qemu-devel.nongnu.org archive mirror
 help / color / mirror / Atom feed
From: Kevin Wolf <kwolf@redhat.com>
To: Peter Lieven <pl@kamp.de>
Cc: famz@redhat.com, benoit@irqsave.net, ming.lei@canonical.com,
	armbru@redhat.com, qemu-devel@nongnu.org, stefanha@redhat.com,
	pbonzini@redhat.com, Max Reitz <mreitz@redhat.com>
Subject: Re: [Qemu-devel] [PATCH v3 4/4] virtio-blk: introduce multiread
Date: Fri, 30 Jan 2015 22:15:01 +0100	[thread overview]
Message-ID: <20150130211501.GG24537@noname.redhat.com> (raw)
In-Reply-To: <54CBF21C.6040102@kamp.de>

Am 30.01.2015 um 22:05 hat Peter Lieven geschrieben:
> Am 30.01.2015 um 18:16 schrieb Max Reitz:
> > On 2015-01-30 at 09:33, Peter Lieven wrote:
> >> this patch finally introduces multiread support to virtio-blk. While
> >> multiwrite support was there for a long time, read support was missing.
> >>
> >> The complete merge logic is moved into virtio-blk.c which has
> >> been the only user of request merging ever since. This is required
> >> to be able to merge chunks of requests and immediately invoke callbacks
> >> for those requests. Secondly, this is required to switch to
> >> direct invocation of coroutines which is planned at a later stage.
> >>
> >> The following benchmarks show the performance of running fio with
> >> 4 worker threads on a local ram disk. The numbers show the average
> >> of 10 test runs after 1 run as warmup phase.
> >>
> >>                |        4k        |       64k        |        4k
> >> MB/s          | rd seq | rd rand | rd seq | rd rand | wr seq | wr rand
> >> --------------+--------+---------+--------+---------+--------+--------
> >> master        | 1221   | 1187    | 4178   | 4114    | 1745   | 1213
> >> multiread     | 1829   | 1189    | 4639   | 4110    | 1894   | 1216
> >>
> >> Signed-off-by: Peter Lieven <pl@kamp.de>
> >> ---
> >>   hw/block/dataplane/virtio-blk.c |   8 +-
> >>   hw/block/virtio-blk.c           | 288 +++++++++++++++++++++++++++-------------
> >>   include/hw/virtio/virtio-blk.h  |  19 +--
> >>   trace-events                    |   1 +
> >>   4 files changed, 210 insertions(+), 106 deletions(-)

> >> +    int64_t sector_num = 0;
> >> +
> >> +    if (mrb->num_reqs == 1) {
> >> +        virtio_submit_multireq2(blk, mrb, 0, 1, -1);
> >> +        mrb->num_reqs = 0;
> >>           return;
> >>       }
> >>   -    ret = blk_aio_multiwrite(blk, mrb->blkreq, mrb->num_writes);
> >> -    if (ret != 0) {
> >> -        for (i = 0; i < mrb->num_writes; i++) {
> >> -            if (mrb->blkreq[i].error) {
> >> -                virtio_blk_rw_complete(mrb->blkreq[i].opaque, -EIO);
> >> +    max_xfer_len = blk_get_max_transfer_length(mrb->reqs[0]->dev->blk);
> >> +    max_xfer_len = MIN_NON_ZERO(max_xfer_len, INT_MAX);
> >> +
> >> +    qsort(mrb->reqs, mrb->num_reqs, sizeof(*mrb->reqs),
> >> +          &virtio_multireq_compare);
> >> +
> >> +    for (i = 0; i < mrb->num_reqs; i++) {
> >> +        VirtIOBlockReq *req = mrb->reqs[i];
> >> +        if (num_reqs > 0) {
> >> +            bool merge = true;
> >> +
> >> +            /* merge would exceed maximum number of IOVs */
> >> +            if (niov + req->qiov.niov + 1 > IOV_MAX) {
> >
> > Hm, why the +1?
> 
> A really good question. I copied this piece from the old merge routine. It seems
> definetely wrong.

The old code merged requests even if they were overlapping. This could
result in one area being split in two.

I think you don't support this here, so removing the + 1 is probably
okay.

Kevin

  reply	other threads:[~2015-01-30 21:15 UTC|newest]

Thread overview: 14+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2015-01-30 14:32 [Qemu-devel] [PATCH v3 0/4] virtio-blk: add multiread support Peter Lieven
2015-01-30 14:32 ` [Qemu-devel] [PATCH v3 1/4] block: add accounting for merged requests Peter Lieven
2015-01-30 15:53   ` Max Reitz
2015-01-30 14:33 ` [Qemu-devel] [PATCH v3 2/4] hw/virtio-blk: add a constant for max number of " Peter Lieven
2015-01-30 16:01   ` Max Reitz
2015-01-30 14:33 ` [Qemu-devel] [PATCH v3 3/4] block-backend: expose bs->bl.max_transfer_length Peter Lieven
2015-01-30 16:04   ` Max Reitz
2015-01-30 14:33 ` [Qemu-devel] [PATCH v3 4/4] virtio-blk: introduce multiread Peter Lieven
2015-01-30 17:16   ` Max Reitz
2015-01-30 21:05     ` Peter Lieven
2015-01-30 21:15       ` Kevin Wolf [this message]
2015-01-31 19:46         ` Peter Lieven
2015-01-30 21:30       ` Max Reitz
2015-01-31 19:45         ` Peter Lieven

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=20150130211501.GG24537@noname.redhat.com \
    --to=kwolf@redhat.com \
    --cc=armbru@redhat.com \
    --cc=benoit@irqsave.net \
    --cc=famz@redhat.com \
    --cc=ming.lei@canonical.com \
    --cc=mreitz@redhat.com \
    --cc=pbonzini@redhat.com \
    --cc=pl@kamp.de \
    --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).