March 2010 - Xquartz-dev - lists.macosforge.org

Re: [Xquartz-dev] Bugfixes for 2.5.0_rc1
by Brandon Allbery 21 Mar '10

21 Mar '10

On Mar 21, 2010, at 13:49 , Itai Seggev wrote: > On Thu, Mar 18, 2010 at 11:01:26PM -0700, Jeremy Huddleston wrote: >> - This is an issue with the system X11 and the fix must come from >> Apple. > > Are you not part of Apple? Or are you saying this can only be fixed > via an OS > update? The latter; it's okay for MacOSForge XQuartz to be released on an irregular schedule and wothout regard for anything else, but a proper OS update needs regression testing against the entire OS distribution and has to be careful not to make major changes that customers don't expect that could break their workflow. -- brandon s. allbery [solaris,freebsd,perl,pugs,haskell] allbery(a)kf8nh.com system administrator [openafs,heimdal,too many hats] allbery(a)ece.cmu.edu electrical and computer engineering, carnegie mellon university KF8NH

1 0

2.5.0 SL and SL reinstalls
by Jack Howarth 21 Mar '10

21 Mar '10

Jeremy, I understand that the Xquartz 2.5.0 release on Snow Leopard, since it resides in /opt, will survive Software Updates without the need for reinstallation. Is same also true if a user reinstalls 10.6 over an installation with 10.6.2 and Xquartz 2.5.0? Hopefully Xquartz 2.5.0 will still be intact after the Software Updates brings the machine back to 10.6.2. Is this correct? Jack

1 0

Bugfixes for 2.5.0_rc1
by Jeremy Huddleston 20 Mar '10

20 Mar '10

So the main issues I've seen reported from 2.5.0_rc1 are: $PATH should be reordered - Fixed locally Display config changes result in clipped pointer events (can't move window some places) - I know a fix, checking upstream to see if it's the best fix Xephyr and Xnest have bad keymaps - Fixed locally ... you can use these binaries: http://static.macosforge.org/xquartz/downloads/testing/Xnest http://static.macosforge.org/xquartz/downloads/testing/Xephyr System fontconfig segfaults after updating to XQuartz - The only thing this can be is a conflict in the caches. I can't really "fix" this in an XQuartz update since it's the system fontconfig crashing System X11.app starts as "default" X11. - You need to log out for that. I don't want to force logout in the installer. After logging out and back in, system X11.app won't run - This is an issue with the system X11 and the fix must come from Apple. If I missed an issue, please report it (preferably with a bug report at http://xquartz.macosforge.org) As it is now, I think I have mostly eliminated the reported issues. Thanks for testing. --Jeremy

4 4

Re: [Xquartz-dev] Bugfixes for 2.5.0_rc1
by Eeri Kask 19 Mar '10

19 Mar '10

Am 03/19/2010 09:38 AM, Jeremy Huddleston <jeremyhu(a)apple.com> schrieb: > So the main issues I've seen reported from 2.5.0_rc1 are: > > $PATH should be reordered > - Fixed locally > > Display config changes result in clipped pointer events (can't move window some places) > - I know a fix, checking upstream to see if it's the best fix > > Xephyr and Xnest have bad keymaps > - Fixed locally ... you can use these binaries: > http://static.macosforge.org/xquartz/downloads/testing/Xnest > http://static.macosforge.org/xquartz/downloads/testing/Xephyr > > System fontconfig segfaults after updating to XQuartz > - The only thing this can be is a conflict in the caches. I can't really "fix" this > in an XQuartz update since it's the system fontconfig crashing > > System X11.app starts as "default" X11. > - You need to log out for that. I don't want to force logout in the installer. > > After logging out and back in, system X11.app won't run > - This is an issue with the system X11 and the fix must come from Apple. > > If I missed an issue, please report it (preferably with a bug report > at http://xquartz.macosforge.org) As it is now, I think I have > mostly eliminated the reported issues. Thanks for testing. Here come few too: (1) Each time I log in (and launch X11) I am asked (sorry for sloppy English translation) something about ... "Do I want to allow remote applications to access X11, an option which can be tuned in firewall options too" ... or something. So each day in the office begins with a click on "yes, I do". How to automate this buttonclick? (2) xmodmap, which in previous X11-versions configured some keys out of .xinitrc now doesn't do it from there anymore, out of no apparent reason. So each day in the office now begins with running this xmodmap by hand out of some xterm; e.g. in order to change left-alt-option from mode-switch back to Alt_L, and few others. How to revert this change? (3) This is older bug: after logging in X11 starts xterm at top-left, despite the presence of ${HOME}/.xinitrc which is run too and sets up the session correctly. How to disable this xterm-starting? (4) New bug: Starting remote X11 applications by ssh -f remotehost remote-application now apparently succeeds only with the first remote application, further applications often result in something like Warning: No xauth data; using fake authentication data for X11 forwarding. /usr/bin/xauth: error in locking authority file /home/eeri/.Xauthority X11 connection rejected because of wrong authentication. Error: cannot open display: localhost:11.0 I think this bug (or some analogous one) was reported already too. It is not sure if this is a result of upgrading to 2.5.0_rc1 or putting ForwardX11Trusted yes into /etc/ssh_config. Without this ssh-option apparently Xquartz+ssh didn't send umlauted ISO-Latin characters (which are accessed by some "modeswitch"-ed keycombination, like modeswitch+a resulting in ä) to remote X11 clients. (5) The "border-issue" shows up movement but unfortunately not yet in direction of a solution. (6) Some applications leave clutter on screen analogous to the bug I reported http://lists.freedesktop.org/archives/xorg/2008-August/037802.html but now without xcompmgr running. So it appears it is probably Xorg but not a compositing-extension issue... (7) Something like once per week or few weeks the old server 2.4.? could be crashed by the ancient xless, if this has changed now remains to be observed. :-) Greetings, Eeri Kask

2 1

XQuartz 2.5.0_rc1
by Jeremy Huddleston 18 Mar '10

18 Mar '10

We've come a long way since 2.4.0 was released, and the end of the road is in sight for 2.5.0. 2.5.0_rc1 looks quite solid, and I'd like to get more people using it. This is the first pre-release of 2.5.0 that has been built for both Leopard and SnowLeopard. The Leopard version still installs to /usr/X11 and X11.app while the SL version installs to /opt/X11 and XQuartz.app to avoid the Software Update collisions that were so common in the Leopard days. With this release, I've created a separate Software Update feed that will contain beta versions (as well as final release versions). To switch to this feed rather than the final-release feed, please execute the following: SnowLeopard: defaults write org.macosforge.xquartz.X11 SUFeedURL http://static.macosforge.org/xquartz/downloads/sparkle/beta.xml Leopard: defaults write org.x.X11 SUFeedURL http://static.macosforge.org/xquartz/downloads/sparkle/beta.xml Assuming you have 2.4.0 on Leopard or one of the earlier 2.5.0 betas for SL, you should see an auto-update notification the next time you start X11 after running the above command. If you don't want to use the SU, you can just download the installer from these links: SnowLeopard: http://static.macosforge.org/xquartz/downloads/SL/XQuartz-2.5.0_rc1.dmg Leopard: http://static.macosforge.org/xquartz/downloads/Leopard/X11-2.5.0_rc1.dmg The full ChangeLog from 2.4.0 can be seen in the draft release notes: http://xquartz.macosforge.org/trac/wiki/X112.5.0 Please hammer this release. I feel confident that it is release quality, but I want to get a wider audience first. Thanks, Jeremy

7 12

Re: [Xquartz-dev] XQuartz 2.5.0_rc1
by Brandon Allbery 18 Mar '10

18 Mar '10

On Mar 17, 2010, at 17:15 , Bill Campbell wrote: > The clusterssh program is a perl::Tk script which executes > multiple xterms via ssh to the remote systems. It is not a > problem with the remote perl::Tk routing not getting screen size > change info as I always shut that connection down before putting > the Macbook Pro to sleep. One thing I have not tried doing is > opening a new xterm in the running X11, and executing the > commands from it to see if it gets new size information on > startup while an open xterm does not. > > I have to plead ignorance in how X11 handles display geometry in > one screen, much less multiple monitors. XQuartz doesn't even try currently; it would be nice if monitor changes were passed on as XRandR extension events, but right now the X session isn't informed of any changes and you need to restart XQuartz to get it to see the change. I keep wanting to find time to poke at XRandR support, but truth be told I'm a bit too far behind on X11 internals at this point (my experience is from early-mid 90s, and things have changed quite a lot since then). -- brandon s. allbery [solaris,freebsd,perl,pugs,haskell] allbery(a)kf8nh.com system administrator [openafs,heimdal,too many hats] allbery(a)ece.cmu.edu electrical and computer engineering, carnegie mellon university KF8NH

2 3

Re: [Xquartz-dev] Window focus issues on non-quartz wm
by Peter Dyballa 18 Mar '10

18 Mar '10

Am 18.03.2010 um 00:58 schrieb Jeremy Huddleston: >> The new XQuartz 2.5.0_rc1 (xorg-server 1.7.5) shows the same >> behaviour. It also produces this message: >> >> /usr/X11/lib/X11/fonts/75dpi: failed to write cache > > Odd... can you please investigate this further. What are the > directory permissions there? pete 225 /\ la -t /usr/X11/lib/X11/fonts/ total 0 drwxr-xr-x 3 root wheel 1258 18 Mär 00:13 encodings drwxr-xr-x 2 root wheel 782 18 Mär 00:10 util drwxr-xr-x 2 root wheel 14144 18 Mär 00:10 misc drwxr-xr-x 2 root wheel 3026 18 Mär 00:10 cyrillic drwxr-xr-x 2 root wheel 2142 18 Mär 00:10 Type1 drwxr-xr-x 2 root wheel 986 18 Mär 00:10 TTF drwxr-xr-x 2 root wheel 986 18 Mär 00:10 OTF drwxr-xr-x 2 root wheel 476 18 Mär 00:10 Speedo drwxr-xr-x 2 root wheel 64634 18 Mär 00:10 75dpi drwxr-xr-x 2 root wheel 64634 18 Mär 00:10 100dpi > What are the file permissions for the cache files (ls -l /usr/X11/ > lib/X11/fonts/75dpi/fonts.*) pete 226 /\ la -t /usr/X11/lib/X11/fonts/75dpi/ | head -11 total 32216 -rw-r--r-- 1 root wheel 165343 18 Mär 00:10 fonts.dir -rw-r--r-- 1 root wheel 4268 18 Mär 00:10 encodings.dir -rw-r--r-- 1 root wheel 2 18 Mär 00:10 fonts.scale -rw-r--r-- 1 root wheel 98704 18 Mär 00:09 fonts.list -rw-r--r-- 1 root wheel 3066 16 Mär 05:15 fonts.alias -rw-r--r-- 1 root wheel 4105 16 Mär 05:11 charB08.pcf.gz -rw-r--r-- 1 root wheel 4380 16 Mär 05:11 charB10.pcf.gz > That file should be created by root when it runs 'font_cache -s' > during startx. Try running this manually: > > sudo font_cache -f -s -v Looks OK! See: pete 227 /\ sudo /usr/X11/bin/font_cache -f -s -v Password: font_cache: Scanning system font directories to generate X11 font caches font_cache: Making fonts.dir for updated directories. font_cache: /System/Library/Fonts Unknown Type 1 weight "All" Couldn't determine weight for HelveLTMM Unknown Type 1 weight "All" Couldn't determine weight for TimesLTMM font_cache: /Library/Fonts font_cache: /Library/Fonts/Extrafonts font_cache: /Library/Fonts/Extrafonts/Bauhaus Unknown Type 1 weight "Heavy" Couldn't determine weight for Bauhaus-Heavy 001.000.pfb font_cache: /Library/Fonts/Linux Libertine font_cache: /Library/Fonts/PostScript 3 Unknown Type 1 weight "Compact" Couldn't determine weight for AntiqOliCom Unknown Type 1 weight "Poster" Couldn't determine weight for BodonPos Unknown Type 1 weight "Poster" Couldn't determine weight for BodonPosCom Unknown Type 1 weight "ExtraBold" Couldn't determine weight for CoppeThiThrBC Unknown Type 1 weight "Extra Bold" Couldn't determine weight for GillSanExtBol Unknown Type 1 weight "ExtraBold" Couldn't determine weight for GoudyExtBol font_cache: /Library/Fonts/iTunes font_cache: /Library/Fonts/urwfonts-1.41 font_cache: /usr/X11/share/fonts/100dpi font_cache: /usr/X11/share/fonts/75dpi font_cache: /usr/X11/share/fonts/OTF font_cache: /usr/X11/share/fonts/Speedo font_cache: /usr/X11/share/fonts/TTF font_cache: /usr/X11/share/fonts/Type1 font_cache: /usr/X11/share/fonts/cyrillic font_cache: /usr/X11/share/fonts/misc font_cache: /usr/X11/share/fonts/util font_cache: Updating FC cache /usr/X11/share/fonts: caching, new cache contents: 0 fonts, 10 dirs /usr/X11/share/fonts/100dpi: caching, new cache contents: 398 fonts, 0 dirs /usr/X11/share/fonts/75dpi: caching, new cache contents: 398 fonts, 0 dirs /usr/X11/share/fonts/OTF: caching, new cache contents: 23 fonts, 0 dirs /usr/X11/share/fonts/Speedo: caching, new cache contents: 0 fonts, 0 dirs /usr/X11/share/fonts/TTF: caching, new cache contents: 23 fonts, 0 dirs /usr/X11/share/fonts/Type1: caching, new cache contents: 29 fonts, 0 dirs /usr/X11/share/fonts/cyrillic: caching, new cache contents: 0 fonts, 0 dirs /usr/X11/share/fonts/encodings: caching, new cache contents: 0 fonts, 1 dirs /usr/X11/share/fonts/encodings/large: caching, new cache contents: 0 fonts, 0 dirs /usr/X11/share/fonts/misc: caching, new cache contents: 59 fonts, 0 dirs /usr/X11/share/fonts/util: caching, new cache contents: 0 fonts, 0 dirs /usr/X11R6/lib/X11/fonts: caching, new cache contents: 0 fonts, 10 dirs /usr/X11R6/lib/X11/fonts/100dpi: caching, new cache contents: 398 fonts, 0 dirs /usr/X11R6/lib/X11/fonts/75dpi: caching, new cache contents: 398 fonts, 0 dirs /usr/X11R6/lib/X11/fonts/OTF: caching, new cache contents: 23 fonts, 0 dirs /usr/X11R6/lib/X11/fonts/Speedo: caching, new cache contents: 0 fonts, 0 dirs /usr/X11R6/lib/X11/fonts/TTF: caching, new cache contents: 23 fonts, 0 dirs /usr/X11R6/lib/X11/fonts/Type1: caching, new cache contents: 29 fonts, 0 dirs /usr/X11R6/lib/X11/fonts/cyrillic: caching, new cache contents: 0 fonts, 0 dirs /usr/X11R6/lib/X11/fonts/encodings: caching, new cache contents: 0 fonts, 1 dirs /usr/X11R6/lib/X11/fonts/encodings/large: caching, new cache contents: 0 fonts, 0 dirs /usr/X11R6/lib/X11/fonts/misc: caching, new cache contents: 59 fonts, 0 dirs /usr/X11R6/lib/X11/fonts/util: caching, new cache contents: 0 fonts, 0 dirs /System/Library/Fonts: caching, new cache contents: 50 fonts, 0 dirs /Library/Fonts: caching, new cache contents: 162 fonts, 5 dirs /Library/Fonts/Extrafonts: caching, new cache contents: 248 fonts, 1 dirs /Library/Fonts/Extrafonts/Bauhaus: caching, new cache contents: 5 fonts, 0 dirs /Library/Fonts/Linux Libertine: caching, new cache contents: 7 fonts, 0 dirs /Library/Fonts/PostScript 3: caching, new cache contents: 76 fonts, 0 dirs /Library/Fonts/iTunes: caching, new cache contents: 30 fonts, 0 dirs /Library/Fonts/urwfonts-1.41: caching, new cache contents: 38 fonts, 0 dirs /usr/X11/var/cache/fontconfig: cleaning cache directory /usr/X11/var/cache/fontconfig: invalid cache file: 2c094d1c20dbb99343a61912e63eac3a-m68k.cache-2 /usr/X11/bin/fc-cache: succeeded font_cache: Done Time spent in user mode (CPU seconds) : 41.587s Time spent in kernel mode (CPU seconds) : 11.462s Total time : 1:37.54s CPU utilisation (percentage) : 54.3% The /Library/Fonts/PostScript 3 folder has all the fonts which came with my PS3 printer, still including FFIL font suitcases. Maybe my system setup is a bit too unorthodox, leaving away all stuff for the extremely time consuming path_helper application? ~/.MacOSX/ environment.plist delivers much improved performance. The "invalid cache file" has been removed. There exists: -rw-r--r-- 1 root wheel 58688 16. Dez 19:49 2c094d1c20dbb99343a61912e63eac3a-x86-64.cache-2 It shouldn't, on a PPC G4, should it? -- Greetings Pete Got Mole problems? Call Avogadro 6.02 x 10^23

1 0

Re: [Xquartz-dev] Window focus issues on non-quartz wm
by Peter Dyballa 17 Mar '10

17 Mar '10

Am 17.03.2010 um 15:00 schrieb Dave Ray: > If I'm in another OS-X app and I activate X11 by clicking on one of > the X11 windows, X11 comes to the front, and the window I click on > accepts keyboard input, but all the X11 windows remain behind the non- > X11 windows. Perhaps that is normal for X11 windows that I didn't > click on. Definitely odd for the window I clicked on. I have a similiar experience with Blackbox WM on Leopard, PPC, in one Space and with XQuartz 2.5.0 (xorg-server 1.7.99.901) from MacPorts as well. It happens that windows from Console, Terminal, or Emacs.app come into the front. Then I can click (select) an X11 client. It gains focus and input goes to it, but it stays obscured by an Aqua client. I have to click the X11 icon in dock to put all the X clients in the foreground or change to a another space and return. IMO clicking the X11 icon in dock or changing space must not be necessary. The new XQuartz 2.5.0_rc1 (xorg-server 1.7.5) shows the same behaviour. It also produces this message: /usr/X11/lib/X11/fonts/75dpi: failed to write cache -- Greetings Pete Well begun is half done. – Optimist. Half done is well begun. – Realist. Half begun is well done. – Australian.

2 1

Window focus issues on non-quartz wm
by Dave Ray 17 Mar '10

17 Mar '10

I've had a window focus issue for many months and it would help me to know if anyone else is observing it. it's not clear if it's a bug or a non-quartz window manager issue. Perhaps it is normal behavior and I'm misinterpreting. If I toggle between X11.app and other OS-X apps using command-tab, When X11 comes to the front, all X11 windows come forward, in front of any open non-X11 windows from other OS-X apps. If I'm in another OS-X app and I activate X11 by clicking on one of the X11 windows, X11 comes to the front, and the window I click on accepts keyboard input, but all the X11 windows remain behind the non- X11 windows. Perhaps that is normal for X11 windows that I didn't click on. Definitely odd for the window I clicked on. All the symptoms go away when I command-tab out of X11 and back in again. Environment is Enlightenment 16 window manager, with X11's Full-screen mode off (not checked), Leopard 10.5.8, Macbook Pro Everything else checks out great on the new release. Symptom has existed for at least 2 years through several releases. Jeremy had me file a bug a while ago just to document it. If nobody sees this, that will help me to know my wm is just plain broken and it's time to recompile.

1 0

Re: [Xquartz-dev] XQuartz 2.5.0_rc1
by Brandon Allbery 17 Mar '10

17 Mar '10

On Mar 16, 2010, at 21:43 , Doug McNutt wrote: > At 21:17 -0400 3/16/10, Brandon Allbery wrote: >> I have my HDTV hooked up for occasional use as a second monitor, >> positioned to the left of my iMac's internal screen > > I had a terrible time with multiple monitors. Much of the problem > was the conversion from X11's all-positive screen coordinates into > Apple's plus and minus relative to the top left of the startup screen. Thing is, this works fine with older XQuartz versions. The only oddness I've seen (and I suspect it's unavoidable without tying the OSX window/screen information to XRandR events) is that with the HDTV connected windows that are positioned relative to the bottom of the screen use the bottom of the HDTV instead of the iMac even though the windows are placed on the iMac. Positioning above is probably not going to work well, especially in the absence of XRandR linking as described above. Then again, it's worth considering that multiple monitor support (unless you designate them as logically separate screens, the historical method of managing them) has always been a bit odd in X11. -- brandon s. allbery [solaris,freebsd,perl,pugs,haskell] allbery(a)kf8nh.com system administrator [openafs,heimdal,too many hats] allbery(a)ece.cmu.edu electrical and computer engineering, carnegie mellon university KF8NH

1 0