Re: upstream kernel crashes

From: "Michael S. Tsirkin" <mst@redhat.com>
To: Xuan Zhuo <xuanzhuo@linux.alibaba.com>,
	Jason Wang <jasowang@redhat.com>,
	Andres Freund <andres@anarazel.de>,
	"David S. Miller" <davem@davemloft.net>,
	Eric Dumazet <edumazet@google.com>,
	Jakub Kicinski <kuba@kernel.org>, Paolo Abeni <pabeni@redhat.com>,
	virtualization@lists.linux-foundation.org,
	netdev@vger.kernel.org,
	Linus Torvalds <torvalds@linux-foundation.org>,
	Jens Axboe <axboe@kernel.dk>,
	James Bottomley <James.Bottomley@hansenpartnership.com>,
	"Martin K. Petersen" <martin.petersen@oracle.com>,
	Guenter Roeck <linux@roeck-us.net>,
	linux-kernel@vger.kernel.org,
	Greg KH <gregkh@linuxfoundation.org>,
	c@redhat.com
Subject: Re: upstream kernel crashes
Date: Mon, 15 Aug 2022 11:40:59 -0400	[thread overview]
Message-ID: <20220815113729-mutt-send-email-mst@kernel.org> (raw)
In-Reply-To: <FCDC5DDE-3CDD-4B8A-916F-CA7D87B547CE@anarazel.de>

On Mon, Aug 15, 2022 at 01:34:41AM -0700, Andres Freund wrote:
> Hi, 
> 
> On August 15, 2022 1:28:29 AM PDT, "Michael S. Tsirkin" <mst@redhat.com> wrote:
> >On Mon, Aug 15, 2022 at 01:15:27AM -0700, Andres Freund wrote:
> >> Hi,
> >> 
> >> On 2022-08-15 03:51:34 -0400, Michael S. Tsirkin wrote:
> >> > It is possible that GCP gets confused if ring size is smaller than the
> >> > device maximum simply because no one did it in the past.
> >> > 
> >> > So I pushed just the revert of 762faee5a267 to the test branch.
> >> > Could you give it a spin?
> >> 
> >> Seems to fix the issue, at least to the extent I can determine at 1am... :)
> >> 
> >> Greetings,
> >> 
> >> Andres Freund
> >
> >So you tested this:
> >
> >commit 13df5a7eaeb22561d39354b576bc98a7e2c389f9 (HEAD, kernel.org/test)
> >Author: Michael S. Tsirkin <mst@redhat.com>
> >Date:   Mon Aug 15 03:44:38 2022 -0400
> >
> >    Revert "virtio_net: set the default max ring size by find_vqs()"
> >    
> >    This reverts commit 762faee5a2678559d3dc09d95f8f2c54cd0466a7.
> >    
> >    Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
> >
> >and it fixes both issues right? No crashes no networking issue?
> 
> Correct. I only did limited testing, but it's survived far longer / more reboots than anything since the commit.
> 
> Andres
> -- 
> Sent from my Android device with K-9 Mail. Please excuse my brevity.


OK so this gives us a quick revert as a solution for now.
Next, I would appreciate it if you just try this simple hack.
If it crashes we either have a long standing problem in virtio
code or more likely a gcp bug where it can't handle smaller
rings than what device requestes.
Thanks!

diff --git a/drivers/virtio/virtio_pci_modern.c b/drivers/virtio/virtio_pci_modern.c
index f7965c5dd36b..bdd5f481570b 100644
--- a/drivers/virtio/virtio_pci_modern.c
+++ b/drivers/virtio/virtio_pci_modern.c
@@ -314,6 +314,9 @@ static struct virtqueue *setup_vq(struct virtio_pci_device *vp_dev,
 	if (!size || size > num)
 		size = num;
 
+	if (size > 1024)
+		size = 1024;
+
 	if (size & (size - 1)) {
 		dev_warn(&vp_dev->pci_dev->dev, "bad queue size %u", size);
 		return ERR_PTR(-EINVAL);


-- 
MST

