* Re: [alsa-devel] [PATCH 1/2] dma: Add a PAUSE_RESUME capability
[not found] ` <1369932575.3890.18.camel@vkoul-udesk3>
@ 2013-05-30 19:10 ` Lars-Peter Clausen
0 siblings, 0 replies; only message in thread
From: Lars-Peter Clausen @ 2013-05-30 19:10 UTC (permalink / raw)
To: Vinod Koul; +Cc: linux kernel, vinod.koul
On 05/30/2013 06:49 PM, Vinod Koul wrote:
> On Wed, 2013-05-29 at 13:42 +0200, Lars-Peter Clausen wrote:
>> Add a capability which allows dmaengine consumers to query whether a dmaengine
>> controller supports the DMA_PAUSE and DMA_RESUME commands. Sometimes a user of
>> the dmaengine API needs to know in advance if a channel will support
>> pause/resume. E.g. in ALSA the information whether it is possible to
>> pause/resume a audio stream or whether it is necessary to abort and restart the
>> stream is communicated to userspace so it can act accordingly.
> - This mail-list is not correct for this patch
> - use get_maintainer.pl, if you dont know but you already know!
> - always CC maintainers
Hm, I have no idea what happened with the Cc list. I might have hit the up key
one time to many after --dry-run, since this is the Cc list of a different
patch series.
>
> the enum dma_transaction_type reports what kind of transactions DMA
> controller is capable of. This is not for capabilities of a given
> transaction type, so this approach is not right.
hm, ok.
>
> Unfortunately, this is not supported by dmaengine atm. We had discussed
> this in last plumbers, and discussion with Matt was to add capability
> API (at least for slave usages). The work from Matt had reached a good
> point but unfortunately didn't reach its conclusion. [1]
>
> Let me resurrect those patches and add the pause, resume query to it in
> more generic way
I've pinged Matt about these patches yesterday or so. He has been rather quite
lately, maybe he's been sucked into the Linaro blackhole.
- Lars
^ permalink raw reply [flat|nested] only message in thread