Re: [Qemu-devel] [PATCHv2] macio: handle non-block ATAPI DMA transfers the same as block DMA transfers

[Aurelien Jarno <aurelien@aurel32.net>]

On 2015-08-13 23:03, Mark Cave-Ayland wrote:
> On 01/08/15 19:33, Aurelien Jarno wrote:
> 
> > On 2015-08-01 17:54, Mark Cave-Ayland wrote:
> >> The existing code incorrectly changes the dma_active flag when a non-block
> >> transfer has completed leading to a hang on newer versions of Linux because the
> >> IDE and DMA engines deadlock waiting for each other.
> >>
> >> Instead copy the buffer directly to RAM, set the remaining transfer size to 0 and
> >> then invoke the ATAPI callback manually once again to correctly finish the
> >> transfer in an identical manner to a block transfer.
> >>
> >> Signed-off-by: Mark Cave-Ayland <mark.cave-ayland@ilande.co.uk>
> >> ---

[snip]

> For some reason pmac_ide_dma_end() was being called without an active
> DMA request and so the RUN bit was clear on entry which was causing the
> error flag to be set. Given that this doesn't occur on lenny and wheezy,
> I'm inclined to believe that this is a bug in the squeeze kernel.
> 
> Anyhow if this survives your testing then I'll repost as 2 separate
> patches with a CC to qemu-stable so that this also gets picked up in the
> next 2.4 release.

Thanks for working on that. I don't feel I have enough knowledge of this
part of QEMU to review the patch, that said, I have been able to test
it and I confirm it fixes the issue.

Reviewed-by: Aurelien Jarno <aurelien@aurel32.net>

-- 
Aurelien Jarno                          GPG: 4096R/1DDD8C9B
aurelien@aurel32.net                 http://www.aurel32.net