From: Wenchao Xia <xiawenc@linux.vnet.ibm.com>
To: Kevin Wolf <kwolf@redhat.com>, qemu-devel@nongnu.org
Cc: pbonzini@redhat.com, armbru@redhat.com, stefanha@redhat.com
Subject: Re: [Qemu-devel] [RFC PATCH 03/19] block: Don't use guest sector size for qemu_blockalign()
Date: Tue, 10 Dec 2013 11:18:16 +0800 [thread overview]
Message-ID: <52A687F8.8000205@linux.vnet.ibm.com> (raw)
In-Reply-To: <1386350580-5666-4-git-send-email-kwolf@redhat.com>
于 2013/12/7 1:22, Kevin Wolf 写道:
> bs->buffer_alignment is set by the device emulation and contains the
> logical block size of the guest device. This isn't something that the
> block layer should know, and even less something to use for determining
> the right alignment of buffers to be used for the host.
>
> The new function bdrv_opt_mem_align() allows for hooks in a BlockDriver
> so that it can tell the qemu block layer the optimal alignment to be
> used so that no bounce buffer must be used in the driver.
>
> This patch may change the buffer alignment from 4k to 512 for all
> callers that used qemu_blockalign() with the top-level image format
> BlockDriverState. The value was never propagated to other levels in the
> tree, so in particular raw-posix never required anything else than 512.
>
> While on disks with 4k sectors direct I/O requires a 4k alignment,
> memory may still be okay when aligned to 512 byte boundaries. This is
> what must have happened in practice, because otherwise this would
> already have failed earlier. Therefore I don't expect regressions even
> with this intermediate state. Later, raw-posix can implement the hook
> and expose a different memory alignment requirement.
>
> Signed-off-by: Kevin Wolf <kwolf@redhat.com>
> ---
> block.c | 32 +++++++++++++++++++++++++++++---
> include/block/block.h | 1 +
> include/block/block_int.h | 4 ++++
> 3 files changed, 34 insertions(+), 3 deletions(-)
>
> diff --git a/block.c b/block.c
> index 613201b..669793b 100644
> --- a/block.c
> +++ b/block.c
> @@ -213,6 +213,31 @@ static void bdrv_io_limits_intercept(BlockDriverState *bs,
> qemu_co_queue_next(&bs->throttled_reqs[is_write]);
> }
>
> +size_t bdrv_opt_mem_align(BlockDriverState *bs)
> +{
> + size_t alignment;
> +
> + if (!bs || !bs->drv) {
> + /* 4k should be on the safe side */
> + return 4096;
> + }
> +
> + if (bs->drv->bdrv_opt_mem_align) {
> + return bs->drv->bdrv_opt_mem_align(bs);
> + }
> +
> + if (bs->file) {
> + alignment = bdrv_opt_mem_align(bs->file);
> + } else {
> + alignment = 512;
> + }
> +
> + if (bs->backing_hd) {
> + alignment = MAX(alignment, bdrv_opt_mem_align(bs->backing_hd));
> + }
Maybe I didn't understand the commit message correctly, does this code
intend to get MAX alignment value in a chain? For example:
base(4096)->mid(512)->top(1024) results: 4096
The condition to traver the backing files seems complex.
> + return alignment;
> +}
> +
> /* check if the path starts with "<protocol>:" */
> static int path_has_protocol(const char *path)
> {
> @@ -4335,7 +4360,7 @@ void bdrv_set_buffer_alignment(BlockDriverState *bs, int align)
>
> void *qemu_blockalign(BlockDriverState *bs, size_t size)
> {
> - return qemu_memalign((bs && bs->buffer_alignment) ? bs->buffer_alignment : 512, size);
> + return qemu_memalign(bdrv_opt_mem_align(bs), size);
> }
>
> /*
> @@ -4344,12 +4369,13 @@ void *qemu_blockalign(BlockDriverState *bs, size_t size)
> bool bdrv_qiov_is_aligned(BlockDriverState *bs, QEMUIOVector *qiov)
> {
> int i;
> + size_t alignment = bdrv_opt_mem_align(bs);
>
> for (i = 0; i < qiov->niov; i++) {
> - if ((uintptr_t) qiov->iov[i].iov_base % bs->buffer_alignment) {
> + if ((uintptr_t) qiov->iov[i].iov_base % alignment) {
> return false;
> }
> - if (qiov->iov[i].iov_len % bs->buffer_alignment) {
> + if (qiov->iov[i].iov_len % alignment) {
> return false;
> }
> }
> diff --git a/include/block/block.h b/include/block/block.h
> index 3560deb..d262c0e 100644
> --- a/include/block/block.h
> +++ b/include/block/block.h
> @@ -383,6 +383,7 @@ void bdrv_img_create(const char *filename, const char *fmt,
> char *options, uint64_t img_size, int flags,
> Error **errp, bool quiet);
>
> +size_t bdrv_opt_mem_align(BlockDriverState *bs);
> void bdrv_set_buffer_alignment(BlockDriverState *bs, int align);
> void *qemu_blockalign(BlockDriverState *bs, size_t size);
> bool bdrv_qiov_is_aligned(BlockDriverState *bs, QEMUIOVector *qiov);
> diff --git a/include/block/block_int.h b/include/block/block_int.h
> index 1666066..6a84f83 100644
> --- a/include/block/block_int.h
> +++ b/include/block/block_int.h
> @@ -119,6 +119,10 @@ struct BlockDriver {
> int64_t sector_num, int nb_sectors,
> BlockDriverCompletionFunc *cb, void *opaque);
>
> + /* Returns the alignment in bytes that is required so that no bounce buffer
> + * is required throughout the stack */
> + int (*bdrv_opt_mem_align)(BlockDriverState *bs);
> +
> int coroutine_fn (*bdrv_co_readv)(BlockDriverState *bs,
> int64_t sector_num, int nb_sectors, QEMUIOVector *qiov);
> int coroutine_fn (*bdrv_co_writev)(BlockDriverState *bs,
>
next prev parent reply other threads:[~2013-12-10 3:18 UTC|newest]
Thread overview: 32+ messages / expand[flat|nested] mbox.gz Atom feed top
2013-12-06 17:22 [Qemu-devel] [RFC PATCH 00/19] block: Support for 512b-on-4k emulation Kevin Wolf
2013-12-06 17:22 ` [Qemu-devel] [RFC PATCH 01/19] qemu_memalign: Allow small alignments Kevin Wolf
2013-12-06 17:22 ` [Qemu-devel] [RFC PATCH 02/19] block: Detect unaligned length in bdrv_qiov_is_aligned() Kevin Wolf
2013-12-06 19:12 ` Eric Blake
2013-12-06 17:22 ` [Qemu-devel] [RFC PATCH 03/19] block: Don't use guest sector size for qemu_blockalign() Kevin Wolf
2013-12-10 3:18 ` Wenchao Xia [this message]
2013-12-10 9:42 ` Kevin Wolf
2013-12-11 2:43 ` Wenchao Xia
2013-12-06 17:22 ` [Qemu-devel] [RFC PATCH 04/19] block: rename buffer_alignment to guest_block_size Kevin Wolf
2013-12-10 3:25 ` Wenchao Xia
2013-12-06 17:22 ` [Qemu-devel] [RFC PATCH 05/19] raw: Probe required direct I/O alignment Kevin Wolf
2013-12-06 17:53 ` Paolo Bonzini
2013-12-09 12:58 ` Kevin Wolf
2013-12-09 13:40 ` Paolo Bonzini
2013-12-06 17:22 ` [Qemu-devel] [RFC PATCH 06/19] block: Introduce bdrv_aligned_preadv() Kevin Wolf
2013-12-06 17:22 ` [Qemu-devel] [RFC PATCH 07/19] block: Introduce bdrv_co_do_preadv() Kevin Wolf
2013-12-06 17:22 ` [Qemu-devel] [RFC PATCH 08/19] block: Introduce bdrv_aligned_pwritev() Kevin Wolf
2013-12-06 17:22 ` [Qemu-devel] [RFC PATCH 09/19] block: write: Handle COR dependency after I/O throttling Kevin Wolf
2013-12-06 17:22 ` [Qemu-devel] [RFC PATCH 10/19] block: Introduce bdrv_co_do_pwritev() Kevin Wolf
2013-12-06 17:22 ` [Qemu-devel] [RFC PATCH 11/19] block: Switch BdrvTrackedRequest to byte granularity Kevin Wolf
2013-12-06 17:22 ` [Qemu-devel] [RFC PATCH 12/19] block: Allow waiting for overlapping requests between begin/end Kevin Wolf
2013-12-06 17:22 ` [Qemu-devel] [RFC PATCH 13/19] block: Make zero-after-EOF work with larger alignment Kevin Wolf
2013-12-06 17:22 ` [Qemu-devel] [RFC PATCH 14/19] block: Generalise and optimise COR serialisation Kevin Wolf
2013-12-06 17:22 ` [Qemu-devel] [RFC PATCH 15/19] block: Make overlap range for serialisation dynamic Kevin Wolf
2013-12-06 17:22 ` [Qemu-devel] [RFC PATCH 16/19] block: Align requests in bdrv_co_do_pwritev() Kevin Wolf
2013-12-06 17:22 ` [Qemu-devel] [RFC PATCH 17/19] block: Change coroutine wrapper to byte granularity Kevin Wolf
2013-12-06 17:22 ` [Qemu-devel] [RFC PATCH 18/19] block: Make bdrv_pread() a bdrv_prwv_co() wrapper Kevin Wolf
2013-12-06 17:23 ` [Qemu-devel] [RFC PATCH 19/19] block: Make bdrv_pwrite() " Kevin Wolf
2013-12-06 17:55 ` [Qemu-devel] [RFC PATCH 00/19] block: Support for 512b-on-4k emulation Paolo Bonzini
2013-12-09 11:16 ` Kevin Wolf
2013-12-09 12:51 ` Stefan Hajnoczi
2013-12-09 13:02 ` Kevin Wolf
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=52A687F8.8000205@linux.vnet.ibm.com \
--to=xiawenc@linux.vnet.ibm.com \
--cc=armbru@redhat.com \
--cc=kwolf@redhat.com \
--cc=pbonzini@redhat.com \
--cc=qemu-devel@nongnu.org \
--cc=stefanha@redhat.com \
/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).