Bug 355482 – gnomescan segfaults on startup

After an evaluation, GNOME has moved from Bugzilla to GitLab. Learn more about GitLab.
No new issues can be reported in GNOME Bugzilla anymore.
To report an issue in a GNOME project, go to GNOME GitLab.
Do not go to GNOME Gitlab for: Bluefish, Doxygen, GnuCash, GStreamer, java-gnome, LDTP, NetworkManager, Tomboy.

Bug 355482 - gnomescan segfaults on startup


Summary:	gnomescan segfaults on startup


Status:	RESOLVED FIXED

Product:	gnome-scan
Classification:	Other
Component:	flegita
Version:	0.2.x
Hardware:	Other Linux

Importance:	Normal blocker
Target Milestone:	---
Assigned To:	Étienne Bersac
QA Contact:	Étienne Bersac

URL:
Whiteboard:

Depends on:
Blocks:

Reported:	2006-09-11 19:46 UTC by Joseph Sacco
Modified:	2006-10-17 16:13 UTC

See Also:
GNOME target:	---
GNOME version:	2.15/2.16

Attachments
borked scan image (55.67 KB, application/octet-stream) 2006-10-17 14:09 UTC, Joseph Sacco	Details

Description Joseph Sacco 2006-09-11 19:46:55 UTC

(gdb) b  gnomescanbackend.c:658
Breakpoint 1 at 0xffa6318: file gnomescanbackend.c, line 658.
(gdb) run
The program being debugged has been started already.
Start it from the beginning? (y or n) y

Starting program: /usr/local/src/GARNOME/2.15/garnome-2.15.x/bin/flegita
[Thread debugging using libthread_db enabled]
[New Thread 810913440 (LWP 15280)]
(gnomescan:15280): GnomeScan-DEBUG: sane_init(0x7fd682ac, NULL)
(gnomescan:15280): GnomeScan-DEBUG: SANE Version = 1.0.13
[New Thread 821400800 (LWP 15283)]
[New Thread 829789408 (LWP 15284)]
[Thread 821400800 (LWP 15283) exited]
[Thread 829789408 (LWP 15284) exited]
[New Thread 829789408 (LWP 15285)]
[New Thread 821400800 (LWP 15286)]
[New Thread 882210016 (LWP 15287)]
[New Thread 890598624 (LWP 15288)]
[New Thread 898987232 (LWP 15289)]
[Thread 829789408 (LWP 15285) exited]
[Thread 821400800 (LWP 15286) exited]
[Thread 882210016 (LWP 15287) exited]
[Thread 898987232 (LWP 15289) exited]
[Thread 890598624 (LWP 15288) exited]
[New Thread 890598624 (LWP 15290)]
[New Thread 821400800 (LWP 15291)]
(gnomescan:15280): GnomeScan-DEBUG: sane_get_devices (0x7fd68274, FALSE)
[Thread 890598624 (LWP 15290) exited]
[Thread 821400800 (LWP 15291) exited]
(gnomescan:15280): GnomeScan-DEBUG: Creating flatbed Scanner Epson GT-9700 (epson:/dev/sg3)
(gnomescan:15280): GnomeScan-DEBUG:  type       = gboolean
(gnomescan:15280): GnomeScan-DEBUG:  value      = TRUE

(gnomescan:15280): GLib-GObject-CRITICAL **: g_value_get_int: assertion `G_VALUE_HOLDS_INT (value)' failed
(gnomescan:15280): GnomeScan-DEBUG: Emulating option 'source'

(gnomescan:15280): GnomeScan-CRITICAL **: gnome_scanner_option_get_constraint: assertion `GNOME_IS_SCANNER_OPTION (option)' failed

+ Trace 72146

Thread 810913440 (LWP 15280)

#0 gsb_scanner_get_geometry
at gnomescanbackend.c line 658
#1 gsb_scanner_new
at gnomescanbackend.c line 515
#2 gnome_scan_backend_probe_scanners
at gnomescanbackend.c line 308
#3 gnome_scan_context_probe_scanners
at gnomescancontext.c line 531
#4 gnome_scan_dialog_run
at gnomescandialog.c line 399
#5 main
at main.c line 112
#1 gsb_scanner_new
at gnomescanbackend.c line 515
#2 gnome_scan_backend_probe_scanners
at gnomescanbackend.c line 308

303       }
304       else {
305         for (i = 0; devices[i] && !(*error); i++) {
306           sid = g_new0 (GnomeScannerId, 1);
307           sid->descriptor = devices[i];
308           scanners = g_slist_append (scanners, gsb_scanner_new (backend, sid, error));
309         }
310       }
311
312       return scanners;
(gdb) p scanners
$4 = (GSList *) 0x0
(gdb) p backend
$5 = (GnomeScanBackend *) 0x10020430
(gdb) p *backend
$6 = {parent = {g_type_instance = {g_class = 0x10055860}, ref_count = 2, qdata = 0x0}}

Comment 1 Étienne Bersac 2006-10-04 17:01:42 UTC

*** Bug 359370 has been marked as a duplicate of this bug. ***

Comment 2 Étienne Bersac 2006-10-11 09:48:08 UTC

Hi

(gnomescan:15280): GnomeScan-DEBUG: Creating flatbed Scanner Epson GT-9700
(epson:/dev/sg3)
(gnomescan:15280): GnomeScan-DEBUG:  type       = gboolean
(gnomescan:15280): GnomeScan-DEBUG:  value      = TRUE

(gnomescan:15280): GLib-GObject-CRITICAL **: g_value_get_int: assertion
`G_VALUE_HOLDS_INT (value)' failed

That is very strange. the first option must be an int (according to sane standard) and epson backend provide a boolean O_O. I commited a workaround, can you test latest SVN please ?

Comment 3 Étienne Bersac 2006-10-17 09:56:38 UTC

Does it still happen with 0.3 ?

Comment 4 Joseph Sacco 2006-10-17 14:09:29 UTC

Created attachment 74870 [details]
borked scan image

Comment 5 Joseph Sacco 2006-10-17 14:10:56 UTC

Étienne,

The good news is the app no longer segfaults when started.  The bad news is the scanned image is borked. See attached PDF.

-Joseph

Comment 6 Étienne Bersac 2006-10-17 15:55:10 UTC

Nice,

That's a known bug, gnomescan currently treat all acquisition as RGB. Will fix that in the following days.

Thanks.

Comment 7 Étienne Bersac 2006-10-17 16:13:41 UTC

Please fill another bug for that problem.