Intel-XE Archive on lore.kernel.org
 help / color / mirror / Atom feed
From: Matthew Brost <matthew.brost@intel.com>
To: intel-xe@lists.freedesktop.org
Subject: [PATCH 19/28] drm/xe/vf: Use GUC_HXG_TYPE_EVENT for GuC context register
Date: Tue, 16 Sep 2025 20:46:06 -0700	[thread overview]
Message-ID: <20250917034615.3977603-20-matthew.brost@intel.com> (raw)
In-Reply-To: <20250917034615.3977603-1-matthew.brost@intel.com>

The only case where the GuC submission backend cannot reason 100%
correctly is when a GuC context is registered during VF post-migration
recovery. In this scenario, it's possible that the GuC context register
H2G is processed, but the immediately following schedule-enable H2G gets
lost.

A double register is harmless when using `GUC_HXG_TYPE_EVENT`, as GuC
simply drops the duplicate H2G. To keep things simple, use
`GUC_HXG_TYPE_EVENT` for all context registrations on VFs.

Signed-off-by: Matthew Brost <matthew.brost@intel.com>
---
 drivers/gpu/drm/xe/xe_guc_ct.c | 30 ++++++++++++++++++++++--------
 1 file changed, 22 insertions(+), 8 deletions(-)

diff --git a/drivers/gpu/drm/xe/xe_guc_ct.c b/drivers/gpu/drm/xe/xe_guc_ct.c
index 661ab1bf4502..6a1baed934c2 100644
--- a/drivers/gpu/drm/xe/xe_guc_ct.c
+++ b/drivers/gpu/drm/xe/xe_guc_ct.c
@@ -737,6 +737,24 @@ static u16 next_ct_seqno(struct xe_guc_ct *ct, bool is_g2h_fence)
 	return seqno;
 }
 
+#define MAKE_ACTION(type, __action)				\
+	FIELD_PREP(GUC_HXG_MSG_0_TYPE, type) |			\
+	FIELD_PREP(GUC_HXG_EVENT_MSG_0_ACTION |			\
+		   GUC_HXG_EVENT_MSG_0_DATA0, __action)
+
+static bool vf_action_can_safely_fail(struct xe_device *xe, u32 action)
+{
+	/*
+	 * If we are VF resuming, we can't exactly track if a context
+	 * registration has been completed in the GuC state machine, it is
+	 * harmless to resend as it will just fail silently if
+	 * GUC_HXG_TYPE_EVENT is used.
+	 */
+	return IS_SRIOV_VF(xe) &&
+		(action == XE_GUC_ACTION_REGISTER_CONTEXT_MULTI_LRC ||
+		 action == XE_GUC_ACTION_REGISTER_CONTEXT);
+}
+
 #define H2G_CT_HEADERS (GUC_CTB_HDR_LEN + 1) /* one DW CTB header and one DW HxG header */
 
 static int h2g_write(struct xe_guc_ct *ct, const u32 *action, u32 len,
@@ -808,18 +826,14 @@ static int h2g_write(struct xe_guc_ct *ct, const u32 *action, u32 len,
 		FIELD_PREP(GUC_CTB_MSG_0_NUM_DWORDS, len) |
 		FIELD_PREP(GUC_CTB_MSG_0_FENCE, ct_fence_value);
 	if (want_response) {
-		cmd[1] =
-			FIELD_PREP(GUC_HXG_MSG_0_TYPE, GUC_HXG_TYPE_REQUEST) |
-			FIELD_PREP(GUC_HXG_EVENT_MSG_0_ACTION |
-				   GUC_HXG_EVENT_MSG_0_DATA0, action[0]);
+		cmd[1] = MAKE_ACTION(GUC_HXG_TYPE_REQUEST, action[0]);
+	} else if (vf_action_can_safely_fail(xe, action[0])) {
+		cmd[1] = MAKE_ACTION(GUC_HXG_TYPE_EVENT, action[0]);
 	} else {
 		fast_req_track(ct, ct_fence_value,
 			       FIELD_GET(GUC_HXG_EVENT_MSG_0_ACTION, action[0]));
 
-		cmd[1] =
-			FIELD_PREP(GUC_HXG_MSG_0_TYPE, GUC_HXG_TYPE_FAST_REQUEST) |
-			FIELD_PREP(GUC_HXG_EVENT_MSG_0_ACTION |
-				   GUC_HXG_EVENT_MSG_0_DATA0, action[0]);
+		cmd[1] = MAKE_ACTION(GUC_HXG_TYPE_FAST_REQUEST, action[0]);
 	}
 
 	/* H2G header in cmd[1] replaces action[0] so: */
-- 
2.34.1


  parent reply	other threads:[~2025-09-17  3:46 UTC|newest]

Thread overview: 33+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2025-09-17  3:45 [PATCH 00/28] VF migration redesign Matthew Brost
2025-09-17  3:45 ` [PATCH 01/28] drm/xe/vf: Lock querying GGTT config during driver init Matthew Brost
2025-09-17  3:45 ` [PATCH 02/28] Revert "drm/xe/vf: Rebase exec queue parallel commands during migration recovery" Matthew Brost
2025-09-17  3:45 ` [PATCH 03/28] Revert "drm/xe/vf: Post migration, repopulate ring area for pending request" Matthew Brost
2025-09-17  3:45 ` [PATCH 04/28] Revert "drm/xe/vf: Fixup CTB send buffer messages after migration" Matthew Brost
2025-09-17  3:45 ` [PATCH 05/28] drm/xe: Save off position in ring in which a job was programmed Matthew Brost
2025-09-17  3:45 ` [PATCH 06/28] drm/xe/guc: Track pending-enable source in submission state Matthew Brost
2025-09-17  3:45 ` [PATCH 07/28] drm/xe: Track LR jobs in DRM scheduler pending list Matthew Brost
2025-09-17  3:45 ` [PATCH 08/28] drm/xe: Don't change LRC ring head on job resubmission Matthew Brost
2025-09-17  3:45 ` [PATCH 09/28] drm/xe/guc: Document GuC submission backend Matthew Brost
2025-09-17  3:45 ` [PATCH 10/28] drm/xe/vf: Add xe_gt_sriov_vf_recovery_inprogress helper Matthew Brost
2025-09-17  3:45 ` [PATCH 11/28] drm/xe/vf: Make VF recovery run on per-GT worker Matthew Brost
2025-09-17  3:45 ` [PATCH 12/28] drm/xe/vf: Abort H2G sends during VF post-migration recovery Matthew Brost
2025-09-17  3:46 ` [PATCH 13/28] drm/xe/vf: Remove memory allocations from VF post migration recovery Matthew Brost
2025-09-17  3:46 ` [PATCH 14/28] drm/xe/vf: Close multi-GT GGTT shift race Matthew Brost
2025-09-17  3:46 ` [PATCH 15/28] drm/xe/vf: Teardown VF post migration worker on driver unload Matthew Brost
2025-09-17  3:46 ` [PATCH 16/28] drm/xe/vf: Don't allow GT reset to be queued during VF post migration recovery Matthew Brost
2025-09-17  3:46 ` [PATCH 17/28] drm/xe/vf: Wakeup in GuC backend on " Matthew Brost
2025-09-17  3:46 ` [PATCH 18/28] drm/xe/vf: Extra debug on GGTT shift Matthew Brost
2025-09-17  3:46 ` Matthew Brost [this message]
2025-09-17  3:46 ` [PATCH 20/28] drm/xe/vf: Stop and flush CTs in VF post migration recovery Matthew Brost
2025-09-17  3:46 ` [PATCH 21/28] drm/xe/vf: Reset TLB invalidations during " Matthew Brost
2025-09-17  3:46 ` [PATCH 22/28] drm/xe/vf: Kickstart after resfix in " Matthew Brost
2025-09-17  3:46 ` [PATCH 23/28] drm/xe/vf: Start CTs before resfix " Matthew Brost
2025-09-17  3:46 ` [PATCH 24/28] drm/xe/vf: Abort VF post migration recovery on failure Matthew Brost
2025-09-17  3:46 ` [PATCH 25/28] drm/xe/vf: Replay GuC submission state on pause / unpause Matthew Brost
2025-09-17  3:46 ` [PATCH 26/28] drm/xe: Move queue init before LRC creation Matthew Brost
2025-09-17  3:46 ` [PATCH 27/28] drm/xe/vf: Add debug prints for GuC replaying state during VF recovery Matthew Brost
2025-09-17  3:46 ` [PATCH 28/28] drm/xe/vf: Workaround for race condition in GuC firmware during VF pause Matthew Brost
2025-09-17  3:55 ` ✗ CI.checkpatch: warning for VF migration redesign Patchwork
2025-09-17  3:56 ` ✓ CI.KUnit: success " Patchwork
2025-09-17  4:39 ` ✗ Xe.CI.BAT: failure " Patchwork
2025-09-17  6:38 ` ✗ Xe.CI.Full: " 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=20250917034615.3977603-20-matthew.brost@intel.com \
    --to=matthew.brost@intel.com \
    --cc=intel-xe@lists.freedesktop.org \
    /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