[PATCH 0/5 v2] sh: dmaengine driver improvements

[PATCH 0/5 v2] sh: dmaengine driver improvements

[Guennadi Liakhovetski]

V2 of this patchset addresses comments to v1 - as is easy to guess;) 
Specific chenges are described in respective patches. One more patch is 
added, described in (4) below. Here again the description from v1:

This patch series introduces several improvements to the shdma SH 
dmaengine driver:

1. (v2) fix handling of DMA error interrupts

2. better use platform-dvice infrastructures, resources to describe DMAC 
implementation details, ioremap address ranges, instead of poking at 
physical addresses directly.

3. separate SH headers for the legacy arch/sh/drivers/dma and the 
dmaengine driver

4. add runtime PM support to shdma dmaengine driver

This series applies on top of my previous ASoC (SIU) and SH / dmaengine 
patches.

Thanks
Guennadi
---
Guennadi Liakhovetski, Ph.D.
Freelance Open-Source Software Developer
http://www.open-technology.de/

Re: [PATCH 0/5 v2] sh: dmaengine driver improvements

[Paul Mundt]

On Thu, Feb 11, 2010 at 05:50:00PM +0100, Guennadi Liakhovetski wrote:
> V2 of this patchset addresses comments to v1 - as is easy to guess;) 
> Specific chenges are described in respective patches. One more patch is 
> added, described in (4) below. Here again the description from v1:
> 
> This patch series introduces several improvements to the shdma SH 
> dmaengine driver:
> 
> 1. (v2) fix handling of DMA error interrupts
> 
> 2. better use platform-dvice infrastructures, resources to describe DMAC 
> implementation details, ioremap address ranges, instead of poking at 
> physical addresses directly.
> 
> 3. separate SH headers for the legacy arch/sh/drivers/dma and the 
> dmaengine driver
> 
> 4. add runtime PM support to shdma dmaengine driver
> 
> This series applies on top of my previous ASoC (SIU) and SH / dmaengine 
> patches.
> 
This definitely looks much better than v1, and should give us a pretty
good foundation for killing off the legacy DMA support incrementally.

Given the dependencies on the SIU patches, I'll roll these in for the
second round of updates during the merge window once the ASoC tree has
merged. This leaves plenty of time for Dan or Iwamatsu-san to voice any
objections in the interim.

Re: [PATCH 0/5 v2] sh: dmaengine driver improvements

[Paul Mundt]

On Fri, Feb 12, 2010 at 12:54:58PM +0900, Paul Mundt wrote:
> On Thu, Feb 11, 2010 at 05:50:00PM +0100, Guennadi Liakhovetski wrote:
> > V2 of this patchset addresses comments to v1 - as is easy to guess;) 
> > Specific chenges are described in respective patches. One more patch is 
> > added, described in (4) below. Here again the description from v1:
> > 
> > This patch series introduces several improvements to the shdma SH 
> > dmaengine driver:
> > 
> > 1. (v2) fix handling of DMA error interrupts
> > 
> > 2. better use platform-dvice infrastructures, resources to describe DMAC 
> > implementation details, ioremap address ranges, instead of poking at 
> > physical addresses directly.
> > 
> > 3. separate SH headers for the legacy arch/sh/drivers/dma and the 
> > dmaengine driver
> > 
> > 4. add runtime PM support to shdma dmaengine driver
> > 
> > This series applies on top of my previous ASoC (SIU) and SH / dmaengine 
> > patches.
> > 
> This definitely looks much better than v1, and should give us a pretty
> good foundation for killing off the legacy DMA support incrementally.
> 
> Given the dependencies on the SIU patches, I'll roll these in for the
> second round of updates during the merge window once the ASoC tree has
> merged. This leaves plenty of time for Dan or Iwamatsu-san to voice any
> objections in the interim.

I've applied these now that the ASoC tree has merged.