Regression: perf script report flamegraph dumps core

Regression: perf script report flamegraph dumps core

[Thomas Richter]

Hi all,

Running the following commands on 6.17.0rc1 produces a core dump
sometimes (actually almost on every invocation on my system):

 # perf record -- find / -xdev > /dev/null
 [ perf record: Woken up 2 times to write data ]
 [ perf record: Captured and wrote 0.104 MB perf.data (2321 samples) ]
 #
 # perf script -s /usr/libexec/perf-core/scripts/python/flamegraph.py -i perf.data
 Segmentation fault (core dumped)

 # perf script report flamegraph
 /usr/libexec/perf-core/scripts/python/bin/flamegraph-report: line 3: \
 206665 Segmentation fault      (core dumped)
 #


However when option -d is specified, the commands succeed:

 # perf script report flamegraph -d
 Misordered timestamps: 0
 dumping data to flamegraph.html

 # ll flamegraph.html
 -rw-r--r--. 1 root root 69686 Aug 13 15:02 flamegraph.html

 # perf script -s /usr/libexec/perf-core/scripts/python/flamegraph.py -d -i perf.data
 Misordered timestamps: 0
 dumping data to flamegraph.html

 # ll flamegraph.html
 -rw-r--r--. 1 root root 69686 Aug 13 15:02 flamegraph.html

Again executing
 # perf script -s /usr/libexec/perf-core/scripts/python/flamegraph.py -D -i perf.data
reveals
 ....
 many entries
 ...
 23874506376796 0x42c0 [0x28]: PERF_RECORD_SAMPLE(IP, 0x1): \
                 209863/209863: 0x1da6f281d0c period: 1250000 addr: 0
 ... thread: find:209863
 ...... dso: /usr/lib/debug/lib/modules/6.17.0-20250812.rc1.git4.2810b7c15886.63.fc42.s390x/vmlinux

 0x42e8@perf.data [0x28]: event: 9
.
. ... raw event: size 40 bytes
.  0000:  00 00 00 09 00 01 00 28 00 00 01 d9 ee e3 70 ea  .......(......p.
.  0010:  00 03 33 c7 00 03 33 c7 00 00 15 b6 b7 78 c8 c2  ..3...3......x..
.  0020:  00 00 00 00 00 13 12 d0                          ........

23874506377410 0x42e8 [0x28]: PERF_RECORD_SAMPLE(IP, 0x1): \
                209863/209863: 0x1d9eee370ea period: 1250000 addr: 0
 ... thread: find:209863
 ...... dso:
Segmentation fault (core dumped)


Is this a known issue? Just asking before I start digging.

Thanks a lot.
-- 
Thomas Richter, Dept 3303, IBM s390 Linux Development, Boeblingen, Germany
--
IBM Deutschland Research & Development GmbH

Vorsitzender des Aufsichtsrats: Wolfgang Wendt

Geschäftsführung: David Faller

Sitz der Gesellschaft: Böblingen / Registergericht: Amtsgericht Stuttgart, HRB 243294