Description of problem: # perf record -a -g --call-graph=dwarf -- nbdkit -U - sparse-random size=1T --run 'nbdcopy "$uri" "$uri"' # perf script > out.perf BFD: DWARF error: invalid abstract instance DIE ref BFD: DWARF error: info pointer extends beyond end of attributes BFD: DWARF error: could not find abbrev number 127 BFD: DWARF error: info pointer extends beyond end of attributes BFD: DWARF error: info pointer extends beyond end of attributes BFD: DWARF error: info pointer extends beyond end of attributes BFD: DWARF error: info pointer extends beyond end of attributes BFD: DWARF error: could not find abbrev number 708 BFD: DWARF error: mangled line number section (bad file number) BFD: DWARF error: info pointer extends beyond end of attributes BFD: DWARF error: info pointer extends beyond end of attributes BFD: DWARF error: info pointer extends beyond end of attributes BFD: DWARF error: offset (823271579) greater than or equal to .debug_str size (58644) BFD: DWARF error: could not find abbrev number 114 BFD: DWARF error: offset (10592310) greater than or equal to .debug_str size (58644) BFD: DWARF error: info pointer extends beyond end of attributes BFD: DWARF error: info pointer extends beyond end of attributes BFD: DWARF error: offset (91691187) greater than or equal to .debug_str size (58644) BFD: DWARF error: offset (805306373) greater than or equal to .debug_str size (43389) BFD: DWARF error: mangled line number section (bad file number) BFD: DWARF error: could not find abbrev number 336 [etc] Version-Release number of selected component (if applicable): perf-5.13.0-0.rc2.git0.1.fc35.x86_64 nbdkit-debuginfo-1.25.7-1.fc35.x86_64 libnbd-debuginfo-1.7.11-1.fc35.x86_64 How reproducible: 100% Steps to Reproduce: 1. See above Additional info: Something to do with DWARF5 debugging information?
An even simpler reproducer: # perf record -a -g --call-graph=dwarf -- nbdkit --version # perf script > out.perf detected invalid bpf_prog_info BFD: DWARF error: offset (4722738) greater than or equal to .debug_str size (6255) BFD: DWARF error: could not find abbrev number 80 BFD: DWARF error: offset (18444) greater than or equal to .debug_str size (6255) BFD: DWARF error: offset (9247694) greater than or equal to .debug_str size (6255) BFD: DWARF error: invalid abstract instance DIE ref BFD: DWARF error: offset (21564085) greater than or equal to .debug_str size (160981) BFD: DWARF error: offset (2165245130) greater than or equal to .debug_str size (160981) BFD: DWARF error: offset (395643904) greater than or equal to .debug_str size (160981) BFD: DWARF error: offset (51249821) greater than or equal to .debug_str size (160981) BFD: DWARF error: offset (386400256) greater than or equal to .debug_str size (160981) BFD: DWARF error: could not find abbrev number 287 [etc for pages and pages]
Possibly this linker bug which would lead to corrupted dwarf5 info? https://sourceware.org/bugzilla/show_bug.cgi?id=27587
Also happens in Fedora 34 with: perf-5.12.5-300.fc34.x86_64 nbdkit-1.25.7-1.fc34.x86_64 nbdkit-debuginfo-1.25.7-1.fc34.x86_64 libnbd-1.7.9-1.fc34.x86_64 libnbd-debuginfo-1.7.9-1.fc34.x86_64
It is more likely that perf is expecting dwarf5 and we switched back to dwarf4 because crash doesn't support dwarf5 yet. Let me see what I can find. We flipped CONFIG_DEBUG_INFO_DWARF_TOOLCHAIN_DEFAULT sometime between rc1 and rc2.
Sorry, was reading that wrong. Likely not what I mentioned above. Looking at binutils, that patch is not included in any builds for rawhide or fc34, so it is possible that is the cause.
I filed this bug yesterday and I thought at the time that perf script was failing to produce any output at all. (The messages look a lot like errors, so I didn't bother to check the output.) It turns out that perf script does in fact work to a large extent despite all of these messages. There are some mangled symbols which appear as "??", but by and large perf is producing enough output to still produce meaningful flamegraphs etc. So it's not a total failure to read DWARF5, but some kind of corner case either in one library we're using or in a small subset of DWARF5 data.
This bug appears to have been reported against 'rawhide' during the Fedora 35 development cycle. Changing version to 35.
This message is a reminder that Fedora Linux 35 is nearing its end of life. Fedora will stop maintaining and issuing updates for Fedora Linux 35 on 2022-12-13. It is Fedora's policy to close all bug reports from releases that are no longer maintained. At that time this bug will be closed as EOL if it remains open with a 'version' of '35'. Package Maintainer: If you wish for this bug to remain open because you plan to fix it in a currently maintained version, change the 'version' to a later Fedora Linux version. Thank you for reporting this issue and we are sorry that we were not able to fix it before Fedora Linux 35 is end of life. If you would still like to see this bug fixed and are able to reproduce it against a later version of Fedora Linux, you are encouraged to change the 'version' to a later version prior to this bug being closed.
Fedora Linux 35 entered end-of-life (EOL) status on 2022-12-13. Fedora Linux 35 is no longer maintained, which means that it will not receive any further security or bug fix updates. As a result we are closing this bug. If you can reproduce this bug against a currently maintained version of Fedora Linux please feel free to reopen this bug against that version. Note that the version field may be hidden. Click the "Show advanced fields" button if you do not see the version field. If you are unable to reopen this bug, please file a new report against an active release. Thank you for reporting this bug and we are sorry it could not be fixed.