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: <[email protected]>
Cc: Dave Jiang <[email protected]>
Cc: Jesse Brandeburg <[email protected]>
Cc: Linus Walleij <[email protected]>
Cc: Maciej Sosnowski <[email protected]>
Reported-by: Richard Scobie <[email protected]>
Signed-off-by: Dan Williams <[email protected]>
This commit is contained in:
Dan Williams 2010-10-13 15:43:10 -07:00
commit c50a898fd4

View file

@ -879,7 +879,7 @@ int __devinit ioat2_dma_probe(struct ioatdma_device *device, int dca)
dma->device_issue_pending = ioat2_issue_pending;
dma->device_alloc_chan_resources = ioat2_alloc_chan_resources;
dma->device_free_chan_resources = ioat2_free_chan_resources;
dma->device_tx_status = ioat_tx_status;
dma->device_tx_status = ioat_dma_tx_status;
err = ioat_probe(device);
if (err)