From: Mark McLoughlin <markmc@redhat.com>
To: Anthony Liguori <aliguori@us.ibm.com>
Cc: qemu-devel@nongnu.org, Avi Kivity <avi@redhat.com>
Subject: Re: [Qemu-devel] [STABLE][PATCH 1/4] Use vectored aiocb storage to store vector translation state
Date: Thu, 14 May 2009 12:38:42 +0100 [thread overview]
Message-ID: <1242301122.17366.22.camel@blaa> (raw)
In-Reply-To: <1242298581-30587-2-git-send-email-markmc@redhat.com>
On Thu, 2009-05-14 at 11:56 +0100, Mark McLoughlin wrote:
> From: Avi Kivity <avi@redhat.com>
>
> Now that we have a dedicated acb pool for vector translation acbs, we can
> store the vector translation state in the acbs instead of in an external
> structure.
>
> Signed-off-by: Avi Kivity <avi@redhat.com>
Signed-off-by: Mark McLoughlin <markmc@redhat.com>
> ---
> block.c | 29 ++++++++++++++---------------
> 1 files changed, 14 insertions(+), 15 deletions(-)
>
> diff --git a/block.c b/block.c
> index b12318f..689ea37 100644
> --- a/block.c
> +++ b/block.c
> @@ -1332,31 +1332,32 @@ char *bdrv_snapshot_dump(char *buf, int buf_size, QEMUSnapshotInfo *sn)
> /**************************************************************/
> /* async I/Os */
>
> -typedef struct VectorTranslationState {
> +typedef struct VectorTranslationAIOCB {
> + BlockDriverAIOCB common;
> QEMUIOVector *iov;
> uint8_t *bounce;
> int is_write;
> BlockDriverAIOCB *aiocb;
> - BlockDriverAIOCB *this_aiocb;
> -} VectorTranslationState;
> +} VectorTranslationAIOCB;
>
> -static void bdrv_aio_cancel_vector(BlockDriverAIOCB *acb)
> +static void bdrv_aio_cancel_vector(BlockDriverAIOCB *_acb)
> {
> - VectorTranslationState *s = acb->opaque;
> + VectorTranslationAIOCB *acb
> + = container_of(_acb, VectorTranslationAIOCB, common);
>
> - bdrv_aio_cancel(s->aiocb);
> + bdrv_aio_cancel(acb->aiocb);
> }
>
> static void bdrv_aio_rw_vector_cb(void *opaque, int ret)
> {
> - VectorTranslationState *s = opaque;
> + VectorTranslationAIOCB *s = (VectorTranslationAIOCB *)opaque;
>
> if (!s->is_write) {
> qemu_iovec_from_buffer(s->iov, s->bounce, s->iov->size);
> }
> qemu_vfree(s->bounce);
> - s->this_aiocb->cb(s->this_aiocb->opaque, ret);
> - qemu_aio_release(s->this_aiocb);
> + s->common.cb(s->common.opaque, ret);
> + qemu_aio_release(s);
> }
>
> static BlockDriverAIOCB *bdrv_aio_rw_vector(BlockDriverState *bs,
> @@ -1368,11 +1369,9 @@ static BlockDriverAIOCB *bdrv_aio_rw_vector(BlockDriverState *bs,
> int is_write)
>
> {
> - VectorTranslationState *s = qemu_mallocz(sizeof(*s));
> - BlockDriverAIOCB *aiocb = qemu_aio_get_pool(&vectored_aio_pool, bs,
> - cb, opaque);
> + VectorTranslationAIOCB *s = qemu_aio_get_pool(&vectored_aio_pool, bs,
> + cb, opaque);
>
> - s->this_aiocb = aiocb;
> s->iov = iov;
> s->bounce = qemu_memalign(512, nb_sectors * 512);
> s->is_write = is_write;
> @@ -1384,7 +1383,7 @@ static BlockDriverAIOCB *bdrv_aio_rw_vector(BlockDriverState *bs,
> s->aiocb = bdrv_aio_read(bs, sector_num, s->bounce, nb_sectors,
> bdrv_aio_rw_vector_cb, s);
> }
> - return aiocb;
> + return &s->common;
> }
>
> BlockDriverAIOCB *bdrv_aio_readv(BlockDriverState *bs, int64_t sector_num,
> @@ -1560,7 +1559,7 @@ static int bdrv_write_em(BlockDriverState *bs, int64_t sector_num,
>
> void bdrv_init(void)
> {
> - aio_pool_init(&vectored_aio_pool, sizeof(BlockDriverAIOCB),
> + aio_pool_init(&vectored_aio_pool, sizeof(VectorTranslationAIOCB),
> bdrv_aio_cancel_vector);
>
> bdrv_register(&bdrv_raw);
next prev parent reply other threads:[~2009-05-14 11:38 UTC|newest]
Thread overview: 17+ messages / expand[flat|nested] mbox.gz Atom feed top
2009-04-20 22:13 [Qemu-devel] [STABLE][PATCH 0/3] Rebase aio dma cancellation work Ryan Harper
2009-04-20 22:13 ` [Qemu-devel] [STABLE][PATCH 1/3] Refactor aio callback allocation to use an aiocb pool Ryan Harper
2009-04-20 22:13 ` [Qemu-devel] [STABLE][PATCH 2/3] Convert vectored aio emulation to use a dedicated pool Ryan Harper
2009-04-20 22:13 ` [Qemu-devel] [STABLE][PATCH 3/3] Implement cancellation method for dma async I/O Ryan Harper
2009-04-21 21:35 ` [Qemu-devel] Re: [STABLE][PATCH 0/3] Rebase aio dma cancellation work Anthony Liguori
2009-04-22 11:03 ` Avi Kivity
2009-05-14 10:56 ` [Qemu-devel] [STABLE][PATCH 0/4] Rebase more " Mark McLoughlin
2009-05-14 10:56 ` [Qemu-devel] [STABLE][PATCH 1/4] Use vectored aiocb storage to store vector translation state Mark McLoughlin
2009-05-14 10:56 ` [Qemu-devel] [STABLE][PATCH 2/4] Move block dma helpers aiocb to store dma state Mark McLoughlin
2009-05-14 10:56 ` [Qemu-devel] [STABLE][PATCH 3/4] Fix vectored aio bounce handling immediate errors Mark McLoughlin
2009-05-14 10:56 ` [Qemu-devel] [STABLE][PATCH 4/4] Fix DMA API when handling an immediate error from block layer Mark McLoughlin
2009-05-14 11:39 ` Mark McLoughlin
2009-05-14 11:39 ` [Qemu-devel] [STABLE][PATCH 3/4] Fix vectored aio bounce handling immediate errors Mark McLoughlin
2009-05-14 11:39 ` [Qemu-devel] [STABLE][PATCH 2/4] Move block dma helpers aiocb to store dma state Mark McLoughlin
2009-05-14 11:38 ` Mark McLoughlin [this message]
2009-05-14 11:28 ` [Qemu-devel] Re: [STABLE][PATCH 0/4] Rebase more aio dma cancellation work Avi Kivity
2009-05-14 14:38 ` [Qemu-devel] " Anthony Liguori
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=1242301122.17366.22.camel@blaa \
--to=markmc@redhat.com \
--cc=aliguori@us.ibm.com \
--cc=avi@redhat.com \
--cc=qemu-devel@nongnu.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).