From a9818854ea7870ec5464d37b72c89f5fc198708e Mon Sep 17 00:00:00 2001 From: Jonathan Kim Date: Fri, 26 Aug 2022 22:04:15 -0400 Subject: drm/amdgpu: expose debug api for mes Similar to the F32 HWS, the RS64 HWS for GFX11 now supports a multi-process debug API. The skip_process_ctx_clear ADD_QUEUE requirement is to prevent the MES from clearing the process context when the first queue is added to the scheduler in order to maintain debug mode settings during queue preemption and restore. The MES clears the process context in this case due to an unresolved FW caching bug during normal mode operations. During debug mode, the KFD will hold a reference to the target process so the process context should never go stale and MES can afford to skip this requirement. Signed-off-by: Jonathan Kim Reviewed-by: Felix Kuehling Signed-off-by: Alex Deucher --- drivers/gpu/drm/amd/amdgpu/amdgpu_mes.h | 20 ++++++++++++++++++++ 1 file changed, 20 insertions(+) (limited to 'drivers/gpu/drm/amd/amdgpu/amdgpu_mes.h') diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_mes.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_mes.h index 547ec35691fa..d20df0cf0d88 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_mes.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_mes.h @@ -256,6 +256,7 @@ enum mes_misc_opcode { MES_MISC_OP_READ_REG, MES_MISC_OP_WRM_REG_WAIT, MES_MISC_OP_WRM_REG_WR_WAIT, + MES_MISC_OP_SET_SHADER_DEBUGGER, }; struct mes_misc_op_input { @@ -278,6 +279,20 @@ struct mes_misc_op_input { uint32_t reg0; uint32_t reg1; } wrm_reg; + + struct { + uint64_t process_context_addr; + union { + struct { + uint64_t single_memop : 1; + uint64_t single_alu_op : 1; + uint64_t reserved: 30; + }; + uint32_t u32all; + } flags; + uint32_t spi_gdbg_per_vmid_cntl; + uint32_t tcp_watch_cntl[4]; + } set_shader_debugger; }; }; @@ -340,6 +355,11 @@ int amdgpu_mes_reg_wait(struct amdgpu_device *adev, uint32_t reg, int amdgpu_mes_reg_write_reg_wait(struct amdgpu_device *adev, uint32_t reg0, uint32_t reg1, uint32_t ref, uint32_t mask); +int amdgpu_mes_set_shader_debugger(struct amdgpu_device *adev, + uint64_t process_context_addr, + uint32_t spi_gdbg_per_vmid_cntl, + const uint32_t *tcp_watch_cntl, + uint32_t flags); int amdgpu_mes_add_ring(struct amdgpu_device *adev, int gang_id, int queue_type, int idx, -- cgit From 69a8c3ae2dea84a6d571e4c1aad306f630f3ccfd Mon Sep 17 00:00:00 2001 From: Jonathan Kim Date: Thu, 1 Sep 2022 11:27:15 -0400 Subject: drm/amdkfd: apply trap workaround for gfx11 Due to a HW bug, waves in only half the shader arrays can enter trap. When starting a debug session, relocate all waves to the first shader array of each shader engine and mask off the 2nd shader array as unavailable. When ending a debug session, re-enable the 2nd shader array per shader engine. User CU masking per queue cannot be guaranteed to remain functional if requested during debugging (e.g. user cu mask requests only 2nd shader array as an available resource leading to zero HW resources available) nor can runtime be alerted of any of these changes during execution. Make user CU masking and debugging mutual exclusive with respect to availability. If the debugger tries to attach to a process with a user cu masked queue, return the runtime status as enabled but busy. If the debugger tries to attach and fails to reallocate queue waves to the first shader array of each shader engine, return the runtime status as enabled but with an error. In addition, like any other mutli-process debug supported devices, disable trap temporary setup per-process to avoid performance impact from setup overhead. Signed-off-by: Jonathan Kim Reviewed-by: Felix Kuehling Signed-off-by: Alex Deucher --- drivers/gpu/drm/amd/amdgpu/amdgpu_mes.h | 2 ++ 1 file changed, 2 insertions(+) (limited to 'drivers/gpu/drm/amd/amdgpu/amdgpu_mes.h') diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_mes.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_mes.h index d20df0cf0d88..b5f5eed2b5ef 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_mes.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_mes.h @@ -219,6 +219,8 @@ struct mes_add_queue_input { uint32_t gws_size; uint64_t tba_addr; uint64_t tma_addr; + uint32_t trap_en; + uint32_t skip_process_ctx_clear; uint32_t is_kfd_process; uint32_t is_aql_queue; uint32_t queue_size; -- cgit From 09d49e14ea6fd125a21f89b80f888c09be32a174 Mon Sep 17 00:00:00 2001 From: Jonathan Kim Date: Tue, 23 May 2023 11:57:27 -0400 Subject: drm/amdkfd: fix and enable debugging for gfx11 There are a couple of fixes required to enable gfx11 debugging. First, ADD_QUEUE.trap_en is an inappropriate place to toggle a per-process register so move it to SET_SHADER_DEBUGGER.trap_en. When ADD_QUEUE.skip_process_ctx_clear is set, MES will prioritize the SET_SHADER_DEBUGGER.trap_en setting. Second, to preserve correct save/restore priviledged wave states in coordination with the trap enablement setting, resume suspended waves early in the disable call. Signed-off-by: Jonathan Kim Reviewed-by: Felix Kuehling Signed-off-by: Alex Deucher --- drivers/gpu/drm/amd/amdgpu/amdgpu_mes.h | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) (limited to 'drivers/gpu/drm/amd/amdgpu/amdgpu_mes.h') diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_mes.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_mes.h index b5f5eed2b5ef..2d6ac30b7135 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_mes.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_mes.h @@ -294,6 +294,7 @@ struct mes_misc_op_input { } flags; uint32_t spi_gdbg_per_vmid_cntl; uint32_t tcp_watch_cntl[4]; + uint32_t trap_en; } set_shader_debugger; }; }; @@ -361,7 +362,8 @@ int amdgpu_mes_set_shader_debugger(struct amdgpu_device *adev, uint64_t process_context_addr, uint32_t spi_gdbg_per_vmid_cntl, const uint32_t *tcp_watch_cntl, - uint32_t flags); + uint32_t flags, + bool trap_en); int amdgpu_mes_add_ring(struct amdgpu_device *adev, int gang_id, int queue_type, int idx, -- cgit