Your message dated Sun, 19 Sep 2010 15:44:04 +0200 with message-id <4C9613A4.4030507@dogguy.org> and subject line Re: Bug#597357: unblock: cairo/1.8.10-6 has caused the Debian Bug report #597357, regarding unblock: cairo/1.8.10-6 to be marked as done. This means that you claim that the problem has been dealt with. If this is not the case it is now your responsibility to reopen the Bug report if necessary, and/or fix the problem forthwith. (NB: If you are a system administrator and have no idea what this message is talking about, this may indicate a serious mail system misconfiguration somewhere. Please contact owner@bugs.debian.org immediately.) -- 597357: http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=597357 Debian Bug Tracking System Contact owner@bugs.debian.org with problems
--- Begin Message ---
- To: submit@bugs.debian.org
- Subject: unblock: cairo/1.8.10-6
- From: Josselin Mouette <joss@debian.org>
- Date: Sun, 19 Sep 2010 00:05:10 +0200
- Message-id: <[🔎] 1284847510.26675.8.camel@tomoyo>
Package: release.debian.org Severity: normal User: release.debian.org@packages.debian.org Usertags: freeze-exception Please unblock the latest cairo upload. cairo (1.8.10-6) unstable; urgency=low . * debian/patches/02_iceweasel-buggy-repeat.patch: + Revert logic of the patch for the infamous iceweasel rendering problems. The only broken X11 driver that is left seems to be fglrx and always enabling the workaround code in cairo causes severe performance degradations for everybody (Closes: #594245). Thanks for Josselin Mouette for the patch. Everybody using the fglrx driver should set the MOZ_CAIRO_BUGGY_REPEAT environment variable, everybody else should get a correctly working and fast cairo again. I’m attaching the new version of the patch. Cheers, -- .''`. Josselin Mouette : :' : `. `' “If you behave this way because you are blackmailed by someone, `- […] I will see what I can do for you.” -- Jörg SchillingIndex: cairo/src/cairo-xlib-display.c =================================================================== --- cairo.orig/src/cairo-xlib-display.c 2010-08-17 05:52:51.868996101 +0200 +++ cairo/src/cairo-xlib-display.c 2010-08-17 05:52:54.692995371 +0200 @@ -66,6 +66,8 @@ static cairo_xlib_display_t *_cairo_xlib_display_list; +static int buggy_repeat_force = -1; + static void _cairo_xlib_remove_close_display_hook_internal (cairo_xlib_display_t *display, cairo_xlib_hook_t *hook); @@ -338,6 +340,17 @@ display->buggy_repeat = TRUE; } + /* XXX workaround; see https://bugzilla.mozilla.org/show_bug.cgi?id=413583 */ + if (buggy_repeat_force == -1) { + if (getenv("MOZ_CAIRO_BUGGY_REPEAT")) + buggy_repeat_force = 1; + else + buggy_repeat_force = 0; + } + + if (buggy_repeat_force) + display->buggy_repeat = TRUE; + display->next = _cairo_xlib_display_list; _cairo_xlib_display_list = display;Attachment: signature.asc
Description: This is a digitally signed message part
--- End Message ---
--- Begin Message ---
- To: Josselin Mouette <joss@debian.org>, 597357-done@bugs.debian.org
- Cc: fglrx-driver@packages.qa.debian.org
- Subject: Re: Bug#597357: unblock: cairo/1.8.10-6
- From: Mehdi Dogguy <mehdi@dogguy.org>
- Date: Sun, 19 Sep 2010 15:44:04 +0200
- Message-id: <4C9613A4.4030507@dogguy.org>
- In-reply-to: <[🔎] 1284898282.26675.18.camel@tomoyo>
- References: <[🔎] 1284847510.26675.8.camel@tomoyo> <[🔎] 4C95D402.7080009@dogguy.org> <[🔎] 1284898282.26675.18.camel@tomoyo>
On 09/19/2010 02:11 PM, Josselin Mouette wrote: > Le dimanche 19 septembre 2010 à 11:12 +0200, Mehdi Dogguy a écrit : >> On 09/19/2010 12:05 AM, Josselin Mouette wrote: >>> Everybody using the fglrx driver should set the >>> MOZ_CAIRO_BUGGY_REPEAT environment variable, everybody else >>> should get a correctly working and fast cairo again. >> >> Isn't it worth mentioning such a change in debian/NEWS? So that >> users are actually aware of the behaviour change. > > I think it should be documented in fglrx instead. The behavior of > 1.8.10-5 is buggy anyway, so we just came back and gave fglrx users > an opportunity to have a working gecko/webkit. > > However nobody is going to look for the solution in the cairo > documentation. Ideally, I’d even say this variable should be set by > fglrx in a Xsession startup script if the driver is being used, > instead of being documented (remember, no one reads documentation). > Fair enough. I CC'ed fglrx maintainers so that they are aware of this change and can document it somehow. It think it's worth mentioning in the Release Notes, no? Unblocked. Regards, -- Mehdi Dogguy مهدي الدڤي http://dogguy.org/
--- End Message ---