summaryrefslogtreecommitdiffstats
Commit message (Collapse)AuthorAgeFilesLines
* crypto: n2 - Fix a get/put_cpu() imbalanceThomas Meyer2011-08-151-2/+2
| | | | | | | Fix a get/put_cpu() imbalance in the error case when qp == NULL Signed-off-by: Thomas Meyer <thomas@m3y3r.de> Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
* crypto: sha1 - SSSE3 based SHA1 implementation for x86-64Mathias Krause2011-08-105-0/+819
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | This is an assembler implementation of the SHA1 algorithm using the Supplemental SSE3 (SSSE3) instructions or, when available, the Advanced Vector Extensions (AVX). Testing with the tcrypt module shows the raw hash performance is up to 2.3 times faster than the C implementation, using 8k data blocks on a Core 2 Duo T5500. For the smalest data set (16 byte) it is still 25% faster. Since this implementation uses SSE/YMM registers it cannot safely be used in every situation, e.g. while an IRQ interrupts a kernel thread. The implementation falls back to the generic SHA1 variant, if using the SSE/YMM registers is not possible. With this algorithm I was able to increase the throughput of a single IPsec link from 344 Mbit/s to 464 Mbit/s on a Core 2 Quad CPU using the SSSE3 variant -- a speedup of +34.8%. Saving and restoring SSE/YMM state might make the actual throughput fluctuate when there are FPU intensive userland applications running. For example, meassuring the performance using iperf2 directly on the machine under test gives wobbling numbers because iperf2 uses the FPU for each packet to check if the reporting interval has expired (in the above test I got min/max/avg: 402/484/464 MBit/s). Using this algorithm on a IPsec gateway gives much more reasonable and stable numbers, albeit not as high as in the directly connected case. Here is the result from an RFC 2544 test run with a EXFO Packet Blazer FTB-8510: frame size sha1-generic sha1-ssse3 delta 64 byte 37.5 MBit/s 37.5 MBit/s 0.0% 128 byte 56.3 MBit/s 62.5 MBit/s +11.0% 256 byte 87.5 MBit/s 100.0 MBit/s +14.3% 512 byte 131.3 MBit/s 150.0 MBit/s +14.2% 1024 byte 162.5 MBit/s 193.8 MBit/s +19.3% 1280 byte 175.0 MBit/s 212.5 MBit/s +21.4% 1420 byte 175.0 MBit/s 218.7 MBit/s +25.0% 1518 byte 150.0 MBit/s 181.2 MBit/s +20.8% The throughput for the largest frame size is lower than for the previous size because the IP packets need to be fragmented in this case to make there way through the IPsec tunnel. Signed-off-by: Mathias Krause <minipli@googlemail.com> Cc: Maxim Locktyukhin <maxim.locktyukhin@intel.com> Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
* crypto: sha1 - export sha1_update for reuseMathias Krause2011-08-102-4/+8
| | | | | | | | | | | Export the update function as crypto_sha1_update() to not have the need to reimplement the same algorithm for each SHA-1 implementation. This way the generic SHA-1 implementation can be used as fallback for other implementations that fail to run under certain circumstances, like the need for an FPU context while executing in IRQ context. Signed-off-by: Mathias Krause <minipli@googlemail.com> Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
* crypto: picoxcell - fix possible invalid pointer dereferenceJamie Iles2011-08-101-1/+1
| | | | | | | | | The completion callback will free the request so we must remove it from the completion list before calling the callback. Cc: Herbert Xu <herbert@gondor.apana.org.au> Signed-off-by: Jamie Iles <jamie@jamieiles.com> Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
* crypto: picoxcell - support for device tree matchingJamie Iles2011-08-102-6/+53
| | | | | | | | | Allow the crypto engines to be matched from device tree bindings. Cc: devicetree-discuss@lists.ozlabs.org Cc: Herbert Xu <herbert@gondor.apana.org.au> Signed-off-by: Jamie Iles <jamie@jamieiles.com> Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
* crypto: picoxcell - add connection ID to the clock nameJamie Iles2011-08-101-1/+1
| | | | | | | | | For using the device tree probing we use a connection ID for the clk_get() operation. Cc: Herbert Xu <herbert@gondor.apana.org.au> Signed-off-by: Jamie Iles <jamie@jamieiles.com> Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
* crypto: picoxcell - convert to platform ID tableJamie Iles2011-08-101-60/+33
| | | | | | | | | | | | Use a platform ID table and a single platform_driver. It's neater and makes the device tree addition easier and more consistent. Rename the match values to be inline with what they'll be in the device tree bindings. There aren't any current in-tree users of the existing device names. Cc: Herbert Xu <herbert@gondor.apana.org.au> Signed-off-by: Jamie Iles <jamie@jamieiles.com> Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
* crypto: padlock-aes - Make module loading even quieter when hardware is missingJonathan Nieder2011-08-031-3/+1
| | | | | | | | | | | | | | | | | | When loading aes via the module alias, a padlock module failing to load due to missing hardware is not particularly notable. With v2.6.27-rc1~1107^2~14 (crypto: padlock - Make module loading quieter when hardware isn't available, 2008-07-03), the padlock-aes module suppresses the relevant messages when the "quiet" flag is in use; but better to suppress this particular message completely, since the administrator can already distinguish such errors by the absence of a message indicating initialization failing or succeeding. This avoids occasional messages in syslog of the form padlock_aes: VIA PadLock not detected. Signed-off-by: Jonathan Nieder <jrnieder@gmail.com> Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
* oom: task->mm == NULL doesn't mean the memory was freedOleg Nesterov2011-08-011-1/+3
| | | | | | | | | | | | | | | | | | exit_mm() sets ->mm == NULL then it does mmput()->exit_mmap() which frees the memory. However select_bad_process() checks ->mm != NULL before TIF_MEMDIE, so it continues to kill other tasks even if we have the oom-killed task freeing its memory. Change select_bad_process() to check ->mm after TIF_MEMDIE, but skip the tasks which have already passed exit_notify() to ensure a zombie with TIF_MEMDIE set can't block oom-killer. Alternatively we could probably clear TIF_MEMDIE after exit_mmap(). Signed-off-by: Oleg Nesterov <oleg@redhat.com> Reviewed-by: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
* Merge branch 'for-linus' of ↵Linus Torvalds2011-08-019-202/+311
|\ | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | git://git.kernel.org/pub/scm/linux/kernel/git/lrg/voltage-2.6 * 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/lrg/voltage-2.6: (23 commits) regulator: Improve WM831x DVS VSEL selection algorithm regulator: Bootstrap wm831x DVS VSEL value from ON VSEL if not already set regulator: Set up GPIO for WM831x VSEL before enabling VSEL mode regulator: Add EPEs to the MODULE_ALIAS() for wm831x-dcdc regulator: Fix WM831x DCDC DVS VSEL bootstrapping regulator: Fix WM831x regulator ID lookups for multiple WM831xs regulator: Fix argument format type errors in error prints regulator: Fix memory leak in set_machine_constraints() error paths regulator: Make core more chatty about some errors regulator: tps65910: Fix array access out of bounds bug regulator: tps65910: Add missing breaks in switch/case regulator: tps65910: Fix a memory leak in tps65910_probe error path regulator: TWL: Remove entry of RES_ID for 6030 macros ASoC: tlv320aic3x: Add correct hw registers to Line1 cross connect muxes regulator: Add basic per consumer debugfs regulator: Add rdev_crit() macro regulator: Refactor supply implementation to work as regular consumers regulator: Include the device name in the microamps_requested_ file regulator: Increase the limit on sysfs file names regulator: Properly register dummy regulator driver ...
| * Merge branch 'for-next' into for-linusLiam Girdwood2011-08-0110-215/+332
| |\
| | * regulator: Improve WM831x DVS VSEL selection algorithmMark Brown2011-07-281-37/+15
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Rather than using the maximum voltage we get passed to select the DVS voltage to use remember the highest voltage we've ever seen. This improves how the driver works when the consumer permits higher voltages than it will ever selects in order to support the widest possible voltage range. Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com> Signed-off-by: Liam Girdwood <lrg@slimlogic.co.uk>
| | * regulator: Bootstrap wm831x DVS VSEL value from ON VSEL if not already setMark Brown2011-07-281-0/+14
| | | | | | | | | | | | | | | | | | | | | | | | If we don't have a DVS VSEL value already set when we start up then start it off with the value currently being used for ON. Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com> Signed-off-by: Liam Girdwood <lrg@slimlogic.co.uk>
| | * regulator: Set up GPIO for WM831x VSEL before enabling VSEL modeMark Brown2011-07-281-21/+20
| | | | | | | | | | | | | | | | | | | | | | | | | | | If the VSEL is not in use prior to us starting up then we need to make sure we initialise the GPIO before we push the DVS control to being done by the GPIO. Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com> Signed-off-by: Liam Girdwood <lrg@slimlogic.co.uk>
| | * regulator: Add EPEs to the MODULE_ALIAS() for wm831x-dcdcMark Brown2011-07-281-0/+1
| | | | | | | | | | | | | | | Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com> Signed-off-by: Liam Girdwood <lrg@slimlogic.co.uk>
| | * regulator: Fix WM831x DCDC DVS VSEL bootstrappingMark Brown2011-07-281-1/+1
| | | | | | | | | | | | | | | | | | | | | Read our initial VSEL from the correct register. Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com> Signed-off-by: Liam Girdwood <lrg@slimlogic.co.uk>
| | * regulator: Fix WM831x regulator ID lookups for multiple WM831xsMark Brown2011-07-282-5/+36
| | | | | | | | | | | | | | | | | | | | | | | | | | | With multiple wm831x devices the device IDs used for the regulators will not always be contiguous so simply taking the modulus is not sufficient to look up the ID, we need to reverse the way the ID is generated. Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com> Signed-off-by: Liam Girdwood <lrg@slimlogic.co.uk>
| | * regulator: Fix argument format type errors in error printsMark Brown2011-07-221-2/+2
| | | | | | | | | | | | | | | | | | | | | We need to dereference the pointers to print their values. Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com> Signed-off-by: Liam Girdwood <lrg@slimlogic.co.uk>
| | * regulator: Fix memory leak in set_machine_constraints() error pathsAxel Lin2011-07-221-3/+4
| | | | | | | | | | | | | | | | | | | | | | | | | | | Properly kfree rdev->constraints in all set_machine_constraints() error paths. Also properly kfree rdev->constraints in regulator_register() error paths. Signed-off-by: Axel Lin <axel.lin@gmail.com> Acked-by: Mark Brown <broonie@opensource.wolfsonmicro.com> Signed-off-by: Liam Girdwood <lrg@slimlogic.co.uk>
| | * regulator: Make core more chatty about some errorsMark Brown2011-07-221-2/+9
| | | | | | | | | | | | | | | | | | | | | | | | Prevent some head scratching by making the core log about some rare but possible errors with invalid voltage ranges and modes being set. Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com> Signed-off-by: Liam Girdwood <lrg@slimlogic.co.uk>
| | * regulator: tps65910: Fix array access out of bounds bugAxel Lin2011-07-221-12/+43
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | For tps65910, the number of regulator is 13. ( ARRAY_SIZE(tps65910_regs) is 13) For tps65911, the number of regulator is 12. ( ARRAY_SIZE(tps65911_regs) is 12) If we are using this driver for tps65911, we hit array access out of bounds bug in tps65910_probe() because current implementation always assume the number of regulator is 13 and thus it will access tps65911_regs[12]. Fix it by setting correct num_regulators for both chips in tps65910_probe(), and allocated neccessay memory accordingly. Signed-off-by: Axel Lin <axel.lin@gmail.com> Acked-by: Mark Brown <broonie@opensource.wolfsonmicro.com> Signed-off-by: Liam Girdwood <lrg@slimlogic.co.uk>
| | * regulator: tps65910: Add missing breaks in switch/caseAxel Lin2011-07-221-0/+7
| | | | | | | | | | | | | | | | | | | | | | | | | | | Also add a default case in tps65910_list_voltage_dcdc to silence 'volt' may be used uninitialized in this function warning. Signed-off-by: Axel Lin <axel.lin@gmail.com> Acked-by: Mark Brown <broonie@opensource.wolfsonmicro.com> Signed-off-by: Liam Girdwood <lrg@slimlogic.co.uk>
| | * regulator: tps65910: Fix a memory leak in tps65910_probe error pathAxel Lin2011-07-221-0/+1
| | | | | | | | | | | | | | | | | | | | | | | | Fix a memory leak if chip id is not matched. Signed-off-by: Axel Lin <axel.lin@gmail.com> Acked-by: Mark Brown <broonie@opensource.wolfsonmicro.com> Signed-off-by: Liam Girdwood <lrg@slimlogic.co.uk>
| | * regulator: TWL: Remove entry of RES_ID for 6030 macrosAmbresh K2011-07-221-34/+30
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | RES_ID is only used in 4030, to send PBM singular message to control the state of dedicated resources. In 6030, we don't have concept of PBM, hence removing the definition of RES_ID (num) from macros. Signed-off-by: Ambresh K <ambresh@ti.com> Acked-by: Mark Brown <broonie@opensource.wolfsonmicro.com> Signed-off-by: Liam Girdwood <lrg@slimlogic.co.uk>
| | * ASoC: tlv320aic3x: Add correct hw registers to Line1 cross connect muxesJarkko Nikula2011-07-041-13/+21
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Commit af46800 ("ASoC: Implement mux control sharing") revealed that "Left Line1[L | R] Mux" and "Right Line1[L | R] Mux" widgets were pointing to the same kcontrols and codec registers and thus soc-core falsely detected them as shared controls. This is actually wrong since there are separate registers in hardware that configure Line1L to RADC and Line1R to LADC cross connects so these muxes should not be shared. Signed-off-by: Jarkko Nikula <jhnikula@gmail.com> Acked-by: Mark Brown <broonie@opensource.wolfsonmicro.com> Signed-off-by: Liam Girdwood <lrg@slimlogic.co.uk>
| | * regulator: Add basic per consumer debugfsMark Brown2011-07-041-1/+29
| | | | | | | | | | | | | | | | | | | | | | | | Report the requested load and voltage for each consumer in debugfs when it is enabled. Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com> Signed-off-by: Liam Girdwood <lrg@slimlogic.co.uk>
| | * regulator: Add rdev_crit() macroMark Brown2011-06-091-0/+2
| | | | | | | | | | | | | | | | | | | | | | | | No actual users but provide the macro so there's less surprise when it's not there. Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com> Signed-off-by: Liam Girdwood <lrg@slimlogic.co.uk>
| | * regulator: Refactor supply implementation to work as regular consumersMark Brown2011-06-092-68/+37
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Currently the regulator supply implementation is somewhat complex and fragile as it doesn't look like standard consumers but is instead a parallel implementation. This causes issues with locking and reference counting. Move the implementation over to using standard consumers to address this. Rather than only notifying the supply on the first enable/disable we do so every time the regulator is enabled or disabled, simplifying locking as we don't need to hold a lock on the consumer we are about to enable. Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com> Signed-off-by: Liam Girdwood <lrg@slimlogic.co.uk>
| | * regulator: Include the device name in the microamps_requested_ fileMark Brown2011-06-091-2/+3
| | | | | | | | | | | | | | | | | | | | | | | | | | | We may have multiple devices requesting a supply with the same name so include the device name in the generated filename for microamps_requested to avoid duplicate files. Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com> Signed-off-by: Liam Girdwood <lrg@slimlogic.co.uk>
| | * regulator: Increase the limit on sysfs file namesMark Brown2011-06-091-1/+1
| | | | | | | | | | | | | | | | | | | | | With verbose filenames we can easily hit 32 characters. Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com> Signed-off-by: Liam Girdwood <lrg@slimlogic.co.uk>
| | * regulator: Properly register dummy regulator driverMark Brown2011-06-081-6/+26
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Recent changes in the driver core appear to mean that the data structures for the driver core are not fully initialised unless the driver is bound. Make sure the driver core knows the dummy driver is in use by binding it to a driver. Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com> Signed-off-by: Liam Girdwood <lrg@slimlogic.co.uk>
| | * regulator: twl-regulator: fix n_voltages for twl6030 variable LDOsColin Cross2011-05-301-1/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | The n_voltages initializer for the TWL6030_ADJUSTABLE_LDO macro is off by one, causing the the highest supported voltage to be unreachable. Setting the machine constraints to only allow the highest voltage causes errors: machine_constraints_voltage: VAUX3_6030: unsupportable voltage constraints twl_reg twl_reg.39: can't register VAUX3_6030, -22 twl_reg: probe of twl_reg.39 failed with error -22 This patch fixes the off by one error. Tested by setting VAUX3_6030 to 3.3V. Signed-off-by: Colin Cross <ccross@android.com> Acked-by: Mark Brown <broonie@opensource.wolfsonmicro.com> Signed-off-by: Liam Girdwood <lrg@slimlogic.co.uk>
| | * regulator: Use _cansleep() for WM8994 regulator GPIOsMark Brown2011-05-301-2/+2
| | | | | | | | | | | | | | | | | | | | | | | | The WM8994 regulator driver is perfectly happy if the GPIO used to enable the regulator sleeps so call the appropriate GPIO API. Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com> Signed-off-by: Liam Girdwood <lrg@slimlogic.co.uk>
| | * regulator: Do bulk enables of regulators in parallelMark Brown2011-05-302-6/+29
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | In order to reduce the impact of ramp times rather than enabling the regulators for a device in series use async tasks to run the actual enables. This means that the delays which the enables implement can all run in parallel, though it does mean that the order in which the supplies come on may be unstable. For super bonus fun points if any of the regulators are shared between multiple supplies on the same device (as is rather likely) then this will test our locking. Note that in this case we only delay once for each physical regulator so the threads shouldn't block each other while delaying. It'd be even nicer if we could coalesce writes to a shared enable registers in PMICs but that's definitely future work, and it may also be useful and is certainly more achievable to optimise out the parallelism if none of the regulators implement ramp delays. Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com> Signed-off-by: Liam Girdwood <lrg@slimlogic.co.uk>
* | | Merge branch 'for_linus' of ↵Linus Torvalds2011-08-0122-1972/+2190
|\ \ \ | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | git://git.kernel.org/pub/scm/linux/kernel/git/tytso/ext4 * 'for_linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tytso/ext4: (60 commits) ext4: prevent memory leaks from ext4_mb_init_backend() on error path ext4: use EXT4_BAD_INO for buddy cache to avoid colliding with valid inode # ext4: use ext4_msg() instead of printk in mballoc ext4: use ext4_kvzalloc()/ext4_kvmalloc() for s_group_desc and s_group_info ext4: introduce ext4_kvmalloc(), ext4_kzalloc(), and ext4_kvfree() ext4: use the correct error exit path in ext4_init_inode_table() ext4: add missing kfree() on error return path in add_new_gdb() ext4: change umode_t in tracepoint headers to be an explicit __u16 ext4: fix races in ext4_sync_parent() ext4: Fix overflow caused by missing cast in ext4_fallocate() ext4: add action of moving index in ext4_ext_rm_idx for Punch Hole ext4: simplify parameters of reserve_backup_gdb() ext4: simplify parameters of add_new_gdb() ext4: remove lock_buffer in bclean() and setup_new_group_blocks() ext4: simplify journal handling in setup_new_group_blocks() ext4: let setup_new_group_blocks() set multiple bits at a time ext4: fix a typo in ext4_group_extend() ext4: let ext4_group_add_blocks() handle 0 blocks quickly ext4: let ext4_group_add_blocks() return an error code ext4: rename ext4_add_groupblocks() to ext4_group_add_blocks() ... Fix up conflict in fs/ext4/inode.c: commit aacfc19c626e ("fs: simplify the blockdev_direct_IO prototype") had changed the ext4_ind_direct_IO() function for the new simplified calling convention, while commit dae1e52cb126 ("ext4: move ext4_ind_* functions from inode.c to indirect.c") moved the function to another file.
| * | | ext4: prevent memory leaks from ext4_mb_init_backend() on error pathYu Jian2011-08-011-6/+6
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | In ext4_mb_init(), if the s_locality_group allocation fails it will currently cause the allocations made in ext4_mb_init_backend() to be leaked. Moving the ext4_mb_init_backend() allocation after the s_locality_group allocation avoids that problem. Signed-off-by: Yu Jian <yujian@whamcloud.com> Signed-off-by: Andreas Dilger <adilger@whamcloud.com> Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
| * | | ext4: use EXT4_BAD_INO for buddy cache to avoid colliding with valid inode #Yu Jian2011-08-011-1/+5
| | | | | | | | | | | | | | | | | | | | | | | | Signed-off-by: Yu Jian <yujian@whamcloud.com> Signed-off-by: Andreas Dilger <adilger@whamcloud.com> Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
| * | | ext4: use ext4_msg() instead of printk in mballocTheodore Ts'o2011-08-011-37/+42
| | | | | | | | | | | | | | | | Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
| * | | ext4: use ext4_kvzalloc()/ext4_kvmalloc() for s_group_desc and s_group_infoTheodore Ts'o2011-08-013-13/+15
| | | | | | | | | | | | | | | | Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
| * | | ext4: introduce ext4_kvmalloc(), ext4_kzalloc(), and ext4_kvfree()Theodore Ts'o2011-08-012-18/+39
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Introduce new helper functions which try kmalloc, and then fall back to vmalloc if necessary, and use them for allocating and deallocating s_flex_groups. Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
| * | | ext4: use the correct error exit path in ext4_init_inode_table()Yongqiang Yang2011-08-011-1/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | This patch lets ext4_init_inode_table() handle errors right. ext4_init_inode_table() should down_write() alloc_sem which has been up_write()ed and stop the started journal handle. Signed-off-by: Yongqiang Yang <xiaoqiangnk@gmail.com> Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
| * | | ext4: add missing kfree() on error return path in add_new_gdb()Dan Carpenter2011-07-301-0/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | We added some more error handling in b40971426a "ext4: add error checking to calls to ext4_handle_dirty_metadata()". But we need to call kfree() as well to avoid a memory leak. Signed-off-by: Dan Carpenter <error27@gmail.com> Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
| * | | ext4: change umode_t in tracepoint headers to be an explicit __u16Theodore Ts'o2011-07-301-8/+8
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | As requested by Al Viro, since umode_t may be changing to a u32 for some architectures. Signed-off-by: "Theodore Ts'o" <tytso@mit.edu> Cc: Al Viro <viro@ZenIV.linux.org.uk>
| * | | ext4: fix races in ext4_sync_parent()Theodore Ts'o2011-07-301-5/+21
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Fix problems if fsync() races against a rename of a parent directory as pointed out by Al Viro in his own inimitable way: >While we are at it, could somebody please explain what the hell is ext4 >doing in >static int ext4_sync_parent(struct inode *inode) >{ > struct writeback_control wbc; > struct dentry *dentry = NULL; > int ret = 0; > > while (inode && ext4_test_inode_state(inode, EXT4_STATE_NEWENTRY)) { > ext4_clear_inode_state(inode, EXT4_STATE_NEWENTRY); > dentry = list_entry(inode->i_dentry.next, > struct dentry, d_alias); > if (!dentry || !dentry->d_parent || !dentry->d_parent->d_inode) > break; > inode = dentry->d_parent->d_inode; > ret = sync_mapping_buffers(inode->i_mapping); > ... >Note that dentry obviously can't be NULL there. dentry->d_parent is never >NULL. And dentry->d_parent would better not be negative, for crying out >loud! What's worse, there's no guarantees that dentry->d_parent will >remain our parent over that sync_mapping_buffers() *and* that inode won't >just be freed under us (after rename() and memory pressure leading to >eviction of what used to be our dentry->d_parent)...... Reported-by: Al Viro <viro@ZenIV.linux.org.uk> Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
| * | | ext4: Fix overflow caused by missing cast in ext4_fallocate()Utako Kusaka2011-07-271-1/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | The logical block number in map.l_blk is a __u32, and so before we shift it left, by the block size, we neeed cast it to a 64-bit size. Otherwise i_size can be corrupted on an ENOSPC. # df -T /mnt/mp1 Filesystem Type 1K-blocks Used Available Use% Mounted on /dev/sda6 ext4 9843276 153056 9190200 2% /mnt/mp1 # fallocate -o 0 -l 2199023251456 /mnt/mp1/testfile fallocate: /mnt/mp1/testfile: fallocate failed: No space left on device # stat /mnt/mp1/testfile File: `/mnt/mp1/testfile' Size: 4293656576 Blocks: 19380440 IO Block: 4096 regular file Device: 806h/2054d Inode: 12 Links: 1 Access: (0644/-rw-r--r--) Uid: ( 0/ root) Gid: ( 0/ root) Access: 2011-07-25 13:01:31.414490496 +0900 Modify: 2011-07-25 13:01:31.414490496 +0900 Change: 2011-07-25 13:01:31.454490495 +0900 Signed-off-by: Utako Kusaka <u-kusaka@wm.jp.nec.com> Signed-off-by: "Theodore Ts'o" <tytso@mit.edu> -- fs/ext4/extents.c | 2 +- 1 files changed, 1 insertions(+), 1 deletions(-)
| * | | ext4: add action of moving index in ext4_ext_rm_idx for Punch HoleRobin Dong2011-07-271-2/+7
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | The old function ext4_ext_rm_idx is used only for truncate case because it just remove last index in extent-index-block. When punching hole, it usually needed to remove "middle" index, therefore we must move indexes which after it forward. (I create a file with 1 depth extent tree and punch hole in the middle of it, the last index in index-block strangly gone, so I find out this bug) Signed-off-by: Robin Dong <sanbai@taobao.com> Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
| * | | ext4: simplify parameters of reserve_backup_gdb()Yongqiang Yang2011-07-271-5/+7
| | | | | | | | | | | | | | | | | | | | | | | | | | | | The reserve_backup_gdb() function only needs the block group number; there's no need to pass a pointer to struct ext4_new_group_data to it. Signed-off-by: Yongqiang Yang <xiaoqiangnk@gmail.com>
| * | | ext4: simplify parameters of add_new_gdb()Yongqiang Yang2011-07-271-15/+24
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | add_new_gdb() only needs the block group number; there is no need to pass a pointer to struct ext4_new_group_data to add_new_gdb(). Instead of filling in a pointer the struct buffer_head in add_new_gdb(), it's simpler to have the caller fetch it from the s_group_desc[] array. [Fixed error path to handle the case where struct buffer_head *primary hasn't been set yet. -- Ted] Signed-off-by: Yongqiang Yang <xiaoqiangnk@gmail.com> Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
| * | | ext4: remove lock_buffer in bclean() and setup_new_group_blocks()Yongqiang Yang2011-07-271-4/+0
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | There is no need to lock the buffers since no one else should be touching these buffers besides the file system. Signed-off-by: Yongqiang Yang <xiaoqiangnk@gmail.com> Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
| * | | ext4: simplify journal handling in setup_new_group_blocks()Yongqiang Yang2011-07-261-19/+20
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | This patch simplifies journal handling in setup_new_group_blocks(). In previous code, block bitmap is modified everywhere in setup_new_group_blocks(), ext4_get_write_access() in extend_or_restart_transaction() is used to guarantee that the block bitmap stays in the new handle, this makes things complicated. The previous commit changed things so that the modifications on the block bitmap are batched and done by ext4_set_bits() at the end of the for loop. This allows us to simplify things. Signed-off-by: Yongqiang Yang <xiaoqiangnk@gmail.com> Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
OpenPOWER on IntegriCloud