449546 - [1.9.0 branch] Crash [@ XSend] dismissing print dialog

Status: RESOLVED WORKSFORME

(Core :: Printing: Setup, defect)

Description

[Bob Clary [:bc] (inactive)]

1. start browser
2. javascript:print()
3. cancel print dialog
4. close browser

WARNING: Caller should supply a printer name.: file /work/mozilla/builds/1.9.1/mozilla/widget/src/xpwidgets/nsPrintOptionsImpl.cpp, line 1068
###!!! ASSERTION: null data pointer: 'Not Reached', file /work/mozilla/builds/1.9.1/mozilla/xpcom/string/src/nsTSubstring.cpp, line 667

...

Program received signal SIGSEGV, Segmentation fault.
[Switching to Thread 47780017704512 (LWP 5050)]
0x00000000ffffffff in ?? ()
(gdb) bt
#0  0x00000000ffffffff in ?? ()
#1  0x000000375044b6cf in _XSend () from /usr/lib64/libX11.so.6
#2  0x000000375044b7c1 in _XReply () from /usr/lib64/libX11.so.6
#3  0x00000037504434fa in XSync () from /usr/lib64/libX11.so.6
#4  0x0000003752806859 in XRenderFindDisplay () from /usr/lib64/libXrender.so.1
#5  0x0000003752803143 in XRenderFreeGlyphs () from /usr/lib64/libXrender.so.1
#6  0x0000003753834bc8 in cairo_xlib_surface_get_display () from /usr/lib64/libcairo.so.2
#7  0x0000003753817e82 in cairo_scaled_font_get_font_matrix () from /usr/lib64/libcairo.so.2
#8  0x000000375380e321 in cairo_create () from /usr/lib64/libcairo.so.2
#9  0x0000003753818463 in cairo_scaled_font_get_font_matrix () from /usr/lib64/libcairo.so.2
#10 0x000000375381879f in cairo_scaled_font_destroy () from /usr/lib64/libcairo.so.2
#11 0x000000375380f279 in cairo_font_face_status () from /usr/lib64/libcairo.so.2
#12 0x000000375380efce in cairo_debug_reset_static_data () from /usr/lib64/libcairo.so.2
#13 0x00002b74a5c4a39e in MOZ_gdk_display_close (display=0x6a80000) at /work/mozilla/builds/1.9.1/mozilla/toolkit/xre/nsAppRunner.cpp:2402
#14 0x00002b74a5c5169a in XRE_main (argc=3, argv=0x7fff04e9c9a8, aAppData=0x6a60d50) at /work/mozilla/builds/1.9.1/mozilla/toolkit/xre/nsAppRunner.cpp:3296
#15 0x0000000000401213 in main (argc=3, argv=0x7fff04e9c9a8) at /work/mozilla/builds/1.9.1/mozilla/browser/app/nsBrowserApp.cpp:158

Comment 1

[Karl Tomlinson (:karlt)]

This works for me.
Some cairo changes in the display-close cleanup have landed recently.
Can you still reproduce with a build from Sunday or later?

If so, which gtk theme engine is being used?

  lsof -p <firefox-bin-process> | grep gtk

Comment 2

[Karl Tomlinson (:karlt)]

See bug 469831 comment 40 and following.
There is a patch there that should help with this, but perhaps the cairo_version check may be wrong.  Any idea what the system cairo version was?  ("pkg-config --modversion cairo", or "ls -l /usr/lib/libcairo.so" may give some hints.)

Comment 3

[Jesse Ruderman]

bc, is this still crashing for you?  It isn't crashing for me.

Comment 4

[Bob Clary [:bc] (inactive)]

does not crash 1.9.3-tracemonkey, 1.9.3, 1.9.2, 1.9.1 but does crash 1.9.0.

Comment 5

[Martin Stránský [:stransky] (ni? me)]

It could be related to Bug 546270. When the print dialog is raised from JS, a default printer name is NULL (at least on 1.9.0).

Comment 6

[Wayne Mery (:wsmwk)]

(In reply to comment #5)
> It could be related to Bug 546270. When the print dialog is raised from JS, a
> default printer name is NULL (at least on 1.9.0).

shouldn't be related given comment 4. (?)

Comment 7

[Bob Clary [:bc] (inactive)]

update crash bugs to critical per guidelines.

Comment 8

[Sylvestre Ledru [:Sylvestre]]

Closing because no crash reported since 12 weeks.

Comment 9

[David Baron :dbaron: (⌚️UTC-5, no longer working on Mozilla)]

Reopening because crash bugs **with testcases** should not be resolved **as WONTFIX** based on queries of crash-stats.  Other resolutions may be appropriate for other reasons.

(Crash signatures are not the same as bug identity; they're merely a search aid to find and group similar crashes.  The bug may still be present, but the signature may have changed slightly, or the bug may even still be present with the same signature but there are simply no recent reports of crashes in that function.)

Comment 10

[Andrei Purice]

Hey Bob,
Does this crash still occur for you?

Comment 11

[Bob Clary [:bc] (inactive)]

Looks like javascript: in the url bar was broken in bug Bug 1466801 and the consensus (though not final) in bug 1506100 is that it shouldn't be fixed.

I'm ok with this being called fixed or works for me now but am loath to do so until someone resolves bug 1506100 one way or another. If javascript: in the url bar is ever reintroduced it might be worth having this bug around. Please don't mark it invalid. It was valid when I filed it 13 years ago and it is depressing to be told your efforts were invalid.

Comment 12

[Daniel Holbert [:dholbert]]

Given comment 2,3,4 indicate that this was specific to a particular branch (1.9.0) and didn't affect newer branches, I think we can call this WORKSFORME.

(There have been substantial updates to printing and cairo/graphics over the past 14 years, so I'd be surprised if this persisted through all of that.)

I tried locally using mozregression-launched Nightlies from back when javascript: URLs still worked (and produced a print dialog when visiting javascript:print(), and I couldn't trigger a crash. And it sounds like the newest build where this was ever reproduced was on gecko 1.9.0, ages ago.

(Specifically, I tested builds that are about as old as will run on my system: -- 2010-04-10 and 2011-01-01 -- plus a build from just before we nerfed javascript: URLs, 2018-07-12.)