Skip to contentSkip to conversation

perf/x86/intel/pt: Fix sampling using single range output

PATCH Applied Adrian Hunter 4 participants 7 messages Nov 12, 2022 – Nov 16, 2022

Subsystems (3)
lore
Share & tools
lore

Adrian Hunter

author reviewer
#0 lore raw

PATCH 1/1 perf/x86/intel/pt: Fix sampling using single range output

Deal with errata TGL052, ADL037 and RPL017 "Trace May Contain Incorrect Data When Configured With Single Range Output Larger Than 4KB" by disabling single range output whenever larger than 4KB.

Fixes 670638477aed ("perf/x86/intel/pt: Opportunistically use single range output mode")
---
 arch/x86/events/intel/pt.c | 9 +++++++++
 1 file changed, 9 insertions(+)
arch/x86/events/intel/pt.c +9 -0
index 82ef87e9a897..42a55794004a 100644
@@ -1263,6 +1263,15 @@ static int pt_buffer_try_single(struct pt_buffer *buf, int nr_pages)
 	if (1 << order != nr_pages)
 		goto out;
 
+	/*
+	 * Some processors cannot always support single range for more than
+	 * 4KB - refer errata TGL052, ADL037 and RPL017. Future processors might
+	 * also be affected, so for now rather than trying to keep track of
+	 * which ones, just disable it for all.
+	 */
+	if (nr_pages > 1)
+		goto out;
+
 	buf->single = true;
 	buf->nr_pages = nr_pages;
 	ret = 0;
Signature · 1 line
2.34.1
maintainer
#1 lore raw
Quoted 17 lines · Adrian Hunter

Deal with errata TGL052, ADL037 and RPL017 "Trace May Contain Incorrect Data When Configured With Single Range Output Larger Than 4KB" by disabling single range output whenever larger than 4KB.

Fixes 670638477aed ("perf/x86/intel/pt: Opportunistically use single range output mode")
---
 arch/x86/events/intel/pt.c | 9 +++++++++
 1 file changed, 9 insertions(+)
arch/x86/events/intel/pt.c +0 -0
index 82ef87e9a897..42a55794004a 100644
@@ -1263,6 +1263,15 @@ static int pt_buffer_try_single(struct pt_buffer *buf, int nr_pages)
 	if (1 << order != nr_pages)

arch/x86/events/intel/pt.c static int pt_buffer_try_single(struct pt_buffer *buf, int nr_pages) · line 1273

Quoted 10 lines
arch/x86/events/intel/pt.c +8 -0
 		goto out;
 
+	/*
+	 * Some processors cannot always support single range for more than
+	 * 4KB - refer errata TGL052, ADL037 and RPL017. Future processors might
+	 * also be affected, so for now rather than trying to keep track of
+	 * which ones, just disable it for all.
+	 */
+	if (nr_pages > 1)
+		goto out;

This effectively declares single-output-mode dead? Because I don't think anybody uses PT with a single 4K buffer.

author reviewer
#2 lore raw
Quoted 20 lines · Peter Zijlstra
Quoted 19 lines · Adrian Hunter

Deal with errata TGL052, ADL037 and RPL017 "Trace May Contain Incorrect Data When Configured With Single Range Output Larger Than 4KB" by disabling single range output whenever larger than 4KB.

Fixes 670638477aed ("perf/x86/intel/pt: Opportunistically use single range output mode")
---
 arch/x86/events/intel/pt.c | 9 +++++++++
 1 file changed, 9 insertions(+)
arch/x86/events/intel/pt.c +0 -0
index 82ef87e9a897..42a55794004a 100644
@@ -1263,6 +1263,15 @@ static int pt_buffer_try_single(struct pt_buffer *buf, int nr_pages)
 	if (1 << order != nr_pages)
 		goto out;
 
Quoted 11 lines

arch/x86/events/intel/pt.c static int pt_buffer_try_single(struct pt_buffer *buf, int nr_pages) · line 1273

Quoted 8 lines
arch/x86/events/intel/pt.c +8 -0
+	/*
+	 * Some processors cannot always support single range for more than
+	 * 4KB - refer errata TGL052, ADL037 and RPL017. Future processors might
+	 * also be affected, so for now rather than trying to keep track of
+	 * which ones, just disable it for all.
+	 */
+	if (nr_pages > 1)
+		goto out;

This effectively declares single-output-mode dead? Because I don't think anybody uses PT with a single 4K buffer.

4K is the default size for "sample mode" i.e. stuffing 4KB of Intel PT trace data into a PERF_RECORD_SAMPLE record that has sample_type bit PERF_SAMPLE_AUX

e.g.

$ perf record -vv --aux-sample -e '{intel_pt//u,cycles:u}' uname 2>err.txt
Linux
$ grep aux_sample_size err.txt
  aux_sample_size                  4096
$ 
maintainer
#3 lore raw
Quoted 36 lines · Adrian Hunter
Quoted 20 lines · Peter Zijlstra
Quoted 19 lines · Adrian Hunter

Deal with errata TGL052, ADL037 and RPL017 "Trace May Contain Incorrect Data When Configured With Single Range Output Larger Than 4KB" by disabling single range output whenever larger than 4KB.

Fixes 670638477aed ("perf/x86/intel/pt: Opportunistically use single range output mode")
---
 arch/x86/events/intel/pt.c | 9 +++++++++
 1 file changed, 9 insertions(+)
arch/x86/events/intel/pt.c +0 -0
index 82ef87e9a897..42a55794004a 100644
@@ -1263,6 +1263,15 @@ static int pt_buffer_try_single(struct pt_buffer *buf, int nr_pages)
 	if (1 << order != nr_pages)
 		goto out;
 
Quoted 11 lines

arch/x86/events/intel/pt.c static int pt_buffer_try_single(struct pt_buffer *buf, int nr_pages) · line 1273

Quoted 8 lines
arch/x86/events/intel/pt.c +8 -0
+	/*
+	 * Some processors cannot always support single range for more than
+	 * 4KB - refer errata TGL052, ADL037 and RPL017. Future processors might
+	 * also be affected, so for now rather than trying to keep track of
+	 * which ones, just disable it for all.
+	 */
+	if (nr_pages > 1)
+		goto out;

This effectively declares single-output-mode dead? Because I don't think anybody uses PT with a single 4K buffer.

4K is the default size for "sample mode" i.e. stuffing 4KB of Intel PT trace data into a PERF_RECORD_SAMPLE record that has sample_type bit PERF_SAMPLE_AUX

Quoted 6 lines

e.g.

$ perf record -vv --aux-sample -e '{intel_pt//u,cycles:u}' uname 2>err.txt
Linux
$ grep aux_sample_size err.txt
  aux_sample_size                  4096

Ah, ok. Not as bad then. Anyway, I'll go queue it for perf/urgent I suppose.

#4 lore raw
Quoted 39 lines · Peter Zijlstra
Quoted 38 lines · Adrian Hunter
Quoted 20 lines · Peter Zijlstra
Quoted 19 lines · Adrian Hunter

Deal with errata TGL052, ADL037 and RPL017 "Trace May Contain Incorrect Data When Configured With Single Range Output Larger Than 4KB" by disabling single range output whenever larger than 4KB.

Fixes 670638477aed ("perf/x86/intel/pt: Opportunistically use single range output mode")
---
 arch/x86/events/intel/pt.c | 9 +++++++++
 1 file changed, 9 insertions(+)
arch/x86/events/intel/pt.c +0 -0
index 82ef87e9a897..42a55794004a 100644
@@ -1263,6 +1263,15 @@ static int pt_buffer_try_single(struct pt_buffer *buf, int nr_pages)
 	if (1 << order != nr_pages)
 		goto out;
 
Quoted 11 lines

arch/x86/events/intel/pt.c static int pt_buffer_try_single(struct pt_buffer *buf, int nr_pages) · line 1273

Quoted 8 lines
arch/x86/events/intel/pt.c +8 -0
+	/*
+	 * Some processors cannot always support single range for more than
+	 * 4KB - refer errata TGL052, ADL037 and RPL017. Future processors might
+	 * also be affected, so for now rather than trying to keep track of
+	 * which ones, just disable it for all.
+	 */
+	if (nr_pages > 1)
+		goto out;

This effectively declares single-output-mode dead? Because I don't think anybody uses PT with a single 4K buffer.

4K is the default size for "sample mode" i.e. stuffing 4KB of Intel PT trace data into a PERF_RECORD_SAMPLE record that has sample_type bit PERF_SAMPLE_AUX

e.g.

Quoted 7 lines
Quoted 4 lines
$ perf record -vv --aux-sample -e '{intel_pt//u,cycles:u}' uname 2>err.txt
Linux
$ grep aux_sample_size err.txt
  aux_sample_size                  4096

Ah, ok. Not as bad then. Anyway, I'll go queue it for perf/urgent I suppose.

It would be better to only limit on the CPUs with the bug because switching buffers causes some extra latencies. So this patch may regress PT overhead or tail latencies.

-Andi

author reviewer
#5 lore raw
Quoted 46 lines · Andi Kleen
Quoted 44 lines · Peter Zijlstra
Quoted 42 lines · Adrian Hunter
Quoted 20 lines · Peter Zijlstra
Quoted 19 lines · Adrian Hunter

Deal with errata TGL052, ADL037 and RPL017 "Trace May Contain Incorrect Data When Configured With Single Range Output Larger Than 4KB" by disabling single range output whenever larger than 4KB.

Fixes 670638477aed ("perf/x86/intel/pt: Opportunistically use single range output mode")
---
 arch/x86/events/intel/pt.c | 9 +++++++++
 1 file changed, 9 insertions(+)
arch/x86/events/intel/pt.c +0 -0
index 82ef87e9a897..42a55794004a 100644
@@ -1263,6 +1263,15 @@ static int pt_buffer_try_single(struct pt_buffer *buf, int nr_pages)
 	if (1 << order != nr_pages)
 		goto out;
 
Quoted 11 lines

arch/x86/events/intel/pt.c static int pt_buffer_try_single(struct pt_buffer *buf, int nr_pages) · line 1273

Quoted 8 lines
arch/x86/events/intel/pt.c +8 -0
+	/*
+	 * Some processors cannot always support single range for more than
+	 * 4KB - refer errata TGL052, ADL037 and RPL017. Future processors might
+	 * also be affected, so for now rather than trying to keep track of
+	 * which ones, just disable it for all.
+	 */
+	if (nr_pages > 1)
+		goto out;

This effectively declares single-output-mode dead? Because I don't think anybody uses PT with a single 4K buffer.

4K is the default size for "sample mode" i.e. stuffing 4KB of Intel PT trace data into a PERF_RECORD_SAMPLE record that has sample_type bit PERF_SAMPLE_AUX

e.g.

$ perf record -vv --aux-sample -e '{intel_pt//u,cycles:u}' uname 2>err.txt
Linux
$ grep aux_sample_size err.txt
  aux_sample_size                  4096
Quoted 6 lines
Quoted 2 lines

Ah, ok. Not as bad then. Anyway, I'll go queue it for perf/urgent I suppose.

It would be better to only limit on the CPUs with the bug because switching buffers causes some extra latencies. So this patch may regress PT overhead or tail latencies.

I could whitelist CPUs that do not have the issue, because a blacklist would keep expanding, which would be a bit of a pain to maintain.

merge bot Applied
#6 lore raw
In reply to Adrian Hunter
The following commit has been merged into the perf/urgent branch of tip:

The following commit has been merged into the perf/urgent branch of tip:

Commit-ID:     ce0d998be9274dd3a3d971cbeaa6fe28fd2c3062
Gitweb:        https://git.kernel.org/tip/ce0d998be9274dd3a3d971cbeaa6fe28fd2c3062
Author:        Adrian Hunter <[email protected]>
AuthorDate:    Sat, 12 Nov 2022 17:15:08 +02:00
Committer:     Peter Zijlstra <[email protected]>
CommitterDate: Wed, 16 Nov 2022 10:12:59 +01:00

perf/x86/intel/pt: Fix sampling using single range output

Deal with errata TGL052, ADL037 and RPL017 "Trace May Contain Incorrect Data When Configured With Single Range Output Larger Than 4KB" by disabling single range output whenever larger than 4KB.

Fixes 670638477aed ("perf/x86/intel/pt: Opportunistically use single range output mode")
---
 arch/x86/events/intel/pt.c |  9 +++++++++
 1 file changed, 9 insertions(+)
arch/x86/events/intel/pt.c +9 -0
index 82ef87e..42a5579 100644
@@ -1263,6 +1263,15 @@ static int pt_buffer_try_single(struct pt_buffer *buf, int nr_pages)
 	if (1 << order != nr_pages)
 		goto out;
 
+	/*
+	 * Some processors cannot always support single range for more than
+	 * 4KB - refer errata TGL052, ADL037 and RPL017. Future processors might
+	 * also be affected, so for now rather than trying to keep track of
+	 * which ones, just disable it for all.
+	 */
+	if (nr_pages > 1)
+		goto out;
+
 	buf->single = true;
 	buf->nr_pages = nr_pages;
 	ret = 0;