All of lore.kernel.org
 help / color / mirror / Atom feed
From: Leo Yan <leo.yan@linaro.org>
To: Arnaldo Carvalho de Melo <acme@kernel.org>,
	Peter Zijlstra <peterz@infradead.org>,
	Ingo Molnar <mingo@redhat.com>,
	Mark Rutland <mark.rutland@arm.com>, Jiri Olsa <jolsa@kernel.org>,
	Namhyung Kim <namhyung@kernel.org>,
	Ian Rogers <irogers@google.com>,
	John Garry <john.garry@huawei.com>, Will Deacon <will@kernel.org>,
	James Clark <james.clark@arm.com>,
	German Gomez <german.gomez@arm.com>,
	Ali Saidi <alisaidi@amazon.com>, Joe Mario <jmario@redhat.com>,
	Adam Li <adam.li@amperecomputing.com>,
	linux-perf-users@vger.kernel.org, linux-kernel@vger.kernel.org,
	linux-arm-kernel@lists.infradead.org
Cc: Leo Yan <leo.yan@linaro.org>
Subject: [PATCH v5 04/17] perf arm-spe: Don't set data source if it's not a memory operation
Date: Sat,  4 Jun 2022 12:28:07 +0800	[thread overview]
Message-ID: <20220604042820.2270916-5-leo.yan@linaro.org> (raw)
In-Reply-To: <20220604042820.2270916-1-leo.yan@linaro.org>

Except memory load and store operations, Arm SPE records also can
support other operation types, bug when set the data source field the
current code assumes a record is a either load operation or store
operation, this leads to wrongly synthesize memory samples.

This patch strictly checks the record operation type, it only sets data
source only for the operation types ARM_SPE_LD and ARM_SPE_ST,
otherwise, returns zero for data source.  Therefore, we can synthesize
memory samples only when data source is a non-zero value, the function
arm_spe__is_memory_event() is useless and removed.

Fixes: e55ed3423c1b ("perf arm-spe: Synthesize memory event")
Signed-off-by: Leo Yan <leo.yan@linaro.org>
Reviewed-by: Ali Saidi <alisaidi@amazon.com>
Tested-by: Ali Saidi <alisaidi@amazon.com>
Reviewed-by: German Gomez <german.gomez@arm.com>
---
 tools/perf/util/arm-spe.c | 22 ++++++++--------------
 1 file changed, 8 insertions(+), 14 deletions(-)

diff --git a/tools/perf/util/arm-spe.c b/tools/perf/util/arm-spe.c
index 1a80151baed9..d040406f3314 100644
--- a/tools/perf/util/arm-spe.c
+++ b/tools/perf/util/arm-spe.c
@@ -387,26 +387,16 @@ static int arm_spe__synth_instruction_sample(struct arm_spe_queue *speq,
 	return arm_spe_deliver_synth_event(spe, speq, event, &sample);
 }
 
-#define SPE_MEM_TYPE	(ARM_SPE_L1D_ACCESS | ARM_SPE_L1D_MISS | \
-			 ARM_SPE_LLC_ACCESS | ARM_SPE_LLC_MISS | \
-			 ARM_SPE_REMOTE_ACCESS)
-
-static bool arm_spe__is_memory_event(enum arm_spe_sample_type type)
-{
-	if (type & SPE_MEM_TYPE)
-		return true;
-
-	return false;
-}
-
 static u64 arm_spe__synth_data_source(const struct arm_spe_record *record)
 {
 	union perf_mem_data_src	data_src = { 0 };
 
 	if (record->op == ARM_SPE_LD)
 		data_src.mem_op = PERF_MEM_OP_LOAD;
-	else
+	else if (record->op == ARM_SPE_ST)
 		data_src.mem_op = PERF_MEM_OP_STORE;
+	else
+		return 0;
 
 	if (record->type & (ARM_SPE_LLC_ACCESS | ARM_SPE_LLC_MISS)) {
 		data_src.mem_lvl = PERF_MEM_LVL_L3;
@@ -510,7 +500,11 @@ static int arm_spe_sample(struct arm_spe_queue *speq)
 			return err;
 	}
 
-	if (spe->sample_memory && arm_spe__is_memory_event(record->type)) {
+	/*
+	 * When data_src is zero it means the record is not a memory operation,
+	 * skip to synthesize memory sample for this case.
+	 */
+	if (spe->sample_memory && data_src) {
 		err = arm_spe__synth_mem_sample(speq, spe->memory_id, data_src);
 		if (err)
 			return err;
-- 
2.25.1


WARNING: multiple messages have this Message-ID (diff)
From: Leo Yan <leo.yan@linaro.org>
To: Arnaldo Carvalho de Melo <acme@kernel.org>,
	Peter Zijlstra <peterz@infradead.org>,
	Ingo Molnar <mingo@redhat.com>,
	Mark Rutland <mark.rutland@arm.com>, Jiri Olsa <jolsa@kernel.org>,
	Namhyung Kim <namhyung@kernel.org>,
	Ian Rogers <irogers@google.com>,
	John Garry <john.garry@huawei.com>, Will Deacon <will@kernel.org>,
	James Clark <james.clark@arm.com>,
	German Gomez <german.gomez@arm.com>,
	Ali Saidi <alisaidi@amazon.com>, Joe Mario <jmario@redhat.com>,
	Adam Li <adam.li@amperecomputing.com>,
	linux-perf-users@vger.kernel.org, linux-kernel@vger.kernel.org,
	linux-arm-kernel@lists.infradead.org
Cc: Leo Yan <leo.yan@linaro.org>
Subject: [PATCH v5 04/17] perf arm-spe: Don't set data source if it's not a memory operation
Date: Sat,  4 Jun 2022 12:28:07 +0800	[thread overview]
Message-ID: <20220604042820.2270916-5-leo.yan@linaro.org> (raw)
In-Reply-To: <20220604042820.2270916-1-leo.yan@linaro.org>

Except memory load and store operations, Arm SPE records also can
support other operation types, bug when set the data source field the
current code assumes a record is a either load operation or store
operation, this leads to wrongly synthesize memory samples.

This patch strictly checks the record operation type, it only sets data
source only for the operation types ARM_SPE_LD and ARM_SPE_ST,
otherwise, returns zero for data source.  Therefore, we can synthesize
memory samples only when data source is a non-zero value, the function
arm_spe__is_memory_event() is useless and removed.

Fixes: e55ed3423c1b ("perf arm-spe: Synthesize memory event")
Signed-off-by: Leo Yan <leo.yan@linaro.org>
Reviewed-by: Ali Saidi <alisaidi@amazon.com>
Tested-by: Ali Saidi <alisaidi@amazon.com>
Reviewed-by: German Gomez <german.gomez@arm.com>
---
 tools/perf/util/arm-spe.c | 22 ++++++++--------------
 1 file changed, 8 insertions(+), 14 deletions(-)

diff --git a/tools/perf/util/arm-spe.c b/tools/perf/util/arm-spe.c
index 1a80151baed9..d040406f3314 100644
--- a/tools/perf/util/arm-spe.c
+++ b/tools/perf/util/arm-spe.c
@@ -387,26 +387,16 @@ static int arm_spe__synth_instruction_sample(struct arm_spe_queue *speq,
 	return arm_spe_deliver_synth_event(spe, speq, event, &sample);
 }
 
-#define SPE_MEM_TYPE	(ARM_SPE_L1D_ACCESS | ARM_SPE_L1D_MISS | \
-			 ARM_SPE_LLC_ACCESS | ARM_SPE_LLC_MISS | \
-			 ARM_SPE_REMOTE_ACCESS)
-
-static bool arm_spe__is_memory_event(enum arm_spe_sample_type type)
-{
-	if (type & SPE_MEM_TYPE)
-		return true;
-
-	return false;
-}
-
 static u64 arm_spe__synth_data_source(const struct arm_spe_record *record)
 {
 	union perf_mem_data_src	data_src = { 0 };
 
 	if (record->op == ARM_SPE_LD)
 		data_src.mem_op = PERF_MEM_OP_LOAD;
-	else
+	else if (record->op == ARM_SPE_ST)
 		data_src.mem_op = PERF_MEM_OP_STORE;
+	else
+		return 0;
 
 	if (record->type & (ARM_SPE_LLC_ACCESS | ARM_SPE_LLC_MISS)) {
 		data_src.mem_lvl = PERF_MEM_LVL_L3;
@@ -510,7 +500,11 @@ static int arm_spe_sample(struct arm_spe_queue *speq)
 			return err;
 	}
 
-	if (spe->sample_memory && arm_spe__is_memory_event(record->type)) {
+	/*
+	 * When data_src is zero it means the record is not a memory operation,
+	 * skip to synthesize memory sample for this case.
+	 */
+	if (spe->sample_memory && data_src) {
 		err = arm_spe__synth_mem_sample(speq, spe->memory_id, data_src);
 		if (err)
 			return err;
-- 
2.25.1


_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel

  parent reply	other threads:[~2022-06-04  4:29 UTC|newest]

Thread overview: 51+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2022-06-04  4:28 [PATCH v5 00/17] perf c2c: Support data source and display for Arm64 Leo Yan
2022-06-04  4:28 ` Leo Yan
2022-06-04  4:28 ` [PATCH v5 01/17] perf: Add SNOOP_PEER flag to perf mem data struct Leo Yan
2022-06-04  4:28   ` Leo Yan
2022-07-20 18:45   ` Arnaldo Carvalho de Melo
2022-07-20 18:45     ` Arnaldo Carvalho de Melo
2022-07-20 18:46     ` Arnaldo Carvalho de Melo
2022-07-20 18:46       ` Arnaldo Carvalho de Melo
2022-07-21  0:27       ` Leo Yan
2022-07-21  0:27         ` Leo Yan
2022-08-08 16:40         ` ping " Arnaldo Carvalho de Melo
2022-06-04  4:28 ` [PATCH v5 02/17] perf tools: sync addition of PERF_MEM_SNOOPX_PEER Leo Yan
2022-06-04  4:28   ` Leo Yan
2022-06-04  4:28 ` [PATCH v5 03/17] perf mem: Print snoop peer flag Leo Yan
2022-06-04  4:28   ` Leo Yan
2022-06-04  4:28 ` Leo Yan [this message]
2022-06-04  4:28   ` [PATCH v5 04/17] perf arm-spe: Don't set data source if it's not a memory operation Leo Yan
2022-06-04  4:28 ` [PATCH v5 05/17] perf arm-spe: Use SPE data source for neoverse cores Leo Yan
2022-06-04  4:28   ` Leo Yan
2022-06-04  4:28 ` [PATCH v5 06/17] perf mem: Add statistics for peer snooping Leo Yan
2022-06-04  4:28   ` Leo Yan
2022-06-04  4:28 ` [PATCH v5 07/17] perf c2c: Output " Leo Yan
2022-06-04  4:28   ` Leo Yan
2022-06-04  4:28 ` [PATCH v5 08/17] perf c2c: Add dimensions for peer load operations Leo Yan
2022-06-04  4:28   ` Leo Yan
2022-06-04  4:28 ` [PATCH v5 09/17] perf c2c: Add dimensions of peer metrics for cache line view Leo Yan
2022-06-04  4:28   ` Leo Yan
2022-06-04  4:28 ` [PATCH v5 10/17] perf c2c: Add mean dimensions for peer operations Leo Yan
2022-06-04  4:28   ` Leo Yan
2022-06-04  4:28 ` [PATCH v5 11/17] perf c2c: Use explicit names for display macros Leo Yan
2022-06-04  4:28   ` Leo Yan
2022-06-04  4:28 ` [PATCH v5 12/17] perf c2c: Rename dimension from 'percent_hitm' to 'percent_costly_snoop' Leo Yan
2022-06-04  4:28   ` Leo Yan
2022-06-04  4:28 ` [PATCH v5 13/17] perf c2c: Refactor node header Leo Yan
2022-06-04  4:28   ` Leo Yan
2022-06-04  4:28 ` [PATCH v5 14/17] perf c2c: Refactor display string Leo Yan
2022-06-04  4:28   ` Leo Yan
2022-06-04  4:28 ` [PATCH v5 15/17] perf c2c: Sort on peer snooping for load operations Leo Yan
2022-06-04  4:28   ` Leo Yan
2022-06-04  4:28 ` [PATCH v5 16/17] perf c2c: Use 'peer' as default display for Arm64 Leo Yan
2022-06-04  4:28   ` Leo Yan
2022-06-04  4:28 ` [PATCH v5 17/17] perf c2c: Update documentation for new display option 'peer' Leo Yan
2022-06-04  4:28   ` Leo Yan
2022-08-10 13:37 ` [PATCH v5 00/17] perf c2c: Support data source and display for Arm64 Arnaldo Carvalho de Melo
2022-08-10 13:37   ` Arnaldo Carvalho de Melo
2022-08-11  6:41   ` Leo Yan
2022-08-11  6:41     ` Leo Yan
2022-08-12 12:43     ` Arnaldo Carvalho de Melo
2022-08-12 12:43       ` Arnaldo Carvalho de Melo
2022-08-12 15:20       ` Leo Yan
2022-08-12 15:20         ` Leo Yan

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=20220604042820.2270916-5-leo.yan@linaro.org \
    --to=leo.yan@linaro.org \
    --cc=acme@kernel.org \
    --cc=adam.li@amperecomputing.com \
    --cc=alisaidi@amazon.com \
    --cc=german.gomez@arm.com \
    --cc=irogers@google.com \
    --cc=james.clark@arm.com \
    --cc=jmario@redhat.com \
    --cc=john.garry@huawei.com \
    --cc=jolsa@kernel.org \
    --cc=linux-arm-kernel@lists.infradead.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-perf-users@vger.kernel.org \
    --cc=mark.rutland@arm.com \
    --cc=mingo@redhat.com \
    --cc=namhyung@kernel.org \
    --cc=peterz@infradead.org \
    --cc=will@kernel.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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.