similar to: [PATCH v2] virtio_net: fix lockdep warning on 32 bit

When we fill up a receive VQ, try_fill_recv currently tries to count kicks using a 64 bit stats counter. Turns out, on a 32 bit kernel that uses a seqcount. sequence counts are "lock" constructs where you need to make sure that writers are serialized. In turn, this means that we mustn't run two try_fill_recv concurrently. Which of course we don't. We do run try_fill_recv

On Thu, Jan 16, 2014 at 01:38:46AM -0800, Michael Dalton wrote: > Add initial support for per-rx queue sysfs attributes to virtio-net. If > mergeable packet buffers are enabled, adds a read-only mergeable packet > buffer size sysfs attribute for each RX queue. > > Signed-off-by: Michael Dalton <mwdalton at google.com> > --- > drivers/net/virtio_net.c | 66

On Sat, 2020-10-24 at 10:22 +0800, Hillf Danton wrote: > > Looks like we can break the lock chain by moving ttm bo's release > method out of mmap_lock, see diff below. Ah, the perfect compliment to morning java, a patchlet to wedge in and see what happens. wedge/build/boot <schlurp... ahhh> Mmm, box says no banana... a lot. [ 30.456921] ================================ [

On Thu, Jan 16, 2014 at 01:38:46AM -0800, Michael Dalton wrote: > Add initial support for per-rx queue sysfs attributes to virtio-net. If > mergeable packet buffers are enabled, adds a read-only mergeable packet > buffer size sysfs attribute for each RX queue. > > Signed-off-by: Michael Dalton <mwdalton at google.com> > --- > drivers/net/virtio_net.c | 66

Since sequence count algorithm is done by hypervisor, better to not reuse seqcount. Still concerned that the code is racy. -----Original Message----- From: Thomas Gleixner [mailto:tglx at linutronix.de] Sent: Friday, February 10, 2017 4:28 AM To: Vitaly Kuznetsov <vkuznets at redhat.com> Cc: Stephen Hemminger <sthemmin at microsoft.com>; x86 at kernel.org; Andy Lutomirski <luto at

Since sequence count algorithm is done by hypervisor, better to not reuse seqcount. Still concerned that the code is racy. -----Original Message----- From: Thomas Gleixner [mailto:tglx at linutronix.de] Sent: Friday, February 10, 2017 4:28 AM To: Vitaly Kuznetsov <vkuznets at redhat.com> Cc: Stephen Hemminger <sthemmin at microsoft.com>; x86 at kernel.org; Andy Lutomirski <luto at

Stephen Hemminger <sthemmin at microsoft.com> writes: > Why not use existing seqlock's? > To be honest I don't quite understand how we could use it -- the sequence locking here is done against the page updated by the hypersior, we're not creating new structures (so I don't understand how we could use struct seqcount which we don't have) but I may be

Stephen Hemminger <sthemmin at microsoft.com> writes: > Why not use existing seqlock's? > To be honest I don't quite understand how we could use it -- the sequence locking here is done against the page updated by the hypersior, we're not creating new structures (so I don't understand how we could use struct seqcount which we don't have) but I may be

Satistics counters is useful for debugging and performance optimization, so this patch lets virtio_net driver collect following and export them to userspace through "ethtool -S": - number of packets sent/received - number of bytes sent/received - number of callbacks for tx/rx - number of kick for tx/rx - number of bytes/packets queued for tx As virtnet_stats were per-cpu, so both

Satistics counters is useful for debugging and performance optimization, so this patch lets virtio_net driver collect following and export them to userspace through "ethtool -S": - number of packets sent/received - number of bytes sent/received - number of callbacks for tx/rx - number of kick for tx/rx - number of bytes/packets queued for tx As virtnet_stats were per-cpu, so both

Hi Jason, Michael Sorry for the delay in response. Jason, I agree this patch ended up being larger than expected. The major implementation parts are: (1) Setup directory structure (driver/per-netdev/rx-queue directories) (2) Network device renames (optional, so debugfs dir has the right name) (3) Support resizing the # of RX queues (optional - we could just export max_queue_pairs files and

Hi Jason, Michael Sorry for the delay in response. Jason, I agree this patch ended up being larger than expected. The major implementation parts are: (1) Setup directory structure (driver/per-netdev/rx-queue directories) (2) Network device renames (optional, so debugfs dir has the right name) (3) Support resizing the # of RX queues (optional - we could just export max_queue_pairs files and

Currently, we store the statistics in the independent fields of virtnet_stats, this is not scalable when we want to add more counters. As suggested by Michael, this patch convert it to an array and use the enum as the index to access them. Signed-off-by: Jason Wang <jasowang at redhat.com> --- drivers/net/virtio_net.c | 30 +++++++++++++++++------------- 1 files changed, 17

Currently, we store the statistics in the independent fields of virtnet_stats, this is not scalable when we want to add more counters. As suggested by Michael, this patch convert it to an array and use the enum as the index to access them. Signed-off-by: Jason Wang <jasowang at redhat.com> --- drivers/net/virtio_net.c | 30 +++++++++++++++++------------- 1 files changed, 17

On Fri, 23 Oct 2020 14:07:13 +0200 Mike Galbraith wrote: > On Fri, 2020-10-23 at 11:01 +0200, Sebastian Andrzej Siewior wrote: > > On 2020-10-22 07:28:20 [+0200], Mike Galbraith wrote: > > > I've only as yet seen nouveau lockdep gripage when firing up one of my > > > full distro KVM's. > > > > Could you please check !RT with the `threadirqs'

We don't maintain tx counters in rx stats any more. There's no need for an extra container of rq stats. Cc: Toshiaki Makita <makita.toshiaki at lab.ntt.co.jp> Signed-off-by: Jason Wang <jasowang at redhat.com> --- drivers/net/virtio_net.c | 80 ++++++++++++++++++++++-------------------------- 1 file changed, 36 insertions(+), 44 deletions(-) diff --git

On Wed, Jul 03, 2019 at 11:44:58AM -0700, Christoph Hellwig wrote: > Checking range->valid is trivial and has no meaningful cost, but > nicely simplifies the fastpath in typical callers. It should not be the typical caller.. > hmm_vma_range_done function, which now is a trivial wrapper around > hmm_range_unregister. > > Signed-off-by: Christoph Hellwig <hch at

Add initial support for per-rx queue sysfs attributes to virtio-net. If mergeable packet buffers are enabled, adds a read-only mergeable packet buffer size sysfs attribute for each RX queue. Signed-off-by: Michael Dalton <mwdalton at google.com> --- drivers/net/virtio_net.c | 66 +++++++++++++++++++++++++++++++++++++++++++++--- 1 file changed, 62 insertions(+), 4 deletions(-) diff --git

Hi Michael, On Thu, Jan 16, 2014 at 3:53 AM, Michael S. Tsirkin <mst at redhat.com> wrote: > Hmm this adds overhead just to prevent sysfs from getting wrong value. > Can't sysfs simply disable softirq while it's reading the value? Yes I think this would work, we could call napi_disable(), read the average packet length from the receive_queue, and then call

On Fri 23-02-18 15:47:36, Mark Rutland wrote: > Hi all, > > While fuzzing arm64/v4.16-rc2 with syzkaller, I simultaneously hit a > number of splats in the block layer: > > * inconsistent {HARDIRQ-ON-W} -> {IN-HARDIRQ-R} usage in > jbd2_trans_will_send_data_barrier > > * BUG: sleeping function called from invalid context at mm/mempool.c:320 > > * WARNING: