agd5f: Bertl, yangman: http://www.botchco.com/alex/xorg/no_wait_if_engine_idle.diff
agd5f: see if that lets you start up X or resume in your respective cases.
agd5f: against the drm
mjt: whee. Just got an OOPS while loading X after fresh boot.
nanonyme: Hmm.
mjt: but it's not repeatable
nanonyme: * Current GPU state * SRBM READ ERROR indicated: Target Adr 0x0044 by HI/BIF
nanonyme: agd5f: Should I assume from that that exa test didn't altogether pass?
agd5f: nanonyme: you can ignore that
nanonyme: Right.
nanonyme: Goody, more stuff to ignore. ;) I was mostly just wondering why I get slowdowns in Xv every now and then, was wondering if r600_demo could give any hints.
nanonyme: agd5f: There wouldn't happen to be any possibility of running out of buffer or whatever? The slowdowns aren't that frequent.
agd5f: nanonyme: profile it
nanonyme: Any tips on how?
agd5f: and ErrorF()'s to the driver when it requests a dma buffer see if it has to retry, etc
agd5f: s/and/add/
agd5f: can also use somethig like oprofile
agd5f: or sysprof, etc.
nanonyme: It's not as if it makes playback unusable or anything. I just found it puzzling that the media player doesn't even seem to see the problem since it usually likes to complain of too slow playback.
nanonyme: And it's clearly visible that something happens every now and then for a split second.
nanonyme: Right, I'll take a look at oprofile, will be back at this later.
nanonyme: Going to postpone this for some time, apparently oprofile is broken for now in Debian testing.
nanonyme: agd5f: Just decided to poke still: tried with radeon driver, same result. It's either drm or lower.
nanonyme: Whoa.
nanonyme: Never mind. I thought scrolling got easier. Obviously it didn't. Scrolling still starts taking a lot of CPU if there's lots of pictures on a web page. :)
ma3x: hi
ma3x: anyone around?
ma3x: how do i add xvideo support for ati radeon hd 3400 series on a 2.6.28-1-686 kernel? using debian lenny but got the kernel from kernel-archives.net
rahman: ma3x: better look here: http://wiki.x.org/wiki/radeonhd:r6xx_r7xx_branch
AStorm: ma3x: short version: hard to do
ma3x: rahman: the problem is i can't find linux-headers for 2.6.28-1-686
AStorm: hmm, I guess any 2.6.28 headers will do
ma3x: i was about to install them when i got some dependency errors
ma3x: linux-headers-2.6.28-1-686: Depends: linux-kbuild-2.6.28 but it is not installable
AStorm: I recommend #debian
ma3x: they don't support sid
AStorm: hmmh
ma3x: and the only reason im using 2.6.28-1-686 is because it has the driver iwlagn for my wireless car
ma3x: d
ma3x: iwas using 2.6.26-1-686 before
rahman: ma3x: I dont remember well, but if you compiled your own kernel from scratch then you already have your headers.
ma3x: i didn't i installed the image
AStorm: yeah, but dpkg doesn't know about it
ma3x: what do you mean doesn't know about it?
rahman: ma3x: Ok, Isaw it like kernel.org :) my bad. BTW compiling radeonhd from git easy not complicated
AStorm: about the fact that you do have your headers
ma3x: AStorm: i don't have the headers
ma3x: i can't install them because they depend on kbuild which is broken
AStorm: then fix it or cheat dpkg
ma3x: but do i need kbuild or just the linux-headers
rahman: ma3x: so it seems there is 2 way: compile yor kernel by hand or use a kernel package that already have header packs and compile your wlan driver by hand. Ofcourse I dont know what problems you have with wlan
AStorm: for x11-drm you will need kbuild
AStorm: (the drm kernel module)
ma3x: i see
ma3x: then i need to compile my iwlagn i guess
ma3x: i don't know if that would be possible in 2.6.26-1
ma3x: ill make a little search
ma3x: but you see the wiki suggests that i install 2.6.28-1 http://wiki.debian.org/iwlagn
rahman: ma3x: if driver doesnt have a problem with exact 2.6.26.1 version kernel then it is easy to compile it
rahman: ma3x: I see
ma3x: rahman: what driver?
rahman: ma3x: wlan driver (or better module) you can always compile any kernel module by hand
ma3x: let me boot 2.6.26-1 and try to compile it
ma3x: i think i've done it before but it didn't work
ma3x: i'll try again and come back here later to report
rahman: ma3x: but It will not work as it says on wiki
ma3x: see ya
ma3x: not work what?
ma3x: on 2.6.26.1 ?
rahman: your wlan driver need a recent kernel
ma3x: so it wont work on 2.6.26.1?
rahman: So this is the debian way it is 2.6.26 kernel even on experimental repos. If you try just try with 2.6.28 kernel
ma3x: try what?
rahman: there are good howtos about kernel compiling for debian users. Try to compile 2.6.28 kernel
ma3x: hm
rahman: ma3x: wait a sec
ma3x: okay
rahman: ma3x: I had never tried but
rahman: ma3x: just go to kernel.org and get 2.6.28.1 kernel tarball
ma3x: and then
rahman: ma3x: take headers dir and place it right place
rahman: see if it works
ma3x: right place? where?
ma3x: /lib ?
rahman: it depends on distros
rahman: wait a sec
ma3x: ok
rahman: ma3x: /usr/includes
ma3x: i did that
ma3x: now i'll try http://wiki.x.org/wiki/radeonhd:r6xx_r7xx_branch
rahman: ma3x: do you have x11 dev packs installed
ma3x: no idea let me check
ma3x: is it x-dev package?
rahman: no idea what they call on debian. it should like libx11-dev
ma3x: ma3x@debian:~/drm/linux-core$ make radeon.o drm.o
ma3x: Makefile:153: *** Cannot find a kernel config file. Stop.
ma3x: i have /boot/config-2.6.28-1-686
ma3x: why does it say that
darkjames: readlink /lib/modules/`uname -r`/source
darkjames: or /build don't know.
darkjames: (AFAIR you need kernel sources for your version to build module for your kernel)
rahman: yes thats what you need: /lib/modules/$(RUNNING_KERNEL)/source
ma3x: and where do i get that from
ma3x: i downloaded linux-2.6.28.1 from kernel.org
ma3x: and i don't see any source/ inside
ma3x: or which files exactly should i copy to /lib/modules/2.6.28-1/source?
darkjames: It depends on your distro.
ma3x: debian lenny
ma3x: but the current kernel i run 26.28-1 is from sid i guess
darkjames: try apt-get install linux-source-2.6.28
darkjames: (is there anybody with Debian on board? I've got gentoo, and it's lot easier :>)
ma3x: ok im installing the source
ma3x: ok now what
rahman: ma3x: do you have 2.6.28 source in your apt repo? So why no header packages there?
ma3x: i did this
ma3x: i have the headers package but it's not installable because it depends on kbuild... which is broken/uninstallable
ma3x: ok now I type readlink /lib/modules/`uname -r`/source and then make radeon.o drm.o
ma3x: and i get again
ma3x: Makefile:153: *** Cannot find a kernel config file. Stop.
darkjames: readlink /lib/modules/`uname -r`/source
darkjames: should print somethink, at my workstation: /usr/src/linux-2.6.28
ma3x: prints nothing
darkjames: so the symlink is broken.
ma3x: there is no symlink
ma3x: /lib/modules/2.6.28-1-686/source doesn't exist
ma3x: dlocate says linux-source-2.6.28: /usr/src
ma3x: should i just make a symlink to /usr/src
darkjames: you can do: make LINUXDIR=/path/to/kernel/source radeon.o
darkjames: but before imo you need make prepare (in /path/to/kernel/source) and copy your config file.
darkjames: bbl
ma3x: i did that
ma3x: now i get error: /home/ma3x/drm/linux-core/drm_bufs.c: In function ‘drm_rmmap_locked’:
ma3x: /home/ma3x/drm/linux-core/drm_bufs.c:402: warning: enumeration value ‘_DRM_GEM’ not handled in switch
ma3x: /bin/sh: scripts/genksyms/genksyms: No such file or directory
ma3x: apt-file search scripts/genksyms/genksyms
ma3x: linux-kbuild-2.6.26: /usr/src/linux-kbuild-2.6.26/scripts/genksyms/genksyms
ma3x: and as you know linux-kbuild-2.6.26 is broken :(
ma3x: there isn't any path around this?
ma3x: ah it's not broken i already have it installed
ma3x: /bin/sh: scripts/genksyms/genksyms: No such file or directory
ma3x: how can i tell it it's in /usr/src/linux-kbuild-2.6.26/scripts/genksyms/genksyms
rahman: ma3x: Dont get me wrong but is there a special reason to use debian? Maybe it is good for servers but its simple lack of bleeding edge stuff.
ma3x: what do you use? gentoo?
ma3x: don't you have to compile stuff all day long?
ma3x: anyway
ma3x: is this the kernel headers directory now?
ma3x: /usr/src/linux-source-2.6.28
ma3x: or another directory inside that tree?
rahman: ma3x: kernel headers are in /usr/includes, kernel sources are in /usr/src/linux-source-2.6.28, and also you should have a symlink in /lib/modules/2.6.28.1/source that links to /usr/src/linux-source-2.6.28 . BTW I dont know what the genksyms error is
yangman: agd5f: no go. still freezing on hibernate resume
ech0s7: here doesn't freeze after resume
ech0s7: i have rv620
yangman: agd5f: getting "(EE) RADEONHD(0): RHDCSStop: Command Submission backend is not active!" towards the end of hibernation, actually. failed to notice that before
yangman: agd5f: X actually stops responding to kill -9 after it freezes
ech0s7: yangman: but now we can leave rv6xx branch of radeonhd and return to radeon ?
ech0s7: i have seen that they have done the merge
Bertl: any new about the X uses 100% and hogs down the system (while being unkillable) problem with HD3850?
yangman: Bertl: agd5f left you a patch about 7 hours ago
yangman: ech0s7: huh?
yangman: ech0s7: you mean master?
ech0s7: yes
yangman: ech0s7: yes
ech0s7: ok thanks
nanonyme: Heh, I just decided - out of interest - clone xf86-video-ati tree for comparison. It seems to behave noticeably slower for the stuff I've noted to be somewhat difficult for the drivers all in all (as in, scrolling pages, resizing windows, dragging windows). ^^
nanonyme: To clone even.
Bertl: yangman: ah, thanks, was working all day .. didn't see it
ech0s7: i have changed radeonhd to radeon
ech0s7: now i get
ech0s7: $ glxgears
ech0s7: unknown chip id 0x95c4, can't guess.
Ge0rG: so, what is the future of radeonhd vs. radeon now?
nanonyme: Probably still unknown.
ech0s7: Ge0rG: i haven't understood
nanonyme: (Other than that there will be progress in both :P)
ech0s7: if we have to continue to use radeonhd or radeon
nanonyme: ech0s7: Erm, which card were you running glxgears with?
ech0s7: nanonyme: hd3470 rv620
nanonyme: ech0s7: Are you aware that there is no 3D support for r6xx/r7xx?
ech0s7: yes i know, but with radeonhd glxgear know the chip
nanonyme: (Therefore glxgears doesn't really do anything useful)
nanonyme: Speaking of 3D, I wonder if dri2proto 2.0 is supposed to get into X.org 7.5.
nanonyme: What? They finally pushed X server 1.6 final out? :o
kkuno: yep
kkuno: dri2 is already in xserver 1.6 right?
nanonyme: kkuno: I'm not talking about dri2.
nanonyme: I'm talking about dri2proto version 2.0.
nanonyme: X.org 7.4 had dri2proto 1.1 iirc.
Grauwolf: hi
ech0s7: kkuno: dri2proto is DRI2 protocol headers
Grauwolf: i've got a problem. yesterday i switched from catalyst to xf86-video-radeonhd. since then tvtime doesn't start anymore
Grauwolf: it says "xvoutput: No XVIDEO port found which supports YUY2 images."
nanonyme: kkuno: They're currently in version 1.99.3 which I guess is kinda release candidate 3 for 2.0.
Grauwolf: any ideas?
ech0s7: Grauwolf: what card do you have ?
Grauwolf: http://bit.ly/taSt
Grauwolf: @ ech0s7
Grauwolf: ech0s7: sry wrong text in the clipboard. i've got ATI Radeon HD 3200
nanonyme: Grauwolf: Do you have radeonhd from master branch and drm kernel modules from r6xx-r7xx-support branch?
Grauwolf: nanonyme: i'm using the package from the archlinux reps. http://www.archlinux.org/packages/extra/i686/xf86-video-radeonhd/
nanonyme: That's too old, I think.
ech0s7: yes it's old
nanonyme: There isn't yet a release of radeonhd that has the necessary support.
Grauwolf: well. it's the last release
nanonyme: It will be in the next release.
Grauwolf: nanonyme: is there a roadmap? when will it be released?
nanonyme: Grauwolf: Not that I know of. It's just that the stuff was just moved to main branch from an experimental branch so it will get into the next release, whenever the release will be.
nanonyme: But there is the thing that you need to compile your kernel modules from drm development branch r6xx-r7xx-support in any case since that part of the support will only be in kernel version 2.6.30. ^^
Grauwolf: so what do i need? only the newest branch from git://anongit.freedesktop.org/git/xorg/driver/xf86-video-radeonhd
Grauwolf: ?
nanonyme: And r6xx-r7xx-support branch from git://anongit.freedesktop.org/mesa/drm
nanonyme: There probably is a guide somewhere for this, hum hum.
Grauwolf: nanonyme: there is alreade a package for that in aur http://aur.archlinux.org/packages.php?ID=24236
Grauwolf: thanks i'll try it
nanonyme: Right.
nanonyme: Grauwolf: That package is also likely obsolete. You no longer need r6xx-r7xx-support branch from radeonhd.
nanonyme: (Well, if it actually is a few days old, probably not that bad)
nanonyme: Grauwolf: Anyway, just radeonhd isn't enough. You still need the kernel modules.
Grauwolf: nanonyme: they are in the deps of this package ;)
ech0s7: Grauwolf: you have need of radeon-git and drm-git(branch r6xx-r7xx)
nanonyme: Grauwolf: Right, then it might work.
Grauwolf: i'm really happy that i found xf86-video-radeonhd yesterday. it's already working better on my system than catalyst
nanonyme: That's actually quite neat packaging.
ech0s7: Grauwolf: install from aur drm-radeon-module-git-r6xx-r7xx and xf86-video-radeonhd-git-r6xx-r7xx
nanonyme: Grauwolf: Of course, if you actually encounter bugs you want to help fixing, you probably end up compiling from git yourself anyway. :) But yeah, you should be able to see with that package if it works at all.
Grauwolf: ech0s7: the first one is a dep of xf86-video-radeonhd-git-r6xx-r7xx so installing that is enough ;)
ech0s7: ok
nanonyme: The second one has drm-radeon-module-git-r6xx-r7xx as a dep.
nanonyme: So yeah, it probably works.
Grauwolf: cross your fingers
nanonyme: Oh, it's actually a real git tree it's using instead of a snapshot. Never mind then, you can just put in possible patches in the git tree if necessary.
nanonyme: Grauwolf: You can tweak that if you want.
nanonyme: _branch="r6xx-r7xx-support origin/r6xx-r7xx-support" can probably be replaced with _branch="master"
nanonyme: (That is, in the radeonhd package)
nanonyme: Your distro will probably update it in a few days to that anyway though.
Grauwolf: thanks
Grauwolf: ok just installed the packages but no change. tvtime still says "xvoutput: No XVIDEO port found which supports YUY2 images. tvtime requires hardware YUY2 overlay support from your video card driver"
nanonyme: Grauwolf: Do you have lines Option "DRI" and Option "AccelMethod" "exa" in your device section of xorg.conf?
nanonyme: It's not yet *that* stable that it'd be on by default. :)
Grauwolf: nanonyme: http://attachr.com/private/55c0d74245ce2434abf8a1c73232325f
Grauwolf: shame on me. still having the names of aticonfig :)
nanonyme: Grauwolf: First off, remove extra device sections.
Grauwolf: k did that
Grauwolf: nanonyme: should i add option "DRI" and Option "AccelMethod"?
nanonyme: And "exa" at the end of the second one.
Grauwolf: nanonyme: Driver "radeonhd" "exa" ?
nanonyme: No, Option "AccelMethod" "exa".
Grauwolf: ah ok
nanonyme: You can probably drop the busid if you only have one display adapter card.
Grauwolf: ok i'm restarting x now
Grauwolf: nanonyme: same problem
nanonyme: Care to show your Xorg.0.log?
Grauwolf: jep mom
Grauwolf: wow actually know everything is damn slow. when i want to move a window everything happens in slow motion
Grauwolf: wow http://attachr.com/private/0ab6107ac395d62196fd5d1fab0af805
nanonyme: Grauwolf: The drm and radeon kernel modules likely weren't loaded.
nanonyme: It fails to open /dev/dri/card0, as it says.
nanonyme: So you get no DRI.
nanonyme: Try modprobing radeon, then starting X again.
Grauwolf: radeon is modprobed
nanonyme: Hrm.
nanonyme: (EE) RADEONHD(0): RHDDRIVersionCheck: drmOpen("radeon", "pci:0000:01:05.0") failed. (WW) RADEONHD(0): RHDDRIPreInit: Version check failed. Disabling DRI.
nanonyme: That's the key, actually.
nanonyme: For some reason you haven't got the right drm modules.
yangman: remove the module, modprobe it again, and check dmesg
Grauwolf: i just did modprobe -r radeon && modprobe radeon; killed x and restarted it. no difference
yangman: Grauwolf: dmesg would give hints on why the dri device isn't created
yangman: check there
yangman: probably inproperly installed modules
Grauwolf: [drm] Module unloaded
Grauwolf: [drm] Initialized drm 1.1.0 20060810
Grauwolf: still slow as hell
yangman: pastebin the last few dozens lines of dmesg output
nanonyme: Kinky, I wonder how I managed to get this happen. [drm:drm_mmap_locked] *ERROR* Could not find map
nanonyme: Four of those in dmesg.
Grauwolf: yangman: http://attachr.com/private/5a83f4ff356d3dbb562cc9013f9039e9
yangman: Grauwolf: I don't see radeon being loaded there at all
nanonyme: Hey, wait a minute. I've talked about that before.
Grauwolf: thats the problem
nanonyme: No answer then either though.
Grauwolf: but i don't know why i installed drm-radeon-module-r6xx-r7xx
Grauwolf: is there anything i shoud modprobe for drm?
Grauwolf: anything beside modprobe radeon
nanonyme: yangman: Wrapper for automatically compiling stuff from git on his distro.
nanonyme: Just a note.
Grauwolf: http://bit.ly/P5cLJ
Grauwolf: thats the PKGBUILD of my distro
Grauwolf: for the drm-module
yangman: does lsmod actually show radeon being loaded?
Grauwolf: yes
yangman: hm
yangman: try doing a depmod
Grauwolf: sry i don't know depmod. what do you want me to do?
yangman: just run 'depmod'
Grauwolf: yangman: nothing happened
yangman: remove the modules and try again
nanonyme: Rather: nothing got displayed.
nanonyme: Not the same as nothing happened.
yangman: and you don't need to restart X every time. we need to make sure /dev/dri/card* gets created first
Grauwolf: yangman: after modprobe -r radeon depmod still doesn't show anything
yangman: Grauwolf: depmod wouldn't show anything
yangman: Grauwolf: I just wanted to make sure module dependencies are updated properly. you don't need to run depmod again
Grauwolf: yangman: ah ok
yangman: sorry, wasn't specific enough. I meant try modprobing it again when I said "try again"
Grauwolf: k did that
agd5f: Bertl: Please try the drm patch and let me know if it helps: http://www.botchco.com/alex/xorg/no_wait_if_engine_idle.diff
yangman: check if any nodes are created in /dev/dri/
Grauwolf: yangman: /dev/dri is empty
yangman: pastebin the last bits of dmesg again
Grauwolf: http://bit.ly/snSqT
yangman: hm... well, I'm stumped
nanonyme: Maybe you could try removing the modules with rmmod, then go insmodding them from the git tree directly.
nanonyme: Would give the answer as to if the right modules are being inserted over and over again here.
Grauwolf: well. everything (except of tvtime) is working with the stable version
yangman: Grauwolf: 2D acceleration requires DRM to be working
agd5f: Grauwolf: make sure you don't have old radeon.ko and drm.ko modules floating around
agd5f: drm isn't initialized until X starts
Grauwolf: agd5f: i already did a full reboot
agd5f: Grauwolf: doesn't matter if you have old modules around. make sure they are gone
Eythan: hello, i have "(II) RADEONHD(0): Unknown card detected: 0x9598:0x174B:0xE410." in my /var/log/Xorg.0.log
Eythan: i have a sapphire Radeon HD 3450
Eythan: but, everything seems working perfectly
yangman: Eythan: that message means there's no known modesetting quirks, not that the card isn't recognized
yangman: Eythan: it's a little misleading
Eythan: yes, if i can help reporting this
Grauwolf: ok i just switched back to the stable version. i thin i just have to wait for the next release. until then i have to use xawtv instead of tvtime
Grauwolf: thanks for your help
yangman: Grauwolf: you're going to be waiting until the new drm is released in a new kernel version, which will be a while
Grauwolf: ah back to stable. everything (except of tvtime) is working fine again
Grauwolf: yangman: i'm a patient person ;)
nanonyme: Grauwolf: It's apparently going to be in kernel 2.6.30 and 2.6.29 isn't even released yet.
yangman: that's assuming it's accepted
nanonyme: Yes, assuming that.
nanonyme: Though if the code is good and the time windows are followed so Linus won't get pissed (again), I suppose there's good chances of it getting accepted.
nanonyme: Too many video packets in the buffer: (195 in 8393651 bytes) # kinky
nanonyme: For God's sake, is my CPU actually too slow for decoding 1080p mpeg4?
bridgman: how many cores are you actually using ?
bridgman: I bet there are at least as many cores "watching" as "working"
nanonyme: bridgman: It's an AMD64 Athlon 3500+ so only one core. :P
yangman: run a mplayer benchmark
nanonyme: Erm, how?
yangman: mplayer -benchmark -nosound -vo null
yangman: see how long it has to spend decoding the video
nanonyme: Didn't work.
nanonyme: No such command.
bridgman: nanonyme; mpeg4 has at least 2 standards, one of them is h.264
yangman: so, no mplayer?
nanonyme: yangman: Sure there is. Mplayer just doesn't have the switches you said.
nanonyme: Some of them anyway.
yangman: it definitely should
yangman: you still need to give it a media file, btw
nanonyme: Lemme try again.
mjt: all switches mentioned are here since the stone age...
nanonyme: Well, it started. Still going.
mjt: and mplayer can't use more than single core anyway
yangman: libavcodec has multithreaded decoding option
nanonyme: bridgman: This is h.264, apparently.
yangman: it defaults to 1
mjt: does it actually work with mplayer?
mjt: and how to activate it?
yangman: well, I have it set to 8. lets see if it's any slower with less
yangman: lavdopts=threads=N
yangman: er, that's from config. cmd line argument has different syntax
nanonyme: Would it be a bad sign if the benchmark is still going? :p
yangman: how long is the video?
yangman: it's going to try and run through it, but just not output anything
nanonyme: Oh, right. That's a point.
nanonyme: yangman: It shows a percentage there. Does that show how full buffer currently is?
nanonyme: Or what?
yangman: there's 2 lines
yangman: the first line is stats in seconds
yangman: take the final value (right-most) and divide by the actual runtime of the media. that gives you time spent decoding
nanonyme: Otherwise fine but this was a full length movie. >.<
nanonyme: V: 512.9 12299/12299 85% 0% 0.0% 0 0 # the last stuff it showed before I fed up. Might leave it overnight.
yangman: go download a HD trailer from apple.com and test on that
yangman: that's what I do
yangman: incidentally, I've probably watched the trailer for Up about 3 dozen times by now >.>
nanonyme: yangman: Apple site crashes my browser.
yangman: unfortunate
nanonyme: Plus I can't download anything from there.
nanonyme: Neither with native Firefox nor a Firefox I have installed inside Wine.
yangman: oh, yeah, it uses those wrapped quicktime things...
yangman: I usually let gnome-mplayer open it, then copy the file URL and wget it
nanonyme: Care to pick a sample URL for me? Apparently my system hates apple.com
MostAwesomeDude: example.com
nanonyme: :p
yangman: http://movies.apple.com/movies/wb/watchmen/watchmen-tlr2_h1080p.mov
yangman: 108MB
yangman: er, 196MB
nanonyme: Okay, downloaded.
yangman: that was fast
nanonyme: Well, I do have a 100Mbps Internet connection.
yangman: totally jealous
nanonyme: Come study here, you'll get one too. ;)
nanonyme: It's technically shared but no one uses it on average.
yangman: heh
yangman: I've considered transplanting myself to Europe, but never really seriously
nanonyme: BENCHMARKs: VC: 84.778s VO: 0.019s A: 0.000s Sys: 1.124s = 85.922s
nanonyme: yangman: That's not h.264 though. Maybe my CPU can do full HD but not if it's h.264... :/
yangman: all the 1080p trailers on apple should be H.264
nanonyme: Oh, wait.
nanonyme: Read wrong, it is.
nanonyme: Wonder what the heck is wrong then.
nanonyme: Same encoding, same resolution, same fps.
nanonyme: I'm starting to wonder if the other movie is corrupt.
yangman: well, that's not so good. r6xx merge regressed Xv for r5xx
rehabdoll: screw r5xx
rehabdoll: :D
brandon_: !list
brandon_: Triakis installer?
yangman: brandon_: are you looking for OSX drivers?
brandon_: yes
yangman: #radeonhd on chat.osx86.hu
yangman: you're on the wrong server
brandon_: thanks
Tanktalus: ok, new EXA support compiled in from the main trunk ... but moving windows around is ... slow :-(
yangman: Tanktalus: you still need DRM from r6xx-r7xx-support branch
Tanktalus: yangman: ah... hmm... that wasn't merged back to main?
Tanktalus: though I did get my dual monitors to work... though I'm thinking that I don't really want the two monitors to be wide screen so much as being completely independent desktops...
Tanktalus: oh, apparently, xrandr 1.2 doesn't support this. *sigh*.
yangman: Tanktalus: drm is technically a separate project. it's being prepped for pushing upstream to kernel right now
Tanktalus: libdrm? from dri.freedesktop.org?
yangman: no, drm. it belongs in the kernel
yangman: http://wiki.x.org/wiki/radeonhd%3Ar6xx_r7xx_branch
Tanktalus: hmm... I have a drm module loaded in my kernel according to modprobe...
Tanktalus: but it's probably too old.
yangman: r6xx support in drm hasn't gone upstream yet
yangman: it's not in any release
Tanktalus: yeah, I'll have to see if I can get someone in gentoo-land to help with the ebuild ;-)
AStorm: Tanktalus: what do you need help with?