Merge tag 'amd-drm-next-6.6-2023-07-28' of https://gitlab.freedesktop.org/agd5f/linux into drm-next

amd-drm-next-6.6-2023-07-28:

amdgpu:
- Lots of checkpatch cleanups
- GFX 9.4.3 updates
- Add USB PD and IFWI flashing documentation
- GPUVM updates
- RAS fixes
- DRR fixes
- FAMS fixes
- Virtual display fixes
- Soft IH fixes
- SMU13 fixes
- Rework PSP firmware loading for other IPs
- Kernel doc fixes
- DCN 3.0.1 fixes
- LTTPR fixes
- DP MST fixes
- DCN 3.1.6 fixes
- SubVP fixes
- Display bandwidth calculation fixes
- VCN4 secure submission fixes
- Allow building DC on RISC-V
- Add visible FB info to bo_print_info
- HBR3 fixes
- Add PSP 14.0 support
- GFX9 MCBP fix
- GMC10 vmhub index fix
- GMC11 vmhub index fix
- Create a new doorbell manager
- SR-IOV fixes

amdkfd:
- Cleanup CRIU dma-buf handling
- Use KIQ to unmap HIQ
- GFX 9.4.3 debugger updates
- GFX 9.4.2 debugger fixes
- Enable cooperative groups fof gfx11
- SVM fixes

radeon:
- Lots of checkpatch cleanups

Merge conflicts:
- drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c
	The switch to drm eu helpers in 8a206685d36f ("drm/amdgpu: use
	drm_exec for GEM and CSA handling v2") clashed with the
	cosmetic cleanups from 30953c4d000b ("drm/amdgpu: Fix style
	issues in amdgpu_gem.c"). I
	kept the former since the cleanup up code is gone.
- drivers/gpu/drm/amd/amdgpu/atom.c.
	adf64e214280 ("drm/amd: Avoid reading the VBIOS part number
	twice") removed code that 992b8fe106ab ("drm/radeon: Replace
	all non-returning strlcpy with strscpy") polished.

From: Alex Deucher <[email protected]>
Link: https://patchwork.freedesktop.org/patch/msgid/[email protected]
[sima: some merge conflict wrangling as noted]
Signed-off-by: Daniel Vetter <[email protected]>

1 files changed, 18 insertions, 16 deletions

diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_debug.c b/drivers/gpu/drm/amd/amdkfd/kfd_debug.c
index fff3ccc04fa9..9ec750666382 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_debug.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_debug.c
@@ -302,8 +302,7 @@ static int kfd_dbg_set_queue_workaround(struct queue *q, bool enable)
 	if (!q)
 		return 0;
 
-	if (KFD_GC_VERSION(q->device) < IP_VERSION(11, 0, 0) ||
-	    KFD_GC_VERSION(q->device) >= IP_VERSION(12, 0, 0))
+	if (!kfd_dbg_has_cwsr_workaround(q->device))
 		return 0;
 
 	if (enable && q->properties.is_user_cu_masked)
@@ -345,11 +344,10 @@ unwind:
 	return r;
 }
 
-int kfd_dbg_set_mes_debug_mode(struct kfd_process_device *pdd)
+int kfd_dbg_set_mes_debug_mode(struct kfd_process_device *pdd, bool sq_trap_en)
 {
 	uint32_t spi_dbg_cntl = pdd->spi_dbg_override | pdd->spi_dbg_launch_mode;
 	uint32_t flags = pdd->process->dbg_flags;
-	bool sq_trap_en = !!spi_dbg_cntl;
 
 	if (!kfd_dbg_is_per_vmid_supported(pdd->dev))
 		return 0;
@@ -433,7 +431,7 @@ int kfd_dbg_trap_clear_dev_address_watch(struct kfd_process_device *pdd,
 	if (!pdd->dev->kfd->shared_resources.enable_mes)
 		r = debug_map_and_unlock(pdd->dev->dqm);
 	else
-		r = kfd_dbg_set_mes_debug_mode(pdd);
+		r = kfd_dbg_set_mes_debug_mode(pdd, true);
 
 	kfd_dbg_clear_dev_watch_id(pdd, watch_id);
 
@@ -446,7 +444,8 @@ int kfd_dbg_trap_set_dev_address_watch(struct kfd_process_device *pdd,
 					uint32_t *watch_id,
 					uint32_t watch_mode)
 {
-	int r = kfd_dbg_get_dev_watch_id(pdd, watch_id);
+	int xcc_id, r = kfd_dbg_get_dev_watch_id(pdd, watch_id);
+	uint32_t xcc_mask = pdd->dev->xcc_mask;
 
 	if (r)
 		return r;
@@ -460,19 +459,21 @@ int kfd_dbg_trap_set_dev_address_watch(struct kfd_process_device *pdd,
 	}
 
 	amdgpu_gfx_off_ctrl(pdd->dev->adev, false);
-	pdd->watch_points[*watch_id] = pdd->dev->kfd2kgd->set_address_watch(
+	for_each_inst(xcc_id, xcc_mask)
+		pdd->watch_points[*watch_id] = pdd->dev->kfd2kgd->set_address_watch(
 				pdd->dev->adev,
 				watch_address,
 				watch_address_mask,
 				*watch_id,
 				watch_mode,
-				pdd->dev->vm_info.last_vmid_kfd);
+				pdd->dev->vm_info.last_vmid_kfd,
+				xcc_id);
 	amdgpu_gfx_off_ctrl(pdd->dev->adev, true);
 
 	if (!pdd->dev->kfd->shared_resources.enable_mes)
 		r = debug_map_and_unlock(pdd->dev->dqm);
 	else
-		r = kfd_dbg_set_mes_debug_mode(pdd);
+		r = kfd_dbg_set_mes_debug_mode(pdd, true);
 
 	/* HWS is broken so no point in HW rollback but release the watchpoint anyways */
 	if (r)
@@ -514,7 +515,7 @@ int kfd_dbg_trap_set_flags(struct kfd_process *target, uint32_t *flags)
 		if (!pdd->dev->kfd->shared_resources.enable_mes)
 			r = debug_refresh_runlist(pdd->dev->dqm);
 		else
-			r = kfd_dbg_set_mes_debug_mode(pdd);
+			r = kfd_dbg_set_mes_debug_mode(pdd, true);
 
 		if (r) {
 			target->dbg_flags = prev_flags;
@@ -537,7 +538,7 @@ int kfd_dbg_trap_set_flags(struct kfd_process *target, uint32_t *flags)
 			if (!pdd->dev->kfd->shared_resources.enable_mes)
 				debug_refresh_runlist(pdd->dev->dqm);
 			else
-				kfd_dbg_set_mes_debug_mode(pdd);
+				kfd_dbg_set_mes_debug_mode(pdd, true);
 		}
 	}
 
@@ -599,7 +600,7 @@ void kfd_dbg_trap_deactivate(struct kfd_process *target, bool unwind, int unwind
 		if (!pdd->dev->kfd->shared_resources.enable_mes)
 			debug_refresh_runlist(pdd->dev->dqm);
 		else
-			kfd_dbg_set_mes_debug_mode(pdd);
+			kfd_dbg_set_mes_debug_mode(pdd, !kfd_dbg_has_cwsr_workaround(pdd->dev));
 	}
 
 	kfd_dbg_set_workaround(target, false);
@@ -715,7 +716,7 @@ int kfd_dbg_trap_activate(struct kfd_process *target)
 		if (!pdd->dev->kfd->shared_resources.enable_mes)
 			r = debug_refresh_runlist(pdd->dev->dqm);
 		else
-			r = kfd_dbg_set_mes_debug_mode(pdd);
+			r = kfd_dbg_set_mes_debug_mode(pdd, true);
 
 		if (r) {
 			target->runtime_info.runtime_state =
@@ -751,7 +752,8 @@ int kfd_dbg_trap_enable(struct kfd_process *target, uint32_t fd,
 		if (!KFD_IS_SOC15(pdd->dev))
 			return -ENODEV;
 
-		if (!kfd_dbg_has_gws_support(pdd->dev) && pdd->qpd.num_gws)
+		if (pdd->qpd.num_gws && (!kfd_dbg_has_gws_support(pdd->dev) ||
+					 kfd_dbg_has_cwsr_workaround(pdd->dev)))
 			return -EBUSY;
 	}
 
@@ -848,7 +850,7 @@ int kfd_dbg_trap_set_wave_launch_override(struct kfd_process *target,
 		if (!pdd->dev->kfd->shared_resources.enable_mes)
 			r = debug_refresh_runlist(pdd->dev->dqm);
 		else
-			r = kfd_dbg_set_mes_debug_mode(pdd);
+			r = kfd_dbg_set_mes_debug_mode(pdd, true);
 
 		if (r)
 			break;
@@ -880,7 +882,7 @@ int kfd_dbg_trap_set_wave_launch_mode(struct kfd_process *target,
 		if (!pdd->dev->kfd->shared_resources.enable_mes)
 			r = debug_refresh_runlist(pdd->dev->dqm);
 		else
-			r = kfd_dbg_set_mes_debug_mode(pdd);
+			r = kfd_dbg_set_mes_debug_mode(pdd, true);
 
 		if (r)
 			break;