From: Michal Wajdeczko <michal.wajdeczko@intel.com>
To: intel-xe@lists.freedesktop.org
Cc: Michal Wajdeczko <michal.wajdeczko@intel.com>,
Matthew Brost <matthew.brost@intel.com>
Subject: [PATCH v2 4/4] drm/xe/guc: Don't treat GuC generic CAT error as protocol error
Date: Tue, 5 Nov 2024 21:45:57 +0100 [thread overview]
Message-ID: <20241105204557.1991-1-michal.wajdeczko@intel.com> (raw)
In-Reply-To: <20241105173032.1947-5-michal.wajdeczko@intel.com>
GuC uses GUC_ID_UNKNOWN if it can not map the CAT fault to any
context. We shouldn't treat that as G2H protocol error that would
justify a GT reset, as it may happen due to some VF activity.
Signed-off-by: Michal Wajdeczko <michal.wajdeczko@intel.com>
Cc: Matthew Brost <matthew.brost@intel.com>
---
v2: replaced 16b GUC_ID_MAX with 32b GUC_ID_UNKNOWN
---
drivers/gpu/drm/xe/xe_guc_fwif.h | 1 +
drivers/gpu/drm/xe/xe_guc_submit.c | 9 +++++++++
2 files changed, 10 insertions(+)
diff --git a/drivers/gpu/drm/xe/xe_guc_fwif.h b/drivers/gpu/drm/xe/xe_guc_fwif.h
index 08ffe59f22fa..057153f89b30 100644
--- a/drivers/gpu/drm/xe/xe_guc_fwif.h
+++ b/drivers/gpu/drm/xe/xe_guc_fwif.h
@@ -17,6 +17,7 @@
#define G2H_LEN_DW_TLB_INVALIDATE 3
#define GUC_ID_MAX 65535
+#define GUC_ID_UNKNOWN 0xffffffff
#define GUC_CONTEXT_DISABLE 0
#define GUC_CONTEXT_ENABLE 1
diff --git a/drivers/gpu/drm/xe/xe_guc_submit.c b/drivers/gpu/drm/xe/xe_guc_submit.c
index 147000fd1177..614110dbb527 100644
--- a/drivers/gpu/drm/xe/xe_guc_submit.c
+++ b/drivers/gpu/drm/xe/xe_guc_submit.c
@@ -2023,6 +2023,15 @@ int xe_guc_exec_queue_memory_cat_error_handler(struct xe_guc *guc, u32 *msg,
guc_id = msg[0];
+ if (guc_id == GUC_ID_UNKNOWN) {
+ /*
+ * GuC uses GUC_ID_UNKNOWN if it can not map the CAT fault to any PF/VF
+ * context. In such case only PF will be notified about that fault.
+ */
+ xe_gt_err_ratelimited(gt, "Memory CAT error reported by GuC!\n");
+ return 0;
+ }
+
q = g2h_exec_queue_lookup(guc, guc_id);
if (unlikely(!q))
return -EPROTO;
--
2.43.0
next prev parent reply other threads:[~2024-11-05 20:46 UTC|newest]
Thread overview: 30+ messages / expand[flat|nested] mbox.gz Atom feed top
2024-11-05 17:30 [PATCH 0/4] GuC: Improve handling of the malformed G2H Michal Wajdeczko
2024-11-05 17:30 ` [PATCH 1/4] drm/xe/guc: Log content of the failed G2H message Michal Wajdeczko
2024-11-05 19:30 ` Matthew Brost
2024-11-05 17:30 ` [PATCH 2/4] drm/xe/guc: Drop redundant logs about invalid G2H length Michal Wajdeczko
2024-11-05 19:31 ` Matthew Brost
2024-11-05 17:30 ` [PATCH 3/4] drm/xe/guc: Don't read data from G2H prior to length check Michal Wajdeczko
2024-11-05 19:36 ` Matthew Brost
2024-11-05 20:01 ` Michal Wajdeczko
2024-11-06 1:32 ` Matthew Brost
2024-11-05 17:30 ` [PATCH 4/4] drm/xe/guc: Don't treat GuC generic CAT error as protocol error Michal Wajdeczko
2024-11-05 20:13 ` Michal Wajdeczko
2024-11-05 20:45 ` Michal Wajdeczko [this message]
2024-11-06 1:24 ` [PATCH v2 " Matthew Brost
2024-11-05 17:38 ` ✓ CI.Patch_applied: success for GuC: Improve handling of the malformed G2H Patchwork
2024-11-05 17:38 ` ✓ CI.checkpatch: " Patchwork
2024-11-05 17:39 ` ✓ CI.KUnit: " Patchwork
2024-11-05 17:52 ` ✓ CI.Build: " Patchwork
2024-11-05 17:54 ` ✓ CI.Hooks: " Patchwork
2024-11-05 17:56 ` ✓ CI.checksparse: " Patchwork
2024-11-05 18:37 ` ✓ CI.BAT: " Patchwork
2024-11-05 20:51 ` ✓ CI.Patch_applied: success for GuC: Improve handling of the malformed G2H (rev2) Patchwork
2024-11-05 20:52 ` ✓ CI.checkpatch: " Patchwork
2024-11-05 20:53 ` ✓ CI.KUnit: " Patchwork
2024-11-05 21:04 ` ✓ CI.Build: " Patchwork
2024-11-05 21:07 ` ✓ CI.Hooks: " Patchwork
2024-11-05 21:08 ` ✓ CI.checksparse: " Patchwork
2024-11-05 21:36 ` ✓ CI.BAT: " Patchwork
2024-11-06 22:55 ` ✗ CI.FULL: failure for GuC: Improve handling of the malformed G2H Patchwork
2024-11-07 1:06 ` ✗ CI.FULL: failure for GuC: Improve handling of the malformed G2H (rev2) Patchwork
2024-11-07 9:21 ` Michal Wajdeczko
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=20241105204557.1991-1-michal.wajdeczko@intel.com \
--to=michal.wajdeczko@intel.com \
--cc=intel-xe@lists.freedesktop.org \
--cc=matthew.brost@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