[PATCH] dmaengine: add slave-dma maintainer

Dan Williams dan.j.williams at intel.com
Mon Feb 14 06:58:43 EST 2011

Slave-dma has become the predominant usage model for dmaengine and needs
special attention.  Memory-to-memory dma usage cases will continue to be
maintained by Dan.

Cc: Alan Cox <alan at linux.intel.com>
Acked-by: Vinod Koul <vinod.koul at intel.com>
Signed-off-by: Dan Williams <dan.j.williams at intel.com>

Several factors have contributed to the need to add a co-maintainer
to dmaengine.  First my time is being consumed in other areas [1].
Also, after Russell's review of amba-pl08x it became clear that the
semantics of the slave interface need to specified [2].  My work on the
iop-adma and ioatdma drivers has been limited to the memory-to-memory
use cases (NET_DMA and ASYNC_TX_DMA).  Without direct slave-dma api
experience (client driver or dma driver) it is difficult to parse its
needs outside of its base similarities with the memory-to-memory case.

I have asked my colleague, Vinod Koul, to:
1/ put more definition around the slave api and usages to try to ensure 
consistent behavior across drivers (as much as possible)
2/ help review and merge new drivers according to this definition

Vinod wrote the intel_mid_dma driver and expects to be involved with
slave-dma for the foreseeable future.

I will continue to maintain the memory-to-memory and related async_tx 
usages of dmaengine (that need fixing [3]), but it is clear that this
usage has become a secondary role for the dmaengine infrastructure.


[1]: http://marc.info/?l=linux-scsi&m=129703780424729&w=2
[2]: http://marc.info/?l=linux-arm-kernel&m=129383280413633&w=2
[3]: http://marc.info/?l=linux-arm-kernel&m=129407269402930&w=2

 1 files changed, 1 insertions(+), 0 deletions(-)

index 55592f8..e2f0992 100644
@@ -2124,6 +2124,7 @@ S:	Supported
 F:	fs/dlm/
+M:	Vinod Koul <vinod.koul at intel.com>
 M:	Dan Williams <dan.j.williams at intel.com>
 S:	Supported
 F:	drivers/dma/

More information about the linux-arm-kernel mailing list