[PATCH AUTOSEL 6.0 26/34] drm/amdgpu: set vm_update_mode=0 as default for Sienna Cichlid in SRIOV case

[PATCH AUTOSEL 6.0 26/34] drm/amdgpu: set vm_update_mode=0 as default for Sienna Cichlid in SRIOV case

[Sasha Levin]

From: Danijel Slivka <danijel.slivka@amd.com>

[ Upstream commit 65f8682b9aaae20c2cdee993e6fe52374ad513c9 ]

For asic with VF MMIO access protection avoid using CPU for VM table updates.
CPU pagetable updates have issues with HDP flush as VF MMIO access protection
blocks write to mmBIF_BX_DEV0_EPF0_VF0_HDP_MEM_COHERENCY_FLUSH_CNTL register
during sriov runtime.

v3: introduce virtualization capability flag AMDGPU_VF_MMIO_ACCESS_PROTECT
which indicates that VF MMIO write access is not allowed in sriov runtime

Signed-off-by: Danijel Slivka <danijel.slivka@amd.com>
Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_virt.c | 6 ++++++
 drivers/gpu/drm/amd/amdgpu/amdgpu_virt.h | 4 ++++
 drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c   | 6 +++++-
 3 files changed, 15 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.c
index 9be57389301b..af5aeb0ec2e9 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.c
@@ -726,6 +726,12 @@ void amdgpu_detect_virtualization(struct amdgpu_device *adev)
 			adev->virt.caps |= AMDGPU_PASSTHROUGH_MODE;
 	}
 
+	if (amdgpu_sriov_vf(adev) && adev->asic_type == CHIP_SIENNA_CICHLID)
+		/* VF MMIO access (except mailbox range) from CPU
+		 * will be blocked during sriov runtime
+		 */
+		adev->virt.caps |= AMDGPU_VF_MMIO_ACCESS_PROTECT;
+
 	/* we have the ability to check now */
 	if (amdgpu_sriov_vf(adev)) {
 		switch (adev->asic_type) {
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.h
index 239f232f9c02..617d072275eb 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.h
@@ -31,6 +31,7 @@
 #define AMDGPU_SRIOV_CAPS_IS_VF        (1 << 2) /* this GPU is a virtual function */
 #define AMDGPU_PASSTHROUGH_MODE        (1 << 3) /* thw whole GPU is pass through for VM */
 #define AMDGPU_SRIOV_CAPS_RUNTIME      (1 << 4) /* is out of full access mode */
+#define AMDGPU_VF_MMIO_ACCESS_PROTECT  (1 << 5) /* MMIO write access is not allowed in sriov runtime */
 
 /* flags for indirect register access path supported by rlcg for sriov */
 #define AMDGPU_RLCG_GC_WRITE_LEGACY    (0x8 << 28)
@@ -294,6 +295,9 @@ struct amdgpu_video_codec_info;
 #define amdgpu_passthrough(adev) \
 ((adev)->virt.caps & AMDGPU_PASSTHROUGH_MODE)
 
+#define amdgpu_sriov_vf_mmio_access_protection(adev) \
+((adev)->virt.caps & AMDGPU_VF_MMIO_ACCESS_PROTECT)
+
 static inline bool is_virtual_machine(void)
 {
 #if defined(CONFIG_X86)
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c
index 690fd4f639f1..04130f8813ef 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c
@@ -2301,7 +2301,11 @@ void amdgpu_vm_manager_init(struct amdgpu_device *adev)
 	 */
 #ifdef CONFIG_X86_64
 	if (amdgpu_vm_update_mode == -1) {
-		if (amdgpu_gmc_vram_full_visible(&adev->gmc))
+		/* For asic with VF MMIO access protection
+		 * avoid using CPU for VM table updates
+		 */
+		if (amdgpu_gmc_vram_full_visible(&adev->gmc) &&
+		    !amdgpu_sriov_vf_mmio_access_protection(adev))
 			adev->vm_manager.vm_update_mode =
 				AMDGPU_VM_USE_CPU_FOR_COMPUTE;
 		else
-- 
2.35.1

[PATCH AUTOSEL 6.0 27/34] drm/amd/pm: skip loading pptable from driver on secure board for smu_v13_0_10

[Sasha Levin]

From: Kenneth Feng <kenneth.feng@amd.com>

[ Upstream commit f700486cd1f2bf381671d1c2c7dc9000db10c50e ]

skip loading pptable from driver on secure board since it's loaded from psp.

Signed-off-by: Kenneth Feng <kenneth.feng@amd.com>
Reviewed-by: Guan Yu <Guan.Yu@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
 drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0.c b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0.c
index 93f9b8377539..750d8da84fac 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0.c
+++ b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0.c
@@ -210,7 +210,8 @@ int smu_v13_0_init_pptable_microcode(struct smu_context *smu)
 		return 0;
 
 	if ((adev->ip_versions[MP1_HWIP][0] == IP_VERSION(13, 0, 7)) ||
-	    (adev->ip_versions[MP1_HWIP][0] == IP_VERSION(13, 0, 0)))
+	    (adev->ip_versions[MP1_HWIP][0] == IP_VERSION(13, 0, 0)) ||
+	    (adev->ip_versions[MP1_HWIP][0] == IP_VERSION(13, 0, 10)))
 		return 0;
 
 	/* override pptable_id from driver parameter */
-- 
2.35.1

[PATCH AUTOSEL 6.0 28/34] drm/amdkfd: Fix type of reset_type parameter in hqd_destroy() callback

[Sasha Levin]

From: Nathan Chancellor <nathan@kernel.org>

[ Upstream commit e688ba3e276422aa88eae7a54186a95320836081 ]

When booting a kernel compiled with CONFIG_CFI_CLANG on a machine with
an RX 6700 XT, there is a CFI failure in kfd_destroy_mqd_cp():

  [   12.894543] CFI failure at kfd_destroy_mqd_cp+0x2a/0x40 [amdgpu] (target: hqd_destroy_v10_3+0x0/0x260 [amdgpu]; expected type: 0x8594d794)

Clang's kernel Control Flow Integrity (kCFI) makes sure that all
indirect call targets have a type that exactly matches the function
pointer prototype. In this case, hqd_destroy()'s third parameter,
reset_type, should have a type of 'uint32_t' but every implementation of
this callback has a third parameter type of 'enum kfd_preempt_type'.

Update the function pointer prototype to match reality so that there is
no more CFI violation.

Link: https://github.com/ClangBuiltLinux/linux/issues/1738
Signed-off-by: Nathan Chancellor <nathan@kernel.org>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
 drivers/gpu/drm/amd/include/kgd_kfd_interface.h | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/amd/include/kgd_kfd_interface.h b/drivers/gpu/drm/amd/include/kgd_kfd_interface.h
index e85364dff4e0..5cb3e8634739 100644
--- a/drivers/gpu/drm/amd/include/kgd_kfd_interface.h
+++ b/drivers/gpu/drm/amd/include/kgd_kfd_interface.h
@@ -262,8 +262,9 @@ struct kfd2kgd_calls {
 				uint32_t queue_id);
 
 	int (*hqd_destroy)(struct amdgpu_device *adev, void *mqd,
-				uint32_t reset_type, unsigned int timeout,
-				uint32_t pipe_id, uint32_t queue_id);
+				enum kfd_preempt_type reset_type,
+				unsigned int timeout, uint32_t pipe_id,
+				uint32_t queue_id);
 
 	bool (*hqd_sdma_is_occupied)(struct amdgpu_device *adev, void *mqd);
 
-- 
2.35.1

[PATCH AUTOSEL 6.0 29/34] drm/amdgpu: Program GC registers through RLCG interface in gfx_v11/gmc_v11

[Sasha Levin]

From: Yifan Zha <Yifan.Zha@amd.com>

[ Upstream commit 97a3d6090f5c2a165dc88bda05c1dcf9f08bf886 ]

[Why]
L1 blocks most of GC registers accessing by MMIO.

[How]
Use RLCG interface to program GC registers under SRIOV VF in full access time.

Signed-off-by: Yifan Zha <Yifan.Zha@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Acked-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
 .../gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v11.c |  2 +-
 drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c         |  2 +-
 drivers/gpu/drm/amd/amdgpu/gmc_v11_0.c         | 18 +++++++++++-------
 3 files changed, 13 insertions(+), 9 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v11.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v11.c
index 0b0a72ca5695..7e80caa05060 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v11.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v11.c
@@ -111,7 +111,7 @@ static int init_interrupts_v11(struct amdgpu_device *adev, uint32_t pipe_id)
 
 	lock_srbm(adev, mec, pipe, 0, 0);
 
-	WREG32(SOC15_REG_OFFSET(GC, 0, regCPC_INT_CNTL),
+	WREG32_SOC15(GC, 0, regCPC_INT_CNTL,
 		CP_INT_CNTL_RING0__TIME_STAMP_INT_ENABLE_MASK |
 		CP_INT_CNTL_RING0__OPCODE_ERROR_INT_ENABLE_MASK);
 
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c
index daf8ba8235cd..03775e0a8100 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c
@@ -1729,7 +1729,7 @@ static void gfx_v11_0_init_compute_vmid(struct amdgpu_device *adev)
 		WREG32_SOC15(GC, 0, regSH_MEM_BASES, sh_mem_bases);
 
 		/* Enable trap for each kfd vmid. */
-		data = RREG32(SOC15_REG_OFFSET(GC, 0, regSPI_GDBG_PER_VMID_CNTL));
+		data = RREG32_SOC15(GC, 0, regSPI_GDBG_PER_VMID_CNTL);
 		data = REG_SET_FIELD(data, SPI_GDBG_PER_VMID_CNTL, TRAP_EN, 1);
 	}
 	soc21_grbm_select(adev, 0, 0, 0, 0);
diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v11_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v11_0.c
index 1471bfb9ae38..2475fdbe8010 100644
--- a/drivers/gpu/drm/amd/amdgpu/gmc_v11_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gmc_v11_0.c
@@ -185,6 +185,10 @@ static void gmc_v11_0_flush_vm_hub(struct amdgpu_device *adev, uint32_t vmid,
 	/* Use register 17 for GART */
 	const unsigned eng = 17;
 	unsigned int i;
+	unsigned char hub_ip = 0;
+
+	hub_ip = (vmhub == AMDGPU_GFXHUB_0) ?
+		   GC_HWIP : MMHUB_HWIP;
 
 	spin_lock(&adev->gmc.invalidate_lock);
 	/*
@@ -198,8 +202,8 @@ static void gmc_v11_0_flush_vm_hub(struct amdgpu_device *adev, uint32_t vmid,
 	if (use_semaphore) {
 		for (i = 0; i < adev->usec_timeout; i++) {
 			/* a read return value of 1 means semaphore acuqire */
-			tmp = RREG32_NO_KIQ(hub->vm_inv_eng0_sem +
-					    hub->eng_distance * eng);
+			tmp = RREG32_RLC_NO_KIQ(hub->vm_inv_eng0_sem +
+					    hub->eng_distance * eng, hub_ip);
 			if (tmp & 0x1)
 				break;
 			udelay(1);
@@ -209,12 +213,12 @@ static void gmc_v11_0_flush_vm_hub(struct amdgpu_device *adev, uint32_t vmid,
 			DRM_ERROR("Timeout waiting for sem acquire in VM flush!\n");
 	}
 
-	WREG32_NO_KIQ(hub->vm_inv_eng0_req + hub->eng_distance * eng, inv_req);
+	WREG32_RLC_NO_KIQ(hub->vm_inv_eng0_req + hub->eng_distance * eng, inv_req, hub_ip);
 
 	/* Wait for ACK with a delay.*/
 	for (i = 0; i < adev->usec_timeout; i++) {
-		tmp = RREG32_NO_KIQ(hub->vm_inv_eng0_ack +
-				    hub->eng_distance * eng);
+		tmp = RREG32_RLC_NO_KIQ(hub->vm_inv_eng0_ack +
+				    hub->eng_distance * eng, hub_ip);
 		tmp &= 1 << vmid;
 		if (tmp)
 			break;
@@ -228,8 +232,8 @@ static void gmc_v11_0_flush_vm_hub(struct amdgpu_device *adev, uint32_t vmid,
 		 * add semaphore release after invalidation,
 		 * write with 0 means semaphore release
 		 */
-		WREG32_NO_KIQ(hub->vm_inv_eng0_sem +
-			      hub->eng_distance * eng, 0);
+		WREG32_RLC_NO_KIQ(hub->vm_inv_eng0_sem +
+			      hub->eng_distance * eng, 0, hub_ip);
 
 	/* Issue additional private vm invalidation to MMHUB */
 	if ((vmhub != AMDGPU_GFXHUB_0) &&
-- 
2.35.1

[PATCH AUTOSEL 6.0 30/34] drm/amdgpu: dequeue mes scheduler during fini

[Sasha Levin]

From: YuBiao Wang <YuBiao.Wang@amd.com>

[ Upstream commit 2abe92c7adc9c0397ba51bf74909b85bc0fff84b ]

[Why]
If mes is not dequeued during fini, mes will be in an uncleaned state
during reload, then mes couldn't receive some commands which leads to
reload failure.

[How]
Perform MES dequeue via MMIO after all the unmap jobs are done by mes
and before kiq fini.

v2: Move the dequeue operation inside kiq_hw_fini.

Signed-off-by: YuBiao Wang <YuBiao.Wang@amd.com>
Reviewed-by: Jack Xiao <Jack.Xiao@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
 drivers/gpu/drm/amd/amdgpu/mes_v11_0.c | 42 ++++++++++++++++++++++++--
 1 file changed, 39 insertions(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/mes_v11_0.c b/drivers/gpu/drm/amd/amdgpu/mes_v11_0.c
index f92744b8d79d..2dd827472d6e 100644
--- a/drivers/gpu/drm/amd/amdgpu/mes_v11_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/mes_v11_0.c
@@ -1145,6 +1145,42 @@ static int mes_v11_0_sw_fini(void *handle)
 	return 0;
 }
 
+static void mes_v11_0_kiq_dequeue_sched(struct amdgpu_device *adev)
+{
+	uint32_t data;
+	int i;
+
+	mutex_lock(&adev->srbm_mutex);
+	soc21_grbm_select(adev, 3, AMDGPU_MES_SCHED_PIPE, 0, 0);
+
+	/* disable the queue if it's active */
+	if (RREG32_SOC15(GC, 0, regCP_HQD_ACTIVE) & 1) {
+		WREG32_SOC15(GC, 0, regCP_HQD_DEQUEUE_REQUEST, 1);
+		for (i = 0; i < adev->usec_timeout; i++) {
+			if (!(RREG32_SOC15(GC, 0, regCP_HQD_ACTIVE) & 1))
+				break;
+			udelay(1);
+		}
+	}
+	data = RREG32_SOC15(GC, 0, regCP_HQD_PQ_DOORBELL_CONTROL);
+	data = REG_SET_FIELD(data, CP_HQD_PQ_DOORBELL_CONTROL,
+				DOORBELL_EN, 0);
+	data = REG_SET_FIELD(data, CP_HQD_PQ_DOORBELL_CONTROL,
+				DOORBELL_HIT, 1);
+	WREG32_SOC15(GC, 0, regCP_HQD_PQ_DOORBELL_CONTROL, data);
+
+	WREG32_SOC15(GC, 0, regCP_HQD_PQ_DOORBELL_CONTROL, 0);
+
+	WREG32_SOC15(GC, 0, regCP_HQD_PQ_WPTR_LO, 0);
+	WREG32_SOC15(GC, 0, regCP_HQD_PQ_WPTR_HI, 0);
+	WREG32_SOC15(GC, 0, regCP_HQD_PQ_RPTR, 0);
+
+	soc21_grbm_select(adev, 0, 0, 0, 0);
+	mutex_unlock(&adev->srbm_mutex);
+
+	adev->mes.ring.sched.ready = false;
+}
+
 static void mes_v11_0_kiq_setting(struct amdgpu_ring *ring)
 {
 	uint32_t tmp;
@@ -1196,6 +1232,9 @@ static int mes_v11_0_kiq_hw_init(struct amdgpu_device *adev)
 
 static int mes_v11_0_kiq_hw_fini(struct amdgpu_device *adev)
 {
+	if (adev->mes.ring.sched.ready)
+		mes_v11_0_kiq_dequeue_sched(adev);
+
 	mes_v11_0_enable(adev, false);
 	return 0;
 }
@@ -1251,9 +1290,6 @@ static int mes_v11_0_hw_init(void *handle)
 
 static int mes_v11_0_hw_fini(void *handle)
 {
-	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
-
-	adev->mes.ring.sched.ready = false;
 	return 0;
 }
 
-- 
2.35.1