Ping?
On Mon, Apr 8, 2024 at 3:02 PM Alex Deucher <alexander.deucher@xxxxxxx> wrote:
>
> Makes it easier to review the logs when there are MES
> errors.
>
> v2: use dbg for emitted, add helpers for fetching strings
>
> Signed-off-by: Alex Deucher <alexander.deucher@xxxxxxx>
> ---
> drivers/gpu/drm/amd/amdgpu/mes_v11_0.c | 78 ++++++++++++++++++++++++--
> 1 file changed, 74 insertions(+), 4 deletions(-)
>
> diff --git a/drivers/gpu/drm/amd/amdgpu/mes_v11_0.c b/drivers/gpu/drm/amd/amdgpu/mes_v11_0.c
> index 072c478665ade..69d39ba726e12 100644
> --- a/drivers/gpu/drm/amd/amdgpu/mes_v11_0.c
> +++ b/drivers/gpu/drm/amd/amdgpu/mes_v11_0.c
> @@ -100,18 +100,72 @@ static const struct amdgpu_ring_funcs mes_v11_0_ring_funcs = {
> .insert_nop = amdgpu_ring_insert_nop,
> };
>
> +static const char *mes_v11_0_opcodes[] = {
> + "MES_SCH_API_SET_HW_RSRC",
> + "MES_SCH_API_SET_SCHEDULING_CONFIG",
> + "MES_SCH_API_ADD_QUEUE"
> + "MES_SCH_API_REMOVE_QUEUE"
> + "MES_SCH_API_PERFORM_YIELD"
> + "MES_SCH_API_SET_GANG_PRIORITY_LEVEL"
> + "MES_SCH_API_SUSPEND"
> + "MES_SCH_API_RESUME"
> + "MES_SCH_API_RESET"
> + "MES_SCH_API_SET_LOG_BUFFER"
> + "MES_SCH_API_CHANGE_GANG_PRORITY"
> + "MES_SCH_API_QUERY_SCHEDULER_STATUS"
> + "MES_SCH_API_PROGRAM_GDS"
> + "MES_SCH_API_SET_DEBUG_VMID"
> + "MES_SCH_API_MISC"
> + "MES_SCH_API_UPDATE_ROOT_PAGE_TABLE"
> + "MES_SCH_API_AMD_LOG"
> +};
> +
> +static const char *mes_v11_0_misc_opcodes[] = {
> + "MESAPI_MISC__WRITE_REG",
> + "MESAPI_MISC__INV_GART",
> + "MESAPI_MISC__QUERY_STATUS",
> + "MESAPI_MISC__READ_REG",
> + "MESAPI_MISC__WAIT_REG_MEM",
> + "MESAPI_MISC__SET_SHADER_DEBUGGER",
> +};
> +
> +static const char *mes_v11_0_get_op_string(union MESAPI__MISC *x_pkt)
> +{
> + const char *op_str = NULL;
> +
> + if (x_pkt->header.opcode < ARRAY_SIZE(mes_v11_0_opcodes))
> + op_str = mes_v11_0_opcodes[x_pkt->header.opcode];
> +
> + return op_str;
> +}
> +
> +static const char *mes_v11_0_get_misc_op_string(union MESAPI__MISC *x_pkt)
> +{
> + const char *op_str = NULL;
> +
> + if ((x_pkt->header.opcode == MES_SCH_API_MISC) &&
> + (x_pkt->opcode <= ARRAY_SIZE(mes_v11_0_misc_opcodes)))
> + op_str = mes_v11_0_misc_opcodes[x_pkt->opcode];
> +
> + return op_str;
> +}
> +
> static int mes_v11_0_submit_pkt_and_poll_completion(struct amdgpu_mes *mes,
> void *pkt, int size,
> int api_status_off)
> {
> int ndw = size / 4;
> signed long r;
> - union MESAPI__ADD_QUEUE *x_pkt = pkt;
> + union MESAPI__MISC *x_pkt = pkt;
> struct MES_API_STATUS *api_status;
> struct amdgpu_device *adev = mes->adev;
> struct amdgpu_ring *ring = &mes->ring;
> unsigned long flags;
> signed long timeout = adev->usec_timeout;
> + const char *op_str, *misc_op_str;
> +
> + if (x_pkt->header.opcode >= MES_SCH_API_MAX)
> + return -EINVAL;
>
> if (amdgpu_emu_mode) {
> timeout *= 100;
> @@ -135,13 +189,29 @@ static int mes_v11_0_submit_pkt_and_poll_completion(struct amdgpu_mes *mes,
> amdgpu_ring_commit(ring);
> spin_unlock_irqrestore(&mes->ring_lock, flags);
>
> - DRM_DEBUG("MES msg=%d was emitted\n", x_pkt->header.opcode);
> + op_str = mes_v11_0_get_op_string(x_pkt);
> + misc_op_str = mes_v11_0_get_misc_op_string(x_pkt);
> +
> + if (misc_op_str)
> + dev_dbg(adev->dev, "MES msg=%s (%s) was emitted\n", op_str, misc_op_str);
> + else if (op_str)
> + dev_dbg(adev->dev, "MES msg=%s was emitted\n", op_str);
> + else
> + dev_dbg(adev->dev, "MES msg=%d was emitted\n", x_pkt->header.opcode);
>
> r = amdgpu_fence_wait_polling(ring, ring->fence_drv.sync_seq,
> timeout);
> if (r < 1) {
> - DRM_ERROR("MES failed to response msg=%d\n",
> - x_pkt->header.opcode);
> +
> + if (misc_op_str)
> + dev_err(adev->dev, "MES failed to respond to msg=%s (%s)\n",
> + op_str, misc_op_str);
> + else if (op_str)
> + dev_err(adev->dev, "MES failed to respond to msg=%s\n",
> + op_str);
> + else
> + dev_err(adev->dev, "MES failed to respond to msg=%d\n",
> + x_pkt->header.opcode);
>
> while (halt_if_hws_hang)
> schedule();
> --
> 2.44.0
>
