diff options
| author | Dave Jiang <dave.jiang@intel.com> | 2016-07-25 19:34:08 +0200 |
|---|---|---|
| committer | Vinod Koul <vinod.koul@intel.com> | 2016-08-08 04:41:43 +0200 |
| commit | 8058e25809f53cadc0438ebb8f920415a0d2ec17 (patch) | |
| tree | d4a90a12d29b88b18d5a9b9db073ca442a86876c | |
| parent | dmaengine: fsl_raid: move unmap to before callback (diff) | |
| download | linux-8058e25809f53cadc0438ebb8f920415a0d2ec17.tar.xz linux-8058e25809f53cadc0438ebb8f920415a0d2ec17.zip | |
dmaengine: mv_xor: move unmap to before callback
Completion callback should happen after dma_descriptor_unmap() has
happened. This allow the cache invalidate to happen and ensure that
the data accessed by the upper layer is in memory that was from DMA
rather than stale data. On some architecture this is done by the
hardware, however we should make the code consistent to not cause
confusion.
Signed-off-by: Dave Jiang <dave.jiang@intel.com>
Signed-off-by: Vinod Koul <vinod.koul@intel.com>
| -rw-r--r-- | drivers/dma/mv_xor.c | 2 |
1 files changed, 1 insertions, 1 deletions
diff --git a/drivers/dma/mv_xor.c b/drivers/dma/mv_xor.c index f8b5e7424b3a..d550efbc7054 100644 --- a/drivers/dma/mv_xor.c +++ b/drivers/dma/mv_xor.c @@ -206,11 +206,11 @@ mv_desc_run_tx_complete_actions(struct mv_xor_desc_slot *desc, if (desc->async_tx.cookie > 0) { cookie = desc->async_tx.cookie; + dma_descriptor_unmap(&desc->async_tx); /* call the callback (must not sleep or submit new * operations to this channel) */ dmaengine_desc_get_callback_invoke(&desc->async_tx, NULL); - dma_descriptor_unmap(&desc->async_tx); } /* run dependent operations */ |