Intel-XE Archive on lore.kernel.org
 help / color / mirror / Atom feed
From: "Summers, Stuart" <stuart.summers@intel.com>
To: "intel-xe@lists.freedesktop.org" <intel-xe@lists.freedesktop.org>,
	"Wajdeczko, Michal" <Michal.Wajdeczko@intel.com>
Cc: "Tauro, Riana" <riana.tauro@intel.com>,
	"De Marchi, Lucas" <lucas.demarchi@intel.com>
Subject: Re: [PATCH 3/3] drm/xe/configfs: Don't expose survivability_mode if not applicable
Date: Tue, 2 Sep 2025 17:25:25 +0000	[thread overview]
Message-ID: <cb4d9b4ea275b9d8dfad37a7e078cea8477e9bda.camel@intel.com> (raw)
In-Reply-To: <20250902131744.5076-4-michal.wajdeczko@intel.com>

On Tue, 2025-09-02 at 15:17 +0200, Michal Wajdeczko wrote:
> The survivability_mode attribute is applicable only for DGFX and
> platforms newer than BATTLEMAGE. Use .is_visible() hook to hide
> this attribute when above conditions are not met. Remove code that
> was trying to fix such configuration during the runtime.
> 
> Signed-off-by: Michal Wajdeczko <michal.wajdeczko@intel.com>
> Cc: Lucas De Marchi <lucas.demarchi@intel.com>
> Cc: Riana Tauro <riana.tauro@intel.com>
> ---
>  drivers/gpu/drm/xe/xe_configfs.c           | 24 ++++++--------------
> --
>  drivers/gpu/drm/xe/xe_configfs.h           |  2 --
>  drivers/gpu/drm/xe/xe_survivability_mode.c | 11 +---------
>  3 files changed, 7 insertions(+), 30 deletions(-)
> 
> diff --git a/drivers/gpu/drm/xe/xe_configfs.c
> b/drivers/gpu/drm/xe/xe_configfs.c
> index 43f000260752..0337811864cd 100644
> --- a/drivers/gpu/drm/xe/xe_configfs.c
> +++ b/drivers/gpu/drm/xe/xe_configfs.c
> @@ -369,7 +369,12 @@ static bool xe_config_device_is_visible(struct
> config_item *item,
>  {
>         struct xe_config_group_device *dev =
> to_xe_config_group_device(item);
>  
> -       return dev->desc; /* shall be always true */
> +       if (attr == &attr_survivability_mode) {
> +               if (!dev->desc->is_dgfx || dev->desc->platform <
> XE_BATTLEMAGE)
> +                       return false;
> +       }
> +
> +       return true;

Why change the return here? Can we either leave this as dev->desc or
otherwise use return true for the initial implementation (previous
patch)?

Everything else in the series looks good to me. This does seem like a
better way to approach this.

Thanks,
Stuart

>  }
>  
>  static struct configfs_group_operations xe_config_device_group_ops =
> {
> @@ -558,23 +563,6 @@ bool xe_configfs_get_survivability_mode(struct
> pci_dev *pdev)
>         return mode;
>  }
>  
> -/**
> - * xe_configfs_clear_survivability_mode - clear configfs
> survivability mode
> - * @pdev: pci device
> - */
> -void xe_configfs_clear_survivability_mode(struct pci_dev *pdev)
> -{
> -       struct xe_config_group_device *dev =
> find_xe_config_group_device(pdev);
> -
> -       if (!dev)
> -               return;
> -
> -       guard(mutex)(&dev->lock);
> -       dev->config.survivability_mode = 0;
> -
> -       config_group_put(&dev->group);
> -}
> -
>  /**
>   * xe_configfs_get_engines_allowed - get engine allowed mask from
> configfs
>   * @pdev: pci device
> diff --git a/drivers/gpu/drm/xe/xe_configfs.h
> b/drivers/gpu/drm/xe/xe_configfs.h
> index 58c8c3164000..1402e863b71c 100644
> --- a/drivers/gpu/drm/xe/xe_configfs.h
> +++ b/drivers/gpu/drm/xe/xe_configfs.h
> @@ -15,7 +15,6 @@ int xe_configfs_init(void);
>  void xe_configfs_exit(void);
>  void xe_configfs_check_device(struct pci_dev *pdev);
>  bool xe_configfs_get_survivability_mode(struct pci_dev *pdev);
> -void xe_configfs_clear_survivability_mode(struct pci_dev *pdev);
>  u64 xe_configfs_get_engines_allowed(struct pci_dev *pdev);
>  bool xe_configfs_get_psmi_enabled(struct pci_dev *pdev);
>  #else
> @@ -23,7 +22,6 @@ static inline int xe_configfs_init(void) { return
> 0; }
>  static inline void xe_configfs_exit(void) { }
>  static inline void xe_configfs_check_device(struct pci_dev *pdev) {
> }
>  static inline bool xe_configfs_get_survivability_mode(struct pci_dev
> *pdev) { return false; }
> -static inline void xe_configfs_clear_survivability_mode(struct
> pci_dev *pdev) { }
>  static inline u64 xe_configfs_get_engines_allowed(struct pci_dev
> *pdev) { return U64_MAX; }
>  static inline bool xe_configfs_get_psmi_enabled(struct pci_dev
> *pdev) { return false; }
>  #endif
> diff --git a/drivers/gpu/drm/xe/xe_survivability_mode.c
> b/drivers/gpu/drm/xe/xe_survivability_mode.c
> index 79426ea46861..19a1732e33d4 100644
> --- a/drivers/gpu/drm/xe/xe_survivability_mode.c
> +++ b/drivers/gpu/drm/xe/xe_survivability_mode.c
> @@ -287,19 +287,10 @@ bool xe_survivability_mode_is_requested(struct
> xe_device *xe)
>         u32 data;
>         bool survivability_mode;
>  
> -       if (!IS_DGFX(xe) || IS_SRIOV_VF(xe))
> +       if (!IS_DGFX(xe) || IS_SRIOV_VF(xe) || xe->info.platform <
> XE_BATTLEMAGE)
>                 return false;
>  
>         survivability_mode =
> xe_configfs_get_survivability_mode(pdev);
> -
> -       if (xe->info.platform < XE_BATTLEMAGE) {
> -               if (survivability_mode) {
> -                       dev_err(&pdev->dev, "Survivability Mode is
> not supported on this card\n");
> -                       xe_configfs_clear_survivability_mode(pdev);
> -               }
> -               return false;
> -       }
> -
>         /* Enable survivability mode if set via configfs */
>         if (survivability_mode)
>                 return true;


  reply	other threads:[~2025-09-02 17:25 UTC|newest]

Thread overview: 24+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2025-09-02 13:17 [PATCH 0/3] Allow to filter-out some configfs attrs Michal Wajdeczko
2025-09-02 13:17 ` [PATCH 1/3] drm/xe/configfs: Don't touch survivability_mode on fini Michal Wajdeczko
2025-09-02 16:37   ` Summers, Stuart
2025-09-02 18:04   ` Lucas De Marchi
2025-09-03  5:39     ` Riana Tauro
2025-09-03  5:49       ` Michal Wajdeczko
2025-09-03 17:50         ` Lucas De Marchi
2025-09-03 19:47           ` Michal Wajdeczko
2025-09-04  4:36             ` Riana Tauro
2025-09-04 10:35   ` [PATCH v2 " Michal Wajdeczko
2025-09-04 15:56     ` Lucas De Marchi
2025-09-09  5:16     ` Riana Tauro
2025-09-02 13:17 ` [PATCH 2/3] drm/xe/configfs: Prepare to filter-out configfs attributes Michal Wajdeczko
2025-09-02 13:17 ` [PATCH 3/3] drm/xe/configfs: Don't expose survivability_mode if not applicable Michal Wajdeczko
2025-09-02 17:25   ` Summers, Stuart [this message]
2025-09-02 18:16     ` Michal Wajdeczko
2025-09-02 18:19       ` Summers, Stuart
2025-09-02 18:10   ` Lucas De Marchi
2025-09-02 15:19 ` ✓ CI.KUnit: success for Allow to filter-out some configfs attrs Patchwork
2025-09-02 15:56 ` ✓ Xe.CI.BAT: " Patchwork
2025-09-02 20:41 ` ✗ Xe.CI.Full: failure " Patchwork
2025-09-04 10:54 ` ✓ CI.KUnit: success for Allow to filter-out some configfs attrs (rev2) Patchwork
2025-09-04 11:30 ` ✓ Xe.CI.BAT: " Patchwork
2025-09-04 23:14 ` ✗ Xe.CI.Full: failure " Patchwork

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=cb4d9b4ea275b9d8dfad37a7e078cea8477e9bda.camel@intel.com \
    --to=stuart.summers@intel.com \
    --cc=Michal.Wajdeczko@intel.com \
    --cc=intel-xe@lists.freedesktop.org \
    --cc=lucas.demarchi@intel.com \
    --cc=riana.tauro@intel.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox