Re: [PATCH] drm/amdgpu: turn back rlcg write for gfx_v10

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

 



[AMD Public Use]


Acked-by: Alex Deucher <alexander.deucher@xxxxxxx>

From: Yintian Tao <yttao@xxxxxxx>
Sent: Tuesday, May 12, 2020 6:17 AM
To: Deucher, Alexander <Alexander.Deucher@xxxxxxx>; Liu, Monk <Monk.Liu@xxxxxxx>; Liu, Shaoyun <Shaoyun.Liu@xxxxxxx>
Cc: amd-gfx@xxxxxxxxxxxxxxxxxxxxx <amd-gfx@xxxxxxxxxxxxxxxxxxxxx>; Tao, Yintian <Yintian.Tao@xxxxxxx>
Subject: [PATCH] drm/amdgpu: turn back rlcg write for gfx_v10
 
There is no need to use amdgpu_mm_wreg_mmio_rlc()
during initialization time because this interface
is only designed for debugfs case to access the
registers which are only permitted by RLCG during
run-time. Therefore, turn back rlcg write for gfx_v10.
If we not turn back it, it will raise amdgpu load failure.
[   54.904333] amdgpu: SMU driver if version not matched
[   54.904393] amdgpu: SMU is initialized successfully!
[   54.905971] [drm] kiq ring mec 2 pipe 1 q 0
[   55.115416] amdgpu 0000:00:06.0: [drm:amdgpu_ring_test_helper [amdgpu]] *ERROR* ring gfx_0.0.0 test failed (-110)
[   55.118877] [drm:amdgpu_device_init [amdgpu]] *ERROR* hw_init of IP block <gfx_v10_0> failed -110
[   55.126587] amdgpu 0000:00:06.0: amdgpu_device_ip_init failed
[   55.133466] amdgpu 0000:00:06.0: Fatal error during GPU init

Signed-off-by: Yintian Tao <yttao@xxxxxxx>
---
 drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c | 14 ++++++--------
 1 file changed, 6 insertions(+), 8 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c
index 449408cfd018..bd5dd4f64311 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c
@@ -4577,13 +4577,11 @@ static int gfx_v10_0_init_csb(struct amdgpu_device *adev)
         adev->gfx.rlc.funcs->get_csb_buffer(adev, adev->gfx.rlc.cs_ptr);
 
         /* csib */
-       /* amdgpu_mm_wreg_mmio_rlc will fall back to mmio if doesn't support rlcg_write */
-       amdgpu_mm_wreg_mmio_rlc(adev, SOC15_REG_OFFSET(GC, 0, mmRLC_CSIB_ADDR_HI),
-                                adev->gfx.rlc.clear_state_gpu_addr >> 32, 0);
-       amdgpu_mm_wreg_mmio_rlc(adev, SOC15_REG_OFFSET(GC, 0, mmRLC_CSIB_ADDR_LO),
-                                adev->gfx.rlc.clear_state_gpu_addr & 0xfffffffc, 0);
-       amdgpu_mm_wreg_mmio_rlc(adev, SOC15_REG_OFFSET(GC, 0, mmRLC_CSIB_LENGTH),
-                                adev->gfx.rlc.clear_state_size, 0);
+       WREG32_SOC15_RLC(GC, 0, mmRLC_CSIB_ADDR_HI,
+                        adev->gfx.rlc.clear_state_gpu_addr >> 32);
+       WREG32_SOC15_RLC(GC, 0, mmRLC_CSIB_ADDR_LO,
+                        adev->gfx.rlc.clear_state_gpu_addr & 0xfffffffc);
+       WREG32_SOC15_RLC(GC, 0, mmRLC_CSIB_LENGTH, adev->gfx.rlc.clear_state_size);
 
         return 0;
 }
@@ -5192,7 +5190,7 @@ static int gfx_v10_0_cp_gfx_enable(struct amdgpu_device *adev, bool enable)
         tmp = REG_SET_FIELD(tmp, CP_ME_CNTL, ME_HALT, enable ? 0 : 1);
         tmp = REG_SET_FIELD(tmp, CP_ME_CNTL, PFP_HALT, enable ? 0 : 1);
         tmp = REG_SET_FIELD(tmp, CP_ME_CNTL, CE_HALT, enable ? 0 : 1);
-       amdgpu_mm_wreg_mmio_rlc(adev, SOC15_REG_OFFSET(GC, 0, mmCP_ME_CNTL), tmp, 0);
+       WREG32_SOC15_RLC(GC, 0, mmCP_ME_CNTL, tmp);
 
         for (i = 0; i < adev->usec_timeout; i++) {
                 if (RREG32_SOC15(GC, 0, mmCP_STAT) == 0)
--
2.17.1

_______________________________________________
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