[Users] Claws Mail GTK3 **beta** unleashed

[Brian Morrison]

On Fri, 2020-10-30 at 10:38 +0000, Brian Morrison wrote:
> On Thu, 2020-10-29 at 22:52 +0000, Paul wrote:
> > On Thu, 29 Oct 2020 18:55:21 +0000
> > Brian Morrison <bdm at fenrir.org.uk> wrote: 
> > 
> > > Thanks for this Paul, another step forward :)
> > > 
> > > I have used the tarball you posted, but I also notice that the
> > > fancy
> > > plugin is missing from the gtk3 branch in git.
> > > 
> > > Will you be adding this back in?
> > 
> > That's odd that you say that, because it is not missing.
> 
> OK, you're right. For reasons I don't understand it didn't show up in
> src/plugins but I have just recreated the tarballs with autogen.sh
> and
> it is there. Odd, but then there is plenty of odd to go around.
> 
> I will also see if I can get my themes sorted out, I have had some
> strangeness with themes lately so maybe a complete clear out is
> needed.
> 
I found some entries in the colors.css file that were invalid, and
these caused the warnings seen before, but it doesn't seem to help. I
also installed several greybird themes and changed to one of them.
Still getting the same crash.

Valgrind reports this:

==9614== Memcheck, a memory error detector
==9614== Copyright (C) 2002-2017, and GNU GPL'd, by Julian Seward et
al.
==9614== Using Valgrind-3.16.1 and LibVEX; rerun with -h for copyright
info
==9614== Command: claws-mail
==9614== 
==9614== Warning: unimplemented fcntl command: 1033
==9614== Invalid read of size 8
==9614==    at 0x7496100: XGetXCBConnection (x11_xcb.c:9)
==9614==    by 0x54BA031: sn_display_new (in /usr/lib64/libstartup-
notification-1.so.0.0.0)
==9614==    by 0x2AA7CF: startup_notification_complete.lto_priv.0
(main.c:296)
==9614==    by 0x234228: main (main.c:1573)
==9614==  Address 0x8370490 is 48 bytes inside a block of size 296
free'd
==9614==    at 0x483A9F5: free (vg_replace_malloc.c:538)
==9614==    by 0x60FBB98: g_free (in /usr/lib64/libglib-
2.0.so.0.6600.2)
==9614==    by 0x61157CF: g_slice_free1 (in /usr/lib64/libglib-
2.0.so.0.6600.2)
==9614==    by 0x40EE7F: hmac_sha1 (pkcs5_pbkdf2.c:65)
==9614==    by 0x41008D: pkcs5_pbkdf2 (pkcs5_pbkdf2.c:116)
==9614==    by 0x2F27A6: _make_key_deriv (password.c:99)
==9614==    by 0x2F2EE2: password_decrypt_gnutls (password.c:467)
==9614==    by 0x2F37D7: passwd_store_get (passwordstore.c:223)
==9614==    by 0x2A303E: UnknownInlinedFun (passwordstore.c:299)
==9614==    by 0x2A303E: imap_session_authenticate.lto_priv.0
(imap.c:1313)
==9614==    by 0x29DEA2: imap_session_get.lto_priv.0 (imap.c:1094)
==9614==    by 0x2A3AA8: imap_get_num_list.lto_priv.0 (imap.c:4670)
==9614==    by 0x286C8F: folder_item_scan_full (folder.c:2186)
==9614==  Block was alloc'd at
==9614==    at 0x4839809: malloc (vg_replace_malloc.c:307)
==9614==    by 0x60FEFC8: g_malloc (in /usr/lib64/libglib-
2.0.so.0.6600.2)
==9614==    by 0x611621D: g_slice_alloc (in /usr/lib64/libglib-
2.0.so.0.6600.2)
==9614==    by 0x60D639A: g_checksum_new (in /usr/lib64/libglib-
2.0.so.0.6600.2)
==9614==    by 0x40EE3A: hmac_sha1 (pkcs5_pbkdf2.c:60)
==9614==    by 0x41008D: pkcs5_pbkdf2 (pkcs5_pbkdf2.c:116)
==9614==    by 0x2F27A6: _make_key_deriv (password.c:99)
==9614==    by 0x2F2EE2: password_decrypt_gnutls (password.c:467)
==9614==    by 0x2F37D7: passwd_store_get (passwordstore.c:223)
==9614==    by 0x2A303E: UnknownInlinedFun (passwordstore.c:299)
==9614==    by 0x2A303E: imap_session_authenticate.lto_priv.0
(imap.c:1313)
==9614==    by 0x29DEA2: imap_session_get.lto_priv.0 (imap.c:1094)
==9614==    by 0x2A3AA8: imap_get_num_list.lto_priv.0 (imap.c:4670)
==9614== 
==9614== Invalid read of size 8
==9614==    at 0x7496107: XGetXCBConnection (x11_xcb.c:9)
==9614==    by 0x54BA031: sn_display_new (in /usr/lib64/libstartup-
notification-1.so.0.0.0)
==9614==    by 0x2AA7CF: startup_notification_complete.lto_priv.0
(main.c:296)
==9614==    by 0x234228: main (main.c:1573)
==9614==  Address 0x0 is not stack'd, malloc'd or (recently) free'd

Looks like something bad is passed to libX11-xcb.so but I don't know
enough to debug much further.

I have installed all the F33 updates, so it seems to be something
passed either by Claws or created by something it is doing.

Added another debuginfo rpm for startup-notification and now I see this
with gdb.

Thread 1 "claws-mail" received signal SIGSEGV, Segmentation fault.
0x00007ffff5378107 in XGetXCBConnection (dpy=dpy at entry=0x555555ac7810)
at x11_xcb.c:9
9	x11_xcb.c: No such file or directory.

Not sure how to find out which file or directory it is unable to find.

Ideas?

-- 

Brian