mbox series

[v4,0/6] dmaengine: fsl-edma: integrate TCD64 support for 64bit physical address

Message ID 20231221153528.1588049-1-Frank.Li@nxp.com (mailing list archive)
Headers show
Series dmaengine: fsl-edma: integrate TCD64 support for 64bit physical address | expand

Message

Frank Li Dec. 21, 2023, 3:35 p.m. UTC
Change from v3 to v4.
Fixed tcd64 type as fsl_edma_hw_tcd64

Change from v2 to v3:
 - fix sparse build warning

Change from v1 to v2:
- fixed mcf-edma-main.c build error.
- fixed readq build error. readq actually is not atomic read in imx95.
So split to two ioread32\iowrite32.
  It needs read at least twice to avoid lower 32 bit part wrap during read
up 32bit part.

first 2 patch is prepare, No function change.
3rd patch is dt-bind doc
4rd patch is actuall support TCD64

Frank Li (6):
  dmaengine: fsl-edma: involve help macro fsl_edma_set(get)_tcd()
  dmaengine: fsl-edma: fix spare build warning
  dmaengine: fsl-edma: add address for channel mux register in
    fsl_edma_chan
  dmaengine: mcf-edma: utilize edma_write_tcdreg() macro for TCD Access
  dt-bindings: fsl-dma: fsl-edma: add fsl,imx95-edma5 compatible string
  dmaengine: fsl-edma: integrate TCD64 support for i.MX95

 .../devicetree/bindings/dma/fsl,edma.yaml     |   2 +
 drivers/dma/fsl-edma-common.c                 | 101 ++++++-----
 drivers/dma/fsl-edma-common.h                 | 161 ++++++++++++++++--
 drivers/dma/fsl-edma-main.c                   |  19 ++-
 drivers/dma/mcf-edma-main.c                   |   2 +-
 5 files changed, 223 insertions(+), 62 deletions(-)

Comments

Frank Li Jan. 22, 2024, 4:55 p.m. UTC | #1
On Thu, Dec 21, 2023 at 10:35:22AM -0500, Frank Li wrote:
> Change from v3 to v4.	
> Fixed tcd64 type as fsl_edma_hw_tcd64
> 
> Change from v2 to v3:
>  - fix sparse build warning
> 
> Change from v1 to v2:
> - fixed mcf-edma-main.c build error.
> - fixed readq build error. readq actually is not atomic read in imx95.
> So split to two ioread32\iowrite32.
>   It needs read at least twice to avoid lower 32 bit part wrap during read
> up 32bit part.
> 
> first 2 patch is prepare, No function change.
> 3rd patch is dt-bind doc
> 4rd patch is actuall support TCD64

@vnod:

	Could you please check these patches? I still have more patches,
which depended on this.
	eDMA is used for cross whole i.MX chips.
		
Frank

> 
> Frank Li (6):
>   dmaengine: fsl-edma: involve help macro fsl_edma_set(get)_tcd()
>   dmaengine: fsl-edma: fix spare build warning
>   dmaengine: fsl-edma: add address for channel mux register in
>     fsl_edma_chan
>   dmaengine: mcf-edma: utilize edma_write_tcdreg() macro for TCD Access
>   dt-bindings: fsl-dma: fsl-edma: add fsl,imx95-edma5 compatible string
>   dmaengine: fsl-edma: integrate TCD64 support for i.MX95
> 
>  .../devicetree/bindings/dma/fsl,edma.yaml     |   2 +
>  drivers/dma/fsl-edma-common.c                 | 101 ++++++-----
>  drivers/dma/fsl-edma-common.h                 | 161 ++++++++++++++++--
>  drivers/dma/fsl-edma-main.c                   |  19 ++-
>  drivers/dma/mcf-edma-main.c                   |   2 +-
>  5 files changed, 223 insertions(+), 62 deletions(-)
> 
> -- 
> 2.34.1
>
Frank Li Jan. 31, 2024, 3:33 p.m. UTC | #2
On Mon, Jan 22, 2024 at 11:55:26AM -0500, Frank Li wrote:
> On Thu, Dec 21, 2023 at 10:35:22AM -0500, Frank Li wrote:
> > Change from v3 to v4.	
> > Fixed tcd64 type as fsl_edma_hw_tcd64
> > 
> > Change from v2 to v3:
> >  - fix sparse build warning
> > 
> > Change from v1 to v2:
> > - fixed mcf-edma-main.c build error.
> > - fixed readq build error. readq actually is not atomic read in imx95.
> > So split to two ioread32\iowrite32.
> >   It needs read at least twice to avoid lower 32 bit part wrap during read
> > up 32bit part.
> > 
> > first 2 patch is prepare, No function change.
> > 3rd patch is dt-bind doc
> > 4rd patch is actuall support TCD64
> 
> @vnod:
> 
> 	Could you please check these patches? I still have more patches,
> which depended on this.
> 	eDMA is used for cross whole i.MX chips.
> 		
> Frank

@Vinod:
	Ping?

Frank

> 
> > 
> > Frank Li (6):
> >   dmaengine: fsl-edma: involve help macro fsl_edma_set(get)_tcd()
> >   dmaengine: fsl-edma: fix spare build warning
> >   dmaengine: fsl-edma: add address for channel mux register in
> >     fsl_edma_chan
> >   dmaengine: mcf-edma: utilize edma_write_tcdreg() macro for TCD Access
> >   dt-bindings: fsl-dma: fsl-edma: add fsl,imx95-edma5 compatible string
> >   dmaengine: fsl-edma: integrate TCD64 support for i.MX95
> > 
> >  .../devicetree/bindings/dma/fsl,edma.yaml     |   2 +
> >  drivers/dma/fsl-edma-common.c                 | 101 ++++++-----
> >  drivers/dma/fsl-edma-common.h                 | 161 ++++++++++++++++--
> >  drivers/dma/fsl-edma-main.c                   |  19 ++-
> >  drivers/dma/mcf-edma-main.c                   |   2 +-
> >  5 files changed, 223 insertions(+), 62 deletions(-)
> > 
> > -- 
> > 2.34.1
> >
Vinod Koul Feb. 7, 2024, 8:49 a.m. UTC | #3
On Thu, 21 Dec 2023 10:35:22 -0500, Frank Li wrote:
> Change from v3 to v4.
> Fixed tcd64 type as fsl_edma_hw_tcd64
> 
> Change from v2 to v3:
>  - fix sparse build warning
> 
> Change from v1 to v2:
> - fixed mcf-edma-main.c build error.
> - fixed readq build error. readq actually is not atomic read in imx95.
> So split to two ioread32\iowrite32.
>   It needs read at least twice to avoid lower 32 bit part wrap during read
> up 32bit part.
> 
> [...]

Applied, thanks!

[1/6] dmaengine: fsl-edma: involve help macro fsl_edma_set(get)_tcd()
      commit: 5dc604455dcf20bdca639bf6b8ea2ea60d39c022
[2/6] dmaengine: fsl-edma: fix spare build warning
      commit: 537df9ab2d72bb782926a7d263a9f0a101e60b2e
[3/6] dmaengine: fsl-edma: add address for channel mux register in fsl_edma_chan
      commit: e0a08ed25492b6437e366b347113db484037b9b9
[4/6] dmaengine: mcf-edma: utilize edma_write_tcdreg() macro for TCD Access
      commit: b51dd7c8aac292396d038d0a9fb9c1589addb515
[5/6] dt-bindings: fsl-dma: fsl-edma: add fsl,imx95-edma5 compatible string
      commit: b7b8715b430ee4431bd675687c5bcda113e7ddd4
[6/6] dmaengine: fsl-edma: integrate TCD64 support for i.MX95
      commit: de7d9cb3b064fdfb2e0e7706d14ffee20b762ad2

Best regards,