Intel-XE Archive on lore.kernel.org
 help / color / mirror / Atom feed
From: John Harrison <john.c.harrison@intel.com>
To: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>,
	<intel-xe@lists.freedesktop.org>
Cc: Julia Filipchuk <julia.filipchuk@intel.com>,
	Michal Wajdeczko <michal.wajdeczko@intel.com>
Subject: Re: [PATCH v4 2/2] drm/xe/guc: Enable the Dynamic Inhibit Context Switch optimization
Date: Fri, 13 Jun 2025 09:16:01 -0700	[thread overview]
Message-ID: <def3f5f0-3bf7-4e30-9089-b0d4567cbf6f@intel.com> (raw)
In-Reply-To: <20250604002454.743404-6-daniele.ceraolospurio@intel.com>

On 6/3/2025 5:24 PM, Daniele Ceraolo Spurio wrote:
> The Dynamic Inhibit Context Switch is an optimization aimed at reducing
> the amount of time the HW is stuck waiting on an unsatisfied semaphore.
> When this optimization is enabled, the GuC will dynamically modify the
> CTX_CTRL_INHIBIT_SYN_CTX_SWITCH in the CTX_CONTEXT_CONTROL register of
> LRCs to enable immediate switching out on an unsatisfied semaphore wait
> when multiple contexts are competing for time on the same engine.
>
> This feature is available on recent HW from GuC 70.40.1 onwards and it
> is enabled via a per-VF feature opt-in.
>
> v2: rebase
> v3: switch to using guc_buf_cache instead of dedicated alloc
> v4: add helper to check for feature availability (Michal), don't enable
> if multi-lrc is possible.
>
> Signed-off-by: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
> Cc: John Harrison <John.C.Harrison@Intel.com>
> Cc: Julia Filipchuk <julia.filipchuk@intel.com>
> Cc: Michal Wajdeczko <michal.wajdeczko@intel.com>
Reviewed-by: John Harrison <John.C.Harrison@Intel.com>

> ---
>   drivers/gpu/drm/xe/abi/guc_klvs_abi.h | 13 ++++++++++++
>   drivers/gpu/drm/xe/xe_guc.c           | 30 +++++++++++++++++++++++++++
>   2 files changed, 43 insertions(+)
>
> diff --git a/drivers/gpu/drm/xe/abi/guc_klvs_abi.h b/drivers/gpu/drm/xe/abi/guc_klvs_abi.h
> index 5b2502bec2dc..0366a9da5977 100644
> --- a/drivers/gpu/drm/xe/abi/guc_klvs_abi.h
> +++ b/drivers/gpu/drm/xe/abi/guc_klvs_abi.h
> @@ -134,11 +134,24 @@ enum  {
>    *      Adds an extra dword to the XE_GUC_ACTION_NOTIFY_MEMORY_CAT_ERROR G2H
>    *      containing the type of the CAT error. On HW that does not support
>    *      reporting the CAT error type, the extra dword is set to 0xdeadbeef.
> + *
> + * _`GUC_KLV_OPT_IN_FEATURE_DYNAMIC_INHIBIT_CONTEXT_SWITCH` : 0x4003
> + *      This KLV enables the Dynamic Inhibit Context Switch optimization, which
> + *      consists in the GuC setting the CTX_CTRL_INHIBIT_SYN_CTX_SWITCH bit to
> + *      zero in the CTX_CONTEXT_CONTROL register of LRCs that are submitted
> + *      to an oversubscribed engine. This will cause those contexts to be
> + *      switched out immediately if they hit an unsatisfied semaphore wait
> + *      (instead of waiting the full timeslice duration). The bit is instead set
> + *      to one if a single context is queued on the engine, to avoid it being
> + *      switched out if there isn't another context that can run in its place.
>    */
>   
>   #define GUC_KLV_OPT_IN_FEATURE_EXT_CAT_ERR_TYPE_KEY 0x4001
>   #define GUC_KLV_OPT_IN_FEATURE_EXT_CAT_ERR_TYPE_LEN 0u
>   
> +#define GUC_KLV_OPT_IN_FEATURE_DYNAMIC_INHIBIT_CONTEXT_SWITCH_KEY 0x4003
> +#define GUC_KLV_OPT_IN_FEATURE_DYNAMIC_INHIBIT_CONTEXT_SWITCH_LEN 0u
> +
>   /**
>    * DOC: GuC VGT Policy KLVs
>    *
> diff --git a/drivers/gpu/drm/xe/xe_guc.c b/drivers/gpu/drm/xe/xe_guc.c
> index 46419861bfcd..b94e7bd865d5 100644
> --- a/drivers/gpu/drm/xe/xe_guc.c
> +++ b/drivers/gpu/drm/xe/xe_guc.c
> @@ -583,6 +583,32 @@ static int __guc_opt_in_features_enable(struct xe_guc *guc, u64 addr, u32 num_dw
>   	return xe_guc_ct_send_block(&guc->ct, action, ARRAY_SIZE(action));
>   }
>   
> +static bool supports_dynamic_ics(struct xe_guc *guc)
> +{
> +	struct xe_device *xe = guc_to_xe(guc);
> +	struct xe_gt *gt = guc_to_gt(guc);
> +
> +	/* Dynamic ICS is available for PVC and Xe2 and newer platforms. */
> +	if (xe->info.platform != XE_PVC && GRAPHICS_VER(xe) < 20)
> +		return false;
> +
> +	/*
> +	 * The feature is currently not compatible with multi-lrc, so the GuC
> +	 * does not support it at all on the media engines (which are the main
> +	 * users of mlrc). On the primary GT side, to avoid it being used in
> +	 * conjunction with mlrc, we only enable it if we are in single CCS
> +	 * mode.
> +	 */
> +	if (xe_gt_is_media_type(gt) || gt->ccs_mode > 1)
> +		return false;
> +
> +	/*
> +	 * Dynamic ICS requires GuC v70.40.1, which maps to compatibility
> +	 * version v1.18.4.
> +	 */
> +	return GUC_SUBMIT_VER(guc) >= MAKE_GUC_VER(1, 18, 4);
> +}
> +
>   #define OPT_IN_MAX_DWORDS 16
>   int xe_guc_opt_in_features_enable(struct xe_guc *guc)
>   {
> @@ -607,6 +633,10 @@ int xe_guc_opt_in_features_enable(struct xe_guc *guc)
>   	if (GUC_SUBMIT_VER(guc) >= MAKE_GUC_VER(1, 7, 0))
>   		klvs[count++] = PREP_GUC_KLV_TAG(OPT_IN_FEATURE_EXT_CAT_ERR_TYPE);
>   
> +
> +	if (supports_dynamic_ics(guc))
> +		klvs[count++] = PREP_GUC_KLV_TAG(OPT_IN_FEATURE_DYNAMIC_INHIBIT_CONTEXT_SWITCH);
> +
>   	if (count) {
>   		xe_assert(xe, count <= OPT_IN_MAX_DWORDS);
>   


  reply	other threads:[~2025-06-13 16:16 UTC|newest]

Thread overview: 20+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2025-06-04  0:24 [PATCH v4 0/2] Enable GuC opt-in features Daniele Ceraolo Spurio
2025-06-04  0:24 ` [PATCH v4 1/2] drm/xe/guc: Enable extended CAT error reporting Daniele Ceraolo Spurio
2025-06-04  0:24 ` [PATCH v4 2/2] drm/xe/guc: Enable the Dynamic Inhibit Context Switch optimization Daniele Ceraolo Spurio
2025-06-13 16:16   ` John Harrison [this message]
2025-06-04  0:33 ` ✓ CI.Patch_applied: success for Enable GuC opt-in features Patchwork
2025-06-04  0:33 ` ✗ CI.checkpatch: warning " Patchwork
2025-06-04  0:34 ` ✗ CI.KUnit: failure " Patchwork
2025-06-05  1:27 ` ✓ CI.Patch_applied: success for Enable GuC opt-in features (rev2) Patchwork
2025-06-05  1:27 ` ✗ CI.checkpatch: warning " Patchwork
2025-06-05  1:28 ` ✓ CI.KUnit: success " Patchwork
2025-06-05  1:39 ` ✓ CI.Build: " Patchwork
2025-06-05  1:42 ` ✓ CI.Hooks: " Patchwork
2025-06-05  1:43 ` ✓ CI.checksparse: " Patchwork
2025-06-05  2:08 ` ✓ Xe.CI.BAT: " Patchwork
2025-06-06  3:59 ` ✗ Xe.CI.Full: failure " Patchwork
2025-06-06 10:43 ` Patchwork
2025-06-17 21:46 ` ✗ CI.checkpatch: warning for Enable GuC opt-in features (rev3) Patchwork
2025-06-17 21:47 ` ✓ CI.KUnit: success " Patchwork
2025-06-17 22:31 ` ✓ Xe.CI.BAT: " Patchwork
2025-06-18  9:51 ` ✗ 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=def3f5f0-3bf7-4e30-9089-b0d4567cbf6f@intel.com \
    --to=john.c.harrison@intel.com \
    --cc=daniele.ceraolospurio@intel.com \
    --cc=intel-xe@lists.freedesktop.org \
    --cc=julia.filipchuk@intel.com \
    --cc=michal.wajdeczko@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