[PATCH v2 0/3] perf record: fix multi-record Zstd compression

[PATCH v2 0/3] perf record: fix multi-record Zstd compression

[Dmitry Ilvokhin]

Patch 1 fixes a 'perf record -z' regression that aborts recording with
"Bad address" and produces an undecompressable perf.data.

While fixing it I found two more latent bugs in the same compressor,
zstd_compress_stream_to_records(): an output-buffer overrun and a broken
zstd-error fallback. Addressed in patches 2 and 3.

v1 -> v2:

- Patch 2 ("perf record: Avoid overrunning the zstd output buffer"):
  rework the overly strict output buffer check to pass the remaining dst
  size to process_comp_header() and bail only when a record won't fit
  (Sashiko).

v1: https://lore.kernel.org/all/cover.1781797544.git.d@ilvokhin.com/

Dmitry Ilvokhin (3):
  perf record: Fix multiple PERF_RECORD_COMPRESSED2 records per push
  perf record: Avoid overrunning the zstd output buffer
  perf record: Don't store raw data on zstd compression failure

 tools/perf/builtin-record.c                   | 48 ++++++++------
 .../record+zstd_comp_decomp_multi_record.sh   | 64 +++++++++++++++++++
 tools/perf/util/compress.h                    |  6 +-
 tools/perf/util/zstd.c                        | 28 +++++---
 4 files changed, 116 insertions(+), 30 deletions(-)
 create mode 100755 tools/perf/tests/shell/record+zstd_comp_decomp_multi_record.sh

-- 
2.53.0-Meta

[PATCH v2 1/3] perf record: Fix multiple PERF_RECORD_COMPRESSED2 records per push

[Dmitry Ilvokhin]

With Zstd compression enabled ('perf record -z'), a single mmap push
whose compressed output exceeds the maximum record size makes
zstd_compress_stream_to_records() emit several PERF_RECORD_COMPRESSED2
records back to back. record__pushfn() however rewrote only the first
record's header to describe the whole blob as one record:

  event->data_size   = compressed - sizeof(struct perf_record_compressed2);
  event->header.size = PERF_ALIGN(compressed, sizeof(u64));
  padding            = event->header.size - compressed;
  ...
  record__write(rec, map, &pad, padding);

perf_event_header::size is a __u16, so once the compressed blob no
longer fits in it the header.size assignment truncates and 'padding'
(size_t) underflows. write() is then handed that bogus length and fails
with EFAULT, aborting the recording:

  failed to write perf data, error: Bad address

The bytes that did reach the file are mis-framed, so reading it back
cannot be decompressed.

This is easy to hit with a high event rate and a large buffer, e.g.:

  perf record -z -F max -m 32M --per-thread -- perf test -w thloop 5 1

The single-record fixup is wrong by construction: because header.size is
16 bits a compressed record cannot exceed 64KB, so the compressor must
split a push into a chain of records, and the session reader already
consumes them as such.

Frame each record where it is produced instead: make
process_comp_header() set the per-record data_size, 8-byte-align
header.size and zero the trailing padding, and let record__pushfn()
write the resulting blob, as the AIO path already does. Reduce
max_record_size by sizeof(u64) so the per-record alignment padding
cannot push header.size past its u16 field.

There is no on-disk format change; a perf.data written by the fixed tool
is still read by existing perf.

Fixes: 208c0e168344 ("perf record: Add 8-byte aligned event type PERF_RECORD_COMPRESSED2")
Reported-by: Farid Zakaria <fmzakari@meta.com>
Signed-off-by: Dmitry Ilvokhin <d@ilvokhin.com>
Cc: stable@vger.kernel.org
---
 tools/perf/builtin-record.c                   | 41 ++++++------
 .../record+zstd_comp_decomp_multi_record.sh   | 64 +++++++++++++++++++
 tools/perf/util/zstd.c                        |  2 +-
 3 files changed, 87 insertions(+), 20 deletions(-)
 create mode 100755 tools/perf/tests/shell/record+zstd_comp_decomp_multi_record.sh

diff --git a/tools/perf/builtin-record.c b/tools/perf/builtin-record.c
index 4a5eba498c02..2562c3177eae 100644
--- a/tools/perf/builtin-record.c
+++ b/tools/perf/builtin-record.c
@@ -652,27 +652,14 @@ static int record__pushfn(struct mmap *map, void *to, void *bf, size_t size)
 	struct record *rec = to;
 
 	if (record__comp_enabled(rec)) {
-		struct perf_record_compressed2 *event = map->data;
-		size_t padding = 0;
-		u8 pad[8] = {0};
 		ssize_t compressed = zstd_compress(rec->session, map, map->data,
 						   mmap__mmap_len(map), bf, size);
 
 		if (compressed < 0)
 			return (int)compressed;
 
-		bf = event;
 		thread->samples++;
-
-		/*
-		 * The record from `zstd_compress` is not 8 bytes aligned, which would cause asan
-		 * error. We make it aligned here.
-		 */
-		event->data_size = compressed - sizeof(struct perf_record_compressed2);
-		event->header.size = PERF_ALIGN(compressed, sizeof(u64));
-		padding = event->header.size - compressed;
-		return record__write(rec, map, bf, compressed) ||
-		       record__write(rec, map, &pad, padding);
+		return record__write(rec, map, map->data, compressed);
 	}
 
 	thread->samples++;
@@ -1590,18 +1577,29 @@ static void record__adjust_affinity(struct record *rec, struct mmap *map)
 	}
 }
 
-static size_t process_comp_header(void *record, size_t increment)
+/*
+ * Called once with data_size == 0 to start a record, then once with
+ * data_size == compressed payload size to finalize and 8-byte-pad it
+ * (unaligned records trip ASan in the reader).
+ */
+static size_t process_comp_header(void *record, size_t data_size)
 {
 	struct perf_record_compressed2 *event = record;
 	size_t size = sizeof(*event);
 
-	if (increment) {
-		event->header.size += increment;
-		return increment;
+	if (data_size) {
+		size_t padding;
+
+		event->data_size = data_size;
+		event->header.size = PERF_ALIGN(size + data_size, sizeof(u64));
+		padding = event->header.size - size - data_size;
+		memset(record + size + data_size, 0, padding);
+		return data_size + padding;
 	}
 
 	event->header.type = PERF_RECORD_COMPRESSED2;
 	event->header.size = size;
+	event->data_size = 0;
 
 	return size;
 }
@@ -1610,7 +1608,12 @@ static ssize_t zstd_compress(struct perf_session *session, struct mmap *map,
 			    void *dst, size_t dst_size, void *src, size_t src_size)
 {
 	ssize_t compressed;
-	size_t max_record_size = PERF_SAMPLE_MAX_SIZE - sizeof(struct perf_record_compressed2) - 1;
+	/*
+	 * Reserve space so per-record PERF_ALIGN() padding keeps header.size
+	 * within u16.
+	 */
+	size_t max_record_size = PERF_SAMPLE_MAX_SIZE
+		- sizeof(struct perf_record_compressed2) - sizeof(u64);
 	struct zstd_data *zstd_data = &session->zstd_data;
 
 	if (map && map->file)
diff --git a/tools/perf/tests/shell/record+zstd_comp_decomp_multi_record.sh b/tools/perf/tests/shell/record+zstd_comp_decomp_multi_record.sh
new file mode 100755
index 000000000000..42efe7260def
--- /dev/null
+++ b/tools/perf/tests/shell/record+zstd_comp_decomp_multi_record.sh
@@ -0,0 +1,64 @@
+#!/bin/bash
+# Zstd perf.data compression/decompression of multi-record data
+
+# SPDX-License-Identifier: GPL-2.0
+
+perfdata=$(mktemp /tmp/__perf_test.perf.data.XXXXX)
+recout=$(mktemp /tmp/__perf_test.zstd.rec.XXXXX)
+injout=$(mktemp /tmp/__perf_test.zstd.inj.XXXXX)
+perf_tool=perf
+
+cleanup() {
+	rm -f "${perfdata}" "${perfdata}".old "${perfdata}".decomp "${recout}" "${injout}"
+}
+trap cleanup EXIT TERM INT
+
+skip_if_no_z_record() {
+	$perf_tool record -h 2>&1 | grep -q -- '-z, --compression-level'
+}
+
+collect_z_record() {
+	echo "Collecting compressed record file:"
+	[ "$(uname -m)" != s390x ] && gflag='-g'
+	$perf_tool record -o "${perfdata}" $gflag -z -F max -m 32M --per-thread -- \
+		$perf_tool test -w thloop 5 1 \
+		>/dev/null 2>"${recout}"
+}
+
+check_record() {
+	echo "Checking record did not fail to write data:"
+	if grep -q "failed to write perf data" "${recout}"; then
+		cat "${recout}"
+		return 1
+	fi
+}
+
+check_decompress() {
+	echo "Checking compressed file decompresses cleanly:"
+	if ! $perf_tool inject -i "${perfdata}" -o "${perfdata}".decomp 2>"${injout}"; then
+		cat "${injout}"
+		return 1
+	fi
+	if grep -Eqi "decompress|corrupt|failed to process type" "${injout}"; then
+		cat "${injout}"
+		return 1
+	fi
+}
+
+skip_if_no_z_record || exit 2
+collect_z_record
+check_record || exit 1
+
+# Need >1 record, else the multi-record path wasn't exercised.
+# Skip rather than pass/fail spuriously.
+nr=$($perf_tool report -i "${perfdata}" --stats 2>/dev/null |
+	awk '/COMPRESSED2 events:/ { print $3 }')
+if [ -z "${nr}" ] || [ "${nr}" -lt 2 ]; then
+	echo "less than two compressed records (${nr:-0}), skipping"
+	exit 2
+fi
+echo "Produced ${nr} compressed records"
+
+check_decompress
+err=$?
+exit $err
diff --git a/tools/perf/util/zstd.c b/tools/perf/util/zstd.c
index 57027e0ac7b6..1955fa2431d1 100644
--- a/tools/perf/util/zstd.c
+++ b/tools/perf/util/zstd.c
@@ -30,7 +30,7 @@ int zstd_fini(struct zstd_data *data)
 
 ssize_t zstd_compress_stream_to_records(struct zstd_data *data, void *dst, size_t dst_size,
 				       void *src, size_t src_size, size_t max_record_size,
-				       size_t process_header(void *record, size_t increment))
+				       size_t process_header(void *record, size_t data_size))
 {
 	size_t ret, size, compressed = 0;
 	ZSTD_inBuffer input = { src, src_size, 0 };
-- 
2.53.0-Meta

Re: [PATCH v2 1/3] perf record: Fix multiple PERF_RECORD_COMPRESSED2 records per push

[sashiko-bot]

Thank you for your contribution! Sashiko AI review found 1 potential issue(s) to consider:
- [Low] Missing explicit `#include <string.h>` for `memset` in `builtin-record.c`.
--

commit cb8ac7d313a36a42601f85ffc5fb9a9061002bc0
Author: Dmitry Ilvokhin <d@ilvokhin.com>

perf record: Fix multiple PERF_RECORD_COMPRESSED2 records per push

This commit frames compressed records individually when produced rather
than rewriting a single header for an entire blob. This prevents truncation
of the header.size field when a compressed blob exceeds 64KB.

> diff --git a/tools/perf/builtin-record.c b/tools/perf/builtin-record.c
> index 4a5eba498c025..2562c3177eae1 100644
> --- a/tools/perf/builtin-record.c
> +++ b/tools/perf/builtin-record.c

[ ... ]

> @@ -1590,18 +1577,29 @@ static void record__adjust_affinity(struct record *rec, struct mmap *map)
>  	}
>  }
>  
> -static size_t process_comp_header(void *record, size_t increment)
> +/*
> + * Called once with data_size == 0 to start a record, then once with
> + * data_size == compressed payload size to finalize and 8-byte-pad it
> + * (unaligned records trip ASan in the reader).
> + */
> +static size_t process_comp_header(void *record, size_t data_size)
>  {
>  	struct perf_record_compressed2 *event = record;
>  	size_t size = sizeof(*event);
>  
> -	if (increment) {
> -		event->header.size += increment;
> -		return increment;
> +	if (data_size) {
> +		size_t padding;
> +
> +		event->data_size = data_size;
> +		event->header.size = PERF_ALIGN(size + data_size, sizeof(u64));
> +		padding = event->header.size - size - data_size;
> +		memset(record + size + data_size, 0, padding);

[Severity: Low]
Does this file need an explicit include for <string.h>?

Relying on implicit header inclusion for POSIX libc functions like memset can
cause build failures on systems using musl libc, where implicit header graphs
differ from glibc.

> +		return data_size + padding;
>  	}
>  
>  	event->header.type = PERF_RECORD_COMPRESSED2;

-- 
Sashiko AI review · https://sashiko.dev/#/patchset/cover.1782743187.git.d@ilvokhin.com?part=1

Re: [PATCH v2 1/3] perf record: Fix multiple PERF_RECORD_COMPRESSED2 records per push

[Namhyung Kim]

Hello,

On Tue, Jun 30, 2026 at 07:17:00AM +0000, Dmitry Ilvokhin wrote:
> With Zstd compression enabled ('perf record -z'), a single mmap push
> whose compressed output exceeds the maximum record size makes
> zstd_compress_stream_to_records() emit several PERF_RECORD_COMPRESSED2
> records back to back. record__pushfn() however rewrote only the first
> record's header to describe the whole blob as one record:
> 
>   event->data_size   = compressed - sizeof(struct perf_record_compressed2);
>   event->header.size = PERF_ALIGN(compressed, sizeof(u64));
>   padding            = event->header.size - compressed;
>   ...
>   record__write(rec, map, &pad, padding);
> 
> perf_event_header::size is a __u16, so once the compressed blob no
> longer fits in it the header.size assignment truncates and 'padding'
> (size_t) underflows. write() is then handed that bogus length and fails
> with EFAULT, aborting the recording:
> 
>   failed to write perf data, error: Bad address
> 
> The bytes that did reach the file are mis-framed, so reading it back
> cannot be decompressed.
> 
> This is easy to hit with a high event rate and a large buffer, e.g.:
> 
>   perf record -z -F max -m 32M --per-thread -- perf test -w thloop 5 1
> 
> The single-record fixup is wrong by construction: because header.size is
> 16 bits a compressed record cannot exceed 64KB, so the compressor must
> split a push into a chain of records, and the session reader already
> consumes them as such.
> 
> Frame each record where it is produced instead: make
> process_comp_header() set the per-record data_size, 8-byte-align
> header.size and zero the trailing padding, and let record__pushfn()
> write the resulting blob, as the AIO path already does. Reduce
> max_record_size by sizeof(u64) so the per-record alignment padding
> cannot push header.size past its u16 field.
> 
> There is no on-disk format change; a perf.data written by the fixed tool
> is still read by existing perf.

Thanks for working on this!

> 
> Fixes: 208c0e168344 ("perf record: Add 8-byte aligned event type PERF_RECORD_COMPRESSED2")
> Reported-by: Farid Zakaria <fmzakari@meta.com>
> Signed-off-by: Dmitry Ilvokhin <d@ilvokhin.com>
> Cc: stable@vger.kernel.org
> ---
>  tools/perf/builtin-record.c                   | 41 ++++++------
>  .../record+zstd_comp_decomp_multi_record.sh   | 64 +++++++++++++++++++
>  tools/perf/util/zstd.c                        |  2 +-
>  3 files changed, 87 insertions(+), 20 deletions(-)
>  create mode 100755 tools/perf/tests/shell/record+zstd_comp_decomp_multi_record.sh
> 
> diff --git a/tools/perf/builtin-record.c b/tools/perf/builtin-record.c
> index 4a5eba498c02..2562c3177eae 100644
> --- a/tools/perf/builtin-record.c
> +++ b/tools/perf/builtin-record.c
> @@ -652,27 +652,14 @@ static int record__pushfn(struct mmap *map, void *to, void *bf, size_t size)
>  	struct record *rec = to;
>  
>  	if (record__comp_enabled(rec)) {
> -		struct perf_record_compressed2 *event = map->data;
> -		size_t padding = 0;
> -		u8 pad[8] = {0};
>  		ssize_t compressed = zstd_compress(rec->session, map, map->data,
>  						   mmap__mmap_len(map), bf, size);
>  
>  		if (compressed < 0)
>  			return (int)compressed;
>  
> -		bf = event;
>  		thread->samples++;
> -
> -		/*
> -		 * The record from `zstd_compress` is not 8 bytes aligned, which would cause asan
> -		 * error. We make it aligned here.
> -		 */
> -		event->data_size = compressed - sizeof(struct perf_record_compressed2);
> -		event->header.size = PERF_ALIGN(compressed, sizeof(u64));
> -		padding = event->header.size - compressed;
> -		return record__write(rec, map, bf, compressed) ||
> -		       record__write(rec, map, &pad, padding);
> +		return record__write(rec, map, map->data, compressed);
>  	}
>  
>  	thread->samples++;
> @@ -1590,18 +1577,29 @@ static void record__adjust_affinity(struct record *rec, struct mmap *map)
>  	}
>  }
>  
> -static size_t process_comp_header(void *record, size_t increment)
> +/*
> + * Called once with data_size == 0 to start a record, then once with
> + * data_size == compressed payload size to finalize and 8-byte-pad it
> + * (unaligned records trip ASan in the reader).
> + */
> +static size_t process_comp_header(void *record, size_t data_size)
>  {
>  	struct perf_record_compressed2 *event = record;
>  	size_t size = sizeof(*event);
>  
> -	if (increment) {
> -		event->header.size += increment;
> -		return increment;
> +	if (data_size) {
> +		size_t padding;
> +
> +		event->data_size = data_size;
> +		event->header.size = PERF_ALIGN(size + data_size, sizeof(u64));
> +		padding = event->header.size - size - data_size;
> +		memset(record + size + data_size, 0, padding);
> +		return data_size + padding;
>  	}
>  
>  	event->header.type = PERF_RECORD_COMPRESSED2;
>  	event->header.size = size;
> +	event->data_size = 0;
>  
>  	return size;
>  }
> @@ -1610,7 +1608,12 @@ static ssize_t zstd_compress(struct perf_session *session, struct mmap *map,
>  			    void *dst, size_t dst_size, void *src, size_t src_size)
>  {
>  	ssize_t compressed;
> -	size_t max_record_size = PERF_SAMPLE_MAX_SIZE - sizeof(struct perf_record_compressed2) - 1;
> +	/*
> +	 * Reserve space so per-record PERF_ALIGN() padding keeps header.size
> +	 * within u16.
> +	 */
> +	size_t max_record_size = PERF_SAMPLE_MAX_SIZE
> +		- sizeof(struct perf_record_compressed2) - sizeof(u64);
>  	struct zstd_data *zstd_data = &session->zstd_data;
>  
>  	if (map && map->file)
> diff --git a/tools/perf/tests/shell/record+zstd_comp_decomp_multi_record.sh b/tools/perf/tests/shell/record+zstd_comp_decomp_multi_record.sh
> new file mode 100755
> index 000000000000..42efe7260def
> --- /dev/null
> +++ b/tools/perf/tests/shell/record+zstd_comp_decomp_multi_record.sh
> @@ -0,0 +1,64 @@
> +#!/bin/bash
> +# Zstd perf.data compression/decompression of multi-record data
> +
> +# SPDX-License-Identifier: GPL-2.0

Can you please remove the blank line here?

Also, this series cannot apply anymore.  Please rebase onto the latest
perf-tools-next.

Thanks,
Namhyung

> +
> +perfdata=$(mktemp /tmp/__perf_test.perf.data.XXXXX)
> +recout=$(mktemp /tmp/__perf_test.zstd.rec.XXXXX)
> +injout=$(mktemp /tmp/__perf_test.zstd.inj.XXXXX)
> +perf_tool=perf
> +
> +cleanup() {
> +	rm -f "${perfdata}" "${perfdata}".old "${perfdata}".decomp "${recout}" "${injout}"
> +}
> +trap cleanup EXIT TERM INT
> +
> +skip_if_no_z_record() {
> +	$perf_tool record -h 2>&1 | grep -q -- '-z, --compression-level'
> +}
> +
> +collect_z_record() {
> +	echo "Collecting compressed record file:"
> +	[ "$(uname -m)" != s390x ] && gflag='-g'
> +	$perf_tool record -o "${perfdata}" $gflag -z -F max -m 32M --per-thread -- \
> +		$perf_tool test -w thloop 5 1 \
> +		>/dev/null 2>"${recout}"
> +}
> +
> +check_record() {
> +	echo "Checking record did not fail to write data:"
> +	if grep -q "failed to write perf data" "${recout}"; then
> +		cat "${recout}"
> +		return 1
> +	fi
> +}
> +
> +check_decompress() {
> +	echo "Checking compressed file decompresses cleanly:"
> +	if ! $perf_tool inject -i "${perfdata}" -o "${perfdata}".decomp 2>"${injout}"; then
> +		cat "${injout}"
> +		return 1
> +	fi
> +	if grep -Eqi "decompress|corrupt|failed to process type" "${injout}"; then
> +		cat "${injout}"
> +		return 1
> +	fi
> +}
> +
> +skip_if_no_z_record || exit 2
> +collect_z_record
> +check_record || exit 1
> +
> +# Need >1 record, else the multi-record path wasn't exercised.
> +# Skip rather than pass/fail spuriously.
> +nr=$($perf_tool report -i "${perfdata}" --stats 2>/dev/null |
> +	awk '/COMPRESSED2 events:/ { print $3 }')
> +if [ -z "${nr}" ] || [ "${nr}" -lt 2 ]; then
> +	echo "less than two compressed records (${nr:-0}), skipping"
> +	exit 2
> +fi
> +echo "Produced ${nr} compressed records"
> +
> +check_decompress
> +err=$?
> +exit $err
> diff --git a/tools/perf/util/zstd.c b/tools/perf/util/zstd.c
> index 57027e0ac7b6..1955fa2431d1 100644
> --- a/tools/perf/util/zstd.c
> +++ b/tools/perf/util/zstd.c
> @@ -30,7 +30,7 @@ int zstd_fini(struct zstd_data *data)
>  
>  ssize_t zstd_compress_stream_to_records(struct zstd_data *data, void *dst, size_t dst_size,
>  				       void *src, size_t src_size, size_t max_record_size,
> -				       size_t process_header(void *record, size_t increment))
> +				       size_t process_header(void *record, size_t data_size))
>  {
>  	size_t ret, size, compressed = 0;
>  	ZSTD_inBuffer input = { src, src_size, 0 };
> -- 
> 2.53.0-Meta
> 

Re: [PATCH v2 1/3] perf record: Fix multiple PERF_RECORD_COMPRESSED2 records per push

[Dmitry Ilvokhin]

On Sat, Jul 04, 2026 at 10:08:03AM -0700, Namhyung Kim wrote:
> > diff --git a/tools/perf/tests/shell/record+zstd_comp_decomp_multi_record.sh b/tools/perf/tests/shell/record+zstd_comp_decomp_multi_record.sh
> > new file mode 100755
> > index 000000000000..42efe7260def
> > --- /dev/null
> > +++ b/tools/perf/tests/shell/record+zstd_comp_decomp_multi_record.sh
> > @@ -0,0 +1,64 @@
> > +#!/bin/bash
> > +# Zstd perf.data compression/decompression of multi-record data
> > +
> > +# SPDX-License-Identifier: GPL-2.0
> 
> Can you please remove the blank line here?
> 
> Also, this series cannot apply anymore.  Please rebase onto the latest
> perf-tools-next.
> 

Thanks for taking a look, Namhyung.

Absolutely. I'll fix it up, rebase on latest perf-tools-next and respin
the series.

[PATCH v2 2/3] perf record: Avoid overrunning the zstd output buffer

[Dmitry Ilvokhin]

zstd_compress_stream_to_records() compresses one mmap chunk into 'dst'
as a chain of records, decrementing 'dst_size' as each record's header
and compressed payload are written, but it never checks that the space
left can still hold a record before writing one.

When the compressed output does not fit in 'dst', 'dst_size -= size'
underflows ('dst_size' is a size_t), the output size is then taken as
max_record_size again, and ZSTD_compressStream() writes past the end
of 'dst'.

Pass the space left in 'dst' to process_comp_header() and let it return
-1 when the record header or its 8-byte alignment padding would not fit.
Stop compressing on -1 instead of writing past 'dst'.

Fixes: f24c1d7523e6 ("perf tools: Introduce Zstd streaming based compression API")
Signed-off-by: Dmitry Ilvokhin <d@ilvokhin.com>
---
 tools/perf/builtin-record.c | 11 +++++++++--
 tools/perf/util/compress.h  |  6 ++++--
 tools/perf/util/zstd.c      | 23 ++++++++++++++++++-----
 3 files changed, 31 insertions(+), 9 deletions(-)

diff --git a/tools/perf/builtin-record.c b/tools/perf/builtin-record.c
index 2562c3177eae..fc89ccc78d4d 100644
--- a/tools/perf/builtin-record.c
+++ b/tools/perf/builtin-record.c
@@ -1581,8 +1581,10 @@ static void record__adjust_affinity(struct record *rec, struct mmap *map)
  * Called once with data_size == 0 to start a record, then once with
  * data_size == compressed payload size to finalize and 8-byte-pad it
  * (unaligned records trip ASan in the reader).
+ * Returns the bytes written, or -1 if it won't fit.
  */
-static size_t process_comp_header(void *record, size_t data_size)
+static ssize_t process_comp_header(void *record, size_t dst_size,
+				   size_t data_size)
 {
 	struct perf_record_compressed2 *event = record;
 	size_t size = sizeof(*event);
@@ -1593,10 +1595,15 @@ static size_t process_comp_header(void *record, size_t data_size)
 		event->data_size = data_size;
 		event->header.size = PERF_ALIGN(size + data_size, sizeof(u64));
 		padding = event->header.size - size - data_size;
+		if (padding > dst_size)
+			return -1;
 		memset(record + size + data_size, 0, padding);
-		return data_size + padding;
+		return padding;
 	}
 
+	if (size > dst_size)
+		return -1;
+
 	event->header.type = PERF_RECORD_COMPRESSED2;
 	event->header.size = size;
 	event->data_size = 0;
diff --git a/tools/perf/util/compress.h b/tools/perf/util/compress.h
index 6cfecfca16f2..ec6c38129e24 100644
--- a/tools/perf/util/compress.h
+++ b/tools/perf/util/compress.h
@@ -54,7 +54,8 @@ int zstd_fini(struct zstd_data *data);
 
 ssize_t zstd_compress_stream_to_records(struct zstd_data *data, void *dst, size_t dst_size,
 				       void *src, size_t src_size, size_t max_record_size,
-				       size_t process_header(void *record, size_t increment));
+				       ssize_t process_header(void *record, size_t dst_size,
+							      size_t data_size));
 
 size_t zstd_decompress_stream(struct zstd_data *data, void *src, size_t src_size,
 			      void *dst, size_t dst_size);
@@ -75,7 +76,8 @@ ssize_t zstd_compress_stream_to_records(struct zstd_data *data __maybe_unused,
 				       void *dst __maybe_unused, size_t dst_size __maybe_unused,
 				       void *src __maybe_unused, size_t src_size __maybe_unused,
 				       size_t max_record_size __maybe_unused,
-				       size_t process_header(void *record, size_t increment) __maybe_unused)
+				       ssize_t process_header(void *record, size_t dst_size,
+							      size_t data_size) __maybe_unused)
 {
 	return 0;
 }
diff --git a/tools/perf/util/zstd.c b/tools/perf/util/zstd.c
index 1955fa2431d1..04910dfed84f 100644
--- a/tools/perf/util/zstd.c
+++ b/tools/perf/util/zstd.c
@@ -30,9 +30,11 @@ int zstd_fini(struct zstd_data *data)
 
 ssize_t zstd_compress_stream_to_records(struct zstd_data *data, void *dst, size_t dst_size,
 				       void *src, size_t src_size, size_t max_record_size,
-				       size_t process_header(void *record, size_t data_size))
+				       ssize_t process_header(void *record, size_t dst_size,
+							      size_t data_size))
 {
-	size_t ret, size, compressed = 0;
+	size_t ret, compressed = 0;
+	ssize_t size;
 	ZSTD_inBuffer input = { src, src_size, 0 };
 	ZSTD_outBuffer output;
 	void *record;
@@ -54,7 +56,11 @@ ssize_t zstd_compress_stream_to_records(struct zstd_data *data, void *dst, size_
 
 	while (input.pos < input.size) {
 		record = dst;
-		size = process_header(record, 0);
+		size = process_header(record, dst_size, 0);
+		if (size < 0) {
+			pr_err("failed to compress: output buffer too small\n");
+			return -1;
+		}
 		compressed += size;
 		dst += size;
 		dst_size -= size;
@@ -68,8 +74,15 @@ ssize_t zstd_compress_stream_to_records(struct zstd_data *data, void *dst, size_
 			memcpy(dst, src, src_size);
 			return src_size;
 		}
-		size = output.pos;
-		size = process_header(record, size);
+		compressed += output.pos;
+		dst += output.pos;
+		dst_size -= output.pos;
+
+		size = process_header(record, dst_size, output.pos);
+		if (size < 0) {
+			pr_err("failed to compress: output buffer too small for padding\n");
+			return -1;
+		}
 		compressed += size;
 		dst += size;
 		dst_size -= size;
-- 
2.53.0-Meta

[PATCH v2 3/3] perf record: Don't store raw data on zstd compression failure

[Dmitry Ilvokhin]

On a ZSTD_compressStream() error, zstd_compress_stream_to_records()
falls back to copying the uncompressed input into the output buffer and
returning its size. That cannot work:

- a COMPRESSED2 payload is always fed to zstd_decompress_stream(), so
  raw bytes placed there can never be read back.

- the record header is never finalized: process_header() is not called
  to write the real header.size and data_size, so they keep their
  initial values and do not match the bytes written.

- the copy is unbounded and can write past the output buffer.

Propagate the error instead.

Fixes: f24c1d7523e6 ("perf tools: Introduce Zstd streaming based compression API")
Signed-off-by: Dmitry Ilvokhin <d@ilvokhin.com>
---
 tools/perf/util/zstd.c | 5 +----
 1 file changed, 1 insertion(+), 4 deletions(-)

diff --git a/tools/perf/util/zstd.c b/tools/perf/util/zstd.c
index 04910dfed84f..cbb150b0f807 100644
--- a/tools/perf/util/zstd.c
+++ b/tools/perf/util/zstd.c
@@ -1,7 +1,5 @@
 // SPDX-License-Identifier: GPL-2.0
 
-#include <string.h>
-
 #include "util/compress.h"
 #include "util/debug.h"
 
@@ -71,8 +69,7 @@ ssize_t zstd_compress_stream_to_records(struct zstd_data *data, void *dst, size_
 		if (ZSTD_isError(ret)) {
 			pr_err("failed to compress %ld bytes: %s\n",
 				(long)src_size, ZSTD_getErrorName(ret));
-			memcpy(dst, src, src_size);
-			return src_size;
+			return -1;
 		}
 		compressed += output.pos;
 		dst += output.pos;
-- 
2.53.0-Meta