[PATCH 2/2] drm/amd/powerplay: shutdown on HW CTF

[Alex Deucher]

On Sat, May 9, 2020 at 3:47 AM Evan Quan <evan.quan at amd.com> wrote:
>
> To prevent further damage.
>
> Change-Id: Iae0399001694bb7446bcc6071cd03d8fa47c9d02
> Signed-off-by: Evan Quan <evan.quan at amd.com>


Series is:
Reviewed-by: Alex Deucher <alexander.deucher at amd.com>

> ---
>  .../gpu/drm/amd/powerplay/hwmgr/smu_helper.c  | 16 +++++++++++++--
>  drivers/gpu/drm/amd/powerplay/smu_v11_0.c     | 20 ++++++++++++++++++-
>  2 files changed, 33 insertions(+), 3 deletions(-)
>
> diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/smu_helper.c b/drivers/gpu/drm/amd/powerplay/hwmgr/smu_helper.c
> index 782f6d295202..4279f95ba779 100644
> --- a/drivers/gpu/drm/amd/powerplay/hwmgr/smu_helper.c
> +++ b/drivers/gpu/drm/amd/powerplay/hwmgr/smu_helper.c
> @@ -612,11 +612,17 @@ int phm_irq_process(struct amdgpu_device *adev,
>                                         PCI_BUS_NUM(adev->pdev->devfn),
>                                         PCI_SLOT(adev->pdev->devfn),
>                                         PCI_FUNC(adev->pdev->devfn));
> -               else if (src_id == VISLANDS30_IV_SRCID_GPIO_19)
> +               else if (src_id == VISLANDS30_IV_SRCID_GPIO_19) {
>                         pr_warn("GPU Critical Temperature Fault detected on PCIe %d:%d.%d!\n",
>                                         PCI_BUS_NUM(adev->pdev->devfn),
>                                         PCI_SLOT(adev->pdev->devfn),
>                                         PCI_FUNC(adev->pdev->devfn));
> +                       /*
> +                        * HW CTF just occurred. Shutdown to prevent further damage.
> +                        */
> +                       dev_emerg(adev->dev, "System is going to shutdown due to HW CTF!\n");
> +                       orderly_poweroff(true);
> +               }
>         } else if (client_id == SOC15_IH_CLIENTID_THM) {
>                 if (src_id == 0) {
>                         pr_warn("GPU over temperature range detected on PCIe %d:%d.%d!\n",
> @@ -634,11 +640,17 @@ int phm_irq_process(struct amdgpu_device *adev,
>                                         PCI_BUS_NUM(adev->pdev->devfn),
>                                         PCI_SLOT(adev->pdev->devfn),
>                                         PCI_FUNC(adev->pdev->devfn));
> -       } else if (client_id == SOC15_IH_CLIENTID_ROM_SMUIO)
> +       } else if (client_id == SOC15_IH_CLIENTID_ROM_SMUIO) {
>                 pr_warn("GPU Critical Temperature Fault detected on PCIe %d:%d.%d!\n",
>                                 PCI_BUS_NUM(adev->pdev->devfn),
>                                 PCI_SLOT(adev->pdev->devfn),
>                                 PCI_FUNC(adev->pdev->devfn));
> +               /*
> +                * HW CTF just occurred. Shutdown to prevent further damage.
> +                */
> +               dev_emerg(adev->dev, "System is going to shutdown due to HW CTF!\n");
> +               orderly_poweroff(true);
> +       }
>
>         return 0;
>  }
> diff --git a/drivers/gpu/drm/amd/powerplay/smu_v11_0.c b/drivers/gpu/drm/amd/powerplay/smu_v11_0.c
> index 5b493f140dda..041aac272a9e 100644
> --- a/drivers/gpu/drm/amd/powerplay/smu_v11_0.c
> +++ b/drivers/gpu/drm/amd/powerplay/smu_v11_0.c
> @@ -1517,6 +1517,8 @@ static int smu_v11_0_ack_ac_dc_interrupt(struct smu_context *smu)
>  #define THM_11_0__SRCID__THM_DIG_THERM_L2H             0               /* ASIC_TEMP > CG_THERMAL_INT.DIG_THERM_INTH  */
>  #define THM_11_0__SRCID__THM_DIG_THERM_H2L             1               /* ASIC_TEMP < CG_THERMAL_INT.DIG_THERM_INTL  */
>
> +#define SMUIO_11_0__SRCID__SMUIO_GPIO19                        83
> +
>  static int smu_v11_0_irq_process(struct amdgpu_device *adev,
>                                  struct amdgpu_irq_src *source,
>                                  struct amdgpu_iv_entry *entry)
> @@ -1556,8 +1558,17 @@ static int smu_v11_0_irq_process(struct amdgpu_device *adev,
>                                 PCI_SLOT(adev->pdev->devfn),
>                                 PCI_FUNC(adev->pdev->devfn));
>                 break;
> -
>                 }
> +       } else if (client_id == SOC15_IH_CLIENTID_ROM_SMUIO) {
> +               pr_warn("GPU Critical Temperature Fault detected on PCIe %d:%d.%d!\n",
> +                               PCI_BUS_NUM(adev->pdev->devfn),
> +                               PCI_SLOT(adev->pdev->devfn),
> +                               PCI_FUNC(adev->pdev->devfn));
> +               /*
> +                * HW CTF just occurred. Shutdown to prevent further damage.
> +                */
> +               dev_emerg(adev->dev, "System is going to shutdown due to HW CTF!\n");
> +               orderly_poweroff(true);
>         } else if (client_id == SOC15_IH_CLIENTID_MP1) {
>                 if (src_id == 0xfe) {
>                         switch (ctxid) {
> @@ -1610,6 +1621,13 @@ int smu_v11_0_register_irq_handler(struct smu_context *smu)
>         if (ret)
>                 return ret;
>
> +       /* Register CTF(GPIO_19) interrupt */
> +       ret = amdgpu_irq_add_id(adev, SOC15_IH_CLIENTID_ROM_SMUIO,
> +                               SMUIO_11_0__SRCID__SMUIO_GPIO19,
> +                               irq_src);
> +       if (ret)
> +               return ret;
> +
>         ret = amdgpu_irq_add_id(adev, SOC15_IH_CLIENTID_MP1,
>                                 0xfe,
>                                 irq_src);
> --
> 2.26.2
>
> _______________________________________________
> amd-gfx mailing list
> amd-gfx at lists.freedesktop.org
> https://lists.freedesktop.org/mailman/listinfo/amd-gfx