Re: [REGRESSION] Massive virtio-net throughput drop in guest VM with Linux 6.8+
From: Markus Fohrer
Date: Fri Apr 04 2025 - 04:52:57 EST
Am Freitag, dem 04.04.2025 um 04:29 -0400 schrieb Michael S. Tsirkin:
> On Fri, Apr 04, 2025 at 10:16:55AM +0200, Markus Fohrer wrote:
> > Am Donnerstag, dem 03.04.2025 um 09:04 -0400 schrieb Michael S.
> > Tsirkin:
> > > On Wed, Apr 02, 2025 at 11:12:07PM +0200, Markus Fohrer wrote:
> > > > Hi,
> > > >
> > > > I'm observing a significant performance regression in KVM guest
> > > > VMs
> > > > using virtio-net with recent Linux kernels (6.8.1+ and 6.14).
> > > >
> > > > When running on a host system equipped with a Broadcom
> > > > NetXtreme-E
> > > > (bnxt_en) NIC and AMD EPYC CPUs, the network throughput in the
> > > > guest drops to 100–200 KB/s. The same guest configuration
> > > > performs
> > > > normally (~100 MB/s) when using kernel 6.8.0 or when the VM is
> > > > moved to a host with Intel NICs.
> > > >
> > > > Test environment:
> > > > - Host: QEMU/KVM, Linux 6.8.1 and 6.14.0
> > > > - Guest: Linux with virtio-net interface
> > > > - NIC: Broadcom BCM57416 (bnxt_en driver, no issues at host
> > > > level)
> > > > - CPU: AMD EPYC
> > > > - Storage: virtio-scsi
> > > > - VM network: virtio-net, virtio-scsi (no CPU or IO
> > > > bottlenecks)
> > > > - Traffic test: iperf3, scp, wget consistently slow in guest
> > > >
> > > > This issue is not present:
> > > > - On 6.8.0
> > > > - On hosts with Intel NICs (same VM config)
> > > >
> > > > I have bisected the issue to the following upstream commit:
> > > >
> > > > 49d14b54a527 ("virtio-net: Suppress tx timeout warning for
> > > > small
> > > > tx")
> > > > https://git.kernel.org/linus/49d14b54a527
> > >
> > > Thanks a lot for the info!
> > >
> > >
> > > both the link and commit point at:
> > >
> > > commit 49d14b54a527289d09a9480f214b8c586322310a
> > > Author: Eric Dumazet <edumazet@xxxxxxxxxx>
> > > Date: Thu Sep 26 16:58:36 2024 +0000
> > >
> > > net: test for not too small csum_start in
> > > virtio_net_hdr_to_skb()
> > >
> > >
> > > is this what you mean?
> > >
> > > I don't know which commit is "virtio-net: Suppress tx timeout
> > > warning
> > > for small tx"
> > >
> > >
> > >
> > > > Reverting this commit restores normal network performance in
> > > > affected guest VMs.
> > > >
> > > > I’m happy to provide more data or assist with testing a
> > > > potential
> > > > fix.
> > > >
> > > > Thanks,
> > > > Markus Fohrer
> > >
> > >
> > > Thanks! First I think it's worth checking what is the setup, e.g.
> > > which offloads are enabled.
> > > Besides that, I'd start by seeing what's doing on. Assuming I'm
> > > right
> > > about
> > > Eric's patch:
> > >
> > > diff --git a/include/linux/virtio_net.h
> > > b/include/linux/virtio_net.h
> > > index 276ca543ef44d8..02a9f4dc594d02 100644
> > > --- a/include/linux/virtio_net.h
> > > +++ b/include/linux/virtio_net.h
> > > @@ -103,8 +103,10 @@ static inline int
> > > virtio_net_hdr_to_skb(struct
> > > sk_buff *skb,
> > >
> > > if (!skb_partial_csum_set(skb, start, off))
> > > return -EINVAL;
> > > + if (skb_transport_offset(skb) < nh_min_len)
> > > + return -EINVAL;
> > >
> > > - nh_min_len = max_t(u32, nh_min_len,
> > > skb_transport_offset(skb));
> > > + nh_min_len = skb_transport_offset(skb);
> > > p_off = nh_min_len + thlen;
> > > if (!pskb_may_pull(skb, p_off))
> > > return -EINVAL;
> > >
> > >
> > > sticking a printk before return -EINVAL to show the offset and
> > > nh_min_len
> > > would be a good 1st step. Thanks!
> > >
> >
> > I added the following printk inside virtio_net_hdr_to_skb():
> >
> > if (skb_transport_offset(skb) < nh_min_len){
> > printk(KERN_INFO "virtio_net: 3 drop, transport_offset=%u,
> > nh_min_len=%u\n",
> > skb_transport_offset(skb), nh_min_len);
> > return -EINVAL;
> > }
> >
> > Built and installed the kernel, then triggered a large download
> > via:
> >
> > wget http://speedtest.belwue.net/10G
> >
> > Relevant output from `dmesg -w`:
> >
> > [ 57.327943] virtio_net: 3 drop, transport_offset=34,
> > nh_min_len=40
> > [ 57.428942] virtio_net: 3 drop, transport_offset=34,
> > nh_min_len=40
> > [ 57.428962] virtio_net: 3 drop, transport_offset=34,
> > nh_min_len=40
> > [ 57.553068] virtio_net: 3 drop, transport_offset=34,
> > nh_min_len=40
> > [ 57.553088] virtio_net: 3 drop, transport_offset=34,
> > nh_min_len=40
> > [ 57.576678] virtio_net: 3 drop, transport_offset=34,
> > nh_min_len=40
> > [ 57.618438] virtio_net: 3 drop, transport_offset=34,
> > nh_min_len=40
> > [ 57.618453] virtio_net: 3 drop, transport_offset=34,
> > nh_min_len=40
> > [ 57.703077] virtio_net: 3 drop, transport_offset=34,
> > nh_min_len=40
> > [ 57.823072] virtio_net: 3 drop, transport_offset=34,
> > nh_min_len=40
> > [ 57.891982] virtio_net: 3 drop, transport_offset=34,
> > nh_min_len=40
> > [ 57.946190] virtio_net: 3 drop, transport_offset=34,
> > nh_min_len=40
> > [ 58.218686] virtio_net: 3 drop, transport_offset=34,
> > nh_min_len=40
>
> Hmm indeed. And what about these values?
> u32 start = __virtio16_to_cpu(little_endian, hdr-
> >csum_start);
> u32 off = __virtio16_to_cpu(little_endian, hdr-
> >csum_offset);
> u32 needed = start + max_t(u32, thlen, off +
> sizeof(__sum16));
> print them too?
>
>
>
> > I would now do the test with commit
> > 49d14b54a527289d09a9480f214b8c586322310a and commit
> > 49d14b54a527289d09a9480f214b8c586322310a~1
> >
>
> Worth checking though it seems likely now the hypervisor is doing
> weird
> things. what kind of backend is it? qemu? tun? vhost-user? vhost-net?
>
Backend: QEMU/KVM hypervisor (Proxmox)
printk output:
[ 58.641906] virtio_net: drop, transport_offset=34 start=34, off=16,
needed=54, nh_min_len=40
[ 58.678048] virtio_net: drop, transport_offset=34 start=34, off=16,
needed=54, nh_min_len=40
[ 58.952871] virtio_net: drop, transport_offset=34 start=34, off=16,
needed=54, nh_min_len=40
[ 58.962157] virtio_net: drop, transport_offset=34 start=34, off=16,
needed=54, nh_min_len=40
[ 59.071645] virtio_net: drop, transport_offset=34 start=34, off=16,
needed=54, nh_min_len=40