[PATCH v2 00/12] dmaengine: Add dmaengine API to avoid call DMA Engine callback directly

Frank.Li@oss.nxp.com posted 12 patches 20 hours ago
crypto/async_tx/async_memcpy.c      |  18 +--
crypto/async_tx/async_pq.c          |  60 ++++----
crypto/async_tx/async_raid6_recov.c |  22 ++-
crypto/async_tx/async_tx.c          |  19 +--
crypto/async_tx/async_xor.c         |  58 ++++---
drivers/dma/dmaengine.c             |   4 +-
drivers/dma/dmatest.c               |   4 +-
drivers/dma/mv_xor.c                |   4 +-
drivers/ntb/ntb_transport.c         |   4 +-
drivers/ntb/test/ntb_perf.c         |   2 +-
include/linux/async_tx.h            |   8 +-
include/linux/dmaengine.h           | 296 +++++++++++++++++++++++++++++++++++-
12 files changed, 383 insertions(+), 116 deletions(-)
[PATCH v2 00/12] dmaengine: Add dmaengine API to avoid call DMA Engine callback directly
Posted by Frank.Li@oss.nxp.com 20 hours ago
Add dmaengine APIs to avoid call DMA Engine callback directly. Prepare to
make dmaengine provider private.

Fix all under crypto/async_tx

Build pass under crypto/async_tx after change dma_device *device to _device.
to make sure all code use dmaengine APIs.

Assisted-by: LLM
Signed-off-by: Frank Li <Frank.Li@nxp.com>
---
Changes in v2:
- Add more patch to cleanup crypto/async_tx
- collect https://lore.kernel.org/imx/b40fa0d7-fbd0-4ee8-b99c-781d08c18838@intel.com/T/#t
- first two patches can be applied without added new dmaengine apis.
- Link to v1: https://patch.msgid.link/20260921-dmaengine_prep_dma_pq-v1-0-9563ed448260@nxp.com

To: Dan Williams <djbw@kernel.org>
To: Herbert Xu <herbert@gondor.apana.org.au>
To: "David S. Miller" <davem@davemloft.net>
To: Vinod Koul <vkoul@kernel.org>
To: Frank Li <Frank.Li@kernel.org>
To: Jon Mason <jdmason@kudzu.us>
To: Dave Jiang <dave.jiang@intel.com>
To: Allen Hubbe <allenbh@gmail.com>
Cc: linux-crypto@vger.kernel.org
Cc: linux-kernel@vger.kernel.org
Cc: dmaengine@vger.kernel.org
Cc: ntb@lists.linux.dev
dmaengine: change dmaengine_get_unmap_data() first arg to dma_chan into this series

---
Frank Li (12):
      async_tx: replace direct ->device_prep*() calls with standard DMA engine API
      async_tx: use dmaengine_get_dma_device() instead of chan->device->dev
      dmaengine: add dmaengine_prep_dma_(pq|pq_val|interrupt|xor)() API
      dmaengine: add dmaengine_is_*_aligned() helpers for DMA consumers
      dmaengine: add dmaengine_get_copy_align() and related alignment getter helpers
      dmaengine: add dmaengine_get_cap_mask() and dmaengine_has_cap() helpers
      dmaengine: add dmaengine_get_max_xor() helper
      dmaengine: change dmaengine_get_unmap_data() first arg to dma_chan
      dmaengine: replace dma_maxqp() with dmaengine_maxpq() taking struct dma_chan *
      async_tx: use dmaengine_prep_dma_(pq|pq_val|interrupt|xor) instead of direct callback
      async_tx: convert to dmaengine_is_*_aligned() helpers
      async_tx: replace open-coded cap_mask accesses with dmaengine_has_cap()

 crypto/async_tx/async_memcpy.c      |  18 +--
 crypto/async_tx/async_pq.c          |  60 ++++----
 crypto/async_tx/async_raid6_recov.c |  22 ++-
 crypto/async_tx/async_tx.c          |  19 +--
 crypto/async_tx/async_xor.c         |  58 ++++---
 drivers/dma/dmaengine.c             |   4 +-
 drivers/dma/dmatest.c               |   4 +-
 drivers/dma/mv_xor.c                |   4 +-
 drivers/ntb/ntb_transport.c         |   4 +-
 drivers/ntb/test/ntb_perf.c         |   2 +-
 include/linux/async_tx.h            |   8 +-
 include/linux/dmaengine.h           | 296 +++++++++++++++++++++++++++++++++++-
 12 files changed, 383 insertions(+), 116 deletions(-)
---
base-commit: 5c4d4169604b335c38bbc79bc1fc03042981fc6f
change-id: 20260921-dmaengine_prep_dma_pq-7eba91a609ed

Best regards,
--  
Frank Li <Frank.Li@nxp.com>