From: Stefan Hajnoczi <stefanha@gmail.com>
To: Kevin Wolf <kwolf@redhat.com>
Cc: qemu-devel@nongnu.org, famz@redhat.com, den@virtuozzo.com,
qemu-block@nongnu.org, Pavel Butsykin <pbutsykin@virtuozzo.com>,
mreitz@redhat.com, stefanha@redhat.com,
Anton Nefedov <anton.nefedov@virtuozzo.com>
Subject: Re: [Qemu-devel] [PATCH v2 1/4] qcow2: multiple clusters write compressed
Date: Mon, 20 Nov 2017 14:53:11 +0000 [thread overview]
Message-ID: <20171120145311.GG4516@stefanha-x1.localdomain> (raw)
In-Reply-To: <1510851298-59922-2-git-send-email-anton.nefedov@virtuozzo.com>
[-- Attachment #1: Type: text/plain, Size: 4978 bytes --]
On Thu, Nov 16, 2017 at 07:54:55PM +0300, Anton Nefedov wrote:
> From: Pavel Butsykin <pbutsykin@virtuozzo.com>
>
> At the moment, qcow2_co_pwritev_compressed can process the requests size
> less than or equal to one cluster. This patch added possibility to write
> compressed data in the QCOW2 more than one cluster. The implementation
> is simple, we just split large requests into separate clusters and write
> using existing functionality.
>
> Signed-off-by: Anton Nefedov <anton.nefedov@virtuozzo.com>
> ---
> block/qcow2.c | 73 ++++++++++++++++++++++++++++++++++++++++++-----------------
> 1 file changed, 52 insertions(+), 21 deletions(-)
Kevin: qcow2_alloc_compressed_cluster_offset() sets up an L2 table entry
before doing the compressed data write. Can the following scenario
occur if there is another request racing with the compressed write?
1. Compressed cluster L2 table entry added to qcow2 in-memory cache
2. Another request forces cached L2 table to be written to disk
3. Power failure or crash before compressed data is written
Now there is an L2 table entry pointing to garbage data. This violates
qcow2's data integrity model.
I'm not sure if compressed writes are safe... It may have been okay for
qemu-img convert but the risk is increased when running a VM.
> diff --git a/block/qcow2.c b/block/qcow2.c
> index c276b24..f1e2759 100644
> --- a/block/qcow2.c
> +++ b/block/qcow2.c
> @@ -3332,11 +3332,9 @@ static int qcow2_truncate(BlockDriverState *bs, int64_t offset,
> return 0;
> }
>
> -/* XXX: put compressed sectors first, then all the cluster aligned
> - tables to avoid losing bytes in alignment */
> static coroutine_fn int
> -qcow2_co_pwritev_compressed(BlockDriverState *bs, uint64_t offset,
> - uint64_t bytes, QEMUIOVector *qiov)
> +qcow2_co_pwritev_cluster_compressed(BlockDriverState *bs, uint64_t offset,
> + uint64_t bytes, QEMUIOVector *qiov)
> {
> BDRVQcow2State *s = bs->opaque;
> QEMUIOVector hd_qiov;
> @@ -3346,25 +3344,12 @@ qcow2_co_pwritev_compressed(BlockDriverState *bs, uint64_t offset,
> uint8_t *buf, *out_buf;
> int64_t cluster_offset;
>
> - if (bytes == 0) {
> - /* align end of file to a sector boundary to ease reading with
> - sector based I/Os */
> - cluster_offset = bdrv_getlength(bs->file->bs);
> - if (cluster_offset < 0) {
> - return cluster_offset;
> - }
> - return bdrv_truncate(bs->file, cluster_offset, PREALLOC_MODE_OFF, NULL);
> - }
> -
> - if (offset_into_cluster(s, offset)) {
> - return -EINVAL;
> - }
> + assert(bytes <= s->cluster_size);
> + assert(!offset_into_cluster(s, offset));
>
> buf = qemu_blockalign(bs, s->cluster_size);
> - if (bytes != s->cluster_size) {
> - if (bytes > s->cluster_size ||
> - offset + bytes != bs->total_sectors << BDRV_SECTOR_BITS)
> - {
> + if (bytes < s->cluster_size) {
> + if (offset + bytes != bs->total_sectors << BDRV_SECTOR_BITS) {
> qemu_vfree(buf);
> return -EINVAL;
> }
> @@ -3444,6 +3429,52 @@ fail:
> return ret;
> }
>
> +/* XXX: put compressed sectors first, then all the cluster aligned
> + tables to avoid losing bytes in alignment */
> +static coroutine_fn int
> +qcow2_co_pwritev_compressed(BlockDriverState *bs, uint64_t offset,
> + uint64_t bytes, QEMUIOVector *qiov)
> +{
> + BDRVQcow2State *s = bs->opaque;
> + QEMUIOVector hd_qiov;
> + uint64_t curr_off = 0;
> + int ret;
> +
> + if (bytes == 0) {
> + /* align end of file to a sector boundary to ease reading with
> + sector based I/Os */
> + int64_t cluster_offset = bdrv_getlength(bs->file->bs);
> + if (cluster_offset < 0) {
> + return cluster_offset;
> + }
> + return bdrv_truncate(bs->file, cluster_offset, PREALLOC_MODE_OFF, NULL);
> + }
> +
> + if (offset_into_cluster(s, offset)) {
> + return -EINVAL;
> + }
> +
> + qemu_iovec_init(&hd_qiov, qiov->niov);
> + do {
> + uint32_t chunk_size;
> +
> + qemu_iovec_reset(&hd_qiov);
> + chunk_size = MIN(bytes, s->cluster_size);
> + qemu_iovec_concat(&hd_qiov, qiov, curr_off, chunk_size);
> +
> + ret = qcow2_co_pwritev_cluster_compressed(bs, offset + curr_off,
> + chunk_size, &hd_qiov);
> + if (ret < 0) {
> + break;
> + }
> + curr_off += chunk_size;
> + bytes -= chunk_size;
> + } while (bytes);
> + qemu_iovec_destroy(&hd_qiov);
> +
> + return ret;
> +}
> +
> static int make_completely_empty(BlockDriverState *bs)
> {
> BDRVQcow2State *s = bs->opaque;
> --
> 2.7.4
>
>
[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 455 bytes --]
next prev parent reply other threads:[~2017-11-20 14:53 UTC|newest]
Thread overview: 12+ messages / expand[flat|nested] mbox.gz Atom feed top
2017-11-16 16:54 [Qemu-devel] [PATCH v2 0/4] compressed block-stream Anton Nefedov
2017-11-16 16:54 ` [Qemu-devel] [PATCH v2 1/4] qcow2: multiple clusters write compressed Anton Nefedov
2017-11-20 14:53 ` Stefan Hajnoczi [this message]
2017-11-20 15:03 ` Denis V. Lunev
2017-11-21 11:07 ` Stefan Hajnoczi
2017-11-16 16:54 ` [Qemu-devel] [PATCH v2 2/4] block: support compressed write for copy-on-read Anton Nefedov
2017-11-20 14:53 ` [Qemu-devel] [Qemu-block] " Stefan Hajnoczi
2017-11-16 16:54 ` [Qemu-devel] [PATCH v2 3/4] block-stream: add compress option Anton Nefedov
2017-11-20 14:53 ` [Qemu-devel] [Qemu-block] " Stefan Hajnoczi
2017-11-16 16:54 ` [Qemu-devel] [PATCH v2 4/4] iotests: 030: add compressed block-stream test Anton Nefedov
2017-11-20 14:38 ` Stefan Hajnoczi
2017-11-20 15:39 ` Anton Nefedov
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=20171120145311.GG4516@stefanha-x1.localdomain \
--to=stefanha@gmail.com \
--cc=anton.nefedov@virtuozzo.com \
--cc=den@virtuozzo.com \
--cc=famz@redhat.com \
--cc=kwolf@redhat.com \
--cc=mreitz@redhat.com \
--cc=pbutsykin@virtuozzo.com \
--cc=qemu-block@nongnu.org \
--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).