linux.git/drivers/dma, branch v2.6.36 Linux kernel source tree ioat2: fix performance regression 2010-10-13T22:43:10+00:00 Dan Williams dan.j.williams@intel.com 2010-10-13T22:43:10+00:00 c50a898fd4e736623ee175920db056194e0bb483 Commit 0793448 "DMAENGINE: generic channel status v2" changed the interface for how dma channel progress is retrieved. It inadvertently exported an internal helper function ioat_tx_status() instead of ioat_dma_tx_status(). The latter polls the hardware to get the latest completion state, while the helper just evaluates the current state without touching hardware. The effect is that we end up waiting for completion timeouts or descriptor allocation errors before the completion state is updated. iperf (before fix): [SUM] 0.0-41.3 sec 364 MBytes 73.9 Mbits/sec iperf (after fix): [SUM] 0.0- 4.5 sec 499 MBytes 940 Mbits/sec This is a regression starting with 2.6.35. Cc: <stable@kernel.org> Cc: Dave Jiang <dave.jiang@intel.com> Cc: Jesse Brandeburg <jesse.brandeburg@intel.com> Cc: Linus Walleij <linus.walleij@stericsson.com> Cc: Maciej Sosnowski <maciej.sosnowski@intel.com> Reported-by: Richard Scobie <richard@sauce.co.nz> Signed-off-by: Dan Williams <dan.j.williams@intel.com> 
Commit 0793448 "DMAENGINE: generic channel status v2" changed the interface for
how dma channel progress is retrieved.  It inadvertently exported an internal
helper function ioat_tx_status() instead of ioat_dma_tx_status().  The latter
polls the hardware to get the latest completion state, while the helper just
evaluates the current state without touching hardware.  The effect is that we
end up waiting for completion timeouts or descriptor allocation errors before
the completion state is updated.

iperf (before fix):
[SUM]  0.0-41.3 sec   364 MBytes  73.9 Mbits/sec

iperf (after fix):
[SUM]  0.0- 4.5 sec   499 MBytes   940 Mbits/sec

This is a regression starting with 2.6.35.

Cc: <stable@kernel.org>
Cc: Dave Jiang <dave.jiang@intel.com>
Cc: Jesse Brandeburg <jesse.brandeburg@intel.com>
Cc: Linus Walleij <linus.walleij@stericsson.com>
Cc: Maciej Sosnowski <maciej.sosnowski@intel.com>
Reported-by: Richard Scobie <richard@sauce.co.nz>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
dmaengine: fix interrupt clearing for mv_xor 2010-09-23T21:14:22+00:00 Simon Guinot sguinot@lacie.com 2010-09-17T21:33:51+00:00 cc60f8878eab892c03d06b10f389232b9b66bd83 When using simultaneously the two DMA channels on a same engine, some transfers are never completed. For example, an endless lock can occur while writing heavily on a RAID5 array (with async-tx offload support enabled). Note that this issue can also be reproduced by using the DMA test client. On a same engine, the interrupt cause register is shared between two DMA channels. This patch make sure that the cause bit is only cleared for the requested channel. Signed-off-by: Simon Guinot <sguinot@lacie.com> Tested-by: Luc Saillard <luc@saillard.org> Acked-by: saeed bishara <saeed.bishara@gmail.com> Cc: <stable@kernel.org> Signed-off-by: Dan Williams <dan.j.williams@intel.com> 
When using simultaneously the two DMA channels on a same engine, some
transfers are never completed. For example, an endless lock can occur
while writing heavily on a RAID5 array (with async-tx offload support
enabled).

Note that this issue can also be reproduced by using the DMA test
client.

On a same engine, the interrupt cause register is shared between two
DMA channels. This patch make sure that the cause bit is only cleared
for the requested channel.

Signed-off-by: Simon Guinot <sguinot@lacie.com>
Tested-by: Luc Saillard <luc@saillard.org>
Acked-by: saeed bishara <saeed.bishara@gmail.com>
Cc: <stable@kernel.org>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
dma/shdma: move dereference below the NULL check 2010-09-22T22:29:17+00:00 Dan Carpenter error27@gmail.com 2010-08-14T09:01:45+00:00 9f9ff20d46c6728b092f34b6a642e1e81ab5e254 "param" can be NULL here, so only dereference it after the check. Signed-off-by: Dan Carpenter <error27@gmail.com> Signed-off-by: Dan Williams <dan.j.williams@intel.com> 
"param" can be NULL here, so only dereference it after the check.

Signed-off-by: Dan Carpenter <error27@gmail.com>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
Merge branch 'next-devicetree' of git://git.secretlab.ca/git/linux-2.6 2010-08-12T16:11:31+00:00 Linus Torvalds torvalds@linux-foundation.org 2010-08-12T16:11:31+00:00 58d4ea65b98f154f3326b038eecda32f90b46ea8 * 'next-devicetree' of git://git.secretlab.ca/git/linux-2.6: mmc_spi: Fix unterminated of_match_table of/sparc: fix build regression from of_device changes of/device: Replace struct of_device with struct platform_device 
* 'next-devicetree' of git://git.secretlab.ca/git/linux-2.6:
  mmc_spi: Fix unterminated of_match_table
  of/sparc: fix build regression from of_device changes
  of/device: Replace struct of_device with struct platform_device
Merge branches 'master' and 'devel' into for-linus 2010-08-10T22:17:52+00:00 Russell King rmk+kernel@arm.linux.org.uk 2010-08-10T22:17:52+00:00 0b019a41553a919965bb02d07d54e3e6c57a796d Conflicts: arch/arm/Kconfig arch/arm/mm/Kconfig 
Conflicts:
	arch/arm/Kconfig
	arch/arm/mm/Kconfig
Merge branch 'next' of git://git.kernel.org/pub/scm/linux/kernel/git/djbw/async_tx 2010-08-10T04:00:07+00:00 Linus Torvalds torvalds@linux-foundation.org 2010-08-10T04:00:07+00:00 dcded10f6dce10411b16134ce9cc87bfdf75c13f * 'next' of git://git.kernel.org/pub/scm/linux/kernel/git/djbw/async_tx: (30 commits) DMAENGINE: at_hdmac: locking fixlet DMAENGINE: pch_dma: kill another usage of __raw_{read|write}l dma: dmatest: fix potential sign bug ioat2: catch and recover from broken vtd configurations v6 DMAENGINE: add runtime slave control to COH 901 318 v3 DMAENGINE: add runtime slave config to DMA40 v3 DMAENGINE: generic slave channel control v3 dmaengine: Driver for Topcliff PCH DMA controller intel_mid: Add Mrst & Mfld DMA Drivers drivers/dma: Eliminate a NULL pointer dereference dma/timb_dma: compile warning on 32 bit DMAENGINE: ste_dma40: support older silicon DMAENGINE: ste_dma40: support disabling physical channels DMAENGINE: ste_dma40: no disabled phy channels on ux500 DMAENGINE: ste_dma40: fix suspend bug DMAENGINE: ste_dma40: add DB8500 memcpy channels DMAENGINE: ste_dma40: no flow control on memcpy DMAENGINE: ste_dma40: arch updates for LCLA and LCPA DMAENGINE: ste_dma40: allocate LCLA dynamically DMAENGINE: ste_dma40: no premature stop ... Fix up trivial conflicts in arch/arm/mach-ux500/devices-db8500.c 
* 'next' of git://git.kernel.org/pub/scm/linux/kernel/git/djbw/async_tx: (30 commits)
  DMAENGINE: at_hdmac: locking fixlet
  DMAENGINE: pch_dma: kill another usage of __raw_{read|write}l
  dma: dmatest: fix potential sign bug
  ioat2: catch and recover from broken vtd configurations v6
  DMAENGINE: add runtime slave control to COH 901 318 v3
  DMAENGINE: add runtime slave config to DMA40 v3
  DMAENGINE: generic slave channel control v3
  dmaengine: Driver for Topcliff PCH DMA controller
  intel_mid: Add Mrst & Mfld DMA Drivers
  drivers/dma: Eliminate a NULL pointer dereference
  dma/timb_dma: compile warning on 32 bit
  DMAENGINE: ste_dma40: support older silicon
  DMAENGINE: ste_dma40: support disabling physical channels
  DMAENGINE: ste_dma40: no disabled phy channels on ux500
  DMAENGINE: ste_dma40: fix suspend bug
  DMAENGINE: ste_dma40: add DB8500 memcpy channels
  DMAENGINE: ste_dma40: no flow control on memcpy
  DMAENGINE: ste_dma40: arch updates for LCLA and LCPA
  DMAENGINE: ste_dma40: allocate LCLA dynamically
  DMAENGINE: ste_dma40: no premature stop
  ...

Fix up trivial conflicts in arch/arm/mach-ux500/devices-db8500.c
Merge master.kernel.org:/pub/scm/linux/kernel/git/lethal/genesis-2.6 into devel-stable 2010-08-06T17:13:19+00:00 Russell King rmk+kernel@arm.linux.org.uk 2010-08-06T17:13:19+00:00 500b9fc922cbec572f4fd1436533bfaed5011262 Conflicts: drivers/net/irda/sh_irda.c 
Conflicts:
	drivers/net/irda/sh_irda.c
of/device: Replace struct of_device with struct platform_device 2010-08-06T15:25:50+00:00 Grant Likely grant.likely@secretlab.ca 2010-08-06T15:25:50+00:00 2dc11581376829303b98eadb2de253bee065a56a of_device is just an alias for platform_device, so remove it entirely. Also replace to_of_device() with to_platform_device() and update comment blocks. This patch was initially generated from the following semantic patch, and then edited by hand to pick up the bits that coccinelle didn't catch. @@ @@ -struct of_device +struct platform_device Signed-off-by: Grant Likely <grant.likely@secretlab.ca> Reviewed-by: David S. Miller <davem@davemloft.net> 
of_device is just an alias for platform_device, so remove it entirely.  Also
replace to_of_device() with to_platform_device() and update comment blocks.

This patch was initially generated from the following semantic patch, and then
edited by hand to pick up the bits that coccinelle didn't catch.

@@
@@
-struct of_device
+struct platform_device

Signed-off-by: Grant Likely <grant.likely@secretlab.ca>
Reviewed-by: David S. Miller <davem@davemloft.net>
DMAENGINE: at_hdmac: locking fixlet 2010-08-05T05:09:17+00:00 Yong Wang yong.y.wang@linux.intel.com 2010-08-05T02:40:08+00:00 b0ebeb9c09cb5b84bb2f33927c84c7648845fcec atc_chain_complete shall be called with atchan->lock held and bh disabled. Signed-off-by: Yong Wang <yong.y.wang@intel.com> Signed-off-by: Dan Williams <dan.j.williams@intel.com> 
atc_chain_complete shall be called with atchan->lock held
and bh disabled.

Signed-off-by: Yong Wang <yong.y.wang@intel.com>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
DMAENGINE: pch_dma: kill another usage of __raw_{read|write}l 2010-08-05T05:08:45+00:00 Yong Wang yong.y.wang@linux.intel.com 2010-08-05T02:38:43+00:00 61cd2203769a2bf35d41f8682f6ef865fe2d23ff Use {read|write}l instead of __raw_{read|write}l since PCH DMA controller is PCI device. Signed-off-by: Yong Wang <yong.y.wang@intel.com> Signed-off-by: Dan Williams <dan.j.williams@intel.com> 
Use {read|write}l instead of __raw_{read|write}l since PCH DMA
controller is PCI device.

Signed-off-by: Yong Wang <yong.y.wang@intel.com>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>