Re: [PATCH 0/9] dmaengine: Support bus widths of 32 bytes and above

On Fri, Jul 31, 2026 at 04:34:16PM +0100, Nuno Sá wrote:
> On Thu, Jul 30, 2026 at 02:05:43PM -0500, Frank Li wrote:
> > On Thu, Jul 30, 2026 at 03:23:07PM +0100, Nuno Sá via B4 Relay wrote:
> > > [You don't often get email from [email protected]. Learn why this is important at https://aka.ms/LearnAboutSenderIdentification ]
> > >
> > > The DMA engine slave capabilities advertise the supported source and
> > > destination bus widths through src_addr_widths / dst_addr_widths. These
> > > are plain u32 bitmasks where a set bit's position equals the
> > > corresponding enum dma_slave_buswidth value, e.g.
> > > DMA_SLAVE_BUSWIDTH_4_BYTES sets bit 4.
> > >
> > > The consequence is that widths of 32 bytes and above cannot be
> > > represented at all: DMA_SLAVE_BUSWIDTH_32/64/128_BYTES would need bits
> > > 32, 64 and 128, which do not fit in a u32. Hardware with wider data
> > > paths is becoming common, so add a representation that can express these
> > > widths while still using enum dma_slave_buswidth.
> > >
> > > This series switches consumers and a small set of producers to bitmap
> > > based bus width capabilities. The legacy dma_device u32 fields are kept
> > > for now so the remaining DMA controller drivers can be converted
> > > incrementally. dma_get_slave_caps() folds legacy producer masks into the
> > > new bitmap representation returned to consumers.
> > >
> > > Once the remaining producers are converted, the legacy dma_device
> > > src/dst_addr_widths fields can be removed as a final cleanup.
> > >
> > > This issue was discussed before here:
> > >
> > > https://lore.kernel.org/dmaengine/abkoXXbaxaiqbBuX@vaman/
> > >
> > > ---
> > > Changes in v1:
> >
> > this should be v2, RFC is v1
>
> Ups! Sorry about that! But I guess now we keep this path? I mean, I
> should not bump to v3 next right?

Needn't bump now. wait for other persion's review.

Frank

>
> - Nuno Sá
>
> >
> > Frank
> > > - Drop RFC tag.
> > > - Rename the new capability representation and helpers from address
> > >   width terminology to bus width terminology.
> > > - Convert all current dma_get_slave_caps() consumers that used the
> > >   legacy src/dst_addr_widths fields.
> > > - Convert a small set of DMA controller producers and the stm32-dma3
> > >   per-channel capability callback.
> > > - Add helper documentation for the new public API.
> > > - Link to RFC: https://lore.kernel.org/dmaengine/20260616-dmaengine-support-wider-dma-masks-v1-0-da23a8dcb756@analog.com/
> > >
> > > ---
> > > Nuno Sá (9):
> > >       dmaengine: Support bus widths of 32 bytes and above
> > >       dmaengine: dma-axi-dmac: Use bus width capability helpers
> > >       dmaengine: dw-axi-dmac: Use bus width capability helpers
> > >       dmaengine: qcom: gpi: Use bus width capability helpers
> > >       dmaengine: stm32-dma3: Use bus width capability helpers
> > >       iio: buffer-dmaengine: Use dma_slave_caps bus width accessors
> > >       ALSA: pcm_dmaengine: Use dma_slave_caps bus width helpers
> > >       spi: dw: Use dma_slave_caps bus width helpers
> > >       dmaengine: Drop legacy bus width fields from dma_slave_caps
> > >
> > >  drivers/dma/dma-axi-dmac.c                         |  13 +-
> > >  drivers/dma/dmaengine.c                            |  20 +-
> > >  drivers/dma/dw-axi-dmac/dw-axi-dmac-platform.c     |  39 ++--
> > >  drivers/dma/qcom/gpi.c                             |  11 +-
> > >  drivers/dma/stm32/stm32-dma3.c                     |  32 +--
> > >  drivers/iio/buffer/industrialio-buffer-dmaengine.c |  15 +-
> > >  drivers/spi/spi-dw-dma.c                           |   6 +-
> > >  drivers/spi/spi-dw.h                               |   3 +-
> > >  include/linux/dmaengine.h                          | 234 ++++++++++++++++++++-
> > >  sound/core/pcm_dmaengine.c                         |  14 +-
> > >  10 files changed, 323 insertions(+), 64 deletions(-)
> > > ---
> > > base-commit: 7524fe142b5a772f8421aeee2132cf7e21a00103
> > > change-id: 20260615-dmaengine-support-wider-dma-masks-5aac12497e27
> > > --
> > >
> > > Thanks!
> > > - Nuno Sá
> > >
> > >