Dan Williams b228f7fdef raid6: fix recovery performance regression ...

commit 5157b4aa5b 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>

2010-05-12 14:57:05 -07:00

..

async_memcpy.c

dmaengine, async_tx: support alignment checks

2009-09-08 17:42:53 -07:00

async_memset.c

dmaengine, async_tx: support alignment checks

2009-09-08 17:42:53 -07:00

async_pq.c

async_tx: build-time toggling of async_{syndrome,xor}_val dma support

2009-11-19 23:21:03 -07:00

async_raid6_recov.c

raid6: fix recovery performance regression

2010-05-12 14:57:05 -07:00

async_tx.c

dmaengine, async_tx: add a "no channel switch" allocator

2009-09-08 17:42:51 -07:00

async_xor.c

async_tx: build-time toggling of async_{syndrome,xor}_val dma support

2009-11-19 23:21:03 -07:00

Kconfig

async_tx: build-time toggling of async_{syndrome,xor}_val dma support

2009-11-19 23:21:03 -07:00

Makefile

async_tx: raid6 recovery self test

2009-08-29 19:09:28 -07:00

raid6test.c

raid6test: fix stack overflow

2009-09-16 21:03:29 -07:00