diff options
| author | Ujjal Singh <ujjal.singh@intel.com> | 2017-08-23 03:31:18 +0300 |
|---|---|---|
| committer | Vinod Koul <vinod.koul@intel.com> | 2017-08-23 19:02:21 +0300 |
| commit | 268e2519f5b7101d707a0df32e628e9990bc0da6 (patch) | |
| tree | e7bd1da9ec7fe99d166646c4121504997f4a7cd2 /drivers/dma/ioat/dma.h | |
| parent | 2e442003ab2fbc386bc04b243f6aaaebdbfbb821 (diff) | |
| download | linux-268e2519f5b7101d707a0df32e628e9990bc0da6.tar.xz | |
dmaengine: ioatdma: Add intr_coalesce sysfs entry
We observed performance increase with DMA copy from memory
to MMIO by changing the interrupt coalescing value to 0.
The previous set value was projected on the C5xxx Xeon
platform and no longer holds true. Removing hard coded
value and providing a tune-able in sysfs in order to allow
user to tune this on a per channel basis. By default this
value will be set to 0.
Example of sysfs variable importing for interrupt coalescing
value from command line:
echo 5> /sys/devices/pci0000:00/0000:00:04.0/dma/dma0chan0/
quickdata/intr_coalesce
Reported-by: Nithin Sujir <nsujir@tintri.com>
Signed-off-by: Ujjal Singh <ujjal.singh@intel.com>
Acked-by: Dave Jiang <dave.jiang@intel.com>
Signed-off-by: Vinod Koul <vinod.koul@intel.com>
Diffstat (limited to 'drivers/dma/ioat/dma.h')
| -rw-r--r-- | drivers/dma/ioat/dma.h | 3 |
1 files changed, 3 insertions, 0 deletions
diff --git a/drivers/dma/ioat/dma.h b/drivers/dma/ioat/dma.h index a9bc1a15b0d1..56200eefcf5e 100644 --- a/drivers/dma/ioat/dma.h +++ b/drivers/dma/ioat/dma.h @@ -142,11 +142,14 @@ struct ioatdma_chan { spinlock_t prep_lock; struct ioat_descs descs[2]; int desc_chunks; + int intr_coalesce; + int prev_intr_coalesce; }; struct ioat_sysfs_entry { struct attribute attr; ssize_t (*show)(struct dma_chan *, char *); + ssize_t (*store)(struct dma_chan *, const char *, size_t); }; /** |