aboutsummaryrefslogtreecommitdiff
AgeCommit message (Collapse)Author
2009-09-16writeback: separate starting of sync vs opportunistic writebackJens Axboe
bdi_start_writeback() is currently split into two paths, one for WB_SYNC_NONE and one for WB_SYNC_ALL. Add bdi_sync_writeback() for WB_SYNC_ALL writeback and let bdi_start_writeback() handle only WB_SYNC_NONE. Push down the writeback_control allocation and only accept the parameters that make sense for each function. This cleans up the API considerably. Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
2009-09-16writeback: inline allocation failure handling in bdi_alloc_queue_work()Jens Axboe
This gets rid of work == NULL in bdi_queue_work() and puts the OOM handling where it belongs. Acked-by: Jan Kara <jack@suse.cz> Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
2009-09-16writeback: use RCU to protect bdi_listJens Axboe
Now that bdi_writeback_all() no longer handles integrity writeback, it doesn't have to block anymore. This means that we can switch bdi_list reader side protection to RCU. Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
2009-09-16writeback: only use bdi_writeback_all() for WB_SYNC_NONE writeoutJens Axboe
Data integrity writeback must use bdi_start_writeback() and ensure that wbc->sb and wbc->bdi are set. Acked-by: Jan Kara <jack@suse.cz> Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
2009-09-16fs: Assign bdi in super_blockJens Axboe
We do this automatically in get_sb_bdev() from the set_bdev_super() callback. Filesystems that have their own private backing_dev_info must assign that in ->fill_super(). Note that ->s_bdi assignment is required for proper writeback! Acked-by: Christoph Hellwig <hch@infradead.org> Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
2009-09-16writeback: make wb_writeback() take an argument structureJens Axboe
We need to be able to pass in range_cyclic as well, so instead of growing yet another argument, split the arguments into a struct wb_writeback_args structure that we can use internally. Also makes it easier to just copy all members to an on-stack struct, since we can't access work after clearing the pending bit. Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
2009-09-16writeback: merely wakeup flusher thread if work allocation fails for ↵Christoph Hellwig
WB_SYNC_NONE Since it's an opportunistic writeback and not a data integrity action, don't punt to blocking writeback. Just wakeup the thread and it will flush old data. Acked-by: Jan Kara <jack@suse.cz> Signed-off-by: Christoph Hellwig <hch@infradead.org> Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
2009-09-16writeback: get rid of wbc->for_writepagesJens Axboe
It's only set, it's never checked. Kill it. Acked-by: Jan Kara <jack@suse.cz> Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
2009-09-16fs: remove bdev->bd_inode_backing_dev_infoJens Axboe
It has been unused since it was introduced in: commit 520808bf20e90fdbdb320264ba7dd5cf9d47dcac Author: Andrew Morton <akpm@osdl.org> Date: Fri May 21 00:46:17 2004 -0700 [PATCH] block device layer: separate backing_dev_info infrastructure So lets just kill it. Acked-by: Jan Kara <jack@suse.cz> Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
2009-09-16amd64_edac: check NB MCE bank enable on the current node properlyBorislav Petkov
The old code was using smp_call_function_many which skips the current cpu if it is in the supplied cpumask. Switch to the rdmsr_on_cpus() interface which takes care of that. In addition, add get_cpus_on_this_dct_cpumask helper which computes a cpumask of all the cores on a node and thus on a DCT. Signed-off-by: Borislav Petkov <borislav.petkov@amd.com>
2009-09-16amd64_edac: Rewrite unganged mode code of f10_early_channel_countWan Wei
Simplify the procedure by checking if there is any DIMM in each channel. This patch will fix the bugs such as when there is no DIMMs under certain node, two DIMMs in the same channel, and only one DIMM in each channel of the node. Borislav: minor fixups Signed-off-by: Wan Wei <wanwei@mail.dawning.com.cn> Signed-off-by: Borislav Petkov <borislav.petkov@amd.com>
2009-09-16amd64_edac: cleanup amd64_check_ecc_enabledBorislav Petkov
Simplify code flow and make sure return value is always valid since further driver init depends on it. Carve out long warning string and make code more readable. Shorten some names, while at it. There should be no functional change resulting from this patch. Signed-off-by: Borislav Petkov <borislav.petkov@amd.com>
2009-09-16x86, EDAC: Provide function to return NodeId of a CPUAndreas Herrmann
Signed-off-by: Andreas Herrmann <andreas.herrmann3@amd.com> Signed-off-by: Borislav Petkov <borislav.petkov@amd.com> Acked-by: H. Peter Anvin <hpa@zytor.com>
2009-09-16amd64_edac: build driver only on AMD hardwareIngo Molnar
-tip testing found the following build failure (config attached): drivers/built-in.o: In function `amd64_check': amd64_edac.c:(.text+0x3e9491): undefined reference to `amd_decode_nb_mce' drivers/built-in.o: In function `amd64_init_2nd_stage': amd64_edac.c:(.text+0x3e9b46): undefined reference to `amd_report_gart_errors' amd64_edac.c:(.text+0x3e9b55): undefined reference to `amd_register_ecc_decoder' drivers/built-in.o: In function `amd64_nbea_store': amd64_edac_dbg.c:(.text+0x3ea22e): undefined reference to `amd_decode_nb_mce' drivers/built-in.o: In function `amd64_remove_one_instance': amd64_edac.c:(.devexit.text+0x3eea): undefined reference to `amd_report_gart_errors' amd64_edac.c:(.devexit.text+0x3ef6): undefined reference to `amd_unregister_ecc_decoder' the AMD EDAC code has a dependency on CONFIG_CPU_SUP_AMD facilities. The patch below solves the problem here. Signed-off-by: Ingo Molnar <mingo@elte.hu> Signed-off-by: Borislav Petkov <borislav.petkov@amd.com>
2009-09-15Merge branch 'master' of git://oss.sgi.com/xfs/xfs into for-linusAlex Elder
Conflicts: fs/xfs/linux-2.6/xfs_lrw.c
2009-09-16m68knommu: Make PAGE_SIZE available to assembly files.Tim Abbott
Signed-off-by: Tim Abbott <tabbott@ksplice.com> Signed-off-by: Greg Ungerer <gerg@uclinux.org>
2009-09-16m68knommu: fix ColdFire definition of CLOCK_TICK_RATEPhilippe De Muyter
The good definition of CLOCK_TICK_RATE for coldfires has been lost in the merge of m68k and m68knommu include files. Restore it. Culprit : commit ebafc17468d58bd903c886175ca84a4edc69ae1d Signed-off-by: Philippe De Muyter <phdm@macqel.be> Signed-off-by: Greg Ungerer <gerg@uclinux.org>
2009-09-16m68knommu: set multi-function pins for ethernet when enabledGreg Ungerer
The ethernet pins on the 532x ColdFire CPU family are multi-function pins. We need to enable them as ethernet pins when using the FEC ethernet driver. Bug report, and older patch, from timothee@manaud.net. Signed-off-by: Greg Ungerer <gerg@uclinux.org>
2009-09-16m68knommu: remove special interrupt handling code for ne2k supportGreg Ungerer
The improved interrupt support for ColdFire CPU cores means we no longer need all the interrupt setup and ack hacks to support the NE2000 driver on ColdFire platforms. Remove all that code. Signed-off-by: Greg Ungerer <gerg@uclinux.org>
2009-09-16m68knommu: relax IO_SPACE_LIMIT settingGreg Ungerer
There is really no limit to the addresses which can be used by the in*() and out*() family of IO space calls in m68k non-MMU environments. So don't impose an artificial address limit, allow the full 32bit range. Signed-off-by: Greg Ungerer <gerg@uclinux.org>
2009-09-16m68knommu: remove ColdFire direct interrupt register accessGreg Ungerer
Now that the ColdFire 5272 has full interrupt controller functionality we can remove all the interrupt masking and acking code from the FEC ethernet driver. Signed-off-by: Greg Ungerer <gerg@uclinux.org>
2009-09-16m68knommu: create a speciailized ColdFire 5272 interrupt controllerGreg Ungerer
The ColdFire 5272 CPU has a very different interrupt controller than any of the other ColdFire parts. It needs its own controller code to correctly setup and ack interrupts. Signed-off-by: Greg Ungerer <gerg@uclinux.org>
2009-09-16m68knommu: add support for second interrupt controller of ColdFire 5249Greg Ungerer
The ColdFire 5249 CPU has a second (compleletly different) interrupt controller. It is the only ColdFire CPU that has this type. It controlls GPIO interrupts amongst a number of interrupts from other internal peripherals. Add support code for it. Signed-off-by: Greg Ungerer <gerg@uclinux.org>
2009-09-16m68knommu: clean up old ColdFire timer irq setupGreg Ungerer
The recent changes to the old ColdFire interrupt controller code means we no longer need to manually unmask the timer interrupt. That is now done in the interrupt controller code proper. Signed-off-by: Greg Ungerer <gerg@uclinux.org>
2009-09-16m68knommu: map ColdFire interrupts to correct masking bitsGreg Ungerer
The older simple ColdFire interrupt controller has no one-to-one mapping of interrupt numbers to bits in the interrupt mask register. Create a mapping array that each ColdFire CPU type can populate with its available interrupts and the bits that each use in the interrupt mask register. Signed-off-by: Greg Ungerer <gerg@uclinux.org>
2009-09-16m68knommu: clean up ColdFire 532x CPU timer setupGreg Ungerer
The newer ColdFire 532x family of CPU's uses the old timer, but has a newer interrupt controller. It doesn't need the special timer setup that was required when using the older interrupt controller. Remove the dead timer irq and level setting code, and define the hard coded vector. Signed-off-by: Greg Ungerer <gerg@uclinux.org>
2009-09-16m68knommu: simplify ColdFire "timers" clock initializationGreg Ungerer
The ColdFire "timers" clock setup can be simplified. There is really no need for the flexible per-platform setup code. The clock interrupt can be hard defined per CPU platform (in CPU include files). This makes the actual timer code simpler. Signed-off-by: Greg Ungerer <gerg@uclinux.org>
2009-09-16m68knommu: support code to mask external interrupts on old ColdFire CPU'sGreg Ungerer
The external interrupts used on the old Coldfire parts with the old style interrupt controller can be properly mask/unmasked in the interrupt handling code. Signed-off-by: Greg Ungerer <gerg@uclinux.org>
2009-09-16m68knommu: merge old ColdFire interrupt controller masking macrosGreg Ungerer
Currently the code that supports setting the old style ColdFire interrupt controller mask registers is macros in the include files of each of the CPU types. Merge all these into a set of real masking functions in the old Coldfire interrupt controller code proper. All the macros are basically the same (excepting a register size difference on really early parts). Signed-off-by: Greg Ungerer <gerg@uclinux.org>
2009-09-16m68knommu: remove duplicate ColdFire mcf_autovector() codeGreg Ungerer
Each of the ColdFire CPU platform code that used the old style interrupt controller had its own copy of the mcf_autovector() function. They are all the same, remove them all and create a single function in the common coldfire/intc.c code. Signed-off-by: Greg Ungerer <gerg@uclinux.org>
2009-09-16m68knommu: move ColdFire INTC definitions to new include fileGreg Ungerer
Create an mcfintc.h include file with the definitions for the old style ColdFire interrupt controller. They are only needed on CPU's that use this old controller - so isolate them on their own. Signed-off-by: Greg Ungerer <gerg@uclinux.org>
2009-09-16m68knommu: mask off all interrupts in ColdFire intc-simr controllerGreg Ungerer
The ColdFire intc-simr interrupt controller should mask off all interrupt sources at init time. Doing it here instead of separately in each platform setup. Signed-off-by: Greg Ungerer <gerg@uclinux.org>
2009-09-16m68knommu: remove timer device interrupt setup for ColdFire 532xGreg Ungerer
With fully implemented interrupt controller code we don't need to do the custom interrupt setup for the timer device of the ColdFire 532x. Remove that code. Signed-off-by: Greg Ungerer <gerg@uclinux.org>
2009-09-16m68knommu: remove interrupt masking from ColdFire pit timerGreg Ungerer
With proper interrupt controller code in place there is no need for devices like the timers to have custom interrupt masking code. Remove it (and the defines that go along with it). Signed-off-by: Greg Ungerer <gerg@uclinux.org>
2009-09-16m68knommu: remove unecessary interrupt level setting in ColdFire 520x setupGreg Ungerer
The new code for the interrupt controller in the ColdFire 520x takes care of all the interrupt controller setup. No manual config of the level registers (ICR) is required by the platform device setup code. So remove it. Signed-off-by: Greg Ungerer <gerg@uclinux.org>
2009-09-16m68knommu: remove the common interrupt controller structureGreg Ungerer
Each different m68knommu CPU interrupt controller type has its own interrupt controller data structures now. Remove the old, and now not used, common irq structs and init code from here. Signed-off-by: Greg Ungerer <gerg@uclinux.org>
2009-09-16m68knommu: complete interrupt controller code for the 68360 CPUGreg Ungerer
Define the interrupt controller structures along with the interrupt controller code for the 68360 CPU. This brings the interrupt setup and control into one place for this CPU family. Signed-off-by: Greg Ungerer <gerg@uclinux.org>
2009-09-16m68knommu: complete interrupt controller code for the 68328 CPU'sGreg Ungerer
Define the interrupt controller structures along with the interrupt controller code for the 68328 CPU family. This brings the interrupt setup and control into one place for this CPU family. Signed-off-by: Greg Ungerer <gerg@uclinux.org>
2009-09-16m68knommu: use common interrupt controller code for older ColdFire CPU'sGreg Ungerer
The old ColdFire CPU's (5206, 5307, 5407, 5249 etc) use a simple interrupt controller. Use common setup code for them. This addition means that all ColdFire CPU's now have some specific type of interrupt controller code. Signed-off-by: Greg Ungerer <gerg@uclinux.org>
2009-09-16m68knommu: clean up ColdFire 532x interrupt setupGreg Ungerer
With the common intc-simr interrupt controller code in place the ColdFire 532x family startup code can be greatly simplified. Remove all the interrupt masking code, and the per-device interrupt config here. Signed-off-by: Greg Ungerer <gerg@uclinux.org>
2009-09-16m68knommu: general interrupt controller for ColdFire 532x partsGreg Ungerer
The ColdFire 532x family of parts uses 2 of the same INTC interrupt controlers used in the ColdFire 520x family. So modify the code to support both parts. The extra code for the second INTC controler in the case of the 520x is easily optimized away to nothing. Signed-off-by: Greg Ungerer <gerg@uclinux.org>
2009-09-16m68knommu: clean up ColdFire 523x interrupt setupGreg Ungerer
With the common intc-2 interrupt controller code in place the ColdFire 523x family startup code can be greatly simplified. Remove all the interrupt masking code, and the per-device interrupt config here. Signed-off-by: Greg Ungerer <gerg@uclinux.org>
2009-09-16m68knommu: clean up ColdFire 528x interrupt setupGreg Ungerer
With the common intc-2 interrupt controller code in place the ColdFire 528x family startup code can be greatly simplified. Remove all the interrupt masking code, and the per-device interrupt config here. Signed-off-by: Greg Ungerer <gerg@uclinux.org>
2009-09-16m68knommu: clean up ColdFire 527x interrupt setupGreg Ungerer
With the common intc-2 interrupt controller code in place the ColdFire 527x family startup code can be greatly simplified. Remove all the interrupt masking code, and the per-device interrupt config here. Signed-off-by: Greg Ungerer <gerg@uclinux.org>
2009-09-16m68knommu: general interrupt controller for ColdFire many 52xx partsGreg Ungerer
Create general interrupt controller code for the many ColdFire version 2 cores that use the two region INTC interrupt controller. This includes the 523x family, 5270, 5271, 5274, 5275, and the 528x families. This code does proper masking and unmasking of interrupts. With this in place some of the driver hacks in place to support ColdFire interrupts can finally go away. Signed-off-by: Greg Ungerer <gerg@uclinux.org>
2009-09-16m68knommu: remove per device interrupt mask setting for ColdFire 520xGreg Ungerer
With general interrupt controller code in place we don't need specific unmasking code for the internal ColdFire 520x UARTs or ethernet (FEC). Signed-off-by: Greg Ungerer <gerg@uclinux.org>
2009-09-16m68knommu: use general interrupt controller for ColdFire 520x familyGreg Ungerer
Create general interrupt controller code for the ColdFire 520x family, that does proper masking and unmasking of interrupts. With this in place some of the driver hacks in place to support ColdFire interrupts can finally go away. Within the ColdFire family there is a variety of different interrupt controllers in use. Some are used on multiple parts, some on only one. There is quite some differences in some varients, so much so that common code for all ColdFire parts would be impossible. This commit introduces code to support one of the newer interrupt controllers in the ColdFire 5208 and 5207 parts. It has very simple mask and unmask operations, so is one of the easiest to support. Signed-off-by: Greg Ungerer <gerg@uclinux.org>
2009-09-16m68k: clean up comment delimiters in dma.hGreg Ungerer
Change C99 style comments to traditional K&R style. Signed-off-by: Greg Ungerer <gerg@uclinux.org>
2009-09-16m68k: merge mmu and non-mmu versions of dma.hGreg Ungerer
The non-mmu version of dma.h contains a lot of ColdFire specific DMA support, but also all of the base m68k support. So use the non-mmu version of dma.h for all. Signed-off-by: Greg Ungerer <gerg@uclinux.org>
2009-09-16m68k: merge the mmu and non-mmu versions of checksum.hGreg Ungerer
The mmu and non-mmu versions of checksum.h are mostly the same, merge them. Signed-off-by: Greg Ungerer <gerg@uclinux.org>