"etags" from universal-ctags segmentation fault on launch

Bug #1924821 reported by Stephen Brennan
8
This bug affects 1 person
Affects Status Importance Assigned to Milestone
universal-ctags (Ubuntu)
Confirmed
Undecided
Unassigned

Bug Description

Installed universal-ctags and ran the "etags" command (which update-alternatives automatically provided from the package). It immediately crashed with a segmentation fault. I installed debug symbols and ran it within valgrind:

$ valgrind etags
==9733== Memcheck, a memory error detector
==9733== Copyright (C) 2002-2017, and GNU GPL'd, by Julian Seward et al.
==9733== Using Valgrind-3.16.1 and LibVEX; rerun with -h for copyright info
==9733== Command: etags
==9733==
==9733== Invalid read of size 8
==9733== at 0x1349AA: enableLanguage (parse.c:1745)
==9733== by 0x12E2F8: setEtagsMode (options.c:813)
==9733== by 0x12E2F8: testEtagsInvocation (options.c:828)
==9733== by 0x12ACF2: ctags_cli_main (main.c:577)
==9733== by 0x4A99CB1: (below main) (libc-start.c:314)
==9733== Address 0x58 is not stack'd, malloc'd or (recently) free'd
==9733==
==9733==
==9733== Process terminating with default action of signal 11 (SIGSEGV)
==9733== Access not within mapped region at address 0x58
==9733== at 0x1349AA: enableLanguage (parse.c:1745)
==9733== by 0x12E2F8: setEtagsMode (options.c:813)
==9733== by 0x12E2F8: testEtagsInvocation (options.c:828)
==9733== by 0x12ACF2: ctags_cli_main (main.c:577)
==9733== by 0x4A99CB1: (below main) (libc-start.c:314)
==9733== If you believe this happened as a result of a stack
==9733== overflow in your program's main thread (unlikely but
==9733== possible), you can try to increase the size of the
==9733== main thread stack using the --main-stacksize= flag.
==9733== The main thread stack size used in this run was 8388608.
==9733==
==9733== HEAP SUMMARY:
==9733== in use at exit: 5,334 bytes in 22 blocks
==9733== total heap usage: 23 allocs, 1 frees, 78,038 bytes allocated
==9733==
==9733== LEAK SUMMARY:
==9733== definitely lost: 6 bytes in 1 blocks
==9733== indirectly lost: 0 bytes in 0 blocks
==9733== possibly lost: 0 bytes in 0 blocks
==9733== still reachable: 5,328 bytes in 21 blocks
==9733== suppressed: 0 bytes in 0 blocks
==9733== Rerun with --leak-check=full to see details of leaked memory
==9733==
==9733== For lists of detected and suppressed errors, rerun with: -s
==9733== ERROR SUMMARY: 1 errors from 1 contexts (suppressed: 0 from 0)
Segmentation fault (core dumped)

My guess is that universal-ctags either does not expect to be run as "etags", or cannot support it. In any case, the normal execution of "ctags" and "ctags-universal" both exit successfully (with a message requesting input files).

ProblemType: Bug
DistroRelease: Ubuntu 20.10
Package: universal-ctags 0+git20200824-1
ProcVersionSignature: Ubuntu 5.8.0-50.56-generic 5.8.18
Uname: Linux 5.8.0-50-generic x86_64
ApportVersion: 2.20.11-0ubuntu50.5
Architecture: amd64
CasperMD5CheckResult: skip
CurrentDesktop: KDE
Date: Fri Apr 16 16:19:18 2021
InstallationDate: Installed on 2020-05-12 (339 days ago)
InstallationMedia: Kubuntu 20.04 LTS "Focal Fossa" - Release amd64 (20200423)
SourcePackage: universal-ctags
UpgradeStatus: Upgraded to groovy on 2021-01-13 (93 days ago)

Revision history for this message
Stephen Brennan (brenns10) wrote :
Revision history for this message
Hans Joachim Desserud (hjd) wrote :

Thanks for taking your time to report this issue and help making Ubuntu better.

I can confirm the issue with universal-ctags 0+git20200824-1 on Ubuntu 21.04. Indeed `etags` crash with a segmentation fault, though running `ctags-universal` directly results in a message requesting input files.

Changed in universal-ctags (Ubuntu):
status: New → Confirmed
tags: added: hirsute impish
To post a comment you must log in.
This report contains Public information  
Everyone can see this information.

Other bug subscribers

Remote bug watches

Bug watches keep track of this bug in other bug trackers.