| |
| |
| |
|
Comments:
<0> alanc: the logging performed to generate the tag push message also appears to truncate things, I think.
<1> anholt: yeah, looked like it... otoh if it hadn't, your last release's log would have been huge :)
<2> btw does the new release supports exa?
<0> nope
<1> or more importantly, aiglx
<0> the exa code is totally broken
<0> aiglx = yes
<1> cool
<2> jbarnes does programs know how to use aiglx?
<1> nakee: yeah, all the fedora gnome composting stuff uses it afaik
<3> it's just GL
<3> glXCreateContext(dpy, 0);
<3> boom, you're using aiglx.
<1> but it needs aiglx to actually perform right?
<3> right
<2> ajax so what special about aiglx? couldn't program run in gl context before?
<1> nakee: dude, you're out of the loop :)
<1> nakee: aiglx makes indirect rendering fast
<2> jbarnes yea, I know that part, I just don't understand why programs has easier time to use indirect rendering rather than the old opengl that uses direct rendering accel?
<3> you would only use indirect glx for features that you can only get that way
<1> nakee: the compositing manager wants to be indirect
<3> like, say, GLX_EXT_texture_from_pixmap with the open drivers
<1> saves texture copies aiui
<3> plus it means you can run GL apps over the network and have them go fast, which is kinda cool
<4> it means you can have fast 3d card in low cpu near desktop, and the 3d app running in noisy huge remote server
<2> it's indirect because it's talking to the xserver (loading pixmaps to it) rather than the screen card?
<4> you are going to put cable extender makers out of business (i remember a studio setup in which they run lots of meters from box to input/output devices)
<2> actually it's very useful if you learn in university and knows the names of the cool graphic cl*** computers;)
<5> ajax: it looks like spectrah has a converter board "coming soon" that can deal with 1920x1200 and convert dvi to lvds
<6> I want to run linux (ubuntu) native in windows.
<7> yo
<5> anyone know how to tell which file a font comes from, with fontconfig?
<8> !h
<8> !help
<8> !command
<8> !commands
<9> ajax, you rock :)
<3> Jes_: it worked?
<3> also note, i810 driver is now stock 1.6.4; modesetting driver is called "intel" now.
<3> which i'm not happy about, but.
<9> ajax: yeah I noted that - only thing that doesn't work is vesa blanking of the monitor
<3> odd
<9> so it doesn't power down my monitor when I leave it
<9> but thats a minor nuisiance
<9> that also used to work with the old radeon card
<3> hm. interesting. i'll have to try that on mine.
<3> but i guess that means the edid quirks thing works; way coo.
<9> I'm now running 1600x1200 - so yeah
<3> can you send me the log?
<9> sent
<3> awesome. not only does it tell me the fixup triggered, but also my build id string hack worked
<9> kewl
<3> i like not having to ask people to rpm -q xorg-x11-server-Xorg
<9> :)
<9> if you figure out the blanking problem, you'll have me off your back .... for a while at least :)
<3> hmm. does 'xset dpms force off' work for you?
<9> that'll blank until I hit a key?
<9> nope
<3> yeah
<9> seems to keep it powered up
<3> does it at least flicker a bit?
<3> 'cause uh, mine works.
<3> well. works. the green light does stay on, one would think it should go orange.
<3> mmm, will poke
<9> it goes black, but it doesn't seem to turn the display down
<9> normally it sayes 'no signal' and then the led goes orange
<10> How do you do the equivalent of cvs update with git?
<3> alright, same as mine does then. cool, reproducible.
<3> ssp: git-fetch && git-rebase, i believe
<9> ssp, git-pull
<3> Jes_: git-pull flips out if you have local changes
<9> ssp, if you just want to pull down the latest
<9> ajax, you're making local changes without registering them in git?
<3> which is very much not the same as cvs up, which will attempt to merge bits
<9> thats what quilt is for :)
<9> quilt pop -a ; git-pull ; quilt push -a
<3> boo, quilt isn't in the developer install image in fedora!
<11> there's also stacked git, would be interested to hear a comparison
<3> aaw, it's in extras.
<3> that's sad
<10> ajax: git-fetch && git-rebase gives me "same as branch 'master', then "fatal: needed a single revision", then "invalid upstream"
<3> neat
<10> Jes_: git-pull says "Already up-to-date", which is not true
<10> Oh well, "git diff" works, and "git clone" works, so a simple script that computes the diff, deletes the repository, fetches a new clone, then applies the patch, should work
<3> or, diffs to a file, backs it out, pulls, and reapplies
<3> would be faster
<9> ajax, the people doing FC are utterly clueless
<9> ajax, thats what quilt does basically
<3> Jes_: hey, i try to keep X from ****ing at least
<9> ajax, I am not talking about the developers (well all of them at least)
<9> but I must admit, that I have seen BZ's be ignored by the person ***igned for > 3 months
<3> i think they gave me s-c-d as a form of hazing
<10> ajax: well the person that had s-c-d before you was certainly also guilty of ignoring it ...
<9> s-c-d?
<3> system-config-display
<3> System / Administration / Display if you're using gnome
<9> ajax, like there's a bz against xchat crashing in FC5 which the maintainer never even posted to
<9> yeah I know what it is :)
<10> alright, after patching local changes away, git-pull still claims "Already up-to-date", which is still not true
<10> It also says that "* refs/heads/origin: same as branch 'master' of git://git.freedesktop.org/git/xorg/app/compiz"
<9> something really weird about FC6T2's Firefox - I click 'install flash' it claims to install it, but afterwards it still pretends it's not installed
<3> feature?
<9> well not sure why it does it
Return to
#freedesktop
or
Go to some related
logs:
#web
smtpd_mumble_restrictions postfix
#perl
kylix sed: error while loading shared libraries: libc.so.6:
#gentoo
#perl
#lisp
FC5 clamd yum
#linux
dell inspiron 9300 gentoo ipw2200
|
|