qemu-devel.nongnu.org archive mirror
 help / color / mirror / Atom feed
From: Dor Laor <dlaor@redhat.com>
To: Anthony Liguori <aliguori@us.ibm.com>
Cc: Uri Lublin <uril@redhat.com>, qemu-devel@nongnu.org
Subject: Re: [Qemu-devel] [PATCH] migrate.c: migrate_fd_put_buffer: Do not busyloop: stop writing if EWOULDBLOCK
Date: Sat, 23 May 2009 18:55:03 +0300	[thread overview]
Message-ID: <4A181C57.50108@redhat.com> (raw)
In-Reply-To: <4A16A696.8070902@us.ibm.com>

[-- Attachment #1: Type: text/plain, Size: 123 bytes --]

Anthony Liguori wrote:
>
> This is whitespace damaged, presumably from copy/paste, can you resend?
>
Here is a better one.

[-- Attachment #2: 0001-The-migration-code-is-non-blocking-designed-for-liv.patch --]
[-- Type: text/plain, Size: 1913 bytes --]

>From 2d875410cb168fa728dcde96885a3e4ddfd0bb58 Mon Sep 17 00:00:00 2001
From: Dor Laor <dor@redhat.com>
Date: Thu, 21 May 2009 12:29:39 +0300
Subject: [PATCH] The migration code is non-blocking, designed for live migration.

Practically migrate_fd_put_buffer busy-loops trying to write, as
on many machines EWOULDBLOCK==EAGAIN (look in include/asm-generic/errno.h).

Based on Uri Lublin's patch.
Signed-off-by: Dor Laor <dor@redhat.com>
---
 qemu/buffered_file.c |    4 ++--
 qemu/migration.c     |    2 +-
 2 files changed, 3 insertions(+), 3 deletions(-)

diff --git a/qemu/buffered_file.c b/qemu/buffered_file.c
index be5baea..34cde6e 100644
--- a/qemu/buffered_file.c
+++ b/qemu/buffered_file.c
@@ -86,7 +86,7 @@ static void buffered_flush(QEMUFileBuffered *s)
 
         ret = s->put_buffer(s->opaque, s->buffer + offset,
                             s->buffer_size - offset);
-        if (ret == -EAGAIN) {
+        if (ret == -EAGAIN || ret == -EWOULDBLOCK) {
             dprintf("backend not ready, freezing\n");
             s->freeze_output = 1;
             break;
@@ -132,7 +132,7 @@ static int buffered_put_buffer(void *opaque, const uint8_t *buf, int64_t pos, in
         }
 
         ret = s->put_buffer(s->opaque, buf + offset, size - offset);
-        if (ret == -EAGAIN) {
+        if (ret == -EAGAIN || ret ==-EWOULDBLOCK) {
             dprintf("backend not ready, freezing\n");
             s->freeze_output = 1;
             break;
diff --git a/qemu/migration.c b/qemu/migration.c
index 85757c5..8846eee 100644
--- a/qemu/migration.c
+++ b/qemu/migration.c
@@ -174,7 +174,7 @@ ssize_t migrate_fd_put_buffer(void *opaque, const void *data, size_t size)
     if (ret == -1)
         ret = -(s->get_error(s));
 
-    if (ret == -EAGAIN)
+    if (ret == -EAGAIN || ret == -EWOULDBLOCK)
         qemu_set_fd_handler2(s->fd, NULL, NULL, migrate_fd_put_notify, s);
 
     return ret;
-- 
1.5.6.6


      reply	other threads:[~2009-05-23 15:54 UTC|newest]

Thread overview: 4+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2009-05-19 11:08 [Qemu-devel] [PATCH] migrate.c: migrate_fd_put_buffer: Do not busyloop: stop writing if EWOULDBLOCK Uri Lublin
2009-05-21  9:32 ` Dor Laor
2009-05-22 13:20   ` Anthony Liguori
2009-05-23 15:55     ` Dor Laor [this message]

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=4A181C57.50108@redhat.com \
    --to=dlaor@redhat.com \
    --cc=aliguori@us.ibm.com \
    --cc=qemu-devel@nongnu.org \
    --cc=uril@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).