From 1f64643aa5f5a17f1723f7ea0f17b7a3a8f632b3 Mon Sep 17 00:00:00 2001 From: Luke Browning Date: Thu, 5 Jun 2008 17:30:25 +0800 Subject: powerpc/spufs: remove class_0_dsisr from spu exception handling According to the CBEA, the SPU dsisr is not updated for class 0 exceptions. spu_stopped() is testing the dsisr that was passed to it from the class 0 exception handler, so we return a false positive here. This patch cleans up the interrupt handler and erroneous tests in spu_stopped. It also removes the fields from the csa since it is not needed to process class 0 events. Signed-off-by: Luke Browning Signed-off-by: Jeremy Kerr --- include/asm-powerpc/spu.h | 1 - include/asm-powerpc/spu_csa.h | 2 +- 2 files changed, 1 insertion(+), 2 deletions(-) (limited to 'include') diff --git a/include/asm-powerpc/spu.h b/include/asm-powerpc/spu.h index 6abead6e681..99348c1f4ca 100644 --- a/include/asm-powerpc/spu.h +++ b/include/asm-powerpc/spu.h @@ -131,7 +131,6 @@ struct spu { u64 flags; u64 class_0_pending; u64 class_0_dar; - u64 class_0_dsisr; u64 class_1_dar; u64 class_1_dsisr; size_t ls_size; diff --git a/include/asm-powerpc/spu_csa.h b/include/asm-powerpc/spu_csa.h index 129ec148d45..a40fd491250 100644 --- a/include/asm-powerpc/spu_csa.h +++ b/include/asm-powerpc/spu_csa.h @@ -254,7 +254,7 @@ struct spu_state { u64 spu_chnldata_RW[32]; u32 spu_mailbox_data[4]; u32 pu_mailbox_data[1]; - u64 class_0_dar, class_0_dsisr, class_0_pending; + u64 class_0_dar, class_0_pending; u64 class_1_dar, class_1_dsisr; unsigned long suspend_time; spinlock_t register_lock; -- cgit v1.2.3 From 598056d5af8fef1dbe8f96f5c2b641a528184e5a Mon Sep 17 00:00:00 2001 From: Nick Piggin Date: Thu, 22 May 2008 00:10:56 +1000 Subject: [POWERPC] Fix rmb to order cacheable vs. noncacheable lwsync is explicitly defined not to have any effect on the ordering of accesses to device memory, so it cannot be used for rmb(). sync appears to be the only barrier which fits the bill. Signed-off-by: Nick Piggin Acked-by: Benjamin Herrenschmidt Signed-off-by: Paul Mackerras --- include/asm-powerpc/system.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include') diff --git a/include/asm-powerpc/system.h b/include/asm-powerpc/system.h index 2b6559a6d11..5235f875b93 100644 --- a/include/asm-powerpc/system.h +++ b/include/asm-powerpc/system.h @@ -34,7 +34,7 @@ * SMP since it is only used to order updates to system memory. */ #define mb() __asm__ __volatile__ ("sync" : : : "memory") -#define rmb() __asm__ __volatile__ (__stringify(LWSYNC) : : : "memory") +#define rmb() __asm__ __volatile__ ("sync" : : : "memory") #define wmb() __asm__ __volatile__ ("sync" : : : "memory") #define read_barrier_depends() do { } while(0) -- cgit v1.2.3 From 305c73687157d677bee6f2a5dbee438d844a028c Mon Sep 17 00:00:00 2001 From: Adrian Bunk Date: Wed, 11 Jun 2008 03:47:45 +1000 Subject: [POWERPC] Build fix for drivers/macintosh/mediabay.c This fixes the following build error with CONFIG_BLK_DEV_IDE_PMAC=n: <-- snip --> ... CC drivers/macintosh/mediabay.o /home/bunk/linux/kernel-2.6/git/linux-2.6/drivers/macintosh/mediabay.c: In function 'check_media_bay': /home/bunk/linux/kernel-2.6/git/linux-2.6/drivers/macintosh/mediabay.c:428: error: 'struct media_bay_info' has no member named 'cd_index' make[3]: *** [drivers/macintosh/mediabay.o] Error 1 <-- snip --> Reported-by: Adrian Bunk Signed-off-by: Adrian Bunk Signed-off-by: Paul Mackerras --- include/asm-powerpc/mediabay.h | 12 ++++++++++-- 1 file changed, 10 insertions(+), 2 deletions(-) (limited to 'include') diff --git a/include/asm-powerpc/mediabay.h b/include/asm-powerpc/mediabay.h index df111c362a7..b2efb332580 100644 --- a/include/asm-powerpc/mediabay.h +++ b/include/asm-powerpc/mediabay.h @@ -17,8 +17,6 @@ #define MB_POWER 6 /* media bay contains a Power device (???) */ #define MB_NO 7 /* media bay contains nothing */ -int check_media_bay(struct device_node *which_bay, int what); - /* Number of bays in the machine or 0 */ extern int media_bay_count; @@ -29,6 +27,16 @@ int check_media_bay_by_base(unsigned long base, int what); /* called by IDE PMAC host driver to register IDE controller for media bay */ int media_bay_set_ide_infos(struct device_node *which_bay, unsigned long base, int irq, ide_hwif_t *hwif); + +int check_media_bay(struct device_node *which_bay, int what); + +#else + +static inline int check_media_bay(struct device_node *which_bay, int what) +{ + return -ENODEV; +} + #endif #endif /* __KERNEL__ */ -- cgit v1.2.3 From 6b7d0b2fb6da3f6182f2bd162bf02ed4aed1813f Mon Sep 17 00:00:00 2001 From: bruno randolf Date: Fri, 6 Jun 2008 16:42:03 +0200 Subject: [MIPS] Alchemy: Add au1500 reserved interrupt In the conversion done in the commits 95c4eb3ef4484ca85da5c98780d358cffd546b90 9d360ab4a7568a8d177280f651a8a772ae52b9b9 [MIPS] Alchemy: Renumber interrupts so irq_cpu can work. one reserved interrupt on au1500 was missed. this broke the au1000 ethernet driver. Signed-off-by: Bruno Randolf Signed-off-by: Ralf Baechle --- include/asm-mips/mach-au1x00/au1000.h | 1 + 1 file changed, 1 insertion(+) (limited to 'include') diff --git a/include/asm-mips/mach-au1x00/au1000.h b/include/asm-mips/mach-au1x00/au1000.h index 1b5064dac00..0d302bad449 100644 --- a/include/asm-mips/mach-au1x00/au1000.h +++ b/include/asm-mips/mach-au1x00/au1000.h @@ -615,6 +615,7 @@ enum soc_au1500_ints { AU1000_RTC_MATCH1_INT, AU1000_RTC_MATCH2_INT, AU1500_PCI_ERR_INT, + AU1500_RESERVED_INT, AU1000_USB_DEV_REQ_INT, AU1000_USB_DEV_SUS_INT, AU1000_USB_HOST_INT, -- cgit v1.2.3 From 89052bd7b393434f7c573ce6a3b88c5f143586d2 Mon Sep 17 00:00:00 2001 From: Ralf Baechle Date: Thu, 12 Jun 2008 17:26:02 +0100 Subject: [MIPS] Fix build for PNX platforms. Build error was caused by commit 351336929ccf222ae38ff0cb7a8dd5fd5c6236a0. Signed-off-by: Ralf Baechle --- include/asm-mips/pgtable-bits.h | 2 -- 1 file changed, 2 deletions(-) (limited to 'include') diff --git a/include/asm-mips/pgtable-bits.h b/include/asm-mips/pgtable-bits.h index 60e2f9338fc..51b34a48c84 100644 --- a/include/asm-mips/pgtable-bits.h +++ b/include/asm-mips/pgtable-bits.h @@ -134,6 +134,4 @@ #define _PAGE_CHG_MASK (PAGE_MASK | _PAGE_ACCESSED | _PAGE_MODIFIED | _CACHE_MASK) -#define CONF_CM_DEFAULT (PAGE_CACHABLE_DEFAULT>>_CACHE_SHIFT) - #endif /* _ASM_PGTABLE_BITS_H */ -- cgit v1.2.3 From d6c3048cad3c9eb312c070e11fdbea56498255ed Mon Sep 17 00:00:00 2001 From: Chris Dearman Date: Fri, 16 May 2008 17:29:54 -0700 Subject: [MIPS] vpe_id is required for VSMP and SMTC builds Signed-off-by: Chris Dearman Signed-off-by: Ralf Baechle --- include/asm-mips/cpu-info.h | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'include') diff --git a/include/asm-mips/cpu-info.h b/include/asm-mips/cpu-info.h index 0c5a358863f..2de73dbb2e9 100644 --- a/include/asm-mips/cpu-info.h +++ b/include/asm-mips/cpu-info.h @@ -56,7 +56,7 @@ struct cpuinfo_mips { struct cache_desc tcache; /* Tertiary/split secondary cache */ int srsets; /* Shadow register sets */ int core; /* physical core number */ -#if defined(CONFIG_MIPS_MT_SMTC) +#if defined(CONFIG_MIPS_MT_SMP) || defined(CONFIG_MIPS_MT_SMTC) /* * In the MIPS MT "SMTC" model, each TC is considered * to be a "CPU" for the purposes of scheduling, but @@ -64,7 +64,7 @@ struct cpuinfo_mips { * to all TCs within the same VPE. */ int vpe_id; /* Virtual Processor number */ -#endif /* CONFIG_MIPS_MT */ +#endif #ifdef CONFIG_MIPS_MT_SMTC int tc_id; /* Thread Context number */ #endif -- cgit v1.2.3 From 0c3bd83b0974238a5808d342663c6407512564d0 Mon Sep 17 00:00:00 2001 From: Thomas Horsten Date: Sat, 14 Jun 2008 02:32:42 +0100 Subject: [MIPS] Lasat: bring back from the dead After the common MIPS CPU interrupt controller (for irq0-7) was introduced the Lasat boards didn't get their interrupts right, so nothing worked. The old routines need to be offset by the new 8 hardware interrupts common to all MIPS CPU's. Signed-off-by: Thomas Horsten Signed-off-by: Ralf Baechle --- include/asm-mips/lasat/serial.h | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'include') diff --git a/include/asm-mips/lasat/serial.h b/include/asm-mips/lasat/serial.h index bafe68b1061..1c37d70579b 100644 --- a/include/asm-mips/lasat/serial.h +++ b/include/asm-mips/lasat/serial.h @@ -4,10 +4,10 @@ #define LASAT_BASE_BAUD_100 (7372800 / 16) #define LASAT_UART_REGS_BASE_100 0x1c8b0000 #define LASAT_UART_REGS_SHIFT_100 2 -#define LASATINT_UART_100 8 +#define LASATINT_UART_100 16 /* * LASAT 200 boards serial configuration */ #define LASAT_BASE_BAUD_200 (100000000 / 16 / 12) #define LASAT_UART_REGS_BASE_200 (Vrc5074_PHYS_BASE + 0x0300) #define LASAT_UART_REGS_SHIFT_200 3 -#define LASATINT_UART_200 13 +#define LASATINT_UART_200 21 -- cgit v1.2.3 From 938b2b14172bd098972df2a5157bfabf161c90e5 Mon Sep 17 00:00:00 2001 From: Thomas Bogendoerfer Date: Thu, 29 May 2008 22:05:07 +0200 Subject: [MIPS] Malta: Fix build errors for 64-bit kernels Fix 64-bit Malta by using CKSEG0ADDR and correct casts. Signed-off-by: Thomas Bogendoerfer Signed-off-by: Ralf Baechle --- include/asm-mips/gic.h | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'include') diff --git a/include/asm-mips/gic.h b/include/asm-mips/gic.h index 3a492f225f0..954807d9d66 100644 --- a/include/asm-mips/gic.h +++ b/include/asm-mips/gic.h @@ -24,8 +24,8 @@ #define MSK(n) ((1 << (n)) - 1) #define REG32(addr) (*(volatile unsigned int *) (addr)) -#define REG(base, offs) REG32((unsigned int)(base) + offs##_##OFS) -#define REGP(base, phys) REG32((unsigned int)(base) + (phys)) +#define REG(base, offs) REG32((unsigned long)(base) + offs##_##OFS) +#define REGP(base, phys) REG32((unsigned long)(base) + (phys)) /* Accessors */ #define GIC_REG(segment, offset) \ -- cgit v1.2.3 From 732a675a6303156d1a197dc780b0712bd4b49d46 Mon Sep 17 00:00:00 2001 From: Jack Steiner Date: Sat, 14 Jun 2008 07:57:25 -0500 Subject: [IA64] Fix CONFIG_IA64_SGI_UV build error Fix build error in CONFIG_IA64_SGI_UV config. (GENERIC builds are ok). Signed-off-by: Jack Steiner Signed-off-by: Tony Luck --- include/asm-ia64/sn/simulator.h | 7 ++++++- 1 file changed, 6 insertions(+), 1 deletion(-) (limited to 'include') diff --git a/include/asm-ia64/sn/simulator.h b/include/asm-ia64/sn/simulator.h index c3fd3eb2576..c2611f6cfe3 100644 --- a/include/asm-ia64/sn/simulator.h +++ b/include/asm-ia64/sn/simulator.h @@ -8,7 +8,7 @@ #ifndef _ASM_IA64_SN_SIMULATOR_H #define _ASM_IA64_SN_SIMULATOR_H - +#if defined(CONFIG_IA64_GENERIC) || defined(CONFIG_IA64_SGI_SN2) || defined(CONFIG_IA64_SGI_UV) #define SNMAGIC 0xaeeeeeee8badbeefL #define IS_MEDUSA() ({long sn; asm("mov %0=cpuid[%1]" : "=r"(sn) : "r"(2)); sn == SNMAGIC;}) @@ -16,5 +16,10 @@ #define IS_RUNNING_ON_SIMULATOR() (sn_prom_type) #define IS_RUNNING_ON_FAKE_PROM() (sn_prom_type == 2) extern int sn_prom_type; /* 0=hardware, 1=medusa/realprom, 2=medusa/fakeprom */ +#else +#define IS_MEDUSA() 0 +#define SIMULATOR_SLEEP() +#define IS_RUNNING_ON_SIMULATOR() 0 +#endif #endif /* _ASM_IA64_SN_SIMULATOR_H */ -- cgit v1.2.3 From 9a8ea36967afad617d9b0930b6fe7592b9ed9772 Mon Sep 17 00:00:00 2001 From: David Woodhouse Date: Mon, 16 Jun 2008 12:18:24 +0100 Subject: Remove #ifdef CONFIG_ARCH_SUPPORTS_AOUT from This file is only included where it makes sense now, so there's no need for the CONFIG_ARCH_SUPPORTS_AOUT conditional -- and that conditional is bad, because we want to export to userspace. Signed-off-by: David Woodhouse Acked-by: Peter Korsgaard Signed-off-by: Linus Torvalds --- include/linux/a.out.h | 8 -------- 1 file changed, 8 deletions(-) (limited to 'include') diff --git a/include/linux/a.out.h b/include/linux/a.out.h index 208f4e8ed30..e86dfca4458 100644 --- a/include/linux/a.out.h +++ b/include/linux/a.out.h @@ -1,8 +1,6 @@ #ifndef __A_OUT_GNU_H__ #define __A_OUT_GNU_H__ -#ifdef CONFIG_ARCH_SUPPORTS_AOUT - #define __GNU_EXEC_MACROS__ #ifndef __STRUCT_EXEC_OVERRIDE__ @@ -277,10 +275,4 @@ struct relocation_info #endif /* no N_RELOCATION_INFO_DECLARED. */ #endif /*__ASSEMBLY__ */ -#else /* CONFIG_ARCH_SUPPORTS_AOUT */ -#ifndef __ASSEMBLY__ -struct exec { -}; -#endif -#endif /* CONFIG_ARCH_SUPPORTS_AOUT */ #endif /* __A_OUT_GNU_H__ */ -- cgit v1.2.3 From e53d6a152793a38aa334d6f7a4850642ae45cedc Mon Sep 17 00:00:00 2001 From: David Woodhouse Date: Mon, 16 Jun 2008 12:24:17 +0100 Subject: Export to userspace again. This seems to have been removed accidentally in commit ed7b1889da256977574663689b598d88950bbd23 ("Unexport asm/page.h"), but wasn't supposed to have been -- the original patch at http://lkml.org/lkml/2007/10/30/144 just moved it from $(header-y) to $(unifdef-y) Signed-off-by: David Woodhouse Acked-by: Peter Korsgaard Signed-off-by: Linus Torvalds --- include/linux/Kbuild | 3 +++ 1 file changed, 3 insertions(+) (limited to 'include') diff --git a/include/linux/Kbuild b/include/linux/Kbuild index 93b98856007..b6fbb2573e8 100644 --- a/include/linux/Kbuild +++ b/include/linux/Kbuild @@ -166,6 +166,9 @@ unifdef-y += acct.h unifdef-y += adb.h unifdef-y += adfs_fs.h unifdef-y += agpgart.h +ifeq ($(wildcard include/asm-$(SRCARCH)/a.out.h),include/asm-$(SRCARCH)/a.out.h) +unifdef-y += a.out.h +endif unifdef-y += apm_bios.h unifdef-y += atalk.h unifdef-y += atmdev.h -- cgit v1.2.3 From 2b4743bd6be9fedaa560f8c6dc3997e9ec21b99b Mon Sep 17 00:00:00 2001 From: YOSHIFUJI Hideaki Date: Mon, 16 Jun 2008 16:48:20 -0700 Subject: ipv6 sit: Avoid extra need for compat layer in PRL management. We've introduced extra need of compat layer for ip_tunnel_prl{} for PRL (Potential Router List) management. Though compat_ioctl is still missing in ipv4/ipv6, let's make the interface more straight-forward and eliminate extra need for nasty compat layer anyway since the interface is new for 2.6.26. Signed-off-by: YOSHIFUJI Hideaki Signed-off-by: David S. Miller --- include/linux/if_tunnel.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include') diff --git a/include/linux/if_tunnel.h b/include/linux/if_tunnel.h index f1fbe9c930d..d4efe401470 100644 --- a/include/linux/if_tunnel.h +++ b/include/linux/if_tunnel.h @@ -41,7 +41,7 @@ struct ip_tunnel_prl { __u16 __reserved; __u32 datalen; __u32 __reserved2; - void __user *data; + /* data follows */ }; /* PRL flags */ -- cgit v1.2.3 From 68b80f11380889996aa7eadba29dbbb5c29a5864 Mon Sep 17 00:00:00 2001 From: Patrick McHardy Date: Tue, 17 Jun 2008 15:51:47 -0700 Subject: netfilter: nf_nat: fix RCU races Fix three ct_extend/NAT extension related races: - When cleaning up the extension area and removing it from the bysource hash, the nat->ct pointer must not be set to NULL since it may still be used in a RCU read side - When replacing a NAT extension area in the bysource hash, the nat->ct pointer must be assigned before performing the replacement - When reallocating extension storage in ct_extend, the old memory must not be freed immediately since it may still be used by a RCU read side Possibly fixes https://bugzilla.redhat.com/show_bug.cgi?id=449315 and/or http://bugzilla.kernel.org/show_bug.cgi?id=10875 Signed-off-by: Patrick McHardy Signed-off-by: David S. Miller --- include/net/netfilter/nf_conntrack_extend.h | 1 + 1 file changed, 1 insertion(+) (limited to 'include') diff --git a/include/net/netfilter/nf_conntrack_extend.h b/include/net/netfilter/nf_conntrack_extend.h index f736e842977..f80c0ed6d87 100644 --- a/include/net/netfilter/nf_conntrack_extend.h +++ b/include/net/netfilter/nf_conntrack_extend.h @@ -15,6 +15,7 @@ enum nf_ct_ext_id /* Extensions: optional stuff which isn't permanently in struct. */ struct nf_ct_ext { + struct rcu_head rcu; u8 offset[NF_CT_EXT_NUM]; u8 len; char data[0]; -- cgit v1.2.3 From c72580129209aaa509ace81c1f2ee1caa9c9774b Mon Sep 17 00:00:00 2001 From: Joe Perches Date: Wed, 26 Mar 2008 14:10:02 -0700 Subject: drivers/char/agp - use bool Use boolean in AGP instead of having own TRUE/FALSE -- Signed-off-by: Joe Perches Signed-off-by: Dave Airlie --- include/linux/agp_backend.h | 14 +++----------- 1 file changed, 3 insertions(+), 11 deletions(-) (limited to 'include') diff --git a/include/linux/agp_backend.h b/include/linux/agp_backend.h index 661d90d6cf7..09b4478ffac 100644 --- a/include/linux/agp_backend.h +++ b/include/linux/agp_backend.h @@ -30,14 +30,6 @@ #ifndef _AGP_BACKEND_H #define _AGP_BACKEND_H 1 -#ifndef TRUE -#define TRUE 1 -#endif - -#ifndef FALSE -#define FALSE 0 -#endif - enum chipset_type { NOT_SUPPORTED, SUPPORTED, @@ -57,7 +49,7 @@ struct agp_kern_info { size_t aper_size; int max_memory; /* In pages */ int current_memory; - int cant_use_aperture; + bool cant_use_aperture; unsigned long page_mask; struct vm_operations_struct *vm_ops; }; @@ -83,8 +75,8 @@ struct agp_memory { off_t pg_start; u32 type; u32 physical; - u8 is_bound; - u8 is_flushed; + bool is_bound; + bool is_flushed; u8 vmalloc_flag; }; -- cgit v1.2.3 From 9516b030b484fc99cf24213caf88df01f99248dd Mon Sep 17 00:00:00 2001 From: Dave Airlie Date: Thu, 19 Jun 2008 10:42:17 +1000 Subject: agp: more boolean conversions. Signed-off-by: Dave Airlie --- include/linux/agp_backend.h | 2 +- include/linux/agpgart.h | 4 ++-- 2 files changed, 3 insertions(+), 3 deletions(-) (limited to 'include') diff --git a/include/linux/agp_backend.h b/include/linux/agp_backend.h index 09b4478ffac..972b12bcfb3 100644 --- a/include/linux/agp_backend.h +++ b/include/linux/agp_backend.h @@ -77,7 +77,7 @@ struct agp_memory { u32 physical; bool is_bound; bool is_flushed; - u8 vmalloc_flag; + bool vmalloc_flag; }; #define AGP_NORMAL_MEMORY 0 diff --git a/include/linux/agpgart.h b/include/linux/agpgart.h index 62aef589eb9..c8fdb6e658e 100644 --- a/include/linux/agpgart.h +++ b/include/linux/agpgart.h @@ -206,8 +206,8 @@ struct agp_front_data { struct agp_controller *current_controller; struct agp_controller *controllers; struct agp_file_private *file_priv_list; - u8 used_by_controller; - u8 backend_acquired; + bool used_by_controller; + bool backend_acquired; }; #endif /* __KERNEL__ */ -- cgit v1.2.3 From ad524d46f36bbc32033bb72ba42958f12bf49b06 Mon Sep 17 00:00:00 2001 From: Jeremy Fitzhardinge Date: Fri, 6 Jun 2008 10:21:39 +0100 Subject: x86: set PAE PHYSICAL_MASK_SHIFT to 44 bits. When a 64-bit x86 processor runs in 32-bit PAE mode, a pte can potentially have the same number of physical address bits as the 64-bit host ("Enhanced Legacy PAE Paging"). This means, in theory, we could have up to 52 bits of physical address in a pte. The 32-bit kernel uses a 32-bit unsigned long to represent a pfn. This means that it can only represent physical addresses up to 32+12=44 bits wide. Rather than widening pfns everywhere, just set 2^44 as the Linux x86_32-PAE architectural limit for physical address size. This is a bugfix for two cases: 1. running a 32-bit PAE kernel on a machine with more than 64GB RAM. 2. running a 32-bit PAE Xen guest on a host machine with more than 64GB RAM In both cases, a pte could need to have more than 36 bits of physical, and masking it to 36-bits will cause fairly severe havoc. Signed-off-by: Jeremy Fitzhardinge Cc: Jan Beulich Cc: Signed-off-by: Ingo Molnar --- include/asm-x86/page_32.h | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) (limited to 'include') diff --git a/include/asm-x86/page_32.h b/include/asm-x86/page_32.h index 424e82f8ae2..ccf0ba3c3ab 100644 --- a/include/asm-x86/page_32.h +++ b/include/asm-x86/page_32.h @@ -14,7 +14,8 @@ #define __PAGE_OFFSET _AC(CONFIG_PAGE_OFFSET, UL) #ifdef CONFIG_X86_PAE -#define __PHYSICAL_MASK_SHIFT 36 +/* 44=32+12, the limit we can fit into an unsigned long pfn */ +#define __PHYSICAL_MASK_SHIFT 44 #define __VIRTUAL_MASK_SHIFT 32 #define PAGETABLE_LEVELS 3 -- cgit v1.2.3 From ec64b6c8763c83899908fdd62746435c19211686 Mon Sep 17 00:00:00 2001 From: Sonic Zhang Date: Thu, 19 Jun 2008 17:07:15 +0800 Subject: Blackfin arch: fix typo error in bf548 serial header file Signed-off-by: Sonic Zhang Signed-off-by: Bryan Wu --- include/asm-blackfin/mach-bf548/bfin_serial_5xx.h | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'include') diff --git a/include/asm-blackfin/mach-bf548/bfin_serial_5xx.h b/include/asm-blackfin/mach-bf548/bfin_serial_5xx.h index 59b4ad4e5b4..aabe5b92f11 100644 --- a/include/asm-blackfin/mach-bf548/bfin_serial_5xx.h +++ b/include/asm-blackfin/mach-bf548/bfin_serial_5xx.h @@ -187,7 +187,7 @@ static void bfin_serial_hw_init(struct bfin_serial_port *uart) #ifdef CONFIG_BFIN_UART1_CTSRTS peripheral_request(P_UART1_RTS, DRIVER_NAME); - peripheral_request(P_UART1_CTS DRIVER_NAME); + peripheral_request(P_UART1_CTS, DRIVER_NAME); #endif #endif @@ -202,7 +202,7 @@ static void bfin_serial_hw_init(struct bfin_serial_port *uart) #ifdef CONFIG_BFIN_UART3_CTSRTS peripheral_request(P_UART3_RTS, DRIVER_NAME); - peripheral_request(P_UART3_CTS DRIVER_NAME); + peripheral_request(P_UART3_CTS, DRIVER_NAME); #endif #endif SSYNC(); -- cgit v1.2.3 From f30ac0ce34f32bb998ac87e37b251374de03e603 Mon Sep 17 00:00:00 2001 From: Sonic Zhang Date: Thu, 19 Jun 2008 17:46:39 +0800 Subject: Blackfin Serial Driver: Use timer to poll CTS PIN instead of workqueue. This allows other threads to run when the serial driver polls the CTS PIN in a loop. Signed-off-by: Sonic Zhang Signed-off-by: Bryan Wu --- include/asm-blackfin/mach-bf527/bfin_serial_5xx.h | 2 +- include/asm-blackfin/mach-bf533/bfin_serial_5xx.h | 2 +- include/asm-blackfin/mach-bf537/bfin_serial_5xx.h | 2 +- include/asm-blackfin/mach-bf548/bfin_serial_5xx.h | 2 +- include/asm-blackfin/mach-bf561/bfin_serial_5xx.h | 2 +- 5 files changed, 5 insertions(+), 5 deletions(-) (limited to 'include') diff --git a/include/asm-blackfin/mach-bf527/bfin_serial_5xx.h b/include/asm-blackfin/mach-bf527/bfin_serial_5xx.h index 96bd09e31e3..2526b6ed6fa 100644 --- a/include/asm-blackfin/mach-bf527/bfin_serial_5xx.h +++ b/include/asm-blackfin/mach-bf527/bfin_serial_5xx.h @@ -96,7 +96,7 @@ struct bfin_serial_port { struct work_struct tx_dma_workqueue; #endif #ifdef CONFIG_SERIAL_BFIN_CTSRTS - struct work_struct cts_workqueue; + struct timer_list cts_timer; int cts_pin; int rts_pin; #endif diff --git a/include/asm-blackfin/mach-bf533/bfin_serial_5xx.h b/include/asm-blackfin/mach-bf533/bfin_serial_5xx.h index e924569ad1d..ebf592b59aa 100644 --- a/include/asm-blackfin/mach-bf533/bfin_serial_5xx.h +++ b/include/asm-blackfin/mach-bf533/bfin_serial_5xx.h @@ -88,7 +88,7 @@ struct bfin_serial_port { # endif #endif #ifdef CONFIG_SERIAL_BFIN_CTSRTS - struct work_struct cts_workqueue; + struct timer_list cts_timer; int cts_pin; int rts_pin; #endif diff --git a/include/asm-blackfin/mach-bf537/bfin_serial_5xx.h b/include/asm-blackfin/mach-bf537/bfin_serial_5xx.h index 41d7b6490bb..1bf56ffa22f 100644 --- a/include/asm-blackfin/mach-bf537/bfin_serial_5xx.h +++ b/include/asm-blackfin/mach-bf537/bfin_serial_5xx.h @@ -96,7 +96,7 @@ struct bfin_serial_port { struct work_struct tx_dma_workqueue; #endif #ifdef CONFIG_SERIAL_BFIN_CTSRTS - struct work_struct cts_workqueue; + struct timer_list cts_timer; int cts_pin; int rts_pin; #endif diff --git a/include/asm-blackfin/mach-bf548/bfin_serial_5xx.h b/include/asm-blackfin/mach-bf548/bfin_serial_5xx.h index aabe5b92f11..5e29446a8e0 100644 --- a/include/asm-blackfin/mach-bf548/bfin_serial_5xx.h +++ b/include/asm-blackfin/mach-bf548/bfin_serial_5xx.h @@ -99,7 +99,7 @@ struct bfin_serial_port { struct work_struct tx_dma_workqueue; #endif #ifdef CONFIG_SERIAL_BFIN_CTSRTS - struct work_struct cts_workqueue; + struct timer_list cts_timer; int cts_pin; int rts_pin; #endif diff --git a/include/asm-blackfin/mach-bf561/bfin_serial_5xx.h b/include/asm-blackfin/mach-bf561/bfin_serial_5xx.h index 30d90b580f1..8aa02780e64 100644 --- a/include/asm-blackfin/mach-bf561/bfin_serial_5xx.h +++ b/include/asm-blackfin/mach-bf561/bfin_serial_5xx.h @@ -88,7 +88,7 @@ struct bfin_serial_port { # endif #endif #ifdef CONFIG_SERIAL_BFIN_CTSRTS - struct work_struct cts_workqueue; + struct timer_list cts_timer; int cts_pin; int rts_pin; #endif -- cgit v1.2.3 From ffe6e1da86d21d7855495b5a772c93f050258f6e Mon Sep 17 00:00:00 2001 From: Jordan Crouse Date: Wed, 18 Jun 2008 11:34:38 -0600 Subject: x86, geode: add a VSA2 ID for General Software General Software writes their own VSA2 module for their version of the Geode BIOS, which returns a different ID then the standard VSA2. This was causing the framebuffer driver to break for most GSW boards. Signed-off-by: Jordan Crouse Cc: tglx@linutronix.de Cc: linux-geode@lists.infradead.org Signed-off-by: Ingo Molnar --- include/asm-x86/geode.h | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'include') diff --git a/include/asm-x86/geode.h b/include/asm-x86/geode.h index 6e6458853a3..bb06027fc83 100644 --- a/include/asm-x86/geode.h +++ b/include/asm-x86/geode.h @@ -112,8 +112,8 @@ extern int geode_get_dev_base(unsigned int dev); #define VSA_VR_UNLOCK 0xFC53 /* unlock virtual register */ #define VSA_VR_SIGNATURE 0x0003 #define VSA_VR_MEM_SIZE 0x0200 -#define VSA_SIG 0x4132 /* signature is ascii 'VSA2' */ - +#define AMD_VSA_SIG 0x4132 /* signature is ascii 'VSA2' */ +#define GSW_VSA_SIG 0x534d /* General Software signature */ /* GPIO */ #define GPIO_OUTPUT_VAL 0x00 -- cgit v1.2.3 From f630e43a215a3129d0c1173cae0bce6ea4855cf7 Mon Sep 17 00:00:00 2001 From: YOSHIFUJI Hideaki Date: Thu, 19 Jun 2008 16:33:57 -0700 Subject: ipv6: Drop packets for loopback address from outside of the box. [ Based upon original report and patch by Karsten Keil. Karsten has verified that this fixes the TAHI test case "ICMPv6 test v6LC.5.1.2 Part F". -DaveM ] Signed-off-by: YOSHIFUJI Hideaki Signed-off-by: David S. Miller --- include/net/ipv6.h | 6 ++++++ 1 file changed, 6 insertions(+) (limited to 'include') diff --git a/include/net/ipv6.h b/include/net/ipv6.h index e0a612bc9c4..f422f7218e1 100644 --- a/include/net/ipv6.h +++ b/include/net/ipv6.h @@ -367,6 +367,12 @@ static inline int ipv6_addr_any(const struct in6_addr *a) a->s6_addr32[2] | a->s6_addr32[3] ) == 0); } +static inline int ipv6_addr_loopback(const struct in6_addr *a) +{ + return ((a->s6_addr32[0] | a->s6_addr32[1] | + a->s6_addr32[2] | (a->s6_addr32[3] ^ htonl(1))) == 0); +} + static inline int ipv6_addr_v4mapped(const struct in6_addr *a) { return ((a->s6_addr32[0] | a->s6_addr32[1] | -- cgit v1.2.3 From 9267b4b3880d00dc2dab90f1d817c856939114f7 Mon Sep 17 00:00:00 2001 From: Ivan Kokshaysky Date: Sat, 21 Jun 2008 03:25:39 +0400 Subject: alpha: fix module load failures on smp (bug #10926) To calculate addresses of locally defined variables, GCC uses 32-bit displacement from the GP. Which doesn't work for per cpu variables in modules, as an offset to the kernel per cpu area is way above 4G. The workaround is to force allocation of a GOT entry for per cpu variable using ldq instruction with a 'literal' relocation. I had to use custom asm/percpu.h, as a required argument magic doesn't work with asm-generic/percpu.h macros. Signed-off-by: Ivan Kokshaysky Signed-off-by: Linus Torvalds --- include/asm-alpha/percpu.h | 72 +++++++++++++++++++++++++++++++++++++++++++++- 1 file changed, 71 insertions(+), 1 deletion(-) (limited to 'include') diff --git a/include/asm-alpha/percpu.h b/include/asm-alpha/percpu.h index 48348fe34c1..82e8a94b4b2 100644 --- a/include/asm-alpha/percpu.h +++ b/include/asm-alpha/percpu.h @@ -1,6 +1,76 @@ #ifndef __ALPHA_PERCPU_H #define __ALPHA_PERCPU_H +#include +#include -#include +/* + * Determine the real variable name from the name visible in the + * kernel sources. + */ +#define per_cpu_var(var) per_cpu__##var + +#ifdef CONFIG_SMP + +/* + * per_cpu_offset() is the offset that has to be added to a + * percpu variable to get to the instance for a certain processor. + */ +extern unsigned long __per_cpu_offset[NR_CPUS]; + +#define per_cpu_offset(x) (__per_cpu_offset[x]) + +#define __my_cpu_offset per_cpu_offset(raw_smp_processor_id()) +#ifdef CONFIG_DEBUG_PREEMPT +#define my_cpu_offset per_cpu_offset(smp_processor_id()) +#else +#define my_cpu_offset __my_cpu_offset +#endif + +#ifndef MODULE +#define SHIFT_PERCPU_PTR(var, offset) RELOC_HIDE(&per_cpu_var(var), (offset)) +#define PER_CPU_ATTRIBUTES +#else +/* + * To calculate addresses of locally defined variables, GCC uses 32-bit + * displacement from the GP. Which doesn't work for per cpu variables in + * modules, as an offset to the kernel per cpu area is way above 4G. + * + * This forces allocation of a GOT entry for per cpu variable using + * ldq instruction with a 'literal' relocation. + */ +#define SHIFT_PERCPU_PTR(var, offset) ({ \ + extern int simple_identifier_##var(void); \ + unsigned long __ptr, tmp_gp; \ + asm ( "br %1, 1f \n\ + 1: ldgp %1, 0(%1) \n\ + ldq %0, per_cpu__" #var"(%1)\t!literal" \ + : "=&r"(__ptr), "=&r"(tmp_gp)); \ + (typeof(&per_cpu_var(var)))(__ptr + (offset)); }) + +#define PER_CPU_ATTRIBUTES __used + +#endif /* MODULE */ + +/* + * A percpu variable may point to a discarded regions. The following are + * established ways to produce a usable pointer from the percpu variable + * offset. + */ +#define per_cpu(var, cpu) \ + (*SHIFT_PERCPU_PTR(var, per_cpu_offset(cpu))) +#define __get_cpu_var(var) \ + (*SHIFT_PERCPU_PTR(var, my_cpu_offset)) +#define __raw_get_cpu_var(var) \ + (*SHIFT_PERCPU_PTR(var, __my_cpu_offset)) + +#else /* ! SMP */ + +#define per_cpu(var, cpu) (*((void)(cpu), &per_cpu_var(var))) +#define __get_cpu_var(var) per_cpu_var(var) +#define __raw_get_cpu_var(var) per_cpu_var(var) + +#endif /* SMP */ + +#define DECLARE_PER_CPU(type, name) extern __typeof__(type) per_cpu_var(name) #endif /* __ALPHA_PERCPU_H */ -- cgit v1.2.3 From d559d4a24a3fed75bd890abcc1f95cd8d8dad6e1 Mon Sep 17 00:00:00 2001 From: Ivan Kokshaysky Date: Sat, 21 Jun 2008 03:28:31 +0400 Subject: alpha: fix compile failures with gcc-4.3 (bug #10438) Vast majority of these build failures are gcc-4.3 warnings about static functions and objects being referenced from non-static (read: "extern inline") functions, in conjunction with our -Werror. We cannot just convert "extern inline" to "static inline", as people keep suggesting all the time, because "extern inline" logic is crucial for generic kernel build. So - just make sure that all callees of critical "extern inline" functions are also "extern inline"; - use "static inline", wherever it's possible. traps.c: work around gcc-4.3 being too smart about array bounds-checking. TODO: add "gnu_inline" attribute to all our "extern inline" functions to ensure desired behaviour with future compilers. Signed-off-by: Ivan Kokshaysky Signed-off-by: Linus Torvalds --- include/asm-alpha/core_mcpcia.h | 2 +- include/asm-alpha/core_t2.h | 14 +++++++------- include/asm-alpha/io.h | 6 +++--- include/asm-alpha/mmu_context.h | 6 +++--- include/asm-alpha/system.h | 10 +++++----- include/asm-alpha/vga.h | 6 +++--- 6 files changed, 22 insertions(+), 22 deletions(-) (limited to 'include') diff --git a/include/asm-alpha/core_mcpcia.h b/include/asm-alpha/core_mcpcia.h index 525b4f6a7ac..acf55b48347 100644 --- a/include/asm-alpha/core_mcpcia.h +++ b/include/asm-alpha/core_mcpcia.h @@ -261,7 +261,7 @@ struct el_MCPCIA_uncorrected_frame_mcheck { } #endif -static inline int __mcpcia_is_mmio(unsigned long addr) +extern inline int __mcpcia_is_mmio(unsigned long addr) { return (addr & 0x80000000UL) == 0; } diff --git a/include/asm-alpha/core_t2.h b/include/asm-alpha/core_t2.h index 90e6b5d6c21..46bfff58f67 100644 --- a/include/asm-alpha/core_t2.h +++ b/include/asm-alpha/core_t2.h @@ -356,13 +356,13 @@ struct el_t2_frame_corrected { #define vip volatile int * #define vuip volatile unsigned int * -static inline u8 t2_inb(unsigned long addr) +extern inline u8 t2_inb(unsigned long addr) { long result = *(vip) ((addr << 5) + T2_IO + 0x00); return __kernel_extbl(result, addr & 3); } -static inline void t2_outb(u8 b, unsigned long addr) +extern inline void t2_outb(u8 b, unsigned long addr) { unsigned long w; @@ -371,13 +371,13 @@ static inline void t2_outb(u8 b, unsigned long addr) mb(); } -static inline u16 t2_inw(unsigned long addr) +extern inline u16 t2_inw(unsigned long addr) { long result = *(vip) ((addr << 5) + T2_IO + 0x08); return __kernel_extwl(result, addr & 3); } -static inline void t2_outw(u16 b, unsigned long addr) +extern inline void t2_outw(u16 b, unsigned long addr) { unsigned long w; @@ -386,12 +386,12 @@ static inline void t2_outw(u16 b, unsigned long addr) mb(); } -static inline u32 t2_inl(unsigned long addr) +extern inline u32 t2_inl(unsigned long addr) { return *(vuip) ((addr << 5) + T2_IO + 0x18); } -static inline void t2_outl(u32 b, unsigned long addr) +extern inline void t2_outl(u32 b, unsigned long addr) { *(vuip) ((addr << 5) + T2_IO + 0x18) = b; mb(); @@ -435,7 +435,7 @@ static inline void t2_outl(u32 b, unsigned long addr) set_hae(msb); \ } -static DEFINE_SPINLOCK(t2_hae_lock); +extern spinlock_t t2_hae_lock; /* * NOTE: take T2_DENSE_MEM off in each readX/writeX routine, since diff --git a/include/asm-alpha/io.h b/include/asm-alpha/io.h index 38f18cf18c9..e971ab000f9 100644 --- a/include/asm-alpha/io.h +++ b/include/asm-alpha/io.h @@ -35,7 +35,7 @@ * register not being up-to-date with respect to the hardware * value. */ -static inline void __set_hae(unsigned long new_hae) +extern inline void __set_hae(unsigned long new_hae) { unsigned long flags; local_irq_save(flags); @@ -49,7 +49,7 @@ static inline void __set_hae(unsigned long new_hae) local_irq_restore(flags); } -static inline void set_hae(unsigned long new_hae) +extern inline void set_hae(unsigned long new_hae) { if (new_hae != alpha_mv.hae_cache) __set_hae(new_hae); @@ -176,7 +176,7 @@ REMAP2(u64, writeq, volatile) #undef REMAP1 #undef REMAP2 -static inline void __iomem *generic_ioportmap(unsigned long a) +extern inline void __iomem *generic_ioportmap(unsigned long a) { return alpha_mv.mv_ioportmap(a); } diff --git a/include/asm-alpha/mmu_context.h b/include/asm-alpha/mmu_context.h index 6a5be1f7deb..86c08a02d23 100644 --- a/include/asm-alpha/mmu_context.h +++ b/include/asm-alpha/mmu_context.h @@ -23,7 +23,7 @@ #endif -extern inline unsigned long +static inline unsigned long __reload_thread(struct pcb_struct *pcb) { register unsigned long a0 __asm__("$16"); @@ -114,7 +114,7 @@ extern unsigned long last_asn; #define __MMU_EXTERN_INLINE #endif -static inline unsigned long +extern inline unsigned long __get_new_mm_context(struct mm_struct *mm, long cpu) { unsigned long asn = cpu_last_asn(cpu); @@ -226,7 +226,7 @@ ev4_activate_mm(struct mm_struct *prev_mm, struct mm_struct *next_mm) # endif #endif -extern inline int +static inline int init_new_context(struct task_struct *tsk, struct mm_struct *mm) { int i; diff --git a/include/asm-alpha/system.h b/include/asm-alpha/system.h index ed221d6408f..afe20fa58c9 100644 --- a/include/asm-alpha/system.h +++ b/include/asm-alpha/system.h @@ -184,7 +184,7 @@ enum amask_enum { __amask; }) #define __CALL_PAL_R0(NAME, TYPE) \ -static inline TYPE NAME(void) \ +extern inline TYPE NAME(void) \ { \ register TYPE __r0 __asm__("$0"); \ __asm__ __volatile__( \ @@ -196,7 +196,7 @@ static inline TYPE NAME(void) \ } #define __CALL_PAL_W1(NAME, TYPE0) \ -static inline void NAME(TYPE0 arg0) \ +extern inline void NAME(TYPE0 arg0) \ { \ register TYPE0 __r16 __asm__("$16") = arg0; \ __asm__ __volatile__( \ @@ -207,7 +207,7 @@ static inline void NAME(TYPE0 arg0) \ } #define __CALL_PAL_W2(NAME, TYPE0, TYPE1) \ -static inline void NAME(TYPE0 arg0, TYPE1 arg1) \ +extern inline void NAME(TYPE0 arg0, TYPE1 arg1) \ { \ register TYPE0 __r16 __asm__("$16") = arg0; \ register TYPE1 __r17 __asm__("$17") = arg1; \ @@ -219,7 +219,7 @@ static inline void NAME(TYPE0 arg0, TYPE1 arg1) \ } #define __CALL_PAL_RW1(NAME, RTYPE, TYPE0) \ -static inline RTYPE NAME(TYPE0 arg0) \ +extern inline RTYPE NAME(TYPE0 arg0) \ { \ register RTYPE __r0 __asm__("$0"); \ register TYPE0 __r16 __asm__("$16") = arg0; \ @@ -232,7 +232,7 @@ static inline RTYPE NAME(TYPE0 arg0) \ } #define __CALL_PAL_RW2(NAME, RTYPE, TYPE0, TYPE1) \ -static inline RTYPE NAME(TYPE0 arg0, TYPE1 arg1) \ +extern inline RTYPE NAME(TYPE0 arg0, TYPE1 arg1) \ { \ register RTYPE __r0 __asm__("$0"); \ register TYPE0 __r16 __asm__("$16") = arg0; \ diff --git a/include/asm-alpha/vga.h b/include/asm-alpha/vga.h index e8df1e7aae6..c00106bac52 100644 --- a/include/asm-alpha/vga.h +++ b/include/asm-alpha/vga.h @@ -13,7 +13,7 @@ #define VT_BUF_HAVE_MEMSETW #define VT_BUF_HAVE_MEMCPYW -extern inline void scr_writew(u16 val, volatile u16 *addr) +static inline void scr_writew(u16 val, volatile u16 *addr) { if (__is_ioaddr(addr)) __raw_writew(val, (volatile u16 __iomem *) addr); @@ -21,7 +21,7 @@ extern inline void scr_writew(u16 val, volatile u16 *addr) *addr = val; } -extern inline u16 scr_readw(volatile const u16 *addr) +static inline u16 scr_readw(volatile const u16 *addr) { if (__is_ioaddr(addr)) return __raw_readw((volatile const u16 __iomem *) addr); @@ -29,7 +29,7 @@ extern inline u16 scr_readw(volatile const u16 *addr) return *addr; } -extern inline void scr_memsetw(u16 *s, u16 c, unsigned int count) +static inline void scr_memsetw(u16 *s, u16 c, unsigned int count) { if (__is_ioaddr(s)) memsetw_io((u16 __iomem *) s, c, count); -- cgit v1.2.3 From b9f75f45a6b46a0ab4eb0857d437a0845871f314 Mon Sep 17 00:00:00 2001 From: "Eric W. Biederman" Date: Fri, 20 Jun 2008 22:16:51 -0700 Subject: netns: Don't receive new packets in a dead network namespace. Alexey Dobriyan writes: > Subject: ICMP sockets destruction vs ICMP packets oops > After icmp_sk_exit() nuked ICMP sockets, we get an interrupt. > icmp_reply() wants ICMP socket. > > Steps to reproduce: > > launch shell in new netns > move real NIC to netns > setup routing > ping -i 0 > exit from shell > > BUG: unable to handle kernel NULL pointer dereference at 0000000000000000 > IP: [] icmp_sk+0x17/0x30 > PGD 17f3cd067 PUD 17f3ce067 PMD 0 > Oops: 0000 [1] PREEMPT SMP DEBUG_PAGEALLOC > CPU 0 > Modules linked in: usblp usbcore > Pid: 0, comm: swapper Not tainted 2.6.26-rc6-netns-ct #4 > RIP: 0010:[] [] icmp_sk+0x17/0x30 > RSP: 0018:ffffffff8057fc30 EFLAGS: 00010286 > RAX: 0000000000000000 RBX: 0000000000000000 RCX: ffff81017c7db900 > RDX: 0000000000000034 RSI: ffff81017c7db900 RDI: ffff81017dc41800 > RBP: ffffffff8057fc40 R08: 0000000000000001 R09: 000000000000a815 > R10: 0000000000000000 R11: 0000000000000001 R12: ffffffff8057fd28 > R13: ffffffff8057fd00 R14: ffff81017c7db938 R15: ffff81017dc41800 > FS: 0000000000000000(0000) GS:ffffffff80525000(0000) knlGS:0000000000000000 > CS: 0010 DS: 0018 ES: 0018 CR0: 000000008005003b > CR2: 0000000000000000 CR3: 000000017fcda000 CR4: 00000000000006e0 > DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 > DR3: 0000000000000000 DR6: 00000000ffff0ff0 DR7: 0000000000000400 > Process swapper (pid: 0, threadinfo ffffffff8053a000, task ffffffff804fa4a0) > Stack: 0000000000000000 ffff81017c7db900 ffffffff8057fcf0 ffffffff803fcfe4 > ffffffff804faa38 0000000000000246 0000000000005a40 0000000000000246 > 000000000001ffff ffff81017dd68dc0 0000000000005a40 0000000055342436 > Call Trace: > [] icmp_reply+0x44/0x1e0 > [] ? ip_route_input+0x23a/0x1360 > [] icmp_echo+0x65/0x70 > [] icmp_rcv+0x180/0x1b0 > [] ip_local_deliver+0xf4/0x1f0 > [] ip_rcv+0x33b/0x650 > [] netif_receive_skb+0x27a/0x340 > [] process_backlog+0x9d/0x100 > [] net_rx_action+0x18d/0x250 > [] __do_softirq+0x75/0x100 > [] call_softirq+0x1c/0x30 > [] do_softirq+0x65/0xa0 > [] irq_exit+0x97/0xa0 > [] do_IRQ+0xa8/0x130 > [] ? mwait_idle+0x0/0x60 > [] ret_from_intr+0x0/0xf > [] ? mwait_idle+0x4c/0x60 > [] ? mwait_idle+0x43/0x60 > [] ? cpu_idle+0x57/0xa0 > [] ? rest_init+0x70/0x80 > Code: 10 5b 41 5c 41 5d 41 5e c9 c3 66 2e 0f 1f 84 00 00 00 00 00 55 48 89 e5 53 > 48 83 ec 08 48 8b 9f 78 01 00 00 e8 2b c7 f1 ff 89 c0 <48> 8b 04 c3 48 83 c4 08 > 5b c9 c3 66 66 66 66 66 2e 0f 1f 84 00 > RIP [] icmp_sk+0x17/0x30 > RSP > CR2: 0000000000000000 > ---[ end trace ea161157b76b33e8 ]--- > Kernel panic - not syncing: Aiee, killing interrupt handler! Receiving packets while we are cleaning up a network namespace is a racy proposition. It is possible when the packet arrives that we have removed some but not all of the state we need to fully process it. We have the choice of either playing wack-a-mole with the cleanup routines or simply dropping packets when we don't have a network namespace to handle them. Since the check looks inexpensive in netif_receive_skb let's just drop the incoming packets. Signed-off-by: Eric W. Biederman Signed-off-by: David S. Miller --- include/net/net_namespace.h | 11 +++++++++++ 1 file changed, 11 insertions(+) (limited to 'include') diff --git a/include/net/net_namespace.h b/include/net/net_namespace.h index aa540e6be50..d9dd0f70729 100644 --- a/include/net/net_namespace.h +++ b/include/net/net_namespace.h @@ -95,6 +95,11 @@ extern struct list_head net_namespace_list; #ifdef CONFIG_NET_NS extern void __put_net(struct net *net); +static inline int net_alive(struct net *net) +{ + return net && atomic_read(&net->count); +} + static inline struct net *get_net(struct net *net) { atomic_inc(&net->count); @@ -125,6 +130,12 @@ int net_eq(const struct net *net1, const struct net *net2) return net1 == net2; } #else + +static inline int net_alive(struct net *net) +{ + return 1; +} + static inline struct net *get_net(struct net *net) { return net; -- cgit v1.2.3 From 71c2742f5e6348d76ee62085cf0a13e5eff0f00e Mon Sep 17 00:00:00 2001 From: Bernhard Walle Date: Sat, 21 Jun 2008 19:01:02 +0200 Subject: Add return value to reserve_bootmem_node() This patch changes the function reserve_bootmem_node() from void to int, returning -ENOMEM if the allocation fails. This fixes a build problem on x86 with CONFIG_KEXEC=y and CONFIG_NEED_MULTIPLE_NODES=y Signed-off-by: Bernhard Walle Reported-by: Adrian Bunk Signed-off-by: Linus Torvalds --- include/linux/bootmem.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include') diff --git a/include/linux/bootmem.h b/include/linux/bootmem.h index 6a5dbdc8a7d..686895bacd9 100644 --- a/include/linux/bootmem.h +++ b/include/linux/bootmem.h @@ -94,7 +94,7 @@ extern unsigned long init_bootmem_node(pg_data_t *pgdat, unsigned long freepfn, unsigned long startpfn, unsigned long endpfn); -extern void reserve_bootmem_node(pg_data_t *pgdat, +extern int reserve_bootmem_node(pg_data_t *pgdat, unsigned long physaddr, unsigned long size, int flags); -- cgit v1.2.3 From 36c7343b4ecac2432430f5393314f1bdc2c219a5 Mon Sep 17 00:00:00 2001 From: Alan Cox Date: Mon, 23 Jun 2008 12:06:52 +0100 Subject: tty_driver: Update required method documentation Some of the requirement rules are now more relaxed. Also correct a contradiction in the previous update Signed-off-by: Alan Cox Signed-off-by: Linus Torvalds --- include/linux/tty_driver.h | 5 ++--- 1 file changed, 2 insertions(+), 3 deletions(-) (limited to 'include') diff --git a/include/linux/tty_driver.h b/include/linux/tty_driver.h index 59f1c0bd8f9..d2a00358676 100644 --- a/include/linux/tty_driver.h +++ b/include/linux/tty_driver.h @@ -27,8 +27,7 @@ * This routine is called by the kernel to write a series of * characters to the tty device. The characters may come from * user space or kernel space. This routine will return the - * number of characters actually accepted for writing. This - * routine is mandatory. + * number of characters actually accepted for writing. * * Optional: Required for writable devices. * @@ -134,7 +133,7 @@ * This routine notifies the tty driver that it should hangup the * tty device. * - * Required: + * Optional: * * void (*break_ctl)(struct tty_stuct *tty, int state); * -- cgit v1.2.3 From 72c6e251ed84b3a9cdfde6711191155c47bb2b9c Mon Sep 17 00:00:00 2001 From: Thorsten Kranzkowski Date: Mon, 23 Jun 2008 20:57:22 +0000 Subject: alpha: fix compile error in arch/alpha/mm/init.c MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Commit 9267b4b3880d00dc2dab90f1d817c856939114f7 ("alpha: fix module load failures on smp (bug #10926)") causes a regression for my ev4 uniprocessor build: CC arch/alpha/mm/init.o /export/data/repositories/linux-2.6/arch/alpha/mm/init.c:34: error: expected ‘=’, ‘,’, ‘;’, ‘asm’ or ‘__attribute__’ before ‘typeof’ make[2]: *** [arch/alpha/mm/init.o] Error 1 make[1]: *** [arch/alpha/mm] Error 2 make: *** [sub-make] Error 2 This fixes it for me (compile and boot tested): Signed-off-by: Thorsten Kranzkowski Acked-by: Ivan Kokshaysky Signed-off-by: Linus Torvalds --- include/asm-alpha/percpu.h | 2 ++ 1 file changed, 2 insertions(+) (limited to 'include') diff --git a/include/asm-alpha/percpu.h b/include/asm-alpha/percpu.h index 82e8a94b4b2..3495e8e00d7 100644 --- a/include/asm-alpha/percpu.h +++ b/include/asm-alpha/percpu.h @@ -69,6 +69,8 @@ extern unsigned long __per_cpu_offset[NR_CPUS]; #define __get_cpu_var(var) per_cpu_var(var) #define __raw_get_cpu_var(var) per_cpu_var(var) +#define PER_CPU_ATTRIBUTES + #endif /* SMP */ #define DECLARE_PER_CPU(type, name) extern __typeof__(type) per_cpu_var(name) -- cgit v1.2.3 From 06e05645661211b9eaadaf6344c335d2e80f0ba2 Mon Sep 17 00:00:00 2001 From: Marcelo Tosatti Date: Fri, 6 Jun 2008 16:37:36 -0300 Subject: KVM: close timer injection race window in __vcpu_run If a timer fires after kvm_inject_pending_timer_irqs() but before local_irq_disable() the code will enter guest mode and only inject such timer interrupt the next time an unrelated event causes an exit. It would be simpler if the timer->pending irq conversion could be done with IRQ's disabled, so that the above problem cannot happen. For now introduce a new vcpu requests bit to cancel guest entry. Signed-off-by: Marcelo Tosatti Signed-off-by: Avi Kivity --- include/linux/kvm_host.h | 1 + 1 file changed, 1 insertion(+) (limited to 'include') diff --git a/include/linux/kvm_host.h b/include/linux/kvm_host.h index 092b1b25291..de9d1df4bba 100644 --- a/include/linux/kvm_host.h +++ b/include/linux/kvm_host.h @@ -33,6 +33,7 @@ #define KVM_REQ_REPORT_TPR_ACCESS 2 #define KVM_REQ_MMU_RELOAD 3 #define KVM_REQ_TRIPLE_FAULT 4 +#define KVM_REQ_PENDING_TIMER 5 struct kvm_vcpu; extern struct kmem_cache *kvm_vcpu_cache; -- cgit v1.2.3 From 28499143933f19b28008a556ed59255d6009391a Mon Sep 17 00:00:00 2001 From: Jeremy Fitzhardinge Date: Fri, 9 May 2008 12:05:57 +0100 Subject: xen: remove support for non-PAE 32-bit Non-PAE operation has been deprecated in Xen for a while, and is rarely tested or used. xen-unstable has now officially dropped non-PAE support. Since Xen/pvops' non-PAE support has also been broken for a while, we may as well completely drop it altogether. Signed-off-by: Jeremy Fitzhardinge Signed-off-by: Ingo Molnar Signed-off-by: Thomas Gleixner Signed-off-by: Ingo Molnar --- include/asm-x86/xen/page.h | 4 ---- 1 file changed, 4 deletions(-) (limited to 'include') diff --git a/include/asm-x86/xen/page.h b/include/asm-x86/xen/page.h index baf3a4dce28..e11f24038b1 100644 --- a/include/asm-x86/xen/page.h +++ b/include/asm-x86/xen/page.h @@ -150,13 +150,9 @@ static inline pte_t __pte_ma(pteval_t x) return (pte_t) { .pte = x }; } -#ifdef CONFIG_X86_PAE #define pmd_val_ma(v) ((v).pmd) #define pud_val_ma(v) ((v).pgd.pgd) #define __pmd_ma(x) ((pmd_t) { (x) } ) -#else /* !X86_PAE */ -#define pmd_val_ma(v) ((v).pud.pgd.pgd) -#endif /* CONFIG_X86_PAE */ #define pgd_val_ma(x) ((x).pgd) -- cgit v1.2.3 From 7af192c954017499ec163bc9dbaaee2e593d7ef2 Mon Sep 17 00:00:00 2001 From: Gerd Hoffmann Date: Tue, 3 Jun 2008 16:17:29 +0200 Subject: x86: Add structs and functions for paravirt clocksource This patch adds structs for the paravirt clocksource ABI used by both xen and kvm (pvclock-abi.h). It also adds some helper functions to read system time and wall clock time from a paravirtual clocksource (pvclock.[ch]). They are based on the xen code. They are enabled using CONFIG_PARAVIRT_CLOCK. Subsequent patches of this series will put the code in use. Signed-off-by: Gerd Hoffmann Acked-by: Jeremy Fitzhardinge Signed-off-by: Avi Kivity --- include/asm-x86/pvclock-abi.h | 42 ++++++++++++++++++++++++++++++++++++++++++ include/asm-x86/pvclock.h | 13 +++++++++++++ 2 files changed, 55 insertions(+) create mode 100644 include/asm-x86/pvclock-abi.h create mode 100644 include/asm-x86/pvclock.h (limited to 'include') diff --git a/include/asm-x86/pvclock-abi.h b/include/asm-x86/pvclock-abi.h new file mode 100644 index 00000000000..6857f840b24 --- /dev/null +++ b/include/asm-x86/pvclock-abi.h @@ -0,0 +1,42 @@ +#ifndef _ASM_X86_PVCLOCK_ABI_H_ +#define _ASM_X86_PVCLOCK_ABI_H_ +#ifndef __ASSEMBLY__ + +/* + * These structs MUST NOT be changed. + * They are the ABI between hypervisor and guest OS. + * Both Xen and KVM are using this. + * + * pvclock_vcpu_time_info holds the system time and the tsc timestamp + * of the last update. So the guest can use the tsc delta to get a + * more precise system time. There is one per virtual cpu. + * + * pvclock_wall_clock references the point in time when the system + * time was zero (usually boot time), thus the guest calculates the + * current wall clock by adding the system time. + * + * Protocol for the "version" fields is: hypervisor raises it (making + * it uneven) before it starts updating the fields and raises it again + * (making it even) when it is done. Thus the guest can make sure the + * time values it got are consistent by checking the version before + * and after reading them. + */ + +struct pvclock_vcpu_time_info { + u32 version; + u32 pad0; + u64 tsc_timestamp; + u64 system_time; + u32 tsc_to_system_mul; + s8 tsc_shift; + u8 pad[3]; +} __attribute__((__packed__)); /* 32 bytes */ + +struct pvclock_wall_clock { + u32 version; + u32 sec; + u32 nsec; +} __attribute__((__packed__)); + +#endif /* __ASSEMBLY__ */ +#endif /* _ASM_X86_PVCLOCK_ABI_H_ */ diff --git a/include/asm-x86/pvclock.h b/include/asm-x86/pvclock.h new file mode 100644 index 00000000000..85b1bba8e0a --- /dev/null +++ b/include/asm-x86/pvclock.h @@ -0,0 +1,13 @@ +#ifndef _ASM_X86_PVCLOCK_H_ +#define _ASM_X86_PVCLOCK_H_ + +#include +#include + +/* some helper functions for xen and kvm pv clock sources */ +cycle_t pvclock_clocksource_read(struct pvclock_vcpu_time_info *src); +void pvclock_read_wallclock(struct pvclock_wall_clock *wall, + struct pvclock_vcpu_time_info *vcpu, + struct timespec *ts); + +#endif /* _ASM_X86_PVCLOCK_H_ */ -- cgit v1.2.3 From 1c7b67f7576c4ca2a344379a4a29eec8fe8e7935 Mon Sep 17 00:00:00 2001 From: Gerd Hoffmann Date: Tue, 3 Jun 2008 16:17:30 +0200 Subject: x86: Make xen use the paravirt clocksource structs and functions This patch updates the xen guest to use the pvclock structs and helper functions. Signed-off-by: Gerd Hoffmann Acked-by: Jeremy Fitzhardinge Signed-off-by: Avi Kivity --- include/xen/interface/xen.h | 7 +++---- 1 file changed, 3 insertions(+), 4 deletions(-) (limited to 'include') diff --git a/include/xen/interface/xen.h b/include/xen/interface/xen.h index 9b018da48cf..819a0331cda 100644 --- a/include/xen/interface/xen.h +++ b/include/xen/interface/xen.h @@ -10,6 +10,7 @@ #define __XEN_PUBLIC_XEN_H__ #include +#include /* * XEN "SYSTEM CALLS" (a.k.a. HYPERCALLS). @@ -336,7 +337,7 @@ struct vcpu_info { uint8_t evtchn_upcall_mask; unsigned long evtchn_pending_sel; struct arch_vcpu_info arch; - struct vcpu_time_info time; + struct pvclock_vcpu_time_info time; }; /* 64 bytes (x86) */ /* @@ -384,9 +385,7 @@ struct shared_info { * Wallclock time: updated only by control software. Guests should base * their gettimeofday() syscall on this wallclock-base value. */ - uint32_t wc_version; /* Version counter: see vcpu_time_info_t. */ - uint32_t wc_sec; /* Secs 00:00:00 UTC, Jan 1, 1970. */ - uint32_t wc_nsec; /* Nsecs 00:00:00 UTC, Jan 1, 1970. */ + struct pvclock_wall_clock wc; struct arch_shared_info arch; -- cgit v1.2.3 From 50d0a0f987b83a8dadb1134d834e35ec410392b5 Mon Sep 17 00:00:00 2001 From: Gerd Hoffmann Date: Tue, 3 Jun 2008 16:17:31 +0200 Subject: KVM: Make kvm host use the paravirt clocksource structs This patch updates the kvm host code to use the pvclock structs. It also makes the paravirt clock compatible with Xen. Signed-off-by: Gerd Hoffmann Signed-off-by: Avi Kivity --- include/asm-x86/kvm_host.h | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) (limited to 'include') diff --git a/include/asm-x86/kvm_host.h b/include/asm-x86/kvm_host.h index 1d8cd01fa51..844f2a89afb 100644 --- a/include/asm-x86/kvm_host.h +++ b/include/asm-x86/kvm_host.h @@ -18,6 +18,7 @@ #include #include +#include #include #define KVM_MAX_VCPUS 16 @@ -282,7 +283,8 @@ struct kvm_vcpu_arch { struct x86_emulate_ctxt emulate_ctxt; gpa_t time; - struct kvm_vcpu_time_info hv_clock; + struct pvclock_vcpu_time_info hv_clock; + unsigned int hv_clock_tsc_khz; unsigned int time_offset; struct page *time_page; }; -- cgit v1.2.3 From 6b1ed9086592fd4b066daae222751bb6757ca5eb Mon Sep 17 00:00:00 2001 From: Gerd Hoffmann Date: Tue, 3 Jun 2008 16:17:33 +0200 Subject: KVM: Remove now unused structs from kvm_para.h The kvm_* structs are obsoleted by the pvclock_* ones. Now all users have been switched over and the old structs can be dropped. Signed-off-by: Gerd Hoffmann Signed-off-by: Avi Kivity --- include/asm-x86/kvm_para.h | 18 ------------------ 1 file changed, 18 deletions(-) (limited to 'include') diff --git a/include/asm-x86/kvm_para.h b/include/asm-x86/kvm_para.h index 50984594207..bfd9900742b 100644 --- a/include/asm-x86/kvm_para.h +++ b/include/asm-x86/kvm_para.h @@ -48,24 +48,6 @@ struct kvm_mmu_op_release_pt { #ifdef __KERNEL__ #include -/* xen binary-compatible interface. See xen headers for details */ -struct kvm_vcpu_time_info { - uint32_t version; - uint32_t pad0; - uint64_t tsc_timestamp; - uint64_t system_time; - uint32_t tsc_to_system_mul; - int8_t tsc_shift; - int8_t pad[3]; -} __attribute__((__packed__)); /* 32 bytes */ - -struct kvm_wall_clock { - uint32_t wc_version; - uint32_t wc_sec; - uint32_t wc_nsec; -} __attribute__((__packed__)); - - extern void kvmclock_init(void); -- cgit v1.2.3