Intel-XE Archive on lore.kernel.org
 help / color / mirror / Atom feed
From: Ashutosh Dixit <ashutosh.dixit@intel.com>
To: intel-xe@lists.freedesktop.org
Subject: [Intel-xe] [PATCH 17/17] drm/xe/oa: Remove filtering reports on context id
Date: Thu,  7 Sep 2023 21:23:48 -0700	[thread overview]
Message-ID: <20230908042348.1592535-18-ashutosh.dixit@intel.com> (raw)
In-Reply-To: <20230908042348.1592535-1-ashutosh.dixit@intel.com>

At present XE OA code does not obtain context id's from GuC. Even if these
context id's were available it is not clear if included reports for
userspace should be filtered on context id's. Till these issues are
resolved remove filtering reports based on context id's.

Suggested-by: Umesh Nerlige Ramappa <umesh.nerlige.ramappa@intel.com>
Signed-off-by: Ashutosh Dixit <ashutosh.dixit@intel.com>
---
 drivers/gpu/drm/xe/xe_oa.c       | 33 +++-----------------------------
 drivers/gpu/drm/xe/xe_oa_types.h |  9 ---------
 2 files changed, 3 insertions(+), 39 deletions(-)

diff --git a/drivers/gpu/drm/xe/xe_oa.c b/drivers/gpu/drm/xe/xe_oa.c
index eab19264b2f70..178722f0833ab 100644
--- a/drivers/gpu/drm/xe/xe_oa.c
+++ b/drivers/gpu/drm/xe/xe_oa.c
@@ -184,13 +184,6 @@ static void oa_timestamp_clear(struct xe_oa_stream *stream, u32 *report)
 		report[1] = 0;
 }
 
-static u32 oa_context_id(struct xe_oa_stream *stream, u32 *report)
-{
-	u32 ctx_id = oa_report_header_64bit(stream) ? report[4] : report[2];
-
-	return ctx_id & stream->specific_ctx_id_mask;
-}
-
 static void oa_context_id_squash(struct xe_oa_stream *stream, u32 *report)
 {
 	if (oa_report_header_64bit(stream))
@@ -368,7 +361,7 @@ static int xe_oa_append_reports(struct xe_oa_stream *stream, char __user *buf,
 
 	for (; OA_TAKEN(tail, head); head = (head + report_size) & mask) {
 		u8 *report = oa_buf_base + head;
-		u32 ctx_id, *report32 = (void *)report;
+		u32 *report32 = (void *)report;
 		u64 reason;
 
 		/*
@@ -379,7 +372,6 @@ static int xe_oa_append_reports(struct xe_oa_stream *stream, char __user *buf,
 		 * this field and it is valid to have a reason field of zero.
 		 */
 		reason = oa_report_reason(stream, report);
-		ctx_id = oa_context_id(stream, report32);
 
 		/*
 		 * Squash whatever is in the CTX_ID field if it's marked as invalid to be
@@ -388,10 +380,8 @@ static int xe_oa_append_reports(struct xe_oa_stream *stream, char __user *buf,
 		 * Note: we don't clear the valid_ctx_bit so userspace can understand that
 		 * the ID has been squashed by the kernel.
 		 */
-		if (oa_report_ctx_invalid(stream, report)) {
-			ctx_id = INVALID_CTX_ID;
+		if (oa_report_ctx_invalid(stream, report))
 			oa_context_id_squash(stream, report32);
-		}
 
 		/*
 		 * NB: The OA unit does not support clock gating off for a specific
@@ -419,21 +409,10 @@ static int xe_oa_append_reports(struct xe_oa_stream *stream, char __user *buf,
 		 * since it's not-uncommon for periodic samples to identify a switch
 		 * before any 'context switch' report.
 		 */
-		if (!stream->exec_q || stream->specific_ctx_id == ctx_id ||
-		    stream->oa_buffer.last_ctx_id == stream->specific_ctx_id ||
-		    reason & OAREPORT_REASON_CTX_SWITCH) {
-			/*
-			 * While filtering for a single context we avoid
-			 * leaking the IDs of other contexts.
-			 */
-			if (stream->exec_q && stream->specific_ctx_id != ctx_id)
-				oa_context_id_squash(stream, report32);
-
+		if (!stream->exec_q || reason & OAREPORT_REASON_CTX_SWITCH) {
 			ret = xe_oa_append_sample(stream, buf, count, offset, report);
 			if (ret)
 				break;
-
-			stream->oa_buffer.last_ctx_id = ctx_id;
 		}
 
 		if (is_power_of_2(report_size)) {
@@ -497,12 +476,6 @@ static void xe_oa_init_oa_buffer(struct xe_oa_stream *stream)
 	/* Mark that we need updated tail pointers to read from... */
 	stream->oa_buffer.tail = 0;
 
-	/*
-	 * Reset state used to recognise context switches, affecting which reports we will
-	 * forward to userspace while filtering for a single context.
-	 */
-	stream->oa_buffer.last_ctx_id = INVALID_CTX_ID;
-
 	spin_unlock_irqrestore(&stream->oa_buffer.ptr_lock, flags);
 
 	/* Zero out the OA buffer since we rely on zero report id and timestamp fields */
diff --git a/drivers/gpu/drm/xe/xe_oa_types.h b/drivers/gpu/drm/xe/xe_oa_types.h
index 3cc1d88fe4a51..7566fef55b0ab 100644
--- a/drivers/gpu/drm/xe/xe_oa_types.h
+++ b/drivers/gpu/drm/xe/xe_oa_types.h
@@ -222,12 +222,6 @@ struct xe_oa_stream {
 	 */
 	struct llist_head oa_config_bos;
 
-	/** @specific_ctx_id: id of the context used for filtering reports */
-	u32 specific_ctx_id;
-
-	/** @specific_ctx_id_mask: The mask used to masking specific_ctx_id bits */
-	u32 specific_ctx_id_mask;
-
 	/**
 	 * @poll_check_timer: High resolution timer that will periodically
 	 * check for data in the circular OA buffer for notifying userspace
@@ -261,9 +255,6 @@ struct xe_oa_stream {
 		/** @vaddr: mapped vaddr of the OA buffer */
 		u8 *vaddr;
 
-		/** @last_ctx_id: last context id for OA data added */
-		u32 last_ctx_id;
-
 		/**
 		 * @ptr_lock: Locks reads and writes to all head/tail state
 		 *
-- 
2.41.0


  parent reply	other threads:[~2023-09-08  4:24 UTC|newest]

Thread overview: 29+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2023-09-08  4:23 [Intel-xe] [PATCH 00/17] Add OA functionality to Xe Ashutosh Dixit
2023-09-08  4:23 ` [Intel-xe] [PATCH 01/17] drm/xe/oa: Introduce OA (observability architecture) uapi Ashutosh Dixit
2023-09-08  4:23 ` [Intel-xe] [PATCH 02/17] drm/xe/oa: Add OA types Ashutosh Dixit
2023-09-08  4:23 ` [Intel-xe] [PATCH 03/17] drm/xe/oa: Add registers and GPU commands used by OA Ashutosh Dixit
2023-09-08  4:23 ` [Intel-xe] [PATCH 04/17] drm/xe/oa: Module init/exit and probe/remove Ashutosh Dixit
2023-09-08  4:23 ` [Intel-xe] [PATCH 05/17] drm/xe/oa: Add/remove config ioctl's Ashutosh Dixit
2023-09-08  4:23 ` [Intel-xe] [PATCH 06/17] drm/xe/oa: Start implementing OA stream open ioctl Ashutosh Dixit
2023-09-08  4:23 ` [Intel-xe] [PATCH 07/17] drm/xe/oa: OA stream initialization Ashutosh Dixit
2023-09-08  4:23 ` [Intel-xe] [PATCH 08/17] drm/xe/oa: Expose OA stream fd Ashutosh Dixit
2023-09-08  4:23 ` [Intel-xe] [PATCH 09/17] drm/xe/oa: Read file_operation Ashutosh Dixit
2023-09-08  4:23 ` [Intel-xe] [PATCH 10/17] drm/xe/oa: Implement queries Ashutosh Dixit
2023-09-08  4:23 ` [Intel-xe] [PATCH 11/17] drm/xe/oa: Override GuC RC with OA on PVC Ashutosh Dixit
2023-09-08  4:23 ` [Intel-xe] [PATCH 12/17] drm/xe/perf: "Perf" layer to support multiple perf counter stream types Ashutosh Dixit
2023-09-08  4:23 ` [Intel-xe] [PATCH 13/17] drm/xe/oa: Multiplex PERF ops through a single PERF ioctl Ashutosh Dixit
2023-09-08  4:23 ` [Intel-xe] [PATCH 14/17] drm/xe/oa: Simplify OA configs in uapi Ashutosh Dixit
2023-09-08  4:23 ` [Intel-xe] [PATCH 15/17] drm/xe/oa: Remove OA format names from OA uapi Ashutosh Dixit
2023-09-08  4:23 ` [Intel-xe] [PATCH 16/17] drm/xe/oa: Make xe_oa_timestamp_frequency per gt Ashutosh Dixit
2023-09-08  4:23 ` Ashutosh Dixit [this message]
2023-09-09  1:24   ` [Intel-xe] [PATCH 17/17] drm/xe/oa: Remove filtering reports on context id Dixit, Ashutosh
2023-11-13 20:22     ` Dixit, Ashutosh
2023-09-08  4:26 ` [Intel-xe] ✓ CI.Patch_applied: success for Add OA functionality to Xe (rev5) Patchwork
2023-09-08  4:27 ` [Intel-xe] ✗ CI.checkpatch: warning " Patchwork
2023-09-08  4:28 ` [Intel-xe] ✓ CI.KUnit: success " Patchwork
2023-09-08  4:35 ` [Intel-xe] ✓ CI.Build: " Patchwork
2023-09-08  4:35 ` [Intel-xe] ✗ CI.Hooks: failure " Patchwork
2023-09-19 21:18   ` Dixit, Ashutosh
2023-09-08  4:36 ` [Intel-xe] ✓ CI.checksparse: success " Patchwork
2023-09-08  5:10 ` [Intel-xe] ✗ CI.BAT: failure " Patchwork
2023-09-19 21:27   ` Dixit, Ashutosh

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=20230908042348.1592535-18-ashutosh.dixit@intel.com \
    --to=ashutosh.dixit@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