level1: hi, I can't figure out how to get S-Video out to work. Is this something that is likely to work at this stage or has it not been implemented?
mcgreg: hi, do you know who the maintainer of the radeonhd.org page is? the link: the link for xf86-video-ati -> http://gitweb.freedesktop.org/?p=xorg/driver/xf86-video-ati;a=summary doesnt work : 403 Forbidden - No such directory.
mcgreg: guess it shpuld be this: http://gitweb.freedesktop.org/?p=xorg/driver/xf86-video-ati.git;a=summary
so1: hi
so1: did something worthwhile happen in the last few days to check out? :-P
vladimir_: Hi, i've run into some annoying issue with radeonhd - if mode switch occures or screensaver blank the screen then gamma value is reset to 1.0
vladimir_: xgamma report gamma to be set to correct value
egbert: vladimir_: this doesn't sound like an issue of the driver. the ddx layer should really handle this and reload the gamma palette.
vladimir_: egbert, ok, thanks for info
rmh3093: i know this is not the right channel for this, but is there an official channel for radeonfb?
rehabdoll: rmh3093: no
annonygmouse: Hi folks!
annonygmouse: I'm having problems with last version of radeonhd (I tested 0.0.4 and also updated an hour ago or so).
egbert: annonygmouse: tell...
annonygmouse: With this new version X run OK, but the mouse cursor doesn't work and keyboard not fully works either...
annonygmouse: Here you have the Xorg.log http://pastebin.com/m150dfeb5
egbert: hrm, but you see a mouse cursor on the screen?
annonygmouse: yes
egbert: i don't think this has anything to do with the output driver.
annonygmouse: but when I physically move it, it doesn't reflect the movement on the screen...
egbert: this seems to be an input problem.
annonygmouse: using an older (the one I'm actually using) works...
annonygmouse: (I meant an older radeonhd version)
egbert: annonygmouse: strange....
annonygmouse: I would gess there's something strange going on at radeonhd level, I've used loglevel 7 and I think it should
annonygmouse: show more things after "(II) RADEONHD(0): FUNCTION: TMDSARestore" ...
annonygmouse: shouldn't?
annonygmouse: at least the radeonhd version that works ends this way:
annonygmouse: (II) RADEONHD(0): FUNCTION: RHDShadowCloseScreen
annonygmouse: (II) RADEONHD(0): FUNCTION: rhdUnmapFB
annonygmouse: (II) RADEONHD(0): FUNCTION: rhdUnmapMMIO
annonygmouse: FreeFontPath: FPE "/usr/share/fonts/X11/misc" refcount is 2, should be 1; fixing.
egbert: this is when you shut down the server.
annonygmouse: (after (II) RADEONHD(0): FUNCTION: TMDSARestore)
annonygmouse: ok
egbert: this should not affect how it performs after you start it.
egbert: still strange.
egbert: can you find out which version worked for you?
egbert: when did you last update before?
annonygmouse: after the randr merge it didn't work at all, and since a couple of days ago it works but no mouse and no keyboard.
annonygmouse: I've just made git pull an hour ago or so.
Tigerchen: annonygmouse: no mouse could be caused by evdev
Tigerchen: got that problem yesterday
annonygmouse: this is the radeonhd version that works for me:
annonygmouse: (II) Loading /usr/lib/xorg/modules/drivers//radeonhd_drv.so
annonygmouse: (II) Module radeonhd: vendor="AMD GPG"
annonygmouse: compiled for 1.4.0, module version = 0.0.2
egbert: Tigerchen: but appearantly it had been working before the randr merge. unless something else has changed, too.
egbert: annonygmouse: this output isn't very helpful
annonygmouse: maybe this is more accurate: RADEONHD: version 0.0.2, built from git branch master, commit dbb073b8..
Tigerchen: egbert: well, yesterday i got the no-mouse-problem and unmerging evdev solved it
egbert: ok.
Tigerchen: because evdev claimed mouse and keyboard and ignored the config i made
egbert: Tigerchen: what changed so that it was working before?
Tigerchen: egbert: i think a new version of some xorg-related package, dunno what exctly. and for me just unmerging xf86-input-evdev solved it so i didn't bother to investigate further
egbert: Tigerchen: ok.
annonygmouse: egbert, tigerchen: Is there any more info I can give you?
egbert: annonygmouse: so when you just switch the radeonhd driver and don't change anything else it changes for you?
Tigerchen: egbert: but i think that problem is not related to raeonhd, because eisnerd who is using fglrx had it too
egbert: how do you switch the driver.
annonygmouse: giving modulepath to "X"
egbert: Tigerchen: yeah, i believe so, too. bot annonygmouse said that one driver is working for him, the other isn't.
Tigerchen: oh, k
annonygmouse: let me try something...
Tigerchen: should learn not to interfere in communications of other people
egbert: annonygmouse: ok, do you have other modules in the radeonhd path (of the non-working version), too?
egbert: could be that the modules just differ.
egbert: Tigerchen: no problem. your input was helpful.
annonygmouse: mmm yes
annonygmouse: as the "make install" puts radeonhd at /usr/local/xorg/modules/drivers/ in order to use the modulepath switch of X I had
annonygmouse: to copy some the modules from /usr/lib/xorg/modules/*.so to /usr/local/lib/xorg/modules
annonygmouse: but I've checked them right know and they are from 29/09/2007... so no change... for a while...
egbert: afair you can separate module directories with a ',' in the path. so you can use your standard path and just prepend it with a directory that only contains radeonhd.
annonygmouse: mmm don't worry, to be sure there's nothing strange from running using modulepath, I'll copy the radeonhd.* to /usr/lib/xorg/modules/drivers and reboot X.
annonygmouse: ok?
egbert: annonygmouse: ok, but you don't have to reboot.
egbert: :)
annonygmouse: (I meant "reboot" X :)
egbert: annonygmouse: :)
annonygmouse: be right back
annonygmouse: egbert?
egbert: annonygmouse: yup!
annonygmouse: you were right! I'm using: (II) RADEONHD: version 0.0.4, built from git branch master, commit 8c5f3454
annonygmouse: and all seems OK.
egbert: annonygmouse: :)
annonygmouse: Maybe it was some problem with the input drivers or something strange because I was having X running (kdm) and tried to use X :1 .....
annonygmouse: no idea...
annonygmouse: but it works! :>
annonygmouse: thanks for the help and sorry to disturb you with this nonBug :O
alessandro: theres some weirdness going on here
alessandro: when i start the Xserver with VGA plugged in
alessandro: it detects the external monitor nicely
alessandro: whereas when i plug it in later
alessandro: i only get up to the resolution of the laptop
egbert: alessandro: this is because it wasn't detected.
annonygmouse: bye!
alessandro: egbert: what do you mean by detected?
alessandro: i mean xrandr display is it nicely
alessandro: but only with reso up to 1400x1050 as on the lap
egbert: alessandro: the monitors are detected and the data about the resolution is collected.
egbert: alessandro: so you mean the mode has been added to the list of available modes but cannot be set?
egbert: ... added when you plugged in the display?
alessandro: yes the modes have been added
alessandro: but not the external monitor's native
egbert: alessandro: this may be an randr problem. maybe you want to wait until emmes comes back.
alessandro: alright
vjm: hi, I just recently changed to radeonhd and it works fine, but does complain about unknown card in xorg.log. So should I send the conntest data to the radeonhd-list?
egbert: vjm: if it works fine you don't have to.
vjm: ok, it works fine...I'm now much happier than with fglrx.. =)
yaneti: somebody care to add rhd_mc.h to Makefile.am ..
egbert: yaneti: ok, sorry.
yaneti: egbert: thx, nbd, it just helps testing
egbert: yaneti: if you give me your email i'll mention you as the author :)
yaneti: no need
egbert: ok, pushed.
yaneti: still has that DDC crash from the other day ..
egbert: yaneti: has emmes looked into this?
yaneti: egbert: i think he said its probably not randr related and was gonna consult with you
egbert: hrm, ok. do you have a log for me? describe the symptoms?
yaneti: http://www.declera.com/~yaneti/radeonhd/xorg.crash.log , backtrace
yaneti: http://www.declera.com/~yaneti/radeonhd/xorg.crash.1124.log more recnt one
yaneti: corresponding verbose x log http://www.declera.com/~yaneti/radeonhd/Xorg.0.log.112401.ce2607d1
egbert: yaneti: ok.
yaneti: i just tried with the latest tree and its the same
yaneti: have a gdb prompt if you need
yaneti: i am wondering if x itslef has some problems in that area that were fixed recently
egbert: yaneti: ok, so you are in the crash?
yaneti: yep
egbert: do you have the sources around?
yaneti: i have them as in the debuginfo package
yaneti: ?
egbert: ok, can you go one frame down and do a list, please?
egbert: my sources may differ from your upstream sources.
egbert: (X sources)
yaneti: which function
egbert: xf86I2CProbeAddress
egbert: xf86i2c.c:400
yaneti: ok
yaneti: 395 d.StartTimeout = b->StartTimeout;
yaneti: 396 d.SlaveAddr = addr;
yaneti: 397 d.pI2CBus = b;
yaneti: 398 d.NextDev = NULL;
yaneti: 399
yaneti: 400 r = b->I2CAddress(&d, addr);
yaneti: 401
yaneti: 402 if (r) b->I2CStop(&d);
yaneti: 403
yaneti: 404 return r;
egbert: hrm, can you print out b and see if I2CAddress is set?
yaneti: ) p b->I2CAddress
yaneti: $2 = (Bool (*)(I2CDevPtr, I2CSlaveAddr)) 0
egbert: bummer!
egbert: yaneti: i will check this.
dvandyk: evening
yaneti: egbert: the machine will be available for testing
dvandyk: can somebody tell me if there is an ETA on the hardware info for 3d support?
egbert: yaneti: ok. i should be able to find this easily.
dvandyk: nobody?
yangman: dvandyk: there isn't one, afaik
airlied: dvandyk: no
dvandyk: that's a pitty
airlied: dvandyk: writing a 3d driver could take ~6months from hw release info in any case
dvandyk: well, i don't want the 3d driver anyway, i just want the hardware information :-)
dvandyk: think HPC on AMD GPUs :-)
airlied: dvandyk: you'll still need a driver..
airlied: dvandyk: and a compiler of some sort
dvandyk: but not a 3d driver :-)
dvandyk: airlied: the latter will be the bigger problem, yes
airlied: dvandyk: most 3D drivers are jujst compilers nowadays
dvandyk: nod
LoneTech: is it known that video out just dies if you don't disable aiglx in the config file?
yaneti: egbert: just downgraded X to the fedora 8 packages, xorg-x11-server-Xorg-1.3.0.0-33.fc8 rebuilt the driver and this time it starts
yaneti: the rawhide server is at 1.4.99.1-0.10.fc9
airlied: yaneti: rawhide X is nsfw :-)
yaneti: airlied: well, since its all bleeding edge i thought i can try the radeonhd progress with it
yaneti: for something so bleeding the rawhide x package is awfully stale
airlied: yaneti: how stale?
airlied: yaneti: it shouldn't be more than a week or so from master.
yaneti: Wed 14 Nov 2007
airlied: yaneti: yeah thanksgiving slows down the US ppl working for Red HAt.
airlied: yaneti: I could rebase it but I'm trying to avoid rawhiding my machines until I fix F8 problems..
yaneti: airlied: cant you or ajax just throuw a built-from git command at something and leave the pain to the rawhide crowd
airlied: yaneti: not yet.. we have to rebase patches etc..
airlied: yaneti: ajax is pushing most of them upstream though.. but we always have aa few Fedora/RH psecific ones..
yaneti: if you could leave all that to ndim, I am sure it will be take care of :)
yaneti: egbert: sorry if i had you chasing ghosts
egbert: yaneti: i'm not sure if it's a ghost, yet.
yaneti: good
egbert: yaneti: the i2c code is inconsistent. when supplying a writeread one doesn't have to supply a probeaddress. however this is not emulated. of course i can supply a probeaddress myself.
yaneti: egbert: sorry, this was more chinese than I can handle ;)
egbert: yaneti: doesn't matter :) the function that was NULL should have been supplied by the driver.
ndim: reads backlog
ndim: yaneti: Ah, there you are. I thought for a second you were afraid of me. :)
yaneti: ndim: nah, silly power
yaneti: ndim: so you mean you don't want to package X for fedora ?:P
yaneti: bleeding edge X that is
ndim: yaneti: I had to make sure that there is at least one configuration it works with (e.g. mine).
ndim: There is no sense in shipping a package if I can't at least be confident that it works at least on SOME system.
yaneti: ndim: you are doing a great job with xorg-x11-drv-radeonhd, it'll be great if all fast moving targets are followed that way
ndim: yaneti: current test builds are at http://radeonhd.lauft.net/#my-rpms
ndim: Is readeonhd intentionally adding all those blank lines to Xorg.${N}.log? It seems to be the only Xserver module to do so...
EnsGabe: Hi- I'm using the latest xrandr and radeonhd gits and I can't seem to get multiple monitors to work how I'd like. I can enable my external monitor only as a clone of the primary one- when I try to set it as a display to the left or right, X crashes. I have a Thinkpad T60 with a Mobility X1400 and run ubuntu gutsy (so it might be a problem with the xserver- it's not the current version.) Is this something that should be working and