linux/drivers/dma/ioat
Dan Williams c50a898fd4 ioat2: fix performance regression
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>
2010-10-13 15:43:10 -07:00
..
dca.c ioat3: dca and raid operations are incompatible 2009-11-19 17:07:10 -07:00
dma.c Merge branch 'ioat' into dmaengine 2010-05-17 16:30:58 -07:00
dma.h ioat2: catch and recover from broken vtd configurations v6 2010-08-04 14:18:17 -07:00
dma_v2.c ioat2: fix performance regression 2010-10-13 15:43:10 -07:00
dma_v2.h ioat2,3: convert to producer/consumer locking 2010-05-01 15:22:55 -07:00
dma_v3.c ioat2: catch and recover from broken vtd configurations v6 2010-08-04 14:18:17 -07:00
hw.h ioat3: dca and raid operations are incompatible 2009-11-19 17:07:10 -07:00
Makefile ioat3: split ioat3 support to its own file, add memset 2009-09-08 17:42:55 -07:00
pci.c ioat: Remove duplicated devm_kzalloc() calls for ioatdma_device 2010-05-02 12:52:35 -07:00
registers.h ioat3: interrupt coalescing 2010-03-03 21:21:13 -07:00