| |
| |
| |
|
Comments:
<0> I have to say I dont quite remember how has what cards, and that information is usually available in our wiki :)
<0> so it might be interesting, but I can't really say right now, sorry
<1> I might even be able to loan it for a limited time. I'll check wiki next week or get back to me :)
<0> shawn_home: well, you could run the tests on it anyway, and let me know
<2> sure, as long as it doesn't hose the box :)
<2> does X have to be not running?
<0> leio: ok, I'll bug you if I need it
<0> shawn_home: X up & running with proprietary drivers. and no, it doesn't hang the box
<2> yes got the binary blob
<2> just compile the tool and run?
<0> well, no, that's the catch :)
<2> er?
<0> edit re.c, uncomment OUTPUT_MULTIPLE_FILES
<2> edit, then compile
<0> edit main.c, uncomment all tests except a couple that don't compile
<0> and then compile :)
<2> is it commented which don't compile?
<0> no, because it depends what's in your glext.h
<2> so compile until it builds
<0> yeah :)
<0> after you run, it'll output one .txt file per test. it's those files we're interested in
<2> ok logging to my work box.. (I love NX Machine)
<0> and really, we're discussing this in the wrong chan
<3> hi krh. i was wondering how akamaru relates to the physics model in compiz. are they completely independent implementations?
<4> alp: the compiz model is based on the wobbly windows from luminocity, which was based on the sproing prototype from the cairo-demo repository
<4> alp: akamaru is a generalized version of that model, which also uses a differenct numerical integration technique
<3> krh: ok, i just looked at the compiz model for the first time and was surprised to see how similar it was to my akamaru clone in compiz-sharp. guess i reached the same conclusion independently which is a good sign
<3> thanks, i've updated http://www.ndesk.org/CompizSharp with this information and a screenshot
<4> alp: nice
<2> krh: I just tried your aiglx compiz patches
<2> (with some hacking for some of them)
<4> shawn_home: did it work for you?
<2> well the GPU didn't wedge
<2> but, the display didn't update
<2> im going to grab an ubuntu compiz-aiglx package instead of the one from git
<2> krh: im not sure if anyone has tried compiz aiglx with the r300
<2> imtrying to find out who
<2> http://lovesunix.net/blog/?p=97
<2> hmm
<4> shawn_home: I've been using compiz on an r300 for the last couple of weeks
<4> shawn_home: you need mesa cvs
<2> im trying your src.rpm and manually patching it in got mesa CVS
<2> replaced the distro's mesa libs and GL/include
<3> whatever happened to the texman branches? is that stalled?
<2> krh: unless the kde window decorator is broken
<2> but it seems to crash even before that
<2> compiling
<4> shawn_home: oh... I haven't tried the kde decorator
<2> i can build the gnome one just need the dev stuff
<2> yeah i'll get the gnome stuff
<2> compiling
<2> now to run compiz --replace...
<5> alp: the new memory manager ones?
<3> airlied: yep. is glib licensing still an issue?
<5> alp: it would be to link glib into the X server..
<5> alp: the new memoey manager work in stil lgoing..
<5> there is a new i915 texman branch from last week..
<5> I'm expecting new cleanedup ttm code for the kernel by the end of this week..
<3> airlied: the mono project is now developing an mit licensed reimplementation of the core features in glib called eglib
<3> it's api compatible, but will lack gobject, key file parsing and all that fluff
<5> alp: yeah I noticed that.. we might have to see how it goes..
<3> it's become a cause celebre and it has the glib test suite to help out, think it will work out. in the meantime i might give i915 texman another try.
<3> there is a mesa branch as well, right?
<6> hrm
<6> DISPATCH ERROR! _glapi_add_dispatch failed to add glProgramLocalParameters4fvEXT!
<6> DISPATCH ERROR! _glapi_add_dispatch failed to add glProgramEnvParameters4fvEXT!
<6> it doesn't wedge, but the display is not updated
<6> my guess is debian's Xorg of 7.1.1 is still 'old'
<6> so mesa cvs head + compiz-aiglx patches from src.rpm still won't work with my combination
<6> krh: is it possible the r300 mobile series doesn't work ?
<6> do you have a r300 mobile or desktop card?
<5> alp: yup in the i915 driver,....
<5> I don't think there are many core mesa changes..
<6> i guess im gonna have to wait til debian experimental has compiz-aiglx merged in (soon)
<3> will have to remeber cvs branch co syntax, heh. been a while
<3> or is this in one of the new mesa git imports?
<5> alp: I don't think new branches aeppeared in krh's mesa...
<5> alp: I also have a kernel tree with the kernel stuff on kernel.org
<3> airlied: i noticed. it's really such a big change from the old days
<4> alp: the texmem_0_2_branch?
<3> krh: looks like it
<5> krh: yup that branch...
<2> airlied: EXA for r300 in current ati driver is greatly improved the delay from displaying a window full screen from offscreen buffer, can't wait for EXA memory manager
<4> hmm, not sure why that branch is missing...
<7> hi there
<8> hi
<8> hmm.. is the listserver broken?
<8> I see messages on gmane that was posted to hal@lists.freedesktop.org
<8> but it appears they are sent to (all?) list members including myself
<8> s/are sent/are not sent/
<8> any admins around to look at the mail queues for this problem?
Return to
#freedesktop
or
Go to some related
logs:
libdvdcss2.deb
#lisp
wusb11 v3 knoppix
#perl
#linux
#lisp
xp write ext3
tightvnc connection refused 111
socket reconnect python
#debian
|
|