Re: [PATCH v3 5/5] Revert "vfio/pci: Disable INTx fast path if using split irqchip"

2020-03-17 Thread Alex Williamson
On Tue, 17 Mar 2020 15:50:42 -0400
Peter Xu  wrote:

> With the resamplefd list introduced, we can savely enable VFIO INTx
> fast path again with split irqchip so it can still be faster than the
> complete slow path.
> 
> Reviewed-by: Eric Auger 
> Signed-off-by: Peter Xu 
> ---
>  hw/vfio/pci.c | 12 
>  1 file changed, 12 deletions(-)

Reviewed-by: Alex Williamson 
Acked-by: Alex Williamson 

> diff --git a/hw/vfio/pci.c b/hw/vfio/pci.c
> index 09703362df..1c0aa27386 100644
> --- a/hw/vfio/pci.c
> +++ b/hw/vfio/pci.c
> @@ -124,18 +124,6 @@ static void vfio_intx_enable_kvm(VFIOPCIDevice *vdev, 
> Error **errp)
>  return;
>  }
>  
> -if (kvm_irqchip_is_split()) {
> -/*
> - * VFIO INTx is currently not working with split kernel
> - * irqchip for level triggered interrupts.  Go the slow path
> - * as long as split is enabled so we can be at least
> - * functional (even with poor performance).
> - *
> - * TODO: Remove this after all things fixed up.
> - */
> -return;
> -}
> -
>  /* Get to a known interrupt state */
>  qemu_set_fd_handler(irq_fd, NULL, NULL, vdev);
>  vfio_mask_single_irqindex(>vbasedev, VFIO_PCI_INTX_IRQ_INDEX);




[PATCH v3 5/5] Revert "vfio/pci: Disable INTx fast path if using split irqchip"

2020-03-17 Thread Peter Xu
With the resamplefd list introduced, we can savely enable VFIO INTx
fast path again with split irqchip so it can still be faster than the
complete slow path.

Reviewed-by: Eric Auger 
Signed-off-by: Peter Xu 
---
 hw/vfio/pci.c | 12 
 1 file changed, 12 deletions(-)

diff --git a/hw/vfio/pci.c b/hw/vfio/pci.c
index 09703362df..1c0aa27386 100644
--- a/hw/vfio/pci.c
+++ b/hw/vfio/pci.c
@@ -124,18 +124,6 @@ static void vfio_intx_enable_kvm(VFIOPCIDevice *vdev, 
Error **errp)
 return;
 }
 
-if (kvm_irqchip_is_split()) {
-/*
- * VFIO INTx is currently not working with split kernel
- * irqchip for level triggered interrupts.  Go the slow path
- * as long as split is enabled so we can be at least
- * functional (even with poor performance).
- *
- * TODO: Remove this after all things fixed up.
- */
-return;
-}
-
 /* Get to a known interrupt state */
 qemu_set_fd_handler(irq_fd, NULL, NULL, vdev);
 vfio_mask_single_irqindex(>vbasedev, VFIO_PCI_INTX_IRQ_INDEX);
-- 
2.24.1