All of lore.kernel.org
 help / color / mirror / Atom feed
From: Stefano Stabellini <stefano.stabellini@eu.citrix.com>
To: David Vrabel <david.vrabel@citrix.com>
Cc: xen-devel@lists.xensource.com, liuling-it@360.cn,
	qemu-devel@nongnu.org,
	Stefano Stabellini <stefano.stabellini@eu.citrix.com>
Subject: Re: [Qemu-devel] [Xen-devel] [PATCH] xenfb.c: avoid expensive loops when prod <= out_cons
Date: Wed, 6 Jan 2016 14:14:18 +0000	[thread overview]
Message-ID: <alpine.DEB.2.02.1601061412000.31111@kaball.uk.xensource.com> (raw)
In-Reply-To: <568D071A.7050009@citrix.com>

On Wed, 6 Jan 2016, David Vrabel wrote:
> On 06/01/16 12:08, Stefano Stabellini wrote:
> > If the frontend sets out_cons to a value higher than out_prod, it will
> > cause xenfb_handle_events to loop about 2^32 times. Avoid that by using
> > better checks at the beginning of the function.
> 
> You can't use less than to compare prod and cons because they wrap.
> 
> You need to compare (prod - cons) against ring size (or similar) to
> check for overflow.  See RING_REQUEST_PROD_OVERFLOW() etc.

Yes, you are right. I think that the right fix should be:


diff --git a/hw/display/xenfb.c b/hw/display/xenfb.c
index 4e2a27a..594baff 100644
--- a/hw/display/xenfb.c
+++ b/hw/display/xenfb.c
@@ -789,8 +789,9 @@ static void xenfb_handle_events(struct XenFB *xenfb)
 
     prod = page->out_prod;
     out_cons = page->out_cons;
-    if (prod == out_cons)
-	return;
+    if (prod - out_cons >= XENFB_OUT_RING_LEL) {
+        return;
+    }
     xen_rmb();		/* ensure we see ring contents up to prod */
     for (cons = out_cons; cons != prod; cons++) {
 	union xenfb_out_event *event = &XENFB_OUT_RING_REF(page, cons);

WARNING: multiple messages have this Message-ID (diff)
From: Stefano Stabellini <stefano.stabellini@eu.citrix.com>
To: David Vrabel <david.vrabel@citrix.com>
Cc: xen-devel@lists.xensource.com, liuling-it@360.cn,
	qemu-devel@nongnu.org,
	Stefano Stabellini <stefano.stabellini@eu.citrix.com>
Subject: Re: [Xen-devel] [PATCH] xenfb.c: avoid expensive loops when prod <= out_cons
Date: Wed, 6 Jan 2016 14:14:18 +0000	[thread overview]
Message-ID: <alpine.DEB.2.02.1601061412000.31111@kaball.uk.xensource.com> (raw)
In-Reply-To: <568D071A.7050009@citrix.com>

On Wed, 6 Jan 2016, David Vrabel wrote:
> On 06/01/16 12:08, Stefano Stabellini wrote:
> > If the frontend sets out_cons to a value higher than out_prod, it will
> > cause xenfb_handle_events to loop about 2^32 times. Avoid that by using
> > better checks at the beginning of the function.
> 
> You can't use less than to compare prod and cons because they wrap.
> 
> You need to compare (prod - cons) against ring size (or similar) to
> check for overflow.  See RING_REQUEST_PROD_OVERFLOW() etc.

Yes, you are right. I think that the right fix should be:


diff --git a/hw/display/xenfb.c b/hw/display/xenfb.c
index 4e2a27a..594baff 100644
--- a/hw/display/xenfb.c
+++ b/hw/display/xenfb.c
@@ -789,8 +789,9 @@ static void xenfb_handle_events(struct XenFB *xenfb)
 
     prod = page->out_prod;
     out_cons = page->out_cons;
-    if (prod == out_cons)
-	return;
+    if (prod - out_cons >= XENFB_OUT_RING_LEL) {
+        return;
+    }
     xen_rmb();		/* ensure we see ring contents up to prod */
     for (cons = out_cons; cons != prod; cons++) {
 	union xenfb_out_event *event = &XENFB_OUT_RING_REF(page, cons);

  reply	other threads:[~2016-01-06 14:14 UTC|newest]

Thread overview: 8+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2016-01-06 12:08 [Qemu-devel] [PATCH] xenfb.c: avoid expensive loops when prod <= out_cons Stefano Stabellini
2016-01-06 12:08 ` Stefano Stabellini
2016-01-06 12:22 ` [Qemu-devel] [Xen-devel] " David Vrabel
2016-01-06 12:22   ` David Vrabel
2016-01-06 14:14   ` Stefano Stabellini [this message]
2016-01-06 14:14     ` Stefano Stabellini
2016-01-06 12:27 ` [Qemu-devel] " Paul Durrant
2016-01-06 12:27   ` Paul Durrant

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=alpine.DEB.2.02.1601061412000.31111@kaball.uk.xensource.com \
    --to=stefano.stabellini@eu.citrix.com \
    --cc=david.vrabel@citrix.com \
    --cc=liuling-it@360.cn \
    --cc=qemu-devel@nongnu.org \
    --cc=xen-devel@lists.xensource.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 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.