intel-xe.lists.freedesktop.org archive mirror
 help / color / mirror / Atom feed
From: "Belgaumkar, Vinay" <vinay.belgaumkar@intel.com>
To: Lucas De Marchi <lucas.demarchi@intel.com>,
	<intel-xe@lists.freedesktop.org>
Cc: <prashanth.kumar@intel.com>, <dnyaneshwar.bhadane@intel.com>,
	Matt Roper <matthew.d.roper@intel.com>,
	Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>,
	John Harrison <John.C.Harrison@Intel.com>
Subject: Re: [PATCH v3 01/13] drm/xe/psmi: Add GuC flag to enable PSMI
Date: Tue, 12 Aug 2025 17:38:14 -0700	[thread overview]
Message-ID: <1749abbd-5b10-4684-b1fb-c340fe57a215@intel.com> (raw)
In-Reply-To: <20250808-psmi-v3-1-a111e9f1e4b7@intel.com>


On 8/8/2025 10:29 AM, Lucas De Marchi wrote:
> PSMI allows to capture data from the GPU useful for early
> validation. From the kernel side there isn't much to be done, just a few
> things:
>
> 	1) Toggle the feature support in GuC
> 	2) Enable some additional WAs
> 	3) Allocate buffers
>
> Here is the first step, with the next ones to follow. For now everything
> is disabled through a check in configfs that is currently hardcoded to
> disabled.
>
> Cc: Matt Roper <matthew.d.roper@intel.com>
> Cc: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
> Cc: John Harrison <John.C.Harrison@Intel.com>
> Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com>
> ---
>   drivers/gpu/drm/xe/xe_configfs.h | 2 ++
>   drivers/gpu/drm/xe/xe_guc.c      | 7 ++++++-
>   drivers/gpu/drm/xe/xe_guc_fwif.h | 1 +
>   3 files changed, 9 insertions(+), 1 deletion(-)
>
> diff --git a/drivers/gpu/drm/xe/xe_configfs.h b/drivers/gpu/drm/xe/xe_configfs.h
> index fb87640080896..c14588b86e833 100644
> --- a/drivers/gpu/drm/xe/xe_configfs.h
> +++ b/drivers/gpu/drm/xe/xe_configfs.h
> @@ -16,12 +16,14 @@ void xe_configfs_exit(void);
>   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);
> +static inline bool xe_configfs_get_psmi_enabled(struct pci_dev *pdev) { return false; }
>   #else
>   static inline int xe_configfs_init(void) { return 0; }
>   static inline void xe_configfs_exit(void) { }
>   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
>   
>   #endif
> diff --git a/drivers/gpu/drm/xe/xe_guc.c b/drivers/gpu/drm/xe/xe_guc.c
> index 9e34401e4489f..cb757a53de856 100644
> --- a/drivers/gpu/drm/xe/xe_guc.c
> +++ b/drivers/gpu/drm/xe/xe_guc.c
> @@ -16,6 +16,7 @@
>   #include "regs/xe_guc_regs.h"
>   #include "regs/xe_irq_regs.h"
>   #include "xe_bo.h"
> +#include "xe_configfs.h"
>   #include "xe_device.h"
>   #include "xe_force_wake.h"
>   #include "xe_gt.h"
> @@ -81,11 +82,15 @@ static u32 guc_ctl_debug_flags(struct xe_guc *guc)
>   
>   static u32 guc_ctl_feature_flags(struct xe_guc *guc)
>   {
> +	struct xe_device *xe = guc_to_xe(guc);
>   	u32 flags = GUC_CTL_ENABLE_LITE_RESTORE;
>   
> -	if (!guc_to_xe(guc)->info.skip_guc_pc)
> +	if (!xe->info.skip_guc_pc)
>   		flags |= GUC_CTL_ENABLE_SLPC;
>   
> +	if (xe_configfs_get_psmi_enabled(to_pci_dev(xe->drm.dev)))
> +		flags |= GUC_CTL_ENABLE_PSMI;
> +
>   	return flags;
>   }
>   
> diff --git a/drivers/gpu/drm/xe/xe_guc_fwif.h b/drivers/gpu/drm/xe/xe_guc_fwif.h
> index ca9f999d38d1e..4dc000c977faf 100644
> --- a/drivers/gpu/drm/xe/xe_guc_fwif.h
> +++ b/drivers/gpu/drm/xe/xe_guc_fwif.h
> @@ -112,6 +112,7 @@ struct guc_update_exec_queue_policy {
>   #define GUC_CTL_FEATURE			2
>   #define   GUC_CTL_ENABLE_SLPC		BIT(2)
>   #define   GUC_CTL_ENABLE_LITE_RESTORE	BIT(4)
> +#define   GUC_CTL_ENABLE_PSMI		BIT(7)

Should we have this as GUC_CTL_ENABLE_PSMI_LOGGING to match the GuC 
nomenclature?

Thanks,

Vinay.

>   #define   GUC_CTL_DISABLE_SCHEDULER	BIT(14)
>   
>   #define GUC_CTL_DEBUG			3
>

  reply	other threads:[~2025-08-13  0:38 UTC|newest]

Thread overview: 33+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2025-08-08 17:29 [PATCH v3 00/13] drm/xe: Add psmi support Lucas De Marchi
2025-08-08 17:29 ` [PATCH v3 01/13] drm/xe/psmi: Add GuC flag to enable PSMI Lucas De Marchi
2025-08-13  0:38   ` Belgaumkar, Vinay [this message]
2025-08-15 21:34     ` Lucas De Marchi
2025-08-08 17:29 ` [PATCH v3 02/13] drm/xe/psmi: Add debugfs interface for PSMI Lucas De Marchi
2025-08-13  1:41   ` Belgaumkar, Vinay
2025-08-13 10:42   ` Matthew Auld
2025-08-15 21:35     ` Lucas De Marchi
2025-08-08 17:29 ` [PATCH v3 03/13] drm/xe/rtp: Add match for psmi Lucas De Marchi
2025-08-14 21:28   ` Belgaumkar, Vinay
2025-08-08 17:29 ` [PATCH v3 04/13] drm/xe/psmi: Add Wa_14020001231 Lucas De Marchi
2025-08-13 17:44   ` Riana Tauro
2025-08-14 11:13     ` Lucas De Marchi
2025-08-08 17:29 ` [PATCH v3 05/13] drm/xe/psmi: Add Wa_16023683509 Lucas De Marchi
2025-08-13 11:15   ` Bhadane, Dnyaneshwar
2025-08-08 17:29 ` [PATCH v3 06/13] drm/xe/configfs: Simplify kernel doc Lucas De Marchi
2025-08-13  6:23   ` Riana Tauro
2025-08-08 17:29 ` [PATCH v3 07/13] drm/xe/configfs: Allow to enable PSMI Lucas De Marchi
2025-08-13  6:58   ` Riana Tauro
2025-08-13 11:23     ` Lucas De Marchi
2025-08-13 17:38       ` Riana Tauro
2025-08-08 17:29 ` [PATCH v3 08/13] drm/xe/configfs: Use guard() for dev->lock Lucas De Marchi
2025-08-12 10:23   ` Bhadane, Dnyaneshwar
2025-08-08 17:29 ` [PATCH v3 09/13] drm/xe/configfs: Block runtime attribute changes Lucas De Marchi
2025-08-13 11:03   ` Riana Tauro
2025-08-08 17:29 ` [PATCH v3 10/13] drm/xe/configfs: Use tree-like output in documentation Lucas De Marchi
2025-08-14 21:31   ` Bhadane, Dnyaneshwar
2025-08-08 17:29 ` [PATCH v3 11/13] drm/xe/configfs: Improve documentation steps Lucas De Marchi
2025-08-13 11:08   ` Riana Tauro
2025-08-08 17:29 ` [PATCH v3 12/13] drm/xe/configfs: Minor fixes to documentation Lucas De Marchi
2025-08-12 10:24   ` Bhadane, Dnyaneshwar
2025-08-08 17:29 ` [PATCH v3 13/13] drm/xe/configfs: Dump custom settings when binding Lucas De Marchi
2025-08-15  0:48   ` Belgaumkar, Vinay

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=1749abbd-5b10-4684-b1fb-c340fe57a215@intel.com \
    --to=vinay.belgaumkar@intel.com \
    --cc=John.C.Harrison@Intel.com \
    --cc=daniele.ceraolospurio@intel.com \
    --cc=dnyaneshwar.bhadane@intel.com \
    --cc=intel-xe@lists.freedesktop.org \
    --cc=lucas.demarchi@intel.com \
    --cc=matthew.d.roper@intel.com \
    --cc=prashanth.kumar@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;
as well as URLs for NNTP newsgroup(s).