diff mbox series

[RFC,QEMU,v3,1/1] xen: Use gsi instead of irq for mapping pirq

Message ID 20231210165240.1551450-2-Jiqian.Chen@amd.com (mailing list archive)
State Superseded
Headers show
Series Support device passthrough when dom0 is PVH on Xen | expand

Commit Message

Chen, Jiqian Dec. 10, 2023, 4:52 p.m. UTC
In PVH dom0, it uses the linux local interrupt mechanism,
when it allocs irq for a gsi, it is dynamic, and follow
the principle of applying first, distributing first. And
the irq number is alloced from small to large, but the
applying gsi number is not, may gsi 38 comes before gsi
28, that causes the irq number is not equal with the gsi
number. And when passthrough a device, qemu wants to use
gsi to map pirq, xen_pt_realize->xc_physdev_map_pirq, but
the gsi number is got from file
/sys/bus/pci/devices/<sbdf>/irq in current code, so it
will fail when mapping.

Use real gsi number read from gsi sysfs.

Co-developed-by: Huang Rui <ray.huang@amd.com>
Signed-off-by: Jiqian Chen <Jiqian.Chen@amd.com>
---
 hw/xen/xen-host-pci-device.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

Comments

Roger Pau Monné Dec. 11, 2023, 3:33 p.m. UTC | #1
On Mon, Dec 11, 2023 at 12:52:40AM +0800, Jiqian Chen wrote:
> In PVH dom0, it uses the linux local interrupt mechanism,
> when it allocs irq for a gsi, it is dynamic, and follow
> the principle of applying first, distributing first. And
> the irq number is alloced from small to large, but the
> applying gsi number is not, may gsi 38 comes before gsi
> 28, that causes the irq number is not equal with the gsi
> number. And when passthrough a device, qemu wants to use
> gsi to map pirq, xen_pt_realize->xc_physdev_map_pirq, but
> the gsi number is got from file
> /sys/bus/pci/devices/<sbdf>/irq in current code, so it
> will fail when mapping.
> 
> Use real gsi number read from gsi sysfs.
> 
> Co-developed-by: Huang Rui <ray.huang@amd.com>
> Signed-off-by: Jiqian Chen <Jiqian.Chen@amd.com>
> ---
>  hw/xen/xen-host-pci-device.c | 2 +-
>  1 file changed, 1 insertion(+), 1 deletion(-)
> 
> diff --git a/hw/xen/xen-host-pci-device.c b/hw/xen/xen-host-pci-device.c
> index 8c6e9a1716..e270ac2631 100644
> --- a/hw/xen/xen-host-pci-device.c
> +++ b/hw/xen/xen-host-pci-device.c
> @@ -364,7 +364,7 @@ void xen_host_pci_device_get(XenHostPCIDevice *d, uint16_t domain,
>      }
>      d->device_id = v;
>  
> -    xen_host_pci_get_dec_value(d, "irq", &v, errp);
> +    xen_host_pci_get_dec_value(d, "gsi", &v, errp);

Don't you need to fallthrough to use the irq number on failure?
Otherwise passthrough won't work on older Linux versions that don't
expose the gsi node.

Thanks, Roger.
Chen, Jiqian Dec. 12, 2023, 6:37 a.m. UTC | #2
On 2023/12/11 23:33, Roger Pau Monné wrote:
> On Mon, Dec 11, 2023 at 12:52:40AM +0800, Jiqian Chen wrote:
>> In PVH dom0, it uses the linux local interrupt mechanism,
>> when it allocs irq for a gsi, it is dynamic, and follow
>> the principle of applying first, distributing first. And
>> the irq number is alloced from small to large, but the
>> applying gsi number is not, may gsi 38 comes before gsi
>> 28, that causes the irq number is not equal with the gsi
>> number. And when passthrough a device, qemu wants to use
>> gsi to map pirq, xen_pt_realize->xc_physdev_map_pirq, but
>> the gsi number is got from file
>> /sys/bus/pci/devices/<sbdf>/irq in current code, so it
>> will fail when mapping.
>>
>> Use real gsi number read from gsi sysfs.
>>
>> Co-developed-by: Huang Rui <ray.huang@amd.com>
>> Signed-off-by: Jiqian Chen <Jiqian.Chen@amd.com>
>> ---
>>  hw/xen/xen-host-pci-device.c | 2 +-
>>  1 file changed, 1 insertion(+), 1 deletion(-)
>>
>> diff --git a/hw/xen/xen-host-pci-device.c b/hw/xen/xen-host-pci-device.c
>> index 8c6e9a1716..e270ac2631 100644
>> --- a/hw/xen/xen-host-pci-device.c
>> +++ b/hw/xen/xen-host-pci-device.c
>> @@ -364,7 +364,7 @@ void xen_host_pci_device_get(XenHostPCIDevice *d, uint16_t domain,
>>      }
>>      d->device_id = v;
>>  
>> -    xen_host_pci_get_dec_value(d, "irq", &v, errp);
>> +    xen_host_pci_get_dec_value(d, "gsi", &v, errp);
> 
> Don't you need to fallthrough to use the irq number on failure?
> Otherwise passthrough won't work on older Linux versions that don't
> expose the gsi node.
You are right, I will use the irq if there isn't a gsi sysfs, in next version. Thank you.

> 
> Thanks, Roger.
diff mbox series

Patch

diff --git a/hw/xen/xen-host-pci-device.c b/hw/xen/xen-host-pci-device.c
index 8c6e9a1716..e270ac2631 100644
--- a/hw/xen/xen-host-pci-device.c
+++ b/hw/xen/xen-host-pci-device.c
@@ -364,7 +364,7 @@  void xen_host_pci_device_get(XenHostPCIDevice *d, uint16_t domain,
     }
     d->device_id = v;
 
-    xen_host_pci_get_dec_value(d, "irq", &v, errp);
+    xen_host_pci_get_dec_value(d, "gsi", &v, errp);
     if (*errp) {
         goto error;
     }