From: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com>
To: qemu-block@nongnu.org
Cc: kwolf@redhat.com, vsementsov@virtuozzo.com,
qemu-devel@nongnu.org, mreitz@redhat.com, den@openvz.org,
jsnow@redhat.com
Subject: [PATCH 1/6] block/block-copy: allocate buffer in block_copy_with_bounce_buffer
Date: Thu, 3 Oct 2019 20:15:34 +0300 [thread overview]
Message-ID: <20191003171539.12327-2-vsementsov@virtuozzo.com> (raw)
In-Reply-To: <20191003171539.12327-1-vsementsov@virtuozzo.com>
Move bounce_buffer allocation block_copy_with_bounce_buffer. This
commit simplifies further work on implementing copying by larger chunks
(of different size) and further asynchronous handling of block_copy
iterations (with help of block/aio_task API).
Allocation works fast, a lot faster than disk io, so it's not a problem
that we now allocate/free bounce_buffer more times. And we anyway will
have to allocate several bounce_buffers for parallel execution of loop
iterations in future.
Signed-off-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com>
---
block/block-copy.c | 21 ++++++++-------------
1 file changed, 8 insertions(+), 13 deletions(-)
diff --git a/block/block-copy.c b/block/block-copy.c
index 5404bc921d..aca0f893d7 100644
--- a/block/block-copy.c
+++ b/block/block-copy.c
@@ -126,20 +126,17 @@ void block_copy_set_callbacks(
static int coroutine_fn block_copy_with_bounce_buffer(BlockCopyState *s,
int64_t start,
int64_t end,
- bool *error_is_read,
- void **bounce_buffer)
+ bool *error_is_read)
{
int ret;
int nbytes;
+ void *bounce_buffer = qemu_blockalign(s->source->bs, s->cluster_size);
assert(QEMU_IS_ALIGNED(start, s->cluster_size));
bdrv_reset_dirty_bitmap(s->copy_bitmap, start, s->cluster_size);
nbytes = MIN(s->cluster_size, s->len - start);
- if (!*bounce_buffer) {
- *bounce_buffer = qemu_blockalign(s->source->bs, s->cluster_size);
- }
- ret = bdrv_co_pread(s->source, start, nbytes, *bounce_buffer, 0);
+ ret = bdrv_co_pread(s->source, start, nbytes, bounce_buffer, 0);
if (ret < 0) {
trace_block_copy_with_bounce_buffer_read_fail(s, start, ret);
if (error_is_read) {
@@ -148,7 +145,7 @@ static int coroutine_fn block_copy_with_bounce_buffer(BlockCopyState *s,
goto fail;
}
- ret = bdrv_co_pwrite(s->target, start, nbytes, *bounce_buffer,
+ ret = bdrv_co_pwrite(s->target, start, nbytes, bounce_buffer,
s->write_flags);
if (ret < 0) {
trace_block_copy_with_bounce_buffer_write_fail(s, start, ret);
@@ -158,8 +155,11 @@ static int coroutine_fn block_copy_with_bounce_buffer(BlockCopyState *s,
goto fail;
}
+ qemu_vfree(bounce_buffer);
+
return nbytes;
fail:
+ qemu_vfree(bounce_buffer);
bdrv_set_dirty_bitmap(s->copy_bitmap, start, s->cluster_size);
return ret;
@@ -271,7 +271,6 @@ int coroutine_fn block_copy(BlockCopyState *s,
{
int ret = 0;
int64_t end = bytes + start; /* bytes */
- void *bounce_buffer = NULL;
int64_t status_bytes;
BlockCopyInFlightReq req;
@@ -324,7 +323,7 @@ int coroutine_fn block_copy(BlockCopyState *s,
}
if (!s->use_copy_range) {
ret = block_copy_with_bounce_buffer(s, start, dirty_end,
- error_is_read, &bounce_buffer);
+ error_is_read);
}
if (ret < 0) {
break;
@@ -335,10 +334,6 @@ int coroutine_fn block_copy(BlockCopyState *s,
ret = 0;
}
- if (bounce_buffer) {
- qemu_vfree(bounce_buffer);
- }
-
block_copy_inflight_req_end(&req);
return ret;
--
2.21.0
next prev parent reply other threads:[~2019-10-03 17:25 UTC|newest]
Thread overview: 23+ messages / expand[flat|nested] mbox.gz Atom feed top
2019-10-03 17:15 [PATCH 0/6] block-copy: memory limit Vladimir Sementsov-Ogievskiy
2019-10-03 17:15 ` Vladimir Sementsov-Ogievskiy [this message]
2019-10-07 13:30 ` [PATCH 1/6] block/block-copy: allocate buffer in block_copy_with_bounce_buffer Max Reitz
2019-10-03 17:15 ` [PATCH 2/6] block/block-copy: limit copy_range_size to 16 MiB Vladimir Sementsov-Ogievskiy
2019-10-07 13:40 ` Max Reitz
2019-10-03 17:15 ` [PATCH 3/6] block/block-copy: refactor copying Vladimir Sementsov-Ogievskiy
2019-10-07 14:17 ` Max Reitz
2019-10-07 16:29 ` Vladimir Sementsov-Ogievskiy
2019-10-03 17:15 ` [PATCH 4/6] util: introduce co-shared-amount Vladimir Sementsov-Ogievskiy
2019-10-07 15:22 ` Max Reitz
2019-10-07 16:49 ` Vladimir Sementsov-Ogievskiy
2019-10-03 17:15 ` [PATCH 5/6] block/block-copy: add memory limit Vladimir Sementsov-Ogievskiy
2019-10-07 15:27 ` Max Reitz
2019-10-07 17:10 ` Vladimir Sementsov-Ogievskiy
2019-10-08 9:03 ` Max Reitz
2019-10-08 9:15 ` Vladimir Sementsov-Ogievskiy
2019-10-08 9:20 ` Vladimir Sementsov-Ogievskiy
2019-10-08 9:57 ` Max Reitz
2019-10-08 9:56 ` Max Reitz
2019-10-03 17:15 ` [PATCH 6/6] block/block-copy: increase buffered copy request Vladimir Sementsov-Ogievskiy
2019-10-07 15:47 ` Max Reitz
2019-10-07 15:48 ` Max Reitz
2019-10-07 16:22 ` Vladimir Sementsov-Ogievskiy
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=20191003171539.12327-2-vsementsov@virtuozzo.com \
--to=vsementsov@virtuozzo.com \
--cc=den@openvz.org \
--cc=jsnow@redhat.com \
--cc=kwolf@redhat.com \
--cc=mreitz@redhat.com \
--cc=qemu-block@nongnu.org \
--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).