Re: [PATCH] habanalabs: add print upon clock change

[Oded Gabbay]

On Mon, Mar 9, 2020 at 1:25 PM Omer Shpigelman <oshpigelman@xxxxxxxxx> wrote:
>
> Add print upon clock slow down due to power consumption or overheating.
> In addition, add print when back to optimal clock.
>
> Signed-off-by: Omer Shpigelman <oshpigelman@xxxxxxxxx>
> ---
>  drivers/misc/habanalabs/goya/goya.c           | 50 ++++++++++++++++++-
>  .../include/goya/goya_async_events.h          |  4 ++
>  2 files changed, 53 insertions(+), 1 deletion(-)
>
> diff --git a/drivers/misc/habanalabs/goya/goya.c b/drivers/misc/habanalabs/goya/goya.c
> index 9d4295cc83cf..1ccf2ed9a8dc 100644
> --- a/drivers/misc/habanalabs/goya/goya.c
> +++ b/drivers/misc/habanalabs/goya/goya.c
> @@ -324,7 +324,11 @@ static u32 goya_all_events[] = {
>         GOYA_ASYNC_EVENT_ID_DMA_BM_CH1,
>         GOYA_ASYNC_EVENT_ID_DMA_BM_CH2,
>         GOYA_ASYNC_EVENT_ID_DMA_BM_CH3,
> -       GOYA_ASYNC_EVENT_ID_DMA_BM_CH4
> +       GOYA_ASYNC_EVENT_ID_DMA_BM_CH4,
> +       GOYA_ASYNC_EVENT_ID_POWER_ENV_S,
> +       GOYA_ASYNC_EVENT_ID_POWER_ENV_E,
> +       GOYA_ASYNC_EVENT_ID_THERMAL_ENV_S,
> +       GOYA_ASYNC_EVENT_ID_THERMAL_ENV_E
>  };
>
>  static int goya_mmu_clear_pgt_range(struct hl_device *hdev);
> @@ -4389,6 +4393,14 @@ static const char *_goya_get_event_desc(u16 event_type)
>                 return "TPC%d_bmon_spmu";
>         case GOYA_ASYNC_EVENT_ID_DMA_BM_CH0 ... GOYA_ASYNC_EVENT_ID_DMA_BM_CH4:
>                 return "DMA_bm_ch%d";
> +       case GOYA_ASYNC_EVENT_ID_POWER_ENV_S:
> +               return "POWER_ENV_S";
> +       case GOYA_ASYNC_EVENT_ID_POWER_ENV_E:
> +               return "POWER_ENV_E";
> +       case GOYA_ASYNC_EVENT_ID_THERMAL_ENV_S:
> +               return "THERMAL_ENV_S";
> +       case GOYA_ASYNC_EVENT_ID_THERMAL_ENV_E:
> +               return "THERMAL_ENV_E";
>         default:
>                 return "N/A";
>         }
> @@ -4619,6 +4631,33 @@ static int goya_unmask_irq(struct hl_device *hdev, u16 event_type)
>         return rc;
>  }
>
> +static void goya_print_clk_change_info(struct hl_device *hdev, u16 event_type)
> +{
> +       switch (event_type) {
> +       case GOYA_ASYNC_EVENT_ID_POWER_ENV_S:
> +               dev_info_ratelimited(hdev->dev,
> +                       "Clock throttling due to power consumption\n");
> +               break;
> +       case GOYA_ASYNC_EVENT_ID_POWER_ENV_E:
> +               dev_info_ratelimited(hdev->dev,
> +                       "Power envelop is safe, back to optimal clock\n");
> +               break;
> +       case GOYA_ASYNC_EVENT_ID_THERMAL_ENV_S:
> +               dev_info_ratelimited(hdev->dev,
> +                       "Clock throttling due to overheating\n");
> +               break;
> +       case GOYA_ASYNC_EVENT_ID_THERMAL_ENV_E:
> +               dev_info_ratelimited(hdev->dev,
> +                       "Thermal envelop is safe, back to optimal clock\n");
> +               break;
> +
> +       default:
> +               dev_err(hdev->dev, "Received invalid clock change event %d\n",
> +                       event_type);
> +               break;
> +       }
> +}
> +
>  void goya_handle_eqe(struct hl_device *hdev, struct hl_eq_entry *eq_entry)
>  {
>         u32 ctl = le32_to_cpu(eq_entry->hdr.ctl);
> @@ -4702,6 +4741,15 @@ void goya_handle_eqe(struct hl_device *hdev, struct hl_eq_entry *eq_entry)
>                 goya_unmask_irq(hdev, event_type);
>                 break;
>
> +       case GOYA_ASYNC_EVENT_ID_POWER_ENV_S:
> +       case GOYA_ASYNC_EVENT_ID_POWER_ENV_E:
> +       case GOYA_ASYNC_EVENT_ID_THERMAL_ENV_S:
> +       case GOYA_ASYNC_EVENT_ID_THERMAL_ENV_E:
> +               goya_print_irq_info(hdev, event_type, false);
> +               goya_print_clk_change_info(hdev, event_type);
> +               goya_unmask_irq(hdev, event_type);
> +               break;
> +
>         default:
>                 dev_err(hdev->dev, "Received invalid H/W interrupt %d\n",
>                                 event_type);
> diff --git a/drivers/misc/habanalabs/include/goya/goya_async_events.h b/drivers/misc/habanalabs/include/goya/goya_async_events.h
> index bb7a1aa3279e..6be41a846c99 100644
> --- a/drivers/misc/habanalabs/include/goya/goya_async_events.h
> +++ b/drivers/misc/habanalabs/include/goya/goya_async_events.h
> @@ -188,6 +188,10 @@ enum goya_async_event_id {
>         GOYA_ASYNC_EVENT_ID_HALT_MACHINE = 485,
>         GOYA_ASYNC_EVENT_ID_INTS_REGISTER = 486,
>         GOYA_ASYNC_EVENT_ID_SOFT_RESET = 487,
> +       GOYA_ASYNC_EVENT_ID_POWER_ENV_S = 507,
> +       GOYA_ASYNC_EVENT_ID_POWER_ENV_E = 508,
> +       GOYA_ASYNC_EVENT_ID_THERMAL_ENV_S = 509,
> +       GOYA_ASYNC_EVENT_ID_THERMAL_ENV_E = 510,
>         GOYA_ASYNC_EVENT_ID_LAST_VALID_ID = 1023,
>         GOYA_ASYNC_EVENT_ID_SIZE
>  };
> --
> 2.17.1
>

This patch is:
Reviewed-by: Oded Gabbay <oded.gabbay@xxxxxxxxx>