Re: [PATCH] drm/amdgpu/sriov: Use VF-accessible register for gpu_clock_count

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



Let's just switch everyone over.  I think this should work for both bare metal and SR-IOV.  Fewer code pathes to maintain.  Feel free to do it as a follow up patch.

Thanks,

Alex



From: Zhao, Jiange <Jiange.Zhao@xxxxxxx>
Sent: Thursday, February 27, 2020 10:02 PM
To: Deucher, Alexander <Alexander.Deucher@xxxxxxx>; amd-gfx@xxxxxxxxxxxxxxxxxxxxx <amd-gfx@xxxxxxxxxxxxxxxxxxxxx>
Cc: Deng, Emily <Emily.Deng@xxxxxxx>; Liu, Monk <Monk.Liu@xxxxxxx>
Subject: Re: [PATCH] drm/amdgpu/sriov: Use VF-accessible register for gpu_clock_count
 
Hi,

I got feedback from Linux team and they simply don't want to change.

I believe that it would work for bare metal as well.

Jiange

From: Deucher, Alexander <Alexander.Deucher@xxxxxxx>
Sent: Thursday, February 27, 2020 10:23 PM
To: Zhao, Jiange <Jiange.Zhao@xxxxxxx>; amd-gfx@xxxxxxxxxxxxxxxxxxxxx <amd-gfx@xxxxxxxxxxxxxxxxxxxxx>
Cc: Deng, Emily <Emily.Deng@xxxxxxx>; Liu, Monk <Monk.Liu@xxxxxxx>
Subject: Re: [PATCH] drm/amdgpu/sriov: Use VF-accessible register for gpu_clock_count
 

[AMD Public Use]


Is there any reason to not just use this for bare metal as well?

Alex


From: amd-gfx <amd-gfx-bounces@xxxxxxxxxxxxxxxxxxxxx> on behalf of jianzh <Jiange.Zhao@xxxxxxx>
Sent: Thursday, February 27, 2020 6:21 AM
To: amd-gfx@xxxxxxxxxxxxxxxxxxxxx <amd-gfx@xxxxxxxxxxxxxxxxxxxxx>
Cc: Deng, Emily <Emily.Deng@xxxxxxx>; Zhao, Jiange <Jiange.Zhao@xxxxxxx>; Liu, Monk <Monk.Liu@xxxxxxx>
Subject: [PATCH] drm/amdgpu/sriov: Use VF-accessible register for gpu_clock_count
 
Navi12 VK CTS subtest timestamp.calibrated.dev_domain_test failed
because mmRLC_CAPTURE_GPU_CLOCK_COUNT register cannot be
written in VF due to security policy.

Solution: use a VF-accessible timestamp register pair
mmGOLDEN_TSC_COUNT_LOWER/UPPER for SRIOV case.

Signed-off-by: jianzh <Jiange.Zhao@xxxxxxx>
---
 drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c | 13 ++++++++++---
 1 file changed, 10 insertions(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c
index 44f00ec..8787a46 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c
@@ -35,6 +35,8 @@
 
 #include "gc/gc_10_1_0_offset.h"
 #include "gc/gc_10_1_0_sh_mask.h"
+#include "smuio/smuio_11_0_0_offset.h"
+#include "smuio/smuio_11_0_0_sh_mask.h"
 #include "navi10_enum.h"
 #include "hdp/hdp_5_0_0_offset.h"
 #include "ivsrcid/gfx/irqsrcs_gfx_10_1.h"
@@ -3920,9 +3922,14 @@ static uint64_t gfx_v10_0_get_gpu_clock_counter(struct amdgpu_device *adev)
 
         amdgpu_gfx_off_ctrl(adev, false);
         mutex_lock(&adev->gfx.gpu_clock_mutex);
-       WREG32_SOC15(GC, 0, mmRLC_CAPTURE_GPU_CLOCK_COUNT, 1);
-       clock = (uint64_t)RREG32_SOC15(GC, 0, mmRLC_GPU_CLOCK_COUNT_LSB) |
-               ((uint64_t)RREG32_SOC15(GC, 0, mmRLC_GPU_CLOCK_COUNT_MSB) << 32ULL);
+       if (!amdgpu_sriov_vf(adev)) {
+               WREG32_SOC15(GC, 0, mmRLC_CAPTURE_GPU_CLOCK_COUNT, 1);
+               clock = (uint64_t)RREG32_SOC15(GC, 0, mmRLC_GPU_CLOCK_COUNT_LSB) |
+                       ((uint64_t)RREG32_SOC15(GC, 0, mmRLC_GPU_CLOCK_COUNT_MSB) << 32ULL);
+       } else {
+               clock = (uint64_t)RREG32_SOC15(SMUIO, 0, mmGOLDEN_TSC_COUNT_LOWER) |
+                       ((uint64_t)RREG32_SOC15(SMUIO, 0, mmGOLDEN_TSC_COUNT_UPPER) << 32ULL);
+       }
         mutex_unlock(&adev->gfx.gpu_clock_mutex);
         amdgpu_gfx_off_ctrl(adev, true);
         return clock;
--
2.7.4

_______________________________________________
amd-gfx mailing list
amd-gfx@xxxxxxxxxxxxxxxxxxxxx
https://nam11.safelinks.protection.outlook.com/?url="">
_______________________________________________
amd-gfx mailing list
amd-gfx@xxxxxxxxxxxxxxxxxxxxx
https://lists.freedesktop.org/mailman/listinfo/amd-gfx

[Index of Archives]     [Linux USB Devel]     [Linux Audio Users]     [Yosemite News]     [Linux Kernel]     [Linux SCSI]

  Powered by Linux