* [PATCH] drm/amd: Adjust ASPM support quirk to cover more Intel hosts
@ 2026-04-19 4:16 Mario Limonciello
2026-04-19 20:28 ` Timur Kristóf
2026-04-20 1:32 ` Wang, Yang(Kevin)
0 siblings, 2 replies; 4+ messages in thread
From: Mario Limonciello @ 2026-04-19 4:16 UTC (permalink / raw)
To: amd-gfx; +Cc: Mario Limonciello
Some of the same issues identified in commit c770ef19673fb
("drm/amd/amdgpu: disable ASPM in some situations") also affect
Tiger Lake systems with GFX11 connected over USB4. Widen the net
to also match these hosts.
Fixes: d9b3a066dfcd3 ("drm/amd: Exclude dGPUs in eGPU enclosures from DPM quirks")
Closes: https://gitlab.freedesktop.org/drm/amd/-/work_items/5145
Signed-off-by: Mario Limonciello <mario.limonciello@amd.com>
---
drivers/gpu/drm/amd/amdgpu/amdgpu_device.c | 9 +++------
1 file changed, 3 insertions(+), 6 deletions(-)
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
index bc7e96b58d3fc..b11c4b5fa8fcd 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
@@ -1334,18 +1334,15 @@ static bool amdgpu_device_aspm_support_quirk(struct amdgpu_device *adev)
#if IS_ENABLED(CONFIG_X86)
struct cpuinfo_x86 *c = &cpu_data(0);
- if (!(amdgpu_ip_version(adev, GC_HWIP, 0) == IP_VERSION(12, 0, 0) ||
- amdgpu_ip_version(adev, GC_HWIP, 0) == IP_VERSION(12, 0, 1)))
- return false;
-
- if (c->x86 == 6 &&
- adev->pm.pcie_gen_mask & CAIL_PCIE_LINK_SPEED_SUPPORT_GEN5) {
+ if (c->x86_vendor == X86_VENDOR_INTEL) {
switch (c->x86_model) {
case VFM_MODEL(INTEL_ALDERLAKE):
case VFM_MODEL(INTEL_ALDERLAKE_L):
case VFM_MODEL(INTEL_RAPTORLAKE):
case VFM_MODEL(INTEL_RAPTORLAKE_P):
case VFM_MODEL(INTEL_RAPTORLAKE_S):
+ case VFM_MODEL(INTEL_TIGERLAKE):
+ case VFM_MODEL(INTEL_TIGERLAKE_L):
return true;
default:
return false;
--
2.43.0
^ permalink raw reply related [flat|nested] 4+ messages in thread
* Re: [PATCH] drm/amd: Adjust ASPM support quirk to cover more Intel hosts
2026-04-19 4:16 [PATCH] drm/amd: Adjust ASPM support quirk to cover more Intel hosts Mario Limonciello
@ 2026-04-19 20:28 ` Timur Kristóf
2026-04-20 2:56 ` Mario Limonciello
2026-04-20 1:32 ` Wang, Yang(Kevin)
1 sibling, 1 reply; 4+ messages in thread
From: Timur Kristóf @ 2026-04-19 20:28 UTC (permalink / raw)
To: amd-gfx; +Cc: Mario Limonciello, Mario Limonciello
On Sunday, April 19, 2026 6:16:52 AM Central European Summer Time Mario
Limonciello wrote:
> Some of the same issues identified in commit c770ef19673fb
> ("drm/amd/amdgpu: disable ASPM in some situations") also affect
> Tiger Lake systems with GFX11 connected over USB4. Widen the net
> to also match these hosts.
Wouldn't it be better to check that the connection is through USB4 rather than
entirely disable ASPM accross all these families of CPUs?
>
> Fixes: d9b3a066dfcd3 ("drm/amd: Exclude dGPUs in eGPU enclosures from DPM
> quirks") Closes: https://gitlab.freedesktop.org/drm/amd/-/work_items/5145
> Signed-off-by: Mario Limonciello <mario.limonciello@amd.com>
> ---
> drivers/gpu/drm/amd/amdgpu/amdgpu_device.c | 9 +++------
> 1 file changed, 3 insertions(+), 6 deletions(-)
>
> diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
> b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c index
> bc7e96b58d3fc..b11c4b5fa8fcd 100644
> --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
> +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
> @@ -1334,18 +1334,15 @@ static bool amdgpu_device_aspm_support_quirk(struct
> amdgpu_device *adev) #if IS_ENABLED(CONFIG_X86)
> struct cpuinfo_x86 *c = &cpu_data(0);
>
> - if (!(amdgpu_ip_version(adev, GC_HWIP, 0) == IP_VERSION(12, 0, 0)
||
> - amdgpu_ip_version(adev, GC_HWIP, 0) == IP_VERSION(12,
0, 1)))
> - return false;
> -
> - if (c->x86 == 6 &&
> - adev->pm.pcie_gen_mask &
CAIL_PCIE_LINK_SPEED_SUPPORT_GEN5) {
> + if (c->x86_vendor == X86_VENDOR_INTEL) {
> switch (c->x86_model) {
> case VFM_MODEL(INTEL_ALDERLAKE):
> case VFM_MODEL(INTEL_ALDERLAKE_L):
> case VFM_MODEL(INTEL_RAPTORLAKE):
> case VFM_MODEL(INTEL_RAPTORLAKE_P):
> case VFM_MODEL(INTEL_RAPTORLAKE_S):
> + case VFM_MODEL(INTEL_TIGERLAKE):
> + case VFM_MODEL(INTEL_TIGERLAKE_L):
> return true;
> default:
> return false;
^ permalink raw reply [flat|nested] 4+ messages in thread
* RE: [PATCH] drm/amd: Adjust ASPM support quirk to cover more Intel hosts
2026-04-19 4:16 [PATCH] drm/amd: Adjust ASPM support quirk to cover more Intel hosts Mario Limonciello
2026-04-19 20:28 ` Timur Kristóf
@ 2026-04-20 1:32 ` Wang, Yang(Kevin)
1 sibling, 0 replies; 4+ messages in thread
From: Wang, Yang(Kevin) @ 2026-04-20 1:32 UTC (permalink / raw)
To: Limonciello, Mario, amd-gfx@lists.freedesktop.org,
Deucher, Alexander, Feng, Kenneth
Cc: Limonciello, Mario
[AMD Official Use Only - AMD Internal Distribution Only]
Cc @Deucher, Alexander @Feng, Kenneth,
Reviewed-by: Yang Wang <kevinyang.wang@amd.com>
Best Regards,
Kevin
-----Original Message-----
From: amd-gfx <amd-gfx-bounces@lists.freedesktop.org> On Behalf Of Mario Limonciello
Sent: Sunday, April 19, 2026 12:17
To: amd-gfx@lists.freedesktop.org
Cc: Limonciello, Mario <Mario.Limonciello@amd.com>
Subject: [PATCH] drm/amd: Adjust ASPM support quirk to cover more Intel hosts
Some of the same issues identified in commit c770ef19673fb
("drm/amd/amdgpu: disable ASPM in some situations") also affect Tiger Lake systems with GFX11 connected over USB4. Widen the net to also match these hosts.
Fixes: d9b3a066dfcd3 ("drm/amd: Exclude dGPUs in eGPU enclosures from DPM quirks")
Closes: https://gitlab.freedesktop.org/drm/amd/-/work_items/5145
Signed-off-by: Mario Limonciello <mario.limonciello@amd.com>
---
drivers/gpu/drm/amd/amdgpu/amdgpu_device.c | 9 +++------
1 file changed, 3 insertions(+), 6 deletions(-)
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
index bc7e96b58d3fc..b11c4b5fa8fcd 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
@@ -1334,18 +1334,15 @@ static bool amdgpu_device_aspm_support_quirk(struct amdgpu_device *adev) #if IS_ENABLED(CONFIG_X86)
struct cpuinfo_x86 *c = &cpu_data(0);
- if (!(amdgpu_ip_version(adev, GC_HWIP, 0) == IP_VERSION(12, 0, 0) ||
- amdgpu_ip_version(adev, GC_HWIP, 0) == IP_VERSION(12, 0, 1)))
- return false;
-
- if (c->x86 == 6 &&
- adev->pm.pcie_gen_mask & CAIL_PCIE_LINK_SPEED_SUPPORT_GEN5) {
+ if (c->x86_vendor == X86_VENDOR_INTEL) {
switch (c->x86_model) {
case VFM_MODEL(INTEL_ALDERLAKE):
case VFM_MODEL(INTEL_ALDERLAKE_L):
case VFM_MODEL(INTEL_RAPTORLAKE):
case VFM_MODEL(INTEL_RAPTORLAKE_P):
case VFM_MODEL(INTEL_RAPTORLAKE_S):
+ case VFM_MODEL(INTEL_TIGERLAKE):
+ case VFM_MODEL(INTEL_TIGERLAKE_L):
return true;
default:
return false;
--
2.43.0
^ permalink raw reply related [flat|nested] 4+ messages in thread
* Re: [PATCH] drm/amd: Adjust ASPM support quirk to cover more Intel hosts
2026-04-19 20:28 ` Timur Kristóf
@ 2026-04-20 2:56 ` Mario Limonciello
0 siblings, 0 replies; 4+ messages in thread
From: Mario Limonciello @ 2026-04-20 2:56 UTC (permalink / raw)
To: Timur Kristóf, amd-gfx
On 4/19/26 15:28, Timur Kristóf wrote:
> On Sunday, April 19, 2026 6:16:52 AM Central European Summer Time Mario
> Limonciello wrote:
>> Some of the same issues identified in commit c770ef19673fb
>> ("drm/amd/amdgpu: disable ASPM in some situations") also affect
>> Tiger Lake systems with GFX11 connected over USB4. Widen the net
>> to also match these hosts.
>
> Wouldn't it be better to check that the connection is through USB4 rather than
> entirely disable ASPM accross all these families of CPUs?
Historically we've had a lot of problems with Intel hosts where the root
ports can't handle the link changing.
I don't currently have any evidence to say the root ports used for PCIe
tunneling are any different than root ports for tradiionally connected
device.
So I am leaning on being more cautious here. We can peel it back later
if that turned out to be wrong.
>
>>
>> Fixes: d9b3a066dfcd3 ("drm/amd: Exclude dGPUs in eGPU enclosures from DPM
>> quirks") Closes: https://gitlab.freedesktop.org/drm/amd/-/work_items/5145
>> Signed-off-by: Mario Limonciello <mario.limonciello@amd.com>
>> ---
>> drivers/gpu/drm/amd/amdgpu/amdgpu_device.c | 9 +++------
>> 1 file changed, 3 insertions(+), 6 deletions(-)
>>
>> diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
>> b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c index
>> bc7e96b58d3fc..b11c4b5fa8fcd 100644
>> --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
>> +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
>> @@ -1334,18 +1334,15 @@ static bool amdgpu_device_aspm_support_quirk(struct
>> amdgpu_device *adev) #if IS_ENABLED(CONFIG_X86)
>> struct cpuinfo_x86 *c = &cpu_data(0);
>>
>> - if (!(amdgpu_ip_version(adev, GC_HWIP, 0) == IP_VERSION(12, 0, 0)
> ||
>> - amdgpu_ip_version(adev, GC_HWIP, 0) == IP_VERSION(12,
> 0, 1)))
>> - return false;
>> -
>> - if (c->x86 == 6 &&
>> - adev->pm.pcie_gen_mask &
> CAIL_PCIE_LINK_SPEED_SUPPORT_GEN5) {
>> + if (c->x86_vendor == X86_VENDOR_INTEL) {
>> switch (c->x86_model) {
>> case VFM_MODEL(INTEL_ALDERLAKE):
>> case VFM_MODEL(INTEL_ALDERLAKE_L):
>> case VFM_MODEL(INTEL_RAPTORLAKE):
>> case VFM_MODEL(INTEL_RAPTORLAKE_P):
>> case VFM_MODEL(INTEL_RAPTORLAKE_S):
>> + case VFM_MODEL(INTEL_TIGERLAKE):
>> + case VFM_MODEL(INTEL_TIGERLAKE_L):
>> return true;
>> default:
>> return false;
>
>
>
>
^ permalink raw reply [flat|nested] 4+ messages in thread
end of thread, other threads:[~2026-04-20 2:56 UTC | newest]
Thread overview: 4+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2026-04-19 4:16 [PATCH] drm/amd: Adjust ASPM support quirk to cover more Intel hosts Mario Limonciello
2026-04-19 20:28 ` Timur Kristóf
2026-04-20 2:56 ` Mario Limonciello
2026-04-20 1:32 ` Wang, Yang(Kevin)
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.