All of lore.kernel.org
 help / color / mirror / Atom feed
* [PATCH 1/2] drm/amdgpu: improve MSI-X handling
@ 2019-10-03 14:13 Alex Deucher
       [not found] ` <20191003141326.7600-1-alexander.deucher-5C7GfCeVMHo@public.gmane.org>
  0 siblings, 1 reply; 4+ messages in thread
From: Alex Deucher @ 2019-10-03 14:13 UTC (permalink / raw)
  To: amd-gfx-PD4FTy7X32lNgt0PjOBp9y5qC8QIuHrW; +Cc: Alex Deucher

Check the number of supported vectors and fall back to MSI if
we return or error or 0 MSI-X vectors.

Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_irq.c | 14 +++++++++++---
 1 file changed, 11 insertions(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_irq.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_irq.c
index 50771b2757dc..98aa28edba6a 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_irq.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_irq.c
@@ -245,11 +245,19 @@ int amdgpu_irq_init(struct amdgpu_device *adev)
 	adev->irq.msi_enabled = false;
 
 	if (amdgpu_msi_ok(adev)) {
-		int nvec = pci_alloc_irq_vectors(adev->pdev, 1, pci_msix_vec_count(adev->pdev),
-					PCI_IRQ_MSI | PCI_IRQ_MSIX);
+		unsigned int flags;
+		int nvec = pci_msix_vec_count(adev->pdev);
+
+		if (nvec <= 0) {
+			flags = PCI_IRQ_MSI;
+			nvec = 1;
+		} else {
+			flags = PCI_IRQ_MSI | PCI_IRQ_MSIX;
+		}
+		nvec = pci_alloc_irq_vectors(adev->pdev, 1, nvec, flags);
 		if (nvec > 0) {
 			adev->irq.msi_enabled = true;
-			dev_dbg(adev->dev, "amdgpu: using MSI.\n");
+			dev_dbg(adev->dev, "amdgpu: using MSI/MSI-X.\n");
 		}
 	}
 
-- 
2.20.1

_______________________________________________
amd-gfx mailing list
amd-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/amd-gfx

^ permalink raw reply related	[flat|nested] 4+ messages in thread

* [PATCH 2/2] drm/amdgpu: disable MSI-X on APUs
       [not found] ` <20191003141326.7600-1-alexander.deucher-5C7GfCeVMHo@public.gmane.org>
@ 2019-10-03 14:13   ` Alex Deucher
       [not found]     ` <20191003141326.7600-2-alexander.deucher-5C7GfCeVMHo@public.gmane.org>
  2019-10-03 15:07   ` [PATCH 1/2] drm/amdgpu: improve MSI-X handling Huang, Ray
  1 sibling, 1 reply; 4+ messages in thread
From: Alex Deucher @ 2019-10-03 14:13 UTC (permalink / raw)
  To: amd-gfx-PD4FTy7X32lNgt0PjOBp9y5qC8QIuHrW; +Cc: Alex Deucher, Tom St Denis

Raven claims to support them, but seems to have problems.  Stick
with MSIs for now on APUs.

Tested-by: Tom St Denis <tom.stdenis@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_irq.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_irq.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_irq.c
index 98aa28edba6a..8f2236bd7d0e 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_irq.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_irq.c
@@ -248,7 +248,8 @@ int amdgpu_irq_init(struct amdgpu_device *adev)
 		unsigned int flags;
 		int nvec = pci_msix_vec_count(adev->pdev);
 
-		if (nvec <= 0) {
+		/* Raven claims to support MSI-X, but seems to have problems */
+		if ((nvec <= 0) || (adev->flags & AMD_IS_APU)) {
 			flags = PCI_IRQ_MSI;
 			nvec = 1;
 		} else {
-- 
2.20.1

_______________________________________________
amd-gfx mailing list
amd-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/amd-gfx

^ permalink raw reply related	[flat|nested] 4+ messages in thread

* Re: [PATCH 2/2] drm/amdgpu: disable MSI-X on APUs
       [not found]     ` <20191003141326.7600-2-alexander.deucher-5C7GfCeVMHo@public.gmane.org>
@ 2019-10-03 14:27       ` Liu, Shaoyun
  0 siblings, 0 replies; 4+ messages in thread
From: Liu, Shaoyun @ 2019-10-03 14:27 UTC (permalink / raw)
  To: Alex Deucher, amd-gfx-PD4FTy7X32lNgt0PjOBp9y5qC8QIuHrW
  Cc: Deucher, Alexander, StDenis, Tom

Thanks Alex and  Tom to catch and  revolve the issue .  I didn't do 
enough test on original test.

We don't need to enable msix  on APU .   The serials is reviewed by 
shaoyun.liu <shaoyun.liu@amd.com>


Regards

shaoyun.liu



On 2019-10-03 10:13 a.m., Alex Deucher wrote:
> Raven claims to support them, but seems to have problems.  Stick
> with MSIs for now on APUs.
>
> Tested-by: Tom St Denis <tom.stdenis@amd.com>
> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
> ---
>   drivers/gpu/drm/amd/amdgpu/amdgpu_irq.c | 3 ++-
>   1 file changed, 2 insertions(+), 1 deletion(-)
>
> diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_irq.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_irq.c
> index 98aa28edba6a..8f2236bd7d0e 100644
> --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_irq.c
> +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_irq.c
> @@ -248,7 +248,8 @@ int amdgpu_irq_init(struct amdgpu_device *adev)
>   		unsigned int flags;
>   		int nvec = pci_msix_vec_count(adev->pdev);
>   
> -		if (nvec <= 0) {
> +		/* Raven claims to support MSI-X, but seems to have problems */
> +		if ((nvec <= 0) || (adev->flags & AMD_IS_APU)) {
>   			flags = PCI_IRQ_MSI;
>   			nvec = 1;
>   		} else {
_______________________________________________
amd-gfx mailing list
amd-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/amd-gfx

^ permalink raw reply	[flat|nested] 4+ messages in thread

* RE: [PATCH 1/2] drm/amdgpu: improve MSI-X handling
       [not found] ` <20191003141326.7600-1-alexander.deucher-5C7GfCeVMHo@public.gmane.org>
  2019-10-03 14:13   ` [PATCH 2/2] drm/amdgpu: disable MSI-X on APUs Alex Deucher
@ 2019-10-03 15:07   ` Huang, Ray
  1 sibling, 0 replies; 4+ messages in thread
From: Huang, Ray @ 2019-10-03 15:07 UTC (permalink / raw)
  To: Alex Deucher, amd-gfx-PD4FTy7X32lNgt0PjOBp9y5qC8QIuHrW; +Cc: Deucher, Alexander

Series are Reviewed-by: Huang Rui <ray.huang@amd.com>

-----Original Message-----
From: amd-gfx <amd-gfx-bounces@lists.freedesktop.org> On Behalf Of Alex Deucher
Sent: Thursday, October 3, 2019 10:13 AM
To: amd-gfx@lists.freedesktop.org
Cc: Deucher, Alexander <Alexander.Deucher@amd.com>
Subject: [PATCH 1/2] drm/amdgpu: improve MSI-X handling

Check the number of supported vectors and fall back to MSI if we return or error or 0 MSI-X vectors.

Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_irq.c | 14 +++++++++++---
 1 file changed, 11 insertions(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_irq.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_irq.c
index 50771b2757dc..98aa28edba6a 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_irq.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_irq.c
@@ -245,11 +245,19 @@ int amdgpu_irq_init(struct amdgpu_device *adev)
 	adev->irq.msi_enabled = false;
 
 	if (amdgpu_msi_ok(adev)) {
-		int nvec = pci_alloc_irq_vectors(adev->pdev, 1, pci_msix_vec_count(adev->pdev),
-					PCI_IRQ_MSI | PCI_IRQ_MSIX);
+		unsigned int flags;
+		int nvec = pci_msix_vec_count(adev->pdev);
+
+		if (nvec <= 0) {
+			flags = PCI_IRQ_MSI;
+			nvec = 1;
+		} else {
+			flags = PCI_IRQ_MSI | PCI_IRQ_MSIX;
+		}
+		nvec = pci_alloc_irq_vectors(adev->pdev, 1, nvec, flags);
 		if (nvec > 0) {
 			adev->irq.msi_enabled = true;
-			dev_dbg(adev->dev, "amdgpu: using MSI.\n");
+			dev_dbg(adev->dev, "amdgpu: using MSI/MSI-X.\n");
 		}
 	}
 
--
2.20.1

_______________________________________________
amd-gfx mailing list
amd-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/amd-gfx
_______________________________________________
amd-gfx mailing list
amd-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/amd-gfx

^ permalink raw reply related	[flat|nested] 4+ messages in thread

end of thread, other threads:[~2019-10-03 15:07 UTC | newest]

Thread overview: 4+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2019-10-03 14:13 [PATCH 1/2] drm/amdgpu: improve MSI-X handling Alex Deucher
     [not found] ` <20191003141326.7600-1-alexander.deucher-5C7GfCeVMHo@public.gmane.org>
2019-10-03 14:13   ` [PATCH 2/2] drm/amdgpu: disable MSI-X on APUs Alex Deucher
     [not found]     ` <20191003141326.7600-2-alexander.deucher-5C7GfCeVMHo@public.gmane.org>
2019-10-03 14:27       ` Liu, Shaoyun
2019-10-03 15:07   ` [PATCH 1/2] drm/amdgpu: improve MSI-X handling Huang, Ray

This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.