On Sun, 2009-05-24 at 19:45 -0700, Josh Wilsdon wrote:
> the kernel refuses to boot with the error message:
>
> "This kernel requires an x86-64 CPU, but only detected an i686 CPU.
> Unable to boot - please use a kernel appropriate for your CPU."
This should fix it:
http://git.kernel.
On Tue, 2009-05-26 at 13:11 +0300, Avi Kivity wrote:
> Gleb Natapov wrote:
> > On Tue, May 26, 2009 at 01:05:35PM +0300, Avi Kivity wrote:
> >
> >> Gleb Natapov wrote:
> >>
> >>> It gets packet without virtio header and adds it if needed. Allows to
> >>> inject packets to vlan from outside
On Tue, 2009-05-26 at 13:03 +0300, Gleb Natapov wrote:
> It gets packet without virtio header and adds it if needed. Allows to
> inject packets to vlan from outside. To send gracious arp for instance.
Acked-by: Mark McLoughlin
This isn't ideal, but neither is the current VNET_HDR s
On Wed, 2009-05-27 at 15:11 +0300, Avi Kivity wrote:
> Multiple cookies on the same address are required by virtio. You can't
> mux since the data doesn't go anywhere.
>
> Virtio can survive by checking all rings on a notify, and we can later
> add a mechanism that has a distinct address for e
On Wed, 2009-05-27 at 13:40 -0400, Gregory Haskins wrote:
> Mark McLoughlin wrote:
> > On Wed, 2009-05-27 at 15:11 +0300, Avi Kivity wrote:
> >
> >
> >> Multiple cookies on the same address are required by virtio. You can't
> >> mux since the da
On Wed, 2009-05-27 at 16:45 -0400, Gregory Haskins wrote:
> Mark McLoughlin wrote:
> > The virtio ABI is fixed, so we couldn't e.g. have the guest use a cookie
> > to identify a queue - it's just going to continue using a per-device
> > queue number.
>
> Act
On Thu, 2009-06-04 at 09:20 +0200, Gilles PIETRI wrote:
> I'm quite pissed off. I just upgraded to kvm-86 on a host that has
> worked nicely on kvm-78 for quite some time. But since I was fearing the
> qcow2 corruption issues, I wanted to upgrade kvm-86. After testing the
> performance, I decid
Hi Greg,
On Wed, 2009-06-03 at 18:04 -0400, Gregory Haskins wrote:
> Hi Mark,
> So with the v5 release of iosignalfd, we now have the notion of a
> "trigger", the API of which is as follows:
>
> ---
> /*!
> * \brief Assign an eventfd to an IO port (PIO or MMIO)
> *
> * As
On Tue, 2009-06-09 at 11:06 +, Fischer, Anna wrote:
> I am testing network throughput between two guests residing on the
> same physical machine. I use a bridge to pass packets between those
> guests and the virtio NIC model. I am wondering why the throughput
> only goes up to about 970Mbps. S
nitor: Remove unused variable
monitor: Introduce get_command_name()
Fix "defined but not used" warning
Mark McLoughlin (22):
Revert "Fix output of uninitialized strings"
net: fix error reporting for some net parameter checks
net: factor tap_read_packet()
On Wed, 2009-06-10 at 20:27 +0100, Jamie Lokier wrote:
> Michael S. Tsirkin wrote:
> > > I think the right long term answer to all this is a way to get QEMU to
> > > dump it's current machine configuration in glorious detail as a file
> > > which can be reloaded as a machine configuration.
> >
> >
On Fri, 2009-06-12 at 16:59 +0300, Michael S. Tsirkin wrote:
> On Fri, Jun 12, 2009 at 09:43:29AM +0100, Mark McLoughlin wrote:
> > = Solution - Separate configuration from compat hints =
> >
> > As I suggested before:
> >
> > - Allow the VM manager to du
On Fri, 2009-06-12 at 09:51 -0500, Anthony Liguori wrote:
> Mark McLoughlin wrote:
> > On Wed, 2009-06-10 at 20:27 +0100, Jamie Lokier wrote:
> >
> >> Michael S. Tsirkin wrote:
> >>
> >>>> I think the right long term answer to all this
On Fri, 2009-06-12 at 09:55 -0500, Anthony Liguori wrote:
> Mark McLoughlin wrote:
> > On Wed, 2009-06-10 at 20:27 +0100, Jamie Lokier wrote:
> >
> > = Solution - Separate configuration from compat hints =
> >
> > As I suggested before:
> >
> >
On Fri, 2009-06-12 at 11:11 -0500, Anthony Liguori wrote:
> Mark McLoughlin wrote:
> > On Fri, 2009-06-12 at 09:51 -0500, Anthony Liguori wrote:
> >
> >> Mark McLoughlin wrote:
> >>
> >>> On Wed, 2009-06-10 at 20:27 +0100, Jamie Lokier wrote:
&g
On Fri, 2009-06-12 at 11:12 -0500, Anthony Liguori wrote:
> Mark McLoughlin wrote:
> > On Fri, 2009-06-12 at 09:55 -0500, Anthony Liguori wrote:
> >
> >> Mark McLoughlin wrote:
> >>
> >>> On Wed, 2009-06-10 at 20:27 +0100, Jamie Lokie
On Fri, 2009-06-12 at 12:00 -0500, Anthony Liguori wrote:
> Mark McLoughlin wrote:
> > So, when libvirt creates a guest for the first time, it makes a copy of
> > the device tree and continues to use that even if qemu is upgraded.
> > That's enough to ensure compat is
On Fri, 2009-06-12 at 20:44 +0300, Blue Swirl wrote:
> If the device has different behavior or different properties from
> guest perspective compared to the old device, it should get a new
> device type so that you could specify in the device tree either the
> old device or the new one.
Yes, that
On Sun, 2009-06-14 at 12:34 +0300, Michael S. Tsirkin wrote:
> On Fri, Jun 12, 2009 at 04:53:27PM +0100, Mark McLoughlin wrote:
> > The other obvious piece to add to it would be PCI addresses, so that
> > even if you remove a device, the addresses assigned to existing devices
&
On Sun, 2009-06-14 at 12:50 +0300, Michael S. Tsirkin wrote:
> On Fri, Jun 12, 2009 at 05:48:23PM +0100, Mark McLoughlin wrote:
> > However, in order to retain compat for that SCSI device (e.g. ensuring
> > the PCI address doesn't change as other devices are added an removed
On Sun, 2009-06-14 at 10:58 +0300, Avi Kivity wrote:
> Mark McLoughlin wrote:
>
>
>
> >> I think the point is that you don't need version numbers if you have a
> >> proper device tree.
> >>
> >
> > How do you add a new attribute
On Mon, 2009-06-15 at 07:48 -0500, Anthony Liguori wrote:
> Avi Kivity wrote:
> > On 06/15/2009 12:09 PM, Mark McLoughlin wrote:
> >>>>> I think the point is that you don't need version numbers if you
> >>>>> have a
> >>>>> prope
On Mon, 2009-06-15 at 18:12 +0300, Dor Laor wrote:
> > It doesn't want to. As Mark said, libvirt just wants to be able to ensure
> > a stable guest ABI, of which stable PCI addresses is one aspect. This does
> > not imply libvirt wants to allocate the PCI addresses, just that it wants
> > a way to
On Mon, 2009-06-15 at 18:05 +0300, Avi Kivity wrote:
> On 06/15/2009 05:24 PM, Anthony Liguori wrote:
> > Dor Laor wrote:
> >> Libvirt does not support r2d. I hope it won't start to support it.
> >
> > It supports mips, sparc, and ppc machines now. I don't see why it
> > wouldn't support r2d. Fo
On Mon, 2009-06-15 at 13:12 -0500, Anthony Liguori wrote:
> Mark McLoughlin wrote:
> > So long as the restrictions would be known to the management app via
> > some "what slots are available" mechanism in qemu, that sounds fine.
> >
>
> I'm not sure a
On Tue, 2009-06-16 at 15:28 +0300, Avi Kivity wrote:
> On 06/16/2009 03:14 PM, Mark McLoughlin wrote:
> > On Mon, 2009-06-15 at 13:12 -0500, Anthony Liguori wrote:
> >
> >> Mark McLoughlin wrote:
> >>
> >>> So long as the restrictions would
On Tue, 2009-06-16 at 19:44 +0100, Jamie Lokier wrote:
> Mark McLoughlin wrote:
> > > Worst case we hardcode those numbers (gasp, faint).
> >
> > Maybe we can just add the open slots to the -help output. That'd be nice
> > and clean.
I was being sarcastic
On Wed, 2009-06-17 at 12:03 +0300, Avi Kivity wrote:
> On 06/17/2009 11:33 AM, Mark McLoughlin wrote:
> >> I particularly don't like the idea of arcane machine-dependent slot
> >> allocation knowledge living in libvirt, because it needs to be in Qemu
> >> anyway
On Wed, 2009-06-17 at 17:44 +0400, Michael Tokarev wrote:
> WARNING: at drivers/base/core.c:122 device_release+0x5f/0x70()
> Device 'virtio8' does not have a release() function, it is broken and
> must be fixed.
This should be fixed in 2.6.29
Cheers,
Mark.
--
To unsubscribe from this list: send
function ‘kvm_get_pit_ch2’
hw/pcspk.c:185: error: too many arguments to function ‘kvm_set_pit_ch2’
hw/pcspk.c:171: warning: unused variable ‘inkernel_state’
Signed-off-by: Mark McLoughlin
Cc: Jan Kiszka
---
hw/pcspk.c |6 +-
1 files changed, 5 insertions(+), 1 deletions(-)
diff --git a
On Thu, 2009-07-02 at 12:41 +0300, Avi Kivity wrote:
> On 07/02/2009 08:48 AM, Avi Kivity wrote:
> >> HOST time (make -j12&& make -j12 modules) with no guest running
> >>
> >> real6m50.936s
> >> user29m12.051s
> >> sys5m
/?func=detail&atid=893831&aid=2817367&group_id=180599
Network doesn't work anymore switching from kvm-86 to kvm-87
Signed-off-by: Mark McLoughlin
---
net.c | 11 ++-
1 files changed, 10 insertions(+), 1 deletions(-)
diff --git a/net.c b/net.c
index ac06ea1..19e
Hi Anthony,
On Mon, 2009-06-22 at 18:57 -0500, Anthony Liguori wrote:
> Hi,
>
> It's getting to be about the time to start thinking about the 0.11.0
> release. 0.10.0 was released on March 2nd so following with the 6 month
> release cycle, that would put 0.11.0 at September 2nd.
>
> Based on
Hi Dor,
On Wed, 2009-07-15 at 18:03 +0300, Dor Laor wrote:
> If the migration socket is full, we get EAGAIN for the write.
> The set_fd_handler2 defers the write for later on. The function
> tries to wake up the iothread by qemu_kvm_notify_work.
> Since this happens in a loop, multiple times, the
On Thu, 2009-07-16 at 15:00 +0300, Dor Laor wrote:
> On 07/16/2009 12:39 PM, Mark McLoughlin wrote:
> > Hi Dor,
> >
> > On Wed, 2009-07-15 at 18:03 +0300, Dor Laor wrote:
> >> If the migration socket is full, we get EAGAIN for the write.
> >> The set_fd_hand
table-0.11 too]
Signed-off-by: Mark McLoughlin
---
net.c |6 ++
1 files changed, 6 insertions(+), 0 deletions(-)
diff --git a/net.c b/net.c
index c7702f8..2428f63 100644
--- a/net.c
+++ b/net.c
@@ -1521,6 +1521,9 @@ int tap_has_vnet_hdr(void *opaque)
VLANClientState *vc = o
On Fri, 2009-08-07 at 13:51 +0200, Arnd Bergmann wrote:
> On Friday 07 August 2009, Mark McLoughlin wrote:
> > slirp has started using VLANClientState::opaque and this has caused the
> > kvm specific tap_has_vnet_hdr() hack to break because we blindly use
> > this opaque poin
On Sun, 2009-08-09 at 12:58 +0300, Avi Kivity wrote:
> On 08/07/2009 03:31 PM, Chris Lalancette wrote:
> > Avi,
> > Trying to use libvirt with development snapshots of qemu-kvm is a bit
> > problematic. The trouble is that for all development snapshots, the value
> > that
> > gets placed in
, 87 etc.) and just uses the 'qemu-kvm-'
string to detect that this is the kvm fork of qemu. The actual kvm
snapshot number is no longer important, it's the qemu version number
which is important.
Signed-off-by: Mark McLoughlin
---
configure |2 +-
1 files changed, 1 i
On Mon, 2009-08-10 at 12:46 +0300, Avi Kivity wrote:
> On 08/10/2009 12:19 PM, Mark McLoughlin wrote:
> >
> > Or we could drop this kvm snapshot numbering system and just use qemu
> > VERSION numbering - i.e. qemu-kvm-devel-88 could have been published as
> > qemu-kv
Hi Michael,
Some random high-level comments:
- I had expected this to be available as:
-net raw,ifname=eth2 -net nic,model=virtio
I'd prefer it this way, because it means you can use this mode even
without vhost and it's ties in better with the way all other qemu
networkin
On Thu, 2009-09-03 at 12:31 -0500, Dustin Kirkland wrote:
> qemu-kvm: fix segfault when running kvm without /dev/kvm, falling back
> to non-accelerated mode
>
> We're seeing segfaults on systems without access to /dev/kvm. It
> looks like the global kvm_allowed is being set just a little too late
On Thu, 2009-09-03 at 19:18 -0300, Glauber Costa wrote:
> On Thu, Sep 3, 2009 at 6:48 PM, Dustin Kirkland wrote:
> > On Thu, Sep 3, 2009 at 3:01 PM, Mark McLoughlin wrote:
> >> On Thu, 2009-09-03 at 12:31 -0500, Dustin Kirkland wrote:
> >>> qemu-kvm: fix segfault whe
On Sat, 2009-07-25 at 12:30 -0300, Marcelo Tosatti wrote:
> On Thu, Jul 23, 2009 at 04:34:13PM +0300, Michael S. Tsirkin wrote:
> > When adding a vector fails, the used counter should
> > not be incremented, otherwise on vector change we will
> > try to update the routing entry.
> >
> > Signed-off
On Tue, 2009-05-05 at 09:56 +0100, Mark McLoughlin wrote:
> This commit:
>
>commit 559a8f45f34cc50d1a60b4f67a06614d506b2e01
>Subject: Remove stray GSO code from virtio_net (Mark McLoughlin)
>
> Removed some GSO code from upstream qemu.git, but it needs to
> be re-ins
On Tue, 2009-09-29 at 21:45 +0100, Mark McLoughlin wrote:
> On Tue, 2009-05-05 at 09:56 +0100, Mark McLoughlin wrote:
> > This commit:
> >
> >commit 559a8f45f34cc50d1a60b4f67a06614d506b2e01
> >Subject: Remove stray GSO code from virtio_net (Mark McLoughlin)
>
On Wed, 2009-09-30 at 08:24 -0300, Glauber Costa wrote:
> On Wed, Sep 30, 2009 at 08:24:18AM +0200, Avi Kivity wrote:
> > On 09/29/2009 10:45 PM, Mark McLoughlin wrote:
> >> On Tue, 2009-05-05 at 09:56 +0100, Mark McLoughlin wrote:
> >>
> >>&
On Wed, 2009-09-30 at 15:55 +0200, Avi Kivity wrote:
> On 09/30/2009 03:51 PM, Christoph Hellwig wrote:
> > I might sound like a broken record, but why isn't the full GSO support
> > for virtio-net upstream in qemu?
> >
>
> IIRC the current hacks are not upstream quality.
>
> The problem (aga
On Wed, 2009-09-30 at 21:15 +0200, Gerd Hoffmann wrote:
> On 09/30/09 15:59, Mark McLoughlin wrote:
> > I'm planning on adding -hostnet and -nic arguments, which would not use
> > vlans by default but rather connect the nic directly to the host side.
>
> No new -nic a
On Thu, 2009-10-01 at 18:49 +0200, Avi Kivity wrote:
> On 09/30/2009 03:59 PM, Mark McLoughlin wrote:
> > I think we should keep the vlan stuff, just de-emphasise it.
> >
>
> Maybe we should do what X.org does, break it silently and remove it some
> time later when
On Wed, 2009-10-07 at 14:50 -0700, Sridhar Samudrala wrote:
> linux 2.6.32 includes UDP fragmentation offload support in software.
> So we can enable UFO on the host tap device if supported and allow setting
> UFO on virtio-net in the guest.
Hmm, we really need to detect whether the host has tunt
On Thu, 2009-10-08 at 15:31 -0700, Sridhar Samudrala wrote:
> On Thu, 2009-10-08 at 11:07 +0100, Mark McLoughlin wrote:
> > On Wed, 2009-10-07 at 14:50 -0700, Sridhar Samudrala wrote:
> > > linux 2.6.32 includes UDP fragmentation offload support in software.
> > > So we
Hi Tom,
On Mon, 2009-10-12 at 17:05 -0500, Tom Lendacky wrote:
> The recent change to QemuOpts for the -net nic option breaks specifying -net
> nic,... more than once. The net_init_nic function's return value in net.c is
> a table index, which is non-zero after the first time it is called. The
an fd and b) that the TUNGETIFF ioctl() is available in
the running kernel.
We add a comment to the "qemu -help" output to indicate to
users that this version of qemu supports IFF_VNET_HDR.
Signed-off-by: Mark McLoughlin <[EMAIL PROTECTED]>
---
qemu/vl.c | 21
On Wed, 2008-08-20 at 18:05 +0100, Mark McLoughlin wrote:
> Allow users to pass an IFF_VNET_HDR tap fd via "-net tap,fd=X"
> by querying the fd with the recently added TUNGETIFF ioctl() to
> see if IFF_VNET_HDR has been enabled.
This was merged by Linus today:
http://git.
On Wed, 2008-08-20 at 20:51 +0300, Avi Kivity wrote:
> Anthony Liguori wrote:
> > Mark McLoughlin wrote:
> >> Allow users to pass an IFF_VNET_HDR tap fd via "-net tap,fd=X"
> >> by querying the fd with the recently added TUNGETIFF ioctl() to
>
Hi,
On Sat, 2008-08-23 at 21:21 +0200, Fabian Deutsch wrote:
> Hey.
>
> When running a guest in the following setup, I observ a "massive"
> performance decrease after a couple of gigabytes of transfrerred data.
>
> other guests keep running fine (no regressions) and nothing unusual is
> observed
On Tue, 2008-09-09 at 12:32 +0200, Fabian Deutsch wrote:
> Am Dienstag, den 09.09.2008, 11:22 +0100 schrieb Mark McLoughlin:
> > Hi,
> >
> > On Sat, 2008-08-23 at 21:21 +0200, Fabian Deutsch wrote:
> > > Hey.
> > >
> > > When running a g
On Tue, 2008-09-09 at 13:11 +0200, Henrik Holst wrote:
> tis 2008-09-09 klockan 11:47 +0100 skrev Mark McLoughlin:
> > On Tue, 2008-09-09 at 12:32 +0200, Fabian Deutsch wrote:
> > > Am Dienstag, den 09.09.2008, 11:22 +0100 schrieb Mark McLoughlin:
> > > > Hi,
> >
gt; +}
> +
> +/hdr->csum_offset = skb->csum_offset/ {
> need_endif = 1;
> }
This looks right to me; I had a slightly uglier version that I forgot to
upstream.
Acked-by: Mark McLoughlin <[EMAIL PROTECTED]>
Cheers,
Mark.
--
To unsubscribe from this list: send th
On Tue, 2008-09-16 at 22:24 +0300, Ben-Ami Yassour wrote:
> On Tue, 2008-09-16 at 09:16 -0500, Anthony Liguori wrote:
> > Ben-Ami Yassour wrote:
> > > I am running virtio with the latest KVM code, and see a significant
> > > performance issue.
> > >
> > > Ping to the host (or any other close machin
In order to de-obfuscate the final patch in this series
Signed-off-by: Mark McLoughlin <[EMAIL PROTECTED]>
---
qemu/hw/virtio-net.c |9 +
1 files changed, 5 insertions(+), 4 deletions(-)
diff --git a/qemu/hw/virtio-net.c b/qemu/hw/virtio-net.c
index bc2ede6..4b4c48b 100644
Split virtqueue_push() into two logical steps - adding an element
to the used ring and notifying the other side of added elements.
Signed-off-by: Mark McLoughlin <[EMAIL PROTECTED]>
---
qemu/hw/virtio.c | 23 ++-
qemu/hw/virtio.h |3 +++
2 files changed, 21 inse
virtio_next_desc() is only used in virtqueue_pop(), so move
it alongside that.
Signed-off-by: Mark McLoughlin <[EMAIL PROTECTED]>
---
qemu/hw/virtio.c | 30 +++---
1 files changed, 15 insertions(+), 15 deletions(-)
diff --git a/qemu/hw/virtio.c b/qemu/hw/virtio.c
benchmark
with GSO disabled on the host side, throughput was seen to increase
from 700Mb/s to 1.7Gb/s.
Based on a patch from Herbert, with the feature renamed from
"datahead" and some re-factoring for readability.
Signed-off-by: Herbert Xu <[EMAIL PROTECTED]>
Signed-off-by: Mark McL
Simplifies the current code, but more especially, the code
in the next patch.
Signed-off-by: Mark McLoughlin <[EMAIL PROTECTED]>
---
qemu/hw/virtio-net.c | 32 +---
1 files changed, 21 insertions(+), 11 deletions(-)
diff --git a/qemu/hw/virtio-net.c b/q
On Sun, 2008-10-12 at 12:00 +0200, Avi Kivity wrote:
> Mark McLoughlin wrote:
> > diff --git a/qemu/hw/virtio-net.c b/qemu/hw/virtio-net.c
> > index 403247b..afa5fe5 100644
> > --- a/qemu/hw/virtio-net.c
> > +++ b/qemu/hw/virtio-net.c
> > @@ -34,9 +34,13 @@
&g
Since 2.6.24, if get_tx_csum, get_sg or get_tso are NULL then the
default ethtool_op_get_xxx functions are called. Before 2.6.24,
however, these need to be manually hooked in.
Signed-off-by: Mark McLoughlin <[EMAIL PROTECTED]>
---
hack-module.awk |6 ++
1 files changed, 6 inse
pping the extra reference taken.
Signed-off-by: Mark McLoughlin <[EMAIL PROTECTED]>
---
hack-module.awk |1 +
1 files changed, 1 insertions(+), 0 deletions(-)
diff --git a/hack-module.awk b/hack-module.awk
index 7cb9381..b401aae 100644
--- a/hack-module.awk
+++ b/hack-module.awk
@@
always.
Signed-off-by: Anthony Liguori <[EMAIL PROTECTED]>
Allowing the weight to be higher should reduce the interrupt
rate in the guest. In practice, though, it doesn't appear to
make much difference.
Signed-off-by: Mark McLoughlin <[EMAIL PROTECTED]>
Cc: Anthony Li
Hi Avi,
Here's a few virtio_net fixes for kvm-guest-drivers-linux.
Cheers,
Mark.
--
To unsubscribe from this list: send the line "unsubscribe kvm" in
the body of a message to [EMAIL PROTECTED]
More majordomo info at http://vger.kernel.org/majordomo-info.html
On Tue, 2008-10-28 at 12:06 +0200, [EMAIL PROTECTED] wrote:
...
> +static int get_real_device(AssignedDevice *pci_dev, uint8_t r_bus,
> + uint8_t r_dev, uint8_t r_func)
> +{
> +char dir[128], name[128];
> +int fd, r = 0;
> +FILE *f;
> +unsigned long long st
On Wed, 2008-10-29 at 12:31 +0200, Muli Ben-Yehuda wrote:
> On Tue, Oct 28, 2008 at 04:55:22PM +0000, Mark McLoughlin wrote:
> > On Tue, 2008-10-28 at 12:06 +0200, [EMAIL PROTECTED] wrote:
> > > +void assigned_dev_set_vector(int irq, int vector);
> > > +void assig
On Wed, 2008-10-29 at 12:31 +0200, Muli Ben-Yehuda wrote:
> On Tue, Oct 28, 2008 at 04:55:22PM +0000, Mark McLoughlin wrote:
> > nr_assigned_devices isn't actually used anywhere.
>
> Nuked.
Still there.
> > > +#define MAX_IO_REGIONS (6)
> >
> > Perha
On Wed, 2008-10-29 at 14:20 +0200, [EMAIL PROTECTED] wrote:
> diff --git a/qemu/hw/piix_pci.c b/qemu/hw/piix_pci.c
> index b9067b8..27d5f02 100644
> --- a/qemu/hw/piix_pci.c
> +++ b/qemu/hw/piix_pci.c
> @@ -246,9 +246,9 @@ static void piix3_set_irq(qemu_irq *pic, int
> irq_num, int level)
> int pi
Keep reading from eventfd until it's empty and aggregate the results.
Only really important when we're emulating eventfd using a pipe as
in that case we may have multiple "1" values queued up, potentially
leading to the pipe buffer filling up and the write side blocking.
Split out code for common eventfd operations to avoid
re-implementation.
Signed-off-by: Mark McLoughlin <[EMAIL PROTECTED]>
---
qemu/compatfd.c | 48
qemu/compatfd.h |2 ++
qemu/qemu-kvm.c | 42 ---
Hey,
The main patch in this series is 5/6 - it just kills off the
virtio_net tx mitigation timer and does all the tx I/O in the
I/O thread.
Below are the results I got from benchmarking guest->host and
host->guest on my machine.
There's enough numbers there to make anyone blind, but basically
t
Remove a variable from virtqueue_pop() which has been unused since the
very start.
Signed-off-by: Mark McLoughlin <[EMAIL PROTECTED]>
---
qemu/hw/virtio.c |3 +--
1 files changed, 1 insertions(+), 2 deletions(-)
diff --git a/qemu/hw/virtio.c b/qemu/hw/virtio.c
index e675f43..8fac354
This allows the guest vcpu thread to exit while the I/O thread is
churning away.
Signed-off-by: Mark McLoughlin <[EMAIL PROTECTED]>
---
qemu/hw/virtio-net.c |2 ++
1 files changed, 2 insertions(+), 0 deletions(-)
diff --git a/qemu/hw/virtio-net.c b/qemu/hw/virtio-net.c
index 0
qemu_eventfd() returns two file descriptors, both of which
must be closed when pipe() is used in the absence of eventfd().
Duplicate the eventfd() file descriptor so that closing
both descriptors will work in that case.
Signed-off-by: Mark McLoughlin <[EMAIL PROTECTED]>
---
qemu/comp
ant.
Signed-off-by: Mark McLoughlin <[EMAIL PROTECTED]>
---
qemu/hw/virtio-net.c | 79 -
1 files changed, 45 insertions(+), 34 deletions(-)
diff --git a/qemu/hw/virtio-net.c b/qemu/hw/virtio-net.c
index bc2ede6..0612f5f 100644
--- a/qemu
der e.g. an I/O thread per device.
Note also that when tuning for a specific workload, which CPU
the I/O thread is pinned to is important.
Signed-off-by: Mark McLoughlin <[EMAIL PROTECTED]>
---
qemu/hw/virtio-net.c | 60 +
1 files changed, 2
On Thu, 2008-10-30 at 14:20 -0500, Anthony Liguori wrote:
> Mark McLoughlin wrote:
> > | guest->host tput |
> > host->guest tput
> > netperf, 10x20s runs (Gb/s) |min/ mean/ max/stddev | mi
retry with a larger buffer. We could add that as
an extension, I guess.
Best short term thing to do is just increase the tap buffer size. Patch
below. A similar patch should go into qemu upstream.
Cheers,
Mark.
From: Mark McLoughlin <[EMAIL PROTECTED]>
Subject: [PATCH] kvm: qemu: increase size o
On Sun, 2008-11-02 at 11:48 +0200, Avi Kivity wrote:
> Mark McLoughlin wrote:
> > Hey,
> >
> > The main patch in this series is 5/6 - it just kills off the
> > virtio_net tx mitigation timer and does all the tx I/O in the
> > I/O thread.
> >
> >
>
On Mon, 2008-11-03 at 14:40 +0200, Avi Kivity wrote:
> Mark McLoughlin wrote:
> > On Sun, 2008-11-02 at 11:48 +0200, Avi Kivity wrote:
> >
> >> Mark McLoughlin wrote:
> >>> The main patch in this series is 5/6 - it just kills off the
> >>> virtio
On Fri, 2008-10-31 at 09:16 +, Mark McLoughlin wrote:
> +static void virtio_net_tx_bh(void *opaque)
> {
> VirtIONet *n = opaque;
>
> -n->tx_timer_active = 0;
> -
> -/* Just in case the driver is not ready on more */
> -if (!(n->vdev.stat
On Wed, 2008-11-05 at 22:49 -0600, Hollis Blanchard wrote:
> - info->queue = kzalloc(PAGE_ALIGN(vring_size(num,PAGE_SIZE)),
> GFP_KERNEL);
> + vring_bytes = PAGE_ALIGN(vring_size(num, VRING_PAGE_SIZE));
> + info->queue = kzalloc(vring_bytes, GFP_KERNEL);
You're still aligning t
On Wed, 2008-11-05 at 22:49 -0600, Hollis Blanchard wrote:
> diff --git a/qemu/hw/virtio.c b/qemu/hw/virtio.c
> --- a/qemu/hw/virtio.c
> +++ b/qemu/hw/virtio.c
> @@ -56,6 +56,10 @@
> */
> #define wmb() do { } while (0)
>
> +#define VRING_PAGE_SIZE (1<<12)
> +
> +#define ALIGN(x, a) (((x)+(a)
Hey,
So, I went off and spent some time gathering more data on this stuff
and putting it together in a more consumable fashion.
Here are some graphs showing the effect some of these changes have on
throughput, cpu utilization and vmexit rate:
http://markmc.fedorapeople.org/virti
On Tue, 2008-11-04 at 08:23 -0700, David S. Ahern wrote:
>
> Mark McLoughlin wrote:
>
> > Note also that when tuning for a specific workload, which CPU
> > the I/O thread is pinned to is important.
> >
>
> Hi Mark:
>
> Can you give an example of when
Hi Avi,
Just thinking about your variable window suggestion ...
On Mon, 2008-11-03 at 14:40 +0200, Avi Kivity wrote:
> Mark McLoughlin wrote:
> > On Sun, 2008-11-02 at 11:48 +0200, Avi Kivity wrote:
> >> Where does the benefit come from?
> >>
> >
> >
On Thu, 2008-11-13 at 13:27 +0100, Emmanuel Lacour wrote:
> Dear kvm users/developpers,
>
> I have a problem here where the network interface of a guest hang
> 2 or 3 times a day. No more packets can be sent out or received, no
> error in guest or host logs. I have to stop networking, remove modul
On Fri, 2008-11-14 at 10:23 +0100, Emmanuel Lacour wrote:
> On Thu, Nov 13, 2008 at 04:24:52PM +0100, Emmanuel Lacour wrote:
> > On Thu, Nov 13, 2008 at 03:12:33PM +0000, Mark McLoughlin wrote:
> > > The fact that re-loading the virtio_net driver fixes things up makes me
> >
On Tue, 2008-11-18 at 19:37 +0100, Emmanuel Lacour wrote:
> On Fri, Nov 14, 2008 at 06:26:44PM +0000, Mark McLoughlin wrote:
> >
> > Right, the tap device tx queue is full because kvm-userspace isn't
> > reading packets from it.
> >
> > This could be
On Wed, 2008-11-19 at 13:13 +, Mark McLoughlin wrote:
> On Tue, 2008-11-18 at 19:37 +0100, Emmanuel Lacour wrote:
> > On Fri, Nov 14, 2008 at 06:26:44PM +0000, Mark McLoughlin wrote:
> > >
> > > Right, the tap device tx queue is full because kvm-userspace isn'
_stat ".../asm/vmx*.h" failed: No such file or directory (2)
Fix by passing the expanded paths returned by the wildcard function to
rsync.
Signed-off-by: Mark McLoughlin <[EMAIL PROTECTED]>
---
kernel/Makefile |2 +-
1 files changed, 1 insertions(+), 1 deletions(-)
diff --gi
Simplifies the current code, but more especially, simplifies the
implementation of the mergeable receive buffers scheme.
Signed-off-by: Mark McLoughlin <[EMAIL PROTECTED]>
---
qemu/hw/virtio-net.c | 27 ++-
1 files changed, 18 insertions(+), 9 deletions(-)
diff -
Hi,
The mergeable receive buffers scheme has been in the Linux
virtio_net driver in net-next-2.6 for a while now, so it's probably
safe to assume what's there now is the final iteration of the ABI.
The following patches implement support for the scheme in
KVM. Only the 5/5 patch c
1 - 100 of 358 matches
Mail list logo
