radv: save the bound pipeline pointers into the trace BO

When a GPU hang is detected in radv_gpu_hang_occured() we know
which command buffer is faulty but the bound pipelines might
have been updated during the execution.

The pointers to the radv_pipeline objects are emitted just
after the second trace ID, that way it would be easy to dump
the active shaders at the moment of the hang.

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>

diff --git a/src/amd/vulkan/radv_cmd_buffer.c b/src/amd/vulkan/radv_cmd_buffer.c
index 4578ac8..2f030a2 100644 (file)
--- a/src/amd/vulkan/radv_cmd_buffer.c
+++ b/src/amd/vulkan/radv_cmd_buffer.c
@@ -330,6 +330,19 @@ radv_cmd_buffer_upload_data(struct radv_cmd_buffer *cmd_buffer,
        return true;
 }
 
+static void
+radv_emit_write_data_packet(struct radeon_winsys_cs *cs, uint64_t va,
+                           unsigned count, const uint32_t *data)
+{
+       radeon_emit(cs, PKT3(PKT3_WRITE_DATA, 2 + count, 0));
+       radeon_emit(cs, S_370_DST_SEL(V_370_MEM_ASYNC) |
+                   S_370_WR_CONFIRM(1) |
+                   S_370_ENGINE_SEL(V_370_ME));
+       radeon_emit(cs, va);
+       radeon_emit(cs, va >> 32);
+       radeon_emit_array(cs, data, count);
+}
+
 void radv_cmd_buffer_trace_emit(struct radv_cmd_buffer *cmd_buffer)
 {
        struct radv_device *device = cmd_buffer->device;
@@ -347,18 +360,47 @@ void radv_cmd_buffer_trace_emit(struct radv_cmd_buffer *cmd_buffer)
 
        ++cmd_buffer->state.trace_id;
        device->ws->cs_add_buffer(cs, device->trace_bo, 8);
-       radeon_emit(cs, PKT3(PKT3_WRITE_DATA, 3, 0));
-       radeon_emit(cs, S_370_DST_SEL(V_370_MEM_ASYNC) |
-                   S_370_WR_CONFIRM(1) |
-                   S_370_ENGINE_SEL(V_370_ME));
-       radeon_emit(cs, va);
-       radeon_emit(cs, va >> 32);
-       radeon_emit(cs, cmd_buffer->state.trace_id);
+       radv_emit_write_data_packet(cs, va, 1, &cmd_buffer->state.trace_id);
        radeon_emit(cs, PKT3(PKT3_NOP, 0, 0));
        radeon_emit(cs, AC_ENCODE_TRACE_POINT(cmd_buffer->state.trace_id));
 }
 
 static void
+radv_save_pipeline(struct radv_cmd_buffer *cmd_buffer,
+                  struct radv_pipeline *pipeline, enum ring_type ring)
+{
+       struct radv_device *device = cmd_buffer->device;
+       struct radeon_winsys_cs *cs = cmd_buffer->cs;
+       uint32_t data[2];
+       uint64_t va;
+
+       if (!device->trace_bo)
+               return;
+
+       va = device->ws->buffer_get_va(device->trace_bo);
+
+       switch (ring) {
+       case RING_GFX:
+               va += 8;
+               break;
+       case RING_COMPUTE:
+               va += 16;
+               break;
+       default:
+               assert(!"invalid ring type");
+       }
+
+       MAYBE_UNUSED unsigned cdw_max = radeon_check_space(device->ws,
+                                                          cmd_buffer->cs, 6);
+
+       data[0] = (uintptr_t)pipeline;
+       data[1] = (uintptr_t)pipeline >> 32;
+
+       device->ws->cs_add_buffer(cs, device->trace_bo, 8);
+       radv_emit_write_data_packet(cs, va, 2, data);
+}
+
+static void
 radv_emit_graphics_blend_state(struct radv_cmd_buffer *cmd_buffer,
                               struct radv_pipeline *pipeline)
 {
@@ -898,6 +940,8 @@ radv_emit_graphics_pipeline(struct radv_cmd_buffer *cmd_buffer)
        }
        radeon_set_context_reg(cmd_buffer->cs, R_028A6C_VGT_GS_OUT_PRIM_TYPE, pipeline->graphics.gs_out);
 
+       radv_save_pipeline(cmd_buffer, pipeline, RING_GFX);
+
        cmd_buffer->state.emitted_pipeline = pipeline;
 }
 
@@ -2330,6 +2374,7 @@ radv_emit_compute_pipeline(struct radv_cmd_buffer *cmd_buffer)
                    S_00B81C_NUM_THREAD_FULL(compute_shader->info.cs.block_size[2]));
 
        assert(cmd_buffer->cs->cdw <= cdw_max);
+       radv_save_pipeline(cmd_buffer, pipeline, RING_COMPUTE);
 }
 
 static void radv_mark_descriptor_sets_dirty(struct radv_cmd_buffer *cmd_buffer)

diff --git a/src/amd/vulkan/radv_debug.c b/src/amd/vulkan/radv_debug.c
index 1df2e35..46679af 100644 (file)
--- a/src/amd/vulkan/radv_debug.c
+++ b/src/amd/vulkan/radv_debug.c
@@ -38,6 +38,8 @@
  *
  * [0]: primary trace ID
  * [1]: secondary trace ID
+ * [2-3]: 64-bit GFX pipeline pointer
+ * [4-5]: 64-bit COMPUTE pipeline pointer
  */
 
 bool