Re: [KVM PATCH v9 0/5] irqfd fixes and enhancements

From: Gregory Haskins <ghaskins@novell.com>
To: Avi Kivity <avi@redhat.com>
Cc: kvm@vger.kernel.org, linux-kernel@vger.kernel.org,
	mst@redhat.com, davidel@xmailserver.org
Subject: Re: [KVM PATCH v9 0/5] irqfd fixes and enhancements
Date: Mon, 06 Jul 2009 10:56:02 -0400	[thread overview]
Message-ID: <4A521082.40209@novell.com> (raw)
In-Reply-To: <4A50723E.6030305@redhat.com>

[-- Attachment #1: Type: text/plain, Size: 2114 bytes --]

Avi Kivity wrote:
> On 07/02/2009 06:50 PM, Avi Kivity wrote:
>> On 07/02/2009 06:37 PM, Gregory Haskins wrote:
>>> (Applies to kvm.git/master:1f9050fd)
>>>
>>> The following is the latest attempt to fix the races in
>>> irqfd/eventfd, as
>>> well as restore DEASSIGN support.  For more details, please read the
>>> patch
>>> headers.
>>>
>>> As always, this series has been tested against the kvm-eventfd unit
>>> test
>>> and everything appears to be functioning properly. You can download
>>> this
>>> test here:
>>
>> Applied, thanks.
>>
>
> ... and unapplied.  There's a refcounting mismatch in irqfd_cleanup: a
> reference is taken for each irqfd, but dropped for each guest.  This
> causes an oops if a guest with no irqfds is created and destroyed:

I was able to reproduce this issue.  The problem turned out to be that I
inadvertently always did a flush_workqueue(), even if the work-queue was
never initialized.   

The following interdiff applied to the reverted patch has been confirmed
to fix the issue:

-------------------

diff --git a/virt/kvm/eventfd.c b/virt/kvm/eventfd.c
index fcc3469..52b0e04 100644
--- a/virt/kvm/eventfd.c
+++ b/virt/kvm/eventfd.c
@@ -318,6 +318,9 @@ kvm_irqfd_deassign(struct kvm *kvm, int fd, int gsi)
        struct _irqfd *irqfd, *tmp;
        struct eventfd_ctx *eventfd;
 
+       if (!kvm->irqfds.init)
+               return -ENOENT;
+
        eventfd = eventfd_ctx_fdget(fd);
        if (IS_ERR(eventfd))
                return PTR_ERR(eventfd);
@@ -360,6 +363,9 @@ kvm_irqfd_release(struct kvm *kvm)
 {
        struct _irqfd *irqfd, *tmp;
 
+       if (!kvm->irqfds.init)
+               return;
+
        spin_lock_irq(&kvm->irqfds.lock);
 
        list_for_each_entry_safe(irqfd, tmp, &kvm->irqfds.items, list)

---------------------

You can pick up this fix folded into the original v9:5/5 patch here:

git pull
git://git.kernel.org/pub/scm/linux/kernel/git/ghaskins/linux-2.6-hacks.git
for-avi

Sorry for the sloppy patch in v9. :(  Will strive to do better next time.

Regards,
-Greg


[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 266 bytes --]
