old_file);
 	toName = build_path_from_dentry(direntry);
-	up(&inode->i_sb->s_vfs_rename_sem);
+	mutex_unlock(&inode->i_sb->s_vfs_rename_mutex);
 	if((fromName == NULL) || (toName == NULL)) {
 		rc = -ENOMEM;
 		goto cifs_hl_exit;
@@ -103,9 +103,9 @@ cifs_follow_link(struct dentry *direntry, struct nameidata *nd)
 
 	xid = GetXid();
 
-	down(&direntry->d_sb->s_vfs_rename_sem);
+	mutex_lock(&direntry->d_sb->s_vfs_rename_mutex);
 	full_path = build_path_from_dentry(direntry);
-	up(&direntry->d_sb->s_vfs_rename_sem);
+	mutex_unlock(&direntry->d_sb->s_vfs_rename_mutex);
 
 	if (!full_path)
 		goto out_no_free;
@@ -164,9 +164,9 @@ cifs_symlink(struct inode *inode, struct dentry *direntry, const char *symname)
 	cifs_sb = CIFS_SB(inode->i_sb);
 	pTcon = cifs_sb->tcon;
 
-	down(&inode->i_sb->s_vfs_rename_sem);
+	mutex_lock(&inode->i_sb->s_vfs_rename_mutex);
 	full_path = build_path_from_dentry(direntry);
-	up(&inode->i_sb->s_vfs_rename_sem);
+	mutex_unlock(&inode->i_sb->s_vfs_rename_mutex);
 
 	if(full_path == NULL) {
 		FreeXid(xid);
@@ -232,9 +232,9 @@ cifs_readlink(struct dentry *direntry, char __user *pBuffer, int buflen)
 
 /* BB would it be safe against deadlock to grab this sem 
       even though rename itself grabs the sem and calls lookup? */
-/*       down(&inode->i_sb->s_vfs_rename_sem);*/
+/*       mutex_lock(&inode->i_sb->s_vfs_rename_mutex);*/
 	full_path = build_path_from_dentry(direntry);
-/*       up(&inode->i_sb->s_vfs_rename_sem);*/
+/*       mutex_unlock(&inode->i_sb->s_vfs_rename_mutex);*/
 
 	if(full_path == NULL) {
 		FreeXid(xid);
diff --git a/fs/cifs/readdir.c b/fs/cifs/readdir.c
index 288cc048d37..edb3b6eb34b 100644
--- a/fs/cifs/readdir.c
+++ b/fs/cifs/readdir.c
@@ -404,9 +404,9 @@ static int initiate_cifs_search(const int xid, struct file *file)
 	if(pTcon == NULL)
 		return -EINVAL;
 
-	down(&file->f_dentry->d_sb->s_vfs_rename_sem);
+	mutex_lock(&file->f_dentry->d_sb->s_vfs_rename_mutex);
 	full_path = build_path_from_dentry(file->f_dentry);
-	up(&file->f_dentry->d_sb->s_vfs_rename_sem);
+	mutex_unlock(&file->f_dentry->d_sb->s_vfs_rename_mutex);
 
 	if(full_path == NULL) {
 		return -ENOMEM;
diff --git a/fs/cifs/xattr.c b/fs/cifs/xattr.c
index 777e3363c2a..3938444d87b 100644
--- a/fs/cifs/xattr.c
+++ b/fs/cifs/xattr.c
@@ -62,9 +62,9 @@ int cifs_removexattr(struct dentry * direntry, const char * ea_name)
 	cifs_sb = CIFS_SB(sb);
 	pTcon = cifs_sb->tcon;
                                                                                      
-	down(&sb->s_vfs_rename_sem);
+	mutex_lock(&sb->s_vfs_rename_mutex);
 	full_path = build_path_from_dentry(direntry);
-	up(&sb->s_vfs_rename_sem);
+	mutex_unlock(&sb->s_vfs_rename_mutex);
 	if(full_path == NULL) {
 		FreeXid(xid);
 		return -ENOMEM;
@@ -116,9 +116,9 @@ int cifs_setxattr(struct dentry * direntry, const char * ea_name,
 	cifs_sb = CIFS_SB(sb);
 	pTcon = cifs_sb->tcon;
 
-	down(&sb->s_vfs_rename_sem);
+	mutex_lock(&sb->s_vfs_rename_mutex);
 	full_path = build_path_from_dentry(direntry);
-	up(&sb->s_vfs_rename_sem);
+	mutex_unlock(&sb->s_vfs_rename_mutex);
 	if(full_path == NULL) {
 		FreeXid(xid);
 		return -ENOMEM;
@@ -223,9 +223,9 @@ ssize_t cifs_getxattr(struct dentry * direntry, const char * ea_name,
 	cifs_sb = CIFS_SB(sb);
 	pTcon = cifs_sb->tcon;
 
-	down(&sb->s_vfs_rename_sem);
+	mutex_lock(&sb->s_vfs_rename_mutex);
 	full_path = build_path_from_dentry(direntry);
-	up(&sb->s_vfs_rename_sem);
+	mutex_unlock(&sb->s_vfs_rename_mutex);
 	if(full_path == NULL) {
 		FreeXid(xid);
 		return -ENOMEM;
@@ -341,9 +341,9 @@ ssize_t cifs_listxattr(struct dentry * direntry, char * data, size_t buf_size)
 	cifs_sb = CIFS_SB(sb);
 	pTcon = cifs_sb->tcon;
 
-	down(&sb->s_vfs_rename_sem);
+	mutex_lock(&sb->s_vfs_rename_mutex);
 	full_path = build_path_from_dentry(direntry);
-	up(&sb->s_vfs_rename_sem);
+	mutex_unlock(&sb->s_vfs_rename_mutex);
 	if(full_path == NULL) {
 		FreeXid(xid);
 		return -ENOMEM;
diff --git a/fs/devpts/inode.c b/fs/devpts/inode.c
index bfb8a230bac..14c5620b5ca 100644
--- a/fs/devpts/inode.c
+++ b/fs/devpts/inode.c
@@ -18,6 +18,7 @@
 #include <linux/mount.h>
 #include <linux/tty.h>
 #include <linux/devpts_fs.h>
+#include <linux/parser.h>
 
 #define DEVPTS_SUPER_MAGIC 0x1cd1
 
@@ -32,39 +33,60 @@ static struct {
 	umode_t mode;
 } config = {.mode = 0600};
 
+enum {
+	Opt_uid, Opt_gid, Opt_mode,
+	Opt_err
+};
+
+static match_table_t tokens = {
+	{Opt_uid, "uid=%u"},
+	{Opt_gid, "gid=%u"},
+	{Opt_mode, "mode=%o"},
+	{Opt_err, NULL}
+};
+
 static int devpts_remount(struct super_block *sb, int *flags, char *data)
 {
-	int setuid = 0;
-	int setgid = 0;
-	uid_t uid = 0;
-	gid_t gid = 0;
-	umode_t mode = 0600;
-	char *this_char;
-
-	this_char = NULL;
-	while ((this_char = strsep(&data, ",")) != NULL) {
-		int n;
-		char dummy;
-		if (!*this_char)
+	char *p;
+
+	config.setuid  = 0;
+	config.setgid  = 0;
+	config.uid     = 0;
+	config.gid     = 0;
+	config.mode    = 0600;
+
+	while ((p = strsep(&data, ",")) != NULL) {
+		substring_t args[MAX_OPT_ARGS];
+		int token;
+		int option;
+
+		if (!*p)
 			continue;
-		if (sscanf(this_char, "uid=%i%c", &n, &dummy) == 1) {
-			setuid = 1;
-			uid = n;
-		} else if (sscanf(this_char, "gid=%i%c", &n, &dummy) == 1) {
-			setgid = 1;
-			gid = n;
-		} else if (sscanf(this_char, "mode=%o%c", &n, &dummy) == 1)
-			mode = n & ~S_IFMT;
-		else {
-			printk("devpts: called with bogus options\n");
+
+		token = match_token(p, tokens, args);
+		switch (token) {
+		case Opt_uid:
+			if (match_int(&args[0], &option))
+				return -EINVAL;
+			config.uid = option;
+			config.setuid = 1;
+			break;
+		case Opt_gid:
+			if (match_int(&args[0], &option))
+				return -EINVAL;
+			config.gid = option;
+			config.setgid = 1;
+			break;
+		case Opt_mode:
+			if (match_octal(&args[0], &option))
+				return -EINVAL;
+			config.mode = option & ~S_IFMT;
+			break;
+		default:
+			printk(KERN_ERR "devpts: called with bogus options\n");
 			return -EINVAL;
 		}
 	}
-	config.setuid  = setuid;
-	config.setgid  = setgid;
-	config.uid     = uid;
-	config.gid     = gid;
-	config.mode    = mode;
 
 	return 0;
 }
diff --git a/fs/dquot.c b/fs/dquot.c
index 1966c890b48..acf07e581f8 100644
--- a/fs/dquot.c
+++ b/fs/dquot.c
@@ -103,12 +103,12 @@
  * (these locking rules also apply for S_NOQUOTA flag in the inode - note that
  * for altering the flag i_mutex is also needed).  If operation is holding
  * reference to dquot in other way (e.g. quotactl ops) it must be guarded by
- * dqonoff_sem.
+ * dqonoff_mutex.
  * This locking assures that:
  *   a) update/access to dquot pointers in inode is serialized
  *   b) everyone is guarded against invalidate_dquots()
  *
- * Each dquot has its dq_lock semaphore. Locked dquots might not be referenced
+ * Each dquot has its dq_lock mutex. Locked dquots might not be referenced
  * from inodes (dquot_alloc_space() and such don't check the dq_lock).
  * Currently dquot is locked only when it is being read to memory (or space for
  * it is being allocated) on the first dqget() and when it is being released on
@@ -118,9 +118,9 @@
  * spinlock to internal buffers before writing.
  *
  * Lock ordering (including related VFS locks) is the following:
- *   i_mutex > dqonoff_sem > iprune_sem > journal_lock > dqptr_sem >
- *   > dquot->dq_lock > dqio_sem
- * i_mutex on quota files is special (it's below dqio_sem)
+ *   i_mutex > dqonoff_sem > journal_lock > dqptr_sem > dquot->dq_lock >
+ *   dqio_mutex
+ * i_mutex on quota files is special (it's below dqio_mutex)
  */
 
 static DEFINE_SPINLOCK(dq_list_lock);
@@ -281,8 +281,8 @@ static inline void remove_inuse(struct dquot *dquot)
 
 static void wait_on_dquot(struct dquot *dquot)
 {
-	down(&dquot->dq_lock);
-	up(&dquot->dq_lock);
+	mutex_lock(&dquot->dq_lock);
+	mutex_unlock(&dquot->dq_lock);
 }
 
 #define mark_dquot_dirty(dquot) ((dquot)->dq_sb->dq_op->mark_dirty(dquot))
@@ -321,8 +321,8 @@ int dquot_acquire(struct dquot *dquot)
 	int ret = 0, ret2 = 0;
 	struct quota_info *dqopt = sb_dqopt(dquot->dq_sb);
 
-	down(&dquot->dq_lock);
-	down(&dqopt->dqio_sem);
+	mutex_lock(&dquot->dq_lock);
+	mutex_lock(&dqopt->dqio_mutex);
 	if (!test_bit(DQ_READ_B, &dquot->dq_flags))
 		ret = dqopt->ops[dquot->dq_type]->read_dqblk(dquot);
 	if (ret < 0)
@@ -343,8 +343,8 @@ int dquot_acquire(struct dquot *dquot)
 	}
 	set_bit(DQ_ACTIVE_B, &dquot->dq_flags);
 out_iolock:
-	up(&dqopt->dqio_sem);
-	up(&dquot->dq_lock);
+	mutex_unlock(&dqopt->dqio_mutex);
+	mutex_unlock(&dquot->dq_lock);
 	return ret;
 }
 
@@ -356,7 +356,7 @@ int dquot_commit(struct dquot *dquot)
 	int ret = 0, ret2 = 0;
 	struct quota_info *dqopt = sb_dqopt(dquot->dq_sb);
 
-	down(&dqopt->dqio_sem);
+	mutex_lock(&dqopt->dqio_mutex);
 	spin_lock(&dq_list_lock);
 	if (!clear_dquot_dirty(dquot)) {
 		spin_unlock(&dq_list_lock);
@@ -373,7 +373,7 @@ int dquot_commit(struct dquot *dquot)
 			ret = ret2;
 	}
 out_sem:
-	up(&dqopt->dqio_sem);
+	mutex_unlock(&dqopt->dqio_mutex);
 	return ret;
 }
 
@@ -385,11 +385,11 @@ int dquot_release(struct dquot *dquot)
 	int ret = 0, ret2 = 0;
 	struct quota_info *dqopt = sb_dqopt(dquot->dq_sb);
 
-	down(&dquot->dq_lock);
+	mutex_lock(&dquot->dq_lock);
 	/* Check whether we are not racing with some other dqget() */
 	if (atomic_read(&dquot->dq_count) > 1)
 		goto out_dqlock;
-	down(&dqopt->dqio_sem);
+	mutex_lock(&dqopt->dqio_mutex);
 	if (dqopt->ops[dquot->dq_type]->release_dqblk) {
 		ret = dqopt->ops[dquot->dq_type]->release_dqblk(dquot);
 		/* Write the info */
@@ -399,31 +399,57 @@ int dquot_release(struct dquot *dquot)
 			ret = ret2;
 	}
 	clear_bit(DQ_ACTIVE_B, &dquot->dq_flags);
-	up(&dqopt->dqio_sem);
+	mutex_unlock(&dqopt->dqio_mutex);
 out_dqlock:
-	up(&dquot->dq_lock);
+	mutex_unlock(&dquot->dq_lock);
 	return ret;
 }
 
 /* Invalidate all dquots on the list. Note that this function is called after
  * quota is disabled and pointers from inodes removed so there cannot be new
- * quota users. Also because we hold dqonoff_sem there can be no quota users
- * for this sb+type at all. */
+ * quota users. There can still be some users of quotas due to inodes being
+ * just deleted or pruned by prune_icache() (those are not attached to any
+ * list). We have to wait for such users.
+ */
 static void invalidate_dquots(struct super_block *sb, int type)
 {
 	struct dquot *dquot, *tmp;
 
+restart:
 	spin_lock(&dq_list_lock);
 	list_for_each_entry_safe(dquot, tmp, &inuse_list, dq_inuse) {
 		if (dquot->dq_sb != sb)
 			continue;
 		if (dquot->dq_type != type)
 			continue;
-#ifdef __DQUOT_PARANOIA
-		if (atomic_read(&dquot->dq_count))
-			BUG();
-#endif
-		/* Quota now has no users and it has been written on last dqput() */
+		/* Wait for dquot users */
+		if (atomic_read(&dquot->dq_count)) {
+			DEFINE_WAIT(wait);
+
+			atomic_inc(&dquot->dq_count);
+			prepare_to_wait(&dquot->dq_wait_unused, &wait,
+					TASK_UNINTERRUPTIBLE);
+			spin_unlock(&dq_list_lock);
+			/* Once dqput() wakes us up, we know it's time to free
+			 * the dquot.
+			 * IMPORTANT: we rely on the fact that there is always
+			 * at most one process waiting for dquot to free.
+			 * Otherwise dq_count would be > 1 and we would never
+			 * wake up.
+			 */
+			if (atomic_read(&dquot->dq_count) > 1)
+				schedule();
+			finish_wait(&dquot->dq_wait_unused, &wait);
+			dqput(dquot);
+			/* At this moment dquot() need not exist (it could be
+			 * reclaimed by prune_dqcache(). Hence we must
+			 * restart. */
+			goto restart;
+		}
+		/*
+		 * Quota now has no users and it has been written on last
+		 * dqput()
+		 */
 		remove_dquot_hash(dquot);
 		remove_free_dquot(dquot);
 		remove_inuse(dquot);
@@ -439,7 +465,7 @@ int vfs_quota_sync(struct super_block *sb, int type)
 	struct quota_info *dqopt = sb_dqopt(sb);
 	int cnt;
 
-	down(&dqopt->dqonoff_sem);
+	mutex_lock(&dqopt->dqonoff_mutex);
 	for (cnt = 0; cnt < MAXQUOTAS; cnt++) {
 		if (type != -1 && cnt != type)
 			continue;
@@ -474,7 +500,7 @@ int vfs_quota_sync(struct super_block *sb, int type)
 	spin_lock(&dq_list_lock);
 	dqstats.syncs++;
 	spin_unlock(&dq_list_lock);
-	up(&dqopt->dqonoff_sem);
+	mutex_unlock(&dqopt->dqonoff_mutex);
 
 	return 0;
 }
@@ -515,7 +541,7 @@ static int shrink_dqcache_memory(int nr, gfp_t gfp_mask)
 /*
  * Put reference to dquot
  * NOTE: If you change this function please check whether dqput_blocks() works right...
- * MUST be called with either dqptr_sem or dqonoff_sem held
+ * MUST be called with either dqptr_sem or dqonoff_mutex held
  */
 static void dqput(struct dquot *dquot)
 {
@@ -540,6 +566,10 @@ we_slept:
 	if (atomic_read(&dquot->dq_count) > 1) {
 		/* We have more than one user... nothing to do */
 		atomic_dec(&dquot->dq_count);
+		/* Releasing dquot during quotaoff phase? */
+		if (!sb_has_quota_enabled(dquot->dq_sb, dquot->dq_type) &&
+		    atomic_read(&dquot->dq_count) == 1)
+			wake_up(&dquot->dq_wait_unused);
 		spin_unlock(&dq_list_lock);
 		return;
 	}
@@ -576,11 +606,12 @@ static struct dquot *get_empty_dquot(struct super_block *sb, int type)
 		return NODQUOT;
 
 	memset((caddr_t)dquot, 0, sizeof(struct dquot));
-	sema_init(&dquot->dq_lock, 1);
+	mutex_init(&dquot->dq_lock);
 	INIT_LIST_HEAD(&dquot->dq_free);
 	INIT_LIST_HEAD(&dquot->dq_inuse);
 	INIT_HLIST_NODE(&dquot->dq_hash);
 	INIT_LIST_HEAD(&dquot->dq_dirty);
+	init_waitqueue_head(&dquot->dq_wait_unused);
 	dquot->dq_sb = sb;
 	dquot->dq_type = type;
 	atomic_set(&dquot->dq_count, 1);
@@ -590,7 +621,7 @@ static struct dquot *get_empty_dquot(struct super_block *sb, int type)
 
 /*
  * Get reference to dquot
- * MUST be called with either dqptr_sem or dqonoff_sem held
+ * MUST be called with either dqptr_sem or dqonoff_mutex held
  */
 static struct dquot *dqget(struct super_block *sb, unsigned int id, int type)
 {
@@ -656,7 +687,7 @@ static int dqinit_needed(struct inode *inode, int type)
 	return 0;
 }
 
-/* This routine is guarded by dqonoff_sem semaphore */
+/* This routine is guarded by dqonoff_mutex mutex */
 static void add_dquot_ref(struct super_block *sb, int type)
 {
 	struct list_head *p;
@@ -732,13 +763,9 @@ static void drop_dquot_ref(struct super_block *sb, int type)
 {
 	LIST_HEAD(tofree_head);
 
-	/* We need to be guarded against prune_icache to reach all the
-	 * inodes - otherwise some can be on the local list of prune_icache */
-	down(&iprune_sem);
 	down_write(&sb_dqopt(sb)->dqptr_sem);
 	remove_dquot_ref(sb, type, &tofree_head);
 	up_write(&sb_dqopt(sb)->dqptr_sem);
-	up(&iprune_sem);
 	put_dquot_list(&tofree_head);
 }
 
@@ -938,8 +965,8 @@ int dquot_initialize(struct inode *inode, int type)
 	unsigned int id = 0;
 	int cnt, ret = 0;
 
-	/* First test before acquiring semaphore - solves deadlocks when we
-         * re-enter the quota code and are already holding the semaphore */
+	/* First test before acquiring mutex - solves deadlocks when we
+         * re-enter the quota code and are already holding the mutex */
 	if (IS_NOQUOTA(inode))
 		return 0;
 	down_write(&sb_dqopt(inode->i_sb)->dqptr_sem);
@@ -1002,8 +1029,8 @@ int dquot_alloc_space(struct inode *inode, qsize_t number, int warn)
 	int cnt, ret = NO_QUOTA;
 	char warntype[MAXQUOTAS];
 
-	/* First test before acquiring semaphore - solves deadlocks when we
-         * re-enter the quota code and are already holding the semaphore */
+	/* First test before acquiring mutex - solves deadlocks when we
+         * re-enter the quota code and are already holding the mutex */
 	if (IS_NOQUOTA(inode)) {
 out_add:
 		inode_add_bytes(inode, number);
@@ -1051,8 +1078,8 @@ int dquot_alloc_inode(const struct inode *inode, unsigned long number)
 	int cnt, ret = NO_QUOTA;
 	char warntype[MAXQUOTAS];
 
-	/* First test before acquiring semaphore - solves deadlocks when we
-         * re-enter the quota code and are already holding the semaphore */
+	/* First test before acquiring mutex - solves deadlocks when we
+         * re-enter the quota code and are already holding the mutex */
 	if (IS_NOQUOTA(inode))
 		return QUOTA_OK;
 	for (cnt = 0; cnt < MAXQUOTAS; cnt++)
@@ -1095,8 +1122,8 @@ int dquot_free_space(struct inode *inode, qsize_t number)
 {
 	unsigned int cnt;
 
-	/* First test before acquiring semaphore - solves deadlocks when we
-         * re-enter the quota code and are already holding the semaphore */
+	/* First test before acquiring mutex - solves deadlocks when we
+         * re-enter the quota code and are already holding the mutex */
 	if (IS_NOQUOTA(inode)) {
 out_sub:
 		inode_sub_bytes(inode, number);
@@ -1131,8 +1158,8 @@ int dquot_free_inode(const struct inode *inode, unsigned long number)
 {
 	unsigned int cnt;
 
-	/* First test before acquiring semaphore - solves deadlocks when we
-         * re-enter the quota code and are already holding the semaphore */
+	/* First test before acquiring mutex - solves deadlocks when we
+         * re-enter the quota code and are already holding the mutex */
 	if (IS_NOQUOTA(inode))
 		return QUOTA_OK;
 	down_read(&sb_dqopt(inode->i_sb)->dqptr_sem);
@@ -1171,8 +1198,8 @@ int dquot_transfer(struct inode *inode, struct iattr *iattr)
 	    chgid = (iattr->ia_valid & ATTR_GID) && inode->i_gid != iattr->ia_gid;
 	char warntype[MAXQUOTAS];
 
-	/* First test before acquiring semaphore - solves deadlocks when we
-         * re-enter the quota code and are already holding the semaphore */
+	/* First test before acquiring mutex - solves deadlocks when we
+         * re-enter the quota code and are already holding the mutex */
 	if (IS_NOQUOTA(inode))
 		return QUOTA_OK;
 	/* Clear the arrays */
@@ -1266,9 +1293,9 @@ int dquot_commit_info(struct super_block *sb, int type)
 	int ret;
 	struct quota_info *dqopt = sb_dqopt(sb);
 
-	down(&dqopt->dqio_sem);
+	mutex_lock(&dqopt->dqio_mutex);
 	ret = dqopt->ops[type]->write_file_info(sb, type);
-	up(&dqopt->dqio_sem);
+	mutex_unlock(&dqopt->dqio_mutex);
 	return ret;
 }
 
@@ -1324,7 +1351,7 @@ int vfs_quota_off(struct super_block *sb, int type)
 	struct inode *toputinode[MAXQUOTAS];
 
 	/* We need to serialize quota_off() for device */
-	down(&dqopt->dqonoff_sem);
+	mutex_lock(&dqopt->dqonoff_mutex);
 	for (cnt = 0; cnt < MAXQUOTAS; cnt++) {
 		toputinode[cnt] = NULL;
 		if (type != -1 && cnt != type)
@@ -1353,7 +1380,7 @@ int vfs_quota_off(struct super_block *sb, int type)
 		dqopt->info[cnt].dqi_bgrace = 0;
 		dqopt->ops[cnt] = NULL;
 	}
-	up(&dqopt->dqonoff_sem);
+	mutex_unlock(&dqopt->dqonoff_mutex);
 	/* Sync the superblock so that buffers with quota data are written to
 	 * disk (and so userspace sees correct data afterwards). */
 	if (sb->s_op->sync_fs)
@@ -1366,7 +1393,7 @@ int vfs_quota_off(struct super_block *sb, int type)
 	 * changes done by userspace on the next quotaon() */
 	for (cnt = 0; cnt < MAXQUOTAS; cnt++)
 		if (toputinode[cnt]) {
-			down(&dqopt->dqonoff_sem);
+			mutex_lock(&dqopt->dqonoff_mutex);
 			/* If quota was reenabled in the meantime, we have
 			 * nothing to do */
 			if (!sb_has_quota_enabled(sb, cnt)) {
@@ -1378,7 +1405,7 @@ int vfs_quota_off(struct super_block *sb, int type)
 				mark_inode_dirty(toputinode[cnt]);
 				iput(toputinode[cnt]);
 			}
-			up(&dqopt->dqonoff_sem);
+			mutex_unlock(&dqopt->dqonoff_mutex);
 		}
 	if (sb->s_bdev)
 		invalidate_bdev(sb->s_bdev, 0);
@@ -1419,7 +1446,7 @@ static int vfs_quota_on_inode(struct inode *inode, int type, int format_id)
 	/* And now flush the block cache so that kernel sees the changes */
 	invalidate_bdev(sb->s_bdev, 0);
 	mutex_lock(&inode->i_mutex);
-	down(&dqopt->dqonoff_sem);
+	mutex_lock(&dqopt->dqonoff_mutex);
 	if (sb_has_quota_enabled(sb, type)) {
 		error = -EBUSY;
 		goto out_lock;
@@ -1444,17 +1471,17 @@ static int vfs_quota_on_inode(struct inode *inode, int type, int format_id)
 	dqopt->ops[type] = fmt->qf_ops;
 	dqopt->info[type].dqi_format = fmt;
 	INIT_LIST_HEAD(&dqopt->info[type].dqi_dirty_list);
-	down(&dqopt->dqio_sem);
+	mutex_lock(&dqopt->dqio_mutex);
 	if ((error = dqopt->ops[type]->read_file_info(sb, type)) < 0) {
-		up(&dqopt->dqio_sem);
+		mutex_unlock(&dqopt->dqio_mutex);
 		goto out_file_init;
 	}
-	up(&dqopt->dqio_sem);
+	mutex_unlock(&dqopt->dqio_mutex);
 	mutex_unlock(&inode->i_mutex);
 	set_enable_flags(dqopt, type);
 
 	add_dquot_ref(sb, type);
-	up(&dqopt->dqonoff_sem);
+	mutex_unlock(&dqopt->dqonoff_mutex);
 
 	return 0;
 
@@ -1462,7 +1489,7 @@ out_file_init:
 	dqopt->files[type] = NULL;
 	iput(inode);
 out_lock:
-	up(&dqopt->dqonoff_sem);
+	mutex_unlock(&dqopt->dqonoff_mutex);
 	if (oldflags != -1) {
 		down_write(&dqopt->dqptr_sem);
 		/* Set the flags back (in the case of accidental quotaon()
@@ -1550,14 +1577,14 @@ int vfs_get_dqblk(struct super_block *sb, int type, qid_t id, struct if_dqblk *d
 {
 	struct dquot *dquot;
 
-	down(&sb_dqopt(sb)->dqonoff_sem);
+	mutex_lock(&sb_dqopt(sb)->dqonoff_mutex);
 	if (!(dquot = dqget(sb, id, type))) {
-		up(&sb_dqopt(sb)->dqonoff_sem);
+		mutex_unlock(&sb_dqopt(sb)->dqonoff_mutex);
 		return -ESRCH;
 	}
 	do_get_dqblk(dquot, di);
 	dqput(dquot);
-	up(&sb_dqopt(sb)->dqonoff_sem);
+	mutex_unlock(&sb_dqopt(sb)->dqonoff_mutex);
 	return 0;
 }
 
@@ -1619,14 +1646,14 @@ int vfs_set_dqblk(struct super_block *sb, int type, qid_t id, struct if_dqblk *d
 {
 	struct dquot *dquot;
 
-	down(&sb_dqopt(sb)->dqonoff_sem);
+	mutex_lock(&sb_dqopt(sb)->dqonoff_mutex);
 	if (!(dquot = dqget(sb, id, type))) {
-		up(&sb_dqopt(sb)->dqonoff_sem);
+		mutex_unlock(&sb_dqopt(sb)->dqonoff_mutex);
 		return -ESRCH;
 	}
 	do_set_dqblk(dquot, di);
 	dqput(dquot);
-	up(&sb_dqopt(sb)->dqonoff_sem);
+	mutex_unlock(&sb_dqopt(sb)->dqonoff_mutex);
 	return 0;
 }
 
@@ -1635,9 +1662,9 @@ int vfs_get_dqinfo(struct super_block *sb, int type, struct if_dqinfo *ii)
 {
 	struct mem_dqinfo *mi;
   
-	down(&sb_dqopt(sb)->dqonoff_sem);
+	mutex_lock(&sb_dqopt(sb)->dqonoff_mutex);
 	if (!sb_has_quota_enabled(sb, type)) {
-		up(&sb_dqopt(sb)->dqonoff_sem);
+		mutex_unlock(&sb_dqopt(sb)->dqonoff_mutex);
 		return -ESRCH;
 	}
 	mi = sb_dqopt(sb)->info + type;
@@ -1647,7 +1674,7 @@ int vfs_get_dqinfo(struct super_block *sb, int type, struct if_dqinfo *ii)
 	ii->dqi_flags = mi->dqi_flags & DQF_MASK;
 	ii->dqi_valid = IIF_ALL;
 	spin_unlock(&dq_data_lock);
-	up(&sb_dqopt(sb)->dqonoff_sem);
+	mutex_unlock(&sb_dqopt(sb)->dqonoff_mutex);
 	return 0;
 }
 
@@ -1656,9 +1683,9 @@ int vfs_set_dqinfo(struct super_block *sb, int type, struct if_dqinfo *ii)
 {
 	struct mem_dqinfo *mi;
 
-	down(&sb_dqopt(sb)->dqonoff_sem);
+	mutex_lock(&sb_dqopt(sb)->dqonoff_mutex);
 	if (!sb_has_quota_enabled(sb, type)) {
-		up(&sb_dqopt(sb)->dqonoff_sem);
+		mutex_unlock(&sb_dqopt(sb)->dqonoff_mutex);
 		return -ESRCH;
 	}
 	mi = sb_dqopt(sb)->info + type;
@@ -1673,7 +1700,7 @@ int vfs_set_dqinfo(struct super_block *sb, int type, struct if_dqinfo *ii)
 	mark_info_dirty(sb, type);
 	/* Force write to disk */
 	sb->dq_op->write_info(sb, type);
-	up(&sb_dqopt(sb)->dqonoff_sem);
+	mutex_unlock(&sb_dqopt(sb)->dqonoff_mutex);
 	return 0;
 }
 
diff --git a/fs/eventpoll.c b/fs/eventpoll.c
index 4284cd31eba..1c2b16fda13 100644
--- a/fs/eventpoll.c
+++ b/fs/eventpoll.c
@@ -34,6 +34,7 @@
 #include <linux/eventpoll.h>
 #include <linux/mount.h>
 #include <linux/bitops.h>
+#include <linux/mutex.h>
 #include <asm/uaccess.h>
 #include <asm/system.h>
 #include <asm/io.h>
@@ -46,7 +47,7 @@
  * LOCKING:
  * There are three level of locking required by epoll :
  *
- * 1) epsem (semaphore)
+ * 1) epmutex (mutex)
  * 2) ep->sem (rw_semaphore)
  * 3) ep->lock (rw_lock)
  *
@@ -67,9 +68,9 @@
  * if a file has been pushed inside an epoll set and it is then
  * close()d without a previous call toepoll_ctl(EPOLL_CTL_DEL).
  * It is possible to drop the "ep->sem" and to use the global
- * semaphore "epsem" (together with "ep->lock") to have it working,
+ * semaphore "epmutex" (together with "ep->lock") to have it working,
  * but having "ep->sem" will make the interface more scalable.
- * Events that require holding "epsem" are very rare, while for
+ * Events that require holding "epmutex" are very rare, while for
  * normal operations the epoll private "ep->sem" will guarantee
  * a greater scalability.
  */
@@ -274,7 +275,7 @@ static struct super_block *eventpollfs_get_sb(struct file_system_type *fs_type,
 /*
  * This semaphore is used to serialize ep_free() and eventpoll_release_file().
  */
-static struct semaphore epsem;
+static struct mutex epmutex;
 
 /* Safe wake up implementation */
 static struct poll_safewake psw;
@@ -451,15 +452,6 @@ static void ep_poll_safewake(struct poll_safewake *psw, wait_queue_head_t *wq)
 }
 
 
-/* Used to initialize the epoll bits inside the "struct file" */
-void eventpoll_init_file(struct file *file)
-{
-
-	INIT_LIST_HEAD(&file->f_ep_links);
-	spin_lock_init(&file->f_ep_lock);
-}
-
-
 /*
  * This is called from eventpoll_release() to unlink files from the eventpoll
  * interface. We need to have this facility to cleanup correctly files that are
@@ -477,10 +469,10 @@ void eventpoll_release_file(struct file *file)
 	 * cleanup path, and this means that noone is using this file anymore.
 	 * The only hit might come from ep_free() but by holding the semaphore
 	 * will correctly serialize the operation. We do need to acquire
-	 * "ep->sem" after "epsem" because ep_remove() requires it when called
+	 * "ep->sem" after "epmutex" because ep_remove() requires it when called
 	 * from anywhere but ep_free().
 	 */
-	down(&epsem);
+	mutex_lock(&epmutex);
 
 	while (!list_empty(lsthead)) {
 		epi = list_entry(lsthead->next, struct epitem, fllink);
@@ -492,7 +484,7 @@ void eventpoll_release_file(struct file *file)
 		up_write(&ep->sem);
 	}
 
-	up(&epsem);
+	mutex_unlock(&epmutex);
 }
 
 
@@ -819,9 +811,9 @@ static void ep_free(struct eventpoll *ep)
 	 * We do not need to hold "ep->sem" here because the epoll file
 	 * is on the way to be removed and no one has references to it
 	 * anymore. The only hit might come from eventpoll_release_file() but
-	 * holding "epsem" is sufficent here.
+	 * holding "epmutex" is sufficent here.
 	 */
-	down(&epsem);
+	mutex_lock(&epmutex);
 
 	/*
 	 * Walks through the whole tree by unregistering poll callbacks.
@@ -843,7 +835,7 @@ static void ep_free(struct eventpoll *ep)
 		ep_remove(ep, epi);
 	}
 
-	up(&epsem);
+	mutex_unlock(&epmutex);
 }
 
 
@@ -1615,7 +1607,7 @@ static int __init eventpoll_init(void)
 {
 	int error;
 
-	init_MUTEX(&epsem);
+	mutex_init(&epmutex);
 
 	/* Initialize the structure used to perform safe poll wait head wake ups */
 	ep_poll_safewake_init(&psw);
diff --git a/fs/ext2/namei.c b/fs/ext2/namei.c
index ad1432a2a62..4ca82498532 100644
--- a/fs/ext2/namei.c
+++ b/fs/ext2/namei.c
@@ -36,22 +36,6 @@
 #include "acl.h"
 #include "xip.h"
 
-/*
- * Couple of helper functions - make the code slightly cleaner.
- */
-
-static inline void ext2_inc_count(struct inode *inode)
-{
-	inode->i_nlink++;
-	mark_inode_dirty(inode);
-}
-
-static inline void ext2_dec_count(struct inode *inode)
-{
-	inode->i_nlink--;
-	mark_inode_dirty(inode);
-}
-
 static inline int ext2_add_nondir(struct dentry *dentry, struct inode *inode)
 {
 	int err = ext2_add_link(dentry, inode);
@@ -59,7 +43,7 @@ static inline int ext2_add_nondir(struct dentry *dentry, struct inode *inode)
 		d_instantiate(dentry, inode);
 		return 0;
 	}
-	ext2_dec_count(inode);
+	inode_dec_link_count(inode);
 	iput(inode);
 	return err;
 }
@@ -201,7 +185,7 @@ out:
 	return err;
 
 out_fail:
-	ext2_dec_count(inode);
+	inode_dec_link_count(inode);
 	iput (inode);
 	goto out;
 }
@@ -215,7 +199,7 @@ static int ext2_link (struct dentry * old_dentry, struct inode * dir,
 		return -EMLINK;
 
 	inode->i_ctime = CURRENT_TIME_SEC;
-	ext2_inc_count(inode);
+	inode_inc_link_count(inode);
 	atomic_inc(&inode->i_count);
 
 	return ext2_add_nondir(dentry, inode);
@@ -229,7 +213,7 @@ static int ext2_mkdir(struct inode * dir, struct dentry * dentry, int mode)
 	if (dir->i_nlink >= EXT2_LINK_MAX)
 		goto out;
 
-	ext2_inc_count(dir);
+	inode_inc_link_count(dir);
 
 	inode = ext2_new_inode (dir, S_IFDIR | mode);
 	err = PTR_ERR(inode);
@@ -243,7 +227,7 @@ static int ext2_mkdir(struct inode * dir, struct dentry * dentry, int mode)
 	else
 		inode->i_mapping->a_ops = &ext2_aops;
 
-	ext2_inc_count(inode);
+	inode_inc_link_count(inode);
 
 	err = ext2_make_empty(inode, dir);
 	if (err)
@@ -258,11 +242,11 @@ out:
 	return err;
 
 out_fail:
-	ext2_dec_count(inode);
-	ext2_dec_count(inode);
+	inode_dec_link_count(inode);
+	inode_dec_link_count(inode);
 	iput(inode);
 out_dir:
-	ext2_dec_count(dir);
+	inode_dec_link_count(dir);
 	goto out;
 }
 
@@ -282,7 +266,7 @@ static int ext2_unlink(struct inode * dir, struct dentry *dentry)
 		goto out;
 
 	inode->i_ctime = dir->i_ctime;
-	ext2_dec_count(inode);
+	inode_dec_link_count(inode);
 	err = 0;
 out:
 	return err;
@@ -297,8 +281,8 @@ static int ext2_rmdir (struct inode * dir, struct dentry *dentry)
 		err = ext2_unlink(dir, dentry);
 		if (!err) {
 			inode->i_size = 0;
-			ext2_dec_count(inode);
-			ext2_dec_count(dir);
+			inode_dec_link_count(inode);
+			inode_dec_link_count(dir);
 		}
 	}
 	return err;
@@ -338,41 +322,41 @@ static int ext2_rename (struct inode * old_dir, struct dentry * old_dentry,
 		new_de = ext2_find_entry (new_dir, new_dentry, &new_page);
 		if (!new_de)
 			goto out_dir;
-		ext2_inc_count(old_inode);
+		inode_inc_link_count(old_inode);
 		ext2_set_link(new_dir, new_de, new_page, old_inode);
 		new_inode->i_ctime = CURRENT_TIME_SEC;
 		if (dir_de)
 			new_inode->i_nlink--;
-		ext2_dec_count(new_inode);
+		inode_dec_link_count(new_inode);
 	} else {
 		if (dir_de) {
 			err = -EMLINK;
 			if (new_dir->i_nlink >= EXT2_LINK_MAX)
 				goto out_dir;
 		}
-		ext2_inc_count(old_inode);
+		inode_inc_link_count(old_inode);
 		err = ext2_add_link(new_dentry, old_inode);
 		if (err) {
-			ext2_dec_count(old_inode);
+			inode_dec_link_count(old_inode);
 			goto out_dir;
 		}
 		if (dir_de)
-			ext2_inc_count(new_dir);
+			inode_inc_link_count(new_dir);
 	}
 
 	/*
 	 * Like most other Unix systems, set the ctime for inodes on a
  	 * rename.
-	 * ext2_dec_count() will mark the inode dirty.
+	 * inode_dec_link_count() will mark the inode dirty.
 	 */
 	old_inode->i_ctime = CURRENT_TIME_SEC;
 
 	ext2_delete_entry (old_de, old_page);
-	ext2_dec_count(old_inode);
+	inode_dec_link_count(old_inode);
 
 	if (dir_de) {
 		ext2_set_link(old_inode, dir_de, dir_page, new_dir);
-		ext2_dec_count(old_dir);
+		inode_dec_link_count(old_dir);
 	}
 	return 0;
 
diff --git a/fs/ext3/dir.c b/fs/ext3/dir.c
index 832867aef3d..773459164bb 100644
--- a/fs/ext3/dir.c
+++ b/fs/ext3/dir.c
@@ -95,11 +95,10 @@ static int ext3_readdir(struct file * filp,
 			 void * dirent, filldir_t filldir)
 {
 	int error = 0;
-	unsigned long offset, blk;
-	int i, num, stored;
-	struct buffer_head * bh, * tmp, * bha[16];
-	struct ext3_dir_entry_2 * de;
-	struct super_block * sb;
+	unsigned long offset;
+	int i, stored;
+	struct ext3_dir_entry_2 *de;
+	struct super_block *sb;
 	int err;
 	struct inode *inode = filp->f_dentry->d_inode;
 	int ret = 0;
@@ -124,12 +123,29 @@ static int ext3_readdir(struct file * filp,
 	}
 #endif
 	stored = 0;
-	bh = NULL;
 	offset = filp->f_pos & (sb->s_blocksize - 1);
 
 	while (!error && !stored && filp->f_pos < inode->i_size) {
-		blk = (filp->f_pos) >> EXT3_BLOCK_SIZE_BITS(sb);
-		bh = ext3_bread(NULL, inode, blk, 0, &err);
+		unsigned long blk = filp->f_pos >> EXT3_BLOCK_SIZE_BITS(sb);
+		struct buffer_head map_bh;
+		struct buffer_head *bh = NULL;
+
+		map_bh.b_state = 0;
+		err = ext3_get_block_handle(NULL, inode, blk, &map_bh, 0, 0);
+		if (!err) {
+			page_cache_readahead(sb->s_bdev->bd_inode->i_mapping,
+				&filp->f_ra,
+				filp,
+				map_bh.b_blocknr >>
+					(PAGE_CACHE_SHIFT - inode->i_blkbits),
+				1);
+			bh = ext3_bread(NULL, inode, blk, 0, &err);
+		}
+
+		/*
+		 * We ignore I/O errors on directories so users have a chance
+		 * of recovering data when there's a bad sector
+		 */
 		if (!bh) {
 			ext3_error (sb, "ext3_readdir",
 				"directory #%lu contains a hole at offset %lu",
@@ -138,26 +154,6 @@ static int ext3_readdir(struct file * filp,
 			continue;
 		}
 
-		/*
-		 * Do the readahead
-		 */
-		if (!offset) {
-			for (i = 16 >> (EXT3_BLOCK_SIZE_BITS(sb) - 9), num = 0;
-			     i > 0; i--) {
-				tmp = ext3_getblk (NULL, inode, ++blk, 0, &err);
-				if (tmp && !buffer_uptodate(tmp) &&
-						!buffer_locked(tmp))
-					bha[num++] = tmp;
-				else
-					brelse (tmp);
-			}
-			if (num) {
-				ll_rw_block (READA, num, bha);
-				for (i = 0; i < num; i++)
-					brelse (bha[i]);
-			}
-		}
-
 revalidate:
 		/* If the dir block has changed since the last call to
 		 * readdir(2), then we might be pointing to an invalid
diff --git a/fs/ext3/file.c b/fs/ext3/file.c
index 98e78345ead..59098ea5671 100644
--- a/fs/ext3/file.c
+++ b/fs/ext3/file.c
@@ -37,9 +37,9 @@ static int ext3_release_file (struct inode * inode, struct file * filp)
 	if ((filp->f_mode & FMODE_WRITE) &&
 			(atomic_read(&inode->i_writecount) == 1))
 	{
-		down(&EXT3_I(inode)->truncate_sem);
+		mutex_lock(&EXT3_I(inode)->truncate_mutex);
 		ext3_discard_reservation(inode);
-		up(&EXT3_I(inode)->truncate_sem);
+		mutex_unlock(&EXT3_I(inode)->truncate_mutex);
 	}
 	if (is_dx(inode) && filp->private_data)
 		ext3_htree_free_dir_info(filp->private_data);
diff --git a/fs/ext3/inode.c b/fs/ext3/inode.c
index 0384e539b88..2c361377e0a 100644
--- a/fs/ext3/inode.c
+++ b/fs/ext3/inode.c
@@ -671,7 +671,7 @@ err_out:
  * The BKL may not be held on entry here.  Be sure to take it early.
  */
 
-static int
+int
 ext3_get_block_handle(handle_t *handle, struct inode *inode, sector_t iblock,
 		struct buffer_head *bh_result, int create, int extend_disksize)
 {
@@ -702,7 +702,7 @@ ext3_get_block_handle(handle_t *handle, struct inode *inode, sector_t iblock,
 	if (!create || err == -EIO)
 		goto cleanup;
 
-	down(&ei->truncate_sem);
+	mutex_lock(&ei->truncate_mutex);
 
 	/*
 	 * If the indirect block is missing while we are reading
@@ -723,7 +723,7 @@ ext3_get_block_handle(handle_t *handle, struct inode *inode, sector_t iblock,
 		}
 		partial = ext3_get_branch(inode, depth, offsets, chain, &err);
 		if (!partial) {
-			up(&ei->truncate_sem);
+			mutex_unlock(&ei->truncate_mutex);
 			if (err)
 				goto cleanup;
 			clear_buffer_new(bh_result);
@@ -759,13 +759,13 @@ ext3_get_block_handle(handle_t *handle, struct inode *inode, sector_t iblock,
 		err = ext3_splice_branch(handle, inode, iblock, chain,
 					 partial, left);
 	/*
-	 * i_disksize growing is protected by truncate_sem.  Don't forget to
+	 * i_disksize growing is protected by truncate_mutex.  Don't forget to
 	 * protect it if you're about to implement concurrent
 	 * ext3_get_block() -bzzz
 	*/
 	if (!err && extend_disksize && inode->i_size > ei->i_disksize)
 		ei->i_disksize = inode->i_size;
-	up(&ei->truncate_sem);
+	mutex_unlock(&ei->truncate_mutex);
 	if (err)
 		goto cleanup;
 
@@ -1227,7 +1227,7 @@ static int journal_dirty_data_fn(handle_t *handle, struct buffer_head *bh)
  *	ext3_file_write() -> generic_file_write() -> __alloc_pages() -> ...
  *
  * Same applies to ext3_get_block().  We will deadlock on various things like
- * lock_journal and i_truncate_sem.
+ * lock_journal and i_truncate_mutex.
  *
  * Setting PF_MEMALLOC here doesn't work - too many internal memory
  * allocations fail.
@@ -2161,7 +2161,7 @@ void ext3_truncate(struct inode * inode)
 	 * From here we block out all ext3_get_block() callers who want to
 	 * modify the block allocation tree.
 	 */
-	down(&ei->truncate_sem);
+	mutex_lock(&ei->truncate_mutex);
 
 	if (n == 1) {		/* direct blocks */
 		ext3_free_data(handle, inode, NULL, i_data+offsets[0],
@@ -2228,7 +2228,7 @@ do_indirects:
 
 	ext3_discard_reservation(inode);
 
-	up(&ei->truncate_sem);
+	mutex_unlock(&ei->truncate_mutex);
 	inode->i_mtime = inode->i_ctime = CURRENT_TIME_SEC;
 	ext3_mark_inode_dirty(handle, inode);
 
diff --git a/fs/ext3/ioctl.c b/fs/ext3/ioctl.c
index 556cd551007..aaf1da17b6d 100644
--- a/fs/ext3/ioctl.c
+++ b/fs/ext3/ioctl.c
@@ -182,7 +182,7 @@ flags_err:
 		 * need to allocate reservation structure for this inode
 		 * before set the window size
 		 */
-		down(&ei->truncate_sem);
+		mutex_lock(&ei->truncate_mutex);
 		if (!ei->i_block_alloc_info)
 			ext3_init_block_alloc_info(inode);
 
@@ -190,7 +190,7 @@ flags_err:
 			struct ext3_reserve_window_node *rsv = &ei->i_block_alloc_info->rsv_window_node;
 			rsv->rsv_goal_size = rsv_window_size;
 		}
-		up(&ei->truncate_sem);
+		mutex_unlock(&ei->truncate_mutex);
 		return 0;
 	}
 	case EXT3_IOC_GROUP_EXTEND: {
diff --git a/fs/ext3/super.c b/fs/ext3/super.c
index 56bf7658601..efe5b20d7a5 100644
--- a/fs/ext3/super.c
+++ b/fs/ext3/super.c
@@ -472,7 +472,7 @@ static void init_once(void * foo, kmem_cache_t * cachep, unsigned long flags)
 #ifdef CONFIG_EXT3_FS_XATTR
 		init_rwsem(&ei->xattr_sem);
 #endif
-		init_MUTEX(&ei->truncate_sem);
+		mutex_init(&ei->truncate_mutex);
 		inode_init_once(&ei->vfs_inode);
 	}
 }
@@ -2382,8 +2382,8 @@ static int ext3_statfs (struct super_block * sb, struct kstatfs * buf)
  * Process 1                         Process 2
  * ext3_create()                     quota_sync()
  *   journal_start()                   write_dquot()
- *   DQUOT_INIT()                        down(dqio_sem)
- *     down(dqio_sem)                    journal_start()
+ *   DQUOT_INIT()                        down(dqio_mutex)
+ *     down(dqio_mutex)                    journal_start()
  *
  */
 
diff --git a/fs/fat/fatent.c b/fs/fat/fatent.c
index a1a9e045121..ab171ea8e86 100644
--- a/fs/fat/fatent.c
+++ b/fs/fat/fatent.c
@@ -267,19 +267,19 @@ static struct fatent_operations fat32_ops = {
 
 static inline void lock_fat(struct msdos_sb_info *sbi)
 {
-	down(&sbi->fat_lock);
+	mutex_lock(&sbi->fat_lock);
 }
 
 static inline void unlock_fat(struct msdos_sb_info *sbi)
 {
-	up(&sbi->fat_lock);
+	mutex_unlock(&sbi->fat_lock);
 }
 
 void fat_ent_access_init(struct super_block *sb)
 {
 	struct msdos_sb_info *sbi = MSDOS_SB(sb);
 
-	init_MUTEX(&sbi->fat_lock);
+	mutex_init(&sbi->fat_lock);
 
 	switch (sbi->fat_bits) {
 	case 32:
diff --git a/fs/fcntl.c b/fs/fcntl.c
index dc4a7007f4e..03c789560fb 100644
--- a/fs/fcntl.c
+++ b/fs/fcntl.c
@@ -73,8 +73,8 @@ repeat:
 	 * orig_start..fdt->next_fd
 	 */
 	start = orig_start;
-	if (start < fdt->next_fd)
-		start = fdt->next_fd;
+	if (start < files->next_fd)
+		start = files->next_fd;
 
 	newfd = start;
 	if (start < fdt->max_fdset) {
@@ -102,9 +102,8 @@ repeat:
 	 * we reacquire the fdtable pointer and use it while holding
 	 * the lock, no one can free it during that time.
 	 */
-	fdt = files_fdtable(files);
-	if (start <= fdt->next_fd)
-		fdt->next_fd = newfd + 1;
+	if (start <= files->next_fd)
+		files->next_fd = newfd + 1;
 
 	error = newfd;
 	
diff --git a/fs/file.c b/fs/file.c
index cea7cbea11d..bbc74331473 100644
--- a/fs/file.c
+++ b/fs/file.c
@@ -125,7 +125,8 @@ static void free_fdtable_rcu(struct rcu_head *rcu)
 		kmem_cache_free(files_cachep, fdt->free_files);
 		return;
 	}
-	if (fdt->max_fdset <= __FD_SETSIZE && fdt->max_fds <= NR_OPEN_DEFAULT) {
+	if (fdt->max_fdset <= EMBEDDED_FD_SET_SIZE &&
+		fdt->max_fds <= NR_OPEN_DEFAULT) {
 		/*
 		 * The fdtable was embedded
 		 */
@@ -155,8 +156,9 @@ static void free_fdtable_rcu(struct rcu_head *rcu)
 
 void free_fdtable(struct fdtable *fdt)
 {
-	if (fdt->free_files || fdt->max_fdset > __FD_SETSIZE ||
-					fdt->max_fds > NR_OPEN_DEFAULT)
+	if (fdt->free_files ||
+		fdt->max_fdset > EMBEDDED_FD_SET_SIZE ||
+		fdt->max_fds > NR_OPEN_DEFAULT)
 		call_rcu(&fdt->rcu, free_fdtable_rcu);
 }
 
@@ -199,7 +201,6 @@ static void copy_fdtable(struct fdtable *nfdt, struct fdtable *fdt)
 		       (nfdt->max_fds - fdt->max_fds) *
 					sizeof(struct file *));
 	}
-	nfdt->next_fd = fdt->next_fd;
 }
 
 /*
@@ -220,11 +221,9 @@ fd_set * alloc_fdset(int num)
 
 void free_fdset(fd_set *array, int num)
 {
-	int size = num / 8;
-
-	if (num <= __FD_SETSIZE) /* Don't free an embedded fdset */
+	if (num <= EMBEDDED_FD_SET_SIZE) /* Don't free an embedded fdset */
 		return;
-	else if (size <= PAGE_SIZE)
+	else if (num <= 8 * PAGE_SIZE)
 		kfree(array);
 	else
 		vfree(array);
@@ -237,22 +236,17 @@ static struct fdtable *alloc_fdtable(int nr)
   	fd_set *new_openset = NULL, *new_execset = NULL;
 	struct file **new_fds;
 
-	fdt = kmalloc(sizeof(*fdt), GFP_KERNEL);
+	fdt = kzalloc(sizeof(*fdt), GFP_KERNEL);
 	if (!fdt)
   		goto out;
-	memset(fdt, 0, sizeof(*fdt));
 
-	nfds = __FD_SETSIZE;
+	nfds = 8 * L1_CACHE_BYTES;
   	/* Expand to the max in easy steps */
-  	do {
-		if (nfds < (PAGE_SIZE * 8))
-			nfds = PAGE_SIZE * 8;
-		else {
-			nfds = nfds * 2;
-			if (nfds > NR_OPEN)
-				nfds = NR_OPEN;
-		}
-	} while (nfds <= nr);
+  	while (nfds <= nr) {
+		nfds = nfds * 2;
+		if (nfds > NR_OPEN)
+			nfds = NR_OPEN;
+	}
 
   	new_openset = alloc_fdset(nfds);
   	new_execset = alloc_fdset(nfds);
diff --git a/fs/file_table.c b/fs/file_table.c
index 44fabeaa941..bcea1998b4d 100644
--- a/fs/file_table.c
+++ b/fs/file_table.c
@@ -88,6 +88,7 @@ int proc_nr_files(ctl_table *table, int write, struct file *filp,
  */
 struct file *get_empty_filp(void)
 {
+	struct task_struct *tsk;
 	static int old_max;
 	struct file * f;
 
@@ -112,13 +113,14 @@ struct file *get_empty_filp(void)
 	if (security_file_alloc(f))
 		goto fail_sec;
 
-	eventpoll_init_file(f);
+	tsk = current;
+	INIT_LIST_HEAD(&f->f_u.fu_list);
 	atomic_set(&f->f_count, 1);
-	f->f_uid = current->fsuid;
-	f->f_gid = current->fsgid;
 	rwlock_init(&f->f_owner.lock);
+	f->f_uid = tsk->fsuid;
+	f->f_gid = tsk->fsgid;
+	eventpoll_init_file(f);
 	/* f->f_version: 0 */
-	INIT_LIST_HEAD(&f->f_u.fu_list);
 	return f;
 
 over:
diff --git a/fs/hpfs/hpfs_fn.h b/fs/hpfs/hpfs_fn.h
index 6628c3b352c..4c6473ab3b3 100644
--- a/fs/hpfs/hpfs_fn.h
+++ b/fs/hpfs/hpfs_fn.h
@@ -9,6 +9,7 @@
 //#define DBG
 //#define DEBUG_LOCKS
 
+#include <linux/mutex.h>
 #include <linux/pagemap.h>
 #include <linux/buffer_head.h>
 #include <linux/hpfs_fs.h>
@@ -57,8 +58,8 @@ struct hpfs_inode_info {
 	unsigned i_ea_uid : 1;	/* file's uid is stored in ea */
 	unsigned i_ea_gid : 1;	/* file's gid is stored in ea */
 	unsigned i_dirty : 1;
-	struct semaphore i_sem;
-	struct semaphore i_parent;
+	struct mutex i_mutex;
+	struct mutex i_parent_mutex;
 	loff_t **i_rddir_off;
 	struct inode vfs_inode;
 };
diff --git a/fs/hpfs/inode.c b/fs/hpfs/inode.c
index e3d17e9ea6c..56f2c338c4d 100644
--- a/fs/hpfs/inode.c
+++ b/fs/hpfs/inode.c
@@ -186,9 +186,9 @@ void hpfs_write_inode(struct inode *i)
 		kfree(hpfs_inode->i_rddir_off);
 		hpfs_inode->i_rddir_off = NULL;
 	}
-	down(&hpfs_inode->i_parent);
+	mutex_lock(&hpfs_inode->i_parent_mutex);
 	if (!i->i_nlink) {
-		up(&hpfs_inode->i_parent);
+		mutex_unlock(&hpfs_inode->i_parent_mutex);
 		return;
 	}
 	parent = iget_locked(i->i_sb, hpfs_inode->i_parent_dir);
@@ -199,14 +199,14 @@ void hpfs_write_inode(struct inode *i)
 			hpfs_read_inode(parent);
 			unlock_new_inode(parent);
 		}
-		down(&hpfs_inode->i_sem);
+		mutex_lock(&hpfs_inode->i_mutex);
 		hpfs_write_inode_nolock(i);
-		up(&hpfs_inode->i_sem);
+		mutex_unlock(&hpfs_inode->i_mutex);
 		iput(parent);
 	} else {
 		mark_inode_dirty(i);
 	}
-	up(&hpfs_inode->i_parent);
+	mutex_unlock(&hpfs_inode->i_parent_mutex);
 }
 
 void hpfs_write_inode_nolock(struct inode *i)
diff --git a/fs/hpfs/namei.c b/fs/hpfs/namei.c
index 8ff8fc433fc..a03abb12c61 100644
--- a/fs/hpfs/namei.c
+++ b/fs/hpfs/namei.c
@@ -60,7 +60,7 @@ static int hpfs_mkdir(struct inode *dir, struct dentry *dentry, int mode)
 	if (dee.read_only)
 		result->i_mode &= ~0222;
 
-	down(&hpfs_i(dir)->i_sem);
+	mutex_lock(&hpfs_i(dir)->i_mutex);
 	r = hpfs_add_dirent(dir, (char *)name, len, &dee, 0);
 	if (r == 1)
 		goto bail3;
@@ -101,11 +101,11 @@ static int hpfs_mkdir(struct inode *dir, struct dentry *dentry, int mode)
 		hpfs_write_inode_nolock(result);
 	}
 	d_instantiate(dentry, result);
-	up(&hpfs_i(dir)->i_sem);
+	mutex_unlock(&hpfs_i(dir)->i_mutex);
 	unlock_kernel();
 	return 0;
 bail3:
-	up(&hpfs_i(dir)->i_sem);
+	mutex_unlock(&hpfs_i(dir)->i_mutex);
 	iput(result);
 bail2:
 	hpfs_brelse4(&qbh0);
@@ -168,7 +168,7 @@ static int hpfs_create(struct inode *dir, struct dentry *dentry, int mode, struc
 	result->i_data.a_ops = &hpfs_aops;
 	hpfs_i(result)->mmu_private = 0;
 
-	down(&hpfs_i(dir)->i_sem);
+	mutex_lock(&hpfs_i(dir)->i_mutex);
 	r = hpfs_add_dirent(dir, (char *)name, len, &dee, 0);
 	if (r == 1)
 		goto bail2;
@@ -193,12 +193,12 @@ static int hpfs_create(struct inode *dir, struct dentry *dentry, int mode, struc
 		hpfs_write_inode_nolock(result);
 	}
 	d_instantiate(dentry, result);
-	up(&hpfs_i(dir)->i_sem);
+	mutex_unlock(&hpfs_i(dir)->i_mutex);
 	unlock_kernel();
 	return 0;
 
 bail2:
-	up(&hpfs_i(dir)->i_sem);
+	mutex_unlock(&hpfs_i(dir)->i_mutex);
 	iput(result);
 bail1:
 	brelse(bh);
@@ -254,7 +254,7 @@ static int hpfs_mknod(struct inode *dir, struct dentry *dentry, int mode, dev_t
 	result->i_blocks = 1;
 	init_special_inode(result, mode, rdev);
 
-	down(&hpfs_i(dir)->i_sem);
+	mutex_lock(&hpfs_i(dir)->i_mutex);
 	r = hpfs_add_dirent(dir, (char *)name, len, &dee, 0);
 	if (r == 1)
 		goto bail2;
@@ -271,12 +271,12 @@ static int hpfs_mknod(struct inode *dir, struct dentry *dentry, int mode, dev_t
 
 	hpfs_write_inode_nolock(result);
 	d_instantiate(dentry, result);
-	up(&hpfs_i(dir)->i_sem);
+	mutex_unlock(&hpfs_i(dir)->i_mutex);
 	brelse(bh);
 	unlock_kernel();
 	return 0;
 bail2:
-	up(&hpfs_i(dir)->i_sem);
+	mutex_unlock(&hpfs_i(dir)->i_mutex);
 	iput(result);
 bail1:
 	brelse(bh);
@@ -333,7 +333,7 @@ static int hpfs_symlink(struct inode *dir, struct dentry *dentry, const char *sy
 	result->i_op = &page_symlink_inode_operations;
 	result->i_data.a_ops = &hpfs_symlink_aops;
 
-	down(&hpfs_i(dir)->i_sem);
+	mutex_lock(&hpfs_i(dir)->i_mutex);
 	r = hpfs_add_dirent(dir, (char *)name, len, &dee, 0);
 	if (r == 1)
 		goto bail2;
@@ -352,11 +352,11 @@ static int hpfs_symlink(struct inode *dir, struct dentry *dentry, const char *sy
 
 	hpfs_write_inode_nolock(result);
 	d_instantiate(dentry, result);
-	up(&hpfs_i(dir)->i_sem);
+	mutex_unlock(&hpfs_i(dir)->i_mutex);
 	unlock_kernel();
 	return 0;
 bail2:
-	up(&hpfs_i(dir)->i_sem);
+	mutex_unlock(&hpfs_i(dir)->i_mutex);
 	iput(result);
 bail1:
 	brelse(bh);
@@ -382,8 +382,8 @@ static int hpfs_unlink(struct inode *dir, struct dentry *dentry)
 	lock_kernel();
 	hpfs_adjust_length((char *)name, &len);
 again:
-	down(&hpfs_i(inode)->i_parent);
-	down(&hpfs_i(dir)->i_sem);
+	mutex_lock(&hpfs_i(inode)->i_parent_mutex);
+	mutex_lock(&hpfs_i(dir)->i_mutex);
 	err = -ENOENT;
 	de = map_dirent(dir, hpfs_i(dir)->i_dno, (char *)name, len, &dno, &qbh);
 	if (!de)
@@ -410,8 +410,8 @@ again:
 		if (rep++)
 			break;
 
-		up(&hpfs_i(dir)->i_sem);
-		up(&hpfs_i(inode)->i_parent);
+		mutex_unlock(&hpfs_i(dir)->i_mutex);
+		mutex_unlock(&hpfs_i(inode)->i_parent_mutex);
 		d_drop(dentry);
 		spin_lock(&dentry->d_lock);
 		if (atomic_read(&dentry->d_count) > 1 ||
@@ -442,8 +442,8 @@ again:
 out1:
 	hpfs_brelse4(&qbh);
 out:
-	up(&hpfs_i(dir)->i_sem);
-	up(&hpfs_i(inode)->i_parent);
+	mutex_unlock(&hpfs_i(dir)->i_mutex);
+	mutex_unlock(&hpfs_i(inode)->i_parent_mutex);
 	unlock_kernel();
 	return err;
 }
@@ -463,8 +463,8 @@ static int hpfs_rmdir(struct inode *dir, struct dentry *dentry)
 
 	hpfs_adjust_length((char *)name, &len);
 	lock_kernel();
-	down(&hpfs_i(inode)->i_parent);
-	down(&hpfs_i(dir)->i_sem);
+	mutex_lock(&hpfs_i(inode)->i_parent_mutex);
+	mutex_lock(&hpfs_i(dir)->i_mutex);
 	err = -ENOENT;
 	de = map_dirent(dir, hpfs_i(dir)->i_dno, (char *)name, len, &dno, &qbh);
 	if (!de)
@@ -502,8 +502,8 @@ static int hpfs_rmdir(struct inode *dir, struct dentry *dentry)
 out1:
 	hpfs_brelse4(&qbh);
 out:
-	up(&hpfs_i(dir)->i_sem);
-	up(&hpfs_i(inode)->i_parent);
+	mutex_unlock(&hpfs_i(dir)->i_mutex);
+	mutex_unlock(&hpfs_i(inode)->i_parent_mutex);
 	unlock_kernel();
 	return err;
 }
@@ -565,12 +565,12 @@ static int hpfs_rename(struct inode *old_dir, struct dentry *old_dentry,
 
 	lock_kernel();
 	/* order doesn't matter, due to VFS exclusion */
-	down(&hpfs_i(i)->i_parent);
+	mutex_lock(&hpfs_i(i)->i_parent_mutex);
 	if (new_inode)
-		down(&hpfs_i(new_inode)->i_parent);
-	down(&hpfs_i(old_dir)->i_sem);
+		mutex_lock(&hpfs_i(new_inode)->i_parent_mutex);
+	mutex_lock(&hpfs_i(old_dir)->i_mutex);
 	if (new_dir != old_dir)
-		down(&hpfs_i(new_dir)->i_sem);
+		mutex_lock(&hpfs_i(new_dir)->i_mutex);
 	
 	/* Erm? Moving over the empty non-busy directory is perfectly legal */
 	if (new_inode && S_ISDIR(new_inode->i_mode)) {
@@ -650,11 +650,11 @@ static int hpfs_rename(struct inode *old_dir, struct dentry *old_dentry,
 	hpfs_decide_conv(i, (char *)new_name, new_len);
 end1:
 	if (old_dir != new_dir)
-		up(&hpfs_i(new_dir)->i_sem);
-	up(&hpfs_i(old_dir)->i_sem);
-	up(&hpfs_i(i)->i_parent);
+		mutex_unlock(&hpfs_i(new_dir)->i_mutex);
+	mutex_unlock(&hpfs_i(old_dir)->i_mutex);
+	mutex_unlock(&hpfs_i(i)->i_parent_mutex);
 	if (new_inode)
-		up(&hpfs_i(new_inode)->i_parent);
+		mutex_unlock(&hpfs_i(new_inode)->i_parent_mutex);
 	unlock_kernel();
 	return err;
 }
diff --git a/fs/hpfs/super.c b/fs/hpfs/super.c
index 63e88d7e2c3..9488a794076 100644
--- a/fs/hpfs/super.c
+++ b/fs/hpfs/super.c
@@ -181,8 +181,8 @@ static void init_once(void * foo, kmem_cache_t * cachep, unsigned long flags)
 
 	if ((flags & (SLAB_CTOR_VERIFY|SLAB_CTOR_CONSTRUCTOR)) ==
 	    SLAB_CTOR_CONSTRUCTOR) {
-		init_MUTEX(&ei->i_sem);
-		init_MUTEX(&ei->i_parent);
+		mutex_init(&ei->i_mutex);
+		mutex_init(&ei->i_parent_mutex);
 		inode_init_once(&ei->vfs_inode);
 	}
 }
diff --git a/fs/inode.c b/fs/inode.c
index d0be6159eb7..25967b67903 100644
--- a/fs/inode.c
+++ b/fs/inode.c
@@ -84,14 +84,14 @@ static struct hlist_head *inode_hashtable;
 DEFINE_SPINLOCK(inode_lock);
 
 /*
- * iprune_sem provides exclusion between the kswapd or try_to_free_pages
+ * iprune_mutex provides exclusion between the kswapd or try_to_free_pages
  * icache shrinking path, and the umount path.  Without this exclusion,
  * by the time prune_icache calls iput for the inode whose pages it has
  * been invalidating, or by the time it calls clear_inode & destroy_inode
  * from its final dispose_list, the struct super_block they refer to
  * (for inode->i_sb->s_op) may already have been freed and reused.
  */
-DECLARE_MUTEX(iprune_sem);
+DEFINE_MUTEX(iprune_mutex);
 
 /*
  * Statistics gathering..
@@ -206,7 +206,7 @@ void inode_init_once(struct inode *inode)
 	i_size_ordered_init(inode);
 #ifdef CONFIG_INOTIFY
 	INIT_LIST_HEAD(&inode->inotify_watches);
-	sema_init(&inode->inotify_sem, 1);
+	mutex_init(&inode->inotify_mutex);
 #endif
 }
 
@@ -319,7 +319,7 @@ static int invalidate_list(struct list_head *head, struct list_head *dispose)
 		/*
 		 * We can reschedule here without worrying about the list's
 		 * consistency because the per-sb list of inodes must not
-		 * change during umount anymore, and because iprune_sem keeps
+		 * change during umount anymore, and because iprune_mutex keeps
 		 * shrink_icache_memory() away.
 		 */
 		cond_resched_lock(&inode_lock);
@@ -355,14 +355,14 @@ int invalidate_inodes(struct super_block * sb)
 	int busy;
 	LIST_HEAD(throw_away);
 
-	down(&iprune_sem);
+	mutex_lock(&iprune_mutex);
 	spin_lock(&inode_lock);
 	inotify_unmount_inodes(&sb->s_inodes);
 	busy = invalidate_list(&sb->s_inodes, &throw_away);
 	spin_unlock(&inode_lock);
 
 	dispose_list(&throw_away);
-	up(&iprune_sem);
+	mutex_unlock(&iprune_mutex);
 
 	return busy;
 }
@@ -377,7 +377,7 @@ int __invalidate_device(struct block_device *bdev)
 	if (sb) {
 		/*
 		 * no need to lock the super, get_super holds the
-		 * read semaphore so the filesystem cannot go away
+		 * read mutex so the filesystem cannot go away
 		 * under us (->put_super runs with the write lock
 		 * hold).
 		 */
@@ -423,7 +423,7 @@ static void prune_icache(int nr_to_scan)
 	int nr_scanned;
 	unsigned long reap = 0;
 
-	down(&iprune_sem);
+	mutex_lock(&iprune_mutex);
 	spin_lock(&inode_lock);
 	for (nr_scanned = 0; nr_scanned < nr_to_scan; nr_scanned++) {
 		struct inode *inode;
@@ -459,7 +459,7 @@ static void prune_icache(int nr_to_scan)
 	spin_unlock(&inode_lock);
 
 	dispose_list(&freeable);
-	up(&iprune_sem);
+	mutex_unlock(&iprune_mutex);
 
 	if (current_is_kswapd())
 		mod_page_state(kswapd_inodesteal, reap);
diff --git a/fs/inotify.c b/fs/inotify.c
index 3041503bde0..0ee39ef591c 100644
--- a/fs/inotify.c
+++ b/fs/inotify.c
@@ -54,10 +54,10 @@ int inotify_max_queued_events;
  * Lock ordering:
  *
  * dentry->d_lock (used to keep d_move() away from dentry->d_parent)
- * iprune_sem (synchronize shrink_icache_memory())
+ * iprune_mutex (synchronize shrink_icache_memory())
  * 	inode_lock (protects the super_block->s_inodes list)
- * 	inode->inotify_sem (protects inode->inotify_watches and watches->i_list)
- * 		inotify_dev->sem (protects inotify_device and watches->d_list)
+ * 	inode->inotify_mutex (protects inode->inotify_watches and watches->i_list)
+ * 		inotify_dev->mutex (protects inotify_device and watches->d_list)
  */
 
 /*
@@ -79,12 +79,12 @@ int inotify_max_queued_events;
 /*
  * struct inotify_device - represents an inotify instance
  *
- * This structure is protected by the semaphore 'sem'.
+ * This structure is protected by the mutex 'mutex'.
  */
 struct inotify_device {
 	wait_queue_head_t 	wq;		/* wait queue for i/o */
 	struct idr		idr;		/* idr mapping wd -> watch */
-	struct semaphore	sem;		/* protects this bad boy */
+	struct mutex		mutex;		/* protects this bad boy */
 	struct list_head 	events;		/* list of queued events */
 	struct list_head	watches;	/* list of watches */
 	atomic_t		count;		/* reference count */
@@ -101,7 +101,7 @@ struct inotify_device {
  * device.  In read(), this list is walked and all events that can fit in the
  * buffer are returned.
  *
- * Protected by dev->sem of the device in which we are queued.
+ * Protected by dev->mutex of the device in which we are queued.
  */
 struct inotify_kernel_event {
 	struct inotify_event	event;	/* the user-space event */
@@ -112,8 +112,8 @@ struct inotify_kernel_event {
 /*
  * struct inotify_watch - represents a watch request on a specific inode
  *
- * d_list is protected by dev->sem of the associated watch->dev.
- * i_list and mask are protected by inode->inotify_sem of the associated inode.
+ * d_list is protected by dev->mutex of the associated watch->dev.
+ * i_list and mask are protected by inode->inotify_mutex of the associated inode.
  * dev, inode, and wd are never written to once the watch is created.
  */
 struct inotify_watch {
@@ -261,7 +261,7 @@ static struct inotify_kernel_event * kernel_event(s32 wd, u32 mask, u32 cookie,
 /*
  * inotify_dev_get_event - return the next event in the given dev's queue
  *
- * Caller must hold dev->sem.
+ * Caller must hold dev->mutex.
  */
 static inline struct inotify_kernel_event *
 inotify_dev_get_event(struct inotify_device *dev)
@@ -272,7 +272,7 @@ inotify_dev_get_event(struct inotify_device *dev)
 /*
  * inotify_dev_queue_event - add a new event to the given device
  *
- * Caller must hold dev->sem.  Can sleep (calls kernel_event()).
+ * Caller must hold dev->mutex.  Can sleep (calls kernel_event()).
  */
 static void inotify_dev_queue_event(struct inotify_device *dev,
 				    struct inotify_watch *watch, u32 mask,
@@ -315,7 +315,7 @@ static void inotify_dev_queue_event(struct inotify_device *dev,
 /*
  * remove_kevent - cleans up and ultimately frees the given kevent
  *
- * Caller must hold dev->sem.
+ * Caller must hold dev->mutex.
  */
 static void remove_kevent(struct inotify_device *dev,
 			  struct inotify_kernel_event *kevent)
@@ -332,7 +332,7 @@ static void remove_kevent(struct inotify_device *dev,
 /*
  * inotify_dev_event_dequeue - destroy an event on the given device
  *
- * Caller must hold dev->sem.
+ * Caller must hold dev->mutex.
  */
 static void inotify_dev_event_dequeue(struct inotify_device *dev)
 {
@@ -346,7 +346,7 @@ static void inotify_dev_event_dequeue(struct inotify_device *dev)
 /*
  * inotify_dev_get_wd - returns the next WD for use by the given dev
  *
- * Callers must hold dev->sem.  This function can sleep.
+ * Callers must hold dev->mutex.  This function can sleep.
  */
 static int inotify_dev_get_wd(struct inotify_device *dev,
 			      struct inotify_watch *watch)
@@ -383,7 +383,7 @@ static int find_inode(const char __user *dirname, struct nameidata *nd,
 /*
  * create_watch - creates a watch on the given device.
  *
- * Callers must hold dev->sem.  Calls inotify_dev_get_wd() so may sleep.
+ * Callers must hold dev->mutex.  Calls inotify_dev_get_wd() so may sleep.
  * Both 'dev' and 'inode' (by way of nameidata) need to be pinned.
  */
 static struct inotify_watch *create_watch(struct inotify_device *dev,
@@ -434,7 +434,7 @@ static struct inotify_watch *create_watch(struct inotify_device *dev,
 /*
  * inotify_find_dev - find the watch associated with the given inode and dev
  *
- * Callers must hold inode->inotify_sem.
+ * Callers must hold inode->inotify_mutex.
  */
 static struct inotify_watch *inode_find_dev(struct inode *inode,
 					    struct inotify_device *dev)
@@ -469,7 +469,7 @@ static void remove_watch_no_event(struct inotify_watch *watch,
  * the IN_IGNORED event to the given device signifying that the inode is no
  * longer watched.
  *
- * Callers must hold both inode->inotify_sem and dev->sem.  We drop a
+ * Callers must hold both inode->inotify_mutex and dev->mutex.  We drop a
  * reference to the inode before returning.
  *
  * The inode is not iput() so as to remain atomic.  If the inode needs to be
@@ -507,21 +507,21 @@ void inotify_inode_queue_event(struct inode *inode, u32 mask, u32 cookie,
 	if (!inotify_inode_watched(inode))
 		return;
 
-	down(&inode->inotify_sem);
+	mutex_lock(&inode->inotify_mutex);
 	list_for_each_entry_safe(watch, next, &inode->inotify_watches, i_list) {
 		u32 watch_mask = watch->mask;
 		if (watch_mask & mask) {
 			struct inotify_device *dev = watch->dev;
 			get_inotify_watch(watch);
-			down(&dev->sem);
+			mutex_lock(&dev->mutex);
 			inotify_dev_queue_event(dev, watch, mask, cookie, name);
 			if (watch_mask & IN_ONESHOT)
 				remove_watch_no_event(watch, dev);
-			up(&dev->sem);
+			mutex_unlock(&dev->mutex);
 			put_inotify_watch(watch);
 		}
 	}
-	up(&inode->inotify_sem);
+	mutex_unlock(&inode->inotify_mutex);
 }
 EXPORT_SYMBOL_GPL(inotify_inode_queue_event);
 
@@ -569,7 +569,7 @@ EXPORT_SYMBOL_GPL(inotify_get_cookie);
  * @list: list of inodes being unmounted (sb->s_inodes)
  *
  * Called with inode_lock held, protecting the unmounting super block's list
- * of inodes, and with iprune_sem held, keeping shrink_icache_memory() at bay.
+ * of inodes, and with iprune_mutex held, keeping shrink_icache_memory() at bay.
  * We temporarily drop inode_lock, however, and CAN block.
  */
 void inotify_unmount_inodes(struct list_head *list)
@@ -618,7 +618,7 @@ void inotify_unmount_inodes(struct list_head *list)
 		 * We can safely drop inode_lock here because we hold
 		 * references on both inode and next_i.  Also no new inodes
 		 * will be added since the umount has begun.  Finally,
-		 * iprune_sem keeps shrink_icache_memory() away.
+		 * iprune_mutex keeps shrink_icache_memory() away.
 		 */
 		spin_unlock(&inode_lock);
 
@@ -626,16 +626,16 @@ void inotify_unmount_inodes(struct list_head *list)
 			iput(need_iput_tmp);
 
 		/* for each watch, send IN_UNMOUNT and then remove it */
-		down(&inode->inotify_sem);
+		mutex_lock(&inode->inotify_mutex);
 		watches = &inode->inotify_watches;
 		list_for_each_entry_safe(watch, next_w, watches, i_list) {
 			struct inotify_device *dev = watch->dev;
-			down(&dev->sem);
+			mutex_lock(&dev->mutex);
 			inotify_dev_queue_event(dev, watch, IN_UNMOUNT,0,NULL);
 			remove_watch(watch, dev);
-			up(&dev->sem);
+			mutex_unlock(&dev->mutex);
 		}
-		up(&inode->inotify_sem);
+		mutex_unlock(&inode->inotify_mutex);
 		iput(inode);		
 
 		spin_lock(&inode_lock);
@@ -651,14 +651,14 @@ void inotify_inode_is_dead(struct inode *inode)
 {
 	struct inotify_watch *watch, *next;
 
-	down(&inode->inotify_sem);
+	mutex_lock(&inode->inotify_mutex);
 	list_for_each_entry_safe(watch, next, &inode->inotify_watches, i_list) {
 		struct inotify_device *dev = watch->dev;
-		down(&dev->sem);
+		mutex_lock(&dev->mutex);
 		remove_watch(watch, dev);
-		up(&dev->sem);
+		mutex_unlock(&dev->mutex);
 	}
-	up(&inode->inotify_sem);
+	mutex_unlock(&inode->inotify_mutex);
 }
 EXPORT_SYMBOL_GPL(inotify_inode_is_dead);
 
@@ -670,10 +670,10 @@ static unsigned int inotify_poll(struct file *file, poll_table *wait)
 	int ret = 0;
 
 	poll_wait(file, &dev->wq, wait);
-	down(&dev->sem);
+	mutex_lock(&dev->mutex);
 	if (!list_empty(&dev->events))
 		ret = POLLIN | POLLRDNORM;
-	up(&dev->sem);
+	mutex_unlock(&dev->mutex);
 
 	return ret;
 }
@@ -695,9 +695,9 @@ static ssize_t inotify_read(struct file *file, char __user *buf,
 
 		prepare_to_wait(&dev->wq, &wait, TASK_INTERRUPTIBLE);
 
-		down(&dev->sem);
+		mutex_lock(&dev->mutex);
 		events = !list_empty(&dev->events);
-		up(&dev->sem);
+		mutex_unlock(&dev->mutex);
 		if (events) {
 			ret = 0;
 			break;
@@ -720,7 +720,7 @@ static ssize_t inotify_read(struct file *file, char __user *buf,
 	if (ret)
 		return ret;
 
-	down(&dev->sem);
+	mutex_lock(&dev->mutex);
 	while (1) {
 		struct inotify_kernel_event *kevent;
 
@@ -750,7 +750,7 @@ static ssize_t inotify_read(struct file *file, char __user *buf,
 
 		remove_kevent(dev, kevent);
 	}
-	up(&dev->sem);
+	mutex_unlock(&dev->mutex);
 
 	return ret;
 }
@@ -763,37 +763,37 @@ static int inotify_release(struct inode *ignored, struct file *file)
 	 * Destroy all of the watches on this device.  Unfortunately, not very
 	 * pretty.  We cannot do a simple iteration over the list, because we
 	 * do not know the inode until we iterate to the watch.  But we need to
-	 * hold inode->inotify_sem before dev->sem.  The following works.
+	 * hold inode->inotify_mutex before dev->mutex.  The following works.
 	 */
 	while (1) {
 		struct inotify_watch *watch;
 		struct list_head *watches;
 		struct inode *inode;
 
-		down(&dev->sem);
+		mutex_lock(&dev->mutex);
 		watches = &dev->watches;
 		if (list_empty(watches)) {
-			up(&dev->sem);
+			mutex_unlock(&dev->mutex);
 			break;
 		}
 		watch = list_entry(watches->next, struct inotify_watch, d_list);
 		get_inotify_watch(watch);
-		up(&dev->sem);
+		mutex_unlock(&dev->mutex);
 
 		inode = watch->inode;
-		down(&inode->inotify_sem);
-		down(&dev->sem);
+		mutex_lock(&inode->inotify_mutex);
+		mutex_lock(&dev->mutex);
 		remove_watch_no_event(watch, dev);
-		up(&dev->sem);
-		up(&inode->inotify_sem);
+		mutex_unlock(&dev->mutex);
+		mutex_unlock(&inode->inotify_mutex);
 		put_inotify_watch(watch);
 	}
 
 	/* destroy all of the events on this device */
-	down(&dev->sem);
+	mutex_lock(&dev->mutex);
 	while (!list_empty(&dev->events))
 		inotify_dev_event_dequeue(dev);
-	up(&dev->sem);
+	mutex_unlock(&dev->mutex);
 
 	/* free this device: the put matching the get in inotify_init() */
 	put_inotify_dev(dev);
@@ -811,26 +811,26 @@ static int inotify_ignore(struct inotify_device *dev, s32 wd)
 	struct inotify_watch *watch;
 	struct inode *inode;
 
-	down(&dev->sem);
+	mutex_lock(&dev->mutex);
 	watch = idr_find(&dev->idr, wd);
 	if (unlikely(!watch)) {
-		up(&dev->sem);
+		mutex_unlock(&dev->mutex);
 		return -EINVAL;
 	}
 	get_inotify_watch(watch);
 	inode = watch->inode;
-	up(&dev->sem);
+	mutex_unlock(&dev->mutex);
 
-	down(&inode->inotify_sem);
-	down(&dev->sem);
+	mutex_lock(&inode->inotify_mutex);
+	mutex_lock(&dev->mutex);
 
 	/* make sure that we did not race */
 	watch = idr_find(&dev->idr, wd);
 	if (likely(watch))
 		remove_watch(watch, dev);
 
-	up(&dev->sem);
-	up(&inode->inotify_sem);
+	mutex_unlock(&dev->mutex);
+	mutex_unlock(&inode->inotify_mutex);
 	put_inotify_watch(watch);
 
 	return 0;
@@ -905,7 +905,7 @@ asmlinkage long sys_inotify_init(void)
 	INIT_LIST_HEAD(&dev->events);
 	INIT_LIST_HEAD(&dev->watches);
 	init_waitqueue_head(&dev->wq);
-	sema_init(&dev->sem, 1);
+	mutex_init(&dev->mutex);
 	dev->event_count = 0;
 	dev->queue_size = 0;
 	dev->max_events = inotify_max_queued_events;
@@ -960,8 +960,8 @@ asmlinkage long sys_inotify_add_watch(int fd, const char __user *path, u32 mask)
 	inode = nd.dentry->d_inode;
 	dev = filp->private_data;
 
-	down(&inode->inotify_sem);
-	down(&dev->sem);
+	mutex_lock(&inode->inotify_mutex);
+	mutex_lock(&dev->mutex);
 
 	if (mask & IN_MASK_ADD)
 		mask_add = 1;
@@ -998,8 +998,8 @@ asmlinkage long sys_inotify_add_watch(int fd, const char __user *path, u32 mask)
 	list_add(&watch->i_list, &inode->inotify_watches);
 	ret = watch->wd;
 out:
-	up(&dev->sem);
-	up(&inode->inotify_sem);
+	mutex_unlock(&dev->mutex);
+	mutex_unlock(&inode->inotify_mutex);
 	path_release(&nd);
 fput_and_out:
 	fput_light(filp, fput_needed);
diff --git a/fs/jbd/checkpoint.c b/fs/jbd/checkpoint.c
index 543ed543d1e..3f5102b069d 100644
--- a/fs/jbd/checkpoint.c
+++ b/fs/jbd/checkpoint.c
@@ -85,7 +85,7 @@ void __log_wait_for_space(journal_t *journal)
 		if (journal->j_flags & JFS_ABORT)
 			return;
 		spin_unlock(&journal->j_state_lock);
-		down(&journal->j_checkpoint_sem);
+		mutex_lock(&journal->j_checkpoint_mutex);
 
 		/*
 		 * Test again, another process may have checkpointed while we
@@ -98,7 +98,7 @@ void __log_wait_for_space(journal_t *journal)
 			log_do_checkpoint(journal);
 			spin_lock(&journal->j_state_lock);
 		}
-		up(&journal->j_checkpoint_sem);
+		mutex_unlock(&journal->j_checkpoint_mutex);
 	}
 }
 
diff --git a/fs/jbd/journal.c b/fs/jbd/journal.c
index e4b516ac498..95a628d8cac 100644
--- a/fs/jbd/journal.c
+++ b/fs/jbd/journal.c
@@ -659,8 +659,8 @@ static journal_t * journal_init_common (void)
 	init_waitqueue_head(&journal->j_wait_checkpoint);
 	init_waitqueue_head(&journal->j_wait_commit);
 	init_waitqueue_head(&journal->j_wait_updates);
-	init_MUTEX(&journal->j_barrier);
-	init_MUTEX(&journal->j_checkpoint_sem);
+	mutex_init(&journal->j_barrier);
+	mutex_init(&journal->j_checkpoint_mutex);
 	spin_lock_init(&journal->j_revoke_lock);
 	spin_lock_init(&journal->j_list_lock);
 	spin_lock_init(&journal->j_state_lock);
diff --git a/fs/jbd/transaction.c b/fs/jbd/transaction.c
index ca917973c2c..5fc40888f4c 100644
--- a/fs/jbd/transaction.c
+++ b/fs/jbd/transaction.c
@@ -455,7 +455,7 @@ void journal_lock_updates(journal_t *journal)
 	 * to make sure that we serialise special journal-locked operations
 	 * too.
 	 */
-	down(&journal->j_barrier);
+	mutex_lock(&journal->j_barrier);
 }
 
 /**
@@ -470,7 +470,7 @@ void journal_unlock_updates (journal_t *journal)
 {
 	J_ASSERT(journal->j_barrier_count != 0);
 
-	up(&journal->j_barrier);
+	mutex_unlock(&journal->j_barrier);
 	spin_lock(&journal->j_state_lock);
 	--journal->j_barrier_count;
 	spin_unlock(&journal->j_state_lock);
diff --git a/fs/jffs/inode-v23.c b/fs/jffs/inode-v23.c
index fc3855a1aef..890d7ff7456 100644
--- a/fs/jffs/inode-v23.c
+++ b/fs/jffs/inode-v23.c
@@ -42,7 +42,7 @@
 #include <linux/quotaops.h>
 #include <linux/highmem.h>
 #include <linux/vfs.h>
-#include <asm/semaphore.h>
+#include <linux/mutex.h>
 #include <asm/byteorder.h>
 #include <asm/uaccess.h>
 
@@ -203,7 +203,7 @@ jffs_setattr(struct dentry *dentry, struct iattr *iattr)
 	fmc = c->fmc;
 
 	D3(printk (KERN_NOTICE "notify_change(): down biglock\n"));
-	down(&fmc->biglock);
+	mutex_lock(&fmc->biglock);
 
 	f = jffs_find_file(c, inode->i_ino);
 
@@ -211,7 +211,7 @@ jffs_setattr(struct dentry *dentry, struct iattr *iattr)
 		printk("jffs_setattr(): Invalid inode number: %lu\n",
 		       inode->i_ino);
 		D3(printk (KERN_NOTICE "notify_change(): up biglock\n"));
-		up(&fmc->biglock);
+		mutex_unlock(&fmc->biglock);
 		res = -EINVAL;
 		goto out;
 	});
@@ -232,7 +232,7 @@ jffs_setattr(struct dentry *dentry, struct iattr *iattr)
 	if (!(new_node = jffs_alloc_node())) {
 		D(printk("jffs_setattr(): Allocation failed!\n"));
 		D3(printk (KERN_NOTICE "notify_change(): up biglock\n"));
-		up(&fmc->biglock);
+		mutex_unlock(&fmc->biglock);
 		res = -ENOMEM;
 		goto out;
 	}
@@ -319,7 +319,7 @@ jffs_setattr(struct dentry *dentry, struct iattr *iattr)
 		D(printk("jffs_notify_change(): The write failed!\n"));
 		jffs_free_node(new_node);
 		D3(printk (KERN_NOTICE "n_c(): up biglock\n"));
-		up(&c->fmc->biglock);
+		mutex_unlock(&c->fmc->biglock);
 		goto out;
 	}
 
@@ -327,7 +327,7 @@ jffs_setattr(struct dentry *dentry, struct iattr *iattr)
 
 	mark_inode_dirty(inode);
 	D3(printk (KERN_NOTICE "n_c(): up biglock\n"));
-	up(&c->fmc->biglock);
+	mutex_unlock(&c->fmc->biglock);
 out:
 	unlock_kernel();
 	return res;
@@ -461,7 +461,7 @@ jffs_rename(struct inode *old_dir, struct dentry *old_dentry,
 		goto jffs_rename_end;
 	}
 	D3(printk (KERN_NOTICE "rename(): down biglock\n"));
-	down(&c->fmc->biglock);
+	mutex_lock(&c->fmc->biglock);
 	/* Create a node and initialize as much as needed.  */
 	result = -ENOMEM;
 	if (!(node = jffs_alloc_node())) {
@@ -555,7 +555,7 @@ jffs_rename(struct inode *old_dir, struct dentry *old_dentry,
 
 jffs_rename_end:
 	D3(printk (KERN_NOTICE "rename(): up biglock\n"));
-	up(&c->fmc->biglock);
+	mutex_unlock(&c->fmc->biglock);
 	unlock_kernel();
 	return result;
 } /* jffs_rename()  */
@@ -574,14 +574,14 @@ jffs_readdir(struct file *filp, void *dirent, filldir_t filldir)
 	int ddino;
 	lock_kernel();
 	D3(printk (KERN_NOTICE "readdir(): down biglock\n"));
-	down(&c->fmc->biglock);
+	mutex_lock(&c->fmc->biglock);
 
 	D2(printk("jffs_readdir(): inode: 0x%p, filp: 0x%p\n", inode, filp));
 	if (filp->f_pos == 0) {
 		D3(printk("jffs_readdir(): \".\" %lu\n", inode->i_ino));
 		if (filldir(dirent, ".", 1, filp->f_pos, inode->i_ino, DT_DIR) < 0) {
 			D3(printk (KERN_NOTICE "readdir(): up biglock\n"));
-			up(&c->fmc->biglock);
+			mutex_unlock(&c->fmc->biglock);
 			unlock_kernel();
 			return 0;
 		}
@@ -598,7 +598,7 @@ jffs_readdir(struct file *filp, void *dirent, filldir_t filldir)
 		D3(printk("jffs_readdir(): \"..\" %u\n", ddino));
 		if (filldir(dirent, "..", 2, filp->f_pos, ddino, DT_DIR) < 0) {
 			D3(printk (KERN_NOTICE "readdir(): up biglock\n"));
-			up(&c->fmc->biglock);
+			mutex_unlock(&c->fmc->biglock);
 			unlock_kernel();
 			return 0;
 		}
@@ -617,7 +617,7 @@ jffs_readdir(struct file *filp, void *dirent, filldir_t filldir)
 		if (filldir(dirent, f->name, f->nsize,
 			    filp->f_pos , f->ino, DT_UNKNOWN) < 0) {
 		        D3(printk (KERN_NOTICE "readdir(): up biglock\n"));
-			up(&c->fmc->biglock);
+			mutex_unlock(&c->fmc->biglock);
 			unlock_kernel();
 			return 0;
 		}
@@ -627,7 +627,7 @@ jffs_readdir(struct file *filp, void *dirent, filldir_t filldir)
 		} while(f && f->deleted);
 	}
 	D3(printk (KERN_NOTICE "readdir(): up biglock\n"));
-	up(&c->fmc->biglock);
+	mutex_unlock(&c->fmc->biglock);
 	unlock_kernel();
 	return filp->f_pos;
 } /* jffs_readdir()  */
@@ -660,7 +660,7 @@ jffs_lookup(struct inode *dir, struct dentry *dentry, struct nameidata *nd)
 	});
 
 	D3(printk (KERN_NOTICE "lookup(): down biglock\n"));
-	down(&c->fmc->biglock);
+	mutex_lock(&c->fmc->biglock);
 
 	r = -ENAMETOOLONG;
 	if (len > JFFS_MAX_NAME_LEN) {
@@ -683,31 +683,31 @@ jffs_lookup(struct inode *dir, struct dentry *dentry, struct nameidata *nd)
 
 	if ((len == 1) && (name[0] == '.')) {
 		D3(printk (KERN_NOTICE "lookup(): up biglock\n"));
-		up(&c->fmc->biglock);
+		mutex_unlock(&c->fmc->biglock);
 		if (!(inode = iget(dir->i_sb, d->ino))) {
 			D(printk("jffs_lookup(): . iget() ==> NULL\n"));
 			goto jffs_lookup_end_no_biglock;
 		}
 		D3(printk (KERN_NOTICE "lookup(): down biglock\n"));
-		down(&c->fmc->biglock);
+		mutex_lock(&c->fmc->biglock);
 	} else if ((len == 2) && (name[0] == '.') && (name[1] == '.')) {
 	        D3(printk (KERN_NOTICE "lookup(): up biglock\n"));
-		up(&c->fmc->biglock);
+		mutex_unlock(&c->fmc->biglock);
  		if (!(inode = iget(dir->i_sb, d->pino))) {
 			D(printk("jffs_lookup(): .. iget() ==> NULL\n"));
 			goto jffs_lookup_end_no_biglock;
 		}
 		D3(printk (KERN_NOTICE "lookup(): down biglock\n"));
-		down(&c->fmc->biglock);
+		mutex_lock(&c->fmc->biglock);
 	} else if ((f = jffs_find_child(d, name, len))) {
 	        D3(printk (KERN_NOTICE "lookup(): up biglock\n"));
-		up(&c->fmc->biglock);
+		mutex_unlock(&c->fmc->biglock);
 		if (!(inode = iget(dir->i_sb, f->ino))) {
 			D(printk("jffs_lookup(): iget() ==> NULL\n"));
 			goto jffs_lookup_end_no_biglock;
 		}
 		D3(printk (KERN_NOTICE "lookup(): down biglock\n"));
-		down(&c->fmc->biglock);
+		mutex_lock(&c->fmc->biglock);
 	} else {
 		D3(printk("jffs_lookup(): Couldn't find the file. "
 			  "f = 0x%p, name = \"%s\", d = 0x%p, d->ino = %u\n",
@@ -717,13 +717,13 @@ jffs_lookup(struct inode *dir, struct dentry *dentry, struct nameidata *nd)
 
 	d_add(dentry, inode);
 	D3(printk (KERN_NOTICE "lookup(): up biglock\n"));
-	up(&c->fmc->biglock);
+	mutex_unlock(&c->fmc->biglock);
 	unlock_kernel();
 	return NULL;
 
 jffs_lookup_end:
 	D3(printk (KERN_NOTICE "lookup(): up biglock\n"));
-	up(&c->fmc->biglock);
+	mutex_unlock(&c->fmc->biglock);
 
 jffs_lookup_end_no_biglock:
 	unlock_kernel();
@@ -753,7 +753,7 @@ jffs_do_readpage_nolock(struct file *file, struct page *page)
 	ClearPageError(page);
 
 	D3(printk (KERN_NOTICE "readpage(): down biglock\n"));
-	down(&c->fmc->biglock);
+	mutex_lock(&c->fmc->biglock);
 
 	read_len = 0;
 	result = 0;
@@ -782,7 +782,7 @@ jffs_do_readpage_nolock(struct file *file, struct page *page)
 	kunmap(page);
 
 	D3(printk (KERN_NOTICE "readpage(): up biglock\n"));
-	up(&c->fmc->biglock);
+	mutex_unlock(&c->fmc->biglock);
 
 	if (result) {
 	        SetPageError(page);
@@ -839,7 +839,7 @@ jffs_mkdir(struct inode *dir, struct dentry *dentry, int mode)
 
 	c = dir_f->c;
 	D3(printk (KERN_NOTICE "mkdir(): down biglock\n"));
-	down(&c->fmc->biglock);
+	mutex_lock(&c->fmc->biglock);
 
 	dir_mode = S_IFDIR | (mode & (S_IRWXUGO|S_ISVTX)
 			      & ~current->fs->umask);
@@ -906,7 +906,7 @@ jffs_mkdir(struct inode *dir, struct dentry *dentry, int mode)
 	result = 0;
 jffs_mkdir_end:
 	D3(printk (KERN_NOTICE "mkdir(): up biglock\n"));
-	up(&c->fmc->biglock);
+	mutex_unlock(&c->fmc->biglock);
 	unlock_kernel();
 	return result;
 } /* jffs_mkdir()  */
@@ -921,10 +921,10 @@ jffs_rmdir(struct inode *dir, struct dentry *dentry)
 	D3(printk("***jffs_rmdir()\n"));
 	D3(printk (KERN_NOTICE "rmdir(): down biglock\n"));
 	lock_kernel();
-	down(&c->fmc->biglock);
+	mutex_lock(&c->fmc->biglock);
 	ret = jffs_remove(dir, dentry, S_IFDIR);
 	D3(printk (KERN_NOTICE "rmdir(): up biglock\n"));
-	up(&c->fmc->biglock);
+	mutex_unlock(&c->fmc->biglock);
 	unlock_kernel();
 	return ret;
 }
@@ -940,10 +940,10 @@ jffs_unlink(struct inode *dir, struct dentry *dentry)
 	lock_kernel();
 	D3(printk("***jffs_unlink()\n"));
 	D3(printk (KERN_NOTICE "unlink(): down biglock\n"));
-	down(&c->fmc->biglock);
+	mutex_lock(&c->fmc->biglock);
 	ret = jffs_remove(dir, dentry, 0);
 	D3(printk (KERN_NOTICE "unlink(): up biglock\n"));
-	up(&c->fmc->biglock);
+	mutex_unlock(&c->fmc->biglock);
 	unlock_kernel();
 	return ret;
 }
@@ -1086,7 +1086,7 @@ jffs_mknod(struct inode *dir, struct dentry *dentry, int mode, dev_t rdev)
 	c = dir_f->c;
 
 	D3(printk (KERN_NOTICE "mknod(): down biglock\n"));
-	down(&c->fmc->biglock);
+	mutex_lock(&c->fmc->biglock);
 
 	/* Create and initialize a new node.  */
 	if (!(node = jffs_alloc_node())) {
@@ -1152,7 +1152,7 @@ jffs_mknod_err:
 
 jffs_mknod_end:
 	D3(printk (KERN_NOTICE "mknod(): up biglock\n"));
-	up(&c->fmc->biglock);
+	mutex_unlock(&c->fmc->biglock);
 	unlock_kernel();
 	return result;
 } /* jffs_mknod()  */
@@ -1203,7 +1203,7 @@ jffs_symlink(struct inode *dir, struct dentry *dentry, const char *symname)
 		return -ENOMEM;
 	}
 	D3(printk (KERN_NOTICE "symlink(): down biglock\n"));
-	down(&c->fmc->biglock);
+	mutex_lock(&c->fmc->biglock);
 
 	node->data_offset = 0;
 	node->removed_size = 0;
@@ -1253,7 +1253,7 @@ jffs_symlink(struct inode *dir, struct dentry *dentry, const char *symname)
 	d_instantiate(dentry, inode);
  jffs_symlink_end:
 	D3(printk (KERN_NOTICE "symlink(): up biglock\n"));
-	up(&c->fmc->biglock);
+	mutex_unlock(&c->fmc->biglock);
 	unlock_kernel();
 	return err;
 } /* jffs_symlink()  */
@@ -1306,7 +1306,7 @@ jffs_create(struct inode *dir, struct dentry *dentry, int mode,
 		return -ENOMEM;
 	}
 	D3(printk (KERN_NOTICE "create(): down biglock\n"));
-	down(&c->fmc->biglock);
+	mutex_lock(&c->fmc->biglock);
 
 	node->data_offset = 0;
 	node->removed_size = 0;
@@ -1359,7 +1359,7 @@ jffs_create(struct inode *dir, struct dentry *dentry, int mode,
 	d_instantiate(dentry, inode);
  jffs_create_end:
 	D3(printk (KERN_NOTICE "create(): up biglock\n"));
-	up(&c->fmc->biglock);
+	mutex_unlock(&c->fmc->biglock);
 	unlock_kernel();
 	return err;
 } /* jffs_create()  */
@@ -1423,7 +1423,7 @@ jffs_file_write(struct file *filp, const char *buf, size_t count,
 	thiscount = min(c->fmc->max_chunk_size - sizeof(struct jffs_raw_inode), count);
 
 	D3(printk (KERN_NOTICE "file_write(): down biglock\n"));
-	down(&c->fmc->biglock);
+	mutex_lock(&c->fmc->biglock);
 
 	/* Urgh. POSIX says we can do short writes if we feel like it. 
 	 * In practice, we can't. Nothing will cope. So we loop until
@@ -1511,7 +1511,7 @@ jffs_file_write(struct file *filp, const char *buf, size_t count,
 	}
  out:
 	D3(printk (KERN_NOTICE "file_write(): up biglock\n"));
-	up(&c->fmc->biglock);
+	mutex_unlock(&c->fmc->biglock);
 
 	/* Fix things in the real inode.  */
 	if (pos > inode->i_size) {
@@ -1567,7 +1567,7 @@ jffs_ioctl(struct inode *inode, struct file *filp, unsigned int cmd,
 		return -EIO;
 	}
 	D3(printk (KERN_NOTICE "ioctl(): down biglock\n"));
-	down(&c->fmc->biglock);
+	mutex_lock(&c->fmc->biglock);
 
 	switch (cmd) {
 	case JFFS_PRINT_HASH:
@@ -1609,7 +1609,7 @@ jffs_ioctl(struct inode *inode, struct file *filp, unsigned int cmd,
 		ret = -ENOTTY;
 	}
 	D3(printk (KERN_NOTICE "ioctl(): up biglock\n"));
-	up(&c->fmc->biglock);
+	mutex_unlock(&c->fmc->biglock);
 	return ret;
 } /* jffs_ioctl()  */
 
@@ -1685,12 +1685,12 @@ jffs_read_inode(struct inode *inode)
 	}
 	c = (struct jffs_control *)inode->i_sb->s_fs_info;
 	D3(printk (KERN_NOTICE "read_inode(): down biglock\n"));
-	down(&c->fmc->biglock);
+	mutex_lock(&c->fmc->biglock);
 	if (!(f = jffs_find_file(c, inode->i_ino))) {
 		D(printk("jffs_read_inode(): No such inode (%lu).\n",
 			 inode->i_ino));
 		D3(printk (KERN_NOTICE "read_inode(): up biglock\n"));
-		up(&c->fmc->biglock);
+		mutex_unlock(&c->fmc->biglock);
 		return;
 	}
 	inode->u.generic_ip = (void *)f;
@@ -1732,7 +1732,7 @@ jffs_read_inode(struct inode *inode)
 	}
 
 	D3(printk (KERN_NOTICE "read_inode(): up biglock\n"));
-	up(&c->fmc->biglock);
+	mutex_unlock(&c->fmc->biglock);
 }
 
 
diff --git a/fs/jffs/intrep.c b/fs/jffs/intrep.c
index ce7b54b0b2b..0ef207dfaf6 100644
--- a/fs/jffs/intrep.c
+++ b/fs/jffs/intrep.c
@@ -62,7 +62,7 @@
 #include <linux/fs.h>
 #include <linux/stat.h>
 #include <linux/pagemap.h>
-#include <asm/semaphore.h>
+#include <linux/mutex.h>
 #include <asm/byteorder.h>
 #include <linux/smp_lock.h>
 #include <linux/time.h>
@@ -3416,7 +3416,7 @@ jffs_garbage_collect_thread(void *ptr)
 		D1(printk (KERN_NOTICE "jffs_garbage_collect_thread(): collecting.\n"));
 
 		D3(printk (KERN_NOTICE "g_c_thread(): down biglock\n"));
-		down(&fmc->biglock);
+		mutex_lock(&fmc->biglock);
 		
 		D1(printk("***jffs_garbage_collect_thread(): round #%u, "
 			  "fmc->dirty_size = %u\n", i++, fmc->dirty_size));
@@ -3447,6 +3447,6 @@ jffs_garbage_collect_thread(void *ptr)
 		
 	gc_end:
 		D3(printk (KERN_NOTICE "g_c_thread(): up biglock\n"));
-		up(&fmc->biglock);
+		mutex_unlock(&fmc->biglock);
 	} /* for (;;) */
 } /* jffs_garbage_collect_thread() */
diff --git a/fs/jffs/jffs_fm.c b/fs/jffs/jffs_fm.c
index 6da13b309bd..7d8ca1aeace 100644
--- a/fs/jffs/jffs_fm.c
+++ b/fs/jffs/jffs_fm.c
@@ -139,7 +139,7 @@ jffs_build_begin(struct jffs_control *c, int unit)
 	fmc->tail = NULL;
 	fmc->head_extra = NULL;
 	fmc->tail_extra = NULL;
-	init_MUTEX(&fmc->biglock);
+	mutex_init(&fmc->biglock);
 	return fmc;
 }
 
diff --git a/fs/jffs/jffs_fm.h b/fs/jffs/jffs_fm.h
index f64151e7412..c794d923df2 100644
--- a/fs/jffs/jffs_fm.h
+++ b/fs/jffs/jffs_fm.h
@@ -20,10 +20,11 @@
 #ifndef __LINUX_JFFS_FM_H__
 #define __LINUX_JFFS_FM_H__
 
+#include <linux/config.h>
 #include <linux/types.h>
 #include <linux/jffs.h>
 #include <linux/mtd/mtd.h>
-#include <linux/config.h>
+#include <linux/mutex.h>
 
 /* The alignment between two nodes in the flash memory.  */
 #define JFFS_ALIGN_SIZE 4
@@ -97,7 +98,7 @@ struct jffs_fmcontrol
 	struct jffs_fm *tail;
 	struct jffs_fm *head_extra;
 	struct jffs_fm *tail_extra;
-	struct semaphore biglock;
+	struct mutex biglock;
 };
 
 /* Notice the two members head_extra and tail_extra in the jffs_control
diff --git a/fs/libfs.c b/fs/libfs.c
index 71fd08fa410..4fdeaceb892 100644
--- a/fs/libfs.c
+++ b/fs/libfs.c
@@ -7,6 +7,8 @@
 #include <linux/pagemap.h>
 #include <linux/mount.h>
 #include <linux/vfs.h>
+#include <linux/mutex.h>
+
 #include <asm/uaccess.h>
 
 int simple_getattr(struct vfsmount *mnt, struct dentry *dentry,
@@ -530,7 +532,7 @@ struct simple_attr {
 	char set_buf[24];
 	void *data;
 	const char *fmt;	/* format for read operation */
-	struct semaphore sem;	/* protects access to these buffers */
+	struct mutex mutex;	/* protects access to these buffers */
 };
 
 /* simple_attr_open is called by an actual attribute open file operation
@@ -549,7 +551,7 @@ int simple_attr_open(struct inode *inode, struct file *file,
 	attr->set = set;
 	attr->data = inode->u.generic_ip;
 	attr->fmt = fmt;
-	init_MUTEX(&attr->sem);
+	mutex_init(&attr->mutex);
 
 	file->private_data = attr;
 
@@ -575,7 +577,7 @@ ssize_t simple_attr_read(struct file *file, char __user *buf,
 	if (!attr->get)
 		return -EACCES;
 
-	down(&attr->sem);
+	mutex_lock(&attr->mutex);
 	if (*ppos) /* continued read */
 		size = strlen(attr->get_buf);
 	else	  /* first read */
@@ -584,7 +586,7 @@ ssize_t simple_attr_read(struct file *file, char __user *buf,
 				 (unsigned long long)attr->get(attr->data));
 
 	ret = simple_read_from_buffer(buf, len, ppos, attr->get_buf, size);
-	up(&attr->sem);
+	mutex_unlock(&attr->mutex);
 	return ret;
 }
 
@@ -602,7 +604,7 @@ ssize_t simple_attr_write(struct file *file, const char __user *buf,
 	if (!attr->set)
 		return -EACCES;
 
-	down(&attr->sem);
+	mutex_lock(&attr->mutex);
 	ret = -EFAULT;
 	size = min(sizeof(attr->set_buf) - 1, len);
 	if (copy_from_user(attr->set_buf, buf, size))
@@ -613,7 +615,7 @@ ssize_t simple_attr_write(struct file *file, const char __user *buf,
 	val = simple_strtol(attr->set_buf, NULL, 0);
 	attr->set(attr->data, val);
 out:
-	up(&attr->sem);
+	mutex_unlock(&attr->mutex);
 	return ret;
 }
 
diff --git a/fs/minix/namei.c b/fs/minix/namei.c
index b25bca5bdb5..5b6a4540a05 100644
--- a/fs/minix/namei.c
+++ b/fs/minix/namei.c
@@ -6,18 +6,6 @@
 
 #include "minix.h"
 
-static inline void inc_count(struct inode *inode)
-{
-	inode->i_nlink++;
-	mark_inode_dirty(inode);
-}
-
-static inline void dec_count(struct inode *inode)
-{
-	inode->i_nlink--;
-	mark_inode_dirty(inode);
-}
-
 static int add_nondir(struct dentry *dentry, struct inode *inode)
 {
 	int err = minix_add_link(dentry, inode);
@@ -25,7 +13,7 @@ static int add_nondir(struct dentry *dentry, struct inode *inode)
 		d_instantiate(dentry, inode);
 		return 0;
 	}
-	dec_count(inode);
+	inode_dec_link_count(inode);
 	iput(inode);
 	return err;
 }
@@ -125,7 +113,7 @@ out:
 	return err;
 
 out_fail:
-	dec_count(inode);
+	inode_dec_link_count(inode);
 	iput(inode);
 	goto out;
 }
@@ -139,7 +127,7 @@ static int minix_link(struct dentry * old_dentry, struct inode * dir,
 		return -EMLINK;
 
 	inode->i_ctime = CURRENT_TIME_SEC;
-	inc_count(inode);
+	inode_inc_link_count(inode);
 	atomic_inc(&inode->i_count);
 	return add_nondir(dentry, inode);
 }
@@ -152,7 +140,7 @@ static int minix_mkdir(struct inode * dir, struct dentry *dentry, int mode)
 	if (dir->i_nlink >= minix_sb(dir->i_sb)->s_link_max)
 		goto out;
 
-	inc_count(dir);
+	inode_inc_link_count(dir);
 
 	inode = minix_new_inode(dir, &err);
 	if (!inode)
@@ -163,7 +151,7 @@ static int minix_mkdir(struct inode * dir, struct dentry *dentry, int mode)
 		inode->i_mode |= S_ISGID;
 	minix_set_inode(inode, 0);
 
-	inc_count(inode);
+	inode_inc_link_count(inode);
 
 	err = minix_make_empty(inode, dir);
 	if (err)
@@ -178,11 +166,11 @@ out:
 	return err;
 
 out_fail:
-	dec_count(inode);
-	dec_count(inode);
+	inode_dec_link_count(inode);
+	inode_dec_link_count(inode);
 	iput(inode);
 out_dir:
-	dec_count(dir);
+	inode_dec_link_count(dir);
 	goto out;
 }
 
@@ -202,7 +190,7 @@ static int minix_unlink(struct inode * dir, struct dentry *dentry)
 		goto end_unlink;
 
 	inode->i_ctime = dir->i_ctime;
-	dec_count(inode);
+	inode_dec_link_count(inode);
 end_unlink:
 	return err;
 }
@@ -215,8 +203,8 @@ static int minix_rmdir(struct inode * dir, struct dentry *dentry)
 	if (minix_empty_dir(inode)) {
 		err = minix_unlink(dir, dentry);
 		if (!err) {
-			dec_count(dir);
-			dec_count(inode);
+			inode_dec_link_count(dir);
+			inode_dec_link_count(inode);
 		}
 	}
 	return err;
@@ -257,34 +245,34 @@ static int minix_rename(struct inode * old_dir, struct dentry *old_dentry,
 		new_de = minix_find_entry(new_dentry, &new_page);
 		if (!new_de)
 			goto out_dir;
-		inc_count(old_inode);
+		inode_inc_link_count(old_inode);
 		minix_set_link(new_de, new_page, old_inode);
 		new_inode->i_ctime = CURRENT_TIME_SEC;
 		if (dir_de)
 			new_inode->i_nlink--;
-		dec_count(new_inode);
+		inode_dec_link_count(new_inode);
 	} else {
 		if (dir_de) {
 			err = -EMLINK;
 			if (new_dir->i_nlink >= info->s_link_max)
 				goto out_dir;
 		}
-		inc_count(old_inode);
+		inode_inc_link_count(old_inode);
 		err = minix_add_link(new_dentry, old_inode);
 		if (err) {
-			dec_count(old_inode);
+			inode_dec_link_count(old_inode);
 			goto out_dir;
 		}
 		if (dir_de)
-			inc_count(new_dir);
+			inode_inc_link_count(new_dir);
 	}
 
 	minix_delete_entry(old_de, old_page);
-	dec_count(old_inode);
+	inode_dec_link_count(old_inode);
 
 	if (dir_de) {
 		minix_set_link(dir_de, dir_page, new_dir);
-		dec_count(old_dir);
+		inode_dec_link_count(old_dir);
 	}
 	return 0;
 
diff --git a/fs/namei.c b/fs/namei.c
index 8dc2b038d5d..c72b940797f 100644
--- a/fs/namei.c
+++ b/fs/namei.c
@@ -104,7 +104,7 @@
  */
 /*
  * [Sep 2001 AV] Single-semaphore locking scheme (kudos to David Holland)
- * implemented.  Let's see if raised priority of ->s_vfs_rename_sem gives
+ * implemented.  Let's see if raised priority of ->s_vfs_rename_mutex gives
  * any extra contention...
  */
 
@@ -1422,7 +1422,7 @@ struct dentry *lock_rename(struct dentry *p1, struct dentry *p2)
 		return NULL;
 	}
 
-	down(&p1->d_inode->i_sb->s_vfs_rename_sem);
+	mutex_lock(&p1->d_inode->i_sb->s_vfs_rename_mutex);
 
 	for (p = p1; p->d_parent != p; p = p->d_parent) {
 		if (p->d_parent == p2) {
@@ -1450,7 +1450,7 @@ void unlock_rename(struct dentry *p1, struct dentry *p2)
 	mutex_unlock(&p1->d_inode->i_mutex);
 	if (p1 != p2) {
 		mutex_unlock(&p2->d_inode->i_mutex);
-		up(&p1->d_inode->i_sb->s_vfs_rename_sem);
+		mutex_unlock(&p1->d_inode->i_sb->s_vfs_rename_mutex);
 	}
 }
 
@@ -2277,17 +2277,17 @@ asmlinkage long sys_link(const char __user *oldname, const char __user *newname)
  *	a) we can get into loop creation. Check is done in is_subdir().
  *	b) race potential - two innocent renames can create a loop together.
  *	   That's where 4.4 screws up. Current fix: serialization on
- *	   sb->s_vfs_rename_sem. We might be more accurate, but that's another
+ *	   sb->s_vfs_rename_mutex. We might be more accurate, but that's another
  *	   story.
  *	c) we have to lock _three_ objects - parents and victim (if it exists).
  *	   And that - after we got ->i_mutex on parents (until then we don't know
  *	   whether the target exists).  Solution: try to be smart with locking
  *	   order for inodes.  We rely on the fact that tree topology may change
- *	   only under ->s_vfs_rename_sem _and_ that parent of the object we
+ *	   only under ->s_vfs_rename_mutex _and_ that parent of the object we
  *	   move will be locked.  Thus we can rank directories by the tree
  *	   (ancestors first) and rank all non-directories after them.
  *	   That works since everybody except rename does "lock parent, lookup,
- *	   lock child" and rename is under ->s_vfs_rename_sem.
+ *	   lock child" and rename is under ->s_vfs_rename_mutex.
  *	   HOWEVER, it relies on the assumption that any object with ->lookup()
  *	   has no more than 1 dentry.  If "hybrid" objects will ever appear,
  *	   we'd better make sure that there's no link(2) for them.
diff --git a/fs/ncpfs/file.c b/fs/ncpfs/file.c
index 973b444d691..ebdad8f6398 100644
--- a/fs/ncpfs/file.c
+++ b/fs/ncpfs/file.c
@@ -46,7 +46,7 @@ int ncp_make_open(struct inode *inode, int right)
 		NCP_FINFO(inode)->volNumber, 
 		NCP_FINFO(inode)->dirEntNum);
 	error = -EACCES;
-	down(&NCP_FINFO(inode)->open_sem);
+	mutex_lock(&NCP_FINFO(inode)->open_mutex);
 	if (!atomic_read(&NCP_FINFO(inode)->opened)) {
 		struct ncp_entry_info finfo;
 		int result;
@@ -93,7 +93,7 @@ int ncp_make_open(struct inode *inode, int right)
 	}
 
 out_unlock:
-	up(&NCP_FINFO(inode)->open_sem);
+	mutex_unlock(&NCP_FINFO(inode)->open_mutex);
 out:
 	return error;
 }
diff --git a/fs/ncpfs/inode.c b/fs/ncpfs/inode.c
index d277a58bd12..0b521d3d97c 100644
--- a/fs/ncpfs/inode.c
+++ b/fs/ncpfs/inode.c
@@ -63,7 +63,7 @@ static void init_once(void * foo, kmem_cache_t * cachep, unsigned long flags)
 
 	if ((flags & (SLAB_CTOR_VERIFY|SLAB_CTOR_CONSTRUCTOR)) ==
 	    SLAB_CTOR_CONSTRUCTOR) {
-		init_MUTEX(&ei->open_sem);
+		mutex_init(&ei->open_mutex);
 		inode_init_once(&ei->vfs_inode);
 	}
 }
@@ -520,7 +520,7 @@ static int ncp_fill_super(struct super_block *sb, void *raw_data, int silent)
 	}
 
 /*	server->lock = 0;	*/
-	init_MUTEX(&server->sem);
+	mutex_init(&server->mutex);
 	server->packet = NULL;
 /*	server->buffer_size = 0;	*/
 /*	server->conn_status = 0;	*/
@@ -557,7 +557,7 @@ static int ncp_fill_super(struct super_block *sb, void *raw_data, int silent)
 	server->dentry_ttl = 0;	/* no caching */
 
 	INIT_LIST_HEAD(&server->tx.requests);
-	init_MUTEX(&server->rcv.creq_sem);
+	mutex_init(&server->rcv.creq_mutex);
 	server->tx.creq		= NULL;
 	server->rcv.creq	= NULL;
 	server->data_ready	= sock->sk->sk_data_ready;
diff --git a/fs/ncpfs/ncplib_kernel.c b/fs/ncpfs/ncplib_kernel.c
index c755e1848a4..d9ebf6439f5 100644
--- a/fs/ncpfs/ncplib_kernel.c
+++ b/fs/ncpfs/ncplib_kernel.c
@@ -291,7 +291,7 @@ ncp_make_closed(struct inode *inode)
 	int err;
 
 	err = 0;
-	down(&NCP_FINFO(inode)->open_sem);	
+	mutex_lock(&NCP_FINFO(inode)->open_mutex);
 	if (atomic_read(&NCP_FINFO(inode)->opened) == 1) {
 		atomic_set(&NCP_FINFO(inode)->opened, 0);
 		err = ncp_close_file(NCP_SERVER(inode), NCP_FINFO(inode)->file_handle);
@@ -301,7 +301,7 @@ ncp_make_closed(struct inode *inode)
 				NCP_FINFO(inode)->volNumber,
 				NCP_FINFO(inode)->dirEntNum, err);
 	}
-	up(&NCP_FINFO(inode)->open_sem);
+	mutex_unlock(&NCP_FINFO(inode)->open_mutex);
 	return err;
 }
 
diff --git a/fs/ncpfs/sock.c b/fs/ncpfs/sock.c
index 6593a5ca88b..8783eb7ec64 100644
--- a/fs/ncpfs/sock.c
+++ b/fs/ncpfs/sock.c
@@ -171,9 +171,9 @@ static inline void __ncp_abort_request(struct ncp_server *server, struct ncp_req
 
 static inline void ncp_abort_request(struct ncp_server *server, struct ncp_request_reply *req, int err)
 {
-	down(&server->rcv.creq_sem);
+	mutex_lock(&server->rcv.creq_mutex);
 	__ncp_abort_request(server, req, err);
-	up(&server->rcv.creq_sem);
+	mutex_unlock(&server->rcv.creq_mutex);
 }
 
 static inline void __ncptcp_abort(struct ncp_server *server)
@@ -303,20 +303,20 @@ static inline void __ncp_start_request(struct ncp_server *server, struct ncp_req
 
 static int ncp_add_request(struct ncp_server *server, struct ncp_request_reply *req)
 {
-	down(&server->rcv.creq_sem);
+	mutex_lock(&server->rcv.creq_mutex);
 	if (!ncp_conn_valid(server)) {
-		up(&server->rcv.creq_sem);
+		mutex_unlock(&server->rcv.creq_mutex);
 		printk(KERN_ERR "ncpfs: tcp: Server died\n");
 		return -EIO;
 	}
 	if (server->tx.creq || server->rcv.creq) {
 		req->status = RQ_QUEUED;
 		list_add_tail(&req->req, &server->tx.requests);
-		up(&server->rcv.creq_sem);
+		mutex_unlock(&server->rcv.creq_mutex);
 		return 0;
 	}
 	__ncp_start_request(server, req);
-	up(&server->rcv.creq_sem);
+	mutex_unlock(&server->rcv.creq_mutex);
 	return 0;
 }
 
@@ -400,7 +400,7 @@ void ncpdgram_rcv_proc(void *s)
 				info_server(server, 0, server->unexpected_packet.data, result);
 				continue;
 			}
-			down(&server->rcv.creq_sem);		
+			mutex_lock(&server->rcv.creq_mutex);
 			req = server->rcv.creq;
 			if (req && (req->tx_type == NCP_ALLOC_SLOT_REQUEST || (server->sequence == reply.sequence && 
 					server->connection == get_conn_number(&reply)))) {
@@ -430,11 +430,11 @@ void ncpdgram_rcv_proc(void *s)
 				     	server->rcv.creq = NULL;
 					ncp_finish_request(req, result);
 					__ncp_next_request(server);
-					up(&server->rcv.creq_sem);
+					mutex_unlock(&server->rcv.creq_mutex);
 					continue;
 				}
 			}
-			up(&server->rcv.creq_sem);
+			mutex_unlock(&server->rcv.creq_mutex);
 		}
 drop:;		
 		_recv(sock, &reply, sizeof(reply), MSG_DONTWAIT);
@@ -472,9 +472,9 @@ static void __ncpdgram_timeout_proc(struct ncp_server *server)
 void ncpdgram_timeout_proc(void *s)
 {
 	struct ncp_server *server = s;
-	down(&server->rcv.creq_sem);
+	mutex_lock(&server->rcv.creq_mutex);
 	__ncpdgram_timeout_proc(server);
-	up(&server->rcv.creq_sem);
+	mutex_unlock(&server->rcv.creq_mutex);
 }
 
 static inline void ncp_init_req(struct ncp_request_reply* req)
@@ -657,18 +657,18 @@ void ncp_tcp_rcv_proc(void *s)
 {
 	struct ncp_server *server = s;
 
-	down(&server->rcv.creq_sem);
+	mutex_lock(&server->rcv.creq_mutex);
 	__ncptcp_rcv_proc(server);
-	up(&server->rcv.creq_sem);
+	mutex_unlock(&server->rcv.creq_mutex);
 }
 
 void ncp_tcp_tx_proc(void *s)
 {
 	struct ncp_server *server = s;
 	
-	down(&server->rcv.creq_sem);
+	mutex_lock(&server->rcv.creq_mutex);
 	__ncptcp_try_send(server);
-	up(&server->rcv.creq_sem);
+	mutex_unlock(&server->rcv.creq_mutex);
 }
 
 static int do_ncp_rpc_call(struct ncp_server *server, int size,
@@ -833,7 +833,7 @@ int ncp_disconnect(struct ncp_server *server)
 
 void ncp_lock_server(struct ncp_server *server)
 {
-	down(&server->sem);
+	mutex_lock(&server->mutex);
 	if (server->lock)
 		printk(KERN_WARNING "ncp_lock_server: was locked!\n");
 	server->lock = 1;
@@ -846,5 +846,5 @@ void ncp_unlock_server(struct ncp_server *server)
 		return;
 	}
 	server->lock = 0;
-	up(&server->sem);
+	mutex_unlock(&server->mutex);
 }
diff --git a/fs/open.c b/fs/open.c
index 70e0230d8e7..1091dadd6c3 100644
--- a/fs/open.c
+++ b/fs/open.c
@@ -973,7 +973,7 @@ repeat:
 	fdt = files_fdtable(files);
  	fd = find_next_zero_bit(fdt->open_fds->fds_bits,
 				fdt->max_fdset,
-				fdt->next_fd);
+				files->next_fd);
 
 	/*
 	 * N.B. For clone tasks sharing a files structure, this test
@@ -998,7 +998,7 @@ repeat:
 
 	FD_SET(fd, fdt->open_fds);
 	FD_CLR(fd, fdt->close_on_exec);
-	fdt->next_fd = fd + 1;
+	files->next_fd = fd + 1;
 #if 1
 	/* Sanity check */
 	if (fdt->fd[fd] != NULL) {
@@ -1019,8 +1019,8 @@ static void __put_unused_fd(struct files_struct *files, unsigned int fd)
 {
 	struct fdtable *fdt = files_fdtable(files);
 	__FD_CLR(fd, fdt->open_fds);
-	if (fd < fdt->next_fd)
-		fdt->next_fd = fd;
+	if (fd < files->next_fd)
+		files->next_fd = fd;
 }
 
 void fastcall put_unused_fd(unsigned int fd)
diff --git a/fs/proc/proc_misc.c b/fs/proc/proc_misc.c
index 1d24fead51a..826c131994c 100644
--- a/fs/proc/proc_misc.c
+++ b/fs/proc/proc_misc.c
@@ -312,7 +312,7 @@ static void *devinfo_next(struct seq_file *f, void *v, loff_t *pos)
 		case BLK_HDR:
 			info->state = BLK_LIST;
 			(*pos)++;
-			break;
+			/*fallthrough*/
 		case BLK_LIST:
 			if (get_blkdev_info(info->blkdev,&idummy,&ndummy)) {
 				/*
diff --git a/fs/qnx4/file.c b/fs/qnx4/file.c
index b471315e24e..c33963fded9 100644
--- a/fs/qnx4/file.c
+++ b/fs/qnx4/file.c
@@ -12,10 +12,7 @@
  * 27-06-1998 by Frank Denis : file overwriting.
  */
 
-#include <linux/config.h>
-#include <linux/types.h>
 #include <linux/fs.h>
-#include <linux/time.h>
 #include <linux/qnx4_fs.h>
 
 /*
diff --git a/fs/quota.c b/fs/quota.c
index ba9e0bf32f6..d6a2be826e2 100644
--- a/fs/quota.c
+++ b/fs/quota.c
@@ -170,10 +170,10 @@ static void quota_sync_sb(struct super_block *sb, int type)
 
 	/* Now when everything is written we can discard the pagecache so
 	 * that userspace sees the changes. We need i_mutex and so we could
-	 * not do it inside dqonoff_sem. Moreover we need to be carefull
+	 * not do it inside dqonoff_mutex. Moreover we need to be carefull
 	 * about races with quotaoff() (that is the reason why we have own
 	 * reference to inode). */
-	down(&sb_dqopt(sb)->dqonoff_sem);
+	mutex_lock(&sb_dqopt(sb)->dqonoff_mutex);
 	for (cnt = 0; cnt < MAXQUOTAS; cnt++) {
 		discard[cnt] = NULL;
 		if (type != -1 && cnt != type)
@@ -182,7 +182,7 @@ static void quota_sync_sb(struct super_block *sb, int type)
 			continue;
 		discard[cnt] = igrab(sb_dqopt(sb)->files[cnt]);
 	}
-	up(&sb_dqopt(sb)->dqonoff_sem);
+	mutex_unlock(&sb_dqopt(sb)->dqonoff_mutex);
 	for (cnt = 0; cnt < MAXQUOTAS; cnt++) {
 		if (discard[cnt]) {
 			mutex_lock(&discard[cnt]->i_mutex);
diff --git a/fs/quota_v2.c b/fs/quota_v2.c
index b4199ec3ece..c519a583e68 100644
--- a/fs/quota_v2.c
+++ b/fs/quota_v2.c
@@ -394,7 +394,7 @@ static int v2_write_dquot(struct dquot *dquot)
 	ssize_t ret;
 	struct v2_disk_dqblk ddquot, empty;
 
-	/* dq_off is guarded by dqio_sem */
+	/* dq_off is guarded by dqio_mutex */
 	if (!dquot->dq_off)
 		if ((ret = dq_insert_tree(dquot)) < 0) {
 			printk(KERN_ERR "VFS: Error %zd occurred while creating quota.\n", ret);
diff --git a/fs/ramfs/file-mmu.c b/fs/ramfs/file-mmu.c
index 2115383dcc8..6ada2095b9a 100644
--- a/fs/ramfs/file-mmu.c
+++ b/fs/ramfs/file-mmu.c
@@ -24,18 +24,7 @@
  * caches is sufficient.
  */
 
-#include <linux/module.h>
 #include <linux/fs.h>
-#include <linux/pagemap.h>
-#include <linux/highmem.h>
-#include <linux/init.h>
-#include <linux/string.h>
-#include <linux/smp_lock.h>
-#include <linux/backing-dev.h>
-#include <linux/ramfs.h>
-
-#include <asm/uaccess.h>
-#include "internal.h"
 
 struct address_space_operations ramfs_aops = {
 	.readpage	= simple_readpage,
diff --git a/fs/seq_file.c b/fs/seq_file.c
index 7c40570b71d..555b9ac04c2 100644
--- a/fs/seq_file.c
+++ b/fs/seq_file.c
@@ -37,7 +37,7 @@ int seq_open(struct file *file, struct seq_operations *op)
 		file->private_data = p;
 	}
 	memset(p, 0, sizeof(*p));
-	sema_init(&p->sem, 1);
+	mutex_init(&p->lock);
 	p->op = op;
 
 	/*
@@ -71,7 +71,7 @@ ssize_t seq_read(struct file *file, char __user *buf, size_t size, loff_t *ppos)
 	void *p;
 	int err = 0;
 
-	down(&m->sem);
+	mutex_lock(&m->lock);
 	/*
 	 * seq_file->op->..m_start/m_stop/m_next may do special actions
 	 * or optimisations based on the file->f_version, so we want to
@@ -164,7 +164,7 @@ Done:
 	else
 		*ppos += copied;
 	file->f_version = m->version;
-	up(&m->sem);
+	mutex_unlock(&m->lock);
 	return copied;
 Enomem:
 	err = -ENOMEM;
@@ -237,7 +237,7 @@ loff_t seq_lseek(struct file *file, loff_t offset, int origin)
 	struct seq_file *m = (struct seq_file *)file->private_data;
 	long long retval = -EINVAL;
 
-	down(&m->sem);
+	mutex_lock(&m->lock);
 	m->version = file->f_version;
 	switch (origin) {
 		case 1:
@@ -260,7 +260,7 @@ loff_t seq_lseek(struct file *file, loff_t offset, int origin)
 				}
 			}
 	}
-	up(&m->sem);
+	mutex_unlock(&m->lock);
 	file->f_version = m->version;
 	return retval;
 }
diff --git a/fs/super.c b/fs/super.c
index e20b5580afd..425861cb1ca 100644
--- a/fs/super.c
+++ b/fs/super.c
@@ -76,9 +76,9 @@ static struct super_block *alloc_super(void)
 		down_write(&s->s_umount);
 		s->s_count = S_BIAS;
 		atomic_set(&s->s_active, 1);
-		sema_init(&s->s_vfs_rename_sem,1);
-		sema_init(&s->s_dquot.dqio_sem, 1);
-		sema_init(&s->s_dquot.dqonoff_sem, 1);
+		mutex_init(&s->s_vfs_rename_mutex);
+		mutex_init(&s->s_dquot.dqio_mutex);
+		mutex_init(&s->s_dquot.dqonoff_mutex);
 		init_rwsem(&s->s_dquot.dqptr_sem);
 		init_waitqueue_head(&s->s_wait_unfrozen);
 		s->s_maxbytes = MAX_NON_LFS;
@@ -693,9 +693,9 @@ struct super_block *get_sb_bdev(struct file_system_type *fs_type,
 	 * will protect the lockfs code from trying to start a snapshot
 	 * while we are mounting
 	 */
-	down(&bdev->bd_mount_sem);
+	mutex_lock(&bdev->bd_mount_mutex);
 	s = sget(fs_type, test_bdev_super, set_bdev_super, bdev);
-	up(&bdev->bd_mount_sem);
+	mutex_unlock(&bdev->bd_mount_mutex);
 	if (IS_ERR(s))
 		goto out;
 
diff --git a/fs/sysv/namei.c b/fs/sysv/namei.c
index 7f0e4b53085..b8a73f716fb 100644
--- a/fs/sysv/namei.c
+++ b/fs/sysv/namei.c
@@ -16,18 +16,6 @@
 #include <linux/smp_lock.h>
 #include "sysv.h"
 
-static inline void inc_count(struct inode *inode)
-{
-	inode->i_nlink++;
-	mark_inode_dirty(inode);
-}
-
-static inline void dec_count(struct inode *inode)
-{
-	inode->i_nlink--;
-	mark_inode_dirty(inode);
-}
-
 static int add_nondir(struct dentry *dentry, struct inode *inode)
 {
 	int err = sysv_add_link(dentry, inode);
@@ -35,7 +23,7 @@ static int add_nondir(struct dentry *dentry, struct inode *inode)
 		d_instantiate(dentry, inode);
 		return 0;
 	}
-	dec_count(inode);
+	inode_dec_link_count(inode);
 	iput(inode);
 	return err;
 }
@@ -124,7 +112,7 @@ out:
 	return err;
 
 out_fail:
-	dec_count(inode);
+	inode_dec_link_count(inode);
 	iput(inode);
 	goto out;
 }
@@ -138,7 +126,7 @@ static int sysv_link(struct dentry * old_dentry, struct inode * dir,
 		return -EMLINK;
 
 	inode->i_ctime = CURRENT_TIME_SEC;
-	inc_count(inode);
+	inode_inc_link_count(inode);
 	atomic_inc(&inode->i_count);
 
 	return add_nondir(dentry, inode);
@@ -151,7 +139,7 @@ static int sysv_mkdir(struct inode * dir, struct dentry *dentry, int mode)
 
 	if (dir->i_nlink >= SYSV_SB(dir->i_sb)->s_link_max) 
 		goto out;
-	inc_count(dir);
+	inode_inc_link_count(dir);
 
 	inode = sysv_new_inode(dir, S_IFDIR|mode);
 	err = PTR_ERR(inode);
@@ -160,7 +148,7 @@ static int sysv_mkdir(struct inode * dir, struct dentry *dentry, int mode)
 
 	sysv_set_inode(inode, 0);
 
-	inc_count(inode);
+	inode_inc_link_count(inode);
 
 	err = sysv_make_empty(inode, dir);
 	if (err)
@@ -175,11 +163,11 @@ out:
 	return err;
 
 out_fail:
-	dec_count(inode);
-	dec_count(inode);
+	inode_dec_link_count(inode);
+	inode_dec_link_count(inode);
 	iput(inode);
 out_dir:
-	dec_count(dir);
+	inode_dec_link_count(dir);
 	goto out;
 }
 
@@ -199,7 +187,7 @@ static int sysv_unlink(struct inode * dir, struct dentry * dentry)
 		goto out;
 
 	inode->i_ctime = dir->i_ctime;
-	dec_count(inode);
+	inode_dec_link_count(inode);
 out:
 	return err;
 }
@@ -213,8 +201,8 @@ static int sysv_rmdir(struct inode * dir, struct dentry * dentry)
 		err = sysv_unlink(dir, dentry);
 		if (!err) {
 			inode->i_size = 0;
-			dec_count(inode);
-			dec_count(dir);
+			inode_dec_link_count(inode);
+			inode_dec_link_count(dir);
 		}
 	}
 	return err;
@@ -258,34 +246,34 @@ static int sysv_rename(struct inode * old_dir, struct dentry * old_dentry,
 		new_de = sysv_find_entry(new_dentry, &new_page);
 		if (!new_de)
 			goto out_dir;
-		inc_count(old_inode);
+		inode_inc_link_count(old_inode);
 		sysv_set_link(new_de, new_page, old_inode);
 		new_inode->i_ctime = CURRENT_TIME_SEC;
 		if (dir_de)
 			new_inode->i_nlink--;
-		dec_count(new_inode);
+		inode_dec_link_count(new_inode);
 	} else {
 		if (dir_de) {
 			err = -EMLINK;
 			if (new_dir->i_nlink >= SYSV_SB(new_dir->i_sb)->s_link_max)
 				goto out_dir;
 		}
-		inc_count(old_inode);
+		inode_inc_link_count(old_inode);
 		err = sysv_add_link(new_dentry, old_inode);
 		if (err) {
-			dec_count(old_inode);
+			inode_dec_link_count(old_inode);
 			goto out_dir;
 		}
 		if (dir_de)
-			inc_count(new_dir);
+			inode_inc_link_count(new_dir);
 	}
 
 	sysv_delete_entry(old_de, old_page);
-	dec_count(old_inode);
+	inode_dec_link_count(old_inode);
 
 	if (dir_de) {
 		sysv_set_link(dir_de, dir_page, new_dir);
-		dec_count(old_dir);
+		inode_dec_link_count(old_dir);
 	}
 	return 0;
 
diff --git a/fs/udf/balloc.c b/fs/udf/balloc.c
index 201049ac8a9..ea521f846d9 100644
--- a/fs/udf/balloc.c
+++ b/fs/udf/balloc.c
@@ -152,7 +152,7 @@ static void udf_bitmap_free_blocks(struct super_block * sb,
 	int bitmap_nr;
 	unsigned long overflow;
 
-	down(&sbi->s_alloc_sem);
+	mutex_lock(&sbi->s_alloc_mutex);
 	if (bloc.logicalBlockNum < 0 ||
 		(bloc.logicalBlockNum + count) > UDF_SB_PARTLEN(sb, bloc.partitionReferenceNum))
 	{
@@ -211,7 +211,7 @@ error_return:
 	sb->s_dirt = 1;
 	if (UDF_SB_LVIDBH(sb))
 		mark_buffer_dirty(UDF_SB_LVIDBH(sb));
-	up(&sbi->s_alloc_sem);
+	mutex_unlock(&sbi->s_alloc_mutex);
 	return;
 }
 
@@ -226,7 +226,7 @@ static int udf_bitmap_prealloc_blocks(struct super_block * sb,
 	int nr_groups, bitmap_nr;
 	struct buffer_head *bh;
 
-	down(&sbi->s_alloc_sem);
+	mutex_lock(&sbi->s_alloc_mutex);
 	if (first_block < 0 || first_block >= UDF_SB_PARTLEN(sb, partition))
 		goto out;
 
@@ -275,7 +275,7 @@ out:
 		mark_buffer_dirty(UDF_SB_LVIDBH(sb));
 	}
 	sb->s_dirt = 1;
-	up(&sbi->s_alloc_sem);
+	mutex_unlock(&sbi->s_alloc_mutex);
 	return alloc_count;
 }
 
@@ -291,7 +291,7 @@ static int udf_bitmap_new_block(struct super_block * sb,
 	int newblock = 0;
 
 	*err = -ENOSPC;
-	down(&sbi->s_alloc_sem);
+	mutex_lock(&sbi->s_alloc_mutex);
 
 repeat:
 	if (goal < 0 || goal >= UDF_SB_PARTLEN(sb, partition))
@@ -364,7 +364,7 @@ repeat:
 	}
 	if (i >= (nr_groups*2))
 	{
-		up(&sbi->s_alloc_sem);
+		mutex_unlock(&sbi->s_alloc_mutex);
 		return newblock;
 	}
 	if (bit < sb->s_blocksize << 3)
@@ -373,7 +373,7 @@ repeat:
 		bit = udf_find_next_one_bit(bh->b_data, sb->s_blocksize << 3, group_start << 3);
 	if (bit >= sb->s_blocksize << 3)
 	{
-		up(&sbi->s_alloc_sem);
+		mutex_unlock(&sbi->s_alloc_mutex);
 		return 0;
 	}
 
@@ -387,7 +387,7 @@ got_block:
 	 */
 	if (inode && DQUOT_ALLOC_BLOCK(inode, 1))
 	{
-		up(&sbi->s_alloc_sem);
+		mutex_unlock(&sbi->s_alloc_mutex);
 		*err = -EDQUOT;
 		return 0;
 	}
@@ -410,13 +410,13 @@ got_block:
 		mark_buffer_dirty(UDF_SB_LVIDBH(sb));
 	}
 	sb->s_dirt = 1;
-	up(&sbi->s_alloc_sem);
+	mutex_unlock(&sbi->s_alloc_mutex);
 	*err = 0;
 	return newblock;
 
 error_return:
 	*err = -EIO;
-	up(&sbi->s_alloc_sem);
+	mutex_unlock(&sbi->s_alloc_mutex);
 	return 0;
 }
 
@@ -433,7 +433,7 @@ static void udf_table_free_blocks(struct super_block * sb,
 	int8_t etype;
 	int i;
 
-	down(&sbi->s_alloc_sem);
+	mutex_lock(&sbi->s_alloc_mutex);
 	if (bloc.logicalBlockNum < 0 ||
 		(bloc.logicalBlockNum + count) > UDF_SB_PARTLEN(sb, bloc.partitionReferenceNum))
 	{
@@ -666,7 +666,7 @@ static void udf_table_free_blocks(struct super_block * sb,
 
 error_return:
 	sb->s_dirt = 1;
-	up(&sbi->s_alloc_sem);
+	mutex_unlock(&sbi->s_alloc_mutex);
 	return;
 }
 
@@ -692,7 +692,7 @@ static int udf_table_prealloc_blocks(struct super_block * sb,
 	else
 		return 0;
 
-	down(&sbi->s_alloc_sem);
+	mutex_lock(&sbi->s_alloc_mutex);
 	extoffset = sizeof(struct unallocSpaceEntry);
 	bloc = UDF_I_LOCATION(table);
 
@@ -736,7 +736,7 @@ static int udf_table_prealloc_blocks(struct super_block * sb,
 		mark_buffer_dirty(UDF_SB_LVIDBH(sb));
 		sb->s_dirt = 1;
 	}
-	up(&sbi->s_alloc_sem);
+	mutex_unlock(&sbi->s_alloc_mutex);
 	return alloc_count;
 }
 
@@ -761,7 +761,7 @@ static int udf_table_new_block(struct super_block * sb,
 	else
 		return newblock;
 
-	down(&sbi->s_alloc_sem);
+	mutex_lock(&sbi->s_alloc_mutex);
 	if (goal < 0 || goal >= UDF_SB_PARTLEN(sb, partition))
 		goal = 0;
 
@@ -811,7 +811,7 @@ static int udf_table_new_block(struct super_block * sb,
 	if (spread == 0xFFFFFFFF)
 	{
 		udf_release_data(goal_bh);
-		up(&sbi->s_alloc_sem);
+		mutex_unlock(&sbi->s_alloc_mutex);
 		return 0;
 	}
 
@@ -827,7 +827,7 @@ static int udf_table_new_block(struct super_block * sb,
 	if (inode && DQUOT_ALLOC_BLOCK(inode, 1))
 	{
 		udf_release_data(goal_bh);
-		up(&sbi->s_alloc_sem);
+		mutex_unlock(&sbi->s_alloc_mutex);
 		*err = -EDQUOT;
 		return 0;
 	}
@@ -846,7 +846,7 @@ static int udf_table_new_block(struct super_block * sb,
 	}
 
 	sb->s_dirt = 1;
-	up(&sbi->s_alloc_sem);
+	mutex_unlock(&sbi->s_alloc_mutex);
 	*err = 0;
 	return newblock;
 }
diff --git a/fs/udf/ialloc.c b/fs/udf/ialloc.c
index c9b707b470c..3873c672cb4 100644
--- a/fs/udf/ialloc.c
+++ b/fs/udf/ialloc.c
@@ -42,7 +42,7 @@ void udf_free_inode(struct inode * inode)
 
 	clear_inode(inode);
 
-	down(&sbi->s_alloc_sem);
+	mutex_lock(&sbi->s_alloc_mutex);
 	if (sbi->s_lvidbh) {
 		if (S_ISDIR(inode->i_mode))
 			UDF_SB_LVIDIU(sb)->numDirs =
@@ -53,7 +53,7 @@ void udf_free_inode(struct inode * inode)
 		
 		mark_buffer_dirty(sbi->s_lvidbh);
 	}
-	up(&sbi->s_alloc_sem);
+	mutex_unlock(&sbi->s_alloc_mutex);
 
 	udf_free_blocks(sb, NULL, UDF_I_LOCATION(inode), 0, 1);
 }
@@ -83,7 +83,7 @@ struct inode * udf_new_inode (struct inode *dir, int mode, int * err)
 		return NULL;
 	}
 
-	down(&sbi->s_alloc_sem);
+	mutex_lock(&sbi->s_alloc_mutex);
 	UDF_I_UNIQUE(inode) = 0;
 	UDF_I_LENEXTENTS(inode) = 0;
 	UDF_I_NEXT_ALLOC_BLOCK(inode) = 0;
@@ -148,7 +148,7 @@ struct inode * udf_new_inode (struct inode *dir, int mode, int * err)
 		UDF_I_CRTIME(inode) = current_fs_time(inode->i_sb);
 	insert_inode_hash(inode);
 	mark_inode_dirty(inode);
-	up(&sbi->s_alloc_sem);
+	mutex_unlock(&sbi->s_alloc_mutex);
 
 	if (DQUOT_ALLOC_INODE(inode))
 	{
diff --git a/fs/udf/super.c b/fs/udf/super.c
index 368d8f81fe5..9303c50c5d5 100644
--- a/fs/udf/super.c
+++ b/fs/udf/super.c
@@ -1515,7 +1515,7 @@ static int udf_fill_super(struct super_block *sb, void *options, int silent)
 	sb->s_fs_info = sbi;
 	memset(UDF_SB(sb), 0x00, sizeof(struct udf_sb_info));
 
-	init_MUTEX(&sbi->s_alloc_sem);
+	mutex_init(&sbi->s_alloc_mutex);
 
 	if (!udf_parse_options((char *)options, &uopt))
 		goto error_out;
diff --git a/fs/ufs/file.c b/fs/ufs/file.c
index ed69d7fe1b5..62ad481810e 100644
--- a/fs/ufs/file.c
+++ b/fs/ufs/file.c
@@ -23,18 +23,8 @@
  *  ext2 fs regular file handling primitives
  */
 
-#include <asm/uaccess.h>
-#include <asm/system.h>
-
-#include <linux/errno.h>
 #include <linux/fs.h>
 #include <linux/ufs_fs.h>
-#include <linux/fcntl.h>
-#include <linux/time.h>
-#include <linux/stat.h>
-#include <linux/mm.h>
-#include <linux/pagemap.h>
-#include <linux/smp_lock.h>
 
 /*
  * We have mostly NULL's here: the current defaults are ok for
diff --git a/fs/ufs/namei.c b/fs/ufs/namei.c
index 2958cde7d3d..8d5f98a01c7 100644
--- a/fs/ufs/namei.c
+++ b/fs/ufs/namei.c
@@ -43,18 +43,6 @@
 #define UFSD(x)
 #endif
 
-static inline void ufs_inc_count(struct inode *inode)
-{
-	inode->i_nlink++;
-	mark_inode_dirty(inode);
-}
-
-static inline void ufs_dec_count(struct inode *inode)
-{
-	inode->i_nlink--;
-	mark_inode_dirty(inode);
-}
-
 static inline int ufs_add_nondir(struct dentry *dentry, struct inode *inode)
 {
 	int err = ufs_add_link(dentry, inode);
@@ -62,7 +50,7 @@ static inline int ufs_add_nondir(struct dentry *dentry, struct inode *inode)
 		d_instantiate(dentry, inode);
 		return 0;
 	}
-	ufs_dec_count(inode);
+	inode_dec_link_count(inode);
 	iput(inode);
 	return err;
 }
@@ -173,7 +161,7 @@ out:
 	return err;
 
 out_fail:
-	ufs_dec_count(inode);
+	inode_dec_link_count(inode);
 	iput(inode);
 	goto out;
 }
@@ -191,7 +179,7 @@ static int ufs_link (struct dentry * old_dentry, struct inode * dir,
 	}
 
 	inode->i_ctime = CURRENT_TIME_SEC;
-	ufs_inc_count(inode);
+	inode_inc_link_count(inode);
 	atomic_inc(&inode->i_count);
 
 	error = ufs_add_nondir(dentry, inode);
@@ -208,7 +196,7 @@ static int ufs_mkdir(struct inode * dir, struct dentry * dentry, int mode)
 		goto out;
 
 	lock_kernel();
-	ufs_inc_count(dir);
+	inode_inc_link_count(dir);
 
 	inode = ufs_new_inode(dir, S_IFDIR|mode);
 	err = PTR_ERR(inode);
@@ -218,7 +206,7 @@ static int ufs_mkdir(struct inode * dir, struct dentry * dentry, int mode)
 	inode->i_op = &ufs_dir_inode_operations;
 	inode->i_fop = &ufs_dir_operations;
 
-	ufs_inc_count(inode);
+	inode_inc_link_count(inode);
 
 	err = ufs_make_empty(inode, dir);
 	if (err)
@@ -234,11 +222,11 @@ out:
 	return err;
 
 out_fail:
-	ufs_dec_count(inode);
-	ufs_dec_count(inode);
+	inode_dec_link_count(inode);
+	inode_dec_link_count(inode);
 	iput (inode);
 out_dir:
-	ufs_dec_count(dir);
+	inode_dec_link_count(dir);
 	unlock_kernel();
 	goto out;
 }
@@ -260,7 +248,7 @@ static int ufs_unlink(struct inode * dir, struct dentry *dentry)
 		goto out;
 
 	inode->i_ctime = dir->i_ctime;
-	ufs_dec_count(inode);
+	inode_dec_link_count(inode);
 	err = 0;
 out:
 	unlock_kernel();
@@ -277,8 +265,8 @@ static int ufs_rmdir (struct inode * dir, struct dentry *dentry)
 		err = ufs_unlink(dir, dentry);
 		if (!err) {
 			inode->i_size = 0;
-			ufs_dec_count(inode);
-			ufs_dec_count(dir);
+			inode_dec_link_count(inode);
+			inode_dec_link_count(dir);
 		}
 	}
 	unlock_kernel();
@@ -319,35 +307,35 @@ static int ufs_rename (struct inode * old_dir, struct dentry * old_dentry,
 		new_de = ufs_find_entry (new_dentry, &new_bh);
 		if (!new_de)
 			goto out_dir;
-		ufs_inc_count(old_inode);
+		inode_inc_link_count(old_inode);
 		ufs_set_link(new_dir, new_de, new_bh, old_inode);
 		new_inode->i_ctime = CURRENT_TIME_SEC;
 		if (dir_de)
 			new_inode->i_nlink--;
-		ufs_dec_count(new_inode);
+		inode_dec_link_count(new_inode);
 	} else {
 		if (dir_de) {
 			err = -EMLINK;
 			if (new_dir->i_nlink >= UFS_LINK_MAX)
 				goto out_dir;
 		}
-		ufs_inc_count(old_inode);
+		inode_inc_link_count(old_inode);
 		err = ufs_add_link(new_dentry, old_inode);
 		if (err) {
-			ufs_dec_count(old_inode);
+			inode_dec_link_count(old_inode);
 			goto out_dir;
 		}
 		if (dir_de)
-			ufs_inc_count(new_dir);
+			inode_inc_link_count(new_dir);
 	}
 
 	ufs_delete_entry (old_dir, old_de, old_bh);
 
-	ufs_dec_count(old_inode);
+	inode_dec_link_count(old_inode);
 
 	if (dir_de) {
 		ufs_set_link(old_inode, dir_de, dir_bh, new_dir);
-		ufs_dec_count(old_dir);
+		inode_dec_link_count(old_dir);
 	}
 	unlock_kernel();
 	return 0;
diff --git a/fs/xfs/linux-2.6/xfs_stats.c b/fs/xfs/linux-2.6/xfs_stats.c
index 8955720a2c6..713e6a7505d 100644
--- a/fs/xfs/linux-2.6/xfs_stats.c
+++ b/fs/xfs/linux-2.6/xfs_stats.c
@@ -62,18 +62,15 @@ xfs_read_xfsstats(
 		while (j < xstats[i].endpoint) {
 			val = 0;
 			/* sum over all cpus */
-			for (c = 0; c < NR_CPUS; c++) {
-				if (!cpu_possible(c)) continue;
+			for_each_cpu(c)
 				val += *(((__u32*)&per_cpu(xfsstats, c) + j));
-			}
 			len += sprintf(buffer + len, " %u", val);
 			j++;
 		}
 		buffer[len++] = '\n';
 	}
 	/* extra precision counters */
-	for (i = 0; i < NR_CPUS; i++) {
-		if (!cpu_possible(i)) continue;
+	for_each_cpu(i) {
 		xs_xstrat_bytes += per_cpu(xfsstats, i).xs_xstrat_bytes;
 		xs_write_bytes += per_cpu(xfsstats, i).xs_write_bytes;
 		xs_read_bytes += per_cpu(xfsstats, i).xs_read_bytes;
diff --git a/fs/xfs/linux-2.6/xfs_sysctl.c b/fs/xfs/linux-2.6/xfs_sysctl.c
index a0256497242..7079cc83721 100644
--- a/fs/xfs/linux-2.6/xfs_sysctl.c
+++ b/fs/xfs/linux-2.6/xfs_sysctl.c
@@ -38,8 +38,7 @@ xfs_stats_clear_proc_handler(
 
 	if (!ret && write && *valp) {
 		printk("XFS Clearing xfsstats\n");
-		for (c = 0; c < NR_CPUS; c++) {
-			if (!cpu_possible(c)) continue;
+		for_each_cpu(c) {
 			preempt_disable();
 			/* save vn_active, it's a universal truth! */
 			vn_active = per_cpu(xfsstats, c).vn_active;
diff --git a/include/asm-alpha/mmu_context.h b/include/asm-alpha/mmu_context.h
index 6f92482cc96..0c017fc181c 100644
--- a/include/asm-alpha/mmu_context.h
+++ b/include/asm-alpha/mmu_context.h
@@ -231,9 +231,8 @@ init_new_context(struct task_struct *tsk, struct mm_struct *mm)
 {
 	int i;
 
-	for (i = 0; i < NR_CPUS; i++)
-		if (cpu_online(i))
-			mm->context[i] = 0;
+	for_each_online_cpu(i)
+		mm->context[i] = 0;
 	if (tsk != current)
 		task_thread_info(tsk)->pcb.ptbr
 		  = ((unsigned long)mm->pgd - IDENT_ADDR) >> PAGE_SHIFT;
diff --git a/include/asm-alpha/topology.h b/include/asm-alpha/topology.h
index eb740e280d9..420ccde6b91 100644
--- a/include/asm-alpha/topology.h
+++ b/include/asm-alpha/topology.h
@@ -27,8 +27,8 @@ static inline cpumask_t node_to_cpumask(int node)
 	cpumask_t node_cpu_mask = CPU_MASK_NONE;
 	int cpu;
 
-	for(cpu = 0; cpu < NR_CPUS; cpu++) {
-		if (cpu_online(cpu) && (cpu_to_node(cpu) == node))
+	for_each_online_cpu(cpu) {
+		if (cpu_to_node(cpu) == node)
 			cpu_set(cpu, node_cpu_mask);
 	}
 
diff --git a/include/asm-generic/bug.h b/include/asm-generic/bug.h
index 400c2b41896..1a565a9d2fa 100644
--- a/include/asm-generic/bug.h
+++ b/include/asm-generic/bug.h
@@ -7,7 +7,7 @@
 #ifdef CONFIG_BUG
 #ifndef HAVE_ARCH_BUG
 #define BUG() do { \
-	printk("kernel BUG at %s:%d!\n", __FILE__, __LINE__); \
+	printk("BUG: failure at %s:%d/%s()!\n", __FILE__, __LINE__, __FUNCTION__); \
 	panic("BUG!"); \
 } while (0)
 #endif
@@ -19,7 +19,7 @@
 #ifndef HAVE_ARCH_WARN_ON
 #define WARN_ON(condition) do { \
 	if (unlikely((condition)!=0)) { \
-		printk("Badness in %s at %s:%d\n", __FUNCTION__, __FILE__, __LINE__); \
+		printk("BUG: warning at %s:%d/%s()\n", __FILE__, __LINE__, __FUNCTION__); \
 		dump_stack(); \
 	} \
 } while (0)
diff --git a/include/asm-generic/percpu.h b/include/asm-generic/percpu.h
index 9044aeb3782..78cf45547e3 100644
--- a/include/asm-generic/percpu.h
+++ b/include/asm-generic/percpu.h
@@ -19,10 +19,9 @@ extern unsigned long __per_cpu_offset[NR_CPUS];
 #define percpu_modcopy(pcpudst, src, size)			\
 do {								\
 	unsigned int __i;					\
-	for (__i = 0; __i < NR_CPUS; __i++)			\
-		if (cpu_possible(__i))				\
-			memcpy((pcpudst)+__per_cpu_offset[__i],	\
-			       (src), (size));			\
+	for_each_cpu(__i)					\
+		memcpy((pcpudst)+__per_cpu_offset[__i],		\
+		       (src), (size));				\
 } while (0)
 #else /* ! SMP */
 
diff --git a/include/asm-i386/alternative.h b/include/asm-i386/alternative.h
new file mode 100644
index 00000000000..e201decea0c
--- /dev/null
+++ b/include/asm-i386/alternative.h
@@ -0,0 +1,129 @@
+#ifndef _I386_ALTERNATIVE_H
+#define _I386_ALTERNATIVE_H
+
+#ifdef __KERNEL__
+
+struct alt_instr {
+	u8 *instr; 		/* original instruction */
+	u8 *replacement;
+	u8  cpuid;		/* cpuid bit set for replacement */
+	u8  instrlen;		/* length of original instruction */
+	u8  replacementlen; 	/* length of new instruction, <= instrlen */
+	u8  pad;
+};
+
+extern void apply_alternatives(struct alt_instr *start, struct alt_instr *end);
+
+struct module;
+extern void alternatives_smp_module_add(struct module *mod, char *name,
+					void *locks, void *locks_end,
+					void *text, void *text_end);
+extern void alternatives_smp_module_del(struct module *mod);
+extern void alternatives_smp_switch(int smp);
+
+#endif
+
+/*
+ * Alternative instructions for different CPU types or capabilities.
+ *
+ * This allows to use optimized instructions even on generic binary
+ * kernels.
+ *
+ * length of oldinstr must be longer or equal the length of newinstr
+ * It can be padded with nops as needed.
+ *
+ * For non barrier like inlines please define new variants
+ * without volatile and memory clobber.
+ */
+#define alternative(oldinstr, newinstr, feature)			\
+	asm volatile ("661:\n\t" oldinstr "\n662:\n" 			\
+		      ".section .altinstructions,\"a\"\n"		\
+		      "  .align 4\n"					\
+		      "  .long 661b\n"            /* label */		\
+		      "  .long 663f\n"		  /* new instruction */	\
+		      "  .byte %c0\n"             /* feature bit */	\
+		      "  .byte 662b-661b\n"       /* sourcelen */	\
+		      "  .byte 664f-663f\n"       /* replacementlen */	\
+		      ".previous\n"					\
+		      ".section .altinstr_replacement,\"ax\"\n"		\
+		      "663:\n\t" newinstr "\n664:\n"   /* replacement */\
+		      ".previous" :: "i" (feature) : "memory")
+
+/*
+ * Alternative inline assembly with input.
+ *
+ * Pecularities:
+ * No memory clobber here.
+ * Argument numbers start with 1.
+ * Best is to use constraints that are fixed size (like (%1) ... "r")
+ * If you use variable sized constraints like "m" or "g" in the
+ * replacement maake sure to pad to the worst case length.
+ */
+#define alternative_input(oldinstr, newinstr, feature, input...)	\
+	asm volatile ("661:\n\t" oldinstr "\n662:\n"			\
+		      ".section .altinstructions,\"a\"\n"		\
+		      "  .align 4\n"					\
+		      "  .long 661b\n"            /* label */		\
+		      "  .long 663f\n"		  /* new instruction */ \
+		      "  .byte %c0\n"             /* feature bit */	\
+		      "  .byte 662b-661b\n"       /* sourcelen */	\
+		      "  .byte 664f-663f\n"       /* replacementlen */ 	\
+		      ".previous\n"					\
+		      ".section .altinstr_replacement,\"ax\"\n"		\
+		      "663:\n\t" newinstr "\n664:\n"   /* replacement */\
+		      ".previous" :: "i" (feature), ##input)
+
+/*
+ * Alternative inline assembly for SMP.
+ *
+ * alternative_smp() takes two versions (SMP first, UP second) and is
+ * for more complex stuff such as spinlocks.
+ *
+ * The LOCK_PREFIX macro defined here replaces the LOCK and
+ * LOCK_PREFIX macros used everywhere in the source tree.
+ *
+ * SMP alternatives use the same data structures as the other
+ * alternatives and the X86_FEATURE_UP flag to indicate the case of a
+ * UP system running a SMP kernel.  The existing apply_alternatives()
+ * works fine for patching a SMP kernel for UP.
+ *
+ * The SMP alternative tables can be kept after boot and contain both
+ * UP and SMP versions of the instructions to allow switching back to
+ * SMP at runtime, when hotplugging in a new CPU, which is especially
+ * useful in virtualized environments.
+ *
+ * The very common lock prefix is handled as special case in a
+ * separate table which is a pure address list without replacement ptr
+ * and size information.  That keeps the table sizes small.
+ */
+
+#ifdef CONFIG_SMP
+#define alternative_smp(smpinstr, upinstr, args...)			\
+	asm volatile ("661:\n\t" smpinstr "\n662:\n" 			\
+		      ".section .smp_altinstructions,\"a\"\n"		\
+		      "  .align 4\n"					\
+		      "  .long 661b\n"            /* label */		\
+		      "  .long 663f\n"		  /* new instruction */	\
+		      "  .byte 0x68\n"            /* X86_FEATURE_UP */	\
+		      "  .byte 662b-661b\n"       /* sourcelen */	\
+		      "  .byte 664f-663f\n"       /* replacementlen */	\
+		      ".previous\n"					\
+		      ".section .smp_altinstr_replacement,\"awx\"\n"   	\
+		      "663:\n\t" upinstr "\n"     /* replacement */	\
+		      "664:\n\t.fill 662b-661b,1,0x42\n" /* space for original */ \
+		      ".previous" : args)
+
+#define LOCK_PREFIX \
+		".section .smp_locks,\"a\"\n"	\
+		"  .align 4\n"			\
+		"  .long 661f\n" /* address */	\
+		".previous\n"			\
+	       	"661:\n\tlock; "
+
+#else /* ! CONFIG_SMP */
+#define alternative_smp(smpinstr, upinstr, args...) \
+	asm volatile (upinstr : args)
+#define LOCK_PREFIX ""
+#endif
+
+#endif /* _I386_ALTERNATIVE_H */
diff --git a/include/asm-i386/arch_hooks.h b/include/asm-i386/arch_hooks.h
index 28b96a6fb9f..238cf4275b9 100644
--- a/include/asm-i386/arch_hooks.h
+++ b/include/asm-i386/arch_hooks.h
@@ -24,4 +24,7 @@ extern void trap_init_hook(void);
 extern void time_init_hook(void);
 extern void mca_nmi_hook(void);
 
+extern int setup_early_printk(char *);
+extern void early_printk(const char *fmt, ...) __attribute__((format(printf,1,2)));
+
 #endif
diff --git a/include/asm-i386/atomic.h b/include/asm-i386/atomic.h
index de649d3aa2d..22d80ece95c 100644
--- a/include/asm-i386/atomic.h
+++ b/include/asm-i386/atomic.h
@@ -10,12 +10,6 @@
  * resource counting etc..
  */
 
-#ifdef CONFIG_SMP
-#define LOCK "lock ; "
-#else
-#define LOCK ""
-#endif
-
 /*
  * Make sure gcc doesn't try to be clever and move things around
  * on us. We need to use _exactly_ the address the user gave us,
@@ -52,7 +46,7 @@ typedef struct { volatile int counter; } atomic_t;
 static __inline__ void atomic_add(int i, atomic_t *v)
 {
 	__asm__ __volatile__(
-		LOCK "addl %1,%0"
+		LOCK_PREFIX "addl %1,%0"
 		:"=m" (v->counter)
 		:"ir" (i), "m" (v->counter));
 }
@@ -67,7 +61,7 @@ static __inline__ void atomic_add(int i, atomic_t *v)
 static __inline__ void atomic_sub(int i, atomic_t *v)
 {
 	__asm__ __volatile__(
-		LOCK "subl %1,%0"
+		LOCK_PREFIX "subl %1,%0"
 		:"=m" (v->counter)
 		:"ir" (i), "m" (v->counter));
 }
@@ -86,7 +80,7 @@ static __inline__ int atomic_sub_and_test(int i, atomic_t *v)
 	unsigned char c;
 
 	__asm__ __volatile__(
-		LOCK "subl %2,%0; sete %1"
+		LOCK_PREFIX "subl %2,%0; sete %1"
 		:"=m" (v->counter), "=qm" (c)
 		:"ir" (i), "m" (v->counter) : "memory");
 	return c;
@@ -101,7 +95,7 @@ static __inline__ int atomic_sub_and_test(int i, atomic_t *v)
 static __inline__ void atomic_inc(atomic_t *v)
 {
 	__asm__ __volatile__(
-		LOCK "incl %0"
+		LOCK_PREFIX "incl %0"
 		:"=m" (v->counter)
 		:"m" (v->counter));
 }
@@ -115,7 +109,7 @@ static __inline__ void atomic_inc(atomic_t *v)
 static __inline__ void atomic_dec(atomic_t *v)
 {
 	__asm__ __volatile__(
-		LOCK "decl %0"
+		LOCK_PREFIX "decl %0"
 		:"=m" (v->counter)
 		:"m" (v->counter));
 }
@@ -133,7 +127,7 @@ static __inline__ int atomic_dec_and_test(atomic_t *v)
 	unsigned char c;
 
 	__asm__ __volatile__(
-		LOCK "decl %0; sete %1"
+		LOCK_PREFIX "decl %0; sete %1"
 		:"=m" (v->counter), "=qm" (c)
 		:"m" (v->counter) : "memory");
 	return c != 0;
@@ -152,7 +146,7 @@ static __inline__ int atomic_inc_and_test(atomic_t *v)
 	unsigned char c;
 
 	__asm__ __volatile__(
-		LOCK "incl %0; sete %1"
+		LOCK_PREFIX "incl %0; sete %1"
 		:"=m" (v->counter), "=qm" (c)
 		:"m" (v->counter) : "memory");
 	return c != 0;
@@ -172,7 +166,7 @@ static __inline__ int atomic_add_negative(int i, atomic_t *v)
 	unsigned char c;
 
 	__asm__ __volatile__(
-		LOCK "addl %2,%0; sets %1"
+		LOCK_PREFIX "addl %2,%0; sets %1"
 		:"=m" (v->counter), "=qm" (c)
 		:"ir" (i), "m" (v->counter) : "memory");
 	return c;
@@ -195,7 +189,7 @@ static __inline__ int atomic_add_return(int i, atomic_t *v)
 	/* Modern 486+ processor */
 	__i = i;
 	__asm__ __volatile__(
-		LOCK "xaddl %0, %1;"
+		LOCK_PREFIX "xaddl %0, %1;"
 		:"=r"(i)
 		:"m"(v->counter), "0"(i));
 	return i + __i;
@@ -231,8 +225,14 @@ static __inline__ int atomic_sub_return(int i, atomic_t *v)
 ({								\
 	int c, old;						\
 	c = atomic_read(v);					\
-	while (c != (u) && (old = atomic_cmpxchg((v), c, c + (a))) != c) \
+	for (;;) {						\
+		if (unlikely(c == (u)))				\
+			break;					\
+		old = atomic_cmpxchg((v), c, c + (a));		\
+		if (likely(old == c))				\
+			break;					\
 		c = old;					\
+	}							\
 	c != (u);						\
 })
 #define atomic_inc_not_zero(v) atomic_add_unless((v), 1, 0)
@@ -242,11 +242,11 @@ static __inline__ int atomic_sub_return(int i, atomic_t *v)
 
 /* These are x86-specific, used by some header files */
 #define atomic_clear_mask(mask, addr) \
-__asm__ __volatile__(LOCK "andl %0,%1" \
+__asm__ __volatile__(LOCK_PREFIX "andl %0,%1" \
 : : "r" (~(mask)),"m" (*addr) : "memory")
 
 #define atomic_set_mask(mask, addr) \
-__asm__ __volatile__(LOCK "orl %0,%1" \
+__asm__ __volatile__(LOCK_PREFIX "orl %0,%1" \
 : : "r" (mask),"m" (*(addr)) : "memory")
 
 /* Atomic operations are already serializing on x86 */
diff --git a/include/asm-i386/bitops.h b/include/asm-i386/bitops.h
index 88e6ca248cd..7d20b95edb3 100644
--- a/include/asm-i386/bitops.h
+++ b/include/asm-i386/bitops.h
@@ -7,6 +7,7 @@
 
 #include <linux/config.h>
 #include <linux/compiler.h>
+#include <asm/alternative.h>
 
 /*
  * These have to be done with inline assembly: that way the bit-setting
@@ -16,12 +17,6 @@
  * bit 0 is the LSB of addr; bit 32 is the LSB of (addr+1).
  */
 
-#ifdef CONFIG_SMP
-#define LOCK_PREFIX "lock ; "
-#else
-#define LOCK_PREFIX ""
-#endif
-
 #define ADDR (*(volatile long *) addr)
 
 /**
diff --git a/include/asm-i386/cache.h b/include/asm-i386/cache.h
index 615911e5bd2..ca15c9c665c 100644
--- a/include/asm-i386/cache.h
+++ b/include/asm-i386/cache.h
@@ -10,4 +10,6 @@
 #define L1_CACHE_SHIFT	(CONFIG_X86_L1_CACHE_SHIFT)
 #define L1_CACHE_BYTES	(1 << L1_CACHE_SHIFT)
 
+#define __read_mostly __attribute__((__section__(".data.read_mostly")))
+
 #endif
diff --git a/include/asm-i386/cpufeature.h b/include/asm-i386/cpufeature.h
index c4ec2a4d8fd..5c0b5876b93 100644
--- a/include/asm-i386/cpufeature.h
+++ b/include/asm-i386/cpufeature.h
@@ -70,6 +70,7 @@
 #define X86_FEATURE_P3		(3*32+ 6) /* P3 */
 #define X86_FEATURE_P4		(3*32+ 7) /* P4 */
 #define X86_FEATURE_CONSTANT_TSC (3*32+ 8) /* TSC ticks at a constant rate */
+#define X86_FEATURE_UP		(3*32+ 9) /* smp kernel running on up */
 
 /* Intel-defined CPU features, CPUID level 0x00000001 (ecx), word 4 */
 #define X86_FEATURE_XMM3	(4*32+ 0) /* Streaming SIMD Extensions-3 */
diff --git a/include/asm-i386/mach-default/do_timer.h b/include/asm-i386/mach-default/do_timer.h
index 56211414fc9..6312c3e7981 100644
--- a/include/asm-i386/mach-default/do_timer.h
+++ b/include/asm-i386/mach-default/do_timer.h
@@ -18,7 +18,7 @@ static inline void do_timer_interrupt_hook(struct pt_regs *regs)
 {
 	do_timer(regs);
 #ifndef CONFIG_SMP
-	update_process_times(user_mode(regs));
+	update_process_times(user_mode_vm(regs));
 #endif
 /*
  * In the SMP case we use the local APIC timer interrupt to do the
diff --git a/include/asm-i386/mach-es7000/mach_mpparse.h b/include/asm-i386/mach-es7000/mach_mpparse.h
index 4a0637a3e20..99f66be240b 100644
--- a/include/asm-i386/mach-es7000/mach_mpparse.h
+++ b/include/asm-i386/mach-es7000/mach_mpparse.h
@@ -30,7 +30,8 @@ static inline int mps_oem_check(struct mp_config_table *mpc, char *oem,
 	return 0;
 }
 
-static inline int es7000_check_dsdt()
+#ifdef CONFIG_ACPI
+static inline int es7000_check_dsdt(void)
 {
 	struct acpi_table_header *header = NULL;
 	if(!acpi_get_table_header_early(ACPI_DSDT, &header))
@@ -54,6 +55,11 @@ static inline int acpi_madt_oem_check(char *oem_id, char *oem_table_id)
 	}
 	return 0;
 }
-
+#else
+static inline int acpi_madt_oem_check(char *oem_id, char *oem_table_id)
+{
+	return 0;
+}
+#endif
 
 #endif /* __ASM_MACH_MPPARSE_H */
diff --git a/include/asm-i386/mach-visws/do_timer.h b/include/asm-i386/mach-visws/do_timer.h
index 92d638fc8b1..95568e6ca91 100644
--- a/include/asm-i386/mach-visws/do_timer.h
+++ b/include/asm-i386/mach-visws/do_timer.h
@@ -11,7 +11,7 @@ static inline void do_timer_interrupt_hook(struct pt_regs *regs)
 
 	do_timer(regs);
 #ifndef CONFIG_SMP
-	update_process_times(user_mode(regs));
+	update_process_times(user_mode_vm(regs));
 #endif
 /*
  * In the SMP case we use the local APIC timer interrupt to do the
diff --git a/include/asm-i386/mach-voyager/do_timer.h b/include/asm-i386/mach-voyager/do_timer.h
index ae510e5d0d7..eaf51809898 100644
--- a/include/asm-i386/mach-voyager/do_timer.h
+++ b/include/asm-i386/mach-voyager/do_timer.h
@@ -5,7 +5,7 @@ static inline void do_timer_interrupt_hook(struct pt_regs *regs)
 {
 	do_timer(regs);
 #ifndef CONFIG_SMP
-	update_process_times(user_mode(regs));
+	update_process_times(user_mode_vm(regs));
 #endif
 
 	voyager_timer_interrupt(regs);
diff --git a/include/asm-i386/mpspec.h b/include/asm-i386/mpspec.h
index 64a0b8e6afe..62113d3bfdc 100644
--- a/include/asm-i386/mpspec.h
+++ b/include/asm-i386/mpspec.h
@@ -22,7 +22,6 @@ extern int mp_bus_id_to_type [MAX_MP_BUSSES];
 extern int mp_irq_entries;
 extern struct mpc_config_intsrc mp_irqs [MAX_IRQ_SOURCES];
 extern int mpc_default_type;
-extern int mp_bus_id_to_pci_bus [MAX_MP_BUSSES];
 extern unsigned long mp_lapic_addr;
 extern int pic_mode;
 extern int using_apic_timer;
diff --git a/include/asm-i386/mtrr.h b/include/asm-i386/mtrr.h
index 5b6ceda68c5..64cf937c7e3 100644
--- a/include/asm-i386/mtrr.h
+++ b/include/asm-i386/mtrr.h
@@ -25,6 +25,7 @@
 
 #include <linux/config.h>
 #include <linux/ioctl.h>
+#include <linux/errno.h>
 
 #define	MTRR_IOCTL_BASE	'M'
 
diff --git a/include/asm-i386/mutex.h b/include/asm-i386/mutex.h
index 9b2199e829f..05a53853122 100644
--- a/include/asm-i386/mutex.h
+++ b/include/asm-i386/mutex.h
@@ -9,6 +9,8 @@
 #ifndef _ASM_MUTEX_H
 #define _ASM_MUTEX_H
 
+#include "asm/alternative.h"
+
 /**
  *  __mutex_fastpath_lock - try to take the lock by moving the count
  *                          from 1 to a 0 value
@@ -27,7 +29,7 @@ do {									\
 	typecheck_fn(fastcall void (*)(atomic_t *), fail_fn);		\
 									\
 	__asm__ __volatile__(						\
-		LOCK	"   decl (%%eax)	\n"			\
+		LOCK_PREFIX "   decl (%%eax)	\n"			\
 			"   js 2f		\n"			\
 			"1:			\n"			\
 									\
@@ -83,7 +85,7 @@ do {									\
 	typecheck_fn(fastcall void (*)(atomic_t *), fail_fn);		\
 									\
 	__asm__ __volatile__(						\
-		LOCK	"   incl (%%eax)	\n"			\
+		LOCK_PREFIX "   incl (%%eax)	\n"			\
 			"   jle 2f		\n"			\
 			"1:			\n"			\
 									\
diff --git a/include/asm-i386/pgtable-2level.h b/include/asm-i386/pgtable-2level.h
index 74ef721b534..27bde973abc 100644
--- a/include/asm-i386/pgtable-2level.h
+++ b/include/asm-i386/pgtable-2level.h
@@ -61,4 +61,6 @@ static inline int pte_exec_kernel(pte_t pte)
 #define __pte_to_swp_entry(pte)		((swp_entry_t) { (pte).pte_low })
 #define __swp_entry_to_pte(x)		((pte_t) { (x).val })
 
+void vmalloc_sync_all(void);
+
 #endif /* _I386_PGTABLE_2LEVEL_H */
diff --git a/include/asm-i386/pgtable-3level.h b/include/asm-i386/pgtable-3level.h
index f1a8b454920..36a5aa63cbb 100644
--- a/include/asm-i386/pgtable-3level.h
+++ b/include/asm-i386/pgtable-3level.h
@@ -152,4 +152,6 @@ static inline pmd_t pfn_pmd(unsigned long page_nr, pgprot_t pgprot)
 
 #define __pmd_free_tlb(tlb, x)		do { } while (0)
 
+#define vmalloc_sync_all() ((void)0)
+
 #endif /* _I386_PGTABLE_3LEVEL_H */
diff --git a/include/asm-i386/rwlock.h b/include/asm-i386/rwlock.h
index b57cc7afdf7..94f00195d54 100644
--- a/include/asm-i386/rwlock.h
+++ b/include/asm-i386/rwlock.h
@@ -21,21 +21,23 @@
 #define RW_LOCK_BIAS_STR	"0x01000000"
 
 #define __build_read_lock_ptr(rw, helper)   \
-	asm volatile(LOCK "subl $1,(%0)\n\t" \
-		     "jns 1f\n" \
-		     "call " helper "\n\t" \
-		     "1:\n" \
-		     ::"a" (rw) : "memory")
+	alternative_smp("lock; subl $1,(%0)\n\t" \
+			"jns 1f\n" \
+			"call " helper "\n\t" \
+			"1:\n", \
+			"subl $1,(%0)\n\t", \
+			:"a" (rw) : "memory")
 
 #define __build_read_lock_const(rw, helper)   \
-	asm volatile(LOCK "subl $1,%0\n\t" \
-		     "jns 1f\n" \
-		     "pushl %%eax\n\t" \
-		     "leal %0,%%eax\n\t" \
-		     "call " helper "\n\t" \
-		     "popl %%eax\n\t" \
-		     "1:\n" \
-		     :"=m" (*(volatile int *)rw) : : "memory")
+	alternative_smp("lock; subl $1,%0\n\t" \
+			"jns 1f\n" \
+			"pushl %%eax\n\t" \
+			"leal %0,%%eax\n\t" \
+			"call " helper "\n\t" \
+			"popl %%eax\n\t" \
+			"1:\n", \
+			"subl $1,%0\n\t", \
+			"=m" (*(volatile int *)rw) : : "memory")
 
 #define __build_read_lock(rw, helper)	do { \
 						if (__builtin_constant_p(rw)) \
@@ -45,21 +47,23 @@
 					} while (0)
 
 #define __build_write_lock_ptr(rw, helper) \
-	asm volatile(LOCK "subl $" RW_LOCK_BIAS_STR ",(%0)\n\t" \
-		     "jz 1f\n" \
-		     "call " helper "\n\t" \
-		     "1:\n" \
-		     ::"a" (rw) : "memory")
+	alternative_smp("lock; subl $" RW_LOCK_BIAS_STR ",(%0)\n\t" \
+			"jz 1f\n" \
+			"call " helper "\n\t" \
+			"1:\n", \
+			"subl $" RW_LOCK_BIAS_STR ",(%0)\n\t", \
+			:"a" (rw) : "memory")
 
 #define __build_write_lock_const(rw, helper) \
-	asm volatile(LOCK "subl $" RW_LOCK_BIAS_STR ",%0\n\t" \
-		     "jz 1f\n" \
-		     "pushl %%eax\n\t" \
-		     "leal %0,%%eax\n\t" \
-		     "call " helper "\n\t" \
-		     "popl %%eax\n\t" \
-		     "1:\n" \
-		     :"=m" (*(volatile int *)rw) : : "memory")
+	alternative_smp("lock; subl $" RW_LOCK_BIAS_STR ",%0\n\t" \
+			"jz 1f\n" \
+			"pushl %%eax\n\t" \
+			"leal %0,%%eax\n\t" \
+			"call " helper "\n\t" \
+			"popl %%eax\n\t" \
+			"1:\n", \
+			"subl $" RW_LOCK_BIAS_STR ",%0\n\t", \
+			"=m" (*(volatile int *)rw) : : "memory")
 
 #define __build_write_lock(rw, helper)	do { \
 						if (__builtin_constant_p(rw)) \
diff --git a/include/asm-i386/semaphore.h b/include/asm-i386/semaphore.h
index 6a42b2142fd..f7a0f310c52 100644
--- a/include/asm-i386/semaphore.h
+++ b/include/asm-i386/semaphore.h
@@ -99,7 +99,7 @@ static inline void down(struct semaphore * sem)
 	might_sleep();
 	__asm__ __volatile__(
 		"# atomic down operation\n\t"
-		LOCK "decl %0\n\t"     /* --sem->count */
+		LOCK_PREFIX "decl %0\n\t"     /* --sem->count */
 		"js 2f\n"
 		"1:\n"
 		LOCK_SECTION_START("")
@@ -123,7 +123,7 @@ static inline int down_interruptible(struct semaphore * sem)
 	might_sleep();
 	__asm__ __volatile__(
 		"# atomic interruptible down operation\n\t"
-		LOCK "decl %1\n\t"     /* --sem->count */
+		LOCK_PREFIX "decl %1\n\t"     /* --sem->count */
 		"js 2f\n\t"
 		"xorl %0,%0\n"
 		"1:\n"
@@ -148,7 +148,7 @@ static inline int down_trylock(struct semaphore * sem)
 
 	__asm__ __volatile__(
 		"# atomic interruptible down operation\n\t"
-		LOCK "decl %1\n\t"     /* --sem->count */
+		LOCK_PREFIX "decl %1\n\t"     /* --sem->count */
 		"js 2f\n\t"
 		"xorl %0,%0\n"
 		"1:\n"
@@ -173,7 +173,7 @@ static inline void up(struct semaphore * sem)
 {
 	__asm__ __volatile__(
 		"# atomic up operation\n\t"
-		LOCK "incl %0\n\t"     /* ++sem->count */
+		LOCK_PREFIX "incl %0\n\t"     /* ++sem->count */
 		"jle 2f\n"
 		"1:\n"
 		LOCK_SECTION_START("")
diff --git a/include/asm-i386/spinlock.h b/include/asm-i386/spinlock.h
index 23604350cdf..d76b7693cf1 100644
--- a/include/asm-i386/spinlock.h
+++ b/include/asm-i386/spinlock.h
@@ -35,31 +35,41 @@
 #define __raw_spin_lock_string_flags \
 	"\n1:\t" \
 	"lock ; decb %0\n\t" \
-	"jns 4f\n\t" \
+	"jns 5f\n" \
 	"2:\t" \
 	"testl $0x200, %1\n\t" \
-	"jz 3f\n\t" \
-	"sti\n\t" \
+	"jz 4f\n\t" \
+	"sti\n" \
 	"3:\t" \
 	"rep;nop\n\t" \
 	"cmpb $0, %0\n\t" \
 	"jle 3b\n\t" \
 	"cli\n\t" \
 	"jmp 1b\n" \
-	"4:\n\t"
+	"4:\t" \
+	"rep;nop\n\t" \
+	"cmpb $0, %0\n\t" \
+	"jg 1b\n\t" \
+	"jmp 4b\n" \
+	"5:\n\t"
+
+#define __raw_spin_lock_string_up \
+	"\n\tdecb %0"
 
 static inline void __raw_spin_lock(raw_spinlock_t *lock)
 {
-	__asm__ __volatile__(
-		__raw_spin_lock_string
-		:"=m" (lock->slock) : : "memory");
+	alternative_smp(
+		__raw_spin_lock_string,
+		__raw_spin_lock_string_up,
+		"=m" (lock->slock) : : "memory");
 }
 
 static inline void __raw_spin_lock_flags(raw_spinlock_t *lock, unsigned long flags)
 {
-	__asm__ __volatile__(
-		__raw_spin_lock_string_flags
-		:"=m" (lock->slock) : "r" (flags) : "memory");
+	alternative_smp(
+		__raw_spin_lock_string_flags,
+		__raw_spin_lock_string_up,
+		"=m" (lock->slock) : "r" (flags) : "memory");
 }
 
 static inline int __raw_spin_trylock(raw_spinlock_t *lock)
@@ -178,12 +188,12 @@ static inline int __raw_write_trylock(raw_rwlock_t *lock)
 
 static inline void __raw_read_unlock(raw_rwlock_t *rw)
 {
-	asm volatile("lock ; incl %0" :"=m" (rw->lock) : : "memory");
+	asm volatile(LOCK_PREFIX "incl %0" :"=m" (rw->lock) : : "memory");
 }
 
 static inline void __raw_write_unlock(raw_rwlock_t *rw)
 {
-	asm volatile("lock ; addl $" RW_LOCK_BIAS_STR ", %0"
+	asm volatile(LOCK_PREFIX "addl $" RW_LOCK_BIAS_STR ", %0"
 				 : "=m" (rw->lock) : : "memory");
 }
 
diff --git a/include/asm-i386/system.h b/include/asm-i386/system.h
index 399145a247f..d0d8d7448d8 100644
--- a/include/asm-i386/system.h
+++ b/include/asm-i386/system.h
@@ -352,67 +352,6 @@ static inline unsigned long long __cmpxchg64(volatile void *ptr, unsigned long l
 
 #endif
     
-#ifdef __KERNEL__
-struct alt_instr { 
-	__u8 *instr; 		/* original instruction */
-	__u8 *replacement;
-	__u8  cpuid;		/* cpuid bit set for replacement */
-	__u8  instrlen;		/* length of original instruction */
-	__u8  replacementlen; 	/* length of new instruction, <= instrlen */ 
-	__u8  pad;
-}; 
-#endif
-
-/* 
- * Alternative instructions for different CPU types or capabilities.
- * 
- * This allows to use optimized instructions even on generic binary
- * kernels.
- * 
- * length of oldinstr must be longer or equal the length of newinstr
- * It can be padded with nops as needed.
- * 
- * For non barrier like inlines please define new variants
- * without volatile and memory clobber.
- */
-#define alternative(oldinstr, newinstr, feature) 	\
-	asm volatile ("661:\n\t" oldinstr "\n662:\n" 		     \
-		      ".section .altinstructions,\"a\"\n"     	     \
-		      "  .align 4\n"				       \
-		      "  .long 661b\n"            /* label */          \
-		      "  .long 663f\n"		  /* new instruction */ 	\
-		      "  .byte %c0\n"             /* feature bit */    \
-		      "  .byte 662b-661b\n"       /* sourcelen */      \
-		      "  .byte 664f-663f\n"       /* replacementlen */ \
-		      ".previous\n"						\
-		      ".section .altinstr_replacement,\"ax\"\n"			\
-		      "663:\n\t" newinstr "\n664:\n"   /* replacement */    \
-		      ".previous" :: "i" (feature) : "memory")  
-
-/*
- * Alternative inline assembly with input.
- * 
- * Pecularities:
- * No memory clobber here. 
- * Argument numbers start with 1.
- * Best is to use constraints that are fixed size (like (%1) ... "r")
- * If you use variable sized constraints like "m" or "g" in the 
- * replacement maake sure to pad to the worst case length.
- */
-#define alternative_input(oldinstr, newinstr, feature, input...)		\
-	asm volatile ("661:\n\t" oldinstr "\n662:\n"				\
-		      ".section .altinstructions,\"a\"\n"			\
-		      "  .align 4\n"						\
-		      "  .long 661b\n"            /* label */			\
-		      "  .long 663f\n"		  /* new instruction */ 	\
-		      "  .byte %c0\n"             /* feature bit */		\
-		      "  .byte 662b-661b\n"       /* sourcelen */		\
-		      "  .byte 664f-663f\n"       /* replacementlen */ 		\
-		      ".previous\n"						\
-		      ".section .altinstr_replacement,\"ax\"\n"			\
-		      "663:\n\t" newinstr "\n664:\n"   /* replacement */ 	\
-		      ".previous" :: "i" (feature), ##input)
-
 /*
  * Force strict CPU ordering.
  * And yes, this is required on UP too when we're talking
@@ -558,5 +497,6 @@ static inline void sched_cacheflush(void)
 }
 
 extern unsigned long arch_align_stack(unsigned long sp);
+extern void free_init_pages(char *what, unsigned long begin, unsigned long end);
 
 #endif
diff --git a/include/asm-i386/uaccess.h b/include/asm-i386/uaccess.h
index 3f1337c3420..371457b1ceb 100644
--- a/include/asm-i386/uaccess.h
+++ b/include/asm-i386/uaccess.h
@@ -197,13 +197,15 @@ extern void __put_user_8(void);
 
 #define put_user(x,ptr)						\
 ({	int __ret_pu;						\
+	__typeof__(*(ptr)) __pu_val;				\
 	__chk_user_ptr(ptr);					\
+	__pu_val = x;						\
 	switch(sizeof(*(ptr))) {				\
-	case 1: __put_user_1(x, ptr); break;			\
-	case 2: __put_user_2(x, ptr); break;			\
-	case 4: __put_user_4(x, ptr); break;			\
-	case 8: __put_user_8(x, ptr); break;			\
-	default:__put_user_X(x, ptr); break;			\
+	case 1: __put_user_1(__pu_val, ptr); break;		\
+	case 2: __put_user_2(__pu_val, ptr); break;		\
+	case 4: __put_user_4(__pu_val, ptr); break;		\
+	case 8: __put_user_8(__pu_val, ptr); break;		\
+	default:__put_user_X(__pu_val, ptr); break;		\
 	}							\
 	__ret_pu;						\
 })
diff --git a/include/asm-i386/unistd.h b/include/asm-i386/unistd.h
index dc81a55dd94..d8afd0e3b81 100644
--- a/include/asm-i386/unistd.h
+++ b/include/asm-i386/unistd.h
@@ -347,9 +347,9 @@ __syscall_return(type,__res); \
 type name(type1 arg1) \
 { \
 long __res; \
-__asm__ volatile ("int $0x80" \
+__asm__ volatile ("push %%ebx ; movl %2,%%ebx ; int $0x80 ; pop %%ebx" \
 	: "=a" (__res) \
-	: "0" (__NR_##name),"b" ((long)(arg1)) : "memory"); \
+	: "0" (__NR_##name),"ri" ((long)(arg1)) : "memory"); \
 __syscall_return(type,__res); \
 }
 
@@ -357,9 +357,10 @@ __syscall_return(type,__res); \
 type name(type1 arg1,type2 arg2) \
 { \
 long __res; \
-__asm__ volatile ("int $0x80" \
+__asm__ volatile ("push %%ebx ; movl %2,%%ebx ; int $0x80 ; pop %%ebx" \
 	: "=a" (__res) \
-	: "0" (__NR_##name),"b" ((long)(arg1)),"c" ((long)(arg2)) : "memory"); \
+	: "0" (__NR_##name),"ri" ((long)(arg1)),"c" ((long)(arg2)) \
+	: "memory"); \
 __syscall_return(type,__res); \
 }
 
@@ -367,9 +368,9 @@ __syscall_return(type,__res); \
 type name(type1 arg1,type2 arg2,type3 arg3) \
 { \
 long __res; \
-__asm__ volatile ("int $0x80" \
+__asm__ volatile ("push %%ebx ; movl %2,%%ebx ; int $0x80 ; pop %%ebx" \
 	: "=a" (__res) \
-	: "0" (__NR_##name),"b" ((long)(arg1)),"c" ((long)(arg2)), \
+	: "0" (__NR_##name),"ri" ((long)(arg1)),"c" ((long)(arg2)), \
 		  "d" ((long)(arg3)) : "memory"); \
 __syscall_return(type,__res); \
 }
@@ -378,9 +379,9 @@ __syscall_return(type,__res); \
 type name (type1 arg1, type2 arg2, type3 arg3, type4 arg4) \
 { \
 long __res; \
-__asm__ volatile ("int $0x80" \
+__asm__ volatile ("push %%ebx ; movl %2,%%ebx ; int $0x80 ; pop %%ebx" \
 	: "=a" (__res) \
-	: "0" (__NR_##name),"b" ((long)(arg1)),"c" ((long)(arg2)), \
+	: "0" (__NR_##name),"ri" ((long)(arg1)),"c" ((long)(arg2)), \
 	  "d" ((long)(arg3)),"S" ((long)(arg4)) : "memory"); \
 __syscall_return(type,__res); \
 } 
@@ -390,10 +391,12 @@ __syscall_return(type,__res); \
 type name (type1 arg1,type2 arg2,type3 arg3,type4 arg4,type5 arg5) \
 { \
 long __res; \
-__asm__ volatile ("int $0x80" \
+__asm__ volatile ("push %%ebx ; movl %2,%%ebx ; movl %1,%%eax ; " \
+                  "int $0x80 ; pop %%ebx" \
 	: "=a" (__res) \
-	: "0" (__NR_##name),"b" ((long)(arg1)),"c" ((long)(arg2)), \
-	  "d" ((long)(arg3)),"S" ((long)(arg4)),"D" ((long)(arg5)) : "memory"); \
+	: "i" (__NR_##name),"ri" ((long)(arg1)),"c" ((long)(arg2)), \
+	  "d" ((long)(arg3)),"S" ((long)(arg4)),"D" ((long)(arg5)) \
+	: "memory"); \
 __syscall_return(type,__res); \
 }
 
@@ -402,11 +405,14 @@ __syscall_return(type,__res); \
 type name (type1 arg1,type2 arg2,type3 arg3,type4 arg4,type5 arg5,type6 arg6) \
 { \
 long __res; \
-__asm__ volatile ("push %%ebp ; movl %%eax,%%ebp ; movl %1,%%eax ; int $0x80 ; pop %%ebp" \
+  struct { long __a1; long __a6; } __s = { (long)arg1, (long)arg6 }; \
+__asm__ volatile ("push %%ebp ; push %%ebx ; movl 4(%2),%%ebp ; " \
+                  "movl 0(%2),%%ebx ; movl %1,%%eax ; int $0x80 ; " \
+                  "pop %%ebx ;  pop %%ebp" \
 	: "=a" (__res) \
-	: "i" (__NR_##name),"b" ((long)(arg1)),"c" ((long)(arg2)), \
-	  "d" ((long)(arg3)),"S" ((long)(arg4)),"D" ((long)(arg5)), \
-	  "0" ((long)(arg6)) : "memory"); \
+	: "i" (__NR_##name),"0" ((long)(&__s)),"c" ((long)(arg2)), \
+	  "d" ((long)(arg3)),"S" ((long)(arg4)),"D" ((long)(arg5)) \
+	: "memory"); \
 __syscall_return(type,__res); \
 }
 
diff --git a/include/asm-ia64/atomic.h b/include/asm-ia64/atomic.h
index d3e0dfa99e1..569ec7574ba 100644
--- a/include/asm-ia64/atomic.h
+++ b/include/asm-ia64/atomic.h
@@ -95,8 +95,14 @@ ia64_atomic64_sub (__s64 i, atomic64_t *v)
 ({								\
 	int c, old;						\
 	c = atomic_read(v);					\
-	while (c != (u) && (old = atomic_cmpxchg((v), c, c + (a))) != c) \
+	for (;;) {						\
+		if (unlikely(c == (u)))				\
+			break;					\
+		old = atomic_cmpxchg((v), c, c + (a));		\
+		if (likely(old == c))				\
+			break;					\
 		c = old;					\
+	}							\
 	c != (u);						\
 })
 #define atomic_inc_not_zero(v) atomic_add_unless((v), 1, 0)
diff --git a/include/asm-ia64/cache.h b/include/asm-ia64/cache.h
index 40dd25195d6..f0a104db8f2 100644
--- a/include/asm-ia64/cache.h
+++ b/include/asm-ia64/cache.h
@@ -25,4 +25,6 @@
 # define SMP_CACHE_BYTES	(1 << 3)
 #endif
 
+#define __read_mostly __attribute__((__section__(".data.read_mostly")))
+
 #endif /* _ASM_IA64_CACHE_H */
diff --git a/include/asm-m68k/atomic.h b/include/asm-m68k/atomic.h
index 862e497c264..732d696d31a 100644
--- a/include/asm-m68k/atomic.h
+++ b/include/asm-m68k/atomic.h
@@ -175,8 +175,14 @@ static inline void atomic_set_mask(unsigned long mask, unsigned long *v)
 ({								\
 	int c, old;						\
 	c = atomic_read(v);					\
-	while (c != (u) && (old = atomic_cmpxchg((v), c, c + (a))) != c) \
+	for (;;) {						\
+		if (unlikely(c == (u)))				\
+			break;					\
+		old = atomic_cmpxchg((v), c, c + (a));		\
+		if (likely(old == c))				\
+			break;					\
 		c = old;					\
+	}							\
 	c != (u);						\
 })
 #define atomic_inc_not_zero(v) atomic_add_unless((v), 1, 0)
diff --git a/include/asm-parisc/cache.h b/include/asm-parisc/cache.h
index 93f179f13ce..ae50f8e12ee 100644
--- a/include/asm-parisc/cache.h
+++ b/include/asm-parisc/cache.h
@@ -29,6 +29,8 @@
 
 #define SMP_CACHE_BYTES L1_CACHE_BYTES
 
+#define __read_mostly __attribute__((__section__(".data.read_mostly")))
+
 extern void flush_data_cache_local(void *);  /* flushes local data-cache only */
 extern void flush_instruction_cache_local(void *); /* flushes local code-cache only */
 #ifdef CONFIG_SMP
diff --git a/include/asm-powerpc/percpu.h b/include/asm-powerpc/percpu.h
index e31922c50e5..464301cd0d0 100644
--- a/include/asm-powerpc/percpu.h
+++ b/include/asm-powerpc/percpu.h
@@ -27,10 +27,9 @@
 #define percpu_modcopy(pcpudst, src, size)			\
 do {								\
 	unsigned int __i;					\
-	for (__i = 0; __i < NR_CPUS; __i++)			\
-		if (cpu_possible(__i))				\
-			memcpy((pcpudst)+__per_cpu_offset(__i),	\
-			       (src), (size));			\
+	for_each_cpu(__i)					\
+		memcpy((pcpudst)+__per_cpu_offset(__i),		\
+		       (src), (size));				\
 } while (0)
 
 extern void setup_per_cpu_areas(void);
diff --git a/include/asm-s390/atomic.h b/include/asm-s390/atomic.h
index be6fefe223d..de1d9926aa6 100644
--- a/include/asm-s390/atomic.h
+++ b/include/asm-s390/atomic.h
@@ -89,10 +89,15 @@ static __inline__ int atomic_cmpxchg(atomic_t *v, int old, int new)
 static __inline__ int atomic_add_unless(atomic_t *v, int a, int u)
 {
 	int c, old;
-
 	c = atomic_read(v);
-	while (c != u && (old = atomic_cmpxchg(v, c, c + a)) != c)
+	for (;;) {
+		if (unlikely(c == u))
+			break;
+		old = atomic_cmpxchg(v, c, c + a);
+		if (likely(old == c))
+			break;
 		c = old;
+	}
 	return c != u;
 }
 
@@ -167,10 +172,15 @@ static __inline__ int atomic64_add_unless(atomic64_t *v,
 					  long long a, long long u)
 {
 	long long c, old;
-
 	c = atomic64_read(v);
-	while (c != u && (old = atomic64_cmpxchg(v, c, c + a)) != c)
+	for (;;) {
+		if (unlikely(c == u))
+			break;
+		old = atomic64_cmpxchg(v, c, c + a);
+		if (likely(old == c))
+			break;
 		c = old;
+	}
 	return c != u;
 }
 
diff --git a/include/asm-s390/percpu.h b/include/asm-s390/percpu.h
index 123fcaca295..e10ed87094f 100644
--- a/include/asm-s390/percpu.h
+++ b/include/asm-s390/percpu.h
@@ -46,10 +46,9 @@ extern unsigned long __per_cpu_offset[NR_CPUS];
 #define percpu_modcopy(pcpudst, src, size)			\
 do {								\
 	unsigned int __i;					\
-	for (__i = 0; __i < NR_CPUS; __i++)			\
-		if (cpu_possible(__i))				\
-			memcpy((pcpudst)+__per_cpu_offset[__i],	\
-			       (src), (size));			\
+	for_each_cpu(__i)					\
+		memcpy((pcpudst)+__per_cpu_offset[__i],		\
+		       (src), (size));				\
 } while (0)
 
 #else /* ! SMP */
diff --git a/include/asm-sparc64/atomic.h b/include/asm-sparc64/atomic.h
index 25256bdc8aa..468eb48d814 100644
--- a/include/asm-sparc64/atomic.h
+++ b/include/asm-sparc64/atomic.h
@@ -78,9 +78,15 @@ extern int atomic64_sub_ret(int, atomic64_t *);
 ({								\
 	int c, old;						\
 	c = atomic_read(v);					\
-	while (c != (u) && (old = atomic_cmpxchg((v), c, c + (a))) != c) \
+	for (;;) {						\
+		if (unlikely(c == (u)))				\
+			break;					\
+		old = atomic_cmpxchg((v), c, c + (a));		\
+		if (likely(old == c))				\
+			break;					\
 		c = old;					\
-	c != (u);						\
+	}							\
+	likely(c != (u));					\
 })
 #define atomic_inc_not_zero(v) atomic_add_unless((v), 1, 0)
 
diff --git a/include/asm-sparc64/cache.h b/include/asm-sparc64/cache.h
index f7d35a2ae9b..e9df17acedd 100644
--- a/include/asm-sparc64/cache.h
+++ b/include/asm-sparc64/cache.h
@@ -13,4 +13,6 @@
 #define        SMP_CACHE_BYTES_SHIFT	6
 #define        SMP_CACHE_BYTES		(1 << SMP_CACHE_BYTES_SHIFT) /* L2 cache line size. */
 
+#define __read_mostly __attribute__((__section__(".data.read_mostly")))
+
 #endif
diff --git a/include/asm-sparc64/percpu.h b/include/asm-sparc64/percpu.h
index aea4e51e7cd..82032e159a7 100644
--- a/include/asm-sparc64/percpu.h
+++ b/include/asm-sparc64/percpu.h
@@ -26,10 +26,9 @@ register unsigned long __local_per_cpu_offset asm("g5");
 #define percpu_modcopy(pcpudst, src, size)			\
 do {								\
 	unsigned int __i;					\
-	for (__i = 0; __i < NR_CPUS; __i++)			\
-		if (cpu_possible(__i))				\
-			memcpy((pcpudst)+__per_cpu_offset(__i),	\
-			       (src), (size));			\
+	for_each_cpu(__i)					\
+		memcpy((pcpudst)+__per_cpu_offset(__i),		\
+		       (src), (size));				\
 } while (0)
 #else /* ! SMP */
 
diff --git a/include/asm-um/alternative.h b/include/asm-um/alternative.h
new file mode 100644
index 00000000000..b6434396bd4
--- /dev/null
+++ b/include/asm-um/alternative.h
@@ -0,0 +1,6 @@
+#ifndef __UM_ALTERNATIVE_H
+#define __UM_ALTERNATIVE_H
+
+#include "asm/arch/alternative.h"
+
+#endif
diff --git a/include/asm-x86_64/atomic.h b/include/asm-x86_64/atomic.h
index 4b5cd553e77..cecbf7baa6a 100644
--- a/include/asm-x86_64/atomic.h
+++ b/include/asm-x86_64/atomic.h
@@ -405,8 +405,14 @@ static __inline__ long atomic64_sub_return(long i, atomic64_t *v)
 ({								\
 	int c, old;						\
 	c = atomic_read(v);					\
-	while (c != (u) && (old = atomic_cmpxchg((v), c, c + (a))) != c) \
+	for (;;) {						\
+		if (unlikely(c == (u)))				\
+			break;					\
+		old = atomic_cmpxchg((v), c, c + (a));		\
+		if (likely(old == c))				\
+			break;					\
 		c = old;					\
+	}							\
 	c != (u);						\
 })
 #define atomic_inc_not_zero(v) atomic_add_unless((v), 1, 0)
diff --git a/include/asm-x86_64/cache.h b/include/asm-x86_64/cache.h
index 263f0a211ed..c8043a16152 100644
--- a/include/asm-x86_64/cache.h
+++ b/include/asm-x86_64/cache.h
@@ -20,6 +20,8 @@
        __attribute__((__section__(".data.page_aligned")))
 #endif
 
+#define __read_mostly __attribute__((__section__(".data.read_mostly")))
+
 #endif
 
 #endif
diff --git a/include/asm-x86_64/percpu.h b/include/asm-x86_64/percpu.h
index 29a6b0408f7..4405b4adeab 100644
--- a/include/asm-x86_64/percpu.h
+++ b/include/asm-x86_64/percpu.h
@@ -26,10 +26,9 @@
 #define percpu_modcopy(pcpudst, src, size)			\
 do {								\
 	unsigned int __i;					\
-	for (__i = 0; __i < NR_CPUS; __i++)			\
-		if (cpu_possible(__i))				\
-			memcpy((pcpudst)+__per_cpu_offset(__i),	\
-			       (src), (size));			\
+	for_each_cpu(__i)					\
+		memcpy((pcpudst)+__per_cpu_offset(__i),		\
+		       (src), (size));				\
 } while (0)
 
 extern void setup_per_cpu_areas(void);
diff --git a/include/linux/cache.h b/include/linux/cache.h
index d22e632f41f..cc4b3aafad9 100644
--- a/include/linux/cache.h
+++ b/include/linux/cache.h
@@ -13,9 +13,7 @@
 #define SMP_CACHE_BYTES L1_CACHE_BYTES
 #endif
 
-#if defined(CONFIG_X86) || defined(CONFIG_SPARC64) || defined(CONFIG_IA64) || defined(CONFIG_PARISC)
-#define __read_mostly __attribute__((__section__(".data.read_mostly")))
-#else
+#ifndef __read_mostly
 #define __read_mostly
 #endif
 
diff --git a/include/linux/cdrom.h b/include/linux/cdrom.h
index b68fdf1f315..3c9b0bc0512 100644
--- a/include/linux/cdrom.h
+++ b/include/linux/cdrom.h
@@ -378,7 +378,6 @@ struct cdrom_generic_command
 #define CDC_MEDIA_CHANGED 	0x80    /* media changed */
 #define CDC_PLAY_AUDIO		0x100   /* audio functions */
 #define CDC_RESET               0x200   /* hard reset device */
-#define CDC_IOCTLS              0x400   /* driver has non-standard ioctls */
 #define CDC_DRIVE_STATUS        0x800   /* driver implements drive status */
 #define CDC_GENERIC_PACKET	0x1000	/* driver implements generic packets */
 #define CDC_CD_R		0x2000	/* drive is a CD-R */
@@ -974,9 +973,7 @@ struct cdrom_device_ops {
 	int (*reset) (struct cdrom_device_info *);
 	/* play stuff */
 	int (*audio_ioctl) (struct cdrom_device_info *,unsigned int, void *);
-	/* dev-specific */
- 	int (*dev_ioctl) (struct cdrom_device_info *,
-			  unsigned int, unsigned long);
+
 /* driver specifications */
 	const int capability;   /* capability flags */
 	int n_minors;           /* number of active minor devices */
diff --git a/include/linux/eventpoll.h b/include/linux/eventpoll.h
index 1289f0ec4c0..1e4bdfcf83a 100644
--- a/include/linux/eventpoll.h
+++ b/include/linux/eventpoll.h
@@ -52,7 +52,12 @@ struct file;
 #ifdef CONFIG_EPOLL
 
 /* Used to initialize the epoll bits inside the "struct file" */
-void eventpoll_init_file(struct file *file);
+static inline void eventpoll_init_file(struct file *file)
+{
+	INIT_LIST_HEAD(&file->f_ep_links);
+	spin_lock_init(&file->f_ep_lock);
+}
+
 
 /* Used to release the epoll bits inside the "struct file" */
 void eventpoll_release_file(struct file *file);
@@ -85,7 +90,6 @@ static inline void eventpoll_release(struct file *file)
 	eventpoll_release_file(file);
 }
 
-
 #else
 
 static inline void eventpoll_init_file(struct file *file) {}
diff --git a/include/linux/ext3_fs.h b/include/linux/ext3_fs.h
index c0272d73ab2..e7239f2f97a 100644
--- a/include/linux/ext3_fs.h
+++ b/include/linux/ext3_fs.h
@@ -772,9 +772,12 @@ extern unsigned long ext3_count_free (struct buffer_head *, unsigned);
 
 
 /* inode.c */
-extern int ext3_forget(handle_t *, int, struct inode *, struct buffer_head *, int);
-extern struct buffer_head * ext3_getblk (handle_t *, struct inode *, long, int, int *);
-extern struct buffer_head * ext3_bread (handle_t *, struct inode *, int, int, int *);
+int ext3_forget(handle_t *, int, struct inode *, struct buffer_head *, int);
+struct buffer_head * ext3_getblk (handle_t *, struct inode *, long, int, int *);
+struct buffer_head * ext3_bread (handle_t *, struct inode *, int, int, int *);
+int ext3_get_block_handle(handle_t *handle, struct inode *inode,
+	sector_t iblock, struct buffer_head *bh_result, int create,
+	int extend_disksize);
 
 extern void ext3_read_inode (struct inode *);
 extern int  ext3_write_inode (struct inode *, int);
diff --git a/include/linux/ext3_fs_i.h b/include/linux/ext3_fs_i.h
index e71dd98dbca..7abf9014718 100644
--- a/include/linux/ext3_fs_i.h
+++ b/include/linux/ext3_fs_i.h
@@ -19,6 +19,7 @@
 #include <linux/rwsem.h>
 #include <linux/rbtree.h>
 #include <linux/seqlock.h>
+#include <linux/mutex.h>
 
 struct ext3_reserve_window {
 	__u32			_rsv_start;	/* First byte reserved */
@@ -122,16 +123,16 @@ struct ext3_inode_info {
 	__u16 i_extra_isize;
 
 	/*
-	 * truncate_sem is for serialising ext3_truncate() against
+	 * truncate_mutex is for serialising ext3_truncate() against
 	 * ext3_getblock().  In the 2.4 ext2 design, great chunks of inode's
 	 * data tree are chopped off during truncate. We can't do that in
 	 * ext3 because whenever we perform intermediate commits during
 	 * truncate, the inode and all the metadata blocks *must* be in a
 	 * consistent state which allows truncation of the orphans to restart
 	 * during recovery.  Hence we must fix the get_block-vs-truncate race
-	 * by other means, so we have truncate_sem.
+	 * by other means, so we have truncate_mutex.
 	 */
-	struct semaphore truncate_sem;
+	struct mutex truncate_mutex;
 	struct inode vfs_inode;
 };
 
diff --git a/include/linux/file.h b/include/linux/file.h
index 9901b850f2e..9f7c2513866 100644
--- a/include/linux/file.h
+++ b/include/linux/file.h
@@ -10,6 +10,7 @@
 #include <linux/compiler.h>
 #include <linux/spinlock.h>
 #include <linux/rcupdate.h>
+#include <linux/types.h>
 
 /*
  * The default fd array needs to be at least BITS_PER_LONG,
@@ -17,10 +18,22 @@
  */
 #define NR_OPEN_DEFAULT BITS_PER_LONG
 
+/*
+ * The embedded_fd_set is a small fd_set,
+ * suitable for most tasks (which open <= BITS_PER_LONG files)
+ */
+struct embedded_fd_set {
+	unsigned long fds_bits[1];
+};
+
+/*
+ * More than this number of fds: we use a separately allocated fd_set
+ */
+#define EMBEDDED_FD_SET_SIZE (BITS_PER_BYTE * sizeof(struct embedded_fd_set))
+
 struct fdtable {
 	unsigned int max_fds;
 	int max_fdset;
-	int next_fd;
 	struct file ** fd;      /* current fd array */
 	fd_set *close_on_exec;
 	fd_set *open_fds;
@@ -33,13 +46,20 @@ struct fdtable {
  * Open file table structure
  */
 struct files_struct {
+  /*
+   * read mostly part
+   */
 	atomic_t count;
 	struct fdtable *fdt;
 	struct fdtable fdtab;
-	fd_set close_on_exec_init;
-	fd_set open_fds_init;
+  /*
+   * written part on a separate cache line in SMP
+   */
+	spinlock_t file_lock ____cacheline_aligned_in_smp;
+	int next_fd;
+	struct embedded_fd_set close_on_exec_init;
+	struct embedded_fd_set open_fds_init;
 	struct file * fd_array[NR_OPEN_DEFAULT];
-	spinlock_t file_lock;     /* Protects concurrent writers.  Nests inside tsk->alloc_lock */
 };
 
 #define files_fdtable(files) (rcu_dereference((files)->fdt))
diff --git a/include/linux/fs.h b/include/linux/fs.h
index 128d0082522..f9c9dea636d 100644
--- a/include/linux/fs.h
+++ b/include/linux/fs.h
@@ -397,8 +397,8 @@ struct block_device {
 	dev_t			bd_dev;  /* not a kdev_t - it's a search key */
 	struct inode *		bd_inode;	/* will die */
 	int			bd_openers;
-	struct semaphore	bd_sem;	/* open/close mutex */
-	struct semaphore	bd_mount_sem;	/* mount mutex */
+	struct mutex		bd_mutex;	/* open/close mutex */
+	struct mutex		bd_mount_mutex;	/* mount mutex */
 	struct list_head	bd_inodes;
 	void *			bd_holder;
 	int			bd_holders;
@@ -509,7 +509,7 @@ struct inode {
 
 #ifdef CONFIG_INOTIFY
 	struct list_head	inotify_watches; /* watches on this inode */
-	struct semaphore	inotify_sem;	/* protects the watches list */
+	struct mutex		inotify_mutex;	/* protects the watches list */
 #endif
 
 	unsigned long		i_state;
@@ -847,7 +847,7 @@ struct super_block {
 	 * The next field is for VFS *only*. No filesystems have any business
 	 * even looking at it. You had been warned.
 	 */
-	struct semaphore s_vfs_rename_sem;	/* Kludge */
+	struct mutex s_vfs_rename_mutex;	/* Kludge */
 
 	/* Granuality of c/m/atime in ns.
 	   Cannot be worse than a second */
@@ -1115,6 +1115,18 @@ static inline void mark_inode_dirty_sync(struct inode *inode)
 	__mark_inode_dirty(inode, I_DIRTY_SYNC);
 }
 
+static inline void inode_inc_link_count(struct inode *inode)
+{
+	inode->i_nlink++;
+	mark_inode_dirty(inode);
+}
+
+static inline void inode_dec_link_count(struct inode *inode)
+{
+	inode->i_nlink--;
+	mark_inode_dirty(inode);
+}
+
 extern void touch_atime(struct vfsmount *mnt, struct dentry *dentry);
 static inline void file_accessed(struct file *file)
 {
@@ -1534,7 +1546,7 @@ extern void destroy_inode(struct inode *);
 extern struct inode *new_inode(struct super_block *);
 extern int remove_suid(struct dentry *);
 extern void remove_dquot_ref(struct super_block *, int, struct list_head *);
-extern struct semaphore iprune_sem;
+extern struct mutex iprune_mutex;
 
 extern void __insert_inode_hash(struct inode *, unsigned long hashval);
 extern void remove_inode_hash(struct inode *);
diff --git a/include/linux/generic_serial.h b/include/linux/generic_serial.h
index 0abe9d9a006..652611a4bdc 100644
--- a/include/linux/generic_serial.h
+++ b/include/linux/generic_serial.h
@@ -12,6 +12,8 @@
 #ifndef GENERIC_SERIAL_H
 #define GENERIC_SERIAL_H
 
+#include <linux/mutex.h>
+
 struct real_driver {
   void                    (*disable_tx_interrupts) (void *);
   void                    (*enable_tx_interrupts) (void *);
@@ -34,7 +36,7 @@ struct gs_port {
   int                     xmit_head;
   int                     xmit_tail;
   int                     xmit_cnt;
-  struct semaphore        port_write_sem;
+  struct mutex            port_write_mutex;
   int                     flags;
   wait_queue_head_t       open_wait;
   wait_queue_head_t       close_wait;
diff --git a/include/linux/genhd.h b/include/linux/genhd.h
index eef5ccdcd73..fd647fde5ec 100644
--- a/include/linux/genhd.h
+++ b/include/linux/genhd.h
@@ -149,22 +149,16 @@ struct disk_attribute {
 ({									\
 	typeof(gendiskp->dkstats->field) res = 0;			\
 	int i;								\
-	for (i=0; i < NR_CPUS; i++) {					\
-		if (!cpu_possible(i))					\
-			continue;					\
+	for_each_cpu(i)							\
 		res += per_cpu_ptr(gendiskp->dkstats, i)->field;	\
-	}								\
 	res;								\
 })
 
 static inline void disk_stat_set_all(struct gendisk *gendiskp, int value)	{
 	int i;
-	for (i=0; i < NR_CPUS; i++) {
-		if (cpu_possible(i)) {
-			memset(per_cpu_ptr(gendiskp->dkstats, i), value,	
-					sizeof (struct disk_stats));
-		}
-	}
+	for_each_cpu(i)
+		memset(per_cpu_ptr(gendiskp->dkstats, i), value,
+				sizeof (struct disk_stats));
 }		
 				
 #else
diff --git a/include/linux/init_task.h b/include/linux/init_task.h
index dcfd2ecccb5..92146f3b742 100644
--- a/include/linux/init_task.h
+++ b/include/linux/init_task.h
@@ -7,11 +7,10 @@
 #define INIT_FDTABLE \
 {							\
 	.max_fds	= NR_OPEN_DEFAULT, 		\
-	.max_fdset	= __FD_SETSIZE, 		\
-	.next_fd	= 0, 				\
+	.max_fdset	= EMBEDDED_FD_SET_SIZE,		\
 	.fd		= &init_files.fd_array[0], 	\
-	.close_on_exec	= &init_files.close_on_exec_init, \
-	.open_fds	= &init_files.open_fds_init, 	\
+	.close_on_exec	= (fd_set *)&init_files.close_on_exec_init, \
+	.open_fds	= (fd_set *)&init_files.open_fds_init, 	\
 	.rcu		= RCU_HEAD_INIT, 		\
 	.free_files	= NULL,		 		\
 	.next		= NULL,		 		\
@@ -20,9 +19,10 @@
 #define INIT_FILES \
 { 							\
 	.count		= ATOMIC_INIT(1), 		\
-	.file_lock	= SPIN_LOCK_UNLOCKED, 		\
 	.fdt		= &init_files.fdtab, 		\
 	.fdtab		= INIT_FDTABLE,			\
+	.file_lock	= SPIN_LOCK_UNLOCKED, 		\
+	.next_fd	= 0, 				\
 	.close_on_exec_init = { { 0, } }, 		\
 	.open_fds_init	= { { 0, } }, 			\
 	.fd_array	= { NULL, } 			\
diff --git a/include/linux/jbd.h b/include/linux/jbd.h
index 41ee79962bb..2ccbfb6340b 100644
--- a/include/linux/jbd.h
+++ b/include/linux/jbd.h
@@ -28,6 +28,7 @@
 #include <linux/journal-head.h>
 #include <linux/stddef.h>
 #include <linux/bit_spinlock.h>
+#include <linux/mutex.h>
 #include <asm/semaphore.h>
 #endif
 
@@ -575,7 +576,7 @@ struct transaction_s
  * @j_wait_checkpoint:  Wait queue to trigger checkpointing
  * @j_wait_commit: Wait queue to trigger commit
  * @j_wait_updates: Wait queue to wait for updates to complete
- * @j_checkpoint_sem: Semaphore for locking against concurrent checkpoints
+ * @j_checkpoint_mutex: Mutex for locking against concurrent checkpoints
  * @j_head: Journal head - identifies the first unused block in the journal
  * @j_tail: Journal tail - identifies the oldest still-used block in the
  *  journal.
@@ -645,7 +646,7 @@ struct journal_s
 	int			j_barrier_count;
 
 	/* The barrier lock itself */
-	struct semaphore	j_barrier;
+	struct mutex		j_barrier;
 
 	/*
 	 * Transactions: The current running transaction...
@@ -687,7 +688,7 @@ struct journal_s
 	wait_queue_head_t	j_wait_updates;
 
 	/* Semaphore for locking against concurrent checkpoints */
-	struct semaphore 	j_checkpoint_sem;
+	struct mutex	 	j_checkpoint_mutex;
 
 	/*
 	 * Journal head: identifies the first unused block in the journal.
diff --git a/include/linux/kernel.h b/include/linux/kernel.h
index 3b507bf05d0..bb6e7ddee2f 100644
--- a/include/linux/kernel.h
+++ b/include/linux/kernel.h
@@ -91,6 +91,9 @@ extern struct notifier_block *panic_notifier_list;
 extern long (*panic_blink)(long time);
 NORET_TYPE void panic(const char * fmt, ...)
 	__attribute__ ((NORET_AND format (printf, 1, 2)));
+extern void oops_enter(void);
+extern void oops_exit(void);
+extern int oops_may_print(void);
 fastcall NORET_TYPE void do_exit(long error_code)
 	ATTRIB_NORET;
 NORET_TYPE void complete_and_exit(struct completion *, long)
diff --git a/include/linux/kprobes.h b/include/linux/kprobes.h
index 669756bc20a..778adc0fa64 100644
--- a/include/linux/kprobes.h
+++ b/include/linux/kprobes.h
@@ -36,6 +36,7 @@
 #include <linux/percpu.h>
 #include <linux/spinlock.h>
 #include <linux/rcupdate.h>
+#include <linux/mutex.h>
 
 #ifdef CONFIG_KPROBES
 #include <asm/kprobes.h>
@@ -152,7 +153,7 @@ struct kretprobe_instance {
 };
 
 extern spinlock_t kretprobe_lock;
-extern struct semaphore kprobe_mutex;
+extern struct mutex kprobe_mutex;
 extern int arch_prepare_kprobe(struct kprobe *p);
 extern void arch_arm_kprobe(struct kprobe *p);
 extern void arch_disarm_kprobe(struct kprobe *p);
diff --git a/include/linux/loop.h b/include/linux/loop.h
index f96506782eb..e76c7611d6c 100644
--- a/include/linux/loop.h
+++ b/include/linux/loop.h
@@ -17,6 +17,7 @@
 #include <linux/bio.h>
 #include <linux/blkdev.h>
 #include <linux/spinlock.h>
+#include <linux/mutex.h>
 
 /* Possible states of device */
 enum {
@@ -60,7 +61,7 @@ struct loop_device {
 	int			lo_state;
 	struct completion	lo_done;
 	struct completion	lo_bh_done;
-	struct semaphore	lo_ctl_mutex;
+	struct mutex		lo_ctl_mutex;
 	int			lo_pending;
 
 	request_queue_t		*lo_queue;
diff --git a/include/linux/msdos_fs.h b/include/linux/msdos_fs.h
index 8bcd9450d92..779e6a5744c 100644
--- a/include/linux/msdos_fs.h
+++ b/include/linux/msdos_fs.h
@@ -184,6 +184,7 @@ struct fat_slot_info {
 #include <linux/string.h>
 #include <linux/nls.h>
 #include <linux/fs.h>
+#include <linux/mutex.h>
 
 struct fat_mount_options {
 	uid_t fs_uid;
@@ -226,7 +227,7 @@ struct msdos_sb_info {
 	unsigned long max_cluster;   /* maximum cluster number */
 	unsigned long root_cluster;  /* first cluster of the root directory */
 	unsigned long fsinfo_sector; /* sector number of FAT32 fsinfo */
-	struct semaphore fat_lock;
+	struct mutex fat_lock;
 	unsigned int prev_free;      /* previously allocated cluster number */
 	unsigned int free_clusters;  /* -1 if undefined */
 	struct fat_mount_options options;
diff --git a/include/linux/nbd.h b/include/linux/nbd.h
index f95d51fae73..a6ce409ec6f 100644
--- a/include/linux/nbd.h
+++ b/include/linux/nbd.h
@@ -38,6 +38,7 @@ enum {
 #ifdef __KERNEL__
 
 #include <linux/wait.h>
+#include <linux/mutex.h>
 
 /* values for flags field */
 #define NBD_READ_ONLY 0x0001
@@ -57,7 +58,7 @@ struct nbd_device {
 	struct request *active_req;
 	wait_queue_head_t active_wq;
 
-	struct semaphore tx_lock;
+	struct mutex tx_lock;
 	struct gendisk *disk;
 	int blksize;
 	u64 bytesize;
diff --git a/include/linux/ncp_fs_i.h b/include/linux/ncp_fs_i.h
index 415be1ec6f9..bdb4c8ae692 100644
--- a/include/linux/ncp_fs_i.h
+++ b/include/linux/ncp_fs_i.h
@@ -19,7 +19,7 @@ struct ncp_inode_info {
 	__le32	DosDirNum;
 	__u8	volNumber;
 	__le32	nwattr;
-	struct semaphore open_sem;
+	struct mutex open_mutex;
 	atomic_t	opened;
 	int	access;
 	int	flags;
diff --git a/include/linux/ncp_fs_sb.h b/include/linux/ncp_fs_sb.h
index cf858eb80f0..b089d950628 100644
--- a/include/linux/ncp_fs_sb.h
+++ b/include/linux/ncp_fs_sb.h
@@ -11,6 +11,7 @@
 #include <linux/types.h>
 #include <linux/ncp_mount.h>
 #include <linux/net.h>
+#include <linux/mutex.h>
 
 #ifdef __KERNEL__
 
@@ -51,7 +52,7 @@ struct ncp_server {
 				   receive replies */
 
 	int lock;		/* To prevent mismatch in protocols. */
-	struct semaphore sem;
+	struct mutex mutex;
 
 	int current_size;	/* for packet preparation */
 	int has_subfunction;
@@ -96,7 +97,7 @@ struct ncp_server {
 	struct {
 		struct work_struct tq;		/* STREAM/DGRAM: data/error ready */
 		struct ncp_request_reply* creq;	/* STREAM/DGRAM: awaiting reply from this request */
-		struct semaphore creq_sem;	/* DGRAM only: lock accesses to rcv.creq */
+		struct mutex creq_mutex;	/* DGRAM only: lock accesses to rcv.creq */
 
 		unsigned int state;		/* STREAM only: receiver state */
 		struct {
diff --git a/include/linux/pm.h b/include/linux/pm.h
index 5be87ba3b7a..6df2585c016 100644
--- a/include/linux/pm.h
+++ b/include/linux/pm.h
@@ -188,6 +188,8 @@ extern void device_power_up(void);
 extern void device_resume(void);
 
 #ifdef CONFIG_PM
+extern suspend_disk_method_t pm_disk_mode;
+
 extern int device_suspend(pm_message_t state);
 
 #define device_set_wakeup_enable(dev,val) \
@@ -215,7 +217,6 @@ static inline int dpm_runtime_suspend(struct device * dev, pm_message_t state)
 
 static inline void dpm_runtime_resume(struct device * dev)
 {
-
 }
 
 #endif
diff --git a/include/linux/profile.h b/include/linux/profile.h
index 026969a5595..1f2fea6640a 100644
--- a/include/linux/profile.h
+++ b/include/linux/profile.h
@@ -14,6 +14,7 @@
 
 struct proc_dir_entry;
 struct pt_regs;
+struct notifier_block;
 
 /* init basic kernel profiler */
 void __init profile_init(void);
@@ -32,7 +33,6 @@ enum profile_type {
 
 #ifdef CONFIG_PROFILING
 
-struct notifier_block;
 struct task_struct;
 struct mm_struct;
 
diff --git a/include/linux/quota.h b/include/linux/quota.h
index f33aeb22c26..8dc2d04a103 100644
--- a/include/linux/quota.h
+++ b/include/linux/quota.h
@@ -38,6 +38,7 @@
 #include <linux/errno.h>
 #include <linux/types.h>
 #include <linux/spinlock.h>
+#include <linux/mutex.h>
 
 #define __DQUOT_VERSION__	"dquot_6.5.1"
 #define __DQUOT_NUM_VERSION__	6*10000+5*100+1
@@ -215,7 +216,7 @@ struct dquot {
 	struct list_head dq_inuse;	/* List of all quotas */
 	struct list_head dq_free;	/* Free list element */
 	struct list_head dq_dirty;	/* List of dirty dquots */
-	struct semaphore dq_lock;	/* dquot IO lock */
+	struct mutex dq_lock;		/* dquot IO lock */
 	atomic_t dq_count;		/* Use count */
 	wait_queue_head_t dq_wait_unused;	/* Wait queue for dquot to become unused */
 	struct super_block *dq_sb;	/* superblock this applies to */
@@ -285,8 +286,8 @@ struct quota_format_type {
 
 struct quota_info {
 	unsigned int flags;			/* Flags for diskquotas on this device */
-	struct semaphore dqio_sem;		/* lock device while I/O in progress */
-	struct semaphore dqonoff_sem;		/* Serialize quotaon & quotaoff */
+	struct mutex dqio_mutex;		/* lock device while I/O in progress */
+	struct mutex dqonoff_mutex;		/* Serialize quotaon & quotaoff */
 	struct rw_semaphore dqptr_sem;		/* serialize ops using quota_info struct, pointers from inode to dquots */
 	struct inode *files[MAXQUOTAS];		/* inodes of quotafiles */
 	struct mem_dqinfo info[MAXQUOTAS];	/* Information for each quota type */
diff --git a/include/linux/raid/raid1.h b/include/linux/raid/raid1.h
index 9d5494aaac0..3009c813d83 100644
--- a/include/linux/raid/raid1.h
+++ b/include/linux/raid/raid1.h
@@ -130,6 +130,6 @@ struct r1bio_s {
  * with failure when last write completes (and all failed).
  * Record that bi_end_io was called with this flag...
  */
-#define	R1BIO_Returned 4
+#define	R1BIO_Returned 6
 
 #endif
diff --git a/include/linux/rcupdate.h b/include/linux/rcupdate.h
index c2ec6c77874..5673008b61e 100644
--- a/include/linux/rcupdate.h
+++ b/include/linux/rcupdate.h
@@ -113,8 +113,6 @@ struct rcu_data {
 
 DECLARE_PER_CPU(struct rcu_data, rcu_data);
 DECLARE_PER_CPU(struct rcu_data, rcu_bh_data);
-extern struct rcu_ctrlblk rcu_ctrlblk;
-extern struct rcu_ctrlblk rcu_bh_ctrlblk;
 
 /*
  * Increment the quiescent state counter.
diff --git a/include/linux/seq_file.h b/include/linux/seq_file.h
index 850a974ee50..b95f6eb7254 100644
--- a/include/linux/seq_file.h
+++ b/include/linux/seq_file.h
@@ -4,7 +4,7 @@
 
 #include <linux/types.h>
 #include <linux/string.h>
-#include <asm/semaphore.h>
+#include <linux/mutex.h>
 
 struct seq_operations;
 struct file;
@@ -19,7 +19,7 @@ struct seq_file {
 	size_t count;
 	loff_t index;
 	loff_t version;
-	struct semaphore sem;
+	struct mutex lock;
 	struct seq_operations *op;
 	void *private;
 };
diff --git a/include/linux/swap.h b/include/linux/swap.h
index 12415dd9445..54eac8a39a4 100644
--- a/include/linux/swap.h
+++ b/include/linux/swap.h
@@ -234,14 +234,15 @@ extern struct page * read_swap_cache_async(swp_entry_t, struct vm_area_struct *v
 /* linux/mm/swapfile.c */
 extern long total_swap_pages;
 extern unsigned int nr_swapfiles;
-extern struct swap_info_struct swap_info[];
 extern void si_swapinfo(struct sysinfo *);
 extern swp_entry_t get_swap_page(void);
-extern swp_entry_t get_swap_page_of_type(int type);
+extern swp_entry_t get_swap_page_of_type(int);
 extern int swap_duplicate(swp_entry_t);
 extern int valid_swaphandles(swp_entry_t, unsigned long *);
 extern void swap_free(swp_entry_t);
 extern void free_swap_and_cache(swp_entry_t);
+extern int swap_type_of(dev_t);
+extern unsigned int count_swap_pages(int, int);
 extern sector_t map_swap_page(struct swap_info_struct *, pgoff_t);
 extern struct swap_info_struct *get_swap_info_struct(unsigned);
 extern int can_share_swap_page(struct page *);
diff --git a/include/linux/tty.h b/include/linux/tty.h
index f45cd74e6f2..f13f49afe19 100644
--- a/include/linux/tty.h
+++ b/include/linux/tty.h
@@ -24,6 +24,7 @@
 #include <linux/tty_driver.h>
 #include <linux/tty_ldisc.h>
 #include <linux/screen_info.h>
+#include <linux/mutex.h>
 
 #include <asm/system.h>
 
@@ -231,8 +232,8 @@ struct tty_struct {
 	int canon_data;
 	unsigned long canon_head;
 	unsigned int canon_column;
-	struct semaphore atomic_read;
-	struct semaphore atomic_write;
+	struct mutex atomic_read_lock;
+	struct mutex atomic_write_lock;
 	unsigned char *write_buf;
 	int write_cnt;
 	spinlock_t read_lock;
@@ -319,8 +320,7 @@ extern void tty_ldisc_put(int);
 extern void tty_wakeup(struct tty_struct *tty);
 extern void tty_ldisc_flush(struct tty_struct *tty);
 
-struct semaphore;
-extern struct semaphore tty_sem;
+extern struct mutex tty_mutex;
 
 /* n_tty.c */
 extern struct tty_ldisc tty_ldisc_N_TTY;
diff --git a/include/linux/tty_flip.h b/include/linux/tty_flip.h
index 222faf97d5f..0c6169fff36 100644
--- a/include/linux/tty_flip.h
+++ b/include/linux/tty_flip.h
@@ -7,14 +7,8 @@ extern int tty_insert_flip_string_flags(struct tty_struct *tty, unsigned char *c
 extern int tty_prepare_flip_string(struct tty_struct *tty, unsigned char **chars, size_t size);
 extern int tty_prepare_flip_string_flags(struct tty_struct *tty, unsigned char **chars, char **flags, size_t size);
 
-#ifdef INCLUDE_INLINE_FUNCS
-#define _INLINE_ extern
-#else
-#define _INLINE_ static __inline__
-#endif
-
-_INLINE_ int tty_insert_flip_char(struct tty_struct *tty,
-				   unsigned char ch, char flag)
+static inline int tty_insert_flip_char(struct tty_struct *tty,
+				       unsigned char ch, char flag)
 {
 	struct tty_buffer *tb = tty->buf.tail;
 	if (tb && tb->active && tb->used < tb->size) {
@@ -25,7 +19,7 @@ _INLINE_ int tty_insert_flip_char(struct tty_struct *tty,
 	return tty_insert_flip_string_flags(tty, &ch, &flag, 1);
 }
 
-_INLINE_ void tty_schedule_flip(struct tty_struct *tty)
+static inline void tty_schedule_flip(struct tty_struct *tty)
 {
 	unsigned long flags;
 	spin_lock_irqsave(&tty->buf.lock, flags);
diff --git a/include/linux/udf_fs_sb.h b/include/linux/udf_fs_sb.h
index b15ff2e99c9..80ae9ef940d 100644
--- a/include/linux/udf_fs_sb.h
+++ b/include/linux/udf_fs_sb.h
@@ -13,7 +13,7 @@
 #ifndef _UDF_FS_SB_H
 #define _UDF_FS_SB_H 1
 
-#include <asm/semaphore.h>
+#include <linux/mutex.h>
 
 #pragma pack(1)
 
@@ -111,7 +111,7 @@ struct udf_sb_info
 	/* VAT inode */
 	struct inode		*s_vat;
 
-	struct semaphore	s_alloc_sem;
+	struct mutex		s_alloc_mutex;
 };
 
 #endif /* _UDF_FS_SB_H */
diff --git a/include/linux/vt_kern.h b/include/linux/vt_kern.h
index fab5aed8ca3..530ae3f4248 100644
--- a/include/linux/vt_kern.h
+++ b/include/linux/vt_kern.h
@@ -73,6 +73,11 @@ int con_copy_unimap(struct vc_data *dst_vc, struct vc_data *src_vc);
 int vt_waitactive(int vt);
 void change_console(struct vc_data *new_vc);
 void reset_vc(struct vc_data *vc);
+#ifdef CONFIG_VT
+int is_console_suspend_safe(void);
+#else
+static inline int is_console_suspend_safe(void) { return 1; }
+#endif
 
 /*
  * vc_screen.c shares this temporary buffer with the console write code so that
diff --git a/init/do_mounts_initrd.c b/init/do_mounts_initrd.c
index a05cabd0fd1..405f9031af8 100644
--- a/init/do_mounts_initrd.c
+++ b/init/do_mounts_initrd.c
@@ -56,6 +56,7 @@ static void __init handle_initrd(void)
 	sys_chroot(".");
 	mount_devfs_fs ();
 
+	current->flags |= PF_NOFREEZE;
 	pid = kernel_thread(do_linuxrc, "/linuxrc", SIGCHLD);
 	if (pid > 0) {
 		while (pid != sys_wait4(-1, NULL, 0, NULL))
diff --git a/init/main.c b/init/main.c
index 4c194c47395..2714e0e7cfe 100644
--- a/init/main.c
+++ b/init/main.c
@@ -325,7 +325,7 @@ static inline void smp_prepare_cpus(unsigned int maxcpus) { }
 #else
 
 #ifdef __GENERIC_PER_CPU
-unsigned long __per_cpu_offset[NR_CPUS];
+unsigned long __per_cpu_offset[NR_CPUS] __read_mostly;
 
 EXPORT_SYMBOL(__per_cpu_offset);
 
@@ -333,6 +333,7 @@ static void __init setup_per_cpu_areas(void)
 {
 	unsigned long size, i;
 	char *ptr;
+	unsigned long nr_possible_cpus = num_possible_cpus();
 
 	/* Copy section for each CPU (we discard the original) */
 	size = ALIGN(__per_cpu_end - __per_cpu_start, SMP_CACHE_BYTES);
@@ -340,12 +341,12 @@ static void __init setup_per_cpu_areas(void)
 	if (size < PERCPU_ENOUGH_ROOM)
 		size = PERCPU_ENOUGH_ROOM;
 #endif
+	ptr = alloc_bootmem(size * nr_possible_cpus);
 
-	ptr = alloc_bootmem(size * NR_CPUS);
-
-	for (i = 0; i < NR_CPUS; i++, ptr += size) {
+	for_each_cpu(i) {
 		__per_cpu_offset[i] = ptr - __per_cpu_start;
 		memcpy(ptr, __per_cpu_start, __per_cpu_end - __per_cpu_start);
+		ptr += size;
 	}
 }
 #endif /* !__GENERIC_PER_CPU */
@@ -438,6 +439,15 @@ void __init parse_early_param(void)
  *	Activate the first processor.
  */
 
+static void __init boot_cpu_init(void)
+{
+	int cpu = smp_processor_id();
+	/* Mark the boot cpu "present", "online" etc for SMP and UP case */
+	cpu_set(cpu, cpu_online_map);
+	cpu_set(cpu, cpu_present_map);
+	cpu_set(cpu, cpu_possible_map);
+}
+
 asmlinkage void __init start_kernel(void)
 {
 	char * command_line;
@@ -447,17 +457,13 @@ asmlinkage void __init start_kernel(void)
  * enable them
  */
 	lock_kernel();
+	boot_cpu_init();
 	page_address_init();
 	printk(KERN_NOTICE);
 	printk(linux_banner);
 	setup_arch(&command_line);
 	setup_per_cpu_areas();
-
-	/*
-	 * Mark the boot cpu "online" so that it can call console drivers in
-	 * printk() and can access its per-cpu storage.
-	 */
-	smp_prepare_boot_cpu();
+	smp_prepare_boot_cpu();	/* arch-specific boot-cpu hooks */
 
 	/*
 	 * Set up the scheduler prior starting any interrupts (such as the
diff --git a/kernel/cpuset.c b/kernel/cpuset.c
index 12815d3f1a0..c86ee051b73 100644
--- a/kernel/cpuset.c
+++ b/kernel/cpuset.c
@@ -53,7 +53,7 @@
 
 #include <asm/uaccess.h>
 #include <asm/atomic.h>
-#include <asm/semaphore.h>
+#include <linux/mutex.h>
 
 #define CPUSET_SUPER_MAGIC		0x27e0eb
 
@@ -168,63 +168,57 @@ static struct vfsmount *cpuset_mount;
 static struct super_block *cpuset_sb;
 
 /*
- * We have two global cpuset semaphores below.  They can nest.
- * It is ok to first take manage_sem, then nest callback_sem.  We also
+ * We have two global cpuset mutexes below.  They can nest.
+ * It is ok to first take manage_mutex, then nest callback_mutex.  We also
  * require taking task_lock() when dereferencing a tasks cpuset pointer.
  * See "The task_lock() exception", at the end of this comment.
  *
- * A task must hold both semaphores to modify cpusets.  If a task
- * holds manage_sem, then it blocks others wanting that semaphore,
- * ensuring that it is the only task able to also acquire callback_sem
+ * A task must hold both mutexes to modify cpusets.  If a task
+ * holds manage_mutex, then it blocks others wanting that mutex,
+ * ensuring that it is the only task able to also acquire callback_mutex
  * and be able to modify cpusets.  It can perform various checks on
  * the cpuset structure first, knowing nothing will change.  It can
- * also allocate memory while just holding manage_sem.  While it is
+ * also allocate memory while just holding manage_mutex.  While it is
  * performing these checks, various callback routines can briefly
- * acquire callback_sem to query cpusets.  Once it is ready to make
- * the changes, it takes callback_sem, blocking everyone else.
+ * acquire callback_mutex to query cpusets.  Once it is ready to make
+ * the changes, it takes callback_mutex, blocking everyone else.
  *
  * Calls to the kernel memory allocator can not be made while holding
- * callback_sem, as that would risk double tripping on callback_sem
+ * callback_mutex, as that would risk double tripping on callback_mutex
  * from one of the callbacks into the cpuset code from within
  * __alloc_pages().
  *
- * If a task is only holding callback_sem, then it has read-only
+ * If a task is only holding callback_mutex, then it has read-only
  * access to cpusets.
  *
  * The task_struct fields mems_allowed and mems_generation may only
  * be accessed in the context of that task, so require no locks.
  *
  * Any task can increment and decrement the count field without lock.
- * So in general, code holding manage_sem or callback_sem can't rely
+ * So in general, code holding manage_mutex or callback_mutex can't rely
  * on the count field not changing.  However, if the count goes to
- * zero, then only attach_task(), which holds both semaphores, can
+ * zero, then only attach_task(), which holds both mutexes, can
  * increment it again.  Because a count of zero means that no tasks
  * are currently attached, therefore there is no way a task attached
  * to that cpuset can fork (the other way to increment the count).
- * So code holding manage_sem or callback_sem can safely assume that
+ * So code holding manage_mutex or callback_mutex can safely assume that
  * if the count is zero, it will stay zero.  Similarly, if a task
- * holds manage_sem or callback_sem on a cpuset with zero count, it
+ * holds manage_mutex or callback_mutex on a cpuset with zero count, it
  * knows that the cpuset won't be removed, as cpuset_rmdir() needs
- * both of those semaphores.
- *
- * A possible optimization to improve parallelism would be to make
- * callback_sem a R/W semaphore (rwsem), allowing the callback routines
- * to proceed in parallel, with read access, until the holder of
- * manage_sem needed to take this rwsem for exclusive write access
- * and modify some cpusets.
+ * both of those mutexes.
  *
  * The cpuset_common_file_write handler for operations that modify
- * the cpuset hierarchy holds manage_sem across the entire operation,
+ * the cpuset hierarchy holds manage_mutex across the entire operation,
  * single threading all such cpuset modifications across the system.
  *
- * The cpuset_common_file_read() handlers only hold callback_sem across
+ * The cpuset_common_file_read() handlers only hold callback_mutex across
  * small pieces of code, such as when reading out possibly multi-word
  * cpumasks and nodemasks.
  *
  * The fork and exit callbacks cpuset_fork() and cpuset_exit(), don't
- * (usually) take either semaphore.  These are the two most performance
+ * (usually) take either mutex.  These are the two most performance
  * critical pieces of code here.  The exception occurs on cpuset_exit(),
- * when a task in a notify_on_release cpuset exits.  Then manage_sem
+ * when a task in a notify_on_release cpuset exits.  Then manage_mutex
  * is taken, and if the cpuset count is zero, a usermode call made
  * to /sbin/cpuset_release_agent with the name of the cpuset (path
  * relative to the root of cpuset file system) as the argument.
@@ -242,9 +236,9 @@ static struct super_block *cpuset_sb;
  *
  * The need for this exception arises from the action of attach_task(),
  * which overwrites one tasks cpuset pointer with another.  It does
- * so using both semaphores, however there are several performance
+ * so using both mutexes, however there are several performance
  * critical places that need to reference task->cpuset without the
- * expense of grabbing a system global semaphore.  Therefore except as
+ * expense of grabbing a system global mutex.  Therefore except as
  * noted below, when dereferencing or, as in attach_task(), modifying
  * a tasks cpuset pointer we use task_lock(), which acts on a spinlock
  * (task->alloc_lock) already in the task_struct routinely used for
@@ -256,8 +250,8 @@ static struct super_block *cpuset_sb;
  * the routine cpuset_update_task_memory_state().
  */
 
-static DECLARE_MUTEX(manage_sem);
-static DECLARE_MUTEX(callback_sem);
+static DEFINE_MUTEX(manage_mutex);
+static DEFINE_MUTEX(callback_mutex);
 
 /*
  * A couple of forward declarations required, due to cyclic reference loop:
@@ -432,7 +426,7 @@ static inline struct cftype *__d_cft(struct dentry *dentry)
 }
 
 /*
- * Call with manage_sem held.  Writes path of cpuset into buf.
+ * Call with manage_mutex held.  Writes path of cpuset into buf.
  * Returns 0 on success, -errno on error.
  */
 
@@ -484,11 +478,11 @@ static int cpuset_path(const struct cpuset *cs, char *buf, int buflen)
  * status of the /sbin/cpuset_release_agent task, so no sense holding
  * our caller up for that.
  *
- * When we had only one cpuset semaphore, we had to call this
+ * When we had only one cpuset mutex, we had to call this
  * without holding it, to avoid deadlock when call_usermodehelper()
  * allocated memory.  With two locks, we could now call this while
- * holding manage_sem, but we still don't, so as to minimize
- * the time manage_sem is held.
+ * holding manage_mutex, but we still don't, so as to minimize
+ * the time manage_mutex is held.
  */
 
 static void cpuset_release_agent(const char *pathbuf)
@@ -520,15 +514,15 @@ static void cpuset_release_agent(const char *pathbuf)
  * cs is notify_on_release() and now both the user count is zero and
  * the list of children is empty, prepare cpuset path in a kmalloc'd
  * buffer, to be returned via ppathbuf, so that the caller can invoke
- * cpuset_release_agent() with it later on, once manage_sem is dropped.
- * Call here with manage_sem held.
+ * cpuset_release_agent() with it later on, once manage_mutex is dropped.
+ * Call here with manage_mutex held.
  *
  * This check_for_release() routine is responsible for kmalloc'ing
  * pathbuf.  The above cpuset_release_agent() is responsible for
  * kfree'ing pathbuf.  The caller of these routines is responsible
  * for providing a pathbuf pointer, initialized to NULL, then
- * calling check_for_release() with manage_sem held and the address
- * of the pathbuf pointer, then dropping manage_sem, then calling
+ * calling check_for_release() with manage_mutex held and the address
+ * of the pathbuf pointer, then dropping manage_mutex, then calling
  * cpuset_release_agent() with pathbuf, as set by check_for_release().
  */
 
@@ -559,7 +553,7 @@ static void check_for_release(struct cpuset *cs, char **ppathbuf)
  * One way or another, we guarantee to return some non-empty subset
  * of cpu_online_map.
  *
- * Call with callback_sem held.
+ * Call with callback_mutex held.
  */
 
 static void guarantee_online_cpus(const struct cpuset *cs, cpumask_t *pmask)
@@ -583,7 +577,7 @@ static void guarantee_online_cpus(const struct cpuset *cs, cpumask_t *pmask)
  * One way or another, we guarantee to return some non-empty subset
  * of node_online_map.
  *
- * Call with callback_sem held.
+ * Call with callback_mutex held.
  */
 
 static void guarantee_online_mems(const struct cpuset *cs, nodemask_t *pmask)
@@ -608,12 +602,12 @@ static void guarantee_online_mems(const struct cpuset *cs, nodemask_t *pmask)
  * current->cpuset if a task has its memory placement changed.
  * Do not call this routine if in_interrupt().
  *
- * Call without callback_sem or task_lock() held.  May be called
- * with or without manage_sem held.  Doesn't need task_lock to guard
+ * Call without callback_mutex or task_lock() held.  May be called
+ * with or without manage_mutex held.  Doesn't need task_lock to guard
  * against another task changing a non-NULL cpuset pointer to NULL,
  * as that is only done by a task on itself, and if the current task
  * is here, it is not simultaneously in the exit code NULL'ing its
- * cpuset pointer.  This routine also might acquire callback_sem and
+ * cpuset pointer.  This routine also might acquire callback_mutex and
  * current->mm->mmap_sem during call.
  *
  * Reading current->cpuset->mems_generation doesn't need task_lock
@@ -658,13 +652,13 @@ void cpuset_update_task_memory_state(void)
 	}
 
 	if (my_cpusets_mem_gen != tsk->cpuset_mems_generation) {
-		down(&callback_sem);
+		mutex_lock(&callback_mutex);
 		task_lock(tsk);
 		cs = tsk->cpuset;	/* Maybe changed when task not locked */
 		guarantee_online_mems(cs, &tsk->mems_allowed);
 		tsk->cpuset_mems_generation = cs->mems_generation;
 		task_unlock(tsk);
-		up(&callback_sem);
+		mutex_unlock(&callback_mutex);
 		mpol_rebind_task(tsk, &tsk->mems_allowed);
 	}
 }
@@ -674,7 +668,7 @@ void cpuset_update_task_memory_state(void)
  *
  * One cpuset is a subset of another if all its allowed CPUs and
  * Memory Nodes are a subset of the other, and its exclusive flags
- * are only set if the other's are set.  Call holding manage_sem.
+ * are only set if the other's are set.  Call holding manage_mutex.
  */
 
 static int is_cpuset_subset(const struct cpuset *p, const struct cpuset *q)
@@ -692,7 +686,7 @@ static int is_cpuset_subset(const struct cpuset *p, const struct cpuset *q)
  * If we replaced the flag and mask values of the current cpuset
  * (cur) with those values in the trial cpuset (trial), would
  * our various subset and exclusive rules still be valid?  Presumes
- * manage_sem held.
+ * manage_mutex held.
  *
  * 'cur' is the address of an actual, in-use cpuset.  Operations
  * such as list traversal that depend on the actual address of the
@@ -746,7 +740,7 @@ static int validate_change(const struct cpuset *cur, const struct cpuset *trial)
  *    exclusive child cpusets
  * Build these two partitions by calling partition_sched_domains
  *
- * Call with manage_sem held.  May nest a call to the
+ * Call with manage_mutex held.  May nest a call to the
  * lock_cpu_hotplug()/unlock_cpu_hotplug() pair.
  */
 
@@ -792,7 +786,7 @@ static void update_cpu_domains(struct cpuset *cur)
 }
 
 /*
- * Call with manage_sem held.  May take callback_sem during call.
+ * Call with manage_mutex held.  May take callback_mutex during call.
  */
 
 static int update_cpumask(struct cpuset *cs, char *buf)
@@ -811,9 +805,9 @@ static int update_cpumask(struct cpuset *cs, char *buf)
 	if (retval < 0)
 		return retval;
 	cpus_unchanged = cpus_equal(cs->cpus_allowed, trialcs.cpus_allowed);
-	down(&callback_sem);
+	mutex_lock(&callback_mutex);
 	cs->cpus_allowed = trialcs.cpus_allowed;
-	up(&callback_sem);
+	mutex_unlock(&callback_mutex);
 	if (is_cpu_exclusive(cs) && !cpus_unchanged)
 		update_cpu_domains(cs);
 	return 0;
@@ -827,7 +821,7 @@ static int update_cpumask(struct cpuset *cs, char *buf)
  * the cpuset is marked 'memory_migrate', migrate the tasks
  * pages to the new memory.
  *
- * Call with manage_sem held.  May take callback_sem during call.
+ * Call with manage_mutex held.  May take callback_mutex during call.
  * Will take tasklist_lock, scan tasklist for tasks in cpuset cs,
  * lock each such tasks mm->mmap_sem, scan its vma's and rebind
  * their mempolicies to the cpusets new mems_allowed.
@@ -862,11 +856,11 @@ static int update_nodemask(struct cpuset *cs, char *buf)
 	if (retval < 0)
 		goto done;
 
-	down(&callback_sem);
+	mutex_lock(&callback_mutex);
 	cs->mems_allowed = trialcs.mems_allowed;
 	atomic_inc(&cpuset_mems_generation);
 	cs->mems_generation = atomic_read(&cpuset_mems_generation);
-	up(&callback_sem);
+	mutex_unlock(&callback_mutex);
 
 	set_cpuset_being_rebound(cs);		/* causes mpol_copy() rebind */
 
@@ -922,7 +916,7 @@ static int update_nodemask(struct cpuset *cs, char *buf)
 	 * tasklist_lock.  Forks can happen again now - the mpol_copy()
 	 * cpuset_being_rebound check will catch such forks, and rebind
 	 * their vma mempolicies too.  Because we still hold the global
-	 * cpuset manage_sem, we know that no other rebind effort will
+	 * cpuset manage_mutex, we know that no other rebind effort will
 	 * be contending for the global variable cpuset_being_rebound.
 	 * It's ok if we rebind the same mm twice; mpol_rebind_mm()
 	 * is idempotent.  Also migrate pages in each mm to new nodes.
@@ -948,7 +942,7 @@ done:
 }
 
 /*
- * Call with manage_sem held.
+ * Call with manage_mutex held.
  */
 
 static int update_memory_pressure_enabled(struct cpuset *cs, char *buf)
@@ -967,7 +961,7 @@ static int update_memory_pressure_enabled(struct cpuset *cs, char *buf)
  * cs:	the cpuset to update
  * buf:	the buffer where we read the 0 or 1
  *
- * Call with manage_sem held.
+ * Call with manage_mutex held.
  */
 
 static int update_flag(cpuset_flagbits_t bit, struct cpuset *cs, char *buf)
@@ -989,12 +983,12 @@ static int update_flag(cpuset_flagbits_t bit, struct cpuset *cs, char *buf)
 		return err;
 	cpu_exclusive_changed =
 		(is_cpu_exclusive(cs) != is_cpu_exclusive(&trialcs));
-	down(&callback_sem);
+	mutex_lock(&callback_mutex);
 	if (turning_on)
 		set_bit(bit, &cs->flags);
 	else
 		clear_bit(bit, &cs->flags);
-	up(&callback_sem);
+	mutex_unlock(&callback_mutex);
 
 	if (cpu_exclusive_changed)
                 update_cpu_domains(cs);
@@ -1104,7 +1098,7 @@ static int fmeter_getrate(struct fmeter *fmp)
  * writing the path of the old cpuset in 'ppathbuf' if it needs to be
  * notified on release.
  *
- * Call holding manage_sem.  May take callback_sem and task_lock of
+ * Call holding manage_mutex.  May take callback_mutex and task_lock of
  * the task 'pid' during call.
  */
 
@@ -1144,13 +1138,13 @@ static int attach_task(struct cpuset *cs, char *pidbuf, char **ppathbuf)
 		get_task_struct(tsk);
 	}
 
-	down(&callback_sem);
+	mutex_lock(&callback_mutex);
 
 	task_lock(tsk);
 	oldcs = tsk->cpuset;
 	if (!oldcs) {
 		task_unlock(tsk);
-		up(&callback_sem);
+		mutex_unlock(&callback_mutex);
 		put_task_struct(tsk);
 		return -ESRCH;
 	}
@@ -1164,7 +1158,7 @@ static int attach_task(struct cpuset *cs, char *pidbuf, char **ppathbuf)
 	from = oldcs->mems_allowed;
 	to = cs->mems_allowed;
 
-	up(&callback_sem);
+	mutex_unlock(&callback_mutex);
 
 	mm = get_task_mm(tsk);
 	if (mm) {
@@ -1221,7 +1215,7 @@ static ssize_t cpuset_common_file_write(struct file *file, const char __user *us
 	}
 	buffer[nbytes] = 0;	/* nul-terminate */
 
-	down(&manage_sem);
+	mutex_lock(&manage_mutex);
 
 	if (is_removed(cs)) {
 		retval = -ENODEV;
@@ -1264,7 +1258,7 @@ static ssize_t cpuset_common_file_write(struct file *file, const char __user *us
 	if (retval == 0)
 		retval = nbytes;
 out2:
-	up(&manage_sem);
+	mutex_unlock(&manage_mutex);
 	cpuset_release_agent(pathbuf);
 out1:
 	kfree(buffer);
@@ -1304,9 +1298,9 @@ static int cpuset_sprintf_cpulist(char *page, struct cpuset *cs)
 {
 	cpumask_t mask;
 
-	down(&callback_sem);
+	mutex_lock(&callback_mutex);
 	mask = cs->cpus_allowed;
-	up(&callback_sem);
+	mutex_unlock(&callback_mutex);
 
 	return cpulist_scnprintf(page, PAGE_SIZE, mask);
 }
@@ -1315,9 +1309,9 @@ static int cpuset_sprintf_memlist(char *page, struct cpuset *cs)
 {
 	nodemask_t mask;
 
-	down(&callback_sem);
+	mutex_lock(&callback_mutex);
 	mask = cs->mems_allowed;
-	up(&callback_sem);
+	mutex_unlock(&callback_mutex);
 
 	return nodelist_scnprintf(page, PAGE_SIZE, mask);
 }
@@ -1598,7 +1592,7 @@ static int pid_array_to_buf(char *buf, int sz, pid_t *a, int npids)
  * Handle an open on 'tasks' file.  Prepare a buffer listing the
  * process id's of tasks currently attached to the cpuset being opened.
  *
- * Does not require any specific cpuset semaphores, and does not take any.
+ * Does not require any specific cpuset mutexes, and does not take any.
  */
 static int cpuset_tasks_open(struct inode *unused, struct file *file)
 {
@@ -1754,7 +1748,7 @@ static int cpuset_populate_dir(struct dentry *cs_dentry)
  *	name:		name of the new cpuset. Will be strcpy'ed.
  *	mode:		mode to set on new inode
  *
- *	Must be called with the semaphore on the parent inode held
+ *	Must be called with the mutex on the parent inode held
  */
 
 static long cpuset_create(struct cpuset *parent, const char *name, int mode)
@@ -1766,7 +1760,7 @@ static long cpuset_create(struct cpuset *parent, const char *name, int mode)
 	if (!cs)
 		return -ENOMEM;
 
-	down(&manage_sem);
+	mutex_lock(&manage_mutex);
 	cpuset_update_task_memory_state();
 	cs->flags = 0;
 	if (notify_on_release(parent))
@@ -1782,28 +1776,28 @@ static long cpuset_create(struct cpuset *parent, const char *name, int mode)
 
 	cs->parent = parent;
 
-	down(&callback_sem);
+	mutex_lock(&callback_mutex);
 	list_add(&cs->sibling, &cs->parent->children);
 	number_of_cpusets++;
-	up(&callback_sem);
+	mutex_unlock(&callback_mutex);
 
 	err = cpuset_create_dir(cs, name, mode);
 	if (err < 0)
 		goto err;
 
 	/*
-	 * Release manage_sem before cpuset_populate_dir() because it
+	 * Release manage_mutex before cpuset_populate_dir() because it
 	 * will down() this new directory's i_mutex and if we race with
 	 * another mkdir, we might deadlock.
 	 */
-	up(&manage_sem);
+	mutex_unlock(&manage_mutex);
 
 	err = cpuset_populate_dir(cs->dentry);
 	/* If err < 0, we have a half-filled directory - oh well ;) */
 	return 0;
 err:
 	list_del(&cs->sibling);
-	up(&manage_sem);
+	mutex_unlock(&manage_mutex);
 	kfree(cs);
 	return err;
 }
@@ -1825,18 +1819,18 @@ static int cpuset_rmdir(struct inode *unused_dir, struct dentry *dentry)
 
 	/* the vfs holds both inode->i_mutex already */
 
-	down(&manage_sem);
+	mutex_lock(&manage_mutex);
 	cpuset_update_task_memory_state();
 	if (atomic_read(&cs->count) > 0) {
-		up(&manage_sem);
+		mutex_unlock(&manage_mutex);
 		return -EBUSY;
 	}
 	if (!list_empty(&cs->children)) {
-		up(&manage_sem);
+		mutex_unlock(&manage_mutex);
 		return -EBUSY;
 	}
 	parent = cs->parent;
-	down(&callback_sem);
+	mutex_lock(&callback_mutex);
 	set_bit(CS_REMOVED, &cs->flags);
 	if (is_cpu_exclusive(cs))
 		update_cpu_domains(cs);
@@ -1848,10 +1842,10 @@ static int cpuset_rmdir(struct inode *unused_dir, struct dentry *dentry)
 	cpuset_d_remove_dir(d);
 	dput(d);
 	number_of_cpusets--;
-	up(&callback_sem);
+	mutex_unlock(&callback_mutex);
 	if (list_empty(&parent->children))
 		check_for_release(parent, &pathbuf);
-	up(&manage_sem);
+	mutex_unlock(&manage_mutex);
 	cpuset_release_agent(pathbuf);
 	return 0;
 }
@@ -1960,19 +1954,19 @@ void cpuset_fork(struct task_struct *child)
  * Description: Detach cpuset from @tsk and release it.
  *
  * Note that cpusets marked notify_on_release force every task in
- * them to take the global manage_sem semaphore when exiting.
+ * them to take the global manage_mutex mutex when exiting.
  * This could impact scaling on very large systems.  Be reluctant to
  * use notify_on_release cpusets where very high task exit scaling
  * is required on large systems.
  *
  * Don't even think about derefencing 'cs' after the cpuset use count
- * goes to zero, except inside a critical section guarded by manage_sem
- * or callback_sem.   Otherwise a zero cpuset use count is a license to
+ * goes to zero, except inside a critical section guarded by manage_mutex
+ * or callback_mutex.   Otherwise a zero cpuset use count is a license to
  * any other task to nuke the cpuset immediately, via cpuset_rmdir().
  *
- * This routine has to take manage_sem, not callback_sem, because
- * it is holding that semaphore while calling check_for_release(),
- * which calls kmalloc(), so can't be called holding callback__sem().
+ * This routine has to take manage_mutex, not callback_mutex, because
+ * it is holding that mutex while calling check_for_release(),
+ * which calls kmalloc(), so can't be called holding callback_mutex().
  *
  * We don't need to task_lock() this reference to tsk->cpuset,
  * because tsk is already marked PF_EXITING, so attach_task() won't
@@ -2022,10 +2016,10 @@ void cpuset_exit(struct task_struct *tsk)
 	if (notify_on_release(cs)) {
 		char *pathbuf = NULL;
 
-		down(&manage_sem);
+		mutex_lock(&manage_mutex);
 		if (atomic_dec_and_test(&cs->count))
 			check_for_release(cs, &pathbuf);
-		up(&manage_sem);
+		mutex_unlock(&manage_mutex);
 		cpuset_release_agent(pathbuf);
 	} else {
 		atomic_dec(&cs->count);
@@ -2046,11 +2040,11 @@ cpumask_t cpuset_cpus_allowed(struct task_struct *tsk)
 {
 	cpumask_t mask;
 
-	down(&callback_sem);
+	mutex_lock(&callback_mutex);
 	task_lock(tsk);
 	guarantee_online_cpus(tsk->cpuset, &mask);
 	task_unlock(tsk);
-	up(&callback_sem);
+	mutex_unlock(&callback_mutex);
 
 	return mask;
 }
@@ -2074,11 +2068,11 @@ nodemask_t cpuset_mems_allowed(struct task_struct *tsk)
 {
 	nodemask_t mask;
 
-	down(&callback_sem);
+	mutex_lock(&callback_mutex);
 	task_lock(tsk);
 	guarantee_online_mems(tsk->cpuset, &mask);
 	task_unlock(tsk);
-	up(&callback_sem);
+	mutex_unlock(&callback_mutex);
 
 	return mask;
 }
@@ -2104,7 +2098,7 @@ int cpuset_zonelist_valid_mems_allowed(struct zonelist *zl)
 
 /*
  * nearest_exclusive_ancestor() - Returns the nearest mem_exclusive
- * ancestor to the specified cpuset.  Call holding callback_sem.
+ * ancestor to the specified cpuset.  Call holding callback_mutex.
  * If no ancestor is mem_exclusive (an unusual configuration), then
  * returns the root cpuset.
  */
@@ -2131,12 +2125,12 @@ static const struct cpuset *nearest_exclusive_ancestor(const struct cpuset *cs)
  * GFP_KERNEL allocations are not so marked, so can escape to the
  * nearest mem_exclusive ancestor cpuset.
  *
- * Scanning up parent cpusets requires callback_sem.  The __alloc_pages()
+ * Scanning up parent cpusets requires callback_mutex.  The __alloc_pages()
  * routine only calls here with __GFP_HARDWALL bit _not_ set if
  * it's a GFP_KERNEL allocation, and all nodes in the current tasks
  * mems_allowed came up empty on the first pass over the zonelist.
  * So only GFP_KERNEL allocations, if all nodes in the cpuset are
- * short of memory, might require taking the callback_sem semaphore.
+ * short of memory, might require taking the callback_mutex mutex.
  *
  * The first loop over the zonelist in mm/page_alloc.c:__alloc_pages()
  * calls here with __GFP_HARDWALL always set in gfp_mask, enforcing
@@ -2171,31 +2165,31 @@ int __cpuset_zone_allowed(struct zone *z, gfp_t gfp_mask)
 		return 1;
 
 	/* Not hardwall and node outside mems_allowed: scan up cpusets */
-	down(&callback_sem);
+	mutex_lock(&callback_mutex);
 
 	task_lock(current);
 	cs = nearest_exclusive_ancestor(current->cpuset);
 	task_unlock(current);
 
 	allowed = node_isset(node, cs->mems_allowed);
-	up(&callback_sem);
+	mutex_unlock(&callback_mutex);
 	return allowed;
 }
 
 /**
  * cpuset_lock - lock out any changes to cpuset structures
  *
- * The out of memory (oom) code needs to lock down cpusets
+ * The out of memory (oom) code needs to mutex_lock cpusets
  * from being changed while it scans the tasklist looking for a
- * task in an overlapping cpuset.  Expose callback_sem via this
+ * task in an overlapping cpuset.  Expose callback_mutex via this
  * cpuset_lock() routine, so the oom code can lock it, before
  * locking the task list.  The tasklist_lock is a spinlock, so
- * must be taken inside callback_sem.
+ * must be taken inside callback_mutex.
  */
 
 void cpuset_lock(void)
 {
-	down(&callback_sem);
+	mutex_lock(&callback_mutex);
 }
 
 /**
@@ -2206,7 +2200,7 @@ void cpuset_lock(void)
 
 void cpuset_unlock(void)
 {
-	up(&callback_sem);
+	mutex_unlock(&callback_mutex);
 }
 
 /**
@@ -2218,7 +2212,7 @@ void cpuset_unlock(void)
  * determine if task @p's memory usage might impact the memory
  * available to the current task.
  *
- * Call while holding callback_sem.
+ * Call while holding callback_mutex.
  **/
 
 int cpuset_excl_nodes_overlap(const struct task_struct *p)
@@ -2289,7 +2283,7 @@ void __cpuset_memory_pressure_bump(void)
  *  - Used for /proc/<pid>/cpuset.
  *  - No need to task_lock(tsk) on this tsk->cpuset reference, as it
  *    doesn't really matter if tsk->cpuset changes after we read it,
- *    and we take manage_sem, keeping attach_task() from changing it
+ *    and we take manage_mutex, keeping attach_task() from changing it
  *    anyway.
  */
 
@@ -2305,7 +2299,7 @@ static int proc_cpuset_show(struct seq_file *m, void *v)
 		return -ENOMEM;
 
 	tsk = m->private;
-	down(&manage_sem);
+	mutex_lock(&manage_mutex);
 	cs = tsk->cpuset;
 	if (!cs) {
 		retval = -EINVAL;
@@ -2318,7 +2312,7 @@ static int proc_cpuset_show(struct seq_file *m, void *v)
 	seq_puts(m, buf);
 	seq_putc(m, '\n');
 out:
-	up(&manage_sem);
+	mutex_unlock(&manage_mutex);
 	kfree(buf);
 	return retval;
 }
diff --git a/kernel/exit.c b/kernel/exit.c
index d1e8d500a7e..8037405e136 100644
--- a/kernel/exit.c
+++ b/kernel/exit.c
@@ -345,9 +345,9 @@ void daemonize(const char *name, ...)
 	exit_mm(current);
 
 	set_special_pids(1, 1);
-	down(&tty_sem);
+	mutex_lock(&tty_mutex);
 	current->signal->tty = NULL;
-	up(&tty_sem);
+	mutex_unlock(&tty_mutex);
 
 	/* Block and flush all signals */
 	sigfillset(&blocked);
diff --git a/kernel/fork.c b/kernel/fork.c
index 9bd7b65ee41..c79ae0b19a4 100644
--- a/kernel/fork.c
+++ b/kernel/fork.c
@@ -607,12 +607,12 @@ static struct files_struct *alloc_files(void)
 	atomic_set(&newf->count, 1);
 
 	spin_lock_init(&newf->file_lock);
+	newf->next_fd = 0;
 	fdt = &newf->fdtab;
-	fdt->next_fd = 0;
 	fdt->max_fds = NR_OPEN_DEFAULT;
-	fdt->max_fdset = __FD_SETSIZE;
-	fdt->close_on_exec = &newf->close_on_exec_init;
-	fdt->open_fds = &newf->open_fds_init;
+	fdt->max_fdset = EMBEDDED_FD_SET_SIZE;
+	fdt->close_on_exec = (fd_set *)&newf->close_on_exec_init;
+	fdt->open_fds = (fd_set *)&newf->open_fds_init;
 	fdt->fd = &newf->fd_array[0];
 	INIT_RCU_HEAD(&fdt->rcu);
 	fdt->free_files = NULL;
diff --git a/kernel/kprobes.c b/kernel/kprobes.c
index fef1af8a73c..1fb9f753ef6 100644
--- a/kernel/kprobes.c
+++ b/kernel/kprobes.c
@@ -48,7 +48,7 @@
 static struct hlist_head kprobe_table[KPROBE_TABLE_SIZE];
 static struct hlist_head kretprobe_inst_table[KPROBE_TABLE_SIZE];
 
-DECLARE_MUTEX(kprobe_mutex);		/* Protects kprobe_table */
+DEFINE_MUTEX(kprobe_mutex);		/* Protects kprobe_table */
 DEFINE_SPINLOCK(kretprobe_lock);	/* Protects kretprobe_inst_table */
 static DEFINE_PER_CPU(struct kprobe *, kprobe_instance) = NULL;
 
@@ -460,7 +460,7 @@ static int __kprobes __register_kprobe(struct kprobe *p,
 	}
 
 	p->nmissed = 0;
-	down(&kprobe_mutex);
+	mutex_lock(&kprobe_mutex);
 	old_p = get_kprobe(p->addr);
 	if (old_p) {
 		ret = register_aggr_kprobe(old_p, p);
@@ -477,7 +477,7 @@ static int __kprobes __register_kprobe(struct kprobe *p,
   	arch_arm_kprobe(p);
 
 out:
-	up(&kprobe_mutex);
+	mutex_unlock(&kprobe_mutex);
 
 	if (ret && probed_mod)
 		module_put(probed_mod);
@@ -496,10 +496,10 @@ void __kprobes unregister_kprobe(struct kprobe *p)
 	struct kprobe *old_p, *list_p;
 	int cleanup_p;
 
-	down(&kprobe_mutex);
+	mutex_lock(&kprobe_mutex);
 	old_p = get_kprobe(p->addr);
 	if (unlikely(!old_p)) {
-		up(&kprobe_mutex);
+		mutex_unlock(&kprobe_mutex);
 		return;
 	}
 	if (p != old_p) {
@@ -507,7 +507,7 @@ void __kprobes unregister_kprobe(struct kprobe *p)
 			if (list_p == p)
 			/* kprobe p is a valid probe */
 				goto valid_p;
-		up(&kprobe_mutex);
+		mutex_unlock(&kprobe_mutex);
 		return;
 	}
 valid_p:
@@ -523,7 +523,7 @@ valid_p:
 		cleanup_p = 0;
 	}
 
-	up(&kprobe_mutex);
+	mutex_unlock(&kprobe_mutex);
 
 	synchronize_sched();
 	if (p->mod_refcounted &&
diff --git a/kernel/kthread.c b/kernel/kthread.c
index e75950a1092..6a5373868a9 100644
--- a/kernel/kthread.c
+++ b/kernel/kthread.c
@@ -12,6 +12,7 @@
 #include <linux/unistd.h>
 #include <linux/file.h>
 #include <linux/module.h>
+#include <linux/mutex.h>
 #include <asm/semaphore.h>
 
 /*
@@ -41,7 +42,7 @@ struct kthread_stop_info
 
 /* Thread stopping is done by setthing this var: lock serializes
  * multiple kthread_stop calls. */
-static DECLARE_MUTEX(kthread_stop_lock);
+static DEFINE_MUTEX(kthread_stop_lock);
 static struct kthread_stop_info kthread_stop_info;
 
 int kthread_should_stop(void)
@@ -173,7 +174,7 @@ int kthread_stop_sem(struct task_struct *k, struct semaphore *s)
 {
 	int ret;
 
-	down(&kthread_stop_lock);
+	mutex_lock(&kthread_stop_lock);
 
 	/* It could exit after stop_info.k set, but before wake_up_process. */
 	get_task_struct(k);
@@ -194,7 +195,7 @@ int kthread_stop_sem(struct task_struct *k, struct semaphore *s)
 	wait_for_completion(&kthread_stop_info.done);
 	kthread_stop_info.k = NULL;
 	ret = kthread_stop_info.err;
-	up(&kthread_stop_lock);
+	mutex_unlock(&kthread_stop_lock);
 
 	return ret;
 }
diff --git a/kernel/module.c b/kernel/module.c
index 77764f22f02..fb404299082 100644
--- a/kernel/module.c
+++ b/kernel/module.c
@@ -39,6 +39,7 @@
 #include <linux/device.h>
 #include <linux/string.h>
 #include <linux/sched.h>
+#include <linux/mutex.h>
 #include <asm/uaccess.h>
 #include <asm/semaphore.h>
 #include <asm/cacheflush.h>
@@ -60,18 +61,18 @@
 static DEFINE_SPINLOCK(modlist_lock);
 
 /* List of modules, protected by module_mutex AND modlist_lock */
-static DECLARE_MUTEX(module_mutex);
+static DEFINE_MUTEX(module_mutex);
 static LIST_HEAD(modules);
 
-static DECLARE_MUTEX(notify_mutex);
+static DEFINE_MUTEX(notify_mutex);
 static struct notifier_block * module_notify_list;
 
 int register_module_notifier(struct notifier_block * nb)
 {
 	int err;
-	down(&notify_mutex);
+	mutex_lock(&notify_mutex);
 	err = notifier_chain_register(&module_notify_list, nb);
-	up(&notify_mutex);
+	mutex_unlock(&notify_mutex);
 	return err;
 }
 EXPORT_SYMBOL(register_module_notifier);
@@ -79,9 +80,9 @@ EXPORT_SYMBOL(register_module_notifier);
 int unregister_module_notifier(struct notifier_block * nb)
 {
 	int err;
-	down(&notify_mutex);
+	mutex_lock(&notify_mutex);
 	err = notifier_chain_unregister(&module_notify_list, nb);
-	up(&notify_mutex);
+	mutex_unlock(&notify_mutex);
 	return err;
 }
 EXPORT_SYMBOL(unregister_module_notifier);
@@ -601,7 +602,7 @@ static void free_module(struct module *mod);
 static void wait_for_zero_refcount(struct module *mod)
 {
 	/* Since we might sleep for some time, drop the semaphore first */
-	up(&module_mutex);
+	mutex_unlock(&module_mutex);
 	for (;;) {
 		DEBUGP("Looking at refcount...\n");
 		set_current_state(TASK_UNINTERRUPTIBLE);
@@ -610,7 +611,7 @@ static void wait_for_zero_refcount(struct module *mod)
 		schedule();
 	}
 	current->state = TASK_RUNNING;
-	down(&module_mutex);
+	mutex_lock(&module_mutex);
 }
 
 asmlinkage long
@@ -627,7 +628,7 @@ sys_delete_module(const char __user *name_user, unsigned int flags)
 		return -EFAULT;
 	name[MODULE_NAME_LEN-1] = '\0';
 
-	if (down_interruptible(&module_mutex) != 0)
+	if (mutex_lock_interruptible(&module_mutex) != 0)
 		return -EINTR;
 
 	mod = find_module(name);
@@ -676,14 +677,14 @@ sys_delete_module(const char __user *name_user, unsigned int flags)
 
 	/* Final destruction now noone is using it. */
 	if (mod->exit != NULL) {
-		up(&module_mutex);
+		mutex_unlock(&module_mutex);
 		mod->exit();
-		down(&module_mutex);
+		mutex_lock(&module_mutex);
 	}
 	free_module(mod);
 
  out:
-	up(&module_mutex);
+	mutex_unlock(&module_mutex);
 	return ret;
 }
 
@@ -1972,13 +1973,13 @@ sys_init_module(void __user *umod,
 		return -EPERM;
 
 	/* Only one module load at a time, please */
-	if (down_interruptible(&module_mutex) != 0)
+	if (mutex_lock_interruptible(&module_mutex) != 0)
 		return -EINTR;
 
 	/* Do all the hard work */
 	mod = load_module(umod, len, uargs);
 	if (IS_ERR(mod)) {
-		up(&module_mutex);
+		mutex_unlock(&module_mutex);
 		return PTR_ERR(mod);
 	}
 
@@ -1987,11 +1988,11 @@ sys_init_module(void __user *umod,
 	stop_machine_run(__link_module, mod, NR_CPUS);
 
 	/* Drop lock so they can recurse */
-	up(&module_mutex);
+	mutex_unlock(&module_mutex);
 
-	down(&notify_mutex);
+	mutex_lock(&notify_mutex);
 	notifier_call_chain(&module_notify_list, MODULE_STATE_COMING, mod);
-	up(&notify_mutex);
+	mutex_unlock(&notify_mutex);
 
 	/* Start the module */
 	if (mod->init != NULL)
@@ -2006,15 +2007,15 @@ sys_init_module(void __user *umod,
 			       mod->name);
 		else {
 			module_put(mod);
-			down(&module_mutex);
+			mutex_lock(&module_mutex);
 			free_module(mod);
-			up(&module_mutex);
+			mutex_unlock(&module_mutex);
 		}
 		return ret;
 	}
 
 	/* Now it's a first class citizen! */
-	down(&module_mutex);
+	mutex_lock(&module_mutex);
 	mod->state = MODULE_STATE_LIVE;
 	/* Drop initial reference. */
 	module_put(mod);
@@ -2022,7 +2023,7 @@ sys_init_module(void __user *umod,
 	mod->module_init = NULL;
 	mod->init_size = 0;
 	mod->init_text_size = 0;
-	up(&module_mutex);
+	mutex_unlock(&module_mutex);
 
 	return 0;
 }
@@ -2112,7 +2113,7 @@ struct module *module_get_kallsym(unsigned int symnum,
 {
 	struct module *mod;
 
-	down(&module_mutex);
+	mutex_lock(&module_mutex);
 	list_for_each_entry(mod, &modules, list) {
 		if (symnum < mod->num_symtab) {
 			*value = mod->symtab[symnum].st_value;
@@ -2120,12 +2121,12 @@ struct module *module_get_kallsym(unsigned int symnum,
 			strncpy(namebuf,
 				mod->strtab + mod->symtab[symnum].st_name,
 				127);
-			up(&module_mutex);
+			mutex_unlock(&module_mutex);
 			return mod;
 		}
 		symnum -= mod->num_symtab;
 	}
-	up(&module_mutex);
+	mutex_unlock(&module_mutex);
 	return NULL;
 }
 
@@ -2168,7 +2169,7 @@ static void *m_start(struct seq_file *m, loff_t *pos)
 	struct list_head *i;
 	loff_t n = 0;
 
-	down(&module_mutex);
+	mutex_lock(&module_mutex);
 	list_for_each(i, &modules) {
 		if (n++ == *pos)
 			break;
@@ -2189,7 +2190,7 @@ static void *m_next(struct seq_file *m, void *p, loff_t *pos)
 
 static void m_stop(struct seq_file *m, void *p)
 {
-	up(&module_mutex);
+	mutex_unlock(&module_mutex);
 }
 
 static int m_show(struct seq_file *m, void *p)
diff --git a/kernel/panic.c b/kernel/panic.c
index 126dc43f1c7..acd95adddb9 100644
--- a/kernel/panic.c
+++ b/kernel/panic.c
@@ -20,10 +20,13 @@
 #include <linux/nmi.h>
 #include <linux/kexec.h>
 
-int panic_timeout;
 int panic_on_oops;
 int tainted;
+static int pause_on_oops;
+static int pause_on_oops_flag;
+static DEFINE_SPINLOCK(pause_on_oops_lock);
 
+int panic_timeout;
 EXPORT_SYMBOL(panic_timeout);
 
 struct notifier_block *panic_notifier_list;
@@ -174,3 +177,95 @@ void add_taint(unsigned flag)
 	tainted |= flag;
 }
 EXPORT_SYMBOL(add_taint);
+
+static int __init pause_on_oops_setup(char *str)
+{
+	pause_on_oops = simple_strtoul(str, NULL, 0);
+	return 1;
+}
+__setup("pause_on_oops=", pause_on_oops_setup);
+
+static void spin_msec(int msecs)
+{
+	int i;
+
+	for (i = 0; i < msecs; i++) {
+		touch_nmi_watchdog();
+		mdelay(1);
+	}
+}
+
+/*
+ * It just happens that oops_enter() and oops_exit() are identically
+ * implemented...
+ */
+static void do_oops_enter_exit(void)
+{
+	unsigned long flags;
+	static int spin_counter;
+
+	if (!pause_on_oops)
+		return;
+
+	spin_lock_irqsave(&pause_on_oops_lock, flags);
+	if (pause_on_oops_flag == 0) {
+		/* This CPU may now print the oops message */
+		pause_on_oops_flag = 1;
+	} else {
+		/* We need to stall this CPU */
+		if (!spin_counter) {
+			/* This CPU gets to do the counting */
+			spin_counter = pause_on_oops;
+			do {
+				spin_unlock(&pause_on_oops_lock);
+				spin_msec(MSEC_PER_SEC);
+				spin_lock(&pause_on_oops_lock);
+			} while (--spin_counter);
+			pause_on_oops_flag = 0;
+		} else {
+			/* This CPU waits for a different one */
+			while (spin_counter) {
+				spin_unlock(&pause_on_oops_lock);
+				spin_msec(1);
+				spin_lock(&pause_on_oops_lock);
+			}
+		}
+	}
+	spin_unlock_irqrestore(&pause_on_oops_lock, flags);
+}
+
+/*
+ * Return true if the calling CPU is allowed to print oops-related info.  This
+ * is a bit racy..
+ */
+int oops_may_print(void)
+{
+	return pause_on_oops_flag == 0;
+}
+
+/*
+ * Called when the architecture enters its oops handler, before it prints
+ * anything.  If this is the first CPU to oops, and it's oopsing the first time
+ * then let it proceed.
+ *
+ * This is all enabled by the pause_on_oops kernel boot option.  We do all this
+ * to ensure that oopses don't scroll off the screen.  It has the side-effect
+ * of preventing later-oopsing CPUs from mucking up the display, too.
+ *
+ * It turns out that the CPU which is allowed to print ends up pausing for the
+ * right duration, whereas all the other CPUs pause for twice as long: once in
+ * oops_enter(), once in oops_exit().
+ */
+void oops_enter(void)
+{
+	do_oops_enter_exit();
+}
+
+/*
+ * Called when the architecture exits its oops handler, after printing
+ * everything.
+ */
+void oops_exit(void)
+{
+	do_oops_enter_exit();
+}
diff --git a/kernel/posix-timers.c b/kernel/posix-timers.c
index fa895fc2ecf..9944379360b 100644
--- a/kernel/posix-timers.c
+++ b/kernel/posix-timers.c
@@ -35,6 +35,7 @@
 #include <linux/interrupt.h>
 #include <linux/slab.h>
 #include <linux/time.h>
+#include <linux/mutex.h>
 
 #include <asm/uaccess.h>
 #include <asm/semaphore.h>
diff --git a/kernel/power/Makefile b/kernel/power/Makefile
index 04be7d0d96a..8d0af3d37a4 100644
--- a/kernel/power/Makefile
+++ b/kernel/power/Makefile
@@ -5,7 +5,7 @@ endif
 
 obj-y				:= main.o process.o console.o
 obj-$(CONFIG_PM_LEGACY)		+= pm.o
-obj-$(CONFIG_SOFTWARE_SUSPEND)	+= swsusp.o disk.o snapshot.o
+obj-$(CONFIG_SOFTWARE_SUSPEND)	+= swsusp.o disk.o snapshot.o swap.o user.o
 
 obj-$(CONFIG_SUSPEND_SMP)	+= smp.o
 
diff --git a/kernel/power/disk.c b/kernel/power/disk.c
index 0b43847dc98..81d4d982f3f 100644
--- a/kernel/power/disk.c
+++ b/kernel/power/disk.c
@@ -22,17 +22,6 @@
 #include "power.h"
 
 
-extern suspend_disk_method_t pm_disk_mode;
-
-extern int swsusp_shrink_memory(void);
-extern int swsusp_suspend(void);
-extern int swsusp_write(struct pbe *pblist, unsigned int nr_pages);
-extern int swsusp_check(void);
-extern int swsusp_read(struct pbe **pblist_ptr);
-extern void swsusp_close(void);
-extern int swsusp_resume(void);
-
-
 static int noresume = 0;
 char resume_file[256] = CONFIG_PM_STD_PARTITION;
 dev_t swsusp_resume_device;
@@ -70,10 +59,6 @@ static void power_down(suspend_disk_method_t mode)
 	while(1);
 }
 
-
-static int in_suspend __nosavedata = 0;
-
-
 static inline void platform_finish(void)
 {
 	if (pm_disk_mode == PM_DISK_PLATFORM) {
@@ -87,7 +72,6 @@ static int prepare_processes(void)
 	int error;
 
 	pm_prepare_console();
-	sys_sync();
 	disable_nonboot_cpus();
 
 	if (freeze_processes()) {
@@ -145,7 +129,7 @@ int pm_suspend_disk(void)
 	if (in_suspend) {
 		device_resume();
 		pr_debug("PM: writing image.\n");
-		error = swsusp_write(pagedir_nosave, nr_copy_pages);
+		error = swsusp_write();
 		if (!error)
 			power_down(pm_disk_mode);
 		else {
@@ -216,7 +200,7 @@ static int software_resume(void)
 
 	pr_debug("PM: Reading swsusp image.\n");
 
-	if ((error = swsusp_read(&pagedir_nosave))) {
+	if ((error = swsusp_read())) {
 		swsusp_free();
 		goto Thaw;
 	}
diff --git a/kernel/power/main.c b/kernel/power/main.c
index 9cb235cba4a..ee371f50cca 100644
--- a/kernel/power/main.c
+++ b/kernel/power/main.c
@@ -103,7 +103,7 @@ static int suspend_prepare(suspend_state_t state)
 }
 
 
-static int suspend_enter(suspend_state_t state)
+int suspend_enter(suspend_state_t state)
 {
 	int error = 0;
 	unsigned long flags;
diff --git a/kernel/power/pm.c b/kernel/power/pm.c
index 33c508e857d..0f6908cce1d 100644
--- a/kernel/power/pm.c
+++ b/kernel/power/pm.c
@@ -25,6 +25,7 @@
 #include <linux/pm.h>
 #include <linux/pm_legacy.h>
 #include <linux/interrupt.h>
+#include <linux/mutex.h>
 
 int pm_active;
 
@@ -40,7 +41,7 @@ int pm_active;
  *	until a resume but that will be fine.
  */
  
-static DECLARE_MUTEX(pm_devs_lock);
+static DEFINE_MUTEX(pm_devs_lock);
 static LIST_HEAD(pm_devs);
 
 /**
@@ -67,9 +68,9 @@ struct pm_dev *pm_register(pm_dev_t type,
 		dev->id = id;
 		dev->callback = callback;
 
-		down(&pm_devs_lock);
+		mutex_lock(&pm_devs_lock);
 		list_add(&dev->entry, &pm_devs);
-		up(&pm_devs_lock);
+		mutex_unlock(&pm_devs_lock);
 	}
 	return dev;
 }
@@ -85,9 +86,9 @@ struct pm_dev *pm_register(pm_dev_t type,
 void pm_unregister(struct pm_dev *dev)
 {
 	if (dev) {
-		down(&pm_devs_lock);
+		mutex_lock(&pm_devs_lock);
 		list_del(&dev->entry);
-		up(&pm_devs_lock);
+		mutex_unlock(&pm_devs_lock);
 
 		kfree(dev);
 	}
@@ -118,7 +119,7 @@ void pm_unregister_all(pm_callback callback)
 	if (!callback)
 		return;
 
-	down(&pm_devs_lock);
+	mutex_lock(&pm_devs_lock);
 	entry = pm_devs.next;
 	while (entry != &pm_devs) {
 		struct pm_dev *dev = list_entry(entry, struct pm_dev, entry);
@@ -126,7 +127,7 @@ void pm_unregister_all(pm_callback callback)
 		if (dev->callback == callback)
 			__pm_unregister(dev);
 	}
-	up(&pm_devs_lock);
+	mutex_unlock(&pm_devs_lock);
 }
 
 /**
@@ -234,7 +235,7 @@ int pm_send_all(pm_request_t rqst, void *data)
 {
 	struct list_head *entry;
 	
-	down(&pm_devs_lock);
+	mutex_lock(&pm_devs_lock);
 	entry = pm_devs.next;
 	while (entry != &pm_devs) {
 		struct pm_dev *dev = list_entry(entry, struct pm_dev, entry);
@@ -246,13 +247,13 @@ int pm_send_all(pm_request_t rqst, void *data)
 				 */
 				if (rqst == PM_SUSPEND)
 					pm_undo_all(dev);
-				up(&pm_devs_lock);
+				mutex_unlock(&pm_devs_lock);
 				return status;
 			}
 		}
 		entry = entry->next;
 	}
-	up(&pm_devs_lock);
+	mutex_unlock(&pm_devs_lock);
 	return 0;
 }
 
diff --git a/kernel/power/power.h b/kernel/power/power.h
index 388dba68084..f06f12f2176 100644
--- a/kernel/power/power.h
+++ b/kernel/power/power.h
@@ -8,6 +8,7 @@ struct swsusp_info {
 	int			cpus;
 	unsigned long		image_pages;
 	unsigned long		pages;
+	unsigned long		size;
 } __attribute__((aligned(PAGE_SIZE)));
 
 
@@ -37,21 +38,79 @@ extern struct subsystem power_subsys;
 /* References to section boundaries */
 extern const void __nosave_begin, __nosave_end;
 
-extern unsigned int nr_copy_pages;
 extern struct pbe *pagedir_nosave;
 
 /* Preferred image size in bytes (default 500 MB) */
 extern unsigned long image_size;
+extern int in_suspend;
+extern dev_t swsusp_resume_device;
 
 extern asmlinkage int swsusp_arch_suspend(void);
 extern asmlinkage int swsusp_arch_resume(void);
 
 extern unsigned int count_data_pages(void);
-extern void free_pagedir(struct pbe *pblist);
-extern void release_eaten_pages(void);
-extern struct pbe *alloc_pagedir(unsigned nr_pages, gfp_t gfp_mask, int safe_needed);
+
+struct snapshot_handle {
+	loff_t		offset;
+	unsigned int	page;
+	unsigned int	page_offset;
+	unsigned int	prev;
+	struct pbe	*pbe;
+	void		*buffer;
+	unsigned int	buf_offset;
+};
+
+#define data_of(handle)	((handle).buffer + (handle).buf_offset)
+
+extern int snapshot_read_next(struct snapshot_handle *handle, size_t count);
+extern int snapshot_write_next(struct snapshot_handle *handle, size_t count);
+int snapshot_image_loaded(struct snapshot_handle *handle);
+
+#define SNAPSHOT_IOC_MAGIC	'3'
+#define SNAPSHOT_FREEZE			_IO(SNAPSHOT_IOC_MAGIC, 1)
+#define SNAPSHOT_UNFREEZE		_IO(SNAPSHOT_IOC_MAGIC, 2)
+#define SNAPSHOT_ATOMIC_SNAPSHOT	_IOW(SNAPSHOT_IOC_MAGIC, 3, void *)
+#define SNAPSHOT_ATOMIC_RESTORE		_IO(SNAPSHOT_IOC_MAGIC, 4)
+#define SNAPSHOT_FREE			_IO(SNAPSHOT_IOC_MAGIC, 5)
+#define SNAPSHOT_SET_IMAGE_SIZE		_IOW(SNAPSHOT_IOC_MAGIC, 6, unsigned long)
+#define SNAPSHOT_AVAIL_SWAP		_IOR(SNAPSHOT_IOC_MAGIC, 7, void *)
+#define SNAPSHOT_GET_SWAP_PAGE		_IOR(SNAPSHOT_IOC_MAGIC, 8, void *)
+#define SNAPSHOT_FREE_SWAP_PAGES	_IO(SNAPSHOT_IOC_MAGIC, 9)
+#define SNAPSHOT_SET_SWAP_FILE		_IOW(SNAPSHOT_IOC_MAGIC, 10, unsigned int)
+#define SNAPSHOT_S2RAM			_IO(SNAPSHOT_IOC_MAGIC, 11)
+#define SNAPSHOT_IOC_MAXNR	11
+
+/**
+ *	The bitmap is used for tracing allocated swap pages
+ *
+ *	The entire bitmap consists of a number of bitmap_page
+ *	structures linked with the help of the .next member.
+ *	Thus each page can be allocated individually, so we only
+ *	need to make 0-order memory allocations to create
+ *	the bitmap.
+ */
+
+#define BITMAP_PAGE_SIZE	(PAGE_SIZE - sizeof(void *))
+#define BITMAP_PAGE_CHUNKS	(BITMAP_PAGE_SIZE / sizeof(long))
+#define BITS_PER_CHUNK		(sizeof(long) * 8)
+#define BITMAP_PAGE_BITS	(BITMAP_PAGE_CHUNKS * BITS_PER_CHUNK)
+
+struct bitmap_page {
+	unsigned long		chunks[BITMAP_PAGE_CHUNKS];
+	struct bitmap_page	*next;
+};
+
+extern void free_bitmap(struct bitmap_page *bitmap);
+extern struct bitmap_page *alloc_bitmap(unsigned int nr_bits);
+extern unsigned long alloc_swap_page(int swap, struct bitmap_page *bitmap);
+extern void free_all_swap_pages(int swap, struct bitmap_page *bitmap);
+
+extern int swsusp_check(void);
+extern int swsusp_shrink_memory(void);
 extern void swsusp_free(void);
-extern int alloc_data_pages(struct pbe *pblist, gfp_t gfp_mask, int safe_needed);
-extern unsigned int snapshot_nr_pages(void);
-extern struct pbe *snapshot_pblist(void);
-extern void snapshot_pblist_set(struct pbe *pblist);
+extern int swsusp_suspend(void);
+extern int swsusp_resume(void);
+extern int swsusp_read(void);
+extern int swsusp_write(void);
+extern void swsusp_close(void);
+extern int suspend_enter(suspend_state_t state);
diff --git a/kernel/power/process.c b/kernel/power/process.c
index 28de118f7a0..8ac7c35fad7 100644
--- a/kernel/power/process.c
+++ b/kernel/power/process.c
@@ -12,11 +12,12 @@
 #include <linux/interrupt.h>
 #include <linux/suspend.h>
 #include <linux/module.h>
+#include <linux/syscalls.h>
 
 /* 
  * Timeout for stopping processes
  */
-#define TIMEOUT	(6 * HZ)
+#define TIMEOUT	(20 * HZ)
 
 
 static inline int freezeable(struct task_struct * p)
@@ -54,38 +55,62 @@ void refrigerator(void)
 	current->state = save;
 }
 
+static inline void freeze_process(struct task_struct *p)
+{
+	unsigned long flags;
+
+	if (!freezing(p)) {
+		freeze(p);
+		spin_lock_irqsave(&p->sighand->siglock, flags);
+		signal_wake_up(p, 0);
+		spin_unlock_irqrestore(&p->sighand->siglock, flags);
+	}
+}
+
 /* 0 = success, else # of processes that we failed to stop */
 int freeze_processes(void)
 {
-	int todo;
+	int todo, nr_user, user_frozen;
 	unsigned long start_time;
 	struct task_struct *g, *p;
 	unsigned long flags;
 
 	printk( "Stopping tasks: " );
 	start_time = jiffies;
+	user_frozen = 0;
 	do {
-		todo = 0;
+		nr_user = todo = 0;
 		read_lock(&tasklist_lock);
 		do_each_thread(g, p) {
 			if (!freezeable(p))
 				continue;
 			if (frozen(p))
 				continue;
-
-			freeze(p);
-			spin_lock_irqsave(&p->sighand->siglock, flags);
-			signal_wake_up(p, 0);
-			spin_unlock_irqrestore(&p->sighand->siglock, flags);
-			todo++;
+			if (p->mm && !(p->flags & PF_BORROWED_MM)) {
+				/* The task is a user-space one.
+				 * Freeze it unless there's a vfork completion
+				 * pending
+				 */
+				if (!p->vfork_done)
+					freeze_process(p);
+				nr_user++;
+			} else {
+				/* Freeze only if the user space is frozen */
+				if (user_frozen)
+					freeze_process(p);
+				todo++;
+			}
 		} while_each_thread(g, p);
 		read_unlock(&tasklist_lock);
+		todo += nr_user;
+		if (!user_frozen && !nr_user) {
+			sys_sync();
+			start_time = jiffies;
+		}
+		user_frozen = !nr_user;
 		yield();			/* Yield is okay here */
-		if (todo && time_after(jiffies, start_time + TIMEOUT)) {
-			printk( "\n" );
-			printk(KERN_ERR " stopping tasks failed (%d tasks remaining)\n", todo );
+		if (todo && time_after(jiffies, start_time + TIMEOUT))
 			break;
-		}
 	} while(todo);
 
 	/* This does not unfreeze processes that are already frozen
@@ -94,8 +119,14 @@ int freeze_processes(void)
 	 * but it cleans up leftover PF_FREEZE requests.
 	 */
 	if (todo) {
+		printk( "\n" );
+		printk(KERN_ERR " stopping tasks timed out "
+			"after %d seconds (%d tasks remaining):\n",
+			TIMEOUT / HZ, todo);
 		read_lock(&tasklist_lock);
-		do_each_thread(g, p)
+		do_each_thread(g, p) {
+			if (freezeable(p) && !frozen(p))
+				printk(KERN_ERR "  %s\n", p->comm);
 			if (freezing(p)) {
 				pr_debug("  clean up: %s\n", p->comm);
 				p->flags &= ~PF_FREEZE;
@@ -103,7 +134,7 @@ int freeze_processes(void)
 				recalc_sigpending_tsk(p);
 				spin_unlock_irqrestore(&p->sighand->siglock, flags);
 			}
-		while_each_thread(g, p);
+		} while_each_thread(g, p);
 		read_unlock(&tasklist_lock);
 		return todo;
 	}
diff --git a/kernel/power/snapshot.c b/kernel/power/snapshot.c
index 8d5a5986d62..c5863d02c89 100644
--- a/kernel/power/snapshot.c
+++ b/kernel/power/snapshot.c
@@ -10,6 +10,7 @@
  */
 
 
+#include <linux/version.h>
 #include <linux/module.h>
 #include <linux/mm.h>
 #include <linux/suspend.h>
@@ -34,7 +35,9 @@
 #include "power.h"
 
 struct pbe *pagedir_nosave;
-unsigned int nr_copy_pages;
+static unsigned int nr_copy_pages;
+static unsigned int nr_meta_pages;
+static unsigned long *buffer;
 
 #ifdef CONFIG_HIGHMEM
 unsigned int count_highmem_pages(void)
@@ -80,7 +83,7 @@ static int save_highmem_zone(struct zone *zone)
 		void *kaddr;
 		unsigned long pfn = zone_pfn + zone->zone_start_pfn;
 
-		if (!(pfn%1000))
+		if (!(pfn%10000))
 			printk(".");
 		if (!pfn_valid(pfn))
 			continue;
@@ -119,13 +122,15 @@ int save_highmem(void)
 	struct zone *zone;
 	int res = 0;
 
-	pr_debug("swsusp: Saving Highmem\n");
+	pr_debug("swsusp: Saving Highmem");
+	drain_local_pages();
 	for_each_zone (zone) {
 		if (is_highmem(zone))
 			res = save_highmem_zone(zone);
 		if (res)
 			return res;
 	}
+	printk("\n");
 	return 0;
 }
 
@@ -235,7 +240,7 @@ static void copy_data_pages(struct pbe *pblist)
  *	free_pagedir - free pages allocated with alloc_pagedir()
  */
 
-void free_pagedir(struct pbe *pblist)
+static void free_pagedir(struct pbe *pblist)
 {
 	struct pbe *pbe;
 
@@ -301,7 +306,7 @@ struct eaten_page {
 
 static struct eaten_page *eaten_pages = NULL;
 
-void release_eaten_pages(void)
+static void release_eaten_pages(void)
 {
 	struct eaten_page *p, *q;
 
@@ -376,7 +381,6 @@ struct pbe *alloc_pagedir(unsigned int nr_pages, gfp_t gfp_mask, int safe_needed
 	if (!nr_pages)
 		return NULL;
 
-	pr_debug("alloc_pagedir(): nr_pages = %d\n", nr_pages);
 	pblist = alloc_image_page(gfp_mask, safe_needed);
 	/* FIXME: rewrite this ugly loop */
 	for (pbe = pblist, num = PBES_PER_PAGE; pbe && num < nr_pages;
@@ -388,7 +392,7 @@ struct pbe *alloc_pagedir(unsigned int nr_pages, gfp_t gfp_mask, int safe_needed
 		free_pagedir(pblist);
 		pblist = NULL;
         } else
-        	create_pbe_list(pblist, nr_pages);
+		create_pbe_list(pblist, nr_pages);
 	return pblist;
 }
 
@@ -414,6 +418,10 @@ void swsusp_free(void)
 				}
 			}
 	}
+	nr_copy_pages = 0;
+	nr_meta_pages = 0;
+	pagedir_nosave = NULL;
+	buffer = NULL;
 }
 
 
@@ -437,7 +445,7 @@ static int enough_free_mem(unsigned int nr_pages)
 		(nr_pages + PBES_PER_PAGE - 1) / PBES_PER_PAGE);
 }
 
-int alloc_data_pages(struct pbe *pblist, gfp_t gfp_mask, int safe_needed)
+static int alloc_data_pages(struct pbe *pblist, gfp_t gfp_mask, int safe_needed)
 {
 	struct pbe *p;
 
@@ -504,7 +512,318 @@ asmlinkage int swsusp_save(void)
 	 */
 
 	nr_copy_pages = nr_pages;
+	nr_meta_pages = (nr_pages * sizeof(long) + PAGE_SIZE - 1) >> PAGE_SHIFT;
 
 	printk("swsusp: critical section/: done (%d pages copied)\n", nr_pages);
 	return 0;
 }
+
+static void init_header(struct swsusp_info *info)
+{
+	memset(info, 0, sizeof(struct swsusp_info));
+	info->version_code = LINUX_VERSION_CODE;
+	info->num_physpages = num_physpages;
+	memcpy(&info->uts, &system_utsname, sizeof(system_utsname));
+	info->cpus = num_online_cpus();
+	info->image_pages = nr_copy_pages;
+	info->pages = nr_copy_pages + nr_meta_pages + 1;
+	info->size = info->pages;
+	info->size <<= PAGE_SHIFT;
+}
+
+/**
+ *	pack_orig_addresses - the .orig_address fields of the PBEs from the
+ *	list starting at @pbe are stored in the array @buf[] (1 page)
+ */
+
+static inline struct pbe *pack_orig_addresses(unsigned long *buf, struct pbe *pbe)
+{
+	int j;
+
+	for (j = 0; j < PAGE_SIZE / sizeof(long) && pbe; j++) {
+		buf[j] = pbe->orig_address;
+		pbe = pbe->next;
+	}
+	if (!pbe)
+		for (; j < PAGE_SIZE / sizeof(long); j++)
+			buf[j] = 0;
+	return pbe;
+}
+
+/**
+ *	snapshot_read_next - used for reading the system memory snapshot.
+ *
+ *	On the first call to it @handle should point to a zeroed
+ *	snapshot_handle structure.  The structure gets updated and a pointer
+ *	to it should be passed to this function every next time.
+ *
+ *	The @count parameter should contain the number of bytes the caller
+ *	wants to read from the snapshot.  It must not be zero.
+ *
+ *	On success the function returns a positive number.  Then, the caller
+ *	is allowed to read up to the returned number of bytes from the memory
+ *	location computed by the data_of() macro.  The number returned
+ *	may be smaller than @count, but this only happens if the read would
+ *	cross a page boundary otherwise.
+ *
+ *	The function returns 0 to indicate the end of data stream condition,
+ *	and a negative number is returned on error.  In such cases the
+ *	structure pointed to by @handle is not updated and should not be used
+ *	any more.
+ */
+
+int snapshot_read_next(struct snapshot_handle *handle, size_t count)
+{
+	if (handle->page > nr_meta_pages + nr_copy_pages)
+		return 0;
+	if (!buffer) {
+		/* This makes the buffer be freed by swsusp_free() */
+		buffer = alloc_image_page(GFP_ATOMIC, 0);
+		if (!buffer)
+			return -ENOMEM;
+	}
+	if (!handle->offset) {
+		init_header((struct swsusp_info *)buffer);
+		handle->buffer = buffer;
+		handle->pbe = pagedir_nosave;
+	}
+	if (handle->prev < handle->page) {
+		if (handle->page <= nr_meta_pages) {
+			handle->pbe = pack_orig_addresses(buffer, handle->pbe);
+			if (!handle->pbe)
+				handle->pbe = pagedir_nosave;
+		} else {
+			handle->buffer = (void *)handle->pbe->address;
+			handle->pbe = handle->pbe->next;
+		}
+		handle->prev = handle->page;
+	}
+	handle->buf_offset = handle->page_offset;
+	if (handle->page_offset + count >= PAGE_SIZE) {
+		count = PAGE_SIZE - handle->page_offset;
+		handle->page_offset = 0;
+		handle->page++;
+	} else {
+		handle->page_offset += count;
+	}
+	handle->offset += count;
+	return count;
+}
+
+/**
+ *	mark_unsafe_pages - mark the pages that cannot be used for storing
+ *	the image during resume, because they conflict with the pages that
+ *	had been used before suspend
+ */
+
+static int mark_unsafe_pages(struct pbe *pblist)
+{
+	struct zone *zone;
+	unsigned long zone_pfn;
+	struct pbe *p;
+
+	if (!pblist) /* a sanity check */
+		return -EINVAL;
+
+	/* Clear page flags */
+	for_each_zone (zone) {
+		for (zone_pfn = 0; zone_pfn < zone->spanned_pages; ++zone_pfn)
+			if (pfn_valid(zone_pfn + zone->zone_start_pfn))
+				ClearPageNosaveFree(pfn_to_page(zone_pfn +
+					zone->zone_start_pfn));
+	}
+
+	/* Mark orig addresses */
+	for_each_pbe (p, pblist) {
+		if (virt_addr_valid(p->orig_address))
+			SetPageNosaveFree(virt_to_page(p->orig_address));
+		else
+			return -EFAULT;
+	}
+
+	return 0;
+}
+
+static void copy_page_backup_list(struct pbe *dst, struct pbe *src)
+{
+	/* We assume both lists contain the same number of elements */
+	while (src) {
+		dst->orig_address = src->orig_address;
+		dst = dst->next;
+		src = src->next;
+	}
+}
+
+static int check_header(struct swsusp_info *info)
+{
+	char *reason = NULL;
+
+	if (info->version_code != LINUX_VERSION_CODE)
+		reason = "kernel version";
+	if (info->num_physpages != num_physpages)
+		reason = "memory size";
+	if (strcmp(info->uts.sysname,system_utsname.sysname))
+		reason = "system type";
+	if (strcmp(info->uts.release,system_utsname.release))
+		reason = "kernel release";
+	if (strcmp(info->uts.version,system_utsname.version))
+		reason = "version";
+	if (strcmp(info->uts.machine,system_utsname.machine))
+		reason = "machine";
+	if (reason) {
+		printk(KERN_ERR "swsusp: Resume mismatch: %s\n", reason);
+		return -EPERM;
+	}
+	return 0;
+}
+
+/**
+ *	load header - check the image header and copy data from it
+ */
+
+static int load_header(struct snapshot_handle *handle,
+                              struct swsusp_info *info)
+{
+	int error;
+	struct pbe *pblist;
+
+	error = check_header(info);
+	if (!error) {
+		pblist = alloc_pagedir(info->image_pages, GFP_ATOMIC, 0);
+		if (!pblist)
+			return -ENOMEM;
+		pagedir_nosave = pblist;
+		handle->pbe = pblist;
+		nr_copy_pages = info->image_pages;
+		nr_meta_pages = info->pages - info->image_pages - 1;
+	}
+	return error;
+}
+
+/**
+ *	unpack_orig_addresses - copy the elements of @buf[] (1 page) to
+ *	the PBEs in the list starting at @pbe
+ */
+
+static inline struct pbe *unpack_orig_addresses(unsigned long *buf,
+                                                struct pbe *pbe)
+{
+	int j;
+
+	for (j = 0; j < PAGE_SIZE / sizeof(long) && pbe; j++) {
+		pbe->orig_address = buf[j];
+		pbe = pbe->next;
+	}
+	return pbe;
+}
+
+/**
+ *	create_image - use metadata contained in the PBE list
+ *	pointed to by pagedir_nosave to mark the pages that will
+ *	be overwritten in the process of restoring the system
+ *	memory state from the image and allocate memory for
+ *	the image avoiding these pages
+ */
+
+static int create_image(struct snapshot_handle *handle)
+{
+	int error = 0;
+	struct pbe *p, *pblist;
+
+	p = pagedir_nosave;
+	error = mark_unsafe_pages(p);
+	if (!error) {
+		pblist = alloc_pagedir(nr_copy_pages, GFP_ATOMIC, 1);
+		if (pblist)
+			copy_page_backup_list(pblist, p);
+		free_pagedir(p);
+		if (!pblist)
+			error = -ENOMEM;
+	}
+	if (!error)
+		error = alloc_data_pages(pblist, GFP_ATOMIC, 1);
+	if (!error) {
+		release_eaten_pages();
+		pagedir_nosave = pblist;
+	} else {
+		pagedir_nosave = NULL;
+		handle->pbe = NULL;
+		nr_copy_pages = 0;
+		nr_meta_pages = 0;
+	}
+	return error;
+}
+
+/**
+ *	snapshot_write_next - used for writing the system memory snapshot.
+ *
+ *	On the first call to it @handle should point to a zeroed
+ *	snapshot_handle structure.  The structure gets updated and a pointer
+ *	to it should be passed to this function every next time.
+ *
+ *	The @count parameter should contain the number of bytes the caller
+ *	wants to write to the image.  It must not be zero.
+ *
+ *	On success the function returns a positive number.  Then, the caller
+ *	is allowed to write up to the returned number of bytes to the memory
+ *	location computed by the data_of() macro.  The number returned
+ *	may be smaller than @count, but this only happens if the write would
+ *	cross a page boundary otherwise.
+ *
+ *	The function returns 0 to indicate the "end of file" condition,
+ *	and a negative number is returned on error.  In such cases the
+ *	structure pointed to by @handle is not updated and should not be used
+ *	any more.
+ */
+
+int snapshot_write_next(struct snapshot_handle *handle, size_t count)
+{
+	int error = 0;
+
+	if (handle->prev && handle->page > nr_meta_pages + nr_copy_pages)
+		return 0;
+	if (!buffer) {
+		/* This makes the buffer be freed by swsusp_free() */
+		buffer = alloc_image_page(GFP_ATOMIC, 0);
+		if (!buffer)
+			return -ENOMEM;
+	}
+	if (!handle->offset)
+		handle->buffer = buffer;
+	if (handle->prev < handle->page) {
+		if (!handle->prev) {
+			error = load_header(handle, (struct swsusp_info *)buffer);
+			if (error)
+				return error;
+		} else if (handle->prev <= nr_meta_pages) {
+			handle->pbe = unpack_orig_addresses(buffer, handle->pbe);
+			if (!handle->pbe) {
+				error = create_image(handle);
+				if (error)
+					return error;
+				handle->pbe = pagedir_nosave;
+				handle->buffer = (void *)handle->pbe->address;
+			}
+		} else {
+			handle->pbe = handle->pbe->next;
+			handle->buffer = (void *)handle->pbe->address;
+		}
+		handle->prev = handle->page;
+	}
+	handle->buf_offset = handle->page_offset;
+	if (handle->page_offset + count >= PAGE_SIZE) {
+		count = PAGE_SIZE - handle->page_offset;
+		handle->page_offset = 0;
+		handle->page++;
+	} else {
+		handle->page_offset += count;
+	}
+	handle->offset += count;
+	return count;
+}
+
+int snapshot_image_loaded(struct snapshot_handle *handle)
+{
+	return !(!handle->pbe || handle->pbe->next || !nr_copy_pages ||
+		handle->page <= nr_meta_pages + nr_copy_pages);
+}
diff --git a/kernel/power/swap.c b/kernel/power/swap.c
new file mode 100644
index 00000000000..9177f3f73a6
--- /dev/null
+++ b/kernel/power/swap.c
@@ -0,0 +1,544 @@
+/*
+ * linux/kernel/power/swap.c
+ *
+ * This file provides functions for reading the suspend image from
+ * and writing it to a swap partition.
+ *
+ * Copyright (C) 1998,2001-2005 Pavel Machek <pavel@suse.cz>
+ * Copyright (C) 2006 Rafael J. Wysocki <rjw@sisk.pl>
+ *
+ * This file is released under the GPLv2.
+ *
+ */
+
+#include <linux/module.h>
+#include <linux/smp_lock.h>
+#include <linux/file.h>
+#include <linux/utsname.h>
+#include <linux/version.h>
+#include <linux/delay.h>
+#include <linux/bitops.h>
+#include <linux/genhd.h>
+#include <linux/device.h>
+#include <linux/buffer_head.h>
+#include <linux/bio.h>
+#include <linux/swap.h>
+#include <linux/swapops.h>
+#include <linux/pm.h>
+
+#include "power.h"
+
+extern char resume_file[];
+
+#define SWSUSP_SIG	"S1SUSPEND"
+
+static struct swsusp_header {
+	char reserved[PAGE_SIZE - 20 - sizeof(swp_entry_t)];
+	swp_entry_t image;
+	char	orig_sig[10];
+	char	sig[10];
+} __attribute__((packed, aligned(PAGE_SIZE))) swsusp_header;
+
+/*
+ * Saving part...
+ */
+
+static unsigned short root_swap = 0xffff;
+
+static int mark_swapfiles(swp_entry_t start)
+{
+	int error;
+
+	rw_swap_page_sync(READ,
+			  swp_entry(root_swap, 0),
+			  virt_to_page((unsigned long)&swsusp_header));
+	if (!memcmp("SWAP-SPACE",swsusp_header.sig, 10) ||
+	    !memcmp("SWAPSPACE2",swsusp_header.sig, 10)) {
+		memcpy(swsusp_header.orig_sig,swsusp_header.sig, 10);
+		memcpy(swsusp_header.sig,SWSUSP_SIG, 10);
+		swsusp_header.image = start;
+		error = rw_swap_page_sync(WRITE,
+					  swp_entry(root_swap, 0),
+					  virt_to_page((unsigned long)
+						       &swsusp_header));
+	} else {
+		pr_debug("swsusp: Partition is not swap space.\n");
+		error = -ENODEV;
+	}
+	return error;
+}
+
+/**
+ *	swsusp_swap_check - check if the resume device is a swap device
+ *	and get its index (if so)
+ */
+
+static int swsusp_swap_check(void) /* This is called before saving image */
+{
+	int res = swap_type_of(swsusp_resume_device);
+
+	if (res >= 0) {
+		root_swap = res;
+		return 0;
+	}
+	return res;
+}
+
+/**
+ *	write_page - Write one page to given swap location.
+ *	@buf:		Address we're writing.
+ *	@offset:	Offset of the swap page we're writing to.
+ */
+
+static int write_page(void *buf, unsigned long offset)
+{
+	swp_entry_t entry;
+	int error = -ENOSPC;
+
+	if (offset) {
+		entry = swp_entry(root_swap, offset);
+		error = rw_swap_page_sync(WRITE, entry, virt_to_page(buf));
+	}
+	return error;
+}
+
+/*
+ *	The swap map is a data structure used for keeping track of each page
+ *	written to a swap partition.  It consists of many swap_map_page
+ *	structures that contain each an array of MAP_PAGE_SIZE swap entries.
+ *	These structures are stored on the swap and linked together with the
+ *	help of the .next_swap member.
+ *
+ *	The swap map is created during suspend.  The swap map pages are
+ *	allocated and populated one at a time, so we only need one memory
+ *	page to set up the entire structure.
+ *
+ *	During resume we also only need to use one swap_map_page structure
+ *	at a time.
+ */
+
+#define MAP_PAGE_ENTRIES	(PAGE_SIZE / sizeof(long) - 1)
+
+struct swap_map_page {
+	unsigned long		entries[MAP_PAGE_ENTRIES];
+	unsigned long		next_swap;
+};
+
+/**
+ *	The swap_map_handle structure is used for handling swap in
+ *	a file-alike way
+ */
+
+struct swap_map_handle {
+	struct swap_map_page *cur;
+	unsigned long cur_swap;
+	struct bitmap_page *bitmap;
+	unsigned int k;
+};
+
+static void release_swap_writer(struct swap_map_handle *handle)
+{
+	if (handle->cur)
+		free_page((unsigned long)handle->cur);
+	handle->cur = NULL;
+	if (handle->bitmap)
+		free_bitmap(handle->bitmap);
+	handle->bitmap = NULL;
+}
+
+static int get_swap_writer(struct swap_map_handle *handle)
+{
+	handle->cur = (struct swap_map_page *)get_zeroed_page(GFP_KERNEL);
+	if (!handle->cur)
+		return -ENOMEM;
+	handle->bitmap = alloc_bitmap(count_swap_pages(root_swap, 0));
+	if (!handle->bitmap) {
+		release_swap_writer(handle);
+		return -ENOMEM;
+	}
+	handle->cur_swap = alloc_swap_page(root_swap, handle->bitmap);
+	if (!handle->cur_swap) {
+		release_swap_writer(handle);
+		return -ENOSPC;
+	}
+	handle->k = 0;
+	return 0;
+}
+
+static int swap_write_page(struct swap_map_handle *handle, void *buf)
+{
+	int error;
+	unsigned long offset;
+
+	if (!handle->cur)
+		return -EINVAL;
+	offset = alloc_swap_page(root_swap, handle->bitmap);
+	error = write_page(buf, offset);
+	if (error)
+		return error;
+	handle->cur->entries[handle->k++] = offset;
+	if (handle->k >= MAP_PAGE_ENTRIES) {
+		offset = alloc_swap_page(root_swap, handle->bitmap);
+		if (!offset)
+			return -ENOSPC;
+		handle->cur->next_swap = offset;
+		error = write_page(handle->cur, handle->cur_swap);
+		if (error)
+			return error;
+		memset(handle->cur, 0, PAGE_SIZE);
+		handle->cur_swap = offset;
+		handle->k = 0;
+	}
+	return 0;
+}
+
+static int flush_swap_writer(struct swap_map_handle *handle)
+{
+	if (handle->cur && handle->cur_swap)
+		return write_page(handle->cur, handle->cur_swap);
+	else
+		return -EINVAL;
+}
+
+/**
+ *	save_image - save the suspend image data
+ */
+
+static int save_image(struct swap_map_handle *handle,
+                      struct snapshot_handle *snapshot,
+                      unsigned int nr_pages)
+{
+	unsigned int m;
+	int ret;
+	int error = 0;
+
+	printk("Saving image data pages (%u pages) ...     ", nr_pages);
+	m = nr_pages / 100;
+	if (!m)
+		m = 1;
+	nr_pages = 0;
+	do {
+		ret = snapshot_read_next(snapshot, PAGE_SIZE);
+		if (ret > 0) {
+			error = swap_write_page(handle, data_of(*snapshot));
+			if (error)
+				break;
+			if (!(nr_pages % m))
+				printk("\b\b\b\b%3d%%", nr_pages / m);
+			nr_pages++;
+		}
+	} while (ret > 0);
+	if (!error)
+		printk("\b\b\b\bdone\n");
+	return error;
+}
+
+/**
+ *	enough_swap - Make sure we have enough swap to save the image.
+ *
+ *	Returns TRUE or FALSE after checking the total amount of swap
+ *	space avaiable from the resume partition.
+ */
+
+static int enough_swap(unsigned int nr_pages)
+{
+	unsigned int free_swap = count_swap_pages(root_swap, 1);
+
+	pr_debug("swsusp: free swap pages: %u\n", free_swap);
+	return free_swap > (nr_pages + PAGES_FOR_IO +
+		(nr_pages + PBES_PER_PAGE - 1) / PBES_PER_PAGE);
+}
+
+/**
+ *	swsusp_write - Write entire image and metadata.
+ *
+ *	It is important _NOT_ to umount filesystems at this point. We want
+ *	them synced (in case something goes wrong) but we DO not want to mark
+ *	filesystem clean: it is not. (And it does not matter, if we resume
+ *	correctly, we'll mark system clean, anyway.)
+ */
+
+int swsusp_write(void)
+{
+	struct swap_map_handle handle;
+	struct snapshot_handle snapshot;
+	struct swsusp_info *header;
+	unsigned long start;
+	int error;
+
+	if ((error = swsusp_swap_check())) {
+		printk(KERN_ERR "swsusp: Cannot find swap device, try swapon -a.\n");
+		return error;
+	}
+	memset(&snapshot, 0, sizeof(struct snapshot_handle));
+	error = snapshot_read_next(&snapshot, PAGE_SIZE);
+	if (error < PAGE_SIZE)
+		return error < 0 ? error : -EFAULT;
+	header = (struct swsusp_info *)data_of(snapshot);
+	if (!enough_swap(header->pages)) {
+		printk(KERN_ERR "swsusp: Not enough free swap\n");
+		return -ENOSPC;
+	}
+	error = get_swap_writer(&handle);
+	if (!error) {
+		start = handle.cur_swap;
+		error = swap_write_page(&handle, header);
+	}
+	if (!error)
+		error = save_image(&handle, &snapshot, header->pages - 1);
+	if (!error) {
+		flush_swap_writer(&handle);
+		printk("S");
+		error = mark_swapfiles(swp_entry(root_swap, start));
+		printk("|\n");
+	}
+	if (error)
+		free_all_swap_pages(root_swap, handle.bitmap);
+	release_swap_writer(&handle);
+	return error;
+}
+
+/*
+ *	Using bio to read from swap.
+ *	This code requires a bit more work than just using buffer heads
+ *	but, it is the recommended way for 2.5/2.6.
+ *	The following are to signal the beginning and end of I/O. Bios
+ *	finish asynchronously, while we want them to happen synchronously.
+ *	A simple atomic_t, and a wait loop take care of this problem.
+ */
+
+static atomic_t io_done = ATOMIC_INIT(0);
+
+static int end_io(struct bio *bio, unsigned int num, int err)
+{
+	if (!test_bit(BIO_UPTODATE, &bio->bi_flags))
+		panic("I/O error reading memory image");
+	atomic_set(&io_done, 0);
+	return 0;
+}
+
+static struct block_device *resume_bdev;
+
+/**
+ *	submit - submit BIO request.
+ *	@rw:	READ or WRITE.
+ *	@off	physical offset of page.
+ *	@page:	page we're reading or writing.
+ *
+ *	Straight from the textbook - allocate and initialize the bio.
+ *	If we're writing, make sure the page is marked as dirty.
+ *	Then submit it and wait.
+ */
+
+static int submit(int rw, pgoff_t page_off, void *page)
+{
+	int error = 0;
+	struct bio *bio;
+
+	bio = bio_alloc(GFP_ATOMIC, 1);
+	if (!bio)
+		return -ENOMEM;
+	bio->bi_sector = page_off * (PAGE_SIZE >> 9);
+	bio->bi_bdev = resume_bdev;
+	bio->bi_end_io = end_io;
+
+	if (bio_add_page(bio, virt_to_page(page), PAGE_SIZE, 0) < PAGE_SIZE) {
+		printk("swsusp: ERROR: adding page to bio at %ld\n",page_off);
+		error = -EFAULT;
+		goto Done;
+	}
+
+	atomic_set(&io_done, 1);
+	submit_bio(rw | (1 << BIO_RW_SYNC), bio);
+	while (atomic_read(&io_done))
+		yield();
+	if (rw == READ)
+		bio_set_pages_dirty(bio);
+ Done:
+	bio_put(bio);
+	return error;
+}
+
+static int bio_read_page(pgoff_t page_off, void *page)
+{
+	return submit(READ, page_off, page);
+}
+
+static int bio_write_page(pgoff_t page_off, void *page)
+{
+	return submit(WRITE, page_off, page);
+}
+
+/**
+ *	The following functions allow us to read data using a swap map
+ *	in a file-alike way
+ */
+
+static void release_swap_reader(struct swap_map_handle *handle)
+{
+	if (handle->cur)
+		free_page((unsigned long)handle->cur);
+	handle->cur = NULL;
+}
+
+static int get_swap_reader(struct swap_map_handle *handle,
+                                      swp_entry_t start)
+{
+	int error;
+
+	if (!swp_offset(start))
+		return -EINVAL;
+	handle->cur = (struct swap_map_page *)get_zeroed_page(GFP_ATOMIC);
+	if (!handle->cur)
+		return -ENOMEM;
+	error = bio_read_page(swp_offset(start), handle->cur);
+	if (error) {
+		release_swap_reader(handle);
+		return error;
+	}
+	handle->k = 0;
+	return 0;
+}
+
+static int swap_read_page(struct swap_map_handle *handle, void *buf)
+{
+	unsigned long offset;
+	int error;
+
+	if (!handle->cur)
+		return -EINVAL;
+	offset = handle->cur->entries[handle->k];
+	if (!offset)
+		return -EFAULT;
+	error = bio_read_page(offset, buf);
+	if (error)
+		return error;
+	if (++handle->k >= MAP_PAGE_ENTRIES) {
+		handle->k = 0;
+		offset = handle->cur->next_swap;
+		if (!offset)
+			release_swap_reader(handle);
+		else
+			error = bio_read_page(offset, handle->cur);
+	}
+	return error;
+}
+
+/**
+ *	load_image - load the image using the swap map handle
+ *	@handle and the snapshot handle @snapshot
+ *	(assume there are @nr_pages pages to load)
+ */
+
+static int load_image(struct swap_map_handle *handle,
+                      struct snapshot_handle *snapshot,
+                      unsigned int nr_pages)
+{
+	unsigned int m;
+	int ret;
+	int error = 0;
+
+	printk("Loading image data pages (%u pages) ...     ", nr_pages);
+	m = nr_pages / 100;
+	if (!m)
+		m = 1;
+	nr_pages = 0;
+	do {
+		ret = snapshot_write_next(snapshot, PAGE_SIZE);
+		if (ret > 0) {
+			error = swap_read_page(handle, data_of(*snapshot));
+			if (error)
+				break;
+			if (!(nr_pages % m))
+				printk("\b\b\b\b%3d%%", nr_pages / m);
+			nr_pages++;
+		}
+	} while (ret > 0);
+	if (!error)
+		printk("\b\b\b\bdone\n");
+	if (!snapshot_image_loaded(snapshot))
+		error = -ENODATA;
+	return error;
+}
+
+int swsusp_read(void)
+{
+	int error;
+	struct swap_map_handle handle;
+	struct snapshot_handle snapshot;
+	struct swsusp_info *header;
+
+	if (IS_ERR(resume_bdev)) {
+		pr_debug("swsusp: block device not initialised\n");
+		return PTR_ERR(resume_bdev);
+	}
+
+	memset(&snapshot, 0, sizeof(struct snapshot_handle));
+	error = snapshot_write_next(&snapshot, PAGE_SIZE);
+	if (error < PAGE_SIZE)
+		return error < 0 ? error : -EFAULT;
+	header = (struct swsusp_info *)data_of(snapshot);
+	error = get_swap_reader(&handle, swsusp_header.image);
+	if (!error)
+		error = swap_read_page(&handle, header);
+	if (!error)
+		error = load_image(&handle, &snapshot, header->pages - 1);
+	release_swap_reader(&handle);
+
+	blkdev_put(resume_bdev);
+
+	if (!error)
+		pr_debug("swsusp: Reading resume file was successful\n");
+	else
+		pr_debug("swsusp: Error %d resuming\n", error);
+	return error;
+}
+
+/**
+ *      swsusp_check - Check for swsusp signature in the resume device
+ */
+
+int swsusp_check(void)
+{
+	int error;
+
+	resume_bdev = open_by_devnum(swsusp_resume_device, FMODE_READ);
+	if (!IS_ERR(resume_bdev)) {
+		set_blocksize(resume_bdev, PAGE_SIZE);
+		memset(&swsusp_header, 0, sizeof(swsusp_header));
+		if ((error = bio_read_page(0, &swsusp_header)))
+			return error;
+		if (!memcmp(SWSUSP_SIG, swsusp_header.sig, 10)) {
+			memcpy(swsusp_header.sig, swsusp_header.orig_sig, 10);
+			/* Reset swap signature now */
+			error = bio_write_page(0, &swsusp_header);
+		} else {
+			return -EINVAL;
+		}
+		if (error)
+			blkdev_put(resume_bdev);
+		else
+			pr_debug("swsusp: Signature found, resuming\n");
+	} else {
+		error = PTR_ERR(resume_bdev);
+	}
+
+	if (error)
+		pr_debug("swsusp: Error %d check for resume file\n", error);
+
+	return error;
+}
+
+/**
+ *	swsusp_close - close swap device.
+ */
+
+void swsusp_close(void)
+{
+	if (IS_ERR(resume_bdev)) {
+		pr_debug("swsusp: block device not initialised\n");
+		return;
+	}
+
+	blkdev_put(resume_bdev);
+}
diff --git a/kernel/power/swsusp.c b/kernel/power/swsusp.c
index 2d9d08f72f7..c4016cbbd3e 100644
--- a/kernel/power/swsusp.c
+++ b/kernel/power/swsusp.c
@@ -31,41 +31,24 @@
  * Fixed runaway init
  *
  * Rafael J. Wysocki <rjw@sisk.pl>
- * Added the swap map data structure and reworked the handling of swap
+ * Reworked the freeing of memory and the handling of swap
  *
  * More state savers are welcome. Especially for the scsi layer...
  *
  * For TODOs,FIXMEs also look in Documentation/power/swsusp.txt
  */
 
-#include <linux/module.h>
 #include <linux/mm.h>
 #include <linux/suspend.h>
-#include <linux/smp_lock.h>
-#include <linux/file.h>
-#include <linux/utsname.h>
-#include <linux/version.h>
-#include <linux/delay.h>
-#include <linux/bitops.h>
 #include <linux/spinlock.h>
-#include <linux/genhd.h>
 #include <linux/kernel.h>
 #include <linux/major.h>
 #include <linux/swap.h>
 #include <linux/pm.h>
-#include <linux/device.h>
-#include <linux/buffer_head.h>
 #include <linux/swapops.h>
 #include <linux/bootmem.h>
 #include <linux/syscalls.h>
 #include <linux/highmem.h>
-#include <linux/bio.h>
-
-#include <asm/uaccess.h>
-#include <asm/mmu_context.h>
-#include <asm/pgtable.h>
-#include <asm/tlbflush.h>
-#include <asm/io.h>
 
 #include "power.h"
 
@@ -77,6 +60,8 @@
  */
 unsigned long image_size = 500 * 1024 * 1024;
 
+int in_suspend __nosavedata = 0;
+
 #ifdef CONFIG_HIGHMEM
 unsigned int count_highmem_pages(void);
 int save_highmem(void);
@@ -87,471 +72,97 @@ static int restore_highmem(void) { return 0; }
 static unsigned int count_highmem_pages(void) { return 0; }
 #endif
 
-extern char resume_file[];
-
-#define SWSUSP_SIG	"S1SUSPEND"
-
-static struct swsusp_header {
-	char reserved[PAGE_SIZE - 20 - sizeof(swp_entry_t)];
-	swp_entry_t image;
-	char	orig_sig[10];
-	char	sig[10];
-} __attribute__((packed, aligned(PAGE_SIZE))) swsusp_header;
-
-static struct swsusp_info swsusp_info;
-
-/*
- * Saving part...
- */
-
-static unsigned short root_swap = 0xffff;
-
-static int mark_swapfiles(swp_entry_t start)
-{
-	int error;
-
-	rw_swap_page_sync(READ,
-			  swp_entry(root_swap, 0),
-			  virt_to_page((unsigned long)&swsusp_header));
-	if (!memcmp("SWAP-SPACE",swsusp_header.sig, 10) ||
-	    !memcmp("SWAPSPACE2",swsusp_header.sig, 10)) {
-		memcpy(swsusp_header.orig_sig,swsusp_header.sig, 10);
-		memcpy(swsusp_header.sig,SWSUSP_SIG, 10);
-		swsusp_header.image = start;
-		error = rw_swap_page_sync(WRITE,
-					  swp_entry(root_swap, 0),
-					  virt_to_page((unsigned long)
-						       &swsusp_header));
-	} else {
-		pr_debug("swsusp: Partition is not swap space.\n");
-		error = -ENODEV;
-	}
-	return error;
-}
-
-/*
- * Check whether the swap device is the specified resume
- * device, irrespective of whether they are specified by
- * identical names.
- *
- * (Thus, device inode aliasing is allowed.  You can say /dev/hda4
- * instead of /dev/ide/host0/bus0/target0/lun0/part4 [if using devfs]
- * and they'll be considered the same device.  This is *necessary* for
- * devfs, since the resume code can only recognize the form /dev/hda4,
- * but the suspend code would see the long name.)
- */
-static inline int is_resume_device(const struct swap_info_struct *swap_info)
-{
-	struct file *file = swap_info->swap_file;
-	struct inode *inode = file->f_dentry->d_inode;
-
-	return S_ISBLK(inode->i_mode) &&
-		swsusp_resume_device == MKDEV(imajor(inode), iminor(inode));
-}
-
-static int swsusp_swap_check(void) /* This is called before saving image */
-{
-	int i;
-
-	spin_lock(&swap_lock);
-	for (i = 0; i < MAX_SWAPFILES; i++) {
-		if (!(swap_info[i].flags & SWP_WRITEOK))
-			continue;
-		if (!swsusp_resume_device || is_resume_device(swap_info + i)) {
-			spin_unlock(&swap_lock);
-			root_swap = i;
-			return 0;
-		}
-	}
-	spin_unlock(&swap_lock);
-	return -ENODEV;
-}
-
-/**
- *	write_page - Write one page to a fresh swap location.
- *	@addr:	Address we're writing.
- *	@loc:	Place to store the entry we used.
- *
- *	Allocate a new swap entry and 'sync' it. Note we discard -EIO
- *	errors. That is an artifact left over from swsusp. It did not
- *	check the return of rw_swap_page_sync() at all, since most pages
- *	written back to swap would return -EIO.
- *	This is a partial improvement, since we will at least return other
- *	errors, though we need to eventually fix the damn code.
- */
-static int write_page(unsigned long addr, swp_entry_t *loc)
-{
-	swp_entry_t entry;
-	int error = -ENOSPC;
-
-	entry = get_swap_page_of_type(root_swap);
-	if (swp_offset(entry)) {
-		error = rw_swap_page_sync(WRITE, entry, virt_to_page(addr));
-		if (!error || error == -EIO)
-			*loc = entry;
-	}
-	return error;
-}
-
 /**
- *	Swap map-handling functions
- *
- *	The swap map is a data structure used for keeping track of each page
- *	written to the swap.  It consists of many swap_map_page structures
- *	that contain each an array of MAP_PAGE_SIZE swap entries.
- *	These structures are linked together with the help of either the
- *	.next (in memory) or the .next_swap (in swap) member.
+ *	The following functions are used for tracing the allocated
+ *	swap pages, so that they can be freed in case of an error.
  *
- *	The swap map is created during suspend.  At that time we need to keep
- *	it in memory, because we have to free all of the allocated swap
- *	entries if an error occurs.  The memory needed is preallocated
- *	so that we know in advance if there's enough of it.
- *
- *	The first swap_map_page structure is filled with the swap entries that
- *	correspond to the first MAP_PAGE_SIZE data pages written to swap and
- *	so on.  After the all of the data pages have been written, the order
- *	of the swap_map_page structures in the map is reversed so that they
- *	can be read from swap in the original order.  This causes the data
- *	pages to be loaded in exactly the same order in which they have been
- *	saved.
- *
- *	During resume we only need to use one swap_map_page structure
- *	at a time, which means that we only need to use two memory pages for
- *	reading the image - one for reading the swap_map_page structures
- *	and the second for reading the data pages from swap.
+ *	The functions operate on a linked bitmap structure defined
+ *	in power.h
  */
 
-#define MAP_PAGE_SIZE	((PAGE_SIZE - sizeof(swp_entry_t) - sizeof(void *)) \
-			/ sizeof(swp_entry_t))
-
-struct swap_map_page {
-	swp_entry_t		entries[MAP_PAGE_SIZE];
-	swp_entry_t		next_swap;
-	struct swap_map_page	*next;
-};
-
-static inline void free_swap_map(struct swap_map_page *swap_map)
+void free_bitmap(struct bitmap_page *bitmap)
 {
-	struct swap_map_page *swp;
+	struct bitmap_page *bp;
 
-	while (swap_map) {
-		swp = swap_map->next;
-		free_page((unsigned long)swap_map);
-		swap_map = swp;
+	while (bitmap) {
+		bp = bitmap->next;
+		free_page((unsigned long)bitmap);
+		bitmap = bp;
 	}
 }
 
-static struct swap_map_page *alloc_swap_map(unsigned int nr_pages)
+struct bitmap_page *alloc_bitmap(unsigned int nr_bits)
 {
-	struct swap_map_page *swap_map, *swp;
-	unsigned n = 0;
+	struct bitmap_page *bitmap, *bp;
+	unsigned int n;
 
-	if (!nr_pages)
+	if (!nr_bits)
 		return NULL;
 
-	pr_debug("alloc_swap_map(): nr_pages = %d\n", nr_pages);
-	swap_map = (struct swap_map_page *)get_zeroed_page(GFP_ATOMIC);
-	swp = swap_map;
-	for (n = MAP_PAGE_SIZE; n < nr_pages; n += MAP_PAGE_SIZE) {
-		swp->next = (struct swap_map_page *)get_zeroed_page(GFP_ATOMIC);
-		swp = swp->next;
-		if (!swp) {
-			free_swap_map(swap_map);
+	bitmap = (struct bitmap_page *)get_zeroed_page(GFP_KERNEL);
+	bp = bitmap;
+	for (n = BITMAP_PAGE_BITS; n < nr_bits; n += BITMAP_PAGE_BITS) {
+		bp->next = (struct bitmap_page *)get_zeroed_page(GFP_KERNEL);
+		bp = bp->next;
+		if (!bp) {
+			free_bitmap(bitmap);
 			return NULL;
 		}
 	}
-	return swap_map;
+	return bitmap;
 }
 
-/**
- *	reverse_swap_map - reverse the order of pages in the swap map
- *	@swap_map
- */
-
-static inline struct swap_map_page *reverse_swap_map(struct swap_map_page *swap_map)
-{
-	struct swap_map_page *prev, *next;
-
-	prev = NULL;
-	while (swap_map) {
-		next = swap_map->next;
-		swap_map->next = prev;
-		prev = swap_map;
-		swap_map = next;
-	}
-	return prev;
-}
-
-/**
- *	free_swap_map_entries - free the swap entries allocated to store
- *	the swap map @swap_map (this is only called in case of an error)
- */
-static inline void free_swap_map_entries(struct swap_map_page *swap_map)
-{
-	while (swap_map) {
-		if (swap_map->next_swap.val)
-			swap_free(swap_map->next_swap);
-		swap_map = swap_map->next;
-	}
-}
-
-/**
- *	save_swap_map - save the swap map used for tracing the data pages
- *	stored in the swap
- */
-
-static int save_swap_map(struct swap_map_page *swap_map, swp_entry_t *start)
-{
-	swp_entry_t entry = (swp_entry_t){0};
-	int error;
-
-	while (swap_map) {
-		swap_map->next_swap = entry;
-		if ((error = write_page((unsigned long)swap_map, &entry)))
-			return error;
-		swap_map = swap_map->next;
-	}
-	*start = entry;
-	return 0;
-}
-
-/**
- *	free_image_entries - free the swap entries allocated to store
- *	the image data pages (this is only called in case of an error)
- */
-
-static inline void free_image_entries(struct swap_map_page *swp)
+static int bitmap_set(struct bitmap_page *bitmap, unsigned long bit)
 {
-	unsigned k;
+	unsigned int n;
 
-	while (swp) {
-		for (k = 0; k < MAP_PAGE_SIZE; k++)
-			if (swp->entries[k].val)
-				swap_free(swp->entries[k]);
-		swp = swp->next;
+	n = BITMAP_PAGE_BITS;
+	while (bitmap && n <= bit) {
+		n += BITMAP_PAGE_BITS;
+		bitmap = bitmap->next;
 	}
-}
-
-/**
- *	The swap_map_handle structure is used for handling the swap map in
- *	a file-alike way
- */
-
-struct swap_map_handle {
-	struct swap_map_page *cur;
-	unsigned int k;
-};
-
-static inline void init_swap_map_handle(struct swap_map_handle *handle,
-                                        struct swap_map_page *map)
-{
-	handle->cur = map;
-	handle->k = 0;
-}
-
-static inline int swap_map_write_page(struct swap_map_handle *handle,
-                                      unsigned long addr)
-{
-	int error;
-
-	error = write_page(addr, handle->cur->entries + handle->k);
-	if (error)
-		return error;
-	if (++handle->k >= MAP_PAGE_SIZE) {
-		handle->cur = handle->cur->next;
-		handle->k = 0;
+	if (!bitmap)
+		return -EINVAL;
+	n -= BITMAP_PAGE_BITS;
+	bit -= n;
+	n = 0;
+	while (bit >= BITS_PER_CHUNK) {
+		bit -= BITS_PER_CHUNK;
+		n++;
 	}
+	bitmap->chunks[n] |= (1UL << bit);
 	return 0;
 }
 
-/**
- *	save_image_data - save the data pages pointed to by the PBEs
- *	from the list @pblist using the swap map handle @handle
- *	(assume there are @nr_pages data pages to save)
- */
-
-static int save_image_data(struct pbe *pblist,
-                           struct swap_map_handle *handle,
-                           unsigned int nr_pages)
-{
-	unsigned int m;
-	struct pbe *p;
-	int error = 0;
-
-	printk("Saving image data pages (%u pages) ...     ", nr_pages);
-	m = nr_pages / 100;
-	if (!m)
-		m = 1;
-	nr_pages = 0;
-	for_each_pbe (p, pblist) {
-		error = swap_map_write_page(handle, p->address);
-		if (error)
-			break;
-		if (!(nr_pages % m))
-			printk("\b\b\b\b%3d%%", nr_pages / m);
-		nr_pages++;
-	}
-	if (!error)
-		printk("\b\b\b\bdone\n");
-	return error;
-}
-
-static void dump_info(void)
-{
-	pr_debug(" swsusp: Version: %u\n",swsusp_info.version_code);
-	pr_debug(" swsusp: Num Pages: %ld\n",swsusp_info.num_physpages);
-	pr_debug(" swsusp: UTS Sys: %s\n",swsusp_info.uts.sysname);
-	pr_debug(" swsusp: UTS Node: %s\n",swsusp_info.uts.nodename);
-	pr_debug(" swsusp: UTS Release: %s\n",swsusp_info.uts.release);
-	pr_debug(" swsusp: UTS Version: %s\n",swsusp_info.uts.version);
-	pr_debug(" swsusp: UTS Machine: %s\n",swsusp_info.uts.machine);
-	pr_debug(" swsusp: UTS Domain: %s\n",swsusp_info.uts.domainname);
-	pr_debug(" swsusp: CPUs: %d\n",swsusp_info.cpus);
-	pr_debug(" swsusp: Image: %ld Pages\n",swsusp_info.image_pages);
-	pr_debug(" swsusp: Total: %ld Pages\n", swsusp_info.pages);
-}
-
-static void init_header(unsigned int nr_pages)
-{
-	memset(&swsusp_info, 0, sizeof(swsusp_info));
-	swsusp_info.version_code = LINUX_VERSION_CODE;
-	swsusp_info.num_physpages = num_physpages;
-	memcpy(&swsusp_info.uts, &system_utsname, sizeof(system_utsname));
-
-	swsusp_info.cpus = num_online_cpus();
-	swsusp_info.image_pages = nr_pages;
-	swsusp_info.pages = nr_pages +
-		((nr_pages * sizeof(long) + PAGE_SIZE - 1) >> PAGE_SHIFT) + 1;
-}
-
-/**
- *	pack_orig_addresses - the .orig_address fields of the PBEs from the
- *	list starting at @pbe are stored in the array @buf[] (1 page)
- */
-
-static inline struct pbe *pack_orig_addresses(unsigned long *buf,
-                                              struct pbe *pbe)
-{
-	int j;
-
-	for (j = 0; j < PAGE_SIZE / sizeof(long) && pbe; j++) {
-		buf[j] = pbe->orig_address;
-		pbe = pbe->next;
-	}
-	if (!pbe)
-		for (; j < PAGE_SIZE / sizeof(long); j++)
-			buf[j] = 0;
-	return pbe;
-}
-
-/**
- *	save_image_metadata - save the .orig_address fields of the PBEs
- *	from the list @pblist using the swap map handle @handle
- */
-
-static int save_image_metadata(struct pbe *pblist,
-                               struct swap_map_handle *handle)
+unsigned long alloc_swap_page(int swap, struct bitmap_page *bitmap)
 {
-	unsigned long *buf;
-	unsigned int n = 0;
-	struct pbe *p;
-	int error = 0;
+	unsigned long offset;
 
-	printk("Saving image metadata ... ");
-	buf = (unsigned long *)get_zeroed_page(GFP_ATOMIC);
-	if (!buf)
-		return -ENOMEM;
-	p = pblist;
-	while (p) {
-		p = pack_orig_addresses(buf, p);
-		error = swap_map_write_page(handle, (unsigned long)buf);
-		if (error)
-			break;
-		n++;
+	offset = swp_offset(get_swap_page_of_type(swap));
+	if (offset) {
+		if (bitmap_set(bitmap, offset)) {
+			swap_free(swp_entry(swap, offset));
+			offset = 0;
+		}
 	}
-	free_page((unsigned long)buf);
-	if (!error)
-		printk("done (%u pages saved)\n", n);
-	return error;
+	return offset;
 }
 
-/**
- *	enough_swap - Make sure we have enough swap to save the image.
- *
- *	Returns TRUE or FALSE after checking the total amount of swap
- *	space avaiable from the resume partition.
- */
-
-static int enough_swap(unsigned int nr_pages)
+void free_all_swap_pages(int swap, struct bitmap_page *bitmap)
 {
-	unsigned int free_swap = swap_info[root_swap].pages -
-		swap_info[root_swap].inuse_pages;
-
-	pr_debug("swsusp: free swap pages: %u\n", free_swap);
-	return free_swap > (nr_pages + PAGES_FOR_IO +
-		(nr_pages + PBES_PER_PAGE - 1) / PBES_PER_PAGE);
-}
+	unsigned int bit, n;
+	unsigned long test;
 
-/**
- *	swsusp_write - Write entire image and metadata.
- *
- *	It is important _NOT_ to umount filesystems at this point. We want
- *	them synced (in case something goes wrong) but we DO not want to mark
- *	filesystem clean: it is not. (And it does not matter, if we resume
- *	correctly, we'll mark system clean, anyway.)
- */
-
-int swsusp_write(struct pbe *pblist, unsigned int nr_pages)
-{
-	struct swap_map_page *swap_map;
-	struct swap_map_handle handle;
-	swp_entry_t start;
-	int error;
-
-	if ((error = swsusp_swap_check())) {
-		printk(KERN_ERR "swsusp: Cannot find swap device, try swapon -a.\n");
-		return error;
-	}
-	if (!enough_swap(nr_pages)) {
-		printk(KERN_ERR "swsusp: Not enough free swap\n");
-		return -ENOSPC;
+	bit = 0;
+	while (bitmap) {
+		for (n = 0; n < BITMAP_PAGE_CHUNKS; n++)
+			for (test = 1UL; test; test <<= 1) {
+				if (bitmap->chunks[n] & test)
+					swap_free(swp_entry(swap, bit));
+				bit++;
+			}
+		bitmap = bitmap->next;
 	}
-
-	init_header(nr_pages);
-	swap_map = alloc_swap_map(swsusp_info.pages);
-	if (!swap_map)
-		return -ENOMEM;
-	init_swap_map_handle(&handle, swap_map);
-
-	error = swap_map_write_page(&handle, (unsigned long)&swsusp_info);
-	if (!error)
-		error = save_image_metadata(pblist, &handle);
-	if (!error)
-		error = save_image_data(pblist, &handle, nr_pages);
-	if (error)
-		goto Free_image_entries;
-
-	swap_map = reverse_swap_map(swap_map);
-	error = save_swap_map(swap_map, &start);
-	if (error)
-		goto Free_map_entries;
-
-	dump_info();
-	printk( "S" );
-	error = mark_swapfiles(start);
-	printk( "|\n" );
-	if (error)
-		goto Free_map_entries;
-
-Free_swap_map:
-	free_swap_map(swap_map);
-	return error;
-
-Free_map_entries:
-	free_swap_map_entries(swap_map);
-Free_image_entries:
-	free_image_entries(swap_map);
-	goto Free_swap_map;
 }
 
 /**
@@ -660,379 +271,3 @@ int swsusp_resume(void)
 	local_irq_enable();
 	return error;
 }
-
-/**
- *	mark_unsafe_pages - mark the pages that cannot be used for storing
- *	the image during resume, because they conflict with the pages that
- *	had been used before suspend
- */
-
-static void mark_unsafe_pages(struct pbe *pblist)
-{
-	struct zone *zone;
-	unsigned long zone_pfn;
-	struct pbe *p;
-
-	if (!pblist) /* a sanity check */
-		return;
-
-	/* Clear page flags */
-	for_each_zone (zone) {
-		for (zone_pfn = 0; zone_pfn < zone->spanned_pages; ++zone_pfn)
-			if (pfn_valid(zone_pfn + zone->zone_start_pfn))
-				ClearPageNosaveFree(pfn_to_page(zone_pfn +
-					zone->zone_start_pfn));
-	}
-
-	/* Mark orig addresses */
-	for_each_pbe (p, pblist)
-		SetPageNosaveFree(virt_to_page(p->orig_address));
-
-}
-
-static void copy_page_backup_list(struct pbe *dst, struct pbe *src)
-{
-	/* We assume both lists contain the same number of elements */
-	while (src) {
-		dst->orig_address = src->orig_address;
-		dst = dst->next;
-		src = src->next;
-	}
-}
-
-/*
- *	Using bio to read from swap.
- *	This code requires a bit more work than just using buffer heads
- *	but, it is the recommended way for 2.5/2.6.
- *	The following are to signal the beginning and end of I/O. Bios
- *	finish asynchronously, while we want them to happen synchronously.
- *	A simple atomic_t, and a wait loop take care of this problem.
- */
-
-static atomic_t io_done = ATOMIC_INIT(0);
-
-static int end_io(struct bio *bio, unsigned int num, int err)
-{
-	if (!test_bit(BIO_UPTODATE, &bio->bi_flags))
-		panic("I/O error reading memory image");
-	atomic_set(&io_done, 0);
-	return 0;
-}
-
-static struct block_device *resume_bdev;
-
-/**
- *	submit - submit BIO request.
- *	@rw:	READ or WRITE.
- *	@off	physical offset of page.
- *	@page:	page we're reading or writing.
- *
- *	Straight from the textbook - allocate and initialize the bio.
- *	If we're writing, make sure the page is marked as dirty.
- *	Then submit it and wait.
- */
-
-static int submit(int rw, pgoff_t page_off, void *page)
-{
-	int error = 0;
-	struct bio *bio;
-
-	bio = bio_alloc(GFP_ATOMIC, 1);
-	if (!bio)
-		return -ENOMEM;
-	bio->bi_sector = page_off * (PAGE_SIZE >> 9);
-	bio->bi_bdev = resume_bdev;
-	bio->bi_end_io = end_io;
-
-	if (bio_add_page(bio, virt_to_page(page), PAGE_SIZE, 0) < PAGE_SIZE) {
-		printk("swsusp: ERROR: adding page to bio at %ld\n",page_off);
-		error = -EFAULT;
-		goto Done;
-	}
-
-
-	atomic_set(&io_done, 1);
-	submit_bio(rw | (1 << BIO_RW_SYNC), bio);
-	while (atomic_read(&io_done))
-		yield();
-	if (rw == READ)
-		bio_set_pages_dirty(bio);
- Done:
-	bio_put(bio);
-	return error;
-}
-
-static int bio_read_page(pgoff_t page_off, void *page)
-{
-	return submit(READ, page_off, page);
-}
-
-static int bio_write_page(pgoff_t page_off, void *page)
-{
-	return submit(WRITE, page_off, page);
-}
-
-/**
- *	The following functions allow us to read data using a swap map
- *	in a file-alike way
- */
-
-static inline void release_swap_map_reader(struct swap_map_handle *handle)
-{
-	if (handle->cur)
-		free_page((unsigned long)handle->cur);
-	handle->cur = NULL;
-}
-
-static inline int get_swap_map_reader(struct swap_map_handle *handle,
-                                      swp_entry_t start)
-{
-	int error;
-
-	if (!swp_offset(start))
-		return -EINVAL;
-	handle->cur = (struct swap_map_page *)get_zeroed_page(GFP_ATOMIC);
-	if (!handle->cur)
-		return -ENOMEM;
-	error = bio_read_page(swp_offset(start), handle->cur);
-	if (error) {
-		release_swap_map_reader(handle);
-		return error;
-	}
-	handle->k = 0;
-	return 0;
-}
-
-static inline int swap_map_read_page(struct swap_map_handle *handle, void *buf)
-{
-	unsigned long offset;
-	int error;
-
-	if (!handle->cur)
-		return -EINVAL;
-	offset = swp_offset(handle->cur->entries[handle->k]);
-	if (!offset)
-		return -EINVAL;
-	error = bio_read_page(offset, buf);
-	if (error)
-		return error;
-	if (++handle->k >= MAP_PAGE_SIZE) {
-		handle->k = 0;
-		offset = swp_offset(handle->cur->next_swap);
-		if (!offset)
-			release_swap_map_reader(handle);
-		else
-			error = bio_read_page(offset, handle->cur);
-	}
-	return error;
-}
-
-static int check_header(void)
-{
-	char *reason = NULL;
-
-	dump_info();
-	if (swsusp_info.version_code != LINUX_VERSION_CODE)
-		reason = "kernel version";
-	if (swsusp_info.num_physpages != num_physpages)
-		reason = "memory size";
-	if (strcmp(swsusp_info.uts.sysname,system_utsname.sysname))
-		reason = "system type";
-	if (strcmp(swsusp_info.uts.release,system_utsname.release))
-		reason = "kernel release";
-	if (strcmp(swsusp_info.uts.version,system_utsname.version))
-		reason = "version";
-	if (strcmp(swsusp_info.uts.machine,system_utsname.machine))
-		reason = "machine";
-	if (reason) {
-		printk(KERN_ERR "swsusp: Resume mismatch: %s\n", reason);
-		return -EPERM;
-	}
-	return 0;
-}
-
-/**
- *	load_image_data - load the image data using the swap map handle
- *	@handle and store them using the page backup list @pblist
- *	(assume there are @nr_pages pages to load)
- */
-
-static int load_image_data(struct pbe *pblist,
-                           struct swap_map_handle *handle,
-                           unsigned int nr_pages)
-{
-	int error;
-	unsigned int m;
-	struct pbe *p;
-
-	if (!pblist)
-		return -EINVAL;
-	printk("Loading image data pages (%u pages) ...     ", nr_pages);
-	m = nr_pages / 100;
-	if (!m)
-		m = 1;
-	nr_pages = 0;
-	p = pblist;
-	while (p) {
-		error = swap_map_read_page(handle, (void *)p->address);
-		if (error)
-			break;
-		p = p->next;
-		if (!(nr_pages % m))
-			printk("\b\b\b\b%3d%%", nr_pages / m);
-		nr_pages++;
-	}
-	if (!error)
-		printk("\b\b\b\bdone\n");
-	return error;
-}
-
-/**
- *	unpack_orig_addresses - copy the elements of @buf[] (1 page) to
- *	the PBEs in the list starting at @pbe
- */
-
-static inline struct pbe *unpack_orig_addresses(unsigned long *buf,
-                                                struct pbe *pbe)
-{
-	int j;
-
-	for (j = 0; j < PAGE_SIZE / sizeof(long) && pbe; j++) {
-		pbe->orig_address = buf[j];
-		pbe = pbe->next;
-	}
-	return pbe;
-}
-
-/**
- *	load_image_metadata - load the image metadata using the swap map
- *	handle @handle and put them into the PBEs in the list @pblist
- */
-
-static int load_image_metadata(struct pbe *pblist, struct swap_map_handle *handle)
-{
-	struct pbe *p;
-	unsigned long *buf;
-	unsigned int n = 0;
-	int error = 0;
-
-	printk("Loading image metadata ... ");
-	buf = (unsigned long *)get_zeroed_page(GFP_ATOMIC);
-	if (!buf)
-		return -ENOMEM;
-	p = pblist;
-	while (p) {
-		error = swap_map_read_page(handle, buf);
-		if (error)
-			break;
-		p = unpack_orig_addresses(buf, p);
-		n++;
-	}
-	free_page((unsigned long)buf);
-	if (!error)
-		printk("done (%u pages loaded)\n", n);
-	return error;
-}
-
-int swsusp_read(struct pbe **pblist_ptr)
-{
-	int error;
-	struct pbe *p, *pblist;
-	struct swap_map_handle handle;
-	unsigned int nr_pages;
-
-	if (IS_ERR(resume_bdev)) {
-		pr_debug("swsusp: block device not initialised\n");
-		return PTR_ERR(resume_bdev);
-	}
-
-	error = get_swap_map_reader(&handle, swsusp_header.image);
-	if (!error)
-		error = swap_map_read_page(&handle, &swsusp_info);
-	if (!error)
-		error = check_header();
-	if (error)
-		return error;
-	nr_pages = swsusp_info.image_pages;
-	p = alloc_pagedir(nr_pages, GFP_ATOMIC, 0);
-	if (!p)
-		return -ENOMEM;
-	error = load_image_metadata(p, &handle);
-	if (!error) {
-		mark_unsafe_pages(p);
-		pblist = alloc_pagedir(nr_pages, GFP_ATOMIC, 1);
-		if (pblist)
-			copy_page_backup_list(pblist, p);
-		free_pagedir(p);
-		if (!pblist)
-			error = -ENOMEM;
-
-		/* Allocate memory for the image and read the data from swap */
-		if (!error)
-			error = alloc_data_pages(pblist, GFP_ATOMIC, 1);
-		if (!error) {
-			release_eaten_pages();
-			error = load_image_data(pblist, &handle, nr_pages);
-		}
-		if (!error)
-			*pblist_ptr = pblist;
-	}
-	release_swap_map_reader(&handle);
-
-	blkdev_put(resume_bdev);
-
-	if (!error)
-		pr_debug("swsusp: Reading resume file was successful\n");
-	else
-		pr_debug("swsusp: Error %d resuming\n", error);
-	return error;
-}
-
-/**
- *      swsusp_check - Check for swsusp signature in the resume device
- */
-
-int swsusp_check(void)
-{
-	int error;
-
-	resume_bdev = open_by_devnum(swsusp_resume_device, FMODE_READ);
-	if (!IS_ERR(resume_bdev)) {
-		set_blocksize(resume_bdev, PAGE_SIZE);
-		memset(&swsusp_header, 0, sizeof(swsusp_header));
-		if ((error = bio_read_page(0, &swsusp_header)))
-			return error;
-		if (!memcmp(SWSUSP_SIG, swsusp_header.sig, 10)) {
-			memcpy(swsusp_header.sig, swsusp_header.orig_sig, 10);
-			/* Reset swap signature now */
-			error = bio_write_page(0, &swsusp_header);
-		} else {
-			return -EINVAL;
-		}
-		if (error)
-			blkdev_put(resume_bdev);
-		else
-			pr_debug("swsusp: Signature found, resuming\n");
-	} else {
-		error = PTR_ERR(resume_bdev);
-	}
-
-	if (error)
-		pr_debug("swsusp: Error %d check for resume file\n", error);
-
-	return error;
-}
-
-/**
- *	swsusp_close - close swap device.
- */
-
-void swsusp_close(void)
-{
-	if (IS_ERR(resume_bdev)) {
-		pr_debug("swsusp: block device not initialised\n");
-		return;
-	}
-
-	blkdev_put(resume_bdev);
-}
diff --git a/kernel/power/user.c b/kernel/power/user.c
new file mode 100644
index 00000000000..3f1539fbe48
--- /dev/null
+++ b/kernel/power/user.c
@@ -0,0 +1,333 @@
+/*
+ * linux/kernel/power/user.c
+ *
+ * This file provides the user space interface for software suspend/resume.
+ *
+ * Copyright (C) 2006 Rafael J. Wysocki <rjw@sisk.pl>
+ *
+ * This file is released under the GPLv2.
+ *
+ */
+
+#include <linux/suspend.h>
+#include <linux/syscalls.h>
+#include <linux/string.h>
+#include <linux/device.h>
+#include <linux/miscdevice.h>
+#include <linux/mm.h>
+#include <linux/swap.h>
+#include <linux/swapops.h>
+#include <linux/pm.h>
+#include <linux/fs.h>
+
+#include <asm/uaccess.h>
+
+#include "power.h"
+
+#define SNAPSHOT_MINOR	231
+
+static struct snapshot_data {
+	struct snapshot_handle handle;
+	int swap;
+	struct bitmap_page *bitmap;
+	int mode;
+	char frozen;
+	char ready;
+} snapshot_state;
+
+static atomic_t device_available = ATOMIC_INIT(1);
+
+static int snapshot_open(struct inode *inode, struct file *filp)
+{
+	struct snapshot_data *data;
+
+	if (!atomic_add_unless(&device_available, -1, 0))
+		return -EBUSY;
+
+	if ((filp->f_flags & O_ACCMODE) == O_RDWR)
+		return -ENOSYS;
+
+	nonseekable_open(inode, filp);
+	data = &snapshot_state;
+	filp->private_data = data;
+	memset(&data->handle, 0, sizeof(struct snapshot_handle));
+	if ((filp->f_flags & O_ACCMODE) == O_RDONLY) {
+		data->swap = swsusp_resume_device ? swap_type_of(swsusp_resume_device) : -1;
+		data->mode = O_RDONLY;
+	} else {
+		data->swap = -1;
+		data->mode = O_WRONLY;
+	}
+	data->bitmap = NULL;
+	data->frozen = 0;
+	data->ready = 0;
+
+	return 0;
+}
+
+static int snapshot_release(struct inode *inode, struct file *filp)
+{
+	struct snapshot_data *data;
+
+	swsusp_free();
+	data = filp->private_data;
+	free_all_swap_pages(data->swap, data->bitmap);
+	free_bitmap(data->bitmap);
+	if (data->frozen) {
+		down(&pm_sem);
+		thaw_processes();
+		enable_nonboot_cpus();
+		up(&pm_sem);
+	}
+	atomic_inc(&device_available);
+	return 0;
+}
+
+static ssize_t snapshot_read(struct file *filp, char __user *buf,
+                             size_t count, loff_t *offp)
+{
+	struct snapshot_data *data;
+	ssize_t res;
+
+	data = filp->private_data;
+	res = snapshot_read_next(&data->handle, count);
+	if (res > 0) {
+		if (copy_to_user(buf, data_of(data->handle), res))
+			res = -EFAULT;
+		else
+			*offp = data->handle.offset;
+	}
+	return res;
+}
+
+static ssize_t snapshot_write(struct file *filp, const char __user *buf,
+                              size_t count, loff_t *offp)
+{
+	struct snapshot_data *data;
+	ssize_t res;
+
+	data = filp->private_data;
+	res = snapshot_write_next(&data->handle, count);
+	if (res > 0) {
+		if (copy_from_user(data_of(data->handle), buf, res))
+			res = -EFAULT;
+		else
+			*offp = data->handle.offset;
+	}
+	return res;
+}
+
+static int snapshot_ioctl(struct inode *inode, struct file *filp,
+                          unsigned int cmd, unsigned long arg)
+{
+	int error = 0;
+	struct snapshot_data *data;
+	loff_t offset, avail;
+
+	if (_IOC_TYPE(cmd) != SNAPSHOT_IOC_MAGIC)
+		return -ENOTTY;
+	if (_IOC_NR(cmd) > SNAPSHOT_IOC_MAXNR)
+		return -ENOTTY;
+	if (!capable(CAP_SYS_ADMIN))
+		return -EPERM;
+
+	data = filp->private_data;
+
+	switch (cmd) {
+
+	case SNAPSHOT_FREEZE:
+		if (data->frozen)
+			break;
+		down(&pm_sem);
+		disable_nonboot_cpus();
+		if (freeze_processes()) {
+			thaw_processes();
+			enable_nonboot_cpus();
+			error = -EBUSY;
+		}
+		up(&pm_sem);
+		if (!error)
+			data->frozen = 1;
+		break;
+
+	case SNAPSHOT_UNFREEZE:
+		if (!data->frozen)
+			break;
+		down(&pm_sem);
+		thaw_processes();
+		enable_nonboot_cpus();
+		up(&pm_sem);
+		data->frozen = 0;
+		break;
+
+	case SNAPSHOT_ATOMIC_SNAPSHOT:
+		if (data->mode != O_RDONLY || !data->frozen  || data->ready) {
+			error = -EPERM;
+			break;
+		}
+		down(&pm_sem);
+		/* Free memory before shutting down devices. */
+		error = swsusp_shrink_memory();
+		if (!error) {
+			error = device_suspend(PMSG_FREEZE);
+			if (!error) {
+				in_suspend = 1;
+				error = swsusp_suspend();
+				device_resume();
+			}
+		}
+		up(&pm_sem);
+		if (!error)
+			error = put_user(in_suspend, (unsigned int __user *)arg);
+		if (!error)
+			data->ready = 1;
+		break;
+
+	case SNAPSHOT_ATOMIC_RESTORE:
+		if (data->mode != O_WRONLY || !data->frozen ||
+		    !snapshot_image_loaded(&data->handle)) {
+			error = -EPERM;
+			break;
+		}
+		down(&pm_sem);
+		pm_prepare_console();
+		error = device_suspend(PMSG_FREEZE);
+		if (!error) {
+			error = swsusp_resume();
+			device_resume();
+		}
+		pm_restore_console();
+		up(&pm_sem);
+		break;
+
+	case SNAPSHOT_FREE:
+		swsusp_free();
+		memset(&data->handle, 0, sizeof(struct snapshot_handle));
+		data->ready = 0;
+		break;
+
+	case SNAPSHOT_SET_IMAGE_SIZE:
+		image_size = arg;
+		break;
+
+	case SNAPSHOT_AVAIL_SWAP:
+		avail = count_swap_pages(data->swap, 1);
+		avail <<= PAGE_SHIFT;
+		error = put_user(avail, (loff_t __user *)arg);
+		break;
+
+	case SNAPSHOT_GET_SWAP_PAGE:
+		if (data->swap < 0 || data->swap >= MAX_SWAPFILES) {
+			error = -ENODEV;
+			break;
+		}
+		if (!data->bitmap) {
+			data->bitmap = alloc_bitmap(count_swap_pages(data->swap, 0));
+			if (!data->bitmap) {
+				error = -ENOMEM;
+				break;
+			}
+		}
+		offset = alloc_swap_page(data->swap, data->bitmap);
+		if (offset) {
+			offset <<= PAGE_SHIFT;
+			error = put_user(offset, (loff_t __user *)arg);
+		} else {
+			error = -ENOSPC;
+		}
+		break;
+
+	case SNAPSHOT_FREE_SWAP_PAGES:
+		if (data->swap < 0 || data->swap >= MAX_SWAPFILES) {
+			error = -ENODEV;
+			break;
+		}
+		free_all_swap_pages(data->swap, data->bitmap);
+		free_bitmap(data->bitmap);
+		data->bitmap = NULL;
+		break;
+
+	case SNAPSHOT_SET_SWAP_FILE:
+		if (!data->bitmap) {
+			/*
+			 * User space encodes device types as two-byte values,
+			 * so we need to recode them
+			 */
+			if (old_decode_dev(arg)) {
+				data->swap = swap_type_of(old_decode_dev(arg));
+				if (data->swap < 0)
+					error = -ENODEV;
+			} else {
+				data->swap = -1;
+				error = -EINVAL;
+			}
+		} else {
+			error = -EPERM;
+		}
+		break;
+
+	case SNAPSHOT_S2RAM:
+		if (!data->frozen) {
+			error = -EPERM;
+			break;
+		}
+
+		if (down_trylock(&pm_sem)) {
+			error = -EBUSY;
+			break;
+		}
+
+		if (pm_ops->prepare) {
+			error = pm_ops->prepare(PM_SUSPEND_MEM);
+			if (error)
+				goto OutS3;
+		}
+
+		/* Put devices to sleep */
+		error = device_suspend(PMSG_SUSPEND);
+		if (error) {
+			printk(KERN_ERR "Failed to suspend some devices.\n");
+		} else {
+			/* Enter S3, system is already frozen */
+			suspend_enter(PM_SUSPEND_MEM);
+
+			/* Wake up devices */
+			device_resume();
+		}
+
+		if (pm_ops->finish)
+			pm_ops->finish(PM_SUSPEND_MEM);
+
+OutS3:
+		up(&pm_sem);
+		break;
+
+	default:
+		error = -ENOTTY;
+
+	}
+
+	return error;
+}
+
+static struct file_operations snapshot_fops = {
+	.open = snapshot_open,
+	.release = snapshot_release,
+	.read = snapshot_read,
+	.write = snapshot_write,
+	.llseek = no_llseek,
+	.ioctl = snapshot_ioctl,
+};
+
+static struct miscdevice snapshot_device = {
+	.minor = SNAPSHOT_MINOR,
+	.name = "snapshot",
+	.fops = &snapshot_fops,
+};
+
+static int __init snapshot_device_init(void)
+{
+	return misc_register(&snapshot_device);
+};
+
+device_initcall(snapshot_device_init);
diff --git a/kernel/profile.c b/kernel/profile.c
index f89248e6d70..ad81f799a9b 100644
--- a/kernel/profile.c
+++ b/kernel/profile.c
@@ -23,6 +23,7 @@
 #include <linux/cpu.h>
 #include <linux/profile.h>
 #include <linux/highmem.h>
+#include <linux/mutex.h>
 #include <asm/sections.h>
 #include <asm/semaphore.h>
 
@@ -44,7 +45,7 @@ static cpumask_t prof_cpu_mask = CPU_MASK_ALL;
 #ifdef CONFIG_SMP
 static DEFINE_PER_CPU(struct profile_hit *[2], cpu_profile_hits);
 static DEFINE_PER_CPU(int, cpu_profile_flip);
-static DECLARE_MUTEX(profile_flip_mutex);
+static DEFINE_MUTEX(profile_flip_mutex);
 #endif /* CONFIG_SMP */
 
 static int __init profile_setup(char * str)
@@ -243,7 +244,7 @@ static void profile_flip_buffers(void)
 {
 	int i, j, cpu;
 
-	down(&profile_flip_mutex);
+	mutex_lock(&profile_flip_mutex);
 	j = per_cpu(cpu_profile_flip, get_cpu());
 	put_cpu();
 	on_each_cpu(__profile_flip_buffers, NULL, 0, 1);
@@ -259,14 +260,14 @@ static void profile_flip_buffers(void)
 			hits[i].hits = hits[i].pc = 0;
 		}
 	}
-	up(&profile_flip_mutex);
+	mutex_unlock(&profile_flip_mutex);
 }
 
 static void profile_discard_flip_buffers(void)
 {
 	int i, cpu;
 
-	down(&profile_flip_mutex);
+	mutex_lock(&profile_flip_mutex);
 	i = per_cpu(cpu_profile_flip, get_cpu());
 	put_cpu();
 	on_each_cpu(__profile_flip_buffers, NULL, 0, 1);
@@ -274,7 +275,7 @@ static void profile_discard_flip_buffers(void)
 		struct profile_hit *hits = per_cpu(cpu_profile_hits, cpu)[i];
 		memset(hits, 0, NR_PROFILE_HIT*sizeof(struct profile_hit));
 	}
-	up(&profile_flip_mutex);
+	mutex_unlock(&profile_flip_mutex);
 }
 
 void profile_hit(int type, void *__pc)
diff --git a/kernel/rcupdate.c b/kernel/rcupdate.c
index fedf5e36975..6df1559b1c0 100644
--- a/kernel/rcupdate.c
+++ b/kernel/rcupdate.c
@@ -47,15 +47,16 @@
 #include <linux/notifier.h>
 #include <linux/rcupdate.h>
 #include <linux/cpu.h>
+#include <linux/mutex.h>
 
 /* Definition for rcupdate control block. */
-struct rcu_ctrlblk rcu_ctrlblk = {
+static struct rcu_ctrlblk rcu_ctrlblk = {
 	.cur = -300,
 	.completed = -300,
 	.lock = SPIN_LOCK_UNLOCKED,
 	.cpumask = CPU_MASK_NONE,
 };
-struct rcu_ctrlblk rcu_bh_ctrlblk = {
+static struct rcu_ctrlblk rcu_bh_ctrlblk = {
 	.cur = -300,
 	.completed = -300,
 	.lock = SPIN_LOCK_UNLOCKED,
@@ -75,7 +76,7 @@ static int rsinterval = 1000;
 #endif
 
 static atomic_t rcu_barrier_cpu_count;
-static struct semaphore rcu_barrier_sema;
+static DEFINE_MUTEX(rcu_barrier_mutex);
 static struct completion rcu_barrier_completion;
 
 #ifdef CONFIG_SMP
@@ -207,13 +208,13 @@ static void rcu_barrier_func(void *notused)
 void rcu_barrier(void)
 {
 	BUG_ON(in_interrupt());
-	/* Take cpucontrol semaphore to protect against CPU hotplug */
-	down(&rcu_barrier_sema);
+	/* Take cpucontrol mutex to protect against CPU hotplug */
+	mutex_lock(&rcu_barrier_mutex);
 	init_completion(&rcu_barrier_completion);
 	atomic_set(&rcu_barrier_cpu_count, 0);
 	on_each_cpu(rcu_barrier_func, NULL, 0, 1);
 	wait_for_completion(&rcu_barrier_completion);
-	up(&rcu_barrier_sema);
+	mutex_unlock(&rcu_barrier_mutex);
 }
 EXPORT_SYMBOL_GPL(rcu_barrier);
 
@@ -549,7 +550,6 @@ static struct notifier_block __devinitdata rcu_nb = {
  */
 void __init rcu_init(void)
 {
-	sema_init(&rcu_barrier_sema, 1);
 	rcu_cpu_notify(&rcu_nb, CPU_UP_PREPARE,
 			(void *)(long)smp_processor_id());
 	/* Register notifier for non-boot CPUs */
diff --git a/kernel/sched.c b/kernel/sched.c
index 6b6e0d70eb3..7ffaabd64f8 100644
--- a/kernel/sched.c
+++ b/kernel/sched.c
@@ -237,6 +237,7 @@ struct runqueue {
 
 	task_t *migration_thread;
 	struct list_head migration_queue;
+	int cpu;
 #endif
 
 #ifdef CONFIG_SCHEDSTATS
@@ -1654,6 +1655,9 @@ unsigned long nr_iowait(void)
 /*
  * double_rq_lock - safely lock two runqueues
  *
+ * We must take them in cpu order to match code in
+ * dependent_sleeper and wake_dependent_sleeper.
+ *
  * Note this does not disable interrupts like task_rq_lock,
  * you need to do so manually before calling.
  */
@@ -1665,7 +1669,7 @@ static void double_rq_lock(runqueue_t *rq1, runqueue_t *rq2)
 		spin_lock(&rq1->lock);
 		__acquire(rq2->lock);	/* Fake it out ;) */
 	} else {
-		if (rq1 < rq2) {
+		if (rq1->cpu < rq2->cpu) {
 			spin_lock(&rq1->lock);
 			spin_lock(&rq2->lock);
 		} else {
@@ -1701,7 +1705,7 @@ static void double_lock_balance(runqueue_t *this_rq, runqueue_t *busiest)
 	__acquires(this_rq->lock)
 {
 	if (unlikely(!spin_trylock(&busiest->lock))) {
-		if (busiest < this_rq) {
+		if (busiest->cpu < this_rq->cpu) {
 			spin_unlock(&this_rq->lock);
 			spin_lock(&busiest->lock);
 			spin_lock(&this_rq->lock);
@@ -2869,7 +2873,7 @@ asmlinkage void __sched schedule(void)
 	 */
 	if (likely(!current->exit_state)) {
 		if (unlikely(in_atomic())) {
-			printk(KERN_ERR "scheduling while atomic: "
+			printk(KERN_ERR "BUG: scheduling while atomic: "
 				"%s/0x%08x/%d\n",
 				current->comm, preempt_count(), current->pid);
 			dump_stack();
@@ -6029,6 +6033,7 @@ void __init sched_init(void)
 		rq->push_cpu = 0;
 		rq->migration_thread = NULL;
 		INIT_LIST_HEAD(&rq->migration_queue);
+		rq->cpu = i;
 #endif
 		atomic_set(&rq->nr_iowait, 0);
 
@@ -6069,7 +6074,7 @@ void __might_sleep(char *file, int line)
 		if (time_before(jiffies, prev_jiffy + HZ) && prev_jiffy)
 			return;
 		prev_jiffy = jiffies;
-		printk(KERN_ERR "Debug: sleeping function called from invalid"
+		printk(KERN_ERR "BUG: sleeping function called from invalid"
 				" context at %s:%d\n", file, line);
 		printk("in_atomic():%d, irqs_disabled():%d\n",
 			in_atomic(), irqs_disabled());
diff --git a/kernel/signal.c b/kernel/signal.c
index ea154104a00..75f7341b0c3 100644
--- a/kernel/signal.c
+++ b/kernel/signal.c
@@ -1922,6 +1922,8 @@ int get_signal_to_deliver(siginfo_t *info, struct k_sigaction *return_ka,
 	sigset_t *mask = &current->blocked;
 	int signr = 0;
 
+	try_to_freeze();
+
 relock:
 	spin_lock_irq(&current->sighand->siglock);
 	for (;;) {
@@ -2099,10 +2101,11 @@ long do_no_restart_syscall(struct restart_block *param)
 int sigprocmask(int how, sigset_t *set, sigset_t *oldset)
 {
 	int error;
-	sigset_t old_block;
 
 	spin_lock_irq(&current->sighand->siglock);
-	old_block = current->blocked;
+	if (oldset)
+		*oldset = current->blocked;
+
 	error = 0;
 	switch (how) {
 	case SIG_BLOCK:
@@ -2119,8 +2122,7 @@ int sigprocmask(int how, sigset_t *set, sigset_t *oldset)
 	}
 	recalc_sigpending();
 	spin_unlock_irq(&current->sighand->siglock);
-	if (oldset)
-		*oldset = old_block;
+
 	return error;
 }
 
@@ -2307,7 +2309,6 @@ sys_rt_sigtimedwait(const sigset_t __user *uthese,
 
 			timeout = schedule_timeout_interruptible(timeout);
 
-			try_to_freeze();
 			spin_lock_irq(&current->sighand->siglock);
 			sig = dequeue_signal(current, &these, &info);
 			current->blocked = current->real_blocked;
diff --git a/kernel/spinlock.c b/kernel/spinlock.c
index 0375fcd5921..d1b810782bc 100644
--- a/kernel/spinlock.c
+++ b/kernel/spinlock.c
@@ -179,16 +179,16 @@ EXPORT_SYMBOL(_write_lock);
 #define BUILD_LOCK_OPS(op, locktype)					\
 void __lockfunc _##op##_lock(locktype##_t *lock)			\
 {									\
-	preempt_disable();						\
 	for (;;) {							\
+		preempt_disable();					\
 		if (likely(_raw_##op##_trylock(lock)))			\
 			break;						\
 		preempt_enable();					\
+									\
 		if (!(lock)->break_lock)				\
 			(lock)->break_lock = 1;				\
 		while (!op##_can_lock(lock) && (lock)->break_lock)	\
 			cpu_relax();					\
-		preempt_disable();					\
 	}								\
 	(lock)->break_lock = 0;						\
 }									\
@@ -199,19 +199,18 @@ unsigned long __lockfunc _##op##_lock_irqsave(locktype##_t *lock)	\
 {									\
 	unsigned long flags;						\
 									\
-	preempt_disable();						\
 	for (;;) {							\
+		preempt_disable();					\
 		local_irq_save(flags);					\
 		if (likely(_raw_##op##_trylock(lock)))			\
 			break;						\
 		local_irq_restore(flags);				\
-									\
 		preempt_enable();					\
+									\
 		if (!(lock)->break_lock)				\
 			(lock)->break_lock = 1;				\
 		while (!op##_can_lock(lock) && (lock)->break_lock)	\
 			cpu_relax();					\
-		preempt_disable();					\
 	}								\
 	(lock)->break_lock = 0;						\
 	return flags;							\
diff --git a/kernel/sys.c b/kernel/sys.c
index f91218a5463..c0fcad9f826 100644
--- a/kernel/sys.c
+++ b/kernel/sys.c
@@ -1227,7 +1227,7 @@ asmlinkage long sys_setsid(void)
 	struct pid *pid;
 	int err = -EPERM;
 
-	down(&tty_sem);
+	mutex_lock(&tty_mutex);
 	write_lock_irq(&tasklist_lock);
 
 	pid = find_pid(PIDTYPE_PGID, group_leader->pid);
@@ -1241,7 +1241,7 @@ asmlinkage long sys_setsid(void)
 	err = process_group(group_leader);
 out:
 	write_unlock_irq(&tasklist_lock);
-	up(&tty_sem);
+	mutex_unlock(&tty_mutex);
 	return err;
 }
 
@@ -1677,9 +1677,6 @@ asmlinkage long sys_setrlimit(unsigned int resource, struct rlimit __user *rlim)
  * a lot simpler!  (Which we're not doing right now because we're not
  * measuring them yet).
  *
- * This expects to be called with tasklist_lock read-locked or better,
- * and the siglock not locked.  It may momentarily take the siglock.
- *
  * When sampling multiple threads for RUSAGE_SELF, under SMP we might have
  * races with threads incrementing their own counters.  But since word
  * reads are atomic, we either get new values or old values and we don't
@@ -1687,6 +1684,25 @@ asmlinkage long sys_setrlimit(unsigned int resource, struct rlimit __user *rlim)
  * the c* fields from p->signal from races with exit.c updating those
  * fields when reaping, so a sample either gets all the additions of a
  * given child after it's reaped, or none so this sample is before reaping.
+ *
+ * tasklist_lock locking optimisation:
+ * If we are current and single threaded, we do not need to take the tasklist
+ * lock or the siglock.  No one else can take our signal_struct away,
+ * no one else can reap the children to update signal->c* counters, and
+ * no one else can race with the signal-> fields.
+ * If we do not take the tasklist_lock, the signal-> fields could be read
+ * out of order while another thread was just exiting. So we place a
+ * read memory barrier when we avoid the lock.  On the writer side,
+ * write memory barrier is implied in  __exit_signal as __exit_signal releases
+ * the siglock spinlock after updating the signal-> fields.
+ *
+ * We don't really need the siglock when we access the non c* fields
+ * of the signal_struct (for RUSAGE_SELF) even in multithreaded
+ * case, since we take the tasklist lock for read and the non c* signal->
+ * fields are updated only in __exit_signal, which is called with
+ * tasklist_lock taken for write, hence these two threads cannot execute
+ * concurrently.
+ *
  */
 
 static void k_getrusage(struct task_struct *p, int who, struct rusage *r)
@@ -1694,13 +1710,23 @@ static void k_getrusage(struct task_struct *p, int who, struct rusage *r)
 	struct task_struct *t;
 	unsigned long flags;
 	cputime_t utime, stime;
+	int need_lock = 0;
 
 	memset((char *) r, 0, sizeof *r);
+	utime = stime = cputime_zero;
 
-	if (unlikely(!p->signal))
-		return;
+	if (p != current || !thread_group_empty(p))
+		need_lock = 1;
 
-	utime = stime = cputime_zero;
+	if (need_lock) {
+		read_lock(&tasklist_lock);
+		if (unlikely(!p->signal)) {
+			read_unlock(&tasklist_lock);
+			return;
+		}
+	} else
+		/* See locking comments above */
+		smp_rmb();
 
 	switch (who) {
 		case RUSAGE_BOTH:
@@ -1740,6 +1766,8 @@ static void k_getrusage(struct task_struct *p, int who, struct rusage *r)
 			BUG();
 	}
 
+	if (need_lock)
+		read_unlock(&tasklist_lock);
 	cputime_to_timeval(utime, &r->ru_utime);
 	cputime_to_timeval(stime, &r->ru_stime);
 }
@@ -1747,9 +1775,7 @@ static void k_getrusage(struct task_struct *p, int who, struct rusage *r)
 int getrusage(struct task_struct *p, int who, struct rusage __user *ru)
 {
 	struct rusage r;
-	read_lock(&tasklist_lock);
 	k_getrusage(p, who, &r);
-	read_unlock(&tasklist_lock);
 	return copy_to_user(ru, &r, sizeof(r)) ? -EFAULT : 0;
 }
 
diff --git a/lib/reed_solomon/reed_solomon.c b/lib/reed_solomon/reed_solomon.c
index f5fef948a41..f8ac9fa95de 100644
--- a/lib/reed_solomon/reed_solomon.c
+++ b/lib/reed_solomon/reed_solomon.c
@@ -44,12 +44,13 @@
 #include <linux/module.h>
 #include <linux/rslib.h>
 #include <linux/slab.h>
+#include <linux/mutex.h>
 #include <asm/semaphore.h>
 
 /* This list holds all currently allocated rs control structures */
 static LIST_HEAD (rslist);
 /* Protection for the list */
-static DECLARE_MUTEX(rslistlock);
+static DEFINE_MUTEX(rslistlock);
 
 /**
  * rs_init - Initialize a Reed-Solomon codec
@@ -161,7 +162,7 @@ errrs:
  */
 void free_rs(struct rs_control *rs)
 {
-	down(&rslistlock);
+	mutex_lock(&rslistlock);
 	rs->users--;
 	if(!rs->users) {
 		list_del(&rs->list);
@@ -170,7 +171,7 @@ void free_rs(struct rs_control *rs)
 		kfree(rs->genpoly);
 		kfree(rs);
 	}
-	up(&rslistlock);
+	mutex_unlock(&rslistlock);
 }
 
 /**
@@ -201,7 +202,7 @@ struct rs_control *init_rs(int symsize, int gfpoly, int fcr, int prim,
 	if (nroots < 0 || nroots >= (1<<symsize))
 		return NULL;
 
-	down(&rslistlock);
+	mutex_lock(&rslistlock);
 
 	/* Walk through the list and look for a matching entry */
 	list_for_each(tmp, &rslist) {
@@ -228,7 +229,7 @@ struct rs_control *init_rs(int symsize, int gfpoly, int fcr, int prim,
 		list_add(&rs->list, &rslist);
 	}
 out:
-	up(&rslistlock);
+	mutex_unlock(&rslistlock);
 	return rs;
 }
 
diff --git a/mm/readahead.c b/mm/readahead.c
index 301b36c4a0c..0f142a40984 100644
--- a/mm/readahead.c
+++ b/mm/readahead.c
@@ -555,6 +555,7 @@ recheck:
 out:
 	return ra->prev_page + 1;
 }
+EXPORT_SYMBOL_GPL(page_cache_readahead);
 
 /*
  * handle_ra_miss() is called when it is known that a page which should have
diff --git a/mm/swapfile.c b/mm/swapfile.c
index 365ed6ff182..39aa9d12961 100644
--- a/mm/swapfile.c
+++ b/mm/swapfile.c
@@ -45,7 +45,7 @@ static const char Unused_offset[] = "Unused swap offset entry ";
 
 struct swap_list_t swap_list = {-1, -1};
 
-struct swap_info_struct swap_info[MAX_SWAPFILES];
+static struct swap_info_struct swap_info[MAX_SWAPFILES];
 
 static DEFINE_MUTEX(swapon_mutex);
 
@@ -417,6 +417,61 @@ void free_swap_and_cache(swp_entry_t entry)
 	}
 }
 
+#ifdef CONFIG_SOFTWARE_SUSPEND
+/*
+ * Find the swap type that corresponds to given device (if any)
+ *
+ * This is needed for software suspend and is done in such a way that inode
+ * aliasing is allowed.
+ */
+int swap_type_of(dev_t device)
+{
+	int i;
+
+	spin_lock(&swap_lock);
+	for (i = 0; i < nr_swapfiles; i++) {
+		struct inode *inode;
+
+		if (!(swap_info[i].flags & SWP_WRITEOK))
+			continue;
+		if (!device) {
+			spin_unlock(&swap_lock);
+			return i;
+		}
+		inode = swap_info->swap_file->f_dentry->d_inode;
+		if (S_ISBLK(inode->i_mode) &&
+		    device == MKDEV(imajor(inode), iminor(inode))) {
+			spin_unlock(&swap_lock);
+			return i;
+		}
+	}
+	spin_unlock(&swap_lock);
+	return -ENODEV;
+}
+
+/*
+ * Return either the total number of swap pages of given type, or the number
+ * of free pages of that type (depending on @free)
+ *
+ * This is needed for software suspend
+ */
+unsigned int count_swap_pages(int type, int free)
+{
+	unsigned int n = 0;
+
+	if (type < nr_swapfiles) {
+		spin_lock(&swap_lock);
+		if (swap_info[type].flags & SWP_WRITEOK) {
+			n = swap_info[type].pages;
+			if (free)
+				n -= swap_info[type].inuse_pages;
+		}
+		spin_unlock(&swap_lock);
+	}
+	return n;
+}
+#endif
+
 /*
  * No need to decide whether this PTE shares the swap entry with others,
  * just let do_wp_page work it out if a write is requested later - to
diff --git a/security/seclvl.c b/security/seclvl.c
index 8529ea6f7aa..441beaf1bbc 100644
--- a/security/seclvl.c
+++ b/security/seclvl.c
@@ -8,6 +8,7 @@
  * Copyright (c) 2001 WireX Communications, Inc <chris@wirex.com>
  * Copyright (c) 2001 Greg Kroah-Hartman <greg@kroah.com>
  * Copyright (c) 2002 International Business Machines <robb@austin.ibm.com>
+ * Copyright (c) 2006 Davi E. M. Arnaut <davi.arnaut@gmail.com>
  *
  *	This program is free software; you can redistribute it and/or modify
  *	it under the terms of the GNU General Public License as published by
@@ -31,6 +32,7 @@
 #include <linux/kobject.h>
 #include <linux/crypto.h>
 #include <asm/scatterlist.h>
+#include <linux/scatterlist.h>
 #include <linux/gfp.h>
 #include <linux/sysfs.h>
 
@@ -194,35 +196,27 @@ static unsigned char hashedPassword[SHA1_DIGEST_SIZE];
  * people...
  */
 static int
-plaintext_to_sha1(unsigned char *hash, const char *plaintext, int len)
+plaintext_to_sha1(unsigned char *hash, const char *plaintext, unsigned int len)
 {
-	char *pgVirtAddr;
 	struct crypto_tfm *tfm;
-	struct scatterlist sg[1];
+	struct scatterlist sg;
 	if (len > PAGE_SIZE) {
 		seclvl_printk(0, KERN_ERR, "Plaintext password too large (%d "
 			      "characters).  Largest possible is %lu "
 			      "bytes.\n", len, PAGE_SIZE);
-		return -ENOMEM;
+		return -EINVAL;
 	}
 	tfm = crypto_alloc_tfm("sha1", CRYPTO_TFM_REQ_MAY_SLEEP);
 	if (tfm == NULL) {
 		seclvl_printk(0, KERN_ERR,
 			      "Failed to load transform for SHA1\n");
-		return -ENOSYS;
+		return -EINVAL;
 	}
-	// Just get a new page; don't play around with page boundaries
-	// and scatterlists.
-	pgVirtAddr = (char *)__get_free_page(GFP_KERNEL);
-	sg[0].page = virt_to_page(pgVirtAddr);
-	sg[0].offset = 0;
-	sg[0].length = len;
-	strncpy(pgVirtAddr, plaintext, len);
+	sg_init_one(&sg, (u8 *)plaintext, len);
 	crypto_digest_init(tfm);
-	crypto_digest_update(tfm, sg, 1);
+	crypto_digest_update(tfm, &sg, 1);
 	crypto_digest_final(tfm, hash);
 	crypto_free_tfm(tfm);
-	free_page((unsigned long)pgVirtAddr);
 	return 0;
 }
 
@@ -234,11 +228,9 @@ static ssize_t
 passwd_write_file(struct file * file, const char __user * buf,
 				size_t count, loff_t *ppos)
 {
-	int i;
-	unsigned char tmp[SHA1_DIGEST_SIZE];
-	char *page;
-	int rc;
+	char *p;
 	int len;
+	unsigned char tmp[SHA1_DIGEST_SIZE];
 
 	if (!*passwd && !*sha1_passwd) {
 		seclvl_printk(0, KERN_ERR, "Attempt to password-unlock the "
@@ -251,38 +243,39 @@ passwd_write_file(struct file * file, const char __user * buf,
 		return -EINVAL;
 	}
 
-	if (count < 0 || count >= PAGE_SIZE)
+	if (count >= PAGE_SIZE)
 		return -EINVAL;
 	if (*ppos != 0)
 		return -EINVAL;
-	page = (char *)get_zeroed_page(GFP_KERNEL);
-	if (!page)
+	p = kmalloc(count, GFP_KERNEL);
+	if (!p)
 		return -ENOMEM;
 	len = -EFAULT;
-	if (copy_from_user(page, buf, count))
+	if (copy_from_user(p, buf, count))
 		goto out;
 	
-	len = strlen(page);
+	len = count;
 	/* ``echo "secret" > seclvl/passwd'' includes a newline */
-	if (page[len - 1] == '\n')
+	if (p[len - 1] == '\n')
 		len--;
 	/* Hash the password, then compare the hashed values */
-	if ((rc = plaintext_to_sha1(tmp, page, len))) {
+	if ((len = plaintext_to_sha1(tmp, p, len))) {
 		seclvl_printk(0, KERN_ERR, "Error hashing password: rc = "
-			      "[%d]\n", rc);
-		return rc;
-	}
-	for (i = 0; i < SHA1_DIGEST_SIZE; i++) {
-		if (hashedPassword[i] != tmp[i])
-			return -EPERM;
+			      "[%d]\n", len);
+		goto out;
 	}
+
+	len = -EPERM;
+	if (memcmp(hashedPassword, tmp, SHA1_DIGEST_SIZE))
+		goto out;
+
 	seclvl_printk(0, KERN_INFO,
 		      "Password accepted; seclvl reduced to 0.\n");
 	seclvl = 0;
 	len = count;
 
 out:
-	free_page((unsigned long)page);
+	kfree (p);
 	return len;
 }
 
@@ -295,13 +288,11 @@ static struct file_operations passwd_file_ops = {
  */
 static int seclvl_ptrace(struct task_struct *parent, struct task_struct *child)
 {
-	if (seclvl >= 0) {
-		if (child->pid == 1) {
-			seclvl_printk(1, KERN_WARNING, "Attempt to ptrace "
-				      "the init process dissallowed in "
-				      "secure level %d\n", seclvl);
-			return -EPERM;
-		}
+	if (seclvl >= 0 && child->pid == 1) {
+		seclvl_printk(1, KERN_WARNING, "Attempt to ptrace "
+			      "the init process dissallowed in "
+			      "secure level %d\n", seclvl);
+		return -EPERM;
 	}
 	return 0;
 }
@@ -312,55 +303,54 @@ static int seclvl_ptrace(struct task_struct *parent, struct task_struct *child)
  */
 static int seclvl_capable(struct task_struct *tsk, int cap)
 {
+	int rc = 0;
+
 	/* init can do anything it wants */
 	if (tsk->pid == 1)
 		return 0;
 
-	switch (seclvl) {
-	case 2:
-		/* fall through */
-	case 1:
-		if (cap == CAP_LINUX_IMMUTABLE) {
+	if (seclvl > 0) {
+		rc = -EPERM;
+
+		if (cap == CAP_LINUX_IMMUTABLE)
 			seclvl_printk(1, KERN_WARNING, "Attempt to modify "
 				      "the IMMUTABLE and/or APPEND extended "
 				      "attribute on a file with the IMMUTABLE "
 				      "and/or APPEND extended attribute set "
 				      "denied in seclvl [%d]\n", seclvl);
-			return -EPERM;
-		} else if (cap == CAP_SYS_RAWIO) {	// Somewhat broad...
+		else if (cap == CAP_SYS_RAWIO)
 			seclvl_printk(1, KERN_WARNING, "Attempt to perform "
 				      "raw I/O while in secure level [%d] "
 				      "denied\n", seclvl);
-			return -EPERM;
-		} else if (cap == CAP_NET_ADMIN) {
+		else if (cap == CAP_NET_ADMIN)
 			seclvl_printk(1, KERN_WARNING, "Attempt to perform "
 				      "network administrative task while "
 				      "in secure level [%d] denied\n", seclvl);
-			return -EPERM;
-		} else if (cap == CAP_SETUID) {
+		else if (cap == CAP_SETUID)
 			seclvl_printk(1, KERN_WARNING, "Attempt to setuid "
 				      "while in secure level [%d] denied\n",
 				      seclvl);
-			return -EPERM;
-		} else if (cap == CAP_SETGID) {
+		else if (cap == CAP_SETGID)
 			seclvl_printk(1, KERN_WARNING, "Attempt to setgid "
 				      "while in secure level [%d] denied\n",
 				      seclvl);
-		} else if (cap == CAP_SYS_MODULE) {
+		else if (cap == CAP_SYS_MODULE)
 			seclvl_printk(1, KERN_WARNING, "Attempt to perform "
 				      "a module operation while in secure "
 				      "level [%d] denied\n", seclvl);
-			return -EPERM;
-		}
-		break;
-	default:
-		break;
+		else
+			rc = 0;
 	}
-	/* from dummy.c */
-	if (cap_is_fs_cap(cap) ? tsk->fsuid == 0 : tsk->euid == 0)
-		return 0;	/* capability granted */
-	seclvl_printk(1, KERN_WARNING, "Capability denied\n");
-	return -EPERM;		/* capability denied */
+
+	if (!rc) {
+		if (!(cap_is_fs_cap(cap) ? tsk->fsuid == 0 : tsk->euid == 0))
+			rc = -EPERM;
+	}
+
+	if (rc)
+		seclvl_printk(1, KERN_WARNING, "Capability denied\n");
+
+	return rc;
 }
 
 /**
@@ -466,12 +456,9 @@ static int seclvl_inode_setattr(struct dentry *dentry, struct iattr *iattr)
 static void seclvl_file_free_security(struct file *filp)
 {
 	struct dentry *dentry = filp->f_dentry;
-	struct inode *inode = NULL;
 
-	if (dentry) {
-		inode = dentry->d_inode;
-		seclvl_bd_release(inode);
-	}
+	if (dentry)
+		seclvl_bd_release(dentry->d_inode);
 }
 
 /**
@@ -479,9 +466,7 @@ static void seclvl_file_free_security(struct file *filp)
  */
 static int seclvl_umount(struct vfsmount *mnt, int flags)
 {
-	if (current->pid == 1)
-		return 0;
-	if (seclvl == 2) {
+	if (current->pid != 1 && seclvl == 2) {
 		seclvl_printk(1, KERN_WARNING, "Attempt to unmount in secure "
 			      "level %d\n", seclvl);
 		return -EPERM;
@@ -505,8 +490,9 @@ static struct security_operations seclvl_ops = {
 static int processPassword(void)
 {
 	int rc = 0;
-	hashedPassword[0] = '\0';
 	if (*passwd) {
+		char *p;
+
 		if (*sha1_passwd) {
 			seclvl_printk(0, KERN_ERR, "Error: Both "
 				      "passwd and sha1_passwd "
@@ -514,12 +500,16 @@ static int processPassword(void)
 				      "exclusive.\n");
 			return -EINVAL;
 		}
-		if ((rc = plaintext_to_sha1(hashedPassword, passwd,
-					    strlen(passwd)))) {
+
+		p = kstrdup(passwd, GFP_KERNEL);
+		if (p == NULL)
+			return -ENOMEM;
+
+		if ((rc = plaintext_to_sha1(hashedPassword, p, strlen(p))))
 			seclvl_printk(0, KERN_ERR, "Error: SHA1 support not "
 				      "in kernel\n");
-			return rc;
-		}
+
+		kfree (p);
 		/* All static data goes to the BSS, which zero's the
 		 * plaintext password out for us. */
 	} else if (*sha1_passwd) {	// Base 16
@@ -542,7 +532,7 @@ static int processPassword(void)
 			sha1_passwd[i + 2] = tmp;
 		}
 	}
-	return 0;
+	return rc;
 }
 
 /**
@@ -552,28 +542,46 @@ struct dentry *dir_ino, *seclvl_ino, *passwd_ino;
 
 static int seclvlfs_register(void)
 {
+	int rc = 0;
+
 	dir_ino = securityfs_create_dir("seclvl", NULL);
-	if (!dir_ino)
-		return -EFAULT;
+
+	if (IS_ERR(dir_ino))
+		return PTR_ERR(dir_ino);
 
 	seclvl_ino = securityfs_create_file("seclvl", S_IRUGO | S_IWUSR,
 				dir_ino, &seclvl, &seclvl_file_ops);
-	if (!seclvl_ino)
+	if (IS_ERR(seclvl_ino)) {
+		rc = PTR_ERR(seclvl_ino);
 		goto out_deldir;
+	}
 	if (*passwd || *sha1_passwd) {
 		passwd_ino = securityfs_create_file("passwd", S_IRUGO | S_IWUSR,
 				dir_ino, NULL, &passwd_file_ops);
-		if (!passwd_ino)
+		if (IS_ERR(passwd_ino)) {
+			rc = PTR_ERR(passwd_ino);
 			goto out_delf;
+		}
 	}
-	return 0;
+	return rc;
+
+out_delf:
+	securityfs_remove(seclvl_ino);
 
 out_deldir:
 	securityfs_remove(dir_ino);
-out_delf:
+
+	return rc;
+}
+
+static void seclvlfs_unregister(void)
+{
 	securityfs_remove(seclvl_ino);
 
-	return -EFAULT;
+	if (*passwd || *sha1_passwd)
+		securityfs_remove(passwd_ino);
+
+	securityfs_remove(dir_ino);
 }
 
 /**
@@ -582,6 +590,8 @@ out_delf:
 static int __init seclvl_init(void)
 {
 	int rc = 0;
+	static char once;
+
 	if (verbosity < 0 || verbosity > 1) {
 		printk(KERN_ERR "Error: bad verbosity [%d]; only 0 or 1 "
 		       "are valid values\n", verbosity);
@@ -600,6 +610,11 @@ static int __init seclvl_init(void)
 			      "module parameter(s): rc = [%d]\n", rc);
 		goto exit;
 	}
+
+	if ((rc = seclvlfs_register())) {
+		seclvl_printk(0, KERN_ERR, "Error registering with sysfs\n");
+		goto exit;
+	}
 	/* register ourselves with the security framework */
 	if (register_security(&seclvl_ops)) {
 		seclvl_printk(0, KERN_ERR,
@@ -611,20 +626,24 @@ static int __init seclvl_init(void)
 			seclvl_printk(0, KERN_ERR, "seclvl: Failure "
 				      "registering with primary security "
 				      "module.\n");
+			seclvlfs_unregister();
 			goto exit;
 		}		/* if primary module registered */
 		secondary = 1;
 	}			/* if we registered ourselves with the security framework */
-	if ((rc = seclvlfs_register())) {
-		seclvl_printk(0, KERN_ERR, "Error registering with sysfs\n");
-		goto exit;
-	}
+
 	seclvl_printk(0, KERN_INFO, "seclvl: Successfully initialized.\n");
+
+	if (once) {
+		once = 1;
+		seclvl_printk(0, KERN_INFO, "seclvl is going away. It has been "
+				"buggy for ages. Also, be warned that "
+				"Securelevels are useless.");
+	}
  exit:
-	if (rc) {
+	if (rc)
 		printk(KERN_ERR "seclvl: Error during initialization: rc = "
 		       "[%d]\n", rc);
-	}
 	return rc;
 }
 
@@ -633,17 +652,14 @@ static int __init seclvl_init(void)
  */
 static void __exit seclvl_exit(void)
 {
-	securityfs_remove(seclvl_ino);
-	if (*passwd || *sha1_passwd)
-		securityfs_remove(passwd_ino);
-	securityfs_remove(dir_ino);
-	if (secondary == 1) {
+	seclvlfs_unregister();
+
+	if (secondary)
 		mod_unreg_security(MY_NAME, &seclvl_ops);
-	} else if (unregister_security(&seclvl_ops)) {
+	else if (unregister_security(&seclvl_ops))
 		seclvl_printk(0, KERN_INFO,
 			      "seclvl: Failure unregistering with the "
 			      "kernel\n");
-	}
 }
 
 module_init(seclvl_init);
diff --git a/sound/oss/ac97_codec.c b/sound/oss/ac97_codec.c
index fd25aca2512..972327c9764 100644
--- a/sound/oss/ac97_codec.c
+++ b/sound/oss/ac97_codec.c
@@ -55,7 +55,7 @@
 #include <linux/pci.h>
 #include <linux/ac97_codec.h>
 #include <asm/uaccess.h>
-#include <asm/semaphore.h>
+#include <linux/mutex.h>
 
 #define CODEC_ID_BUFSZ 14
 
@@ -304,7 +304,7 @@ static const unsigned int ac97_oss_rm[] = {
 
 static LIST_HEAD(codecs);
 static LIST_HEAD(codec_drivers);
-static DECLARE_MUTEX(codec_sem);
+static DEFINE_MUTEX(codec_mutex);
 
 /* reads the given OSS mixer from the ac97 the caller must have insured that the ac97 knows
    about that given mixer, and should be holding a spinlock for the card */
@@ -769,9 +769,9 @@ void ac97_release_codec(struct ac97_codec *codec)
 {
 	/* Remove from the list first, we don't want to be
 	   "rediscovered" */
-	down(&codec_sem);
+	mutex_lock(&codec_mutex);
 	list_del(&codec->list);
-	up(&codec_sem);
+	mutex_unlock(&codec_mutex);
 	/*
 	 *	The driver needs to deal with internal
 	 *	locking to avoid accidents here. 
@@ -889,7 +889,7 @@ int ac97_probe_codec(struct ac97_codec *codec)
 	 *	callbacks.
 	 */
 	 
-	down(&codec_sem);
+	mutex_lock(&codec_mutex);
 	list_add(&codec->list, &codecs);
 
 	list_for_each(l, &codec_drivers) {
@@ -903,7 +903,7 @@ int ac97_probe_codec(struct ac97_codec *codec)
 		}
 	}
 
-	up(&codec_sem);
+	mutex_unlock(&codec_mutex);
 	return 1;
 }
 
@@ -1439,7 +1439,7 @@ int ac97_register_driver(struct ac97_driver *driver)
 	struct list_head *l;
 	struct ac97_codec *c;
 	
-	down(&codec_sem);
+	mutex_lock(&codec_mutex);
 	INIT_LIST_HEAD(&driver->list);
 	list_add(&driver->list, &codec_drivers);
 	
@@ -1452,7 +1452,7 @@ int ac97_register_driver(struct ac97_driver *driver)
 			continue;
 		c->driver = driver;
 	}
-	up(&codec_sem);
+	mutex_unlock(&codec_mutex);
 	return 0;
 }
 
@@ -1471,7 +1471,7 @@ void ac97_unregister_driver(struct ac97_driver *driver)
 	struct list_head *l;
 	struct ac97_codec *c;
 	
-	down(&codec_sem);
+	mutex_lock(&codec_mutex);
 	list_del_init(&driver->list);
 
 	list_for_each(l, &codecs)
@@ -1483,7 +1483,7 @@ void ac97_unregister_driver(struct ac97_driver *driver)
 		}
 	}
 	
-	up(&codec_sem);
+	mutex_unlock(&codec_mutex);
 }
 
 EXPORT_SYMBOL_GPL(ac97_unregister_driver);
@@ -1494,14 +1494,14 @@ static int swap_headphone(int remove_master)
 	struct ac97_codec *c;
 	
 	if (remove_master) {
-		down(&codec_sem);
+		mutex_lock(&codec_mutex);
 		list_for_each(l, &codecs)
 		{
 			c = list_entry(l, struct ac97_codec, list);
 			if (supported_mixer(c, SOUND_MIXER_PHONEOUT))
 				c->supported_mixers &= ~SOUND_MASK_PHONEOUT;
 		}
-		up(&codec_sem);
+		mutex_unlock(&codec_mutex);
 	} else
 		ac97_hw[SOUND_MIXER_PHONEOUT].offset = AC97_MASTER_VOL_STEREO;
 
diff --git a/sound/oss/aci.c b/sound/oss/aci.c
index 3928c2802cc..3bfac375dbd 100644
--- a/sound/oss/aci.c
+++ b/sound/oss/aci.c
@@ -56,7 +56,8 @@
 #include <linux/module.h> 
 #include <linux/proc_fs.h>
 #include <linux/slab.h>
-#include <asm/semaphore.h>
+#include <linux/mutex.h>
+
 #include <asm/io.h>
 #include <asm/uaccess.h>
 #include "sound_config.h"
@@ -79,7 +80,7 @@ static int aci_micpreamp=3; /* microphone preamp-level that can't be    *
 			 * checked with ACI versions prior to 0xb0	*/
 
 static int mixer_device;
-static struct semaphore aci_sem;
+static struct mutex aci_mutex;
 
 #ifdef MODULE
 static int reset;
@@ -212,7 +213,7 @@ int aci_rw_cmd(int write1, int write2, int write3)
 	int write[] = {write1, write2, write3};
 	int read = -EINTR, i;
 
-	if (down_interruptible(&aci_sem))
+	if (mutex_lock_interruptible(&aci_mutex))
 		goto out;
 
 	for (i=0; i<3; i++) {
@@ -227,7 +228,7 @@ int aci_rw_cmd(int write1, int write2, int write3)
 	}
 	
 	read = aci_rawread();
-out_up:	up(&aci_sem);
+out_up:	mutex_unlock(&aci_mutex);
 out:	return read;
 }
 
@@ -603,7 +604,7 @@ static int __init attach_aci(void)
 	char *boardname;
 	int i, rc = -EBUSY;
 
-	init_MUTEX(&aci_sem);
+	mutex_init(&aci_mutex);
 
 	outb(0xE3, 0xf8f); /* Write MAD16 password */
 	aci_port = (inb(0xf90) & 0x10) ?
diff --git a/sound/oss/ad1889.c b/sound/oss/ad1889.c
index a0d73f34310..54dabf86280 100644
--- a/sound/oss/ad1889.c
+++ b/sound/oss/ad1889.c
@@ -38,6 +38,7 @@
 #include <linux/ac97_codec.h>
 #include <linux/sound.h>
 #include <linux/interrupt.h>
+#include <linux/mutex.h>
 
 #include <asm/delay.h>
 #include <asm/io.h>
@@ -238,7 +239,7 @@ static ad1889_dev_t *ad1889_alloc_dev(struct pci_dev *pci)
 
 	for (i = 0; i < AD_MAX_STATES; i++) {
 		dev->state[i].card = dev;
-		init_MUTEX(&dev->state[i].sem);
+		mutex_init(&dev->state[i].mutex);
 		init_waitqueue_head(&dev->state[i].dmabuf.wait);
 	}
 
@@ -461,7 +462,7 @@ static ssize_t ad1889_write(struct file *file, const char __user *buffer, size_t
 	ssize_t ret = 0;
 	DECLARE_WAITQUEUE(wait, current);
 
-	down(&state->sem);
+	mutex_lock(&state->mutex);
 #if 0
 	if (dmabuf->mapped) {
 		ret = -ENXIO;
@@ -546,7 +547,7 @@ static ssize_t ad1889_write(struct file *file, const char __user *buffer, size_t
 err2:
 	remove_wait_queue(&state->dmabuf.wait, &wait);
 err1:
-	up(&state->sem);
+	mutex_unlock(&state->mutex);
 	return ret;
 }
 
diff --git a/sound/oss/ad1889.h b/sound/oss/ad1889.h
index e04affce1dd..861b3213f30 100644
--- a/sound/oss/ad1889.h
+++ b/sound/oss/ad1889.h
@@ -100,7 +100,7 @@ typedef struct ad1889_state {
 		unsigned int subdivision;
 	} dmabuf;
 
-	struct semaphore sem;
+	struct mutex mutex;
 } ad1889_state_t;
 
 typedef struct ad1889_dev {
diff --git a/sound/oss/ali5455.c b/sound/oss/ali5455.c
index 9c9e6c0410f..62bb936b1f3 100644
--- a/sound/oss/ali5455.c
+++ b/sound/oss/ali5455.c
@@ -64,6 +64,8 @@
 #include <linux/smp_lock.h>
 #include <linux/ac97_codec.h>
 #include <linux/interrupt.h>
+#include <linux/mutex.h>
+
 #include <asm/uaccess.h>
 
 #ifndef PCI_DEVICE_ID_ALI_5455
@@ -234,7 +236,7 @@ struct ali_state {
 	struct ali_card *card;	/* Card info */
 
 	/* single open lock mechanism, only used for recording */
-	struct semaphore open_sem;
+	struct mutex open_mutex;
 	wait_queue_head_t open_wait;
 
 	/* file mode */
@@ -2807,7 +2809,7 @@ found_virt:
 	state->card = card;
 	state->magic = ALI5455_STATE_MAGIC;
 	init_waitqueue_head(&dmabuf->wait);
-	init_MUTEX(&state->open_sem);
+	mutex_init(&state->open_mutex);
 	file->private_data = state;
 	dmabuf->trigger = 0;
 	/* allocate hardware channels */
@@ -3359,7 +3361,7 @@ static void __devinit ali_configure_clocking(void)
 		state->card = card;
 		state->magic = ALI5455_STATE_MAGIC;
 		init_waitqueue_head(&dmabuf->wait);
-		init_MUTEX(&state->open_sem);
+		mutex_init(&state->open_mutex);
 		dmabuf->fmt = ALI5455_FMT_STEREO | ALI5455_FMT_16BIT;
 		dmabuf->trigger = PCM_ENABLE_OUTPUT;
 		ali_set_dac_rate(state, 48000);
diff --git a/sound/oss/au1000.c b/sound/oss/au1000.c
index c407de86cbb..fe54de25aaf 100644
--- a/sound/oss/au1000.c
+++ b/sound/oss/au1000.c
@@ -68,6 +68,8 @@
 #include <linux/smp_lock.h>
 #include <linux/ac97_codec.h>
 #include <linux/interrupt.h>
+#include <linux/mutex.h>
+
 #include <asm/io.h>
 #include <asm/uaccess.h>
 #include <asm/mach-au1x00/au1000.h>
@@ -120,8 +122,8 @@ struct au1000_state {
 	int             no_vra;	// do not use VRA
 
 	spinlock_t      lock;
-	struct semaphore open_sem;
-	struct semaphore sem;
+	struct mutex open_mutex;
+	struct mutex sem;
 	mode_t          open_mode;
 	wait_queue_head_t open_wait;
 
@@ -1106,7 +1108,7 @@ static ssize_t au1000_read(struct file *file, char *buffer,
 
 	count *= db->cnt_factor;
 
-	down(&s->sem);
+	mutex_lock(&s->sem);
 	add_wait_queue(&db->wait, &wait);
 
 	while (count > 0) {
@@ -1125,14 +1127,14 @@ static ssize_t au1000_read(struct file *file, char *buffer,
 						ret = -EAGAIN;
 					goto out;
 				}
-				up(&s->sem);
+				mutex_unlock(&s->sem);
 				schedule();
 				if (signal_pending(current)) {
 					if (!ret)
 						ret = -ERESTARTSYS;
 					goto out2;
 				}
-				down(&s->sem);
+				mutex_lock(&s->sem);
 			}
 		} while (avail <= 0);
 
@@ -1159,7 +1161,7 @@ static ssize_t au1000_read(struct file *file, char *buffer,
 	}			// while (count > 0)
 
 out:
-	up(&s->sem);
+	mutex_unlock(&s->sem);
 out2:
 	remove_wait_queue(&db->wait, &wait);
 	set_current_state(TASK_RUNNING);
@@ -1187,7 +1189,7 @@ static ssize_t au1000_write(struct file *file, const char *buffer,
 
 	count *= db->cnt_factor;
 
-	down(&s->sem);	
+	mutex_lock(&s->sem);
 	add_wait_queue(&db->wait, &wait);
 
 	while (count > 0) {
@@ -1204,14 +1206,14 @@ static ssize_t au1000_write(struct file *file, const char *buffer,
 						ret = -EAGAIN;
 					goto out;
 				}
-				up(&s->sem);
+				mutex_unlock(&s->sem);
 				schedule();
 				if (signal_pending(current)) {
 					if (!ret)
 						ret = -ERESTARTSYS;
 					goto out2;
 				}
-				down(&s->sem);
+				mutex_lock(&s->sem);
 			}
 		} while (avail <= 0);
 
@@ -1240,7 +1242,7 @@ static ssize_t au1000_write(struct file *file, const char *buffer,
 	}			// while (count > 0)
 
 out:
-	up(&s->sem);
+	mutex_unlock(&s->sem);
 out2:
 	remove_wait_queue(&db->wait, &wait);
 	set_current_state(TASK_RUNNING);
@@ -1298,7 +1300,7 @@ static int au1000_mmap(struct file *file, struct vm_area_struct *vma)
 	dbg("%s", __FUNCTION__);
     
 	lock_kernel();
-	down(&s->sem);
+	mutex_lock(&s->sem);
 	if (vma->vm_flags & VM_WRITE)
 		db = &s->dma_dac;
 	else if (vma->vm_flags & VM_READ)
@@ -1324,7 +1326,7 @@ static int au1000_mmap(struct file *file, struct vm_area_struct *vma)
 	vma->vm_flags &= ~VM_IO;
 	db->mapped = 1;
 out:
-	up(&s->sem);
+	mutex_unlock(&s->sem);
 	unlock_kernel();
 	return ret;
 }
@@ -1829,21 +1831,21 @@ static int  au1000_open(struct inode *inode, struct file *file)
 	
 	file->private_data = s;
 	/* wait for device to become free */
-	down(&s->open_sem);
+	mutex_lock(&s->open_mutex);
 	while (s->open_mode & file->f_mode) {
 		if (file->f_flags & O_NONBLOCK) {
-			up(&s->open_sem);
+			mutex_unlock(&s->open_mutex);
 			return -EBUSY;
 		}
 		add_wait_queue(&s->open_wait, &wait);
 		__set_current_state(TASK_INTERRUPTIBLE);
-		up(&s->open_sem);
+		mutex_unlock(&s->open_mutex);
 		schedule();
 		remove_wait_queue(&s->open_wait, &wait);
 		set_current_state(TASK_RUNNING);
 		if (signal_pending(current))
 			return -ERESTARTSYS;
-		down(&s->open_sem);
+		mutex_lock(&s->open_mutex);
 	}
 
 	stop_dac(s);
@@ -1879,8 +1881,8 @@ static int  au1000_open(struct inode *inode, struct file *file)
 	}
 
 	s->open_mode |= file->f_mode & (FMODE_READ | FMODE_WRITE);
-	up(&s->open_sem);
-	init_MUTEX(&s->sem);
+	mutex_unlock(&s->open_mutex);
+	mutex_init(&s->sem);
 	return nonseekable_open(inode, file);
 }
 
@@ -1896,7 +1898,7 @@ static int au1000_release(struct inode *inode, struct file *file)
 		lock_kernel();
 	}
 
-	down(&s->open_sem);
+	mutex_lock(&s->open_mutex);
 	if (file->f_mode & FMODE_WRITE) {
 		stop_dac(s);
 		dealloc_dmabuf(s, &s->dma_dac);
@@ -1906,7 +1908,7 @@ static int au1000_release(struct inode *inode, struct file *file)
 		dealloc_dmabuf(s, &s->dma_adc);
 	}
 	s->open_mode &= ((~file->f_mode) & (FMODE_READ|FMODE_WRITE));
-	up(&s->open_sem);
+	mutex_unlock(&s->open_mutex);
 	wake_up(&s->open_wait);
 	unlock_kernel();
 	return 0;
@@ -1996,7 +1998,7 @@ static int __devinit au1000_probe(void)
 	init_waitqueue_head(&s->dma_adc.wait);
 	init_waitqueue_head(&s->dma_dac.wait);
 	init_waitqueue_head(&s->open_wait);
-	init_MUTEX(&s->open_sem);
+	mutex_init(&s->open_mutex);
 	spin_lock_init(&s->lock);
 	s->codec.private_data = s;
 	s->codec.id = 0;
diff --git a/sound/oss/au1550_ac97.c b/sound/oss/au1550_ac97.c
index bdee0502f3e..6a4956b8025 100644
--- a/sound/oss/au1550_ac97.c
+++ b/sound/oss/au1550_ac97.c
@@ -52,6 +52,8 @@
 #include <linux/spinlock.h>
 #include <linux/smp_lock.h>
 #include <linux/ac97_codec.h>
+#include <linux/mutex.h>
+
 #include <asm/io.h>
 #include <asm/uaccess.h>
 #include <asm/hardirq.h>
@@ -90,8 +92,8 @@ static struct au1550_state {
 	int             no_vra;		/* do not use VRA */
 
 	spinlock_t      lock;
-	struct semaphore open_sem;
-	struct semaphore sem;
+	struct mutex open_mutex;
+	struct mutex sem;
 	mode_t          open_mode;
 	wait_queue_head_t open_wait;
 
@@ -1044,7 +1046,7 @@ au1550_read(struct file *file, char *buffer, size_t count, loff_t *ppos)
 
 	count *= db->cnt_factor;
 
-	down(&s->sem);
+	mutex_lock(&s->sem);
 	add_wait_queue(&db->wait, &wait);
 
 	while (count > 0) {
@@ -1064,14 +1066,14 @@ au1550_read(struct file *file, char *buffer, size_t count, loff_t *ppos)
 						ret = -EAGAIN;
 					goto out;
 				}
-				up(&s->sem);
+				mutex_unlock(&s->sem);
 				schedule();
 				if (signal_pending(current)) {
 					if (!ret)
 						ret = -ERESTARTSYS;
 					goto out2;
 				}
-				down(&s->sem);
+				mutex_lock(&s->sem);
 			}
 		} while (avail <= 0);
 
@@ -1099,7 +1101,7 @@ au1550_read(struct file *file, char *buffer, size_t count, loff_t *ppos)
 	}			/* while (count > 0) */
 
 out:
-	up(&s->sem);
+	mutex_unlock(&s->sem);
 out2:
 	remove_wait_queue(&db->wait, &wait);
 	set_current_state(TASK_RUNNING);
@@ -1125,7 +1127,7 @@ au1550_write(struct file *file, const char *buffer, size_t count, loff_t * ppos)
 
 	count *= db->cnt_factor;
 
-	down(&s->sem);
+	mutex_lock(&s->sem);
 	add_wait_queue(&db->wait, &wait);
 
 	while (count > 0) {
@@ -1143,14 +1145,14 @@ au1550_write(struct file *file, const char *buffer, size_t count, loff_t * ppos)
 						ret = -EAGAIN;
 					goto out;
 				}
-				up(&s->sem);
+				mutex_unlock(&s->sem);
 				schedule();
 				if (signal_pending(current)) {
 					if (!ret)
 						ret = -ERESTARTSYS;
 					goto out2;
 				}
-				down(&s->sem);
+				mutex_lock(&s->sem);
 			}
 		} while (avail <= 0);
 
@@ -1196,7 +1198,7 @@ au1550_write(struct file *file, const char *buffer, size_t count, loff_t * ppos)
 	}			/* while (count > 0) */
 
 out:
-	up(&s->sem);
+	mutex_unlock(&s->sem);
 out2:
 	remove_wait_queue(&db->wait, &wait);
 	set_current_state(TASK_RUNNING);
@@ -1253,7 +1255,7 @@ au1550_mmap(struct file *file, struct vm_area_struct *vma)
 	int ret = 0;
 
 	lock_kernel();
-	down(&s->sem);
+	mutex_lock(&s->sem);
 	if (vma->vm_flags & VM_WRITE)
 		db = &s->dma_dac;
 	else if (vma->vm_flags & VM_READ)
@@ -1279,7 +1281,7 @@ au1550_mmap(struct file *file, struct vm_area_struct *vma)
 	vma->vm_flags &= ~VM_IO;
 	db->mapped = 1;
 out:
-	up(&s->sem);
+	mutex_unlock(&s->sem);
 	unlock_kernel();
 	return ret;
 }
@@ -1790,21 +1792,21 @@ au1550_open(struct inode *inode, struct file *file)
 
 	file->private_data = s;
 	/* wait for device to become free */
-	down(&s->open_sem);
+	mutex_lock(&s->open_mutex);
 	while (s->open_mode & file->f_mode) {
 		if (file->f_flags & O_NONBLOCK) {
-			up(&s->open_sem);
+			mutex_unlock(&s->open_mutex);
 			return -EBUSY;
 		}
 		add_wait_queue(&s->open_wait, &wait);
 		__set_current_state(TASK_INTERRUPTIBLE);
-		up(&s->open_sem);
+		mutex_unlock(&s->open_mutex);
 		schedule();
 		remove_wait_queue(&s->open_wait, &wait);
 		set_current_state(TASK_RUNNING);
 		if (signal_pending(current))
 			return -ERESTARTSYS;
-		down(&s->open_sem);
+		mutex_lock(&s->open_mutex);
 	}
 
 	stop_dac(s);
@@ -1840,8 +1842,8 @@ au1550_open(struct inode *inode, struct file *file)
 	}
 
 	s->open_mode |= file->f_mode & (FMODE_READ | FMODE_WRITE);
-	up(&s->open_sem);
-	init_MUTEX(&s->sem);
+	mutex_unlock(&s->open_mutex);
+	mutex_init(&s->sem);
 	return 0;
 }
 
@@ -1858,7 +1860,7 @@ au1550_release(struct inode *inode, struct file *file)
 		lock_kernel();
 	}
 
-	down(&s->open_sem);
+	mutex_lock(&s->open_mutex);
 	if (file->f_mode & FMODE_WRITE) {
 		stop_dac(s);
 		kfree(s->dma_dac.rawbuf);
@@ -1870,7 +1872,7 @@ au1550_release(struct inode *inode, struct file *file)
 		s->dma_adc.rawbuf = NULL;
 	}
 	s->open_mode &= ((~file->f_mode) & (FMODE_READ|FMODE_WRITE));
-	up(&s->open_sem);
+	mutex_unlock(&s->open_mutex);
 	wake_up(&s->open_wait);
 	unlock_kernel();
 	return 0;
@@ -1902,7 +1904,7 @@ au1550_probe(void)
 	init_waitqueue_head(&s->dma_adc.wait);
 	init_waitqueue_head(&s->dma_dac.wait);
 	init_waitqueue_head(&s->open_wait);
-	init_MUTEX(&s->open_sem);
+	mutex_init(&s->open_mutex);
 	spin_lock_init(&s->lock);
 
 	s->codec = ac97_alloc_codec();
diff --git a/sound/oss/btaudio.c b/sound/oss/btaudio.c
index 4007a5680ac..bfe3b534ef3 100644
--- a/sound/oss/btaudio.c
+++ b/sound/oss/btaudio.c
@@ -32,6 +32,8 @@
 #include <linux/soundcard.h>
 #include <linux/slab.h>
 #include <linux/kdev_t.h>
+#include <linux/mutex.h>
+
 #include <asm/uaccess.h>
 #include <asm/io.h>
 
@@ -108,7 +110,7 @@ struct btaudio {
 
 	/* locking */
 	int            users;
-	struct semaphore lock;
+	struct mutex lock;
 
 	/* risc instructions */
 	unsigned int   risc_size;
@@ -440,7 +442,7 @@ static struct file_operations btaudio_mixer_fops = {
 static int btaudio_dsp_open(struct inode *inode, struct file *file,
 			    struct btaudio *bta, int analog)
 {
-	down(&bta->lock);
+	mutex_lock(&bta->lock);
 	if (bta->users)
 		goto busy;
 	bta->users++;
@@ -452,11 +454,11 @@ static int btaudio_dsp_open(struct inode *inode, struct file *file,
 	bta->read_count = 0;
 	bta->sampleshift = 0;
 
-	up(&bta->lock);
+	mutex_unlock(&bta->lock);
 	return 0;
 
  busy:
-	up(&bta->lock);
+	mutex_unlock(&bta->lock);
 	return -EBUSY;
 }
 
@@ -496,11 +498,11 @@ static int btaudio_dsp_release(struct inode *inode, struct file *file)
 {
 	struct btaudio *bta = file->private_data;
 
-	down(&bta->lock);
+	mutex_lock(&bta->lock);
 	if (bta->recording)
 		stop_recording(bta);
 	bta->users--;
-	up(&bta->lock);
+	mutex_unlock(&bta->lock);
 	return 0;
 }
 
@@ -513,7 +515,7 @@ static ssize_t btaudio_dsp_read(struct file *file, char __user *buffer,
 	DECLARE_WAITQUEUE(wait, current);
 
 	add_wait_queue(&bta->readq, &wait);
-	down(&bta->lock);
+	mutex_lock(&bta->lock);
 	while (swcount > 0) {
 		if (0 == bta->read_count) {
 			if (!bta->recording) {
@@ -528,10 +530,10 @@ static ssize_t btaudio_dsp_read(struct file *file, char __user *buffer,
 					ret = -EAGAIN;
 				break;
 			}
-			up(&bta->lock);
+			mutex_unlock(&bta->lock);
 			current->state = TASK_INTERRUPTIBLE;
 			schedule();
-			down(&bta->lock);
+			mutex_lock(&bta->lock);
 			if(signal_pending(current)) {
 				if (0 == ret)
 					ret = -EINTR;
@@ -604,7 +606,7 @@ static ssize_t btaudio_dsp_read(struct file *file, char __user *buffer,
 		if (bta->read_offset == bta->buf_size)
 			bta->read_offset = 0;
 	}
-	up(&bta->lock);
+	mutex_unlock(&bta->lock);
 	remove_wait_queue(&bta->readq, &wait);
 	current->state = TASK_RUNNING;
 	return ret;
@@ -651,10 +653,10 @@ static int btaudio_dsp_ioctl(struct inode *inode, struct file *file,
 			bta->decimation  = 0;
 		}
 		if (bta->recording) {
-			down(&bta->lock);
+			mutex_lock(&bta->lock);
 			stop_recording(bta);
 			start_recording(bta);
-			up(&bta->lock);
+			mutex_unlock(&bta->lock);
 		}
 		/* fall through */
         case SOUND_PCM_READ_RATE:
@@ -716,10 +718,10 @@ static int btaudio_dsp_ioctl(struct inode *inode, struct file *file,
 			else
 				bta->bits = 16;
 			if (bta->recording) {
-				down(&bta->lock);
+				mutex_lock(&bta->lock);
 				stop_recording(bta);
 				start_recording(bta);
-				up(&bta->lock);
+				mutex_unlock(&bta->lock);
 			}
 		}
 		if (debug)
@@ -736,9 +738,9 @@ static int btaudio_dsp_ioctl(struct inode *inode, struct file *file,
 
         case SNDCTL_DSP_RESET:
 		if (bta->recording) {
-			down(&bta->lock);
+			mutex_lock(&bta->lock);
 			stop_recording(bta);
-			up(&bta->lock);
+			mutex_unlock(&bta->lock);
 		}
 		return 0;
         case SNDCTL_DSP_GETBLKSIZE:
@@ -941,7 +943,7 @@ static int __devinit btaudio_probe(struct pci_dev *pci_dev,
 	if (rate)
 		bta->rate = rate;
 	
-	init_MUTEX(&bta->lock);
+	mutex_init(&bta->lock);
         init_waitqueue_head(&bta->readq);
 
 	if (-1 != latency) {
diff --git a/sound/oss/cmpci.c b/sound/oss/cmpci.c
index 7cfbb08db53..1fbd5137f6d 100644
--- a/sound/oss/cmpci.c
+++ b/sound/oss/cmpci.c
@@ -138,6 +138,8 @@
 #endif
 #ifdef CONFIG_SOUND_CMPCI_JOYSTICK
 #include <linux/gameport.h>
+#include <linux/mutex.h>
+
 #endif
 
 /* --------------------------------------------------------------------- */
@@ -392,7 +394,7 @@ struct cm_state {
 	unsigned char fmt, enable;
 
 	spinlock_t lock;
-	struct semaphore open_sem;
+	struct mutex open_mutex;
 	mode_t open_mode;
 	wait_queue_head_t open_wait;
 
@@ -2825,21 +2827,21 @@ static int cm_open(struct inode *inode, struct file *file)
        	VALIDATE_STATE(s);
 	file->private_data = s;
 	/* wait for device to become free */
-	down(&s->open_sem);
+	mutex_lock(&s->open_mutex);
 	while (s->open_mode & file->f_mode) {
 		if (file->f_flags & O_NONBLOCK) {
-			up(&s->open_sem);
+			mutex_unlock(&s->open_mutex);
 			return -EBUSY;
 		}
 		add_wait_queue(&s->open_wait, &wait);
 		__set_current_state(TASK_INTERRUPTIBLE);
-		up(&s->open_sem);
+		mutex_unlock(&s->open_mutex);
 		schedule();
 		remove_wait_queue(&s->open_wait, &wait);
 		set_current_state(TASK_RUNNING);
 		if (signal_pending(current))
 			return -ERESTARTSYS;
-		down(&s->open_sem);
+		mutex_lock(&s->open_mutex);
 	}
 	if (file->f_mode & FMODE_READ) {
 		s->status &= ~DO_BIGENDIAN_R;
@@ -2867,7 +2869,7 @@ static int cm_open(struct inode *inode, struct file *file)
 	}
 	set_fmt(s, fmtm, fmts);
 	s->open_mode |= file->f_mode & (FMODE_READ | FMODE_WRITE);
-	up(&s->open_sem);
+	mutex_unlock(&s->open_mutex);
 	return nonseekable_open(inode, file);
 }
 
@@ -2879,7 +2881,7 @@ static int cm_release(struct inode *inode, struct file *file)
 	lock_kernel();
 	if (file->f_mode & FMODE_WRITE)
 		drain_dac(s, file->f_flags & O_NONBLOCK);
-	down(&s->open_sem);
+	mutex_lock(&s->open_mutex);
 	if (file->f_mode & FMODE_WRITE) {
 		stop_dac(s);
 
@@ -2903,7 +2905,7 @@ static int cm_release(struct inode *inode, struct file *file)
 		s->status &= ~DO_BIGENDIAN_R;
 	}
 	s->open_mode &= ~(file->f_mode & (FMODE_READ|FMODE_WRITE));
-	up(&s->open_sem);
+	mutex_unlock(&s->open_mutex);
 	wake_up(&s->open_wait);
 	unlock_kernel();
 	return 0;
@@ -3080,7 +3082,7 @@ static int __devinit cm_probe(struct pci_dev *pcidev, const struct pci_device_id
 	init_waitqueue_head(&s->dma_adc.wait);
 	init_waitqueue_head(&s->dma_dac.wait);
 	init_waitqueue_head(&s->open_wait);
-	init_MUTEX(&s->open_sem);
+	mutex_init(&s->open_mutex);
 	spin_lock_init(&s->lock);
 	s->magic = CM_MAGIC;
 	s->dev = pcidev;
diff --git a/sound/oss/cs4281/cs4281m.c b/sound/oss/cs4281/cs4281m.c
index 0720365f643..0004442f9b7 100644
--- a/sound/oss/cs4281/cs4281m.c
+++ b/sound/oss/cs4281/cs4281m.c
@@ -245,9 +245,9 @@ struct cs4281_state {
 	void *tmpbuff;		// tmp buffer for sample conversions
 	unsigned ena;
 	spinlock_t lock;
-	struct semaphore open_sem;
-	struct semaphore open_sem_adc;
-	struct semaphore open_sem_dac;
+	struct mutex open_sem;
+	struct mutex open_sem_adc;
+	struct mutex open_sem_dac;
 	mode_t open_mode;
 	wait_queue_head_t open_wait;
 	wait_queue_head_t open_wait_adc;
@@ -3598,20 +3598,20 @@ static int cs4281_release(struct inode *inode, struct file *file)
 
 	if (file->f_mode & FMODE_WRITE) {
 		drain_dac(s, file->f_flags & O_NONBLOCK);
-		down(&s->open_sem_dac);
+		mutex_lock(&s->open_sem_dac);
 		stop_dac(s);
 		dealloc_dmabuf(s, &s->dma_dac);
 		s->open_mode &= ~FMODE_WRITE;
-		up(&s->open_sem_dac);
+		mutex_unlock(&s->open_sem_dac);
 		wake_up(&s->open_wait_dac);
 	}
 	if (file->f_mode & FMODE_READ) {
 		drain_adc(s, file->f_flags & O_NONBLOCK);
-		down(&s->open_sem_adc);
+		mutex_lock(&s->open_sem_adc);
 		stop_adc(s);
 		dealloc_dmabuf(s, &s->dma_adc);
 		s->open_mode &= ~FMODE_READ;
-		up(&s->open_sem_adc);
+		mutex_unlock(&s->open_sem_adc);
 		wake_up(&s->open_wait_adc);
 	}
 	return 0;
@@ -3651,33 +3651,33 @@ static int cs4281_open(struct inode *inode, struct file *file)
 		return -ENODEV;
 	}
 	if (file->f_mode & FMODE_WRITE) {
-		down(&s->open_sem_dac);
+		mutex_lock(&s->open_sem_dac);
 		while (s->open_mode & FMODE_WRITE) {
 			if (file->f_flags & O_NONBLOCK) {
-				up(&s->open_sem_dac);
+				mutex_unlock(&s->open_sem_dac);
 				return -EBUSY;
 			}
-			up(&s->open_sem_dac);
+			mutex_unlock(&s->open_sem_dac);
 			interruptible_sleep_on(&s->open_wait_dac);
 
 			if (signal_pending(current))
 				return -ERESTARTSYS;
-			down(&s->open_sem_dac);
+			mutex_lock(&s->open_sem_dac);
 		}
 	}
 	if (file->f_mode & FMODE_READ) {
-		down(&s->open_sem_adc);
+		mutex_lock(&s->open_sem_adc);
 		while (s->open_mode & FMODE_READ) {
 			if (file->f_flags & O_NONBLOCK) {
-				up(&s->open_sem_adc);
+				mutex_unlock(&s->open_sem_adc);
 				return -EBUSY;
 			}
-			up(&s->open_sem_adc);
+			mutex_unlock(&s->open_sem_adc);
 			interruptible_sleep_on(&s->open_wait_adc);
 
 			if (signal_pending(current))
 				return -ERESTARTSYS;
-			down(&s->open_sem_adc);
+			mutex_lock(&s->open_sem_adc);
 		}
 	}
 	s->open_mode |= file->f_mode & (FMODE_READ | FMODE_WRITE);
@@ -3691,7 +3691,7 @@ static int cs4281_open(struct inode *inode, struct file *file)
 		s->ena &= ~FMODE_READ;
 		s->dma_adc.ossfragshift = s->dma_adc.ossmaxfrags =
 		    s->dma_adc.subdivision = 0;
-		up(&s->open_sem_adc);
+		mutex_unlock(&s->open_sem_adc);
 
 		if (prog_dmabuf_adc(s)) {
 			CS_DBGOUT(CS_OPEN | CS_ERROR, 2, printk(KERN_ERR
@@ -3711,7 +3711,7 @@ static int cs4281_open(struct inode *inode, struct file *file)
 		s->ena &= ~FMODE_WRITE;
 		s->dma_dac.ossfragshift = s->dma_dac.ossmaxfrags =
 		    s->dma_dac.subdivision = 0;
-		up(&s->open_sem_dac);
+		mutex_unlock(&s->open_sem_dac);
 
 		if (prog_dmabuf_dac(s)) {
 			CS_DBGOUT(CS_OPEN | CS_ERROR, 2, printk(KERN_ERR
@@ -3978,17 +3978,17 @@ static int cs4281_midi_open(struct inode *inode, struct file *file)
 	VALIDATE_STATE(s);
 	file->private_data = s;
 	// wait for device to become free 
-	down(&s->open_sem);
+	mutex_lock(&s->open_sem);
 	while (s->open_mode & (file->f_mode << FMODE_MIDI_SHIFT)) {
 		if (file->f_flags & O_NONBLOCK) {
-			up(&s->open_sem);
+			mutex_unlock(&s->open_sem);
 			return -EBUSY;
 		}
-		up(&s->open_sem);
+		mutex_unlock(&s->open_sem);
 		interruptible_sleep_on(&s->open_wait);
 		if (signal_pending(current))
 			return -ERESTARTSYS;
-		down(&s->open_sem);
+		mutex_lock(&s->open_sem);
 	}
 	spin_lock_irqsave(&s->lock, flags);
 	if (!(s->open_mode & (FMODE_MIDI_READ | FMODE_MIDI_WRITE))) {
@@ -4018,7 +4018,7 @@ static int cs4281_midi_open(struct inode *inode, struct file *file)
 	    (file->
 	     f_mode << FMODE_MIDI_SHIFT) & (FMODE_MIDI_READ |
 					    FMODE_MIDI_WRITE);
-	up(&s->open_sem);
+	mutex_unlock(&s->open_sem);
 	return nonseekable_open(inode, file);
 }
 
@@ -4057,7 +4057,7 @@ static int cs4281_midi_release(struct inode *inode, struct file *file)
 		remove_wait_queue(&s->midi.owait, &wait);
 		current->state = TASK_RUNNING;
 	}
-	down(&s->open_sem);
+	mutex_lock(&s->open_sem);
 	s->open_mode &=
 	    (~(file->f_mode << FMODE_MIDI_SHIFT)) & (FMODE_MIDI_READ |
 						     FMODE_MIDI_WRITE);
@@ -4067,7 +4067,7 @@ static int cs4281_midi_release(struct inode *inode, struct file *file)
 		del_timer(&s->midi.timer);
 	}
 	spin_unlock_irqrestore(&s->lock, flags);
-	up(&s->open_sem);
+	mutex_unlock(&s->open_sem);
 	wake_up(&s->open_wait);
 	return 0;
 }
@@ -4300,9 +4300,9 @@ static int __devinit cs4281_probe(struct pci_dev *pcidev,
 	init_waitqueue_head(&s->open_wait_dac);
 	init_waitqueue_head(&s->midi.iwait);
 	init_waitqueue_head(&s->midi.owait);
-	init_MUTEX(&s->open_sem);
-	init_MUTEX(&s->open_sem_adc);
-	init_MUTEX(&s->open_sem_dac);
+	mutex_init(&s->open_sem);
+	mutex_init(&s->open_sem_adc);
+	mutex_init(&s->open_sem_dac);
 	spin_lock_init(&s->lock);
 	s->pBA0phys = pci_resource_start(pcidev, 0);
 	s->pBA1phys = pci_resource_start(pcidev, 1);
diff --git a/sound/oss/cs46xx.c b/sound/oss/cs46xx.c
index 58e25c82eaf..53881bc91bb 100644
--- a/sound/oss/cs46xx.c
+++ b/sound/oss/cs46xx.c
@@ -90,6 +90,7 @@
 #include <linux/init.h>
 #include <linux/poll.h>
 #include <linux/ac97_codec.h>
+#include <linux/mutex.h>
 
 #include <asm/io.h>
 #include <asm/dma.h>
@@ -238,7 +239,7 @@ struct cs_state {
 	struct cs_card *card;	/* Card info */
 
 	/* single open lock mechanism, only used for recording */
-	struct semaphore open_sem;
+	struct mutex open_mutex;
 	wait_queue_head_t open_wait;
 
 	/* file mode */
@@ -297,7 +298,7 @@ struct cs_state {
 		unsigned subdivision;
 	} dmabuf;
 	/* Guard against mmap/write/read races */
-	struct semaphore sem;
+	struct mutex sem;
 };
 
 struct cs_card {
@@ -375,7 +376,7 @@ struct cs_card {
 		unsigned char ibuf[CS_MIDIINBUF];
 		unsigned char obuf[CS_MIDIOUTBUF];
 		mode_t open_mode;
-		struct semaphore open_sem;
+		struct mutex open_mutex;
 	} midi;
 	struct cs46xx_pm pm;
 };
@@ -1428,9 +1429,9 @@ static int prog_dmabuf(struct cs_state *state)
 {
 	int ret;
 	
-	down(&state->sem);
+	mutex_lock(&state->sem);
 	ret = __prog_dmabuf(state);
-	up(&state->sem);
+	mutex_unlock(&state->sem);
 	
 	return ret;
 }
@@ -1831,17 +1832,17 @@ static int cs_midi_open(struct inode *inode, struct file *file)
 
         file->private_data = card;
         /* wait for device to become free */
-        down(&card->midi.open_sem);
+        mutex_lock(&card->midi.open_mutex);
         while (card->midi.open_mode & file->f_mode) {
                 if (file->f_flags & O_NONBLOCK) {
-                        up(&card->midi.open_sem);
+                        mutex_unlock(&card->midi.open_mutex);
                         return -EBUSY;
                 }
-                up(&card->midi.open_sem);
+                mutex_unlock(&card->midi.open_mutex);
                 interruptible_sleep_on(&card->midi.open_wait);
                 if (signal_pending(current))
                         return -ERESTARTSYS;
-                down(&card->midi.open_sem);
+                mutex_lock(&card->midi.open_mutex);
         }
         spin_lock_irqsave(&card->midi.lock, flags);
         if (!(card->midi.open_mode & (FMODE_READ | FMODE_WRITE))) {
@@ -1859,7 +1860,7 @@ static int cs_midi_open(struct inode *inode, struct file *file)
         }
         spin_unlock_irqrestore(&card->midi.lock, flags);
         card->midi.open_mode |= (file->f_mode & (FMODE_READ | FMODE_WRITE));
-        up(&card->midi.open_sem);
+        mutex_unlock(&card->midi.open_mutex);
         return 0;
 }
 
@@ -1891,9 +1892,9 @@ static int cs_midi_release(struct inode *inode, struct file *file)
                 remove_wait_queue(&card->midi.owait, &wait);
                 current->state = TASK_RUNNING;
         }
-        down(&card->midi.open_sem);
+        mutex_lock(&card->midi.open_mutex);
         card->midi.open_mode &= (~(file->f_mode & (FMODE_READ | FMODE_WRITE)));
-        up(&card->midi.open_sem);
+        mutex_unlock(&card->midi.open_mutex);
         wake_up(&card->midi.open_wait);
         return 0;
 }
@@ -2081,7 +2082,7 @@ static ssize_t cs_read(struct file *file, char __user *buffer, size_t count, lof
 	if (!access_ok(VERIFY_WRITE, buffer, count))
 		return -EFAULT;
 	
-	down(&state->sem);
+	mutex_lock(&state->sem);
 	if (!dmabuf->ready && (ret = __prog_dmabuf(state)))
 		goto out2;
 
@@ -2114,13 +2115,13 @@ static ssize_t cs_read(struct file *file, char __user *buffer, size_t count, lof
 				if (!ret) ret = -EAGAIN;
 				goto out;
  			}
-			up(&state->sem);
+			mutex_unlock(&state->sem);
 			schedule();
 			if (signal_pending(current)) {
 				if(!ret) ret = -ERESTARTSYS;
 				goto out;
 			}
-			down(&state->sem);
+			mutex_lock(&state->sem);
 			if (dmabuf->mapped) 
 			{
 				if(!ret)
@@ -2155,7 +2156,7 @@ static ssize_t cs_read(struct file *file, char __user *buffer, size_t count, lof
 out:
 	remove_wait_queue(&state->dmabuf.wait, &wait);
 out2:
-	up(&state->sem);
+	mutex_unlock(&state->sem);
 	set_current_state(TASK_RUNNING);
 	CS_DBGOUT(CS_WAVE_READ | CS_FUNCTION, 4, 
 		printk("cs46xx: cs_read()- %zd\n",ret) );
@@ -2184,7 +2185,7 @@ static ssize_t cs_write(struct file *file, const char __user *buffer, size_t cou
 		return -EFAULT;
 	dmabuf = &state->dmabuf;
 
-	down(&state->sem);
+	mutex_lock(&state->sem);
 	if (dmabuf->mapped)
 	{
 		ret = -ENXIO;
@@ -2240,13 +2241,13 @@ static ssize_t cs_write(struct file *file, const char __user *buffer, size_t cou
 				if (!ret) ret = -EAGAIN;
 				goto out;
  			}
-			up(&state->sem);
+			mutex_unlock(&state->sem);
 			schedule();
  			if (signal_pending(current)) {
 				if(!ret) ret = -ERESTARTSYS;
 				goto out;
  			}
-			down(&state->sem);
+			mutex_lock(&state->sem);
 			if (dmabuf->mapped)
 			{
 				if(!ret)
@@ -2278,7 +2279,7 @@ static ssize_t cs_write(struct file *file, const char __user *buffer, size_t cou
 		start_dac(state);
 	}
 out:
-	up(&state->sem);
+	mutex_unlock(&state->sem);
 	remove_wait_queue(&state->dmabuf.wait, &wait);
 	set_current_state(TASK_RUNNING);
 
@@ -2411,7 +2412,7 @@ static int cs_mmap(struct file *file, struct vm_area_struct *vma)
 		goto out;
 	}
 
-	down(&state->sem);	
+	mutex_lock(&state->sem);
 	dmabuf = &state->dmabuf;
 	if (cs4x_pgoff(vma) != 0)
 	{
@@ -2438,7 +2439,7 @@ static int cs_mmap(struct file *file, struct vm_area_struct *vma)
 
 	CS_DBGOUT(CS_FUNCTION, 2, printk("cs46xx: cs_mmap()-\n") );
 out:
-	up(&state->sem);
+	mutex_unlock(&state->sem);
 	return ret;	
 }
 
@@ -3200,7 +3201,7 @@ static int cs_open(struct inode *inode, struct file *file)
 			if (state == NULL)
 				return -ENOMEM;
 			memset(state, 0, sizeof(struct cs_state));
-			init_MUTEX(&state->sem);
+			mutex_init(&state->sem);
 			dmabuf = &state->dmabuf;
 			dmabuf->pbuf = (void *)get_zeroed_page(GFP_KERNEL | GFP_DMA);
 			if(dmabuf->pbuf==NULL)
@@ -3241,10 +3242,10 @@ static int cs_open(struct inode *inode, struct file *file)
 		state->virt = 0;
 		state->magic = CS_STATE_MAGIC;
 		init_waitqueue_head(&dmabuf->wait);
-		init_MUTEX(&state->open_sem);
+		mutex_init(&state->open_mutex);
 		file->private_data = card;
 
-		down(&state->open_sem);
+		mutex_lock(&state->open_mutex);
 
 		/* set default sample format. According to OSS Programmer's Guide  /dev/dsp
 		   should be default to unsigned 8-bits, mono, with sample rate 8kHz and
@@ -3260,7 +3261,7 @@ static int cs_open(struct inode *inode, struct file *file)
 		cs_set_divisor(dmabuf);
 
 		state->open_mode |= FMODE_READ;
-		up(&state->open_sem);
+		mutex_unlock(&state->open_mutex);
 	}
 	if(file->f_mode & FMODE_WRITE)
 	{
@@ -3271,7 +3272,7 @@ static int cs_open(struct inode *inode, struct file *file)
 			if (state == NULL)
 				return -ENOMEM;
 			memset(state, 0, sizeof(struct cs_state));
-			init_MUTEX(&state->sem);
+			mutex_init(&state->sem);
 			dmabuf = &state->dmabuf;
 			dmabuf->pbuf = (void *)get_zeroed_page(GFP_KERNEL | GFP_DMA);
 			if(dmabuf->pbuf==NULL)
@@ -3312,10 +3313,10 @@ static int cs_open(struct inode *inode, struct file *file)
 		state->virt = 1;
 		state->magic = CS_STATE_MAGIC;
 		init_waitqueue_head(&dmabuf->wait);
-		init_MUTEX(&state->open_sem);
+		mutex_init(&state->open_mutex);
 		file->private_data = card;
 
-		down(&state->open_sem);
+		mutex_lock(&state->open_mutex);
 
 		/* set default sample format. According to OSS Programmer's Guide  /dev/dsp
 		   should be default to unsigned 8-bits, mono, with sample rate 8kHz and
@@ -3331,7 +3332,7 @@ static int cs_open(struct inode *inode, struct file *file)
 		cs_set_divisor(dmabuf);
 
 		state->open_mode |= FMODE_WRITE;
-		up(&state->open_sem);
+		mutex_unlock(&state->open_mutex);
 		if((ret = prog_dmabuf(state)))
 			return ret;
 	}
@@ -3363,14 +3364,14 @@ static int cs_release(struct inode *inode, struct file *file)
 			cs_clear_tail(state);
 			drain_dac(state, file->f_flags & O_NONBLOCK);
 			/* stop DMA state machine and free DMA buffers/channels */
-			down(&state->open_sem);
+			mutex_lock(&state->open_mutex);
 			stop_dac(state);
 			dealloc_dmabuf(state);
 			state->card->free_pcm_channel(state->card, dmabuf->channel->num);
 			free_page((unsigned long)state->dmabuf.pbuf);
 
-			/* we're covered by the open_sem */
-			up(&state->open_sem);
+			/* we're covered by the open_mutex */
+			mutex_unlock(&state->open_mutex);
 			state->card->states[state->virt] = NULL;
 			state->open_mode &= (~file->f_mode) & (FMODE_READ|FMODE_WRITE);
 
@@ -3395,14 +3396,14 @@ static int cs_release(struct inode *inode, struct file *file)
 		{
 			CS_DBGOUT(CS_RELEASE, 2, printk("cs46xx: cs_release() FMODE_READ\n") );
 			dmabuf = &state->dmabuf;
-			down(&state->open_sem);
+			mutex_lock(&state->open_mutex);
 			stop_adc(state);
 			dealloc_dmabuf(state);
 			state->card->free_pcm_channel(state->card, dmabuf->channel->num);
 			free_page((unsigned long)state->dmabuf.pbuf);
 
-			/* we're covered by the open_sem */
-			up(&state->open_sem);
+			/* we're covered by the open_mutex */
+			mutex_unlock(&state->open_mutex);
 			state->card->states[state->virt] = NULL;
 			state->open_mode &= (~file->f_mode) & (FMODE_READ|FMODE_WRITE);
 
@@ -5507,7 +5508,7 @@ static int __devinit cs46xx_probe(struct pci_dev *pci_dev,
 	}
 
         init_waitqueue_head(&card->midi.open_wait);
-        init_MUTEX(&card->midi.open_sem);
+        mutex_init(&card->midi.open_mutex);
         init_waitqueue_head(&card->midi.iwait);
         init_waitqueue_head(&card->midi.owait);
         cs461x_pokeBA0(card, BA0_MIDCR, MIDCR_MRST);   
diff --git a/sound/oss/dmasound/dmasound_awacs.c b/sound/oss/dmasound/dmasound_awacs.c
index 74f975676cc..a17375141c3 100644
--- a/sound/oss/dmasound/dmasound_awacs.c
+++ b/sound/oss/dmasound/dmasound_awacs.c
@@ -80,7 +80,7 @@
 #include <linux/kmod.h>
 #include <linux/interrupt.h>
 #include <linux/input.h>
-#include <asm/semaphore.h>
+#include <linux/mutex.h>
 #ifdef CONFIG_ADB_CUDA
 #include <linux/cuda.h>
 #endif
@@ -130,7 +130,7 @@ static struct resource awacs_rsrc[3];
 static char awacs_name[64];
 static int awacs_revision;
 static int awacs_sleeping;
-static DECLARE_MUTEX(dmasound_sem);
+static DEFINE_MUTEX(dmasound_mutex);
 
 static int sound_device_id;		/* exists after iMac revA */
 static int hw_can_byteswap = 1 ;	/* most pmac sound h/w can */
@@ -312,11 +312,11 @@ extern int daca_enter_sleep(void);
 extern int daca_leave_sleep(void);
 
 #define TRY_LOCK()	\
-	if ((rc = down_interruptible(&dmasound_sem)) != 0)	\
+	if ((rc = mutex_lock_interruptible(&dmasound_mutex)) != 0)	\
 		return rc;
-#define LOCK()		down(&dmasound_sem);
+#define LOCK()		mutex_lock(&dmasound_mutex);
 
-#define UNLOCK()	up(&dmasound_sem);
+#define UNLOCK()	mutex_unlock(&dmasound_mutex);
 
 /* We use different versions that the ones provided in dmasound.h
  * 
diff --git a/sound/oss/emu10k1/hwaccess.h b/sound/oss/emu10k1/hwaccess.h
index 104223a192a..85e27bda694 100644
--- a/sound/oss/emu10k1/hwaccess.h
+++ b/sound/oss/emu10k1/hwaccess.h
@@ -181,7 +181,7 @@ struct emu10k1_card
 	struct emu10k1_mpuout	*mpuout;
 	struct emu10k1_mpuin	*mpuin;
 
-	struct semaphore	open_sem;
+	struct mutex		open_sem;
 	mode_t			open_mode;
 	wait_queue_head_t	open_wait;
 
diff --git a/sound/oss/emu10k1/main.c b/sound/oss/emu10k1/main.c
index 23241cbdd90..0cd44a6f7ac 100644
--- a/sound/oss/emu10k1/main.c
+++ b/sound/oss/emu10k1/main.c
@@ -1320,7 +1320,7 @@ static int __devinit emu10k1_probe(struct pci_dev *pci_dev, const struct pci_dev
 	card->is_aps = (subsysvid == EMU_APS_SUBID);
 
 	spin_lock_init(&card->lock);
-	init_MUTEX(&card->open_sem);
+	mutex_init(&card->open_sem);
 	card->open_mode = 0;
 	init_waitqueue_head(&card->open_wait);
 
diff --git a/sound/oss/emu10k1/midi.c b/sound/oss/emu10k1/midi.c
index b40b5f97aac..959a96794db 100644
--- a/sound/oss/emu10k1/midi.c
+++ b/sound/oss/emu10k1/midi.c
@@ -110,21 +110,21 @@ match:
 #endif
 
 	/* Wait for device to become free */
-	down(&card->open_sem);
+	mutex_lock(&card->open_sem);
 	while (card->open_mode & (file->f_mode << FMODE_MIDI_SHIFT)) {
 		if (file->f_flags & O_NONBLOCK) {
-			up(&card->open_sem);
+			mutex_unlock(&card->open_sem);
 			return -EBUSY;
 		}
 
-		up(&card->open_sem);
+		mutex_unlock(&card->open_sem);
 		interruptible_sleep_on(&card->open_wait);
 
 		if (signal_pending(current)) {
 			return -ERESTARTSYS;
 		}
 
-		down(&card->open_sem);
+		mutex_lock(&card->open_sem);
 	}
 
 	if ((midi_dev = (struct emu10k1_mididevice *) kmalloc(sizeof(*midi_dev), GFP_KERNEL)) == NULL)
@@ -183,7 +183,7 @@ match:
 
 	card->open_mode |= (file->f_mode << FMODE_MIDI_SHIFT) & (FMODE_MIDI_READ | FMODE_MIDI_WRITE);
 
-	up(&card->open_sem);
+	mutex_unlock(&card->open_sem);
 
 	return nonseekable_open(inode, file);
 }
@@ -234,9 +234,9 @@ static int emu10k1_midi_release(struct inode *inode, struct file *file)
 
 	kfree(midi_dev);
 
-	down(&card->open_sem);
+	mutex_lock(&card->open_sem);
 	card->open_mode &= ~((file->f_mode << FMODE_MIDI_SHIFT) & (FMODE_MIDI_READ | FMODE_MIDI_WRITE));
-	up(&card->open_sem);
+	mutex_unlock(&card->open_sem);
 	wake_up_interruptible(&card->open_wait);
 
 	unlock_kernel();
diff --git a/sound/oss/es1370.c b/sound/oss/es1370.c
index ae55c536613..094f569cc6e 100644
--- a/sound/oss/es1370.c
+++ b/sound/oss/es1370.c
@@ -157,6 +157,7 @@
 #include <linux/gameport.h>
 #include <linux/wait.h>
 #include <linux/dma-mapping.h>
+#include <linux/mutex.h>
 
 #include <asm/io.h>
 #include <asm/page.h>
@@ -346,7 +347,7 @@ struct es1370_state {
 	unsigned sctrl;
 
 	spinlock_t lock;
-	struct semaphore open_sem;
+	struct mutex open_mutex;
 	mode_t open_mode;
 	wait_queue_head_t open_wait;
 
@@ -393,7 +394,7 @@ struct es1370_state {
 	struct gameport *gameport;
 #endif
 
-	struct semaphore sem;
+	struct mutex mutex;
 };
 
 /* --------------------------------------------------------------------- */
@@ -1159,7 +1160,7 @@ static ssize_t es1370_read(struct file *file, char __user *buffer, size_t count,
 		return -ENXIO;
 	if (!access_ok(VERIFY_WRITE, buffer, count))
 		return -EFAULT;
-	down(&s->sem);	
+	mutex_lock(&s->mutex);
 	if (!s->dma_adc.ready && (ret = prog_dmabuf_adc(s)))
 		goto out;
         
@@ -1183,14 +1184,14 @@ static ssize_t es1370_read(struct file *file, char __user *buffer, size_t count,
 					ret = -EAGAIN;
 				goto out;
 			}
-			up(&s->sem);
+			mutex_unlock(&s->mutex);
 			schedule();
 			if (signal_pending(current)) {
 				if (!ret)
 					ret = -ERESTARTSYS;
 				goto out;
 			}
-			down(&s->sem);
+			mutex_lock(&s->mutex);
 			if (s->dma_adc.mapped)
 			{
 				ret = -ENXIO;
@@ -1215,7 +1216,7 @@ static ssize_t es1370_read(struct file *file, char __user *buffer, size_t count,
 			start_adc(s);
 	}
 out:
-	up(&s->sem);
+	mutex_unlock(&s->mutex);
         remove_wait_queue(&s->dma_adc.wait, &wait);
 	set_current_state(TASK_RUNNING);
 	return ret;
@@ -1235,7 +1236,7 @@ static ssize_t es1370_write(struct file *file, const char __user *buffer, size_t
 		return -ENXIO;
 	if (!access_ok(VERIFY_READ, buffer, count))
 		return -EFAULT;
-	down(&s->sem);	
+	mutex_lock(&s->mutex);
 	if (!s->dma_dac2.ready && (ret = prog_dmabuf_dac2(s)))
 		goto out;
 	ret = 0;
@@ -1263,14 +1264,14 @@ static ssize_t es1370_write(struct file *file, const char __user *buffer, size_t
 					ret = -EAGAIN;
 				goto out;
 			}
-			up(&s->sem);
+			mutex_unlock(&s->mutex);
 			schedule();
 			if (signal_pending(current)) {
 				if (!ret)
 					ret = -ERESTARTSYS;
 				goto out;	
 			}
-			down(&s->sem);
+			mutex_lock(&s->mutex);
 			if (s->dma_dac2.mapped)
 			{
 			ret = -ENXIO;
@@ -1296,7 +1297,7 @@ static ssize_t es1370_write(struct file *file, const char __user *buffer, size_t
 			start_dac2(s);
 	}
 out:
-	up(&s->sem);
+	mutex_unlock(&s->mutex);
         remove_wait_queue(&s->dma_dac2.wait, &wait);
 	set_current_state(TASK_RUNNING);
 	return ret;
@@ -1348,7 +1349,7 @@ static int es1370_mmap(struct file *file, struct vm_area_struct *vma)
 
 	VALIDATE_STATE(s);
 	lock_kernel();
-	down(&s->sem);
+	mutex_lock(&s->mutex);
 	if (vma->vm_flags & VM_WRITE) {
 		if ((ret = prog_dmabuf_dac2(s)) != 0) {
 			goto out;
@@ -1380,7 +1381,7 @@ static int es1370_mmap(struct file *file, struct vm_area_struct *vma)
 	}
 	db->mapped = 1;
 out:
-	up(&s->sem);
+	mutex_unlock(&s->mutex);
 	unlock_kernel();
 	return ret;
 }
@@ -1752,21 +1753,21 @@ static int es1370_open(struct inode *inode, struct file *file)
        	VALIDATE_STATE(s);
 	file->private_data = s;
 	/* wait for device to become free */
-	down(&s->open_sem);
+	mutex_lock(&s->open_mutex);
 	while (s->open_mode & file->f_mode) {
 		if (file->f_flags & O_NONBLOCK) {
-			up(&s->open_sem);
+			mutex_unlock(&s->open_mutex);
 			return -EBUSY;
 		}
 		add_wait_queue(&s->open_wait, &wait);
 		__set_current_state(TASK_INTERRUPTIBLE);
-		up(&s->open_sem);
+		mutex_unlock(&s->open_mutex);
 		schedule();
 		remove_wait_queue(&s->open_wait, &wait);
 		set_current_state(TASK_RUNNING);
 		if (signal_pending(current))
 			return -ERESTARTSYS;
-		down(&s->open_sem);
+		mutex_lock(&s->open_mutex);
 	}
 	spin_lock_irqsave(&s->lock, flags);
 	if (!(s->open_mode & (FMODE_READ|FMODE_WRITE)))
@@ -1793,8 +1794,8 @@ static int es1370_open(struct inode *inode, struct file *file)
 	outl(s->ctrl, s->io+ES1370_REG_CONTROL);
 	spin_unlock_irqrestore(&s->lock, flags);
 	s->open_mode |= file->f_mode & (FMODE_READ | FMODE_WRITE);
-	up(&s->open_sem);
-	init_MUTEX(&s->sem);
+	mutex_unlock(&s->open_mutex);
+	mutex_init(&s->mutex);
 	return nonseekable_open(inode, file);
 }
 
@@ -1806,7 +1807,7 @@ static int es1370_release(struct inode *inode, struct file *file)
 	lock_kernel();
 	if (file->f_mode & FMODE_WRITE)
 		drain_dac2(s, file->f_flags & O_NONBLOCK);
-	down(&s->open_sem);
+	mutex_lock(&s->open_mutex);
 	if (file->f_mode & FMODE_WRITE) {
 		stop_dac2(s);
 		synchronize_irq(s->irq);
@@ -1818,7 +1819,7 @@ static int es1370_release(struct inode *inode, struct file *file)
 	}
 	s->open_mode &= ~(file->f_mode & (FMODE_READ|FMODE_WRITE));
 	wake_up(&s->open_wait);
-	up(&s->open_sem);
+	mutex_unlock(&s->open_mutex);
 	unlock_kernel();
 	return 0;
 }
@@ -2198,21 +2199,21 @@ static int es1370_open_dac(struct inode *inode, struct file *file)
 		return -EINVAL;
        	file->private_data = s;
 	/* wait for device to become free */
-	down(&s->open_sem);
+	mutex_lock(&s->open_mutex);
 	while (s->open_mode & FMODE_DAC) {
 		if (file->f_flags & O_NONBLOCK) {
-			up(&s->open_sem);
+			mutex_unlock(&s->open_mutex);
 			return -EBUSY;
 		}
 		add_wait_queue(&s->open_wait, &wait);
 		__set_current_state(TASK_INTERRUPTIBLE);
-		up(&s->open_sem);
+		mutex_unlock(&s->open_mutex);
 		schedule();
 		remove_wait_queue(&s->open_wait, &wait);
 		set_current_state(TASK_RUNNING);
 		if (signal_pending(current))
 			return -ERESTARTSYS;
-		down(&s->open_sem);
+		mutex_lock(&s->open_mutex);
 	}
 	s->dma_dac1.ossfragshift = s->dma_dac1.ossmaxfrags = s->dma_dac1.subdivision = 0;
 	s->dma_dac1.enabled = 1;
@@ -2227,7 +2228,7 @@ static int es1370_open_dac(struct inode *inode, struct file *file)
 	outl(s->ctrl, s->io+ES1370_REG_CONTROL);
 	spin_unlock_irqrestore(&s->lock, flags);
 	s->open_mode |= FMODE_DAC;
-	up(&s->open_sem);
+	mutex_unlock(&s->open_mutex);
 	return nonseekable_open(inode, file);
 }
 
@@ -2238,12 +2239,12 @@ static int es1370_release_dac(struct inode *inode, struct file *file)
 	VALIDATE_STATE(s);
 	lock_kernel();
 	drain_dac1(s, file->f_flags & O_NONBLOCK);
-	down(&s->open_sem);
+	mutex_lock(&s->open_mutex);
 	stop_dac1(s);
 	dealloc_dmabuf(s, &s->dma_dac1);
 	s->open_mode &= ~FMODE_DAC;
 	wake_up(&s->open_wait);
-	up(&s->open_sem);
+	mutex_unlock(&s->open_mutex);
 	unlock_kernel();
 	return 0;
 }
@@ -2430,21 +2431,21 @@ static int es1370_midi_open(struct inode *inode, struct file *file)
        	VALIDATE_STATE(s);
 	file->private_data = s;
 	/* wait for device to become free */
-	down(&s->open_sem);
+	mutex_lock(&s->open_mutex);
 	while (s->open_mode & (file->f_mode << FMODE_MIDI_SHIFT)) {
 		if (file->f_flags & O_NONBLOCK) {
-			up(&s->open_sem);
+			mutex_unlock(&s->open_mutex);
 			return -EBUSY;
 		}
 		add_wait_queue(&s->open_wait, &wait);
 		__set_current_state(TASK_INTERRUPTIBLE);
-		up(&s->open_sem);
+		mutex_unlock(&s->open_mutex);
 		schedule();
 		remove_wait_queue(&s->open_wait, &wait);
 		set_current_state(TASK_RUNNING);
 		if (signal_pending(current))
 			return -ERESTARTSYS;
-		down(&s->open_sem);
+		mutex_lock(&s->open_mutex);
 	}
 	spin_lock_irqsave(&s->lock, flags);
 	if (!(s->open_mode & (FMODE_MIDI_READ | FMODE_MIDI_WRITE))) {
@@ -2465,7 +2466,7 @@ static int es1370_midi_open(struct inode *inode, struct file *file)
 	es1370_handle_midi(s);
 	spin_unlock_irqrestore(&s->lock, flags);
 	s->open_mode |= (file->f_mode << FMODE_MIDI_SHIFT) & (FMODE_MIDI_READ | FMODE_MIDI_WRITE);
-	up(&s->open_sem);
+	mutex_unlock(&s->open_mutex);
 	return nonseekable_open(inode, file);
 }
 
@@ -2499,7 +2500,7 @@ static int es1370_midi_release(struct inode *inode, struct file *file)
 		remove_wait_queue(&s->midi.owait, &wait);
 		set_current_state(TASK_RUNNING);
 	}
-	down(&s->open_sem);
+	mutex_lock(&s->open_mutex);
 	s->open_mode &= ~((file->f_mode << FMODE_MIDI_SHIFT) & (FMODE_MIDI_READ|FMODE_MIDI_WRITE));
 	spin_lock_irqsave(&s->lock, flags);
 	if (!(s->open_mode & (FMODE_MIDI_READ | FMODE_MIDI_WRITE))) {
@@ -2508,7 +2509,7 @@ static int es1370_midi_release(struct inode *inode, struct file *file)
 	}
 	spin_unlock_irqrestore(&s->lock, flags);
 	wake_up(&s->open_wait);
-	up(&s->open_sem);
+	mutex_unlock(&s->open_mutex);
 	unlock_kernel();
 	return 0;
 }
@@ -2638,7 +2639,7 @@ static int __devinit es1370_probe(struct pci_dev *pcidev, const struct pci_devic
 	init_waitqueue_head(&s->open_wait);
 	init_waitqueue_head(&s->midi.iwait);
 	init_waitqueue_head(&s->midi.owait);
-	init_MUTEX(&s->open_sem);
+	mutex_init(&s->open_mutex);
 	spin_lock_init(&s->lock);
 	s->magic = ES1370_MAGIC;
 	s->dev = pcidev;
diff --git a/sound/oss/es1371.c b/sound/oss/es1371.c
index 5c697f16257..4400c853868 100644
--- a/sound/oss/es1371.c
+++ b/sound/oss/es1371.c
@@ -129,6 +129,7 @@
 #include <linux/gameport.h>
 #include <linux/wait.h>
 #include <linux/dma-mapping.h>
+#include <linux/mutex.h>
 
 #include <asm/io.h>
 #include <asm/page.h>
@@ -419,7 +420,7 @@ struct es1371_state {
 	unsigned dac1rate, dac2rate, adcrate;
 
 	spinlock_t lock;
-	struct semaphore open_sem;
+	struct mutex open_mutex;
 	mode_t open_mode;
 	wait_queue_head_t open_wait;
 
@@ -462,7 +463,7 @@ struct es1371_state {
 	struct gameport *gameport;
 #endif
 
-	struct semaphore sem;
+	struct mutex sem;
 };
 
 /* --------------------------------------------------------------------- */
@@ -1346,7 +1347,7 @@ static ssize_t es1371_read(struct file *file, char __user *buffer, size_t count,
 		return -ENXIO;
 	if (!access_ok(VERIFY_WRITE, buffer, count))
 		return -EFAULT;
-	down(&s->sem);
+	mutex_lock(&s->sem);
 	if (!s->dma_adc.ready && (ret = prog_dmabuf_adc(s)))
 		goto out2;
 	
@@ -1370,14 +1371,14 @@ static ssize_t es1371_read(struct file *file, char __user *buffer, size_t count,
 					ret = -EAGAIN;
 				goto out;
 			}
-			up(&s->sem);
+			mutex_unlock(&s->sem);
 			schedule();
 			if (signal_pending(current)) {
 				if (!ret)
 					ret = -ERESTARTSYS;
 				goto out2;
 			}
-			down(&s->sem);
+			mutex_lock(&s->sem);
 			if (s->dma_adc.mapped)
 			{
 				ret = -ENXIO;
@@ -1402,7 +1403,7 @@ static ssize_t es1371_read(struct file *file, char __user *buffer, size_t count,
 			start_adc(s);
 	}
 out:
-	up(&s->sem);
+	mutex_unlock(&s->sem);
 out2:
 	remove_wait_queue(&s->dma_adc.wait, &wait);
 	set_current_state(TASK_RUNNING);
@@ -1423,7 +1424,7 @@ static ssize_t es1371_write(struct file *file, const char __user *buffer, size_t
 		return -ENXIO;
 	if (!access_ok(VERIFY_READ, buffer, count))
 		return -EFAULT;
-	down(&s->sem);	
+	mutex_lock(&s->sem);
 	if (!s->dma_dac2.ready && (ret = prog_dmabuf_dac2(s)))
 		goto out3;
 	ret = 0;
@@ -1451,14 +1452,14 @@ static ssize_t es1371_write(struct file *file, const char __user *buffer, size_t
 					ret = -EAGAIN;
 				goto out;
 			}	
-			up(&s->sem);
+			mutex_unlock(&s->sem);
 			schedule();
 			if (signal_pending(current)) {
 				if (!ret)
 					ret = -ERESTARTSYS;
 				goto out2;
 			}
-			down(&s->sem);
+			mutex_lock(&s->sem);
 			if (s->dma_dac2.mapped)
 			{
 				ret = -ENXIO;
@@ -1484,7 +1485,7 @@ static ssize_t es1371_write(struct file *file, const char __user *buffer, size_t
 			start_dac2(s);
 	}
 out:
-	up(&s->sem);
+	mutex_unlock(&s->sem);
 out2:
 	remove_wait_queue(&s->dma_dac2.wait, &wait);
 out3:	
@@ -1538,7 +1539,7 @@ static int es1371_mmap(struct file *file, struct vm_area_struct *vma)
 
 	VALIDATE_STATE(s);
 	lock_kernel();
-	down(&s->sem);
+	mutex_lock(&s->sem);
 	
 	if (vma->vm_flags & VM_WRITE) {
 		if ((ret = prog_dmabuf_dac2(s)) != 0) {
@@ -1571,7 +1572,7 @@ static int es1371_mmap(struct file *file, struct vm_area_struct *vma)
 	}
 	db->mapped = 1;
 out:
-	up(&s->sem);
+	mutex_unlock(&s->sem);
 	unlock_kernel();
 	return ret;
 }
@@ -1938,21 +1939,21 @@ static int es1371_open(struct inode *inode, struct file *file)
        	VALIDATE_STATE(s);
 	file->private_data = s;
 	/* wait for device to become free */
-	down(&s->open_sem);
+	mutex_lock(&s->open_mutex);
 	while (s->open_mode & file->f_mode) {
 		if (file->f_flags & O_NONBLOCK) {
-			up(&s->open_sem);
+			mutex_unlock(&s->open_mutex);
 			return -EBUSY;
 		}
 		add_wait_queue(&s->open_wait, &wait);
 		__set_current_state(TASK_INTERRUPTIBLE);
-		up(&s->open_sem);
+		mutex_unlock(&s->open_mutex);
 		schedule();
 		remove_wait_queue(&s->open_wait, &wait);
 		set_current_state(TASK_RUNNING);
 		if (signal_pending(current))
 			return -ERESTARTSYS;
-		down(&s->open_sem);
+		mutex_lock(&s->open_mutex);
 	}
 	if (file->f_mode & FMODE_READ) {
 		s->dma_adc.ossfragshift = s->dma_adc.ossmaxfrags = s->dma_adc.subdivision = 0;
@@ -1982,8 +1983,8 @@ static int es1371_open(struct inode *inode, struct file *file)
 	outl(s->sctrl, s->io+ES1371_REG_SERIAL_CONTROL);
 	spin_unlock_irqrestore(&s->lock, flags);
 	s->open_mode |= file->f_mode & (FMODE_READ | FMODE_WRITE);
-	up(&s->open_sem);
-	init_MUTEX(&s->sem);
+	mutex_unlock(&s->open_mutex);
+	mutex_init(&s->sem);
 	return nonseekable_open(inode, file);
 }
 
@@ -1995,7 +1996,7 @@ static int es1371_release(struct inode *inode, struct file *file)
 	lock_kernel();
 	if (file->f_mode & FMODE_WRITE)
 		drain_dac2(s, file->f_flags & O_NONBLOCK);
-	down(&s->open_sem);
+	mutex_lock(&s->open_mutex);
 	if (file->f_mode & FMODE_WRITE) {
 		stop_dac2(s);
 		dealloc_dmabuf(s, &s->dma_dac2);
@@ -2005,7 +2006,7 @@ static int es1371_release(struct inode *inode, struct file *file)
 		dealloc_dmabuf(s, &s->dma_adc);
 	}
 	s->open_mode &= ~(file->f_mode & (FMODE_READ|FMODE_WRITE));
-	up(&s->open_sem);
+	mutex_unlock(&s->open_mutex);
 	wake_up(&s->open_wait);
 	unlock_kernel();
 	return 0;
@@ -2377,21 +2378,21 @@ static int es1371_open_dac(struct inode *inode, struct file *file)
 		return -EINVAL;
        	file->private_data = s;
 	/* wait for device to become free */
-	down(&s->open_sem);
+	mutex_lock(&s->open_mutex);
 	while (s->open_mode & FMODE_DAC) {
 		if (file->f_flags & O_NONBLOCK) {
-			up(&s->open_sem);
+			mutex_unlock(&s->open_mutex);
 			return -EBUSY;
 		}
 		add_wait_queue(&s->open_wait, &wait);
 		__set_current_state(TASK_INTERRUPTIBLE);
-		up(&s->open_sem);
+		mutex_unlock(&s->open_mutex);
 		schedule();
 		remove_wait_queue(&s->open_wait, &wait);
 		set_current_state(TASK_RUNNING);
 		if (signal_pending(current))
 			return -ERESTARTSYS;
-		down(&s->open_sem);
+		mutex_lock(&s->open_mutex);
 	}
 	s->dma_dac1.ossfragshift = s->dma_dac1.ossmaxfrags = s->dma_dac1.subdivision = 0;
 	s->dma_dac1.enabled = 1;
@@ -2405,7 +2406,7 @@ static int es1371_open_dac(struct inode *inode, struct file *file)
 	outl(s->sctrl, s->io+ES1371_REG_SERIAL_CONTROL);
 	spin_unlock_irqrestore(&s->lock, flags);
 	s->open_mode |= FMODE_DAC;
-	up(&s->open_sem);
+	mutex_unlock(&s->open_mutex);
 	return nonseekable_open(inode, file);
 }
 
@@ -2416,11 +2417,11 @@ static int es1371_release_dac(struct inode *inode, struct file *file)
 	VALIDATE_STATE(s);
 	lock_kernel();
 	drain_dac1(s, file->f_flags & O_NONBLOCK);
-	down(&s->open_sem);
+	mutex_lock(&s->open_mutex);
 	stop_dac1(s);
 	dealloc_dmabuf(s, &s->dma_dac1);
 	s->open_mode &= ~FMODE_DAC;
-	up(&s->open_sem);
+	mutex_unlock(&s->open_mutex);
 	wake_up(&s->open_wait);
 	unlock_kernel();
 	return 0;
@@ -2608,21 +2609,21 @@ static int es1371_midi_open(struct inode *inode, struct file *file)
        	VALIDATE_STATE(s);
 	file->private_data = s;
 	/* wait for device to become free */
-	down(&s->open_sem);
+	mutex_lock(&s->open_mutex);
 	while (s->open_mode & (file->f_mode << FMODE_MIDI_SHIFT)) {
 		if (file->f_flags & O_NONBLOCK) {
-			up(&s->open_sem);
+			mutex_unlock(&s->open_mutex);
 			return -EBUSY;
 		}
 		add_wait_queue(&s->open_wait, &wait);
 		__set_current_state(TASK_INTERRUPTIBLE);
-		up(&s->open_sem);
+		mutex_unlock(&s->open_mutex);
 		schedule();
 		remove_wait_queue(&s->open_wait, &wait);
 		set_current_state(TASK_RUNNING);
 		if (signal_pending(current))
 			return -ERESTARTSYS;
-		down(&s->open_sem);
+		mutex_lock(&s->open_mutex);
 	}
 	spin_lock_irqsave(&s->lock, flags);
 	if (!(s->open_mode & (FMODE_MIDI_READ | FMODE_MIDI_WRITE))) {
@@ -2643,7 +2644,7 @@ static int es1371_midi_open(struct inode *inode, struct file *file)
 	es1371_handle_midi(s);
 	spin_unlock_irqrestore(&s->lock, flags);
 	s->open_mode |= (file->f_mode << FMODE_MIDI_SHIFT) & (FMODE_MIDI_READ | FMODE_MIDI_WRITE);
-	up(&s->open_sem);
+	mutex_unlock(&s->open_mutex);
 	return nonseekable_open(inode, file);
 }
 
@@ -2676,7 +2677,7 @@ static int es1371_midi_release(struct inode *inode, struct file *file)
 		remove_wait_queue(&s->midi.owait, &wait);
 		set_current_state(TASK_RUNNING);
 	}
-	down(&s->open_sem);
+	mutex_lock(&s->open_mutex);
 	s->open_mode &= ~((file->f_mode << FMODE_MIDI_SHIFT) & (FMODE_MIDI_READ|FMODE_MIDI_WRITE));
 	spin_lock_irqsave(&s->lock, flags);
 	if (!(s->open_mode & (FMODE_MIDI_READ | FMODE_MIDI_WRITE))) {
@@ -2684,7 +2685,7 @@ static int es1371_midi_release(struct inode *inode, struct file *file)
 		outl(s->ctrl, s->io+ES1371_REG_CONTROL);
 	}
 	spin_unlock_irqrestore(&s->lock, flags);
-	up(&s->open_sem);
+	mutex_unlock(&s->open_mutex);
 	wake_up(&s->open_wait);
 	unlock_kernel();
 	return 0;
@@ -2884,7 +2885,7 @@ static int __devinit es1371_probe(struct pci_dev *pcidev, const struct pci_devic
 	init_waitqueue_head(&s->open_wait);
 	init_waitqueue_head(&s->midi.iwait);
 	init_waitqueue_head(&s->midi.owait);
-	init_MUTEX(&s->open_sem);
+	mutex_init(&s->open_mutex);
 	spin_lock_init(&s->lock);
 	s->magic = ES1371_MAGIC;
 	s->dev = pcidev;
diff --git a/sound/oss/esssolo1.c b/sound/oss/esssolo1.c
index 849b59f67ef..78d3e29ce96 100644
--- a/sound/oss/esssolo1.c
+++ b/sound/oss/esssolo1.c
@@ -105,6 +105,8 @@
 #include <linux/gameport.h>
 #include <linux/wait.h>
 #include <linux/dma-mapping.h>
+#include <linux/mutex.h>
+
 
 #include <asm/io.h>
 #include <asm/page.h>
@@ -191,7 +193,7 @@ struct solo1_state {
 	unsigned ena;
 
 	spinlock_t lock;
-	struct semaphore open_sem;
+	struct mutex open_mutex;
 	mode_t open_mode;
 	wait_queue_head_t open_wait;
 
@@ -1581,7 +1583,7 @@ static int solo1_release(struct inode *inode, struct file *file)
 	lock_kernel();
 	if (file->f_mode & FMODE_WRITE)
 		drain_dac(s, file->f_flags & O_NONBLOCK);
-	down(&s->open_sem);
+	mutex_lock(&s->open_mutex);
 	if (file->f_mode & FMODE_WRITE) {
 		stop_dac(s);
 		outb(0, s->iobase+6);  /* disable DMA */
@@ -1595,7 +1597,7 @@ static int solo1_release(struct inode *inode, struct file *file)
 	}
 	s->open_mode &= ~(FMODE_READ | FMODE_WRITE);
 	wake_up(&s->open_wait);
-	up(&s->open_sem);
+	mutex_unlock(&s->open_mutex);
 	unlock_kernel();
 	return 0;
 }
@@ -1624,21 +1626,21 @@ static int solo1_open(struct inode *inode, struct file *file)
        	VALIDATE_STATE(s);
 	file->private_data = s;
 	/* wait for device to become free */
-	down(&s->open_sem);
+	mutex_lock(&s->open_mutex);
 	while (s->open_mode & (FMODE_READ | FMODE_WRITE)) {
 		if (file->f_flags & O_NONBLOCK) {
-			up(&s->open_sem);
+			mutex_unlock(&s->open_mutex);
 			return -EBUSY;
 		}
 		add_wait_queue(&s->open_wait, &wait);
 		__set_current_state(TASK_INTERRUPTIBLE);
-		up(&s->open_sem);
+		mutex_unlock(&s->open_mutex);
 		schedule();
 		remove_wait_queue(&s->open_wait, &wait);
 		set_current_state(TASK_RUNNING);
 		if (signal_pending(current))
 			return -ERESTARTSYS;
-		down(&s->open_sem);
+		mutex_lock(&s->open_mutex);
 	}
 	s->fmt = AFMT_U8;
 	s->channels = 1;
@@ -1650,7 +1652,7 @@ static int solo1_open(struct inode *inode, struct file *file)
 	s->dma_dac.ossfragshift = s->dma_dac.ossmaxfrags = s->dma_dac.subdivision = 0;
 	s->dma_dac.enabled = 1;
 	s->open_mode |= file->f_mode & (FMODE_READ | FMODE_WRITE);
-	up(&s->open_sem);
+	mutex_unlock(&s->open_mutex);
 	prog_codec(s);
 	return nonseekable_open(inode, file);
 }
@@ -1911,21 +1913,21 @@ static int solo1_midi_open(struct inode *inode, struct file *file)
        	VALIDATE_STATE(s);
 	file->private_data = s;
 	/* wait for device to become free */
-	down(&s->open_sem);
+	mutex_lock(&s->open_mutex);
 	while (s->open_mode & (file->f_mode << FMODE_MIDI_SHIFT)) {
 		if (file->f_flags & O_NONBLOCK) {
-			up(&s->open_sem);
+			mutex_unlock(&s->open_mutex);
 			return -EBUSY;
 		}
 		add_wait_queue(&s->open_wait, &wait);
 		__set_current_state(TASK_INTERRUPTIBLE);
-		up(&s->open_sem);
+		mutex_unlock(&s->open_mutex);
 		schedule();
 		remove_wait_queue(&s->open_wait, &wait);
 		set_current_state(TASK_RUNNING);
 		if (signal_pending(current))
 			return -ERESTARTSYS;
-		down(&s->open_sem);
+		mutex_lock(&s->open_mutex);
 	}
 	spin_lock_irqsave(&s->lock, flags);
 	if (!(s->open_mode & (FMODE_MIDI_READ | FMODE_MIDI_WRITE))) {
@@ -1951,7 +1953,7 @@ static int solo1_midi_open(struct inode *inode, struct file *file)
 	}
 	spin_unlock_irqrestore(&s->lock, flags);
 	s->open_mode |= (file->f_mode << FMODE_MIDI_SHIFT) & (FMODE_MIDI_READ | FMODE_MIDI_WRITE);
-	up(&s->open_sem);
+	mutex_unlock(&s->open_mutex);
 	return nonseekable_open(inode, file);
 }
 
@@ -1985,7 +1987,7 @@ static int solo1_midi_release(struct inode *inode, struct file *file)
 		remove_wait_queue(&s->midi.owait, &wait);
 		set_current_state(TASK_RUNNING);
 	}
-	down(&s->open_sem);
+	mutex_lock(&s->open_mutex);
 	s->open_mode &= ~((file->f_mode << FMODE_MIDI_SHIFT) & (FMODE_MIDI_READ|FMODE_MIDI_WRITE));
 	spin_lock_irqsave(&s->lock, flags);
 	if (!(s->open_mode & (FMODE_MIDI_READ | FMODE_MIDI_WRITE))) {
@@ -1994,7 +1996,7 @@ static int solo1_midi_release(struct inode *inode, struct file *file)
 	}
 	spin_unlock_irqrestore(&s->lock, flags);
 	wake_up(&s->open_wait);
-	up(&s->open_sem);
+	mutex_unlock(&s->open_mutex);
 	unlock_kernel();
 	return 0;
 }
@@ -2132,24 +2134,24 @@ static int solo1_dmfm_open(struct inode *inode, struct file *file)
        	VALIDATE_STATE(s);
 	file->private_data = s;
 	/* wait for device to become free */
-	down(&s->open_sem);
+	mutex_lock(&s->open_mutex);
 	while (s->open_mode & FMODE_DMFM) {
 		if (file->f_flags & O_NONBLOCK) {
-			up(&s->open_sem);
+			mutex_unlock(&s->open_mutex);
 			return -EBUSY;
 		}
 		add_wait_queue(&s->open_wait, &wait);
 		__set_current_state(TASK_INTERRUPTIBLE);
-		up(&s->open_sem);
+		mutex_unlock(&s->open_mutex);
 		schedule();
 		remove_wait_queue(&s->open_wait, &wait);
 		set_current_state(TASK_RUNNING);
 		if (signal_pending(current))
 			return -ERESTARTSYS;
-		down(&s->open_sem);
+		mutex_lock(&s->open_mutex);
 	}
 	if (!request_region(s->sbbase, FMSYNTH_EXTENT, "ESS Solo1")) {
-		up(&s->open_sem);
+		mutex_unlock(&s->open_mutex);
 		printk(KERN_ERR "solo1: FM synth io ports in use, opl3 loaded?\n");
 		return -EBUSY;
 	}
@@ -2161,7 +2163,7 @@ static int solo1_dmfm_open(struct inode *inode, struct file *file)
 	outb(5, s->sbbase+2);
 	outb(1, s->sbbase+3);  /* enable OPL3 */
 	s->open_mode |= FMODE_DMFM;
-	up(&s->open_sem);
+	mutex_unlock(&s->open_mutex);
 	return nonseekable_open(inode, file);
 }
 
@@ -2172,7 +2174,7 @@ static int solo1_dmfm_release(struct inode *inode, struct file *file)
 
 	VALIDATE_STATE(s);
 	lock_kernel();
-	down(&s->open_sem);
+	mutex_lock(&s->open_mutex);
 	s->open_mode &= ~FMODE_DMFM;
 	for (regb = 0xb0; regb < 0xb9; regb++) {
 		outb(regb, s->sbbase);
@@ -2182,7 +2184,7 @@ static int solo1_dmfm_release(struct inode *inode, struct file *file)
 	}
 	release_region(s->sbbase, FMSYNTH_EXTENT);
 	wake_up(&s->open_wait);
-	up(&s->open_sem);
+	mutex_unlock(&s->open_mutex);
 	unlock_kernel();
 	return 0;
 }
@@ -2362,7 +2364,7 @@ static int __devinit solo1_probe(struct pci_dev *pcidev, const struct pci_device
 	init_waitqueue_head(&s->open_wait);
 	init_waitqueue_head(&s->midi.iwait);
 	init_waitqueue_head(&s->midi.owait);
-	init_MUTEX(&s->open_sem);
+	mutex_init(&s->open_mutex);
 	spin_lock_init(&s->lock);
 	s->magic = SOLO1_MAGIC;
 	s->dev = pcidev;
diff --git a/sound/oss/forte.c b/sound/oss/forte.c
index 8406bc90c4f..0294eec8ad9 100644
--- a/sound/oss/forte.c
+++ b/sound/oss/forte.c
@@ -43,6 +43,7 @@
 #include <linux/interrupt.h>
 
 #include <linux/proc_fs.h>
+#include <linux/mutex.h>
 
 #include <asm/uaccess.h>
 #include <asm/io.h>
@@ -185,7 +186,7 @@ struct forte_chip {
 	unsigned long		iobase;
 	int			irq;
 
-	struct semaphore	open_sem; 	/* Device access */
+	struct mutex		open_mutex; 	/* Device access */
 	spinlock_t		lock;		/* State */
 
 	spinlock_t		ac97_lock;
@@ -1242,13 +1243,13 @@ forte_dsp_open (struct inode *inode, struct file *file)
 	struct forte_chip *chip = forte; /* FIXME: HACK FROM HELL! */
 
 	if (file->f_flags & O_NONBLOCK) {
-		if (down_trylock (&chip->open_sem)) {
+		if (!mutex_trylock(&chip->open_mutex)) {
 			DPRINTK ("%s: returning -EAGAIN\n", __FUNCTION__);
 			return -EAGAIN;
 		}
 	}
 	else {
-		if (down_interruptible (&chip->open_sem)) {
+		if (mutex_lock_interruptible(&chip->open_mutex)) {
 			DPRINTK ("%s: returning -ERESTARTSYS\n", __FUNCTION__);
 			return -ERESTARTSYS;
 		}
@@ -1302,7 +1303,7 @@ forte_dsp_release (struct inode *inode, struct file *file)
 		spin_unlock_irq (&chip->lock);
 	}
 
-	up (&chip->open_sem);
+	mutex_unlock(&chip->open_mutex);
 
 	return ret;
 }
@@ -2011,7 +2012,7 @@ forte_probe (struct pci_dev *pci_dev, const struct pci_device_id *pci_id)
 	memset (chip, 0, sizeof (struct forte_chip));
 	chip->pci_dev = pci_dev;
 
-	init_MUTEX(&chip->open_sem);
+	mutex_init(&chip->open_mutex);
 	spin_lock_init (&chip->lock);
 	spin_lock_init (&chip->ac97_lock);
 
diff --git a/sound/oss/hal2.c b/sound/oss/hal2.c
index afe97c4ce06..dd4f59d30a3 100644
--- a/sound/oss/hal2.c
+++ b/sound/oss/hal2.c
@@ -32,6 +32,8 @@
 #include <linux/dma-mapping.h>
 #include <linux/sound.h>
 #include <linux/soundcard.h>
+#include <linux/mutex.h>
+
 
 #include <asm/io.h>
 #include <asm/sgi/hpc3.h>
@@ -92,7 +94,7 @@ struct hal2_codec {
 
 	wait_queue_head_t dma_wait;
 	spinlock_t lock;
-	struct semaphore sem;
+	struct mutex sem;
 
 	int usecount;			/* recording and playback are
 					 * independent */
@@ -1178,7 +1180,7 @@ static ssize_t hal2_read(struct file *file, char *buffer,
 
 	if (!count)
 		return 0;
-	if (down_interruptible(&adc->sem))
+	if (mutex_lock_interruptible(&adc->sem))
 		return -EINTR;
 	if (file->f_flags & O_NONBLOCK) {
 		err = hal2_get_buffer(hal2, buffer, count);
@@ -1217,7 +1219,7 @@ static ssize_t hal2_read(struct file *file, char *buffer,
 			}
 		} while (count > 0 && err >= 0);
 	}
-	up(&adc->sem);
+	mutex_unlock(&adc->sem);
 
 	return err;
 }
@@ -1232,7 +1234,7 @@ static ssize_t hal2_write(struct file *file, const char *buffer,
 
 	if (!count)
 		return 0;
-	if (down_interruptible(&dac->sem))
+	if (mutex_lock_interruptible(&dac->sem))
 		return -EINTR;
 	if (file->f_flags & O_NONBLOCK) {
 		err = hal2_add_buffer(hal2, buf, count);
@@ -1271,7 +1273,7 @@ static ssize_t hal2_write(struct file *file, const char *buffer,
 			}
 		} while (count > 0 && err >= 0);
 	}
-	up(&dac->sem);
+	mutex_unlock(&dac->sem);
 
 	return err;
 }
@@ -1356,20 +1358,20 @@ static int hal2_release(struct inode *inode, struct file *file)
 	if (file->f_mode & FMODE_READ) {
 		struct hal2_codec *adc = &hal2->adc;
 
-		down(&adc->sem);
+		mutex_lock(&adc->sem);
 		hal2_stop_adc(hal2);
 		hal2_free_adc_dmabuf(adc);
 		adc->usecount--;
-		up(&adc->sem);
+		mutex_unlock(&adc->sem);
 	}
 	if (file->f_mode & FMODE_WRITE) {
 		struct hal2_codec *dac = &hal2->dac;
 
-		down(&dac->sem);
+		mutex_lock(&dac->sem);
 		hal2_sync_dac(hal2);
 		hal2_free_dac_dmabuf(dac);
 		dac->usecount--;
-		up(&dac->sem);
+		mutex_unlock(&dac->sem);
 	}
 
 	return 0;
@@ -1400,7 +1402,7 @@ static void hal2_init_codec(struct hal2_codec *codec, struct hpc3_regs *hpc3,
 	codec->pbus.pbusnr = index;
 	codec->pbus.pbus = &hpc3->pbdma[index];
 	init_waitqueue_head(&codec->dma_wait);
-	init_MUTEX(&codec->sem);
+	mutex_init(&codec->sem);
 	spin_lock_init(&codec->lock);
 }
 
diff --git a/sound/oss/i810_audio.c b/sound/oss/i810_audio.c
index abc242abd5b..dd2b871cdac 100644
--- a/sound/oss/i810_audio.c
+++ b/sound/oss/i810_audio.c
@@ -100,6 +100,8 @@
 #include <linux/smp_lock.h>
 #include <linux/ac97_codec.h>
 #include <linux/bitops.h>
+#include <linux/mutex.h>
+
 #include <asm/uaccess.h>
 
 #define DRIVER_VERSION "1.01"
@@ -331,7 +333,7 @@ struct i810_state {
 	struct i810_card *card;	/* Card info */
 
 	/* single open lock mechanism, only used for recording */
-	struct semaphore open_sem;
+	struct mutex open_mutex;
 	wait_queue_head_t open_wait;
 
 	/* file mode */
@@ -2597,7 +2599,7 @@ found_virt:
 	state->card = card;
 	state->magic = I810_STATE_MAGIC;
 	init_waitqueue_head(&dmabuf->wait);
-	init_MUTEX(&state->open_sem);
+	mutex_init(&state->open_mutex);
 	file->private_data = state;
 	dmabuf->trigger = 0;
 
@@ -3213,7 +3215,7 @@ static void __devinit i810_configure_clocking (void)
 		state->card = card;
 		state->magic = I810_STATE_MAGIC;
 		init_waitqueue_head(&dmabuf->wait);
-		init_MUTEX(&state->open_sem);
+		mutex_init(&state->open_mutex);
 		dmabuf->fmt = I810_FMT_STEREO | I810_FMT_16BIT;
 		dmabuf->trigger = PCM_ENABLE_OUTPUT;
 		i810_set_spdif_output(state, -1, 0);
diff --git a/sound/oss/ite8172.c b/sound/oss/ite8172.c
index 8fd2f9a9e66..ffcb910f5c3 100644
--- a/sound/oss/ite8172.c
+++ b/sound/oss/ite8172.c
@@ -71,6 +71,8 @@
 #include <linux/smp_lock.h>
 #include <linux/ac97_codec.h>
 #include <linux/interrupt.h>
+#include <linux/mutex.h>
+
 #include <asm/io.h>
 #include <asm/dma.h>
 #include <asm/uaccess.h>
@@ -304,7 +306,7 @@ struct it8172_state {
 	unsigned dacrate, adcrate;
 
 	spinlock_t lock;
-	struct semaphore open_sem;
+	struct mutex open_mutex;
 	mode_t open_mode;
 	wait_queue_head_t open_wait;
 
@@ -1801,21 +1803,21 @@ static int it8172_open(struct inode *inode, struct file *file)
 	}
 	file->private_data = s;
 	/* wait for device to become free */
-	down(&s->open_sem);
+	mutex_lock(&s->open_mutex);
 	while (s->open_mode & file->f_mode) {
 		if (file->f_flags & O_NONBLOCK) {
-			up(&s->open_sem);
+			mutex_unlock(&s->open_mutex);
 			return -EBUSY;
 		}
 		add_wait_queue(&s->open_wait, &wait);
 		__set_current_state(TASK_INTERRUPTIBLE);
-		up(&s->open_sem);
+		mutex_unlock(&s->open_mutex);
 		schedule();
 		remove_wait_queue(&s->open_wait, &wait);
 		set_current_state(TASK_RUNNING);
 		if (signal_pending(current))
 			return -ERESTARTSYS;
-		down(&s->open_sem);
+		mutex_lock(&s->open_mutex);
 	}
 
 	spin_lock_irqsave(&s->lock, flags);
@@ -1850,7 +1852,7 @@ static int it8172_open(struct inode *inode, struct file *file)
 	spin_unlock_irqrestore(&s->lock, flags);
 
 	s->open_mode |= (file->f_mode & (FMODE_READ | FMODE_WRITE));
-	up(&s->open_sem);
+	mutex_unlock(&s->open_mutex);
 	return nonseekable_open(inode, file);
 }
 
@@ -1864,7 +1866,7 @@ static int it8172_release(struct inode *inode, struct file *file)
 	lock_kernel();
 	if (file->f_mode & FMODE_WRITE)
 		drain_dac(s, file->f_flags & O_NONBLOCK);
-	down(&s->open_sem);
+	mutex_lock(&s->open_mutex);
 	if (file->f_mode & FMODE_WRITE) {
 		stop_dac(s);
 		dealloc_dmabuf(s, &s->dma_dac);
@@ -1874,7 +1876,7 @@ static int it8172_release(struct inode *inode, struct file *file)
 		dealloc_dmabuf(s, &s->dma_adc);
 	}
 	s->open_mode &= ((~file->f_mode) & (FMODE_READ|FMODE_WRITE));
-	up(&s->open_sem);
+	mutex_unlock(&s->open_mutex);
 	wake_up(&s->open_wait);
 	unlock_kernel();
 	return 0;
@@ -1997,7 +1999,7 @@ static int __devinit it8172_probe(struct pci_dev *pcidev,
 	init_waitqueue_head(&s->dma_adc.wait);
 	init_waitqueue_head(&s->dma_dac.wait);
 	init_waitqueue_head(&s->open_wait);
-	init_MUTEX(&s->open_sem);
+	mutex_init(&s->open_mutex);
 	spin_lock_init(&s->lock);
 	s->dev = pcidev;
 	s->io = pci_resource_start(pcidev, 0);
diff --git a/sound/oss/maestro.c b/sound/oss/maestro.c
index d4b569acf76..e647f2f8627 100644
--- a/sound/oss/maestro.c
+++ b/sound/oss/maestro.c
@@ -223,6 +223,8 @@
 #include <linux/reboot.h>
 #include <linux/bitops.h>
 #include <linux/wait.h>
+#include <linux/mutex.h>
+
 
 #include <asm/current.h>
 #include <asm/dma.h>
@@ -397,7 +399,7 @@ struct ess_state {
 	/* this locks around the oss state in the driver */
 	spinlock_t lock;
 	/* only let 1 be opening at a time */
-	struct semaphore open_sem;
+	struct mutex open_mutex;
 	wait_queue_head_t open_wait;
 	mode_t open_mode;
 
@@ -3020,26 +3022,26 @@ ess_open(struct inode *inode, struct file *file)
        	VALIDATE_STATE(s);
 	file->private_data = s;
 	/* wait for device to become free */
-	down(&s->open_sem);
+	mutex_lock(&s->open_mutex);
 	while (s->open_mode & file->f_mode) {
 		if (file->f_flags & O_NONBLOCK) {
-			up(&s->open_sem);
+			mutex_unlock(&s->open_mutex);
 			return -EWOULDBLOCK;
 		}
-		up(&s->open_sem);
+		mutex_unlock(&s->open_mutex);
 		interruptible_sleep_on(&s->open_wait);
 		if (signal_pending(current))
 			return -ERESTARTSYS;
-		down(&s->open_sem);
+		mutex_lock(&s->open_mutex);
 	}
 
 	/* under semaphore.. */
 	if ((s->card->dmapages==NULL) && allocate_buffers(s)) {
-		up(&s->open_sem);
+		mutex_unlock(&s->open_mutex);
 		return -ENOMEM;
 	}
 
-	/* we're covered by the open_sem */
+	/* we're covered by the open_mutex */
 	if( ! s->card->dsps_open )  {
 		maestro_power(s->card,ACPI_D0);
 		start_bob(s);
@@ -3076,7 +3078,7 @@ ess_open(struct inode *inode, struct file *file)
 	set_fmt(s, fmtm, fmts);
 	s->open_mode |= file->f_mode & (FMODE_READ | FMODE_WRITE);
 
-	up(&s->open_sem);
+	mutex_unlock(&s->open_mutex);
 	return nonseekable_open(inode, file);
 }
 
@@ -3089,7 +3091,7 @@ ess_release(struct inode *inode, struct file *file)
 	lock_kernel();
 	if (file->f_mode & FMODE_WRITE)
 		drain_dac(s, file->f_flags & O_NONBLOCK);
-	down(&s->open_sem);
+	mutex_lock(&s->open_mutex);
 	if (file->f_mode & FMODE_WRITE) {
 		stop_dac(s);
 	}
@@ -3098,7 +3100,7 @@ ess_release(struct inode *inode, struct file *file)
 	}
 		
 	s->open_mode &= (~file->f_mode) & (FMODE_READ|FMODE_WRITE);
-	/* we're covered by the open_sem */
+	/* we're covered by the open_mutex */
 	M_printk("maestro: %d dsps now alive\n",s->card->dsps_open-1);
 	if( --s->card->dsps_open <= 0) {
 		s->card->dsps_open = 0;
@@ -3106,7 +3108,7 @@ ess_release(struct inode *inode, struct file *file)
 		free_buffers(s);
 		maestro_power(s->card,ACPI_D2);
 	}
-	up(&s->open_sem);
+	mutex_unlock(&s->open_mutex);
 	wake_up(&s->open_wait);
 	unlock_kernel();
 	return 0;
@@ -3466,7 +3468,7 @@ maestro_probe(struct pci_dev *pcidev,const struct pci_device_id *pdid)
 		init_waitqueue_head(&s->dma_dac.wait);
 		init_waitqueue_head(&s->open_wait);
 		spin_lock_init(&s->lock);
-		init_MUTEX(&s->open_sem);
+		mutex_init(&s->open_mutex);
 		s->magic = ESS_STATE_MAGIC;
 		
 		s->apu[0] = 6*i;
diff --git a/sound/oss/maestro3.c b/sound/oss/maestro3.c
index f3dec70fcb9..66044aff258 100644
--- a/sound/oss/maestro3.c
+++ b/sound/oss/maestro3.c
@@ -144,6 +144,8 @@
 #include <linux/spinlock.h>
 #include <linux/ac97_codec.h>
 #include <linux/wait.h>
+#include <linux/mutex.h>
+
 
 #include <asm/io.h>
 #include <asm/dma.h>
@@ -205,7 +207,7 @@ struct m3_state {
 		when irqhandler uses s->lock
 		and m3_assp_read uses card->lock ?
 		*/
-    struct semaphore open_sem;
+    struct mutex open_mutex;
     wait_queue_head_t open_wait;
     mode_t open_mode;
 
@@ -2013,17 +2015,17 @@ static int m3_open(struct inode *inode, struct file *file)
     file->private_data = s;
 
     /* wait for device to become free */
-    down(&s->open_sem);
+    mutex_lock(&s->open_mutex);
     while (s->open_mode & file->f_mode) {
         if (file->f_flags & O_NONBLOCK) {
-            up(&s->open_sem);
+            mutex_unlock(&s->open_mutex);
             return -EWOULDBLOCK;
         }
-        up(&s->open_sem);
+        mutex_unlock(&s->open_mutex);
         interruptible_sleep_on(&s->open_wait);
         if (signal_pending(current))
             return -ERESTARTSYS;
-        down(&s->open_sem);
+        mutex_lock(&s->open_mutex);
     }
     
     spin_lock_irqsave(&c->lock, flags);
@@ -2047,7 +2049,7 @@ static int m3_open(struct inode *inode, struct file *file)
     set_fmt(s, fmtm, fmts);
     s->open_mode |= file->f_mode & (FMODE_READ | FMODE_WRITE);
 
-    up(&s->open_sem);
+    mutex_unlock(&s->open_mutex);
     spin_unlock_irqrestore(&c->lock, flags);
     return nonseekable_open(inode, file);
 }
@@ -2062,7 +2064,7 @@ static int m3_release(struct inode *inode, struct file *file)
     if (file->f_mode & FMODE_WRITE)
         drain_dac(s, file->f_flags & O_NONBLOCK);
 
-    down(&s->open_sem);
+    mutex_lock(&s->open_mutex);
     spin_lock_irqsave(&card->lock, flags);
 
     if (file->f_mode & FMODE_WRITE) {
@@ -2083,7 +2085,7 @@ static int m3_release(struct inode *inode, struct file *file)
     s->open_mode &= (~file->f_mode) & (FMODE_READ|FMODE_WRITE);
 
     spin_unlock_irqrestore(&card->lock, flags);
-    up(&s->open_sem);
+    mutex_unlock(&s->open_mutex);
     wake_up(&s->open_wait);
 
     return 0;
@@ -2679,7 +2681,7 @@ static int __devinit m3_probe(struct pci_dev *pci_dev, const struct pci_device_i
         init_waitqueue_head(&s->dma_adc.wait);
         init_waitqueue_head(&s->dma_dac.wait);
         init_waitqueue_head(&s->open_wait);
-        init_MUTEX(&(s->open_sem));
+        mutex_init(&(s->open_mutex));
         s->magic = M3_STATE_MAGIC;
 
         m3_assp_client_init(s);
diff --git a/sound/oss/nec_vrc5477.c b/sound/oss/nec_vrc5477.c
index fbb9170e8e0..21c1954d910 100644
--- a/sound/oss/nec_vrc5477.c
+++ b/sound/oss/nec_vrc5477.c
@@ -78,6 +78,8 @@
 #include <linux/spinlock.h>
 #include <linux/smp_lock.h>
 #include <linux/ac97_codec.h>
+#include <linux/mutex.h>
+
 #include <asm/io.h>
 #include <asm/dma.h>
 #include <asm/uaccess.h>
@@ -198,7 +200,7 @@ struct vrc5477_ac97_state {
 	unsigned short extended_status;
 
 	spinlock_t lock;
-	struct semaphore open_sem;
+	struct mutex open_mutex;
 	mode_t open_mode;
 	wait_queue_head_t open_wait;
 
@@ -1617,22 +1619,22 @@ static int vrc5477_ac97_open(struct inode *inode, struct file *file)
 	file->private_data = s;
 
 	/* wait for device to become free */
-	down(&s->open_sem);
+	mutex_lock(&s->open_mutex);
 	while (s->open_mode & file->f_mode) {
 
 		if (file->f_flags & O_NONBLOCK) {
-			up(&s->open_sem);
+			mutex_unlock(&s->open_mutex);
 			return -EBUSY;
 		}
 		add_wait_queue(&s->open_wait, &wait);
 		__set_current_state(TASK_INTERRUPTIBLE);
-		up(&s->open_sem);
+		mutex_unlock(&s->open_mutex);
 		schedule();
 		remove_wait_queue(&s->open_wait, &wait);
 		set_current_state(TASK_RUNNING);
 		if (signal_pending(current))
 			return -ERESTARTSYS;
-		down(&s->open_sem);
+		mutex_lock(&s->open_mutex);
 	}
 
 	spin_lock_irqsave(&s->lock, flags);
@@ -1659,7 +1661,7 @@ static int vrc5477_ac97_open(struct inode *inode, struct file *file)
  bailout:
 	spin_unlock_irqrestore(&s->lock, flags);
 
-	up(&s->open_sem);
+	mutex_unlock(&s->open_mutex);
 	return ret;
 }
 
@@ -1671,7 +1673,7 @@ static int vrc5477_ac97_release(struct inode *inode, struct file *file)
 	lock_kernel();
 	if (file->f_mode & FMODE_WRITE)
 		drain_dac(s, file->f_flags & O_NONBLOCK);
-	down(&s->open_sem);
+	mutex_lock(&s->open_mutex);
 	if (file->f_mode & FMODE_WRITE) {
 		stop_dac(s);
 		dealloc_dmabuf(s, &s->dma_dac);
@@ -1681,7 +1683,7 @@ static int vrc5477_ac97_release(struct inode *inode, struct file *file)
 		dealloc_dmabuf(s, &s->dma_adc);
 	}
 	s->open_mode &= (~file->f_mode) & (FMODE_READ|FMODE_WRITE);
-	up(&s->open_sem);
+	mutex_unlock(&s->open_mutex);
 	wake_up(&s->open_wait);
 	unlock_kernel();
 	return 0;
@@ -1867,7 +1869,7 @@ static int __devinit vrc5477_ac97_probe(struct pci_dev *pcidev,
 	init_waitqueue_head(&s->dma_adc.wait);
 	init_waitqueue_head(&s->dma_dac.wait);
 	init_waitqueue_head(&s->open_wait);
-	init_MUTEX(&s->open_sem);
+	mutex_init(&s->open_mutex);
 	spin_lock_init(&s->lock);
 
 	s->dev = pcidev;
diff --git a/sound/oss/rme96xx.c b/sound/oss/rme96xx.c
index faa0b7919b6..a1ec9d131ab 100644
--- a/sound/oss/rme96xx.c
+++ b/sound/oss/rme96xx.c
@@ -58,6 +58,7 @@ TODO:
 #include <linux/interrupt.h>
 #include <linux/poll.h>
 #include <linux/wait.h>
+#include <linux/mutex.h>
 
 #include <asm/dma.h>
 #include <asm/page.h>
@@ -326,7 +327,7 @@ typedef struct _rme96xx_info {
 
 		/* waiting and locking */
 		wait_queue_head_t wait;
-		struct semaphore  open_sem;
+		struct mutex  open_mutex;
 		wait_queue_head_t open_wait;
 
 	} dma[RME96xx_MAX_DEVS]; 
@@ -842,7 +843,7 @@ static void busmaster_free(void* ptr,int size) {
 
 static int rme96xx_dmabuf_init(rme96xx_info * s,struct dmabuf* dma,int ioffset,int ooffset) {
 
-	init_MUTEX(&dma->open_sem);
+	mutex_init(&dma->open_mutex);
 	init_waitqueue_head(&dma->open_wait);
 	init_waitqueue_head(&dma->wait);
 	dma->s = s; 
@@ -1469,21 +1470,21 @@ static int rme96xx_open(struct inode *in, struct file *f)
 	dma = &s->dma[devnum];
 	f->private_data = dma;
 	/* wait for device to become free */
-	down(&dma->open_sem);
+	mutex_lock(&dma->open_mutex);
 	while (dma->open_mode & f->f_mode) {
 		if (f->f_flags & O_NONBLOCK) {
-			up(&dma->open_sem);
+			mutex_unlock(&dma->open_mutex);
 			return -EBUSY;
 		}
 		add_wait_queue(&dma->open_wait, &wait);
 		__set_current_state(TASK_INTERRUPTIBLE);
-		up(&dma->open_sem);
+		mutex_unlock(&dma->open_mutex);
 		schedule();
 		remove_wait_queue(&dma->open_wait, &wait);
 		set_current_state(TASK_RUNNING);
 		if (signal_pending(current))
 			return -ERESTARTSYS;
-		down(&dma->open_sem);
+		mutex_lock(&dma->open_mutex);
 	}
 
 	COMM                ("hardware open")
@@ -1492,7 +1493,7 @@ static int rme96xx_open(struct inode *in, struct file *f)
 
 	dma->open_mode |= (f->f_mode & (FMODE_READ | FMODE_WRITE));
 	dma->opened = 1;
-	up(&dma->open_sem);
+	mutex_unlock(&dma->open_mutex);
 
 	DBG(printk("device num %d open finished\n",devnum));
 	return 0;
@@ -1524,7 +1525,7 @@ static int rme96xx_release(struct inode *in, struct file *file)
 	}
 
 	wake_up(&dma->open_wait);
-	up(&dma->open_sem);
+	mutex_unlock(&dma->open_mutex);
 
 	return 0;
 }
diff --git a/sound/oss/sonicvibes.c b/sound/oss/sonicvibes.c
index 71b05e2f697..69a4b8778b5 100644
--- a/sound/oss/sonicvibes.c
+++ b/sound/oss/sonicvibes.c
@@ -116,6 +116,8 @@
 #include <linux/spinlock.h>
 #include <linux/smp_lock.h>
 #include <linux/gameport.h>
+#include <linux/mutex.h>
+
 
 #include <asm/io.h>
 #include <asm/uaccess.h>
@@ -328,7 +330,7 @@ struct sv_state {
 	unsigned char fmt, enable;
 
 	spinlock_t lock;
-	struct semaphore open_sem;
+	struct mutex open_mutex;
 	mode_t open_mode;
 	wait_queue_head_t open_wait;
 
@@ -1922,21 +1924,21 @@ static int sv_open(struct inode *inode, struct file *file)
        	VALIDATE_STATE(s);
 	file->private_data = s;
 	/* wait for device to become free */
-	down(&s->open_sem);
+	mutex_lock(&s->open_mutex);
 	while (s->open_mode & file->f_mode) {
 		if (file->f_flags & O_NONBLOCK) {
-			up(&s->open_sem);
+			mutex_unlock(&s->open_mutex);
 			return -EBUSY;
 		}
 		add_wait_queue(&s->open_wait, &wait);
 		__set_current_state(TASK_INTERRUPTIBLE);
-		up(&s->open_sem);
+		mutex_unlock(&s->open_mutex);
 		schedule();
 		remove_wait_queue(&s->open_wait, &wait);
 		set_current_state(TASK_RUNNING);
 		if (signal_pending(current))
 			return -ERESTARTSYS;
-		down(&s->open_sem);
+		mutex_lock(&s->open_mutex);
 	}
 	if (file->f_mode & FMODE_READ) {
 		fmtm &= ~((SV_CFMT_STEREO | SV_CFMT_16BIT) << SV_CFMT_CSHIFT);
@@ -1956,7 +1958,7 @@ static int sv_open(struct inode *inode, struct file *file)
 	}
 	set_fmt(s, fmtm, fmts);
 	s->open_mode |= file->f_mode & (FMODE_READ | FMODE_WRITE);
-	up(&s->open_sem);
+	mutex_unlock(&s->open_mutex);
 	return nonseekable_open(inode, file);
 }
 
@@ -1968,7 +1970,7 @@ static int sv_release(struct inode *inode, struct file *file)
 	lock_kernel();
 	if (file->f_mode & FMODE_WRITE)
 		drain_dac(s, file->f_flags & O_NONBLOCK);
-	down(&s->open_sem);
+	mutex_lock(&s->open_mutex);
 	if (file->f_mode & FMODE_WRITE) {
 		stop_dac(s);
 		dealloc_dmabuf(s, &s->dma_dac);
@@ -1979,7 +1981,7 @@ static int sv_release(struct inode *inode, struct file *file)
 	}
 	s->open_mode &= ~(file->f_mode & (FMODE_READ|FMODE_WRITE));
 	wake_up(&s->open_wait);
-	up(&s->open_sem);
+	mutex_unlock(&s->open_mutex);
 	unlock_kernel();
 	return 0;
 }
@@ -2167,21 +2169,21 @@ static int sv_midi_open(struct inode *inode, struct file *file)
        	VALIDATE_STATE(s);
 	file->private_data = s;
 	/* wait for device to become free */
-	down(&s->open_sem);
+	mutex_lock(&s->open_mutex);
 	while (s->open_mode & (file->f_mode << FMODE_MIDI_SHIFT)) {
 		if (file->f_flags & O_NONBLOCK) {
-			up(&s->open_sem);
+			mutex_unlock(&s->open_mutex);
 			return -EBUSY;
 		}
 		add_wait_queue(&s->open_wait, &wait);
 		__set_current_state(TASK_INTERRUPTIBLE);
-		up(&s->open_sem);
+		mutex_unlock(&s->open_mutex);
 		schedule();
 		remove_wait_queue(&s->open_wait, &wait);
 		set_current_state(TASK_RUNNING);
 		if (signal_pending(current))
 			return -ERESTARTSYS;
-		down(&s->open_sem);
+		mutex_lock(&s->open_mutex);
 	}
 	spin_lock_irqsave(&s->lock, flags);
 	if (!(s->open_mode & (FMODE_MIDI_READ | FMODE_MIDI_WRITE))) {
@@ -2210,7 +2212,7 @@ static int sv_midi_open(struct inode *inode, struct file *file)
 	}
 	spin_unlock_irqrestore(&s->lock, flags);
 	s->open_mode |= (file->f_mode << FMODE_MIDI_SHIFT) & (FMODE_MIDI_READ | FMODE_MIDI_WRITE);
-	up(&s->open_sem);
+	mutex_unlock(&s->open_mutex);
 	return nonseekable_open(inode, file);
 }
 
@@ -2248,7 +2250,7 @@ static int sv_midi_release(struct inode *inode, struct file *file)
 		remove_wait_queue(&s->midi.owait, &wait);
 		set_current_state(TASK_RUNNING);
 	}
-	down(&s->open_sem);
+	mutex_lock(&s->open_mutex);
 	s->open_mode &= ~((file->f_mode << FMODE_MIDI_SHIFT) & (FMODE_MIDI_READ|FMODE_MIDI_WRITE));
 	spin_lock_irqsave(&s->lock, flags);
 	if (!(s->open_mode & (FMODE_MIDI_READ | FMODE_MIDI_WRITE))) {
@@ -2257,7 +2259,7 @@ static int sv_midi_release(struct inode *inode, struct file *file)
 	}
 	spin_unlock_irqrestore(&s->lock, flags);
 	wake_up(&s->open_wait);
-	up(&s->open_sem);
+	mutex_unlock(&s->open_mutex);
 	unlock_kernel();
 	return 0;
 }
@@ -2388,21 +2390,21 @@ static int sv_dmfm_open(struct inode *inode, struct file *file)
        	VALIDATE_STATE(s);
 	file->private_data = s;
 	/* wait for device to become free */
-	down(&s->open_sem);
+	mutex_lock(&s->open_mutex);
 	while (s->open_mode & FMODE_DMFM) {
 		if (file->f_flags & O_NONBLOCK) {
-			up(&s->open_sem);
+			mutex_unlock(&s->open_mutex);
 			return -EBUSY;
 		}
 		add_wait_queue(&s->open_wait, &wait);
 		__set_current_state(TASK_INTERRUPTIBLE);
-		up(&s->open_sem);
+		mutex_unlock(&s->open_mutex);
 		schedule();
 		remove_wait_queue(&s->open_wait, &wait);
 		set_current_state(TASK_RUNNING);
 		if (signal_pending(current))
 			return -ERESTARTSYS;
-		down(&s->open_sem);
+		mutex_lock(&s->open_mutex);
 	}
 	/* init the stuff */
 	outb(1, s->iosynth);
@@ -2412,7 +2414,7 @@ static int sv_dmfm_open(struct inode *inode, struct file *file)
 	outb(5, s->iosynth+2);
 	outb(1, s->iosynth+3);  /* enable OPL3 */
 	s->open_mode |= FMODE_DMFM;
-	up(&s->open_sem);
+	mutex_unlock(&s->open_mutex);
 	return nonseekable_open(inode, file);
 }
 
@@ -2423,7 +2425,7 @@ static int sv_dmfm_release(struct inode *inode, struct file *file)
 
 	VALIDATE_STATE(s);
 	lock_kernel();
-	down(&s->open_sem);
+	mutex_lock(&s->open_mutex);
 	s->open_mode &= ~FMODE_DMFM;
 	for (regb = 0xb0; regb < 0xb9; regb++) {
 		outb(regb, s->iosynth);
@@ -2432,7 +2434,7 @@ static int sv_dmfm_release(struct inode *inode, struct file *file)
 		outb(0, s->iosynth+3);
 	}
 	wake_up(&s->open_wait);
-	up(&s->open_sem);
+	mutex_unlock(&s->open_mutex);
 	unlock_kernel();
 	return 0;
 }
@@ -2582,7 +2584,7 @@ static int __devinit sv_probe(struct pci_dev *pcidev, const struct pci_device_id
 	init_waitqueue_head(&s->open_wait);
 	init_waitqueue_head(&s->midi.iwait);
 	init_waitqueue_head(&s->midi.owait);
-	init_MUTEX(&s->open_sem);
+	mutex_init(&s->open_mutex);
 	spin_lock_init(&s->lock);
 	s->magic = SV_MAGIC;
 	s->dev = pcidev;
diff --git a/sound/oss/swarm_cs4297a.c b/sound/oss/swarm_cs4297a.c
index df4d3771fa8..dce9016cbcf 100644
--- a/sound/oss/swarm_cs4297a.c
+++ b/sound/oss/swarm_cs4297a.c
@@ -76,6 +76,7 @@
 #include <linux/init.h>
 #include <linux/poll.h>
 #include <linux/smp_lock.h>
+#include <linux/mutex.h>
 
 #include <asm/byteorder.h>
 #include <asm/dma.h>
@@ -291,9 +292,9 @@ struct cs4297a_state {
 	unsigned conversion:1;	// conversion from 16 to 8 bit in progress
 	unsigned ena;
 	spinlock_t lock;
-	struct semaphore open_sem;
-	struct semaphore open_sem_adc;
-	struct semaphore open_sem_dac;
+	struct mutex open_mutex;
+	struct mutex open_sem_adc;
+	struct mutex open_sem_dac;
 	mode_t open_mode;
 	wait_queue_head_t open_wait;
 	wait_queue_head_t open_wait_adc;
@@ -2352,20 +2353,20 @@ static int cs4297a_release(struct inode *inode, struct file *file)
 
 	if (file->f_mode & FMODE_WRITE) {
 		drain_dac(s, file->f_flags & O_NONBLOCK);
-		down(&s->open_sem_dac);
+		mutex_lock(&s->open_sem_dac);
 		stop_dac(s);
 		dealloc_dmabuf(s, &s->dma_dac);
 		s->open_mode &= ~FMODE_WRITE;
-		up(&s->open_sem_dac);
+		mutex_unlock(&s->open_sem_dac);
 		wake_up(&s->open_wait_dac);
 	}
 	if (file->f_mode & FMODE_READ) {
 		drain_adc(s, file->f_flags & O_NONBLOCK);
-		down(&s->open_sem_adc);
+		mutex_lock(&s->open_sem_adc);
 		stop_adc(s);
 		dealloc_dmabuf(s, &s->dma_adc);
 		s->open_mode &= ~FMODE_READ;
-		up(&s->open_sem_adc);
+		mutex_unlock(&s->open_sem_adc);
 		wake_up(&s->open_wait_adc);
 	}
 	return 0;
@@ -2413,37 +2414,37 @@ static int cs4297a_open(struct inode *inode, struct file *file)
                                 ;
                 }
           
-		down(&s->open_sem_dac);
+		mutex_lock(&s->open_sem_dac);
 		while (s->open_mode & FMODE_WRITE) {
 			if (file->f_flags & O_NONBLOCK) {
-				up(&s->open_sem_dac);
+				mutex_unlock(&s->open_sem_dac);
 				return -EBUSY;
 			}
-			up(&s->open_sem_dac);
+			mutex_unlock(&s->open_sem_dac);
 			interruptible_sleep_on(&s->open_wait_dac);
 
 			if (signal_pending(current)) {
                                 printk("open - sig pending\n");
 				return -ERESTARTSYS;
                         }
-			down(&s->open_sem_dac);
+			mutex_lock(&s->open_sem_dac);
 		}
 	}
 	if (file->f_mode & FMODE_READ) {
-		down(&s->open_sem_adc);
+		mutex_lock(&s->open_sem_adc);
 		while (s->open_mode & FMODE_READ) {
 			if (file->f_flags & O_NONBLOCK) {
-				up(&s->open_sem_adc);
+				mutex_unlock(&s->open_sem_adc);
 				return -EBUSY;
 			}
-			up(&s->open_sem_adc);
+			mutex_unlock(&s->open_sem_adc);
 			interruptible_sleep_on(&s->open_wait_adc);
 
 			if (signal_pending(current)) {
                                 printk("open - sig pending\n");
 				return -ERESTARTSYS;
                         }
-			down(&s->open_sem_adc);
+			mutex_lock(&s->open_sem_adc);
 		}
 	}
 	s->open_mode |= file->f_mode & (FMODE_READ | FMODE_WRITE);
@@ -2456,7 +2457,7 @@ static int cs4297a_open(struct inode *inode, struct file *file)
 		s->ena &= ~FMODE_READ;
 		s->dma_adc.ossfragshift = s->dma_adc.ossmaxfrags =
 		    s->dma_adc.subdivision = 0;
-		up(&s->open_sem_adc);
+		mutex_unlock(&s->open_sem_adc);
 
 		if (prog_dmabuf_adc(s)) {
 			CS_DBGOUT(CS_OPEN | CS_ERROR, 2, printk(KERN_ERR
@@ -2474,7 +2475,7 @@ static int cs4297a_open(struct inode *inode, struct file *file)
 		s->ena &= ~FMODE_WRITE;
 		s->dma_dac.ossfragshift = s->dma_dac.ossmaxfrags =
 		    s->dma_dac.subdivision = 0;
-		up(&s->open_sem_dac);
+		mutex_unlock(&s->open_sem_dac);
 
 		if (prog_dmabuf_dac(s)) {
 			CS_DBGOUT(CS_OPEN | CS_ERROR, 2, printk(KERN_ERR
@@ -2631,8 +2632,8 @@ static int __init cs4297a_init(void)
 	init_waitqueue_head(&s->open_wait);
 	init_waitqueue_head(&s->open_wait_adc);
 	init_waitqueue_head(&s->open_wait_dac);
-	init_MUTEX(&s->open_sem_adc);
-	init_MUTEX(&s->open_sem_dac);
+	mutex_init(&s->open_sem_adc);
+	mutex_init(&s->open_sem_dac);
 	spin_lock_init(&s->lock);
 
         s->irq = K_INT_SER_1;
diff --git a/sound/oss/trident.c b/sound/oss/trident.c
index a21c663e7e1..e61a454a815 100644
--- a/sound/oss/trident.c
+++ b/sound/oss/trident.c
@@ -190,7 +190,7 @@
  *
  *	Lock order (high->low)
  *		lock	-	hardware lock
- *		open_sem - 	guard opens
+ *		open_mutex - 	guard opens
  *		sem	-	guard dmabuf, write re-entry etc
  */
 
@@ -216,6 +216,8 @@
 #include <linux/pm.h>
 #include <linux/gameport.h>
 #include <linux/kernel.h>
+#include <linux/mutex.h>
+
 #include <asm/uaccess.h>
 #include <asm/io.h>
 #include <asm/dma.h>
@@ -349,7 +351,7 @@ struct trident_state {
 	unsigned chans_num;
 	unsigned long fmt_flag;
 	/* Guard against mmap/write/read races */
-	struct semaphore sem;
+	struct mutex sem;
 
 };
 
@@ -402,7 +404,7 @@ struct trident_card {
 	struct trident_card *next;
 
 	/* single open lock mechanism, only used for recording */
-	struct semaphore open_sem;
+	struct mutex open_mutex;
 
 	/* The trident has a certain amount of cross channel interaction
 	   so we use a single per card lock */
@@ -1881,7 +1883,7 @@ trident_read(struct file *file, char __user *buffer, size_t count, loff_t * ppos
 	if (!access_ok(VERIFY_WRITE, buffer, count))
 		return -EFAULT;
 
-	down(&state->sem);
+	mutex_lock(&state->sem);
 	if (!dmabuf->ready && (ret = prog_dmabuf_record(state)))
 		goto out;
 
@@ -1913,7 +1915,7 @@ trident_read(struct file *file, char __user *buffer, size_t count, loff_t * ppos
 				goto out;
 			}
 
-			up(&state->sem);
+			mutex_unlock(&state->sem);
 			/* No matter how much space left in the buffer, */ 
 			/* we have to wait until CSO == ESO/2 or CSO == ESO */ 
 			/* when address engine interrupts */
@@ -1940,7 +1942,7 @@ trident_read(struct file *file, char __user *buffer, size_t count, loff_t * ppos
 					ret = -ERESTARTSYS;
 				goto out;
 			}
-			down(&state->sem);
+			mutex_lock(&state->sem);
 			if (dmabuf->mapped) {
 				if (!ret)
 					ret = -ENXIO;
@@ -1968,7 +1970,7 @@ trident_read(struct file *file, char __user *buffer, size_t count, loff_t * ppos
 		start_adc(state);
 	}
 out:
-	up(&state->sem);
+	mutex_unlock(&state->sem);
 	return ret;
 }
 
@@ -1996,7 +1998,7 @@ trident_write(struct file *file, const char __user *buffer, size_t count, loff_t
 	 *      Guard against an mmap or ioctl while writing
 	 */
 
-	down(&state->sem);
+	mutex_lock(&state->sem);
 
 	if (dmabuf->mapped) {
 		ret = -ENXIO;
@@ -2045,7 +2047,7 @@ trident_write(struct file *file, const char __user *buffer, size_t count, loff_t
 			tmo = (dmabuf->dmasize * HZ) / (dmabuf->rate * 2);
 			tmo >>= sample_shift[dmabuf->fmt];
 			unlock_set_fmt(state);
-			up(&state->sem);
+			mutex_unlock(&state->sem);
 
 			/* There are two situations when sleep_on_timeout */ 
 			/* returns, one is when the interrupt is serviced */ 
@@ -2073,7 +2075,7 @@ trident_write(struct file *file, const char __user *buffer, size_t count, loff_t
 					ret = -ERESTARTSYS;
 				goto out_nolock;
 			}
-			down(&state->sem);
+			mutex_lock(&state->sem);
 			if (dmabuf->mapped) {
 				if (!ret)
 					ret = -ENXIO;
@@ -2131,7 +2133,7 @@ trident_write(struct file *file, const char __user *buffer, size_t count, loff_t
 		start_dac(state);
 	}
 out:
-	up(&state->sem);
+	mutex_unlock(&state->sem);
 out_nolock:
 	return ret;
 }
@@ -2152,24 +2154,24 @@ trident_poll(struct file *file, struct poll_table_struct *wait)
 	 *      prog_dmabuf events
 	 */
 
-	down(&state->sem);
+	mutex_lock(&state->sem);
 
 	if (file->f_mode & FMODE_WRITE) {
 		if (!dmabuf->ready && prog_dmabuf_playback(state)) {
-			up(&state->sem);
+			mutex_unlock(&state->sem);
 			return 0;
 		}
 		poll_wait(file, &dmabuf->wait, wait);
 	}
 	if (file->f_mode & FMODE_READ) {
 		if (!dmabuf->ready && prog_dmabuf_record(state)) {
-			up(&state->sem);
+			mutex_unlock(&state->sem);
 			return 0;
 		}
 		poll_wait(file, &dmabuf->wait, wait);
 	}
 
-	up(&state->sem);
+	mutex_unlock(&state->sem);
 
 	spin_lock_irqsave(&state->card->lock, flags);
 	trident_update_ptr(state);
@@ -2207,7 +2209,7 @@ trident_mmap(struct file *file, struct vm_area_struct *vma)
 	 *      a read or write against an mmap.
 	 */
 
-	down(&state->sem);
+	mutex_lock(&state->sem);
 
 	if (vma->vm_flags & VM_WRITE) {
 		if ((ret = prog_dmabuf_playback(state)) != 0)
@@ -2232,7 +2234,7 @@ trident_mmap(struct file *file, struct vm_area_struct *vma)
 	dmabuf->mapped = 1;
 	ret = 0;
 out:
-	up(&state->sem);
+	mutex_unlock(&state->sem);
 	return ret;
 }
 
@@ -2429,15 +2431,15 @@ trident_ioctl(struct inode *inode, struct file *file,
 							unlock_set_fmt(state);
 							break;
 						}
-						down(&state->card->open_sem);
+						mutex_lock(&state->card->open_mutex);
 						ret = ali_allocate_other_states_resources(state, 6);
 						if (ret < 0) {
-							up(&state->card->open_sem);
+							mutex_unlock(&state->card->open_mutex);
 							unlock_set_fmt(state);
 							break;
 						}
 						state->card->multi_channel_use_count++;
-						up(&state->card->open_sem);
+						mutex_unlock(&state->card->open_mutex);
 					} else
 						val = 2;	/*yield to 2-channels */
 				} else
@@ -2727,11 +2729,11 @@ trident_open(struct inode *inode, struct file *file)
 
 	/* find an available virtual channel (instance of /dev/dsp) */
 	while (card != NULL) {
-		down(&card->open_sem);
+		mutex_lock(&card->open_mutex);
 		if (file->f_mode & FMODE_READ) {
 			/* Skip opens on cards that are in 6 channel mode */
 			if (card->multi_channel_use_count > 0) {
-				up(&card->open_sem);
+				mutex_unlock(&card->open_mutex);
 				card = card->next;
 				continue;
 			}
@@ -2740,16 +2742,16 @@ trident_open(struct inode *inode, struct file *file)
 			if (card->states[i] == NULL) {
 				state = card->states[i] = kmalloc(sizeof(*state), GFP_KERNEL);
 				if (state == NULL) {
-					up(&card->open_sem);
+					mutex_unlock(&card->open_mutex);
 					return -ENOMEM;
 				}
 				memset(state, 0, sizeof(*state));
-				init_MUTEX(&state->sem);
+				mutex_init(&state->sem);
 				dmabuf = &state->dmabuf;
 				goto found_virt;
 			}
 		}
-		up(&card->open_sem);
+		mutex_unlock(&card->open_mutex);
 		card = card->next;
 	}
 	/* no more virtual channel avaiable */
@@ -2816,7 +2818,7 @@ trident_open(struct inode *inode, struct file *file)
 	}
 
 	state->open_mode |= file->f_mode & (FMODE_READ | FMODE_WRITE);
-	up(&card->open_sem);
+	mutex_unlock(&card->open_mutex);
 
 	pr_debug("trident: open virtual channel %d, hard channel %d\n",
 		 state->virt, dmabuf->channel->num);
@@ -2845,7 +2847,7 @@ trident_release(struct inode *inode, struct file *file)
 		 state->virt, dmabuf->channel->num);
 
 	/* stop DMA state machine and free DMA buffers/channels */
-	down(&card->open_sem);
+	mutex_lock(&card->open_mutex);
 
 	if (file->f_mode & FMODE_WRITE) {
 		stop_dac(state);
@@ -2878,8 +2880,8 @@ trident_release(struct inode *inode, struct file *file)
 	card->states[state->virt] = NULL;
 	kfree(state);
 
-	/* we're covered by the open_sem */
-	up(&card->open_sem);
+	/* we're covered by the open_mutex */
+	mutex_unlock(&card->open_mutex);
 
 	return 0;
 }
@@ -4405,7 +4407,7 @@ trident_probe(struct pci_dev *pci_dev, const struct pci_device_id *pci_id)
 	card->banks[BANK_B].addresses = &bank_b_addrs;
 	card->banks[BANK_B].bitmap = 0UL;
 
-	init_MUTEX(&card->open_sem);
+	mutex_init(&card->open_mutex);
 	spin_lock_init(&card->lock);
 	init_timer(&card->timer);
 
diff --git a/sound/oss/via82cxxx_audio.c b/sound/oss/via82cxxx_audio.c
index 83edda93f0b..1a921ee71ab 100644
--- a/sound/oss/via82cxxx_audio.c
+++ b/sound/oss/via82cxxx_audio.c
@@ -38,7 +38,8 @@
 #include <linux/dma-mapping.h>
 #include <asm/io.h>
 #include <asm/uaccess.h>
-#include <asm/semaphore.h>
+#include <linux/mutex.h>
+
 #include "sound_config.h"
 #include "dev_table.h"
 #include "mpu401.h"
@@ -311,8 +312,8 @@ struct via_info {
 	
 	int mixer_vol;		/* 8233/35 volume  - not yet implemented */
 
-	struct semaphore syscall_sem;
-	struct semaphore open_sem;
+	struct mutex syscall_mutex;
+	struct mutex open_mutex;
 
 	/* The 8233/8235 have 4 DX audio channels, two record and
 	   one six channel out. We bind ch_in to DX 1, ch_out to multichannel
@@ -505,10 +506,10 @@ static inline int via_syscall_down (struct via_info *card, int nonblock)
 	nonblock = 0;
 
 	if (nonblock) {
-		if (down_trylock (&card->syscall_sem))
+		if (!mutex_trylock(&card->syscall_mutex))
 			return -EAGAIN;
 	} else {
-		if (down_interruptible (&card->syscall_sem))
+		if (mutex_lock_interruptible(&card->syscall_mutex))
 			return -ERESTARTSYS;
 	}
 
@@ -1609,7 +1610,7 @@ static int via_mixer_ioctl (struct inode *inode, struct file *file, unsigned int
 #endif
 	rc = codec->mixer_ioctl(codec, cmd, arg);
 
-	up (&card->syscall_sem);
+	mutex_unlock(&card->syscall_mutex);
 
 out:
 	DPRINTK ("EXIT, returning %d\n", rc);
@@ -2228,7 +2229,7 @@ static int via_dsp_mmap(struct file *file, struct vm_area_struct *vma)
 	if (wr)
 		card->ch_out.is_mapped = 1;
 
-	up (&card->syscall_sem);
+	mutex_unlock(&card->syscall_mutex);
 	rc = 0;
 
 out:
@@ -2256,7 +2257,7 @@ handle_one_block:
 	/* Thomas Sailer:
 	 * But also to ourselves, release semaphore if we do so */
 	if (need_resched()) {
-		up(&card->syscall_sem);
+		mutex_unlock(&card->syscall_mutex);
 		schedule ();
 		ret = via_syscall_down (card, nonblock);
 		if (ret)
@@ -2286,7 +2287,7 @@ handle_one_block:
 			break;
 		}
 
-		up(&card->syscall_sem);
+		mutex_unlock(&card->syscall_mutex);
 
 		DPRINTK ("Sleeping on block %d\n", n);
 		schedule();
@@ -2402,7 +2403,7 @@ static ssize_t via_dsp_read(struct file *file, char __user *buffer, size_t count
 	rc = via_dsp_do_read (card, buffer, count, nonblock);
 
 out_up:
-	up (&card->syscall_sem);
+	mutex_unlock(&card->syscall_mutex);
 out:
 	DPRINTK ("EXIT, returning %ld\n",(long) rc);
 	return rc;
@@ -2426,7 +2427,7 @@ handle_one_block:
 	/* Thomas Sailer:
 	 * But also to ourselves, release semaphore if we do so */
 	if (need_resched()) {
-		up(&card->syscall_sem);
+		mutex_unlock(&card->syscall_mutex);
 		schedule ();
 		ret = via_syscall_down (card, nonblock);
 		if (ret)
@@ -2456,7 +2457,7 @@ handle_one_block:
 			break;
 		}
 
-		up(&card->syscall_sem);
+		mutex_unlock(&card->syscall_mutex);
 
 		DPRINTK ("Sleeping on page %d, tmp==%d, ir==%d\n", n, tmp, chan->is_record);
 		schedule();
@@ -2585,7 +2586,7 @@ static ssize_t via_dsp_write(struct file *file, const char __user *buffer, size_
 	rc = via_dsp_do_write (card, buffer, count, nonblock);
 
 out_up:
-	up (&card->syscall_sem);
+	mutex_unlock(&card->syscall_mutex);
 out:
 	DPRINTK ("EXIT, returning %ld\n",(long) rc);
 	return rc;
@@ -2634,7 +2635,7 @@ static unsigned int via_dsp_poll(struct file *file, struct poll_table_struct *wa
  *	Sleeps until all playback has been flushed to the audio
  *	hardware.
  *
- *	Locking: inside card->syscall_sem
+ *	Locking: inside card->syscall_mutex
  */
 
 static int via_dsp_drain_playback (struct via_info *card,
@@ -2692,7 +2693,7 @@ static int via_dsp_drain_playback (struct via_info *card,
 			printk (KERN_ERR "sleeping but not active\n");
 #endif
 
-		up(&card->syscall_sem);
+		mutex_unlock(&card->syscall_mutex);
 
 		DPRINTK ("sleeping, nbufs=%d\n", atomic_read (&chan->n_frags));
 		schedule();
@@ -2748,7 +2749,7 @@ out:
  *
  *	Handles SNDCTL_DSP_GETISPACE and SNDCTL_DSP_GETOSPACE.
  *
- *	Locking: inside card->syscall_sem
+ *	Locking: inside card->syscall_mutex
  */
 
 static int via_dsp_ioctl_space (struct via_info *card,
@@ -2793,7 +2794,7 @@ static int via_dsp_ioctl_space (struct via_info *card,
  *
  *	Handles SNDCTL_DSP_GETIPTR and SNDCTL_DSP_GETOPTR.
  *
- *	Locking: inside card->syscall_sem
+ *	Locking: inside card->syscall_mutex
  */
 
 static int via_dsp_ioctl_ptr (struct via_info *card,
@@ -3221,7 +3222,7 @@ static int via_dsp_ioctl (struct inode *inode, struct file *file,
 		break;
 	}
 
-	up (&card->syscall_sem);
+	mutex_unlock(&card->syscall_mutex);
 	DPRINTK ("EXIT, returning %d\n", rc);
 	return rc;
 }
@@ -3264,12 +3265,12 @@ static int via_dsp_open (struct inode *inode, struct file *file)
 
 match:
 	if (nonblock) {
-		if (down_trylock (&card->open_sem)) {
+		if (!mutex_trylock(&card->open_mutex)) {
 			DPRINTK ("EXIT, returning -EAGAIN\n");
 			return -EAGAIN;
 		}
 	} else {
-		if (down_interruptible (&card->open_sem)) {
+		if (mutex_lock_interruptible(&card->open_mutex)) {
 			DPRINTK ("EXIT, returning -ERESTARTSYS\n");
 			return -ERESTARTSYS;
 		}
@@ -3355,8 +3356,8 @@ static int via_dsp_release(struct inode *inode, struct file *file)
 		via_chan_buffer_free (card, &card->ch_in);
 	}
 
-	up (&card->syscall_sem);
-	up (&card->open_sem);
+	mutex_unlock(&card->syscall_mutex);
+	mutex_unlock(&card->open_mutex);
 
 	DPRINTK ("EXIT, returning 0\n");
 	return 0;
@@ -3414,8 +3415,8 @@ static int __devinit via_init_one (struct pci_dev *pdev, const struct pci_device
 	card->card_num = via_num_cards++;
 	spin_lock_init (&card->lock);
 	spin_lock_init (&card->ac97_lock);
-	init_MUTEX (&card->syscall_sem);
-	init_MUTEX (&card->open_sem);
+	mutex_init(&card->syscall_mutex);
+	mutex_init(&card->open_mutex);
 
 	/* we must init these now, in case the intr handler needs them */
 	via_chan_init_defaults (card, &card->ch_out);
diff --git a/sound/oss/vwsnd.c b/sound/oss/vwsnd.c
index 265423054ca..b372e88e857 100644
--- a/sound/oss/vwsnd.c
+++ b/sound/oss/vwsnd.c
@@ -94,7 +94,7 @@
  *	Open will block until the previous client has closed the
  *	device, unless O_NONBLOCK is specified.
  *
- *	The semaphore devc->io_sema serializes PCM I/O syscalls.  This
+ *	The semaphore devc->io_mutex serializes PCM I/O syscalls.  This
  *	is unnecessary in Linux 2.2, because the kernel lock
  *	serializes read, write, and ioctl globally, but it's there,
  *	ready for the brave, new post-kernel-lock world.
@@ -105,7 +105,7 @@
  *	area it owns and update its pointers.  See pcm_output() and
  *	pcm_input() for most of the gory stuff.
  *
- *	devc->mix_sema serializes all mixer ioctls.  This is also
+ *	devc->mix_mutex serializes all mixer ioctls.  This is also
  *	redundant because of the kernel lock.
  *
  *	The lowest level lock is lith->lithium_lock.  It is a
@@ -148,7 +148,8 @@
 #include <linux/smp_lock.h>
 #include <linux/wait.h>
 #include <linux/interrupt.h>
-#include <asm/semaphore.h>
+#include <linux/mutex.h>
+
 #include <asm/mach-visws/cobalt.h>
 
 #include "sound_config.h"
@@ -1447,11 +1448,11 @@ typedef enum vwsnd_port_flags {
  *
  *	port->lock protects: hwstate, flags, swb_[iu]_avail.
  *
- *	devc->io_sema protects: swstate, sw_*, swb_[iu]_idx.
+ *	devc->io_mutex protects: swstate, sw_*, swb_[iu]_idx.
  *
  *	everything else is only written by open/release or
  *	pcm_{setup,shutdown}(), which are serialized by a
- *	combination of devc->open_sema and devc->io_sema.
+ *	combination of devc->open_mutex and devc->io_mutex.
  */
 
 typedef struct vwsnd_port {
@@ -1507,9 +1508,9 @@ typedef struct vwsnd_dev {
 	int		audio_minor;	/* minor number of audio device */
 	int		mixer_minor;	/* minor number of mixer device */
 
-	struct semaphore open_sema;
-	struct semaphore io_sema;
-	struct semaphore mix_sema;
+	struct mutex open_mutex;
+	struct mutex io_mutex;
+	struct mutex mix_mutex;
 	mode_t		open_mode;
 	wait_queue_head_t open_wait;
 
@@ -1633,7 +1634,7 @@ static __inline__ unsigned int swb_inc_i(vwsnd_port_t *port, int inc)
  * mode-setting ioctls have been done, but before the first I/O is
  * done.
  *
- * Locking: called with devc->io_sema held.
+ * Locking: called with devc->io_mutex held.
  *
  * Returns 0 on success, -errno on failure.
  */
@@ -2319,9 +2320,9 @@ static ssize_t vwsnd_audio_read(struct file *file,
 	vwsnd_dev_t *devc = file->private_data;
 	ssize_t ret;
 
-	down(&devc->io_sema);
+	mutex_lock(&devc->io_mutex);
 	ret = vwsnd_audio_do_read(file, buffer, count, ppos);
-	up(&devc->io_sema);
+	mutex_unlock(&devc->io_mutex);
 	return ret;
 }
 
@@ -2394,9 +2395,9 @@ static ssize_t vwsnd_audio_write(struct file *file,
 	vwsnd_dev_t *devc = file->private_data;
 	ssize_t ret;
 
-	down(&devc->io_sema);
+	mutex_lock(&devc->io_mutex);
 	ret = vwsnd_audio_do_write(file, buffer, count, ppos);
-	up(&devc->io_sema);
+	mutex_unlock(&devc->io_mutex);
 	return ret;
 }
 
@@ -2891,9 +2892,9 @@ static int vwsnd_audio_ioctl(struct inode *inode,
 	vwsnd_dev_t *devc = (vwsnd_dev_t *) file->private_data;
 	int ret;
 
-	down(&devc->io_sema);
+	mutex_lock(&devc->io_mutex);
 	ret = vwsnd_audio_do_ioctl(inode, file, cmd, arg);
-	up(&devc->io_sema);
+	mutex_unlock(&devc->io_mutex);
 	return ret;
 }
 
@@ -2929,9 +2930,9 @@ static int vwsnd_audio_open(struct inode *inode, struct file *file)
 		return -ENODEV;
 	}
 
-	down(&devc->open_sema);
+	mutex_lock(&devc->open_mutex);
 	while (devc->open_mode & file->f_mode) {
-		up(&devc->open_sema);
+		mutex_unlock(&devc->open_mutex);
 		if (file->f_flags & O_NONBLOCK) {
 			DEC_USE_COUNT;
 			return -EBUSY;
@@ -2941,10 +2942,10 @@ static int vwsnd_audio_open(struct inode *inode, struct file *file)
 			DEC_USE_COUNT;
 			return -ERESTARTSYS;
 		}
-		down(&devc->open_sema);
+		mutex_lock(&devc->open_mutex);
 	}
 	devc->open_mode |= file->f_mode & (FMODE_READ | FMODE_WRITE);
-	up(&devc->open_sema);
+	mutex_unlock(&devc->open_mutex);
 
 	/* get default sample format from minor number. */
 
@@ -2960,7 +2961,7 @@ static int vwsnd_audio_open(struct inode *inode, struct file *file)
 
 	/* Initialize vwsnd_ports. */
 
-	down(&devc->io_sema);
+	mutex_lock(&devc->io_mutex);
 	{
 		if (file->f_mode & FMODE_READ) {
 			devc->rport.swstate        = SW_INITIAL;
@@ -2987,7 +2988,7 @@ static int vwsnd_audio_open(struct inode *inode, struct file *file)
 			devc->wport.frag_count     = 0;
 		}
 	}
-	up(&devc->io_sema);
+	mutex_unlock(&devc->io_mutex);
 
 	file->private_data = devc;
 	DBGRV();
@@ -3005,7 +3006,7 @@ static int vwsnd_audio_release(struct inode *inode, struct file *file)
 	int err = 0;
 
 	lock_kernel();
-	down(&devc->io_sema);
+	mutex_lock(&devc->io_mutex);
 	{
 		DBGEV("(inode=0x%p, file=0x%p)\n", inode, file);
 
@@ -3022,13 +3023,13 @@ static int vwsnd_audio_release(struct inode *inode, struct file *file)
 		if (wport)
 			wport->swstate = SW_OFF;
 	}
-	up(&devc->io_sema);
+	mutex_unlock(&devc->io_mutex);
 
-	down(&devc->open_sema);
+	mutex_lock(&devc->open_mutex);
 	{
 		devc->open_mode &= ~file->f_mode;
 	}
-	up(&devc->open_sema);
+	mutex_unlock(&devc->open_mutex);
 	wake_up(&devc->open_wait);
 	DEC_USE_COUNT;
 	DBGR();
@@ -3213,7 +3214,7 @@ static int vwsnd_mixer_ioctl(struct inode *ioctl,
 
 	DBGEV("(devc=0x%p, cmd=0x%x, arg=0x%lx)\n", devc, cmd, arg);
 
-	down(&devc->mix_sema);
+	mutex_lock(&devc->mix_mutex);
 	{
 		if ((cmd & ~nrmask) == MIXER_READ(0))
 			retval = mixer_read_ioctl(devc, nr, (void __user *) arg);
@@ -3222,7 +3223,7 @@ static int vwsnd_mixer_ioctl(struct inode *ioctl,
 		else
 			retval = -EINVAL;
 	}
-	up(&devc->mix_sema);
+	mutex_unlock(&devc->mix_mutex);
 	return retval;
 }
 
@@ -3376,9 +3377,9 @@ static int __init attach_vwsnd(struct address_info *hw_config)
 
 	/* Initialize as much of *devc as possible */
 
-	init_MUTEX(&devc->open_sema);
-	init_MUTEX(&devc->io_sema);
-	init_MUTEX(&devc->mix_sema);
+	mutex_init(&devc->open_mutex);
+	mutex_init(&devc->io_mutex);
+	mutex_init(&devc->mix_mutex);
 	devc->open_mode = 0;
 	spin_lock_init(&devc->rport.lock);
 	init_waitqueue_head(&devc->rport.queue);
diff --git a/sound/oss/ymfpci.c b/sound/oss/ymfpci.c
index f8bd72e46f5..bf90c124a7e 100644
--- a/sound/oss/ymfpci.c
+++ b/sound/oss/ymfpci.c
@@ -1918,10 +1918,10 @@ static int ymf_open(struct inode *inode, struct file *file)
 	if (unit == NULL)
 		return -ENODEV;
 
-	down(&unit->open_sem);
+	mutex_lock(&unit->open_mutex);
 
 	if ((state = ymf_state_alloc(unit)) == NULL) {
-		up(&unit->open_sem);
+		mutex_unlock(&unit->open_mutex);
 		return -ENOMEM;
 	}
 	list_add_tail(&state->chain, &unit->states);
@@ -1956,7 +1956,7 @@ static int ymf_open(struct inode *inode, struct file *file)
 	ymfpci_writeb(unit, YDSXGR_TIMERCTRL,
 	    (YDSXGR_TIMERCTRL_TEN|YDSXGR_TIMERCTRL_TIEN));
 #endif
-	up(&unit->open_sem);
+	mutex_unlock(&unit->open_mutex);
 
 	return nonseekable_open(inode, file);
 
@@ -1974,7 +1974,7 @@ out_nodma:
 	list_del(&state->chain);
 	kfree(state);
 
-	up(&unit->open_sem);
+	mutex_unlock(&unit->open_mutex);
 	return err;
 }
 
@@ -1987,7 +1987,7 @@ static int ymf_release(struct inode *inode, struct file *file)
 	ymfpci_writeb(unit, YDSXGR_TIMERCTRL, 0);
 #endif
 
-	down(&unit->open_sem);
+	mutex_lock(&unit->open_mutex);
 
 	/*
 	 * XXX Solve the case of O_NONBLOCK close - don't deallocate here.
@@ -2004,7 +2004,7 @@ static int ymf_release(struct inode *inode, struct file *file)
 	file->private_data = NULL;	/* Can you tell I programmed Solaris */
 	kfree(state);
 
-	up(&unit->open_sem);
+	mutex_unlock(&unit->open_mutex);
 
 	return 0;
 }
@@ -2532,7 +2532,7 @@ static int __devinit ymf_probe_one(struct pci_dev *pcidev, const struct pci_devi
 	spin_lock_init(&codec->reg_lock);
 	spin_lock_init(&codec->voice_lock);
 	spin_lock_init(&codec->ac97_lock);
-	init_MUTEX(&codec->open_sem);
+	mutex_init(&codec->open_mutex);
 	INIT_LIST_HEAD(&codec->states);
 	codec->pci = pcidev;
 
diff --git a/sound/oss/ymfpci.h b/sound/oss/ymfpci.h
index f810a100c64..ac1785f2b7e 100644
--- a/sound/oss/ymfpci.h
+++ b/sound/oss/ymfpci.h
@@ -22,6 +22,7 @@
  *
  */
 #include <linux/config.h>
+#include <linux/mutex.h>
 
 /*
  *  Direct registers
@@ -279,7 +280,7 @@ struct ymf_unit {
 
 	/* soundcore stuff */
 	int dev_audio;
-	struct semaphore open_sem;
+	struct mutex open_mutex;
 
 	struct list_head ymf_devs;
 	struct list_head states;	/* List of states for this unit */