aboutsummaryrefslogtreecommitdiffstats
Commit message (Collapse)AuthorAge
* ASoC: Define constants for WM8962 GPIO functionsMark Brown2011-04-26
| | | | | Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com> Acked-by: Liam Girdwood <lrg@ti.com>
* ASoC: Move WM8962 FLL configuration to CODECMark Brown2011-04-26
| | | | | | | There's only one DAI anyway. Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com> Acked-by: Liam Girdwood <lrg@ti.com>
* ASoC: Support FLL lock interrupt on WM8962Mark Brown2011-04-26
| | | | | Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com> Acked-by: Liam Girdwood <lrg@ti.com>
* ASoC: Support 24.576MHz MCLKs in WM8915Mark Brown2011-04-26
| | | | | | | We can safely divide these down to within the supported SYSCLK range. Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com> Acked-by: Liam Girdwood <lrg@ti.com>
* Merge branch 'for-2.6.39' into for-2.6.40Mark Brown2011-04-26
|\
| * davinci-mcasp: fix _CBM_CFS pin directionsBen Gardiner2011-04-26
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | The current davinci_mcasp_set_dai_fmt() sets bits ACLKX and ACLKR in the PDIR register for the codec clock-master/frame-slave mode; however, this results in the ACLKX and ACLKR pins being outputs according to SPRUFM1 [1] which conflicts with "codec is clock master." Similarly to the previous patch in this series, "fix _CBM_CFS hw_params" -- For codec clock-master/frame-slave mode (_CMB_CFS), clear bits ACLKX and ACLKR in the PDIR register to set the pins as inputs and hence allow externally sourced bit-clocks. [1] http://www.ti.com/litv/pdf/sprufm1 Signed-off-by: Ben Gardiner <bengardiner@nanometrics.ca> Reviewed-by: James Nuss <jamesnuss@nanometrics.ca> Acked-by: Liam Girdwood <lrg@ti.com> Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
| * davinci-mcasp: fix _CBM_CFS hw_paramsBen Gardiner2011-04-26
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | The current davinci_mcasp_set_dai_fmt() sets bits ACLKXE and ACLKRE (CLKXM and CLKRM as they are reffered to in SPRUFM1 [1]) for codec clock-slave/ frame-slave mode (_CBS_CFS) which selects internally generated bit-clock and frame-sync signals; however, it does the same thing again for codec clock-master/frame-slave mode (_CBM_CFS) in the very next case statement which is incorrectly selecting internally generated bit-clocks in this mode. For codec clock-master/frame-slave mode (_CBM_CFS), clear bits ACLKXE and ACLKRE to select externally-generated bit-clocks. [1] http://www.ti.com/litv/pdf/sprufm1 Signed-off-by: Ben Gardiner <bengardiner@nanometrics.ca> Reviewed-by: James Nuss <jamesnuss@nanometrics.ca> Acked-by: Liam Girdwood <lrg@ti.com> Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
| * davinci-mcasp: use bitfield definitions for PDIRBen Gardiner2011-04-26
| | | | | | | | | | | | | | | | | | | | | | | | | | The current driver creates value for set/clr of PDIR using (x<<26) instead of the #defines that are convieniently made available. Update the driver to use the bitfield definitions of PDIR. There is no functional change introduced by this patch. Signed-off-by: Ben Gardiner <bengardiner@nanometrics.ca> Reviewed-by: James Nuss <jamesnuss@nanometrics.ca> Acked-by: Liam Girdwood <lrg@ti.com> Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
| * ASoC: davinci-mcasp: correct tdm_slots limitBen Gardiner2011-04-26
| | | | | | | | | | | | | | | | | | | | | | | | | | | | The current check for the number of tdm-slots specified by platform data is always true (x >= 2 || x <= 32); therefore the else branch that warns of an incorrect number of slots can never be taken. Check that the number of tdm slots specified by platform data is between 2 and 32, inclusive. Signed-off-by: Ben Gardiner <bengardiner@nanometrics.ca> Reviewed-by: James Nuss <jamesnuss@nanometrics.ca> Acked-by: Liam Girdwood <lrg@ti.com> Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
* | ASoC: sh: fsi: Add module/port clock control functionKuninori Morimoto2011-04-26
| | | | | | | | | | | | | | | | | | | | | | | | The FIFO of each port were always working though it was not used in current FSI driver. This patch add module/port clock control function for fixing it. This patch is also caring suspend/resume. Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com> Reviewed-by: Simon Horman <simon@horms.net> Acked-by: Liam Girdwood <lrg@ti.com> Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
* | ASoC: sh: fsi: add dev_pm_ops :: suspend/resumeKuninori Morimoto2011-04-26
| | | | | | | | | | | | | | | | | | | | | | | | | | Current FSI driver sets important settings when probing. And it are not set again as long as driver is not bind again. This mean FSI driver will lost it from register if suspend/resume are happen. This patch save important settings for suspend/resume. Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com> Reviewed-by: Simon Horman <simon@horms.net> Acked-by: Liam Girdwood <lrg@ti.com> Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
* | ASoC: sh: fsi: add fsi_is_clk_master functionKuninori Morimoto2011-04-26
| | | | | | | | | | | | | | | | | | | | | | | | | | If FSI port is clock master, it use set_rate function which is callback from platform, and it is not necessary to call it if FSI port is clock slave. Current FSI driver called this callback if platform provide it. This patch modify it. Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com> Reviewed-by: Simon Horman <simon@horms.net> Acked-by: Liam Girdwood <lrg@ti.com> Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
* | ASoC: Remove default settings from Tegra KconfigMark Brown2011-04-26
| | | | | | | | | | | | | | There needs to be a strong reason for overriding the Kconfig default. Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com> Acked-by: Liam Girdwood <lrg@ti.com>
* | Merge branch 'tegra' into for-2.6.40Mark Brown2011-04-21
|\ \ | | | | | | | | | | | | | | | | | | Fix up merge with Harmony driver rename. Conflicts: sound/soc/tegra/Kconfig
| * | ARM: Tegra: select MACH_HAS_SND_SOC_TEGRA_WM8903Stephen Warren2011-04-21
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | CONFIG_SND_SOC_TEGRA_WM8903 is useful for many Tegra boards. To avoid the ASoC tegra/Kconfig enumerating them all, instead have the Tegra machine Kconfig select MACH_HAS_SND_SOC_TEGRA_WM8903 where appropriate, and have SND_SOC_TEGRA_WM8903 depend on this. [Redid ASoC diff so it applies. -- broonie] Signed-off-by: Stephen Warren <swarren@nvidia.com> Acked-by: Olof Johansson <olof@lixom.net> Acked-by: Liam Girdwood <lrg@ti.com> Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
* | | ASoC: WM8903: Implement DMIC supportStephen Warren2011-04-20
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | In addition to the currently supported analog capture path, the WM8903 also supports digital mics. The analog and digital capture paths are exclusive; a mux is present to select the capture source. Logically, the mux exists to select the decimator's input, from either the ADC or DMIC block outputs. However, the ADC power domain also includes the DMIC interface. Consequently, this change represents the mux as existing immediately before the ADC, and selecting between the Input PGA and DMIC block outputs. An alternative might be to represent the mux in its correct location, and associate the ADC power enable controls with both the real ADC, and a fake ADC for the DMIC? Signed-off-by: Stephen Warren <swarren@nvidia.com> Acked-by: Liam Girdwood <lrg@ti.com> Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
* | | ASoC: Add EQ and filter to max98095 CODEC driverPeter Hsiang2011-04-20
| | | | | | | | | | | | | | | | | | | | | | | | This patch adds the equalizer and biquad filter controls. Signed-off-by: Peter Hsiang <peter.hsiang@maxim-ic.com> Acked-by: Liam Girdwood <lrg@ti.com> Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
* | | ASoC: Tegra: wm8903: s/code/data/ for control/widget/mapsStephen Warren2011-04-20
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Replace calls to a variety of registration functions by updating struct snd_soc_card snd_soc_tegra_wm8903 to directly point at the various control/widget/map tables instead. The ASoC core now performs any required registration based on these data fields. (Applying Mark's TrimSlice review comments to the existing driver) Signed-off-by: Stephen Warren <swarren@nvidia.com> Acked-by: Liam Girdwood <lrg@ti.com> Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
* | | ASoC: fix two ident style problemsLu Guanqun2011-04-20
| | | | | | | | | | | | | | | | | | Signed-off-by: Lu Guanqun <guanqun.lu@intel.com> Acked-by: Liam Girdwood <lrg@ti.com> Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
* | | ASoC: simple style fixLu Guanqun2011-04-20
| | | | | | | | | | | | | | | | | | | | | | | | | | | replace the tab with spaces, make it align with other paragraphs Signed-off-by: Lu Guanqun <guanqun.lu@intel.com> Acked-by: Liam Girdwood <lrg@ti.com> Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
* | | ASoC: remove unused commentLu Guanqun2011-04-20
| | | | | | | | | | | | | | | | | | | | | | | | | | | `type` parameter is not longer used in `snd_soc_codec_set_cache_io`, so remove this line. Signed-off-by: Lu Guanqun <guanqun.lu@intel.com> Acked-by: Liam Girdwood <lrg@ti.com> Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
* | | ASoC: fix a simple coding style issueLu Guanqun2011-04-20
| | | | | | | | | | | | | | | | | | Signed-off-by: Lu Guanqun <guanqun.lu@intel.com> Acked-by: Liam Girdwood <lrg@ti.com> Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
* | | ASoC: snd_soc_dapm_get_pin_status: Match other contexts tooStephen Warren2011-04-20
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Not all widgets on a card are within the codec's DAPM context. Fix snd_soc_dapm_get_pin_status to search all contexts when looking for a widget. This change is required when modifying tegra_wm8903 to use snd_soc_card.widgets rather than calling snd_soc_dapm_new_controls; the former adds the widgets to the card's DAPM context, whereas tegra_wm8903 uses the codec's DAPM context when calling snd_soc_dapm_new_controls. By code inspection, I suspect this also applies to Samsung Speyside. Signed-off-by: Stephen Warren <swarren@nvidia.com> Acked-by: Liam Girdwood <lrg@ti.com> Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
* | | ASoC: Tegra: Retrieve card from DAPM context not codecStephen Warren2011-04-20
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Card widgets are created in the card's DAPM context, not any codec's DAPM context. Hence, w->codec==NULL. Instead, find the card from the widget through the DAPM context of the widget, not the codec of the widget. Signed-off-by: Stephen Warren <swarren@nvidia.com> Acked-by: Liam Girdwood <lrg@ti.com> Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
* | | ASoC: Tegra: Don't return mclk_changed from utils_set_rateStephen Warren2011-04-20
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Only the clock programming code needs to know whether the clocks changed, and that is encapsulated within tegra_asoc_utils_set_rate(). The machine driver's call to snd_soc_dai_set_sysclk(codec_dai, ...) is safe irrespective of whether the clocks changed. (Applying Mark's TrimSlice review comments to the existing driver) Signed-off-by: Stephen Warren <swarren@nvidia.com> Acked-by: Liam Girdwood <lrg@ti.com> Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
* | | ASoC: Tegra: wm8903: Remove redundant drvdata clearsStephen Warren2011-04-20
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | When the driver is not initialized/registered, nothing should be touching these fields anyway, so there's no point clearing them out. (Applying Mark's TrimSlice review comments to the existing driver) Signed-off-by: Stephen Warren <swarren@nvidia.com> Acked-by: Liam Girdwood <lrg@ti.com> Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
* | | ASoC: Tegra: wm8903 probe: Don't call machine_is_*()Stephen Warren2011-04-20
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | This machine driver is a platform driver, and hence will only be instantiated on the correct machines. Hence, there is no need to check the current machine during probe. (Applying Mark's TrimSlice review comments to the existing driver) Signed-off-by: Stephen Warren <swarren@nvidia.com> Acked-by: Liam Girdwood <lrg@ti.com> Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
* | | ASoC: Tegra: Support more boardsStephen Warren2011-04-18
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | * Ventana is identical to Harmony. * Seaboard, Kaen, and Aebl are all pretty similar, mainly with slightly different sets of GPIOs, and slightly different WM8903 pin connectivity. Signed-off-by: Stephen Warren <swarren@nvidia.com> Acked-by: Liam Girdwood <lrg@ti.com> Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
* | | ASoC: Tegra: Don't store snd_soc_jack_gpio in an arrayStephen Warren2011-04-18
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Storing the struct in an array makes the assignments to the GPIO member a little non-obvious, and is pointless when there's only a single GPIO. (I thought I fixed this during the review cycle when first submitting this driver, but I guess I overlooked that) Signed-off-by: Stephen Warren <swarren@nvidia.com> Acked-by: Liam Girdwood <lrg@ti.com> Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
* | | ASoC: Tegra: Rename Kconfig SND_TEGRA_SOC_* to SND_SOC_TEGRA_*Stephen Warren2011-04-18
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | The previous commit renames SND_TEGRA_SOC_HARMONY to SND_TEGRA_SOC_WM8903. While we're breaking people's .config files, rename all Tegra/SOC-related Kconfig variables to be more consistent with at least the core codec variables. Note that there exist machines that name their variables both ways. Signed-off-by: Stephen Warren <swarren@nvidia.com> Acked-by: Liam Girdwood <lrg@ti.com> Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
* | | ASoC: Tegra: Rename harmony.c to tegra_wm8903.cStephen Warren2011-04-18
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Soon, this machine driver will be updated to handle a number of Tegra boards using the WM8903 codec. Rename the file in advance to reflect this. Fix the content of tegra_wm8903.c to match the rename; replace references to Harmony board with something more generic. * s/struct tegra_harmony/struct tegra_wm8903/ * s/harmony/machine/ # variable name * Similar rename for some functions * Similar comment fix * Similar MODULE_DESCRIPTION fix Signed-off-by: Stephen Warren <swarren@nvidia.com> Acked-by: Liam Girdwood <lrg@ti.com> Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
* | | Merge branch 'tegra' into for-2.6.40Mark Brown2011-04-18
|\| |
| * | ARM: Tegra: Add to tegra_wm8903_platform_dataStephen Warren2011-04-18
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Seaboard derivate Kaen has a GPIO to mute the headphone output. Add a field to tegra_wm8903_platform_data so the board files can pass the GPIO number for that to the ASoC machine driver. Also, initialize this new field to a "not present" value for Harmony. Signed-off-by: Stephen Warren <swarren@nvidia.com> Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
| * | ASoC: Tegra: Rename pdev tegra-snd-harmony to tegra-snd-wm8903Stephen Warren2011-04-18
| | | | | | | | | | | | | | | | | | | | | | | | | | | Soon, this machine driver will be updated to handle a number of Tegra boards using the WM8903 codec. Rename the platform device in advance to reflect this. Signed-off-by: Stephen Warren <swarren@nvidia.com> Acked-by: Liam Girdwood <lrg@slimlogic.co.uk> Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
| * | ARM: Tegra: Rename harmony_audio.h -> tegra_wm8903_pdata.hStephen Warren2011-04-18
| | | | | | | | | | | | | | | | | | | | | | | | The audio driver will soon support more than just the Tegra Harmony board. Rename the platform data header file and data type to reflect this. Signed-off-by: Stephen Warren <swarren@nvidia.com> Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
* | | ASoC: Merge branch 'for-2.6.39' into for-2.6.40Mark Brown2011-04-18
|\ \ \ | | |/ | |/| | | | | | | | | | | | | Fix trivial conflict caused by silly spelling fix patch. Conflicts: sound/soc/codecs/wm8994.c
| * | ASoC: add a module alias to the FSI driverGuennadi Liakhovetski2011-04-18
| | | | | | | | | | | | | | | | | | | | | | | | | | | This patch enables FSI driver autoloading on sh-mobile systems. Signed-off-by: Guennadi Liakhovetski <g.liakhovetski@gmx.de> Reviewed-by: Simon Horman <horms@verge.net.au Acked-by: Liam Girdwood <lrg@ti.com> Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
| * | Merge commit 'v2.6.39-rc3' into for-2.6.39Mark Brown2011-04-18
| |\|
| | * Linux 2.6.39-rc3Linus Torvalds2011-04-11
| | |
| | * Merge branch 'for-linus' of git://oss.sgi.com/xfs/xfsLinus Torvalds2011-04-11
| | |\ | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | * 'for-linus' of git://oss.sgi.com/xfs/xfs: xfs: use proper interfaces for on-stack plugging xfs: fix xfs_debug warnings xfs: fix variable set but not used warnings xfs: convert log tail checking to a warning xfs: catch bad block numbers freeing extents. xfs: push the AIL from memory reclaim and periodic sync xfs: clean up code layout in xfs_trans_ail.c xfs: convert the xfsaild threads to a workqueue xfs: introduce background inode reclaim work xfs: convert ENOSPC inode flushing to use new syncd workqueue xfs: introduce a xfssyncd workqueue xfs: fix extent format buffer allocation size xfs: fix unreferenced var error in xfs_buf.c Also, applied patch from Tony Luck that fixes ia64: xfs_destroy_workqueues() should not be tagged with__exit in the branch before merging.
| | | * xfs_destroy_workqueues() should not be tagged with__exitLuck, Tony2011-04-11
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | ia64 throws away .exit sections for the built-in CONFIG case, so routines that are used in other circumstances should not be tagged as __exit. Signed-off-by: Tony Luck <tony.luck@intel.com> Reviewed-by: Christoph Hellwig <hch@lst.de> Signed-off-by: Alex Elder <aelder@sgi.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
| | | * xfs: use proper interfaces for on-stack pluggingChristoph Hellwig2011-04-08
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Add proper blk_start_plug/blk_finish_plug pairs for the two places where we issue buffer I/O, and remove the blk_flush_plug in xfs_buf_lock and xfs_buf_iowait, given that context switches already flush the per-process plugging lists. Signed-off-by: Christoph Hellwig <hch@lst.de> Signed-off-by: Alex Elder <aelder@sgi.com>
| | | * xfs: fix xfs_debug warningsChristoph Hellwig2011-04-08
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | For a CONFIG_XFS_DEBUG=n build gcc complains about statements with no effect in xfs_debug: fs/xfs/quota/xfs_qm_syscalls.c: In function 'xfs_qm_scall_trunc_qfiles': fs/xfs/quota/xfs_qm_syscalls.c:291:3: warning: statement with no effect The reason for that is that the various new xfs message functions have a return value which is never used, and in case of the non-debug build xfs_debug the macro evaluates to a plain 0 which produces the above warnings. This can be fixed by turning xfs_debug into an inline function instead of a macro, but in addition to that I've also changed all the message helpers to return void as we never use their return values. Signed-off-by: Christoph Hellwig <hch@lst.de> Reviewed-by: Dave Chinner <dchinner@redhat.com> Signed-off-by: Alex Elder <aelder@sgi.com>
| | | * xfs: fix variable set but not used warningsChristoph Hellwig2011-04-08
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | GCC 4.6 now warnings about variables set but not used. Fix the trivially fixable warnings of this sort. Signed-off-by: Christoph Hellwig <hch@lst.de> Signed-off-by: Alex Elder <aelder@sgi.com>
| | | * xfs: convert log tail checking to a warningDave Chinner2011-04-07
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | On the Power platform, the log tail debug checks fire excessively causing the system to panic early in testing. The debug checks are known to be racy, though on x86_64 there is no evidence that they trigger at all. We want to keep the checks active on debug systems to alert us to problems with log space accounting, but we need to reduce the impact of a racy check on testing on the Power platform. As a result, convert the ASSERT conditions to warnings, and allow them to fire only once per filesystem mount. This will prevent false positives from interfering with testing, whilst still providing us with the indication that they may be a problem with log space accounting should that occur. Signed-off-by: Dave Chinner <dchinner@redhat.com> Reviewed-by: Christoph Hellwig <hch@lst.de> Reviewed-by: Alex Elder <aelder@sgi.com>
| | | * xfs: catch bad block numbers freeing extents.Dave Chinner2011-04-07
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | A fuzzed filesystem crashed a kernel when freeing an extent with a block number beyond the end of the filesystem. Convert all the debug asserts in xfs_free_extent() to active checks so that we catch bad extents and return that the filesytsem is corrupted rather than crashing. Signed-off-by: Dave Chinner <dchinner@redhat.com> Reviewed-by: Christoph Hellwig <hch@lst.de> Reviewed-by: Alex Elder <aelder@sgi.com>
| | | * xfs: push the AIL from memory reclaim and periodic syncDave Chinner2011-04-07
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | When we are short on memory, we want to expedite the cleaning of dirty objects. Hence when we run short on memory, we need to kick the AIL flushing into action to clean as many dirty objects as quickly as possible. To implement this, sample the lsn of the log item at the head of the AIL and use that as the push target for the AIL flush. Further, we keep items in the AIL that are dirty that are not tracked any other way, so we can get objects sitting in the AIL that don't get written back until the AIL is pushed. Hence to get the filesystem to the idle state, we might need to push the AIL to flush out any remaining dirty objects sitting in the AIL. This requires the same push mechanism as the reclaim push. This patch also renames xfs_trans_ail_tail() to xfs_ail_min_lsn() to match the new xfs_ail_max_lsn() function introduced in this patch. Similarly for xfs_trans_ail_push -> xfs_ail_push. Signed-off-by: Dave Chinner <dchinner@redhat.com> Reviewed-by: Alex Elder <aelder@sgi.com>
| | | * xfs: clean up code layout in xfs_trans_ail.cDave Chinner2011-04-07
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | This patch rearranges the location of functions in xfs_trans_ail.c to remove the need for forward declarations of those functions in preparation for adding new functions without the need for forward declarations. Signed-off-by: Dave Chinner <dchinner@redhat.com> Reviewed-by: Alex Elder <aelder@sgi.com>
| | | * xfs: convert the xfsaild threads to a workqueueDave Chinner2011-04-07
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Similar to the xfssyncd, the per-filesystem xfsaild threads can be converted to a global workqueue and run periodically by delayed works. This makes sense for the AIL pushing because it uses variable timeouts depending on the work that needs to be done. By removing the xfsaild, we simplify the AIL pushing code and remove the need to spread the code to implement the threading and pushing across multiple files. Signed-off-by: Dave Chinner <dchinner@redhat.com> Reviewed-by: Christoph Hellwig <hch@lst.de> Reviewed-by: Alex Elder <aelder@sgi.com>
| | | * xfs: introduce background inode reclaim workDave Chinner2011-04-07
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Background inode reclaim needs to run more frequently that the XFS syncd work is run as 30s is too long between optimal reclaim runs. Add a new periodic work item to the xfs syncd workqueue to run a fast, non-blocking inode reclaim scan. Background inode reclaim is kicked by the act of marking inodes for reclaim. When an AG is first marked as having reclaimable inodes, the background reclaim work is kicked. It will continue to run periodically untill it detects that there are no more reclaimable inodes. It will be kicked again when the first inode is queued for reclaim. To ensure shrinker based inode reclaim throttles to the inode cleaning and reclaim rate but still reclaim inodes efficiently, make it kick the background inode reclaim so that when we are low on memory we are trying to reclaim inodes as efficiently as possible. This kick shoul d not be necessary, but it will protect against failures to kick the background reclaim when inodes are first dirtied. To provide the rate throttling, make the shrinker pass do synchronous inode reclaim so that it blocks on inodes under IO. This means that the shrinker will reclaim inodes rather than just skipping over them, but it does not adversely affect the rate of reclaim because most dirty inodes are already under IO due to the background reclaim work the shrinker kicked. These two modifications solve one of the two OOM killer invocations Chris Mason reported recently when running a stress testing script. The particular workload trigger for the OOM killer invocation is where there are more threads than CPUs all unlinking files in an extremely memory constrained environment. Unlike other solutions, this one does not have a performance impact on performance when memory is not constrained or the number of concurrent threads operating is <= to the number of CPUs. Signed-off-by: Dave Chinner <dchinner@redhat.com> Reviewed-by: Christoph Hellwig <hch@lst.de> Reviewed-by: Alex Elder <aelder@sgi.com>
generated by cgit v1.2.2 (git 2.25.0) at 2025-02-15 14:21:17 -0500