[PATCH 1/2] drm/amdgpu: add reset register trace dump function for gfx_v10_0

Somalapuram Amaranath Amaranath.Somalapuram at amd.com
Mon Jan 17 06:32:58 UTC 2022


Implementation of register trace dump function on the AMD GPU resets

Signed-off-by: Somalapuram Amaranath <Amaranath.Somalapuram at amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_trace.h |  8 ++++
 drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c    | 53 ++++++++++++++++++++++-
 drivers/gpu/drm/amd/include/amd_shared.h  |  1 +
 3 files changed, 60 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_trace.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_trace.h
index d855cb53c7e0..c97b53b54333 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_trace.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_trace.h
@@ -537,6 +537,14 @@ TRACE_EVENT(amdgpu_ib_pipe_sync,
 		      __entry->seqno)
 );
 
+TRACE_EVENT(gfx_v10_0_reset_reg_dumps,
+	    TP_PROTO(char *reg_dumps),
+	    TP_ARGS(reg_dumps),
+	    TP_STRUCT__entry(__string(dumps, reg_dumps)),
+	    TP_fast_assign(__assign_str(dumps, reg_dumps);),
+	    TP_printk("amdgpu register dump {%s}", __get_str(dumps))
+);
+
 #undef AMDGPU_JOB_GET_TIMELINE_NAME
 #endif
 
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c
index dbe7442fb25c..a63cdd0ad53a 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c
@@ -46,7 +46,7 @@
 #include "v10_structs.h"
 #include "gfx_v10_0.h"
 #include "nbio_v2_3.h"
-
+#include "amdgpu_trace.h"
 /*
  * Navi10 has two graphic rings to share each graphic pipe.
  * 1. Primary ring
@@ -188,6 +188,12 @@
 #define RLCG_ERROR_REPORT_ENABLED(adev) \
 	(amdgpu_sriov_reg_indirect_mmhub(adev) || amdgpu_sriov_reg_indirect_gc(adev))
 
+#define N_REGS (17)
+#define DUMP_REG(addr) do {					\
+		(dump)[i][0] = (addr);				\
+		(dump)[i++][1] =  RREG32_SOC15_IP(GC, addr);	\
+	} while (0)
+
 MODULE_FIRMWARE("amdgpu/navi10_ce.bin");
 MODULE_FIRMWARE("amdgpu/navi10_pfp.bin");
 MODULE_FIRMWARE("amdgpu/navi10_me.bin");
@@ -7466,7 +7472,6 @@ static int gfx_v10_0_hw_init(void *handle)
 {
 	int r;
 	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
-
 	if (!amdgpu_emu_mode)
 		gfx_v10_0_init_golden_registers(adev);
 
@@ -7580,6 +7585,49 @@ static int gfx_v10_0_hw_fini(void *handle)
 	return 0;
 }
 
+static int gfx_v10_0_reset_reg_dumps(void *handle)
+{
+	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
+	uint32_t i = 0;
+	uint32_t (*dump)[2], n_regs = 0;
+	char *reg_dumps;
+
+	dump = kmalloc(N_REGS*2*sizeof(uint32_t), GFP_KERNEL);
+	reg_dumps = kmalloc(1024, GFP_KERNEL);
+
+	if (dump == NULL || reg_dumps == NULL)
+		return -ENOMEM;
+
+	DUMP_REG(mmGRBM_STATUS2);
+	DUMP_REG(mmGRBM_STATUS_SE0);
+	DUMP_REG(mmGRBM_STATUS_SE1);
+	DUMP_REG(mmGRBM_STATUS_SE2);
+	DUMP_REG(mmGRBM_STATUS_SE3);
+	DUMP_REG(mmSDMA0_STATUS_REG);
+	DUMP_REG(mmSDMA1_STATUS_REG);
+	DUMP_REG(mmCP_STAT);
+	DUMP_REG(mmCP_STALLED_STAT1);
+	DUMP_REG(mmCP_STALLED_STAT1);
+	DUMP_REG(mmCP_STALLED_STAT3);
+	DUMP_REG(mmCP_CPC_STATUS);
+	DUMP_REG(mmCP_CPC_BUSY_STAT);
+	DUMP_REG(mmCP_CPC_STALLED_STAT1);
+	DUMP_REG(mmCP_CPF_STATUS);
+	DUMP_REG(mmCP_CPF_BUSY_STAT);
+	DUMP_REG(mmCP_CPF_STALLED_STAT1);
+
+	n_regs = i;
+
+	for (i = 0; i < n_regs; i++)
+		sprintf(reg_dumps + strlen(reg_dumps), "%08x: %08x, ", dump[i][0], dump[i][1]);
+
+	trace_gfx_v10_0_reset_reg_dumps(reg_dumps);
+	kfree(dump);
+	kfree(reg_dumps);
+
+	return 0;
+}
+
 static int gfx_v10_0_suspend(void *handle)
 {
 	return gfx_v10_0_hw_fini(handle);
@@ -9359,6 +9407,7 @@ static const struct amd_ip_funcs gfx_v10_0_ip_funcs = {
 	.set_clockgating_state = gfx_v10_0_set_clockgating_state,
 	.set_powergating_state = gfx_v10_0_set_powergating_state,
 	.get_clockgating_state = gfx_v10_0_get_clockgating_state,
+	.reset_reg_dumps = gfx_v10_0_reset_reg_dumps,
 };
 
 static const struct amdgpu_ring_funcs gfx_v10_0_ring_funcs_gfx = {
diff --git a/drivers/gpu/drm/amd/include/amd_shared.h b/drivers/gpu/drm/amd/include/amd_shared.h
index 4b9e68a79f06..a165680d73a2 100644
--- a/drivers/gpu/drm/amd/include/amd_shared.h
+++ b/drivers/gpu/drm/amd/include/amd_shared.h
@@ -298,6 +298,7 @@ struct amd_ip_funcs {
 				     enum amd_powergating_state state);
 	void (*get_clockgating_state)(void *handle, u32 *flags);
 	int (*enable_umd_pstate)(void *handle, enum amd_dpm_forced_level *level);
+	int (*reset_reg_dumps)(void *handle);
 };
 
 
-- 
2.25.1



More information about the amd-gfx mailing list