linux-toradex.git/crypto/async_tx, branch v2.6.33.17 Linux kernel for Apalis and Colibri modules raid6: fix recovery performance regression 2010-05-12T22:02:42+00:00 Dan Williams dan.j.williams@intel.com 2010-05-05T03:41:56+00:00 d15b5afa5324123c49dd9b19acc56d9223cadb45 commit 5157b4aa5b7de8787b6318e61bcc285031bb9088 upstream. The raid6 recovery code should immediately drop back to the optimized synchronous path when a p+q dma resource is not available. Otherwise we run the non-optimized/multi-pass async code in sync mode. Verified with raid6test (NDISKS=255) Applies to kernels >= 2.6.32. Acked-by: NeilBrown <neilb@suse.de> Reported-by: H. Peter Anvin <hpa@zytor.com> Signed-off-by: Dan Williams <dan.j.williams@intel.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de> 
commit 5157b4aa5b7de8787b6318e61bcc285031bb9088 upstream.

The raid6 recovery code should immediately drop back to the optimized
synchronous path when a p+q dma resource is not available.  Otherwise we
run the non-optimized/multi-pass async code in sync mode.

Verified with raid6test (NDISKS=255)

Applies to kernels >= 2.6.32.

Acked-by: NeilBrown <neilb@suse.de>
Reported-by: H. Peter Anvin <hpa@zytor.com>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
async_tx: expand async raid6 test to cover ioatdma corner case 2009-12-17T20:55:38+00:00 Dan Williams dan.j.williams@intel.com 2009-12-17T20:55:38+00:00 e02a0e47a3f061c1a53fc4376332a988ec047e8a Add explicit 11 and 12 disks cases to exercise the 0 < src_cnt % 8 < 3 corner case in the ioatdma driver. Signed-off-by: Dan Williams <dan.j.williams@intel.com> 
Add explicit 11 and 12 disks cases to exercise the 0 < src_cnt % 8 < 3
corner case in the ioatdma driver.

Signed-off-by: Dan Williams <dan.j.williams@intel.com>
async_tx: build-time toggling of async_{syndrome,xor}_val dma support 2009-11-20T06:21:03+00:00 Dan Williams dan.j.williams@intel.com 2009-11-20T00:10:37+00:00 7b3cc2b1fc2066391e498f3387204908c4eced21 ioat3.2 does not support asynchronous error notifications which makes the driver experience latencies when non-zero pq validate results are expected. Provide a mechanism for turning off async_xor_val and async_syndrome_val via Kconfig. This approach is generally useful for any driver that specifies ASYNC_TX_DISABLE_CHANNEL_SWITCH and would like to force the async_tx api to fall back to the synchronous path for certain operations. Signed-off-by: Dan Williams <dan.j.williams@intel.com> 
ioat3.2 does not support asynchronous error notifications which makes
the driver experience latencies when non-zero pq validate results are
expected.  Provide a mechanism for turning off async_xor_val and
async_syndrome_val via Kconfig.  This approach is generally useful for
any driver that specifies ASYNC_TX_DISABLE_CHANNEL_SWITCH and would like
to force the async_tx api to fall back to the synchronous path for
certain operations.

Signed-off-by: Dan Williams <dan.j.williams@intel.com>
async_tx: fix asynchronous raid6 recovery for ddf layouts 2009-10-20T06:34:46+00:00 Dan Williams dan.j.williams@intel.com 2009-10-19T21:05:12+00:00 da17bf4306fd3a52e938b121df82a7baa10eb282 The raid6 recovery code currently requires special handling of the 4-disk and 5-disk recovery scenarios for the native layout. Quoting from commit 0a82a623: In these situations the default N-disk algorithm will present 0-source or 1-source operations to dma devices. To cover for dma devices where the minimum source count is 2 we implement 4-disk and 5-disk handling in the recovery code. The ddf layout presents disks=6 and disks=7 to the recovery code in these situations. Instead of looking at the number of disks count the number of non-zero sources in the list and call the special case code when the number of non-failed sources is 0 or 1. [neilb@suse.de: replace 'ddf' flag with counting good sources] Signed-off-by: Dan Williams <dan.j.williams@intel.com> 
The raid6 recovery code currently requires special handling of the
4-disk and 5-disk recovery scenarios for the native layout.  Quoting
from commit 0a82a623:

     In these situations the default N-disk algorithm will present
     0-source or 1-source operations to dma devices.  To cover for
     dma devices where the minimum source count is 2 we implement
     4-disk and 5-disk handling in the recovery code.

The ddf layout presents disks=6 and disks=7 to the recovery code in
these situations.  Instead of looking at the number of disks count the
number of non-zero sources in the list and call the special case code
when the number of non-failed sources is 0 or 1.

[neilb@suse.de: replace 'ddf' flag with counting good sources]
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
async_pq: rename scribble page 2009-10-20T06:34:46+00:00 Dan Williams dan.j.williams@intel.com 2009-10-20T01:09:32+00:00 030b07720be0f3bfada12ff6bfa3c61a91212f32 The global scribble page is used as a temporary destination buffer when disabling the P or Q result is requested. The local scribble buffer contains memory for performing address conversions. Rename the global variable to avoid confusion. Signed-off-by: Dan Williams <dan.j.williams@intel.com> 
The global scribble page is used as a temporary destination buffer when
disabling the P or Q result is requested.  The local scribble buffer
contains memory for performing address conversions.  Rename the global
variable to avoid confusion.

Signed-off-by: Dan Williams <dan.j.williams@intel.com>
async_pq: kill a stray dma_map() call and other cleanups 2009-10-20T01:20:20+00:00 Dan Williams dan.j.williams@intel.com 2009-10-20T01:09:32+00:00 5676470f06f783aebf545c8f17ca772911022068 - update the kernel doc for async_syndrome to indicate what NULL in the source list means - whitespace fixups Signed-off-by: Dan Williams <dan.j.williams@intel.com> 
- update the kernel doc for async_syndrome to indicate what NULL in the
  source list means
- whitespace fixups

Signed-off-by: Dan Williams <dan.j.williams@intel.com>
raid6/async_tx: handle holes in block list in async_syndrome_val 2009-10-16T05:40:34+00:00 NeilBrown neilb@suse.de 2009-10-16T05:40:34+00:00 b2141e6951ad56c8f65e70547baeabd5698e390a async_syndrome_val check the P and Q blocks used for RAID6 calculations. With DDF raid6, some of the data blocks might be NULL, so this needs to be handled in the same way that async_gen_syndrome handles it. As async_syndrome_val calls async_xor, also enhance async_xor to detect and skip NULL blocks in the list. Signed-off-by: NeilBrown <neilb@suse.de> 
async_syndrome_val check the P and Q blocks used for RAID6
calculations.
With DDF raid6, some of the data blocks might be NULL, so
this needs to be handled in the same way that async_gen_syndrome
handles it.

As async_syndrome_val calls async_xor, also enhance async_xor
to detect and skip NULL blocks in the list.

Signed-off-by: NeilBrown <neilb@suse.de>
md/async: don't pass a memory pointer as a page pointer. 2009-10-16T05:40:25+00:00 NeilBrown neilb@suse.de 2009-10-16T05:40:25+00:00 5dd33c9a4c29015f6d87568d33521c98931a387e md/raid6 passes a list of 'struct page *' to the async_tx routines, which then either DMA map them for offload, or take the page_address for CPU based calculations. For RAID6 we sometime leave 'blanks' in the list of pages. For CPU based calcs, we want to treat theses as a page of zeros. For offloaded calculations, we simply don't pass a page to the hardware. Currently the 'blanks' are encoded as a pointer to raid6_empty_zero_page. This is a 4096 byte memory region, not a 'struct page'. This is mostly handled correctly but is rather ugly. So change the code to pass and expect a NULL pointer for the blanks. When taking page_address of a page, we need to check for a NULL and in that case use raid6_empty_zero_page. Signed-off-by: NeilBrown <neilb@suse.de> 
md/raid6 passes a list of 'struct page *' to the async_tx routines,
which then either DMA map them for offload, or take the page_address
for CPU based calculations.

For RAID6 we sometime leave 'blanks' in the list of pages.
For CPU based calcs, we want to treat theses as a page of zeros.
For offloaded calculations, we simply don't pass a page to the
hardware.

Currently the 'blanks' are encoded as a pointer to
raid6_empty_zero_page.  This is a 4096 byte memory region, not a
'struct page'.  This is mostly handled correctly but is rather ugly.

So change the code to pass and expect a NULL pointer for the blanks.
When taking page_address of a page, we need to check for a NULL and
in that case use raid6_empty_zero_page.

Signed-off-by: NeilBrown <neilb@suse.de>
async_tx/raid6: add missing dma_unmap calls to the async fail case 2009-09-21T17:47:40+00:00 Dan Williams dan.j.williams@intel.com 2009-09-21T17:47:40+00:00 1f6672d44c1ae7408b43c06170ec34eb0a0e9b9f If we are unable to offload async_mult() or async_sum_product(), then unmap the buffers before falling through to the synchronous path. Signed-off-by: Dan Williams <dan.j.williams@intel.com> 
If we are unable to offload async_mult() or async_sum_product(), then
unmap the buffers before falling through to the synchronous path.

Signed-off-by: Dan Williams <dan.j.williams@intel.com>
raid6test: fix stack overflow 2009-09-17T04:03:29+00:00 Dan Williams dan.j.williams@intel.com 2009-09-17T04:03:29+00:00 1b6df6930994d5d027375b07ac9da63644eb5758 Testing on x86_64 with NDISKS=255 yields: do_IRQ: modprobe near stack overflow (cur:ffff88007d19c000,sp:ffff88007d19c128) ...and eventually general protection fault: 0000 [#1] Moving the scribble buffers off the stack allows the test to complete successfully. Signed-off-by: Dan Williams <dan.j.williams@intel.com> 
Testing on x86_64 with NDISKS=255 yields:

   do_IRQ: modprobe near stack overflow (cur:ffff88007d19c000,sp:ffff88007d19c128)

...and eventually

   general protection fault: 0000 [#1]

Moving the scribble buffers off the stack allows the test to complete
successfully.

Signed-off-by: Dan Williams <dan.j.williams@intel.com>