RE: [PATCHv5] drm/amdgpu: vi: disable ASPM on Intel Alder Lake based systems
From: Limonciello, Mario
Date: Mon May 02 2022 - 22:16:09 EST
[Public]
> -----Original Message-----
> From: Alex Deucher <alexdeucher@xxxxxxxxx>
> Sent: Friday, April 29, 2022 11:14
> To: Gong, Richard <Richard.Gong@xxxxxxx>
> Cc: Deucher, Alexander <Alexander.Deucher@xxxxxxx>; Koenig, Christian
> <Christian.Koenig@xxxxxxx>; Pan, Xinhui <Xinhui.Pan@xxxxxxx>; Dave
> Airlie <airlied@xxxxxxxx>; Daniel Vetter <daniel@xxxxxxxx>; amd-gfx list <amd-
> gfx@xxxxxxxxxxxxxxxxxxxxx>; kernel test robot <lkp@xxxxxxxxx>; LKML <linux-
> kernel@xxxxxxxxxxxxxxx>; Maling list - DRI developers <dri-
> devel@xxxxxxxxxxxxxxxxxxxxx>; Limonciello, Mario
> <Mario.Limonciello@xxxxxxx>
> Subject: Re: [PATCHv5] drm/amdgpu: vi: disable ASPM on Intel Alder Lake
> based systems
>
> Reviewed-by: Alex Deucher <alexander.deucher@xxxxxxx>
>
> On Fri, Apr 29, 2022 at 12:08 PM Richard Gong <richard.gong@xxxxxxx>
> wrote:
> >
> > Active State Power Management (ASPM) feature is enabled since kernel
> 5.14.
> > There are some AMD Volcanic Islands (VI) GFX cards, such as the WX3200
> and
> > RX640, that do not work with ASPM-enabled Intel Alder Lake based
> systems.
> > Using these GFX cards as video/display output, Intel Alder Lake based
> > systems will freeze after suspend/resume.
> >
> > The issue was originally reported on one system (Dell Precision 3660 with
> > BIOS version 0.14.81), but was later confirmed to affect at least 4
> > pre-production Alder Lake based systems.
> >
> > Add an extra check to disable ASPM on Intel Alder Lake based systems with
> > the problematic AMD Volcanic Islands GFX cards.
> >
> > Fixes: 0064b0ce85bb ("drm/amd/pm: enable ASPM by default")
> > Link:
> https://nam11.safelinks.protection.outlook.com/?url=https%3A%2F%2Fgitla
> b.freedesktop.org%2Fdrm%2Famd%2F-
> %2Fissues%2F1885&data=05%7C01%7Cmario.limonciello%40amd.com%
> 7C1fdb6c767a4a4b3f572c08da29fb3f1a%7C3dd8961fe4884e608e11a82d994e1
> 83d%7C0%7C0%7C637868456326825256%7CUnknown%7CTWFpbGZsb3d8eyJ
> WIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%
> 7C3000%7C%7C%7C&sdata=LtV9WqUzB032KFmH2g%2F2BMtX2R6DyfM
> KqxCF1e2rHHg%3D&reserved=0
> > Reported-by: kernel test robot <lkp@xxxxxxxxx>
You should drop this "Reported-by:". That makes more sense when it's a patch
that is already queued up and committee. The bot just caught an intermediary
revision that wasn't committed anywhere, and it doesn't make sense here.
I don't think you need to resend out a new version for review to change commit
message for that, this can just be fixed up when committing the change.
Reviewed-by: Mario Limonciello <mario.limonciello@xxxxxxx>
> > Signed-off-by: Richard Gong <richard.gong@xxxxxxx>
> > ---
> > v5: added vi to commit header and updated commit message
> > rolled back guard with the preprocessor as did in v2 to correct build
> > error on non-x86 systems
> > v4: s/CONFIG_X86_64/CONFIG_X86
> > enhanced check logic
> > v3: s/intel_core_aspm_chk/aspm_support_quirk_check
> > correct build error with W=1 option
> > v2: correct commit description
> > move the check from chip family to problematic platform
> > ---
> > drivers/gpu/drm/amd/amdgpu/vi.c | 17 ++++++++++++++++-
> > 1 file changed, 16 insertions(+), 1 deletion(-)
> >
> > diff --git a/drivers/gpu/drm/amd/amdgpu/vi.c
> b/drivers/gpu/drm/amd/amdgpu/vi.c
> > index 039b90cdc3bc..45f0188c4273 100644
> > --- a/drivers/gpu/drm/amd/amdgpu/vi.c
> > +++ b/drivers/gpu/drm/amd/amdgpu/vi.c
> > @@ -81,6 +81,10 @@
> > #include "mxgpu_vi.h"
> > #include "amdgpu_dm.h"
> >
> > +#if IS_ENABLED(CONFIG_X86)
> > +#include <asm/intel-family.h>
> > +#endif
> > +
> > #define ixPCIE_LC_L1_PM_SUBSTATE 0x100100C6
> > #define
> PCIE_LC_L1_PM_SUBSTATE__LC_L1_SUBSTATES_OVERRIDE_EN_MASK
> 0x00000001L
> > #define
> PCIE_LC_L1_PM_SUBSTATE__LC_PCI_PM_L1_2_OVERRIDE_MASK
> 0x00000002L
> > @@ -1134,13 +1138,24 @@ static void vi_enable_aspm(struct
> amdgpu_device *adev)
> > WREG32_PCIE(ixPCIE_LC_CNTL, data);
> > }
> >
> > +static bool aspm_support_quirk_check(void)
> > +{
> > +#if IS_ENABLED(CONFIG_X86)
> > + struct cpuinfo_x86 *c = &cpu_data(0);
> > +
> > + return !(c->x86 == 6 && c->x86_model == INTEL_FAM6_ALDERLAKE);
> > +#else
> > + return true;
> > +#endif
> > +}
> > +
> > static void vi_program_aspm(struct amdgpu_device *adev)
> > {
> > u32 data, data1, orig;
> > bool bL1SS = false;
> > bool bClkReqSupport = true;
> >
> > - if (!amdgpu_device_should_use_aspm(adev))
> > + if (!amdgpu_device_should_use_aspm(adev) ||
> !aspm_support_quirk_check())
> > return;
> >
> > if (adev->flags & AMD_IS_APU ||
> > --
> > 2.25.1
> >