From: Tim Deegan <tim@xen.org>
To: Jan Beulich <JBeulich@suse.com>
Cc: xen-devel@lists.xen.org
Subject: Re: [PATCH] xen/xenoprof: avoid division by 0
Date: Thu, 14 Feb 2013 17:10:04 +0000 [thread overview]
Message-ID: <20130214171004.GS83752@ocelot.phlegethon.org> (raw)
In-Reply-To: <511D276A02000078000BE73D@nat28.tlf.novell.com>
At 17:05 +0000 on 14 Feb (1360861530), Jan Beulich wrote:
> >>> On 14.02.13 at 17:50, Tim Deegan <tim@xen.org> wrote:
> > At 16:34 +0000 on 14 Feb (1360859678), Jan Beulich wrote:
> >> >>> On 14.02.13 at 17:16, Tim Deegan <tim@xen.org> wrote:
> >> > --- a/xen/common/xenoprof.c Thu Feb 14 15:46:56 2013 +0000
> >> > +++ b/xen/common/xenoprof.c Thu Feb 14 16:16:17 2013 +0000
> >> > @@ -225,7 +225,7 @@ static int alloc_xenoprof_struct(
> >> > #endif
> >> >
> >> > /* reduce max_samples if necessary to limit pages allocated */
> >> > - max_bufsize = (MAX_OPROF_SHARED_PAGES * PAGE_SIZE) / nvcpu;
> >> > + max_bufsize = (MAX_OPROF_SHARED_PAGES * PAGE_SIZE) / (nvcpu ?: 1);
> >> > max_max_samples = ( (max_bufsize - bufsize) / i ) + 1;
> >> > if ( (unsigned)max_samples > max_max_samples )
> >> > max_samples = max_max_samples;
> >>
> >> I think the function would better return an error in that case. After
> >> all there's little point in setting up anything when we for sure don't
> >> know how many vCPU-s a domain is going to have.
> >
> > Grand so:
> >
> > # HG changeset patch
> > # Parent 5a84cc531072378e6e5ff89b4c0e9a35000dc56f
> > xen/xenoprof: avoid division by 0.
> >
> > Signed-off-by: Tim Deegan <tim@xen.org>
> >
> > diff -r 5a84cc531072 xen/common/xenoprof.c
> > --- a/xen/common/xenoprof.c Thu Feb 14 15:46:56 2013 +0000
> > +++ b/xen/common/xenoprof.c Thu Feb 14 16:48:49 2013 +0000
> > @@ -213,6 +213,9 @@ static int alloc_xenoprof_struct(
> > for_each_vcpu ( d, v )
> > nvcpu++;
> >
> > + if ( !nvcpu )
> > + return -EINVAL;
>
> Missing some cleanup here? Or move the preceding loop and this
> addition to the top of the function?
D'oh, yes. Trying again:
# HG changeset patch
# Parent 5a84cc531072378e6e5ff89b4c0e9a35000dc56f
xen/xenoprof: avoid division by 0.
Signed-off-by: Tim Deegan <tim@xen.org>
diff -r 5a84cc531072 xen/common/xenoprof.c
--- a/xen/common/xenoprof.c Thu Feb 14 15:46:56 2013 +0000
+++ b/xen/common/xenoprof.c Thu Feb 14 17:07:41 2013 +0000
@@ -193,6 +193,13 @@ static int alloc_xenoprof_struct(
unsigned max_max_samples;
int i;
+ nvcpu = 0;
+ for_each_vcpu ( d, v )
+ nvcpu++;
+
+ if ( !nvcpu )
+ return -EINVAL;
+
d->xenoprof = xzalloc(struct xenoprof);
if ( d->xenoprof == NULL )
{
@@ -209,10 +216,6 @@ static int alloc_xenoprof_struct(
return -ENOMEM;
}
- nvcpu = 0;
- for_each_vcpu ( d, v )
- nvcpu++;
-
bufsize = sizeof(struct xenoprof_buf);
i = sizeof(struct event_log);
#ifdef CONFIG_COMPAT
next prev parent reply other threads:[~2013-02-14 17:10 UTC|newest]
Thread overview: 7+ messages / expand[flat|nested] mbox.gz Atom feed top
2013-02-14 16:16 [PATCH] xen/xenoprof: avoid division by 0 Tim Deegan
2013-02-14 16:34 ` Jan Beulich
2013-02-14 16:50 ` Tim Deegan
2013-02-14 17:05 ` Jan Beulich
2013-02-14 17:10 ` Tim Deegan [this message]
2013-02-15 8:10 ` Jan Beulich
2013-02-15 8:24 ` Keir Fraser
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=20130214171004.GS83752@ocelot.phlegethon.org \
--to=tim@xen.org \
--cc=JBeulich@suse.com \
--cc=xen-devel@lists.xen.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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.