The MES event-log/MSCRATCH region is only zeroed once at sw_init. On the
resume path (hw_init without sw_init) it retains stale data, so with
mes_log_enable=1 the MES firmware walks stale pointers on activation and
triggers a recoverable CPC page fault:
[gfxhub] page fault (src_id:0 ring:153 vmid:0 pasid:0)
in page starting at address 0x000000855c237000 from client 10
GCVM_L2_PROTECTION_FAULT_STATUS:0x00000B33
Faulty UTCL2 client ID: CPC (0x5)
MORE_FAULTS: 0x1 WALKER_ERROR: 0x1 PERMISSION_FAULTS: 0x3
MAPPING_ERROR: 0x1 RW: 0x0
Re-clear it before unhalting MES.
Signed-off-by: Jesse Zhang <[email protected]>
---
drivers/gpu/drm/amd/amdgpu/mes_v12_0.c | 11 +++++++++++
1 file changed, 11 insertions(+)
diff --git a/drivers/gpu/drm/amd/amdgpu/mes_v12_0.c
b/drivers/gpu/drm/amd/amdgpu/mes_v12_0.c
index ee72afdb0a1d..9e149cf98889 100644
--- a/drivers/gpu/drm/amd/amdgpu/mes_v12_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/mes_v12_0.c
@@ -1306,6 +1306,17 @@ static void mes_v12_0_enable(struct amdgpu_device *adev,
bool enable)
uint32_t pipe, data = 0;
if (enable) {
+ /*
+ * The event-log/MSCRATCH region is only zeroed once at sw_init.
+ * On the resume path (hw_init without sw_init) it still holds
the
+ * previous session's data, so the MES firmware walks stale
+ * pointers on activation and triggers a CPC page fault.
Re-clear
+ * it before unhalting MES.
+ */
+ if (amdgpu_mes_log_enable && adev->mes.event_log_cpu_addr)
+ memset(adev->mes.event_log_cpu_addr, 0,
+ adev->mes.event_log_size);
+
mutex_lock(&adev->srbm_mutex);
for (pipe = 0; pipe < AMDGPU_MAX_MES_PIPES; pipe++) {
soc24_grbm_select(adev, 3, pipe, 0, 0);
--
2.45.1
