This adds a way to check ring empty state after enable_cb outside any
locks. Will be used by virtio_net.
Note: there's room for more optimization: caller is likely to have a
memory barrier already, which means we might be able to get rid of a
barrier here. Deferring this optimization until we do
virtio net called virtqueue_enable_cq on RX path after napi_complete, so
with NAPI_STATE_SCHED clear - outside the implicit napi lock.
This violates the requirement to synchronize virtqueue_enable_cq wrt
virtqueue_add_buf. In particular, used event can move backwards,
causing us to lose interrupts
For small packets we can simplify xmit processing
by linearizing buffers with the header:
most packets seem to have enough head room
we can use for this purpose.
Since existing hypervisors require that header
is the first s/g element, we need a feature bit
for this.
Signed-off-by: Michael S. Tsirk
The following changes since commit 8bb495e3f02401ee6f76d1b1d77f3ac9f079e376:
Linux 3.10 (2013-06-30 15:13:29 -0700)
are available in the git repository at:
git://git.kernel.org/pub/scm/linux/kernel/git/mst/vhost.git tags/for_linus
for you to fetch changes up to 09a34c8404c1d4c5782de319c02e1
Hello.
On 08-07-2013 13:04, Michael S. Tsirkin wrote:
virtio net called virtqueue_enable_cq on RX path after napi_complete, so
with NAPI_STATE_SCHED clear - outside the implicit napi lock.
This violates the requirement to synchronize virtqueue_enable_cq wrt
virtqueue_add_buf. In particular, us
On Mon, Jul 08, 2013 at 04:52:26PM +0400, Sergei Shtylyov wrote:
> Hello.
>
> On 08-07-2013 13:04, Michael S. Tsirkin wrote:
>
> >virtio net called virtqueue_enable_cq on RX path after napi_complete, so
> >with NAPI_STATE_SCHED clear - outside the implicit napi lock.
> >This violates the requirem
On Mon, Jul 08, 2013 at 08:28:24AM +0200, Michael Opdenacker wrote:
> This patch proposes to remove the XEN_PRIVILEGED_GUEST kernel
> configuration parameter defined in arch/x86/xen/Kconfig, but used
> nowhere in the makefiles and source code.
>
> This dummy parameter was added 3 years back, and i
On 07/08/2013 12:28 PM, Konrad Rzeszutek Wilk wrote:
> On Mon, Jul 08, 2013 at 08:28:24AM +0200, Michael Opdenacker wrote:
>> This patch proposes to remove the XEN_PRIVILEGED_GUEST kernel
>> configuration parameter defined in arch/x86/xen/Kconfig, but used
>> nowhere in the makefiles and source cod
On 07/08/2013 12:34 PM, Matt Wilson wrote:
>
> For reference, here's a related Debian bug report:
> http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=633127
>
> Summary: grub2 scripts parse /boot/config-$(uname -r)
>
What. The. Fuck.
-hpa
Paolo Bonzini writes:
> Il 04/07/2013 09:39, Michael S. Tsirkin ha scritto:
>>> > Transport feature bits are still negotiated per device though.
>>> > virtio-scsi devices in QEMU would not negotiate that feature.
>>> >
>>> > Paolo
>>> >
> > So I think it needs to be a per-device feature bit
On Mon, 2013-07-08 at 22:58 +0200, Borislav Petkov wrote:
> On Mon, Jul 08, 2013 at 01:29:40PM -0700, H. Peter Anvin wrote:
> > What. The. Fuck.
>
> This is just marvellous: grub2 has a bunch of scripts in /etc/grub.d
> which rely on the presence of kernel config files in /boot or / and
> greps th
Paul Bolle wrote:
>On Mon, 2013-07-08 at 22:58 +0200, Borislav Petkov wrote:
>> On Mon, Jul 08, 2013 at 01:29:40PM -0700, H. Peter Anvin wrote:
>> > What. The. Fuck.
>>
>> This is just marvellous: grub2 has a bunch of scripts in /etc/grub.d
>> which rely on the presence of kernel config files in
"Michael S. Tsirkin" writes:
> On Mon, Jul 08, 2013 at 10:51:39AM +0930, Rusty Russell wrote:
>> "Michael S. Tsirkin" writes:
>> > On Thu, Jul 04, 2013 at 10:49:42AM +0300, Michael S. Tsirkin wrote:
>> >> > case).
>> >> >
>> >> > Regrettably, initial driver implementations used simple layouts
>>
"Michael S. Tsirkin" writes:
> For small packets we can simplify xmit processing
> by linearizing buffers with the header:
> most packets seem to have enough head room
> we can use for this purpose.
> Since existing hypervisors require that header
> is the first s/g element, we need a feature bit
On 07/08/2013 05:04 PM, Michael S. Tsirkin wrote:
> This adds a way to check ring empty state after enable_cb outside any
> locks. Will be used by virtio_net.
>
> Note: there's room for more optimization: caller is likely to have a
> memory barrier already, which means we might be able to get rid o
On 07/08/2013 05:04 PM, Michael S. Tsirkin wrote:
> virtio net called virtqueue_enable_cq on RX path after napi_complete, so
> with NAPI_STATE_SCHED clear - outside the implicit napi lock.
> This violates the requirement to synchronize virtqueue_enable_cq wrt
> virtqueue_add_buf. In particular, us
virtio net called virtqueue_enable_cq on RX path after napi_complete, so
with NAPI_STATE_SCHED clear - outside the implicit napi lock.
This violates the requirement to synchronize virtqueue_enable_cq wrt
virtqueue_add_buf. In particular, used event can move backwards,
causing us to lose interrupts
This adds a way to check ring empty state after enable_cb outside any
locks. Will be used by virtio_net.
Note: there's room for more optimization: caller is likely to have a
memory barrier already, which means we might be able to get rid of a
barrier here. Deferring this optimization until we do
On Tue, Jul 09, 2013 at 11:46:23AM +0930, Rusty Russell wrote:
> "Michael S. Tsirkin" writes:
> > For small packets we can simplify xmit processing
> > by linearizing buffers with the header:
> > most packets seem to have enough head room
> > we can use for this purpose.
> > Since existing hypervi
19 matches
Mail list logo