Re: vte [was Re: houston, we have a problem- 2.10 showstoppers]

[Richard Hult <richard imendio com>]

Hi,

I get crashes when running minicom and resizing the window with those 
latest patches. It appears to be the ones committed on the 28th that 
breaks things. I don't have the time right now to investigate this any 
further unfortunately.

It's the same assertion as in

http://bugzilla.gnome.org/show_bug.cgi?id=163814

(but on Debian unstable)

Should we really be doing these kinds of changes a week before the release?

/Richard

Kjartan Maraas wrote:
> I went ahead and commited the fedora patches today, and included a
> couple of patches from bugzilla that other distros have been shipping
> with for some time. Can we leave it at that for now and drop the fork?
>
> I think we have a decent set of improvements over what was there in the
> last release now and we can spend some time weeding out any issues
> introduced by these patches and seeing to it that those are reported in
> bugzilla etc.
>
> There is one bugreport with patches that I think should be looked at
> before 2.10:
>
> http://bugzilla.gnome.org/show_bug.cgi?id=137864
>
> It would be very nice if we could ship with the patches in here included
> as well. I got the impression Mandrake has already been using these.
>
> I didn't want to just go ahead without a clear indication of approval,
> but when I got no answer I felt something had to be done. I think that
> this was something "the community" wanted to get some traction on too
> and apologize if I've stepped over the line. Reverting back to where we
> were is only a few clicks on URLs in bonsai away. :-)
>
> Excerpt from ChangeLog with recent changes:
>
> * src/dumpkeys.c: Fix build on NetBSD and Darwin.
> Patch from Adrian Bunk <bunk at stusta de>. Closes
> bug #126377 [http://bugzilla.gnome.org/show_bug.cgi?id=126377]
>
> * src/vteregex.h: Don't include <config.h> here. It breaks
> building on AIX and can cause problems for l10n. Closes
> bug #161352 [http://bugzilla.gnome.org/show_bug.cgi?id=161352]
>
> * src/keymap.c: (_vte_keymap_map): Fix keypad behaviour
> to be like xterm. Closes bug #128099
> [http://bugzilla.gnome.org/show_bug.cgi?id=128099]. Fix is from
> <jylefort at brutele be>.
>
> * src/vte.c: (vte_terminal_visibility_notify): Fix black
> backgrounds in new terminals. Fix from Jon Nettleton
> <nettleton at gmail com>. Closes bug #125364
> [http://bugzilla.gnome.org/show_bug.cgi?id=125364]
>
> * src/vte.c: (vte_terminal_adjust_adjustments),
> (vte_terminal_maybe_scroll_to_bottom),
> (vte_terminal_update_insert_delta), (vte_terminal_handle_scroll),
> (vte_terminal_set_scroll_adjustment), (vte_terminal_size_allocate):
> Apply patch from Fedora to fix some scrolling issues. Closes
> bug #168210 [http://bugzilla.gnome.org/show_bug.cgi?id=168210] and
> possibly others
>
> * src/iso2022.c: (_vte_iso2022_process): Apply another
> patch from Fedora. This should fix some screen corruption
> with multibyte charsets. Closes bug #168211
> [http://bugzilla.gnome.org/show_bug.cgi?id=168211]
>
> * src/vte.c: (vte_terminal_scroll_region),
> (vte_sequence_handler_al), (vte_sequence_handler_dl),
> (vte_sequence_handler_sr), (vte_sequence_handler_insert_lines),
> (vte_sequence_handler_delete_lines), (vte_terminal_handle_scroll):
> Another patch from fedora. Make sure the screen is fully redrawn
> before scrolling. Apparently this patch has some issues, but those
> were found to be small enough to apply in Fedora, so I guess
> the same holds true here.
>
> * src/vte.c: (vte_terminal_unrealize): Fix a crash
> when trying to use input methods. From Fedora.
> Closes bug #168213 [http://bugzilla.gnome.org/show_bug.cgi?id=168213]
>
> * src/vte.c: (vte_terminal_scroll_region): Fix for
> problem with redrawing when scrolling back then forward
> Closes bug #122150 [http://bugzilla.gnome.org/show_bug.cgi?id=122150].
> Patch from Benjamin Otte.
> * src/iso2022.c: (_vte_iso2022_is_ambiguous),
> (_vte_iso2022_find_nextctl):
> * src/uniwidths:
> * src/vte.c: (vte_terminal_catch_child_exited), (vte_terminal_eof),
> (vte_terminal_process_incoming), (vte_terminal_feed),
> (vte_terminal_init), (vte_terminal_finalize), (vte_terminal_reset),
> (add_display_timeout), (add_coalesce_timeout),
> (remove_display_timeout), (remove_coalesce_timeout),
> (vte_terminal_stop_processing), (vte_terminal_start_processing),
> (vte_terminal_is_processing), (display_timeout),
> (coalesce_timeout):
> * src/vtexft.c: (_vte_xft_drawcharfontspec), (_vte_xft_draw_text):
> Apply fix from Fedora package for performance problems. Closes
> bug #143914 [http://bugzilla.gnome.org/show_bug.cgi?id=143914]. Patch
> was written by S�Sandmann.
>
> * src/vte.c: (vte_sequence_handler_send_primary_device_attributes):
> Patch from Mariano to make vte claim to be the right type terminal.
> Closes bug #130671 [http://bugzilla.gnome.org/show_bug.cgi?id=130671].
>
> I'll make a test tarball with these and the ones from
> http://bugzilla.gnome.org/show_bug.cgi?id=137864 so we can get people
> testing and possibly close out a few of the old bugzilla reports that
> are relevant to these fixes.
>
> Nalin, would it be ok with you if this was released as 0.11.12 and could
> you find the time to review the patches in the report mentioned in the
> paragraph above?
>
> Cheers
> Kjartan
>
> _______________________________________________
> gnome-hackers mailing list
> gnome-hackers gnome org
> http://mail.gnome.org/mailman/listinfo/gnome-hackers


--
Imendio AB, http://www.imendio.com/