Bug 2189228

Summary: Emacs 28.2 on Fedora 38 often starts with segmentation fault
Product: [Fedora] Fedora Reporter: dirk
Component: cairoAssignee: Benjamin Otte <otte>
Status: CLOSED ERRATA QA Contact: Fedora Extras Quality Assurance <extras-qa>
Severity: medium Docs Contact:
Priority: unspecified    
Version: 38CC: ajax, amigadave, cagney, dan.cermak, dswbike, dueno, gnome-sig, gordon.messmer, jkeating, levinedl, mclasen, msekleta, otte, phracek, pizza, robinlee.sysu, rstrode, swt, woodard
Target Milestone: ---Keywords: Desktop, Upgrades
Target Release: ---   
Hardware: x86_64   
OS: Linux   
URL: https://retrace.fedoraproject.org/faf/reports/647278/
Whiteboard:
Fixed In Version: cairo-1.17.8-4.fc39 cairo-1.17.8-4.fc38 Doc Type: If docs needed, set a value
Doc Text:
Story Points: ---
Clone Of: Environment:
Last Closed: 2023-05-18 01:57:33 UTC Type: ---
Regression: --- Mount Type: ---
Documentation: --- CRM:
Verified Versions: Category: ---
oVirt Team: --- RHEL 7.3 requirements from Atomic Host:
Cloudforms Team: --- Target Upstream Version:
Embargoed:
Attachments:
Description Flags
File: backtrace none

Description dirk 2023-04-24 13:33:43 UTC
Emacs 28.2 on Fedora 38 often starts with segmentation fault. "Often" means maybe each 2nd or 3rd time, or so. This problem is new to me since upgrading from Fedora 37 to 38. Using Wayland and GNOME 44. 

Reproducible: Sometimes

Steps to Reproduce:
1. Starting Emacs 28.2 either by clicking desktop icon or in the terminal
2. 
3.
Actual Results:  
Each 2nd or 3rd time or so it crashes immediately with segmentation fault. See output in terminal under "Additional Information". See also ABRT report 
https://retrace.fedoraproject.org/faf/reports/647278/

Expected Results:  
Running Emacs.

Problem new to me since system upgrade from F37 to F38 some days ago.

Output (backtrace) in terminal:

$ emacs
Fatal error 11: Segmentation fault
Backtrace:
emacs(emacs_backtrace+0x5c)[0x5889dc]
emacs(terminate_due_to_signal+0xa1)[0x46533c]
emacs[0x465cc6]
emacs[0x6ae2b4]
/lib64/libc.so.6(+0x3dba0)[0x7fe47a05eba0]
/lib64/libcairo.so.2(+0x59c8c)[0x7fe48288fc8c]
/lib64/libcairo.so.2(+0xa8144)[0x7fe4828de144]
/lib64/libcairo.so.2(+0x64127)[0x7fe48289a127]
/lib64/libcairo.so.2(+0x70099)[0x7fe4828a6099]
/lib64/libcairo.so.2(+0x720e9)[0x7fe4828a80e9]
/lib64/libcairo.so.2(+0x2be9a)[0x7fe482861e9a]
/lib64/libcairo.so.2(cairo_show_glyphs+0x2e)[0x7fe4828b645e]
emacs[0x68c1cd]
emacs[0x539ac2]
emacs[0x53c0b4]
emacs[0x4c0f14]
emacs(gui_write_glyphs+0x9c)[0x4ccc9c]
emacs[0x4757a6]
emacs[0x47caac]
emacs[0x47d824]
emacs(update_frame+0x1ed)[0x47e4cd]
emacs[0x4ab34d]
emacs[0x4aca88]
emacs(message3_nolog+0xf6)[0x4a2a86]
emacs(message3+0x308)[0x4a2f18]
emacs(Fmessage+0x46)[0x5fa976]
/usr/bin/../lib64/emacs/28.2/native-lisp/28.2-a66ed3a6/preloaded/mule-3352613d-82b9a969.eln(F6c6f61642d776974682d636f64652d636f6e76657273696f6e_load_with_code_conversion_0+0x31b)[0x7fe4731db51b]
emacs(Ffuncall+0x221)[0x603971]
emacs(Fload+0x3f9)[0x6305c9]
emacs(Ffuncall+0x221)[0x603971]
emacs[0x609b15]
emacs(Fmapc+0x55)[0x609db5]
emacs(eval_sub+0x5c8)[0x602d78]
emacs[0x63001b]
emacs(Feval_buffer+0x17a)[0x63143a]
/usr/bin/../lib64/emacs/28.2/native-lisp/28.2-a66ed3a6/preloaded/mule-3352613d-82b9a969.eln(F6c6f61642d776974682d636f64652d636f6e76657273696f6e_load_with_code_conversion_0+0x27b)[0x7fe4731db47b]
emacs(Ffuncall+0x221)[0x603971]
emacs(Fload+0x3f9)[0x6305c9]
emacs(Ffuncall+0x221)[0x603971]
/usr/bin/../lib64/emacs/28.2/native-lisp/28.2-a66ed3a6/preloaded/startup-bbc6ea72-9c04c94d.eln(F636f6d6d616e642d6c696e65_command_line_0+0x1436)[0x7fe47316fca6]
emacs(Ffuncall+0x221)[0x603971]
...
Speicherzugriffsfehler (Speicherabzug geschrieben)

(Translation of the last line: segmentation fault (core dumped) )

Comment 1 Scott Talbert 2023-04-26 19:03:21 UTC
I can reproduce this bug as well.  However, I think it might actually be a bug in cairo.  I rebuilt cairo-1.17.6-2 for F38, installed it, and I can no longer reproduce the problem.

Comment 2 Scott Talbert 2023-05-16 00:16:06 UTC
Reassigning to cairo.

PR submitted with upstream fix: https://src.fedoraproject.org/rpms/cairo/pull-request/3

Comment 3 Scott Talbert 2023-05-16 00:18:58 UTC
*** Bug 2188678 has been marked as a duplicate of this bug. ***

Comment 4 Scott Talbert 2023-05-16 00:21:40 UTC
*** Bug 2182529 has been marked as a duplicate of this bug. ***

Comment 5 Fedora Update System 2023-05-16 09:01:43 UTC
FEDORA-2023-c3a12fe3c1 has been submitted as an update to Fedora 38. https://bodhi.fedoraproject.org/updates/FEDORA-2023-c3a12fe3c1

Comment 6 Fedora Update System 2023-05-17 01:22:18 UTC
FEDORA-2023-c3a12fe3c1 has been pushed to the Fedora 38 testing repository.
Soon you'll be able to install the update with the following command:
`sudo dnf upgrade --enablerepo=updates-testing --refresh --advisory=FEDORA-2023-c3a12fe3c1`
You can provide feedback for this update here: https://bodhi.fedoraproject.org/updates/FEDORA-2023-c3a12fe3c1

See also https://fedoraproject.org/wiki/QA:Updates_Testing for more information on how to test updates.

Comment 7 Hin-Tak Leung 2023-05-17 02:17:56 UTC
emacs launched by git to edit commit message. Immediately dies. Re-launch right away was okay.


reporter:       libreport-2.17.10
executable:     /usr/bin/emacs-28.2
rootdir:        /
kernel:         6.2.14-300.fc38.x86_64
journald_cursor: s=15e22ab40381453db1322b6a626f8d6b;i=33971;b=f0ed22e081024d6f991ed6a668791908;m=ed6d832e05;t=5fbda02fc7190;x=a8a820ae0f2fc4dc
cgroup:         0::/user.slice/user-1000.slice/user/app.slice/app-org.gnome.Terminal.slice/vte-spawn-841e5717-eda3-49ba-8b25-5b04211d6f49.scope
runlevel:       N 5
uid:            1000
backtrace_rating: 4
cmdline:        /usr/bin/emacs /home/Hin-Tak/git-others/ghostpdl-trunk/.git/COMMIT_EDITMSG
type:           CCpp
reason:         emacs-28.2 killed by SIGSEGV
package:        emacs-1:28.2-4.fc38
comment:        emacs launched by git to edit commit message. Immediately dies. Re-launch right away was okay.
crash_function: handle_fatal_signal

Comment 8 Hin-Tak Leung 2023-05-17 02:17:58 UTC
Created attachment 1964986 [details]
File: backtrace

Comment 9 Sebastian Frenger 2023-05-17 12:57:10 UTC
i installed asci-doc-support in spacemacs by open an file, a few seconds later while installation was running (or finished) it crashed.


reporter:       libreport-2.17.10
type:           CCpp
reason:         emacs-28.2 killed by SIGSEGV
journald_cursor: s=6a399b8f2c8c465fae430cd77790dd91;i=4bcbd;b=711a6ddc316c47bc8c0b66e6045dbdfd;m=f7e183da;t=5fbde2317eb9d;x=4c215a423753e590
executable:     /usr/bin/emacs-28.2
cmdline:        /usr/bin/emacs
cgroup:         0::/user.slice/user-1000.slice/user/app.slice/app-gnome-emacs-2279.scope
rootdir:        /
uid:            1000
kernel:         6.2.15-300.fc38.x86_64
package:        emacs-1:28.2-4.fc38
runlevel:       N 5
backtrace_rating: 4
crash_function: handle_fatal_signal
comment:        i installed asci-doc-support in spacemacs by open an file, a few seconds later while installation was running (or finished) it crashed.

Comment 10 Fedora Update System 2023-05-18 01:57:33 UTC
FEDORA-2023-c3a12fe3c1 has been pushed to the Fedora 38 stable repository.
If problem still persists, please make note of it in this bug report.

Comment 11 David Levine 2023-05-18 12:15:44 UTC
cairo-1.17.8-4 seems to have fixed the problem with emacs startup for me.

Comment 12 dirk 2023-05-18 15:32:52 UTC
cairo-1.17.8-4.fc38 also fixed it for me.

Comment 13 Ben Woodard 2023-05-26 20:52:36 UTC
git commit -s
It popped up and disappeared. Then the next time that I tried it, no problem.


reporter:       libreport-2.17.10
type:           CCpp
reason:         emacs-28.2 killed by SIGSEGV
journald_cursor: s=c0f919bcddcd4756b3f8034fd5176be0;i=2a9df;b=81274ac23b344893a85b93a434ca8769;m=b8f5e389cb;t=5fc9ec515f7a5;x=cb0ec989bdef7e09
executable:     /usr/bin/emacs-28.2
cmdline:        /usr/bin/emacs /home/ben/Shared/Work/libabigail/.git/COMMIT_EDITMSG
cgroup:         0::/user.slice/user-1000.slice/user/app.slice/app-org.gnome.Terminal.slice/vte-spawn-8ec28427-34c3-48a3-8754-56a47385d450.scope
rootdir:        /
uid:            1000
kernel:         6.2.15-300.fc38.x86_64
package:        emacs-1:28.2-4.fc38
runlevel:       N 5
backtrace_rating: 4
crash_function: handle_fatal_signal

Comment 14 Scott Talbert 2023-05-26 20:57:13 UTC
(In reply to Ben Woodard from comment #13)
> git commit -s
> It popped up and disappeared. Then the next time that I tried it, no problem.
> 
> 
> reporter:       libreport-2.17.10
> type:           CCpp
> reason:         emacs-28.2 killed by SIGSEGV
> journald_cursor:
> s=c0f919bcddcd4756b3f8034fd5176be0;i=2a9df;
> b=81274ac23b344893a85b93a434ca8769;m=b8f5e389cb;t=5fc9ec515f7a5;
> x=cb0ec989bdef7e09
> executable:     /usr/bin/emacs-28.2
> cmdline:        /usr/bin/emacs
> /home/ben/Shared/Work/libabigail/.git/COMMIT_EDITMSG
> cgroup:        
> 0::/user.slice/user-1000.slice/user/app.slice/app-org.gnome.
> Terminal.slice/vte-spawn-8ec28427-34c3-48a3-8754-56a47385d450.scope
> rootdir:        /
> uid:            1000
> kernel:         6.2.15-300.fc38.x86_64
> package:        emacs-1:28.2-4.fc38
> runlevel:       N 5
> backtrace_rating: 4
> crash_function: handle_fatal_signal

@woodard are you running with cairo-1.17.8-4.fc38?  If so, it's probably a different bug, please open a new one.

Comment 15 Red Hat Bugzilla 2023-09-24 04:25:08 UTC
The needinfo request[s] on this closed bug have been removed as they have been unresolved for 120 days