[Qemu-devel] [PATCH v2] virtio-pci: Disable virtio-ioeventfd when !CONFIG_IOTHREAD

[Qemu-devel] [PATCH v2] virtio-pci: Disable virtio-ioeventfd when !CONFIG_IOTHREAD

[Stefan Hajnoczi]

It is not possible to use virtio-ioeventfd when building without an I/O
thread.  We rely on a signal to kick us out of vcpu execution.  Timers
and AIO use SIGALRM and SIGUSR2 respectively.  Unfortunately eventfd
does not support O_ASYNC (SIGIO) so eventfd cannot be used in a signal
driven manner.

Signed-off-by: Stefan Hajnoczi <stefanha@linux.vnet.ibm.com>
---
 kvm-all.c |    8 ++++++--
 1 files changed, 6 insertions(+), 2 deletions(-)

diff --git a/kvm-all.c b/kvm-all.c
index 255b6fa..881ea32 100644
--- a/kvm-all.c
+++ b/kvm-all.c
@@ -449,10 +449,14 @@ int kvm_check_extension(KVMState *s, unsigned int extension)
 
 static int kvm_check_many_ioeventfds(void)
 {
-    /* Older kernels have a 6 device limit on the KVM io bus.  Find out so we
+    /* Userspace can use ioeventfd for virtqueue kick.  This requires a host
+     * that supports eventfd(2) and an I/O thread; since eventfd does not
+     * support SIGIO it cannot interrupt the vcpu.
+     *
+     * Older kernels have a 6 device limit on the KVM io bus.  Find out so we
      * can avoid creating too many ioeventfds.
      */
-#ifdef CONFIG_EVENTFD
+#if defined(CONFIG_EVENTFD) && defined(CONFIG_IOTHREAD)
     int ioeventfds[7];
     int i, ret = 0;
     for (i = 0; i < ARRAY_SIZE(ioeventfds); i++) {
-- 
1.7.2.3

[Qemu-devel] Re: [PATCH v2] virtio-pci: Disable virtio-ioeventfd when !CONFIG_IOTHREAD

[Michael S. Tsirkin]

On Tue, Jan 25, 2011 at 02:53:31PM +0000, Stefan Hajnoczi wrote:
> It is not possible to use virtio-ioeventfd when building without an I/O
> thread.  We rely on a signal to kick us out of vcpu execution.  Timers
> and AIO use SIGALRM and SIGUSR2 respectively.  Unfortunately eventfd
> does not support O_ASYNC (SIGIO) so eventfd cannot be used in a signal
> driven manner.
> 
> Signed-off-by: Stefan Hajnoczi <stefanha@linux.vnet.ibm.com>
> ---
>  kvm-all.c |    8 ++++++--
>  1 files changed, 6 insertions(+), 2 deletions(-)
> 
> diff --git a/kvm-all.c b/kvm-all.c
> index 255b6fa..881ea32 100644
> --- a/kvm-all.c
> +++ b/kvm-all.c
> @@ -449,10 +449,14 @@ int kvm_check_extension(KVMState *s, unsigned int extension)
>  
>  static int kvm_check_many_ioeventfds(void)
>  {
> -    /* Older kernels have a 6 device limit on the KVM io bus.  Find out so we
> +    /* Userspace can use ioeventfd for virtqueue kick.  This requires a host

virtqueue kick -> io notifications?

> +     * that supports eventfd(2) and an I/O thread; since eventfd does not
> +     * support SIGIO it cannot interrupt the vcpu.
> +     *
> +     * Older kernels have a 6 device limit on the KVM io bus.  Find out so we
>       * can avoid creating too many ioeventfds.
>       */
> -#ifdef CONFIG_EVENTFD
> +#if defined(CONFIG_EVENTFD) && defined(CONFIG_IOTHREAD)
>      int ioeventfds[7];
>      int i, ret = 0;
>      for (i = 0; i < ARRAY_SIZE(ioeventfds); i++) {
> -- 
> 1.7.2.3

Re: [Qemu-devel] Re: [PATCH v2] virtio-pci: Disable virtio-ioeventfd when !CONFIG_IOTHREAD

[Stefan Hajnoczi]

On Tue, Jan 25, 2011 at 2:56 PM, Michael S. Tsirkin <mst@redhat.com> wrote:
> On Tue, Jan 25, 2011 at 02:53:31PM +0000, Stefan Hajnoczi wrote:
>> It is not possible to use virtio-ioeventfd when building without an I/O
>> thread.  We rely on a signal to kick us out of vcpu execution.  Timers
>> and AIO use SIGALRM and SIGUSR2 respectively.  Unfortunately eventfd
>> does not support O_ASYNC (SIGIO) so eventfd cannot be used in a signal
>> driven manner.
>>
>> Signed-off-by: Stefan Hajnoczi <stefanha@linux.vnet.ibm.com>
>> ---
>>  kvm-all.c |    8 ++++++--
>>  1 files changed, 6 insertions(+), 2 deletions(-)
>>
>> diff --git a/kvm-all.c b/kvm-all.c
>> index 255b6fa..881ea32 100644
>> --- a/kvm-all.c
>> +++ b/kvm-all.c
>> @@ -449,10 +449,14 @@ int kvm_check_extension(KVMState *s, unsigned int extension)
>>
>>  static int kvm_check_many_ioeventfds(void)
>>  {
>> -    /* Older kernels have a 6 device limit on the KVM io bus.  Find out so we
>> +    /* Userspace can use ioeventfd for virtqueue kick.  This requires a host
>
> virtqueue kick -> io notifications?

Fixed in v3.

Stefan