[PATCH v2 1/4] drm/xe: Validate wedged_mode parameter and define enum for modes
Michal Wajdeczko
michal.wajdeczko at intel.com
Mon May 12 17:05:54 UTC 2025
On 12.05.2025 18:34, Lukasz Laguna wrote:
> Check correctness of the wedged_mode parameter input to ensure only
> supported values are accepted. Additionally, replace magic numbers with
> a clearly defined enum.
>
> Fixes: 6b8ef44cc0a9 ("drm/xe: Introduce the wedged_mode debugfs")
> Signed-off-by: Lukasz Laguna <lukasz.laguna at intel.com>
> ---
> drivers/gpu/drm/xe/xe_debugfs.c | 5 +++--
> drivers/gpu/drm/xe/xe_device.c | 26 ++++++++++++++++++++++++--
> drivers/gpu/drm/xe/xe_device.h | 1 +
> drivers/gpu/drm/xe/xe_device_types.h | 6 +++++-
> drivers/gpu/drm/xe/xe_guc_ads.c | 4 ++--
> drivers/gpu/drm/xe/xe_guc_capture.c | 3 ++-
> drivers/gpu/drm/xe/xe_module.c | 5 +++--
> drivers/gpu/drm/xe/xe_module.h | 2 +-
> 8 files changed, 41 insertions(+), 11 deletions(-)
>
> diff --git a/drivers/gpu/drm/xe/xe_debugfs.c b/drivers/gpu/drm/xe/xe_debugfs.c
> index d0503959a8ed..7e14eb037295 100644
> --- a/drivers/gpu/drm/xe/xe_debugfs.c
> +++ b/drivers/gpu/drm/xe/xe_debugfs.c
> @@ -163,8 +163,9 @@ static ssize_t wedged_mode_set(struct file *f, const char __user *ubuf,
> if (ret)
> return ret;
>
> - if (wedged_mode > 2)
> - return -EINVAL;
> + ret = xe_device_wedged_mode_validate(xe, wedged_mode);
> + if (ret)
> + return ret;
>
> if (xe->wedged.mode == wedged_mode)
> return size;
> diff --git a/drivers/gpu/drm/xe/xe_device.c b/drivers/gpu/drm/xe/xe_device.c
> index 399ae5f40321..b2d4f1b7d46a 100644
> --- a/drivers/gpu/drm/xe/xe_device.c
> +++ b/drivers/gpu/drm/xe/xe_device.c
> @@ -729,7 +729,10 @@ int xe_device_probe_early(struct xe_device *xe)
> if (err)
> return err;
>
> - xe->wedged.mode = xe_modparam.wedged_mode;
> + xe->wedged.mode = xe_device_wedged_mode_validate(xe, xe_modparam.wedged_mode) ?
> + XE_WEDGED_MODE_FATAL_ERROR /* default */ :
XE_WEDGED_MODE_DEFAULT (see below)
> + xe_modparam.wedged_mode;
> + drm_dbg(&xe->drm, "wedged_mode: setting mode (%u)\n", xe->wedged.mode);
>
> return 0;
> }
> @@ -1143,7 +1146,7 @@ void xe_device_declare_wedged(struct xe_device *xe)
> struct xe_gt *gt;
> u8 id;
>
> - if (xe->wedged.mode == 0) {
> + if (xe->wedged.mode == XE_WEDGED_MODE_DISABLED) {
> drm_dbg(&xe->drm, "Wedged mode is forcibly disabled\n");
> return;
> }
> @@ -1171,3 +1174,22 @@ void xe_device_declare_wedged(struct xe_device *xe)
> for_each_gt(gt, xe, id)
> xe_gt_declare_wedged(gt);
> }
> +
> +/**
> + * xe_device_wedged_mode_validate() - Check if given mode is supported
> + * @xe: the &xe_device
> + * @mode: requested mode to validate
> + *
> + * Check whether the provided wedged mode is supported.
> + *
> + * Return: 0 if mode is supported, error code otherwise.
> + */
> +int xe_device_wedged_mode_validate(struct xe_device *xe, unsigned int mode)
we are trying to name functions using <object> <verb> <subject>, so:
int xe_device_validate_wedged_mode()
or if we don't care about different error codes:
bool xe_device_is_wedged_mode_valid()
> +{
> + if (mode > XE_WEDGED_MODE_FIRST_HANG) {
> + drm_dbg(&xe->drm, "wedged_mode: invalid value (%u)\n", mode);
> + return -EINVAL;
> + }
> +
> + return 0;
> +}
> diff --git a/drivers/gpu/drm/xe/xe_device.h b/drivers/gpu/drm/xe/xe_device.h
> index 0bc3bc8e6803..a22ebf394450 100644
> --- a/drivers/gpu/drm/xe/xe_device.h
> +++ b/drivers/gpu/drm/xe/xe_device.h
> @@ -191,6 +191,7 @@ static inline bool xe_device_wedged(struct xe_device *xe)
> }
>
> void xe_device_declare_wedged(struct xe_device *xe);
> +int xe_device_wedged_mode_validate(struct xe_device *xe, unsigned int mode);
>
> struct xe_file *xe_file_get(struct xe_file *xef);
> void xe_file_put(struct xe_file *xef);
> diff --git a/drivers/gpu/drm/xe/xe_device_types.h b/drivers/gpu/drm/xe/xe_device_types.h
> index 06c65dace026..1bd96981628a 100644
> --- a/drivers/gpu/drm/xe/xe_device_types.h
> +++ b/drivers/gpu/drm/xe/xe_device_types.h
> @@ -556,7 +556,11 @@ struct xe_device {
> /** @wedged.flag: Xe device faced a critical error and is now blocked. */
> atomic_t flag;
> /** @wedged.mode: Mode controlled by kernel parameter and debugfs */
> - int mode;
> + enum xe_wedged_mode {
likely it can be nameless enum as we don't use it elsewhere (unless we
move it to xe_module.h near actual modparam, where we usually want to
describe module parameters)
> + XE_WEDGED_MODE_DISABLED = 0,
> + XE_WEDGED_MODE_FATAL_ERROR = 1,
> + XE_WEDGED_MODE_FIRST_HANG = 2,
XE_WEDGED_MODE_NEVER = 0,
XE_WEDGED_MODE_UPON_CRITICAL_ERROR = 1,
XE_WEDGED_MODE_UPON_ANY_HANG = 2,
and
XE_WEDGED_MODE_DEFAULT = XE_WEDGED_MODE_UPON_CRITICAL_ERROR,
to match modparam description?
> + } mode;
> } wedged;
>
> /** @bo_device: Struct to control async free of BOs */
> diff --git a/drivers/gpu/drm/xe/xe_guc_ads.c b/drivers/gpu/drm/xe/xe_guc_ads.c
> index 44c1fa2fe7c8..d9b7fff1d2a7 100644
> --- a/drivers/gpu/drm/xe/xe_guc_ads.c
> +++ b/drivers/gpu/drm/xe/xe_guc_ads.c
> @@ -469,7 +469,7 @@ static void guc_policies_init(struct xe_guc_ads *ads)
> ads_blob_write(ads, policies.max_num_work_items,
> GLOBAL_POLICY_MAX_NUM_WI);
>
> - if (xe->wedged.mode == 2)
> + if (xe->wedged.mode == XE_WEDGED_MODE_FIRST_HANG)
> global_flags |= GLOBAL_POLICY_DISABLE_ENGINE_RESET;
>
> ads_blob_write(ads, policies.global_flags, global_flags);
> @@ -1018,7 +1018,7 @@ int xe_guc_ads_scheduler_policy_toggle_reset(struct xe_guc_ads *ads)
> policies->dpc_promote_time = ads_blob_read(ads, policies.dpc_promote_time);
> policies->max_num_work_items = ads_blob_read(ads, policies.max_num_work_items);
> policies->is_valid = 1;
> - if (xe->wedged.mode == 2)
> + if (xe->wedged.mode == XE_WEDGED_MODE_FIRST_HANG)
> policies->global_flags |= GLOBAL_POLICY_DISABLE_ENGINE_RESET;
> else
> policies->global_flags &= ~GLOBAL_POLICY_DISABLE_ENGINE_RESET;
> diff --git a/drivers/gpu/drm/xe/xe_guc_capture.c b/drivers/gpu/drm/xe/xe_guc_capture.c
> index 859a3ba91be5..c592b4f294cb 100644
> --- a/drivers/gpu/drm/xe/xe_guc_capture.c
> +++ b/drivers/gpu/drm/xe/xe_guc_capture.c
> @@ -1856,7 +1856,8 @@ xe_guc_capture_get_matching_and_lock(struct xe_exec_queue *q)
> return NULL;
>
> xe = gt_to_xe(q->gt);
> - if (xe->wedged.mode >= 2 || !xe_device_uc_enabled(xe) || IS_SRIOV_VF(xe))
> + if (xe->wedged.mode == XE_WEDGED_MODE_FIRST_HANG || !xe_device_uc_enabled(xe) ||
> + IS_SRIOV_VF(xe))
> return NULL;
>
> ss = &xe->devcoredump.snapshot;
> diff --git a/drivers/gpu/drm/xe/xe_module.c b/drivers/gpu/drm/xe/xe_module.c
> index 05c7d0ae6d83..fb41bf76d48c 100644
> --- a/drivers/gpu/drm/xe/xe_module.c
> +++ b/drivers/gpu/drm/xe/xe_module.c
> @@ -10,6 +10,7 @@
>
> #include <drm/drm_module.h>
>
> +#include "xe_device.h"
> #include "xe_drv.h"
> #include "xe_configfs.h"
> #include "xe_hw_fence.h"
> @@ -25,7 +26,7 @@ struct xe_modparam xe_modparam = {
> #ifdef CONFIG_PCI_IOV
> .max_vfs = IS_ENABLED(CONFIG_DRM_XE_DEBUG) ? ~0 : 0,
> #endif
> - .wedged_mode = 1,
> + .wedged_mode = XE_WEDGED_MODE_FATAL_ERROR,
> .svm_notifier_size = 512,
> /* the rest are 0 by default */
> };
> @@ -71,7 +72,7 @@ MODULE_PARM_DESC(max_vfs,
> "(0 = no VFs [default]; N = allow up to N VFs)");
> #endif
>
> -module_param_named_unsafe(wedged_mode, xe_modparam.wedged_mode, int, 0600);
> +module_param_named_unsafe(wedged_mode, xe_modparam.wedged_mode, uint, 0600);
> MODULE_PARM_DESC(wedged_mode,
> "Module's default policy for the wedged mode - 0=never, 1=upon-critical-errors[default], 2=upon-any-hang");
>
> diff --git a/drivers/gpu/drm/xe/xe_module.h b/drivers/gpu/drm/xe/xe_module.h
> index 84339e509c80..fb674bc08955 100644
> --- a/drivers/gpu/drm/xe/xe_module.h
> +++ b/drivers/gpu/drm/xe/xe_module.h
> @@ -22,7 +22,7 @@ struct xe_modparam {
> #ifdef CONFIG_PCI_IOV
> unsigned int max_vfs;
> #endif
> - int wedged_mode;
> + unsigned int wedged_mode;
> u32 svm_notifier_size;
> };
>
More information about the Intel-xe
mailing list