[Top][All Lists]
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [Qemu-devel] [PATCH 6/6] ide: Avoid canceling IDE DMA
|
From: |
Kevin Wolf |
|
Subject: |
Re: [Qemu-devel] [PATCH 6/6] ide: Avoid canceling IDE DMA |
|
Date: |
Fri, 18 Mar 2011 12:34:09 +0100 |
|
User-agent: |
Mozilla/5.0 (X11; U; Linux x86_64; en-US; rv:1.9.1.15) Gecko/20101027 Fedora/3.0.10-1.fc12 Thunderbird/3.0.10 |
Am 18.03.2011 12:19, schrieb Christoph Hellwig:
> On Tue, Aug 03, 2010 at 04:44:30PM +0200, Kevin Wolf wrote:
>> Furthermore because the DMA operation is splitted into many synchronous
>> aio_read/write if there's more than one entry in the SG table, without this
>> patch the DMA would be cancelled in the middle, something we've no idea if it
>> happens on real hardware too or not. Overall this seems a great risk for zero
>> gain.
>
> That hasn't been true for a long time when this code was commited, at least
> on kernel supporting preadv/pwritev and/or aio.
So what do you want to tell us? Should the patch be reverted?
I think the request can still be split in multiple bdrv_aio_readv/writev
calls in dma-helpers.c and in non-raw block drivers, so we would still
cancel somewhere in the middle.
Kevin