drm/amdgpu - optimize rlc spm cntl

v1
- driver MMIO read the register to check whether write is required
- if write is required, sriov full time to use rlcg, otherwise use KIQ

v2
- include gfx v11 sriov runtime case

Signed-off-by: Jane Jian <Jane.Jian@amd.com>
Acked-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

3 files changed, 38 insertions, 20 deletions

diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c
index 953df202953a..1f516466ac13 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c
@@ -8062,15 +8062,24 @@ static int gfx_v10_0_update_gfx_clock_gating(struct amdgpu_device *adev,
 static void gfx_v10_0_update_spm_vmid_internal(struct amdgpu_device *adev,
 					       unsigned int vmid)
 {
-	u32 data;
+	u32 reg, pre_data, data;
 
+	reg = SOC15_REG_OFFSET(GC, 0, mmRLC_SPM_MC_CNTL);
 	/* not for *_SOC15 */
-	data = RREG32_SOC15_NO_KIQ(GC, 0, mmRLC_SPM_MC_CNTL);
+	if (amdgpu_sriov_is_pp_one_vf(adev) && !amdgpu_sriov_runtime(adev))
+		pre_data = RREG32_NO_KIQ(reg);
+	else
+		pre_data = RREG32(reg);
 
-	data &= ~RLC_SPM_MC_CNTL__RLC_SPM_VMID_MASK;
+	data = pre_data & (~RLC_SPM_MC_CNTL__RLC_SPM_VMID_MASK);
 	data |= (vmid & RLC_SPM_MC_CNTL__RLC_SPM_VMID_MASK) << RLC_SPM_MC_CNTL__RLC_SPM_VMID__SHIFT;
 
-	WREG32_SOC15_NO_KIQ(GC, 0, mmRLC_SPM_MC_CNTL, data);
+	if (pre_data != data) {
+		if (amdgpu_sriov_is_pp_one_vf(adev) && !amdgpu_sriov_runtime(adev)) {
+			WREG32_SOC15_NO_KIQ(GC, 0, mmRLC_SPM_MC_CNTL, data);
+		} else
+			WREG32_SOC15(GC, 0, mmRLC_SPM_MC_CNTL, data);
+	}
 }
 
 static void gfx_v10_0_update_spm_vmid(struct amdgpu_device *adev, struct amdgpu_ring *ring, unsigned int vmid)
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c
index ad6431013c73..e5882da8332a 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c
@@ -5030,24 +5030,31 @@ static int gfx_v11_0_update_gfx_clock_gating(struct amdgpu_device *adev,
 
 static void gfx_v11_0_update_spm_vmid(struct amdgpu_device *adev, struct amdgpu_ring *ring, unsigned vmid)
 {
-	u32 data;
+	u32 reg, pre_data, data;
 
 	amdgpu_gfx_off_ctrl(adev, false);
+	reg = SOC15_REG_OFFSET(GC, 0, regRLC_SPM_MC_CNTL);
+	if (amdgpu_sriov_is_pp_one_vf(adev) && !amdgpu_sriov_runtime(adev))
+		pre_data = RREG32_NO_KIQ(reg);
+	else
+		pre_data = RREG32(reg);
 
-	data = RREG32_SOC15_NO_KIQ(GC, 0, regRLC_SPM_MC_CNTL);
-
-	data &= ~RLC_SPM_MC_CNTL__RLC_SPM_VMID_MASK;
+	data = pre_data & (~RLC_SPM_MC_CNTL__RLC_SPM_VMID_MASK);
 	data |= (vmid & RLC_SPM_MC_CNTL__RLC_SPM_VMID_MASK) << RLC_SPM_MC_CNTL__RLC_SPM_VMID__SHIFT;
 
-	WREG32_SOC15_NO_KIQ(GC, 0, regRLC_SPM_MC_CNTL, data);
-
+	if (pre_data != data) {
+		if (amdgpu_sriov_is_pp_one_vf(adev) && !amdgpu_sriov_runtime(adev)) {
+			WREG32_SOC15_NO_KIQ(GC, 0, regRLC_SPM_MC_CNTL, data);
+		} else
+			WREG32_SOC15(GC, 0, regRLC_SPM_MC_CNTL, data);
+	}
 	amdgpu_gfx_off_ctrl(adev, true);
 
 	if (ring
 		&& amdgpu_sriov_is_pp_one_vf(adev)
+		&& (pre_data != data)
 		&& ((ring->funcs->type == AMDGPU_RING_TYPE_GFX)
 			|| (ring->funcs->type == AMDGPU_RING_TYPE_COMPUTE))) {
-		uint32_t reg = SOC15_REG_OFFSET(GC, 0, regRLC_SPM_MC_CNTL);
 		amdgpu_ring_emit_wreg(ring, reg, data);
 	}
 }
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
index 07b299ec7169..aecc2bcea145 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
@@ -1395,21 +1395,23 @@ static int gfx_v9_4_3_rlc_resume(struct amdgpu_device *adev)
 static void gfx_v9_4_3_update_spm_vmid(struct amdgpu_device *adev, struct amdgpu_ring *ring,
 				       unsigned vmid)
 {
-	u32 reg, data;
+	u32 reg, pre_data, data;
 
 	reg = SOC15_REG_OFFSET(GC, GET_INST(GC, 0), regRLC_SPM_MC_CNTL);
-	if (amdgpu_sriov_is_pp_one_vf(adev))
-		data = RREG32_NO_KIQ(reg);
+	if (amdgpu_sriov_is_pp_one_vf(adev) && !amdgpu_sriov_runtime(adev))
+		pre_data = RREG32_NO_KIQ(reg);
 	else
-		data = RREG32(reg);
+		pre_data = RREG32(reg);
 
-	data &= ~RLC_SPM_MC_CNTL__RLC_SPM_VMID_MASK;
+	data =	pre_data & (~RLC_SPM_MC_CNTL__RLC_SPM_VMID_MASK);
 	data |= (vmid & RLC_SPM_MC_CNTL__RLC_SPM_VMID_MASK) << RLC_SPM_MC_CNTL__RLC_SPM_VMID__SHIFT;
 
-	if (amdgpu_sriov_is_pp_one_vf(adev))
-		WREG32_SOC15_NO_KIQ(GC, GET_INST(GC, 0), regRLC_SPM_MC_CNTL, data);
-	else
-		WREG32_SOC15(GC, GET_INST(GC, 0), regRLC_SPM_MC_CNTL, data);
+	if (pre_data != data) {
+		if (amdgpu_sriov_is_pp_one_vf(adev) && !amdgpu_sriov_runtime(adev)) {
+			WREG32_SOC15_NO_KIQ(GC, GET_INST(GC, 0), regRLC_SPM_MC_CNTL, data);
+		} else
+			WREG32_SOC15(GC, GET_INST(GC, 0), regRLC_SPM_MC_CNTL, data);
+	}
 }
 
 static const struct soc15_reg_rlcg rlcg_access_gc_9_4_3[] = {