Re: [PATCH 11/12] DMAENGINE: let PL08X memcpy TXDs wait

[Dan Williams]

On Thu, Sep 23, 2010 at 12:46 AM, Linus Walleij
<linus.ml.walleij@xxxxxxxxx> wrote:
> 2010/9/23 Dan Williams <dan.j.williams@xxxxxxxxx>:
>
>> The prep routine is meant to guarantee that all the resources for a
>> transaction have been acquired.
>
> A practicality here:
>
> The assumption to fail for physical channel allocation at .tx_submit()
> is already in the first patch adding the driver.
>
> Should I squash all the icremental improvements into the original
> driver, convert to fail at .prep() and resubmit to make all of this
> easier, or do you want me to refactor the patch stack, or refactor
> from patch 11/12 or...?
>

At this point I am already rebasing my next branch to catch up with
the backlog, so a new v5 of the driver with all the roll-ups works for
me.

Also not sure if you caught this comment from earlier:
> sidenote: this driver needs to be converted to a dma descriptor pool
> (allocated at device_alloc_chan_resources time), or at a minimum use
> GFP_NOWAIT in device_prep_dma_memcpy because that routine may be
> called from atomic contexts.

Thanks,
Dan
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/