--- Log opened Вск Сен 10 00:00:35 2006
00:22 -!- Myrizio [n=Myrizio@host211-98-static.104-80-b.business.telecomitalia.it] has joined #nouveau
00:24 -!- Aexoden [n=Aexoden@uranus.aexoden.com] has quit [Remote closed the connection]
00:31 -!- swany [n=swany@81-234-181-143-o1108.tbon.telia.com] has quit []
00:34 -!- swany [n=swany@81-234-181-143-o1108.tbon.telia.com] has joined #nouveau
00:38 -!- Aexoden [n=Aexoden@uranus.aexoden.com] has joined #nouveau
00:51 -!- Aexoden [n=Aexoden@uranus.aexoden.com] has quit [Remote closed the connection]
00:53 -!- Aexoden [n=Aexoden@uranus.aexoden.com] has joined #nouveau
01:11 -!- open_source_nut [n=billalbr@dialup-4.245.39.47.Dial1.StLouis1.Level3.net] has joined #nouveau
01:12 -!- open_source_nut [n=billalbr@dialup-4.245.39.47.Dial1.StLouis1.Level3.net] has left #nouveau []
01:18 -!- darktama [n=darktama@gentoo/contributor/darktama] has quit [Read error: 110 (Connection timed out)]
01:18 -!- open_source_nut [n=billalbr@dialup-4.245.39.47.Dial1.StLouis1.Level3.net] has joined #nouveau
01:20 -!- EdB [n=EdB@ARennes-251-1-91-200.w86-199.abo.wanadoo.fr] has quit ["Konversation terminated!"]
01:21 -!- predatorfreak [n=predator@adsl-69-212-52-66.dsl.sfldmi.ameritech.net] has quit ["<blank>"]
01:25 -!- open_source_nut [n=billalbr@dialup-4.245.39.47.Dial1.StLouis1.Level3.net] has quit [Client Quit]
01:28 -!- predatorfreak [n=predator@adsl-69-212-52-66.dsl.sfldmi.ameritech.net] has joined #nouveau
01:30 -!- open_source_nut [n=open_sou@dialup-4.245.39.47.Dial1.StLouis1.Level3.net] has joined #nouveau
01:42 -!- `Duke` [n=gnu@ANantes-251-1-175-153.w90-12.abo.wanadoo.fr] has quit ["Fatal signal: Segmentation Fault"]
02:26 -!- maxtoo [n=maxtoo@berryx.homedns.org] has quit ["Les choses que l'on possède, finissent par nous posséder"]
02:34 -!- open_source_nut [n=open_sou@dialup-4.245.39.47.Dial1.StLouis1.Level3.net] has quit ["Leaving"]
03:02 -!- marteus [n=marteus@0x503fbab3.albnxx8.adsl-dhcp.tele.dk] has quit [Read error: 110 (Connection timed out)]
03:07 -!- b33fc0d3 [n=bifter@gentoo/contributor/b33fc0d3] has quit [Read error: 60 (Operation timed out)]
03:58 -!- Unavowed [n=silent@13.t6.ds.pwr.wroc.pl] has quit ["leaving"]
04:10 -!- b33fc0d3 [n=bifter@gentoo/contributor/b33fc0d3] has joined #nouveau
04:21 -!- Unbeliever is now known as K_Flipando
04:39 -!- K_Flipando [n=hazel@84-16-252-194.internetserviceteam.com] has quit [Remote closed the connection]
04:52 -!- open_source_nut [n=open_sou@dialup-4.245.57.75.Dial1.StLouis1.Level3.net] has joined #nouveau
04:53 -!- open_source_nut_ [n=billalbr@dialup-4.245.57.75.Dial1.StLouis1.Level3.net] has joined #nouveau
04:53 -!- open_source_nut_ [n=billalbr@dialup-4.245.57.75.Dial1.StLouis1.Level3.net] has left #nouveau []
05:07 -!- osn_ [n=opensour@dialup-4.245.57.75.Dial1.StLouis1.Level3.net] has joined #nouveau
05:08 -!- osn_ [n=opensour@dialup-4.245.57.75.Dial1.StLouis1.Level3.net] has left #nouveau []
05:13 -!- open_source_nut [n=open_sou@dialup-4.245.57.75.Dial1.StLouis1.Level3.net] has quit ["Leaving"]
05:30 -!- open_source_nut [n=opensour@dialup-4.245.57.75.Dial1.StLouis1.Level3.net] has joined #nouveau
05:41 -!- cetrox [i=cetrox@57-246-222-201.adsl.terra.cl] has quit [Remote closed the connection]
05:41 -!- cetrox [i=cetrox@57-246-222-201.adsl.terra.cl] has joined #nouveau
05:55 -!- cetrox [i=cetrox@57-246-222-201.adsl.terra.cl] has quit [Remote closed the connection]
05:55 -!- cetrox [i=cetrox@57-246-222-201.adsl.terra.cl] has joined #nouveau
06:02 -!- cetrox [i=cetrox@57-246-222-201.adsl.terra.cl] has quit [Remote closed the connection]
06:03 -!- cetrox [i=cetrox@57-246-222-201.adsl.terra.cl] has joined #nouveau
06:16 -!- open_source_nut [n=opensour@dialup-4.245.57.75.Dial1.StLouis1.Level3.net] has left #nouveau []
06:50 -!- etzel [n=thisnuke@west-193.rcn.NMT.EDU] has quit ["electrical storm"]
07:18 -!- etzel [n=thisnuke@west-193.rcn.nmt.edu] has joined #nouveau
07:31 -!- etzel [n=thisnuke@west-193.rcn.nmt.edu] has quit [Remote closed the connection]
07:41 -!- Myrizio [n=Myrizio@host211-98-static.104-80-b.business.telecomitalia.it] has quit ["Goodbye Ruby Tuesday (Perl Friday)"]
07:45 -!- etzel [n=thisnuke@west-193.rcn.nmt.edu] has joined #nouveau
08:36 -!- predatorfreak [n=predator@adsl-69-212-52-66.dsl.sfldmi.ameritech.net] has quit ["<blank>"]
10:27 -!- darktama [n=darktama@gentoo/contributor/darktama] has joined #nouveau
11:10 -!- predatorfreak [n=predator@adsl-69-212-52-66.dsl.sfldmi.ameritech.net] has joined #nouveau
11:58 -!- airlied [n=airlied@skynet.skynet.ie] has joined #nouveau
11:58 -!- airlied [n=airlied@skynet.skynet.ie] has left #nouveau []
11:58 -!- airlied [n=airlied@skynet.skynet.ie] has joined #nouveau
12:02 < darktama> airlied: welcome back! how was your holiday?
12:07 -!- KoalaBR [n=KoalaBR@port-83-236-13-117.dynamic.qsc.de] has joined #nouveau
12:07 < KoalaBR> Hello
12:09 < darktama> hey KoalaBR
12:11 < KoalaBR> darktama: Did you see my meager results?
12:11 < KoalaBR> I've got the suspicion,. that you somehow wanted more / something slightly different
12:12 < darktama> nope, that's what I wanted :) it's a shame we don't really have a good way of testing for other formats though
12:12 < KoalaBR> Ok, fine
12:13 < KoalaBR> Well, I have setup Xorg 7.1 but I can't seem to get nouveau working
12:13 < darktama> I played with FBO's a bit to see if I could get some more surface formats from it, but it seems limited with what formats it'll let you use
12:13 < airlied> darktama: good... much needed time away from PC, hopefully get back up to speed on development this week..
12:13 < darktama> nice :)
12:14 < KoalaBR> arlied: Your experience will be very helpfull :)
12:14 < darktama> KoalaBR: what problem are you having?
12:14 < airlied> much happen here while I was gone? (btw got the channel log URL?
12:14 < KoalaBR> Can't insert the kernel mods
12:14 < airlied> ah linked from wiki..
12:14 < KoalaBR> http://lumag.spb.ru/nouveau/irclogs/
12:14 < darktama> airlied: well, I sort-of got 3D going.. you still need to startx with nvidia first though :(
12:15 < darktama> KoalaBR: error messages?
12:15 < airlied> darktama: cool... 
12:15 < KoalaBR> Missing modules insmod -1 
12:15 < KoalaBR> darktama: I will try again need to setup my laptop so that I can tell you immediately what you need to know
12:16 < darktama> ok
12:22 -!- maxtoo [n=maxtoo@berryx.homedns.org] has joined #nouveau
12:25 -!- KoalaBR__ [n=chatzill@port-83-236-13-117.dynamic.qsc.de] has joined #nouveau
12:26 < KoalaBR__> Hi, there we are...
12:26 -!- KoalaBR [n=KoalaBR@port-83-236-13-117.dynamic.qsc.de] has quit [Remote closed the connection]
12:29 < KoalaBR__> darktama: When trying to insmod drm.ko (from git drm) I get "drm: exports duplicate symbol drm_get_resource_start
12:30 < KoalaBR__> nvidia module is rmmod'ded
12:30 < darktama> do you already have drm built into your kernel?
12:30 < KoalaBR__> Hm, maybe
12:30 < KoalaBR__> Sec
12:31 < KoalaBR__> If I ignore that problem and insert nouveau.ko I get:
12:31 < KoalaBR__> Unknown symbols: drm_calloc, drm_get_dev, drm _cleanup_pci and drm_rmma
12:33 < KoalaBR__> In Kernel: CONFIG_DRM=y, nothing else is set with DRM in the symbol name
12:33 < darktama> yup, turn that off
12:33 < KoalaBR__> Ok
12:35 < KoalaBR__> When I reboot,  will go offline a few minutes, as the laptop is routed through my my workstation
12:37 < KoalaBR__> Rebooting
12:40 -!- KoalaBR__ [n=chatzill@port-83-236-13-117.dynamic.qsc.de] has quit ["ChatZilla 0.9.61 [Mozilla rv:1.7.13/20060417]"]
12:41 -!- KoalaBR [n=KoalaBR_@port-83-236-13-117.dynamic.qsc.de] has joined #nouveau
12:42 < KoalaBR> drm works
12:43 < KoalaBR> nouveau too
12:43 < KoalaBR> Now for X
12:45 < KoalaBR> Works, but I noticed, I still load glx (from NVidia), suppress?
12:46 < darktama> I don't usually worry about it
12:46 < darktama> you could just "eselect opengl set xorg-x11" if it bothers you too much
12:46 < KoalaBR> darktama: Will write a doc about what to do later
12:46 < KoalaBR> No, I was wondering whether it would cause problems...
12:47 < darktama> nope, it shouldn't
12:47 < KoalaBR> So, what can I do to help (either testing or programmng)?
12:48 < KoalaBR> Please keep in mind: I'm a newbie in gfx drivers programming, so nothing too complicated for now, will need to ramp up difficulty :)
12:49 < KoalaBR> (No newbie otherwise)
12:49 < darktama> hmm well, I usually find something that I don't like (or a feature that I want) and work on that
12:50 -!- shenki [n=shenki@ppp110-122.lns4.adl4.internode.on.net] has joined #nouveau
12:51 < KoalaBR> Good idea, but currently I can't seem to find any problems with the driver, 2D using EXA (accoring do Xorg.log) is working without a hitch
12:53 -!- EdB [n=EdB@ARennes-251-1-76-162.w86-195.abo.wanadoo.fr] has joined #nouveau
13:04 -!- K [n=hazel@84-16-252-194.internetserviceteam.com] has joined #nouveau
13:12 < KoalaBR> Well my DDC maximum is 0 Mhz :modeclock <value>Mhz exceeds SSC maximum of 0Mhz
13:12 < KoalaBR> darktam: Where does this message originates from (X11, Drm, nouveau...?)
13:12 < KoalaBR> exceeds DDC not SSC
13:14 < darktama> the message comes from X I believe, but nouveau could be telling X the wrong info
13:14 < KoalaBR> Ok, will try to find this out
13:16 -!- Duke` [n=gnu@ANantes-251-1-129-130.w86-210.abo.wanadoo.fr] has joined #nouveau
13:33 -!- Unavowed [n=silent@13.t6.ds.pwr.wroc.pl] has joined #nouveau
13:41 < KoalaBR> darktama: If I want to print out some values to stdout, is printf() sufficient (xf86nouveau)?
13:42 < darktama> fprintf(stderr, ...) works fine from within the ddx, so I don't see why printf wouldn't
13:42 < darktama> if you want to be sure, xf86DrvMsg(0, X_ERROR, ...) will work too
13:43 < KoalaBR> Ok, thanks :)
13:52 -!- shenki_ [n=shenki@ppp110-122.lns4.adl4.internode.on.net] has joined #nouveau
13:53 -!- shenki [n=shenki@ppp110-122.lns4.adl4.internode.on.net] has quit [Read error: 104 (Connection reset by peer)]
13:57 < KoalaBR> Will be back later
14:00 -!- elsewhereunknown [n=elsewher@82-168-177-167.dsl.ip.tiscali.nl] has joined #nouveau
14:06 < elsewhereunknown> i know glxgears isn't a benchmark, but does anyone have a 6600gt@1280x1024 number
14:07 < elsewhereunknown> (my own computer runs gentoo and gives me 8000 fps, this computer has a 6500 and only gets 600 fps)
14:08 < elsewhereunknown> sorry, wrong channel :-)
14:10 < Duke`> 600 seems low...
14:11 -!- shenki_ [n=shenki@ppp110-122.lns4.adl4.internode.on.net] has quit [Read error: 54 (Connection reset by peer)]
14:11 -!- shenki [n=shenki@ppp110-122.lns4.adl4.internode.on.net] has joined #nouveau
14:13 < Duke`> 915 fps with my GeForce2 MX... are you running glxgears in fullscreen or original window size?
14:14 < EdB> elsewhereunknown, i think off a bad linking to gl.so
14:15 < elsewhereunknown> original size
14:16 < elsewhereunknown> the problem also is that i get 100% cpu usage on one of the cores
14:16 < elsewhereunknown> of the cpu
14:16 < EdB> elsewhereunknown, software rendering
14:16 -!- predatorfreak [n=predator@adsl-69-212-52-66.dsl.sfldmi.ameritech.net] has quit ["<blank>"]
14:17 < Duke`> even in direct rendering it take 100% CPU usage
14:17 < Duke`> *takes
14:17 -!- KoalaBR [n=KoalaBR_@port-83-236-13-117.dynamic.qsc.de] has quit [Read error: 110 (Connection timed out)]
14:17 < elsewhereunknown> if this were a p3 or something, it wouldn't be strange
14:18 < elsewhereunknown> but this a brand new cpu
14:18 < elsewhereunknown> even for software rendering 600 fps is low
14:18 < EdB> got 200 on my 2800xp with soft
14:19 < Duke`> I noticed on my pentium-M that sometimes the CPU freq is still 600 Mhz instead of 1,5 Ghz and so I have low results at glxgears
14:19 < Duke`> (in direct rendering)
14:19 < elsewhereunknown> very strange, i can get 900-1000 fps with software rendering
14:19 < Duke`> and with direct rendering you have only 600?
14:19 < elsewhereunknown> on my system
14:20 < elsewhereunknown> this is my father's system
14:20 < elsewhereunknown> brb
14:20 -!- elsewhereunknown [n=elsewher@82-168-177-167.dsl.ip.tiscali.nl] has quit [Remote closed the connection]
14:22 -!- elsewhereunknown [n=elsewher@82-168-177-167.dsl.ip.tiscali.nl] has joined #nouveau
14:26 -!- pmdata [i=patrice@ANantes-154-1-21-164.w81-53.abo.wanadoo.fr] has joined #nouveau
14:27 -!- elsewhereunknown [n=elsewher@82-168-177-167.dsl.ip.tiscali.nl] has quit [Remote closed the connection]
14:39 -!- elsewhereunknown [n=elsewher@82-168-177-167.dsl.ip.tiscali.nl] has joined #nouveau
14:45 -!- cptn [n=jw@217-162-119-116.dclient.hispeed.ch] has quit ["leaving"]
14:54 -!- stringfellow [n=stringfe@a80-100-96-220.adsl.xs4all.nl] has joined #nouveau
15:10 -!- cetrox_ [i=cetrox@250-246-222-201.adsl.terra.cl] has joined #nouveau
15:19 -!- elsewhereunknown [n=elsewher@82-168-177-167.dsl.ip.tiscali.nl] has quit ["Bye"]
15:25 -!- elsewhereunknown [n=elsewher@82-168-177-167.dsl.ip.tiscali.nl] has joined #nouveau
15:26 -!- cetrox [i=cetrox@57-246-222-201.adsl.terra.cl] has quit [Read error: 110 (Connection timed out)]
15:28 -!- elsewhereunknown [n=elsewher@82-168-177-167.dsl.ip.tiscali.nl] has quit [Client Quit]
15:31 -!- elsewhereunknown [n=elsewher@82-168-177-167.dsl.ip.tiscali.nl] has joined #nouveau
15:59 -!- K is now known as Ko
15:59 -!- Ko is now known as KO
16:12 -!- pmdata [i=patrice@ANantes-154-1-21-164.w81-53.abo.wanadoo.fr] has quit ["using sirc version 2.211+KSIRC/1.3.11"]
17:07 -!- elsewhereunknown [n=elsewher@82-168-177-167.dsl.ip.tiscali.nl] has quit ["Bye"]
17:19 -!- elsewhereunknown [n=elsewher@82-168-177-167.dsl.ip.tiscali.nl] has joined #nouveau
18:00 -!- Unavowed [n=silent@13.t6.ds.pwr.wroc.pl] has quit ["leaving"]
18:01 -!- shenki [n=shenki@ppp110-122.lns4.adl4.internode.on.net] has quit [Read error: 54 (Connection reset by peer)]
18:02 -!- shenki [n=shenki@ppp110-122.lns4.adl4.internode.on.net] has joined #nouveau
18:26 -!- shenki [n=shenki@ppp110-122.lns4.adl4.internode.on.net] has quit ["Leaving"]
18:28 -!- Myrizio [n=Myrizio@host149-101-static.104-80-b.business.telecomitalia.it] has joined #nouveau
18:38 -!- elsewhereunknown [n=elsewher@82-168-177-167.dsl.ip.tiscali.nl] has quit ["Bye"]
18:39 -!- elsewhereunknown [n=elsewher@82-168-177-167.dsl.ip.tiscali.nl] has joined #nouveau
19:13 -!- cetrox_ is now known as cetrox
19:29 -!- marteus [n=marteus@0x503fbab3.albnxx8.adsl-dhcp.tele.dk] has joined #nouveau
19:47 -!- jkolb_ [n=jkolb@wsi-128-136.wsi.com] has joined #nouveau
20:06 -!- alien_ [n=alien@c164229.adsl.hansenet.de] has joined #nouveau
20:08 -!- shenki [n=shenki@ppp110-122.lns4.adl4.internode.on.net] has joined #nouveau
20:13 -!- b33fc0d3 is now known as teddrivesbigtruc
20:14 -!- stringfellow [n=stringfe@a80-100-96-220.adsl.xs4all.nl] has quit [Remote closed the connection]
20:15 -!- teddrivesbigtruc is now known as b33fc0d3
20:45 -!- elsewhereunknown [n=elsewher@82-168-177-167.dsl.ip.tiscali.nl] has quit ["Bye"]
20:49 -!- phh [n=phh@moi.phhusson.info] has joined #nouveau
20:49 -!- _Demo_ [n=Demo@modemcable038.138-201-24.mc.videotron.ca] has joined #nouveau
20:54 -!- pmdata [n=patrice@ANantes-154-1-111-75.w86-214.abo.wanadoo.fr] has joined #nouveau
21:08 < pmdata> just added test for NV_fence extension
21:08 < pmdata> seems it uses what we called *_COUNTER in nv_objects.h
21:08 -!- cptn [n=jw@217-162-119-116.dclient.hispeed.ch] has joined #nouveau
21:09 < darktama> yeah, my guess is that it's a wait for some var to == the _COUNTER val
21:10 < pmdata> you mean the gpu has an internal counter on each object, and the _COUNTER value is a wait for this internal counter to reach that said value?
21:12 < darktama> no, something would have to set the value.  or perhaps _COUNTER is setting the value, and something else waits.. I don't know
21:13 < pmdata> NV_fence doc says Nvidia use it also some fences in their driver, so this is what all these _COUNTER are in the dumps
21:13 < darktama> yup, perhaps it's used to stall the FIFO until a DMA is complete?
21:13 < darktama> that would explain it during texture uploads
21:22 < pmdata> should we rename _COUNTER to _FENCE then?
21:22 < darktama> hm, it definitely has *something* to do with fences.. so yeah probably.. I'd like to know what it actually does though
21:26 < pmdata> I guess: "wait till internal counter for this object reaches _FENCE value"
21:26 -!- Myrizio [n=Myrizio@host149-101-static.104-80-b.business.telecomitalia.it] has quit ["Goodbye Ruby Tuesday (Perl Friday)"]
21:28 < darktama> yeah, I guess so
21:28 < darktama> btw.. I just noticed something strange in my FIFO dump.. one sec, I'll pastebin it
21:30 < darktama> http://sh.nu/p/3126
21:31 < darktama> notice the 0xbee4/0x1000/0x1000.. well, the 0x1000 is always the pitch of the rendering surface..
21:31 < darktama> *but* renouveau is seeing it as a method, not a parameter to a method
21:33 -!- alien_ [n=alien@c164229.adsl.hansenet.de] has quit ["Leaving"]
21:34 < darktama> sorry, I know that it's nothing to do with fences.. but, I noticed it while grepping for 0xbeef7201 :)
21:40 < jkolb_> might be because the 0's were skipped and that throws it off?
21:42 < darktama> hm, perhaps.. but it's odd that there's a command of size 0 just before that too
21:48 < jkolb_> hm
21:50 < darktama> ah damn, the mmtrace version of renouveau doesn't seem to work on amd64
21:50 < darktama> lumag_offline: any ideas?
21:50 < darktama> darktama@araqiel ~/cvs/mmtrace $ valgrind --tool=mmtrace glxgears
21:50 < darktama> valgrind: failed to start tool 'mmtrace' for platform 'amd64-linux': No such file or directory
21:50 < darktama> oops
21:50 < darktama> darktama@araqiel ~/cvs/mmtrace $ valgrind --tool=mmtrace glxgears
21:50 < darktama> valgrind: mmap(0x400000, 16384) failed in UME with error 22.
21:58 < darktama> and no, renouveau didn't misinterpret the FIFO.. a raw dump of it shows the same thing
22:00 -!- KO [n=hazel@84-16-252-194.internetserviceteam.com] has quit [Remote closed the connection]
22:02 < pmdata> 0xbee4 is obviously a different thing
22:02 < darktama> yes, it does seem that way.. doesn't make a lot of sense though
22:04 < darktama> that's why I wanted to use mmtrace, to see if that part of the FIFO was overwritten for some reason after the command was executed..
22:05 < darktama> I might do an x86 install of some distro to check it
22:10 -!- K [n=hazel@84-16-252-194.internetserviceteam.com] has joined #nouveau
22:14 < jkolb_> did someone want dumps of a quadro?
22:15 < pmdata> stereo context?
22:15 < jkolb_> i thought someone did...
22:15 < pmdata> hum what quadro do you have?
22:15 < jkolb_> fx 4500
22:16 < pmdata> can you make dumps with a stereo context? if you can't, we also lack standard quadro dumps
22:17 < jkolb_> how do i do that? just run renouveua?
22:17 < jkolb_> i don't see a test_stereo
22:17 -!- stringfellow [n=stringfe@ip56503c9f.direct-adsl.nl] has joined #nouveau
22:23 < jkolb_> pmdata: i have a pile of dumps
22:23 < jkolb_> what do i do with them?
22:24 < pmdata> in renouveau main.c there is a commented line to create a stereo context (need 2 output video devices)
22:24 < pmdata> if you have standard dumps, post a link to an tar archive there
22:25 < jkolb_> 2 output video devices? i have one monitor. i have the daughterboard that goes along with the card in there. that's hooked up to a tiny broadcast monitor thingy
22:26 < pmdata> stereo = 2 output devices (either glasses, or 2 monitors)
22:26 < pmdata> it's also need to be setup in xorg.conf/xf86config
22:26 < jkolb_> ok. i don't have that
22:27 < pmdata> ok, then go for current dumps
22:27 < jkolb_> ok i have those
22:31 < pmdata> what is your board pci id?
22:33 < jkolb_> 009d?
22:33 < jkolb_> on the output tests i get 10de-009d in the filenames
22:33 < pmdata> ok
22:33 < jkolb_> Quadro FX 4500
22:34 < pmdata> yeah, g70gl, nice
22:35 < jkolb_> it's for work. i have to deal with some of the graphics to video stuff
22:36 < jkolb_> anyway, i have the dumps if you want them
22:36 < pmdata> yep please
22:39 < jkolb_> where?
22:39 < pmdata> can you post a url to them?
22:41 -!- stillunknown [n=stillunk@82-168-177-167.dsl.ip.tiscali.nl] has quit ["Words get written, words get twisted, old meanings change in the drift of time. (sung by Ian Anderson)"]
22:41 < pmdata> or do you have cvs nouveau access?
22:42 < jkolb_> pmdata: i actually do have access but i'm on my work machine
22:42 < jkolb_> i'll post a url to a tarball with them
22:42 < pmdata> yep, I will grab it and put them on nouveau.sf.net/tests
22:43 < jkolb_> pmdata: people.brandeis.edu/~jkolb/nouveau/nv.tar.gz
22:45 -!- stillunknown [n=stillunk@82-168-177-167.dsl.ip.tiscali.nl] has joined #nouveau
22:47 < pmdata> jkolb> thanks, dumps added
22:48 < pmdata> http://nouveau.sourceforge.net/tests/g70gl/card_10de-009d_test_default.txt
22:48 < pmdata> you have same 0xbee4 darktama talked about earlier
22:49 < darktama> pmdata: yup, the nv3x dumps have the same thing
22:59 < jkolb_> pmdata: np
23:04 < jkolb_> i don't know how different the card would be to others we have. there aren't that many differences in a quadro i don't think
23:05 < pmdata> most differences are in the driver that use more hw features
23:05 < pmdata> so if there are differences between g70 and g70gl in dumps, bingo
23:06 < darktama> I'd love some nv40gl dumps
23:07 < darktama> but, since g70 is more or less the same as nv40 (and nv30 for that matter).. the g70gl stuff might apply to nv40 too
23:07 -!- maxtoo [n=maxtoo@berryx.homedns.org] has quit ["Les choses que l'on possède, finissent par nous posséder"]
23:08 -!- phh [n=phh@moi.phhusson.info] has quit ["Quitte"]
23:11 < jkolb_> it's really cool how the cards are so similar. it should make writing drivers a lot easier
23:12 < darktama> hm, vertices are submitted differently on quadros
23:12 < marcheu> jkolb_: ah, it's cool that you have a g70gl
23:12 < marcheu> darktama: yeah, it's always the case
23:12 < pmdata> nvidia always have some nice docs about quadro cards that list the new stuff that is hw accelerated
23:12 < pmdata> you should check for the g70
23:13 < darktama> it's weird, it uses the VTX_ATTR stuff that sets default vtx attributes on "normal" nv30/40/g70
23:13 < jkolb_> marcheu: yeah. i need it for work. have to deal with the nv ctrl extension and stuff
23:13 < marcheu> ah :) we stopped buying quadros and use bare geforces here :)
23:13  * darktama makes a note to try the quadro method in his exa code
23:14 < marcheu> darktama: won't make a difference with code that isn't vertex-limited
23:14 < darktama> yeah, true - but I want to see if I can make it work :)
23:15 < marcheu> yes, that'll work
23:16 < darktama> any other quadro features that'd be useful to us?
23:16 < marcheu> we have to read nvidia's PR
23:16 < marcheu> and see what they use there which they don't on geforces
23:17 -!- cetrox [i=cetrox@250-246-222-201.adsl.terra.cl] has quit [Remote closed the connection]
23:17 -!- cetrox [i=cetrox@250-246-222-201.adsl.terra.cl] has joined #nouveau
23:18 -!- stillunknown [n=stillunk@82-168-177-167.dsl.ip.tiscali.nl] has quit ["Words get written, words get twisted, old meanings change in the drift of time. (sung by Ian Anderson)"]
23:21 < pmdata> we could make a page on wiki that convert nvidia buzzword to hw feature
23:23 < darktama> marcheu: do you have any thoughts on the weirdness I saw in the fifo dumps?
23:24 < pmdata> http://download.nvidia.com/ndemand/Quadro_extranet/Product_Overview/PO_QFX_APR06.pdf
23:26 -!- egore [n=egore@dslb-084-063-034-077.pools.arcor-ip.net] has joined #nouveau
23:27 -!- KoalaBR_ [n=KoalaBR@port-83-236-12-219.dynamic.qsc.de] has joined #nouveau
23:27 < KoalaBR_> Hi
23:29 < KoalaBR_> marcheu: I have nouveau running and want to help with testing and programming. But I need some hints to get into gear:
23:30 < pmdata> so we have hw accelerated AA lines, overlay planes, 2 sided lighting, cliping planes, pixel readback <- this last one is interesting
23:30 < darktama> hmm, yes it is
23:30 < KoalaBR_> Either some links to texts or some very simple programming tasks, as I'm a newbie to driver programming
23:30 < pmdata> quad buffered stereo
23:30 < darktama> pixel readback from where?  blenders, or in a fragment program
23:30 < pmdata> maybe vram -> system ram?
23:31 < darktama> ah, is that what it meant.. damn, got me excited for nothing :)
23:31 < KoalaBR_> BTW: I have added a small "Howto compile / Setup nouveau" to the homepage
23:31 < darktama> if we could read the framebuffer from a fragprog.. that would be awesome
23:32 < pmdata> well, we must compare this doc against the standard g70 one
23:32 < darktama> and, it's already done in the blenders.. that's the whole point of them.. so ignore that part please :)
23:33 < pmdata> http://www.nvidia.com/object/IO_30459.html
23:33  * darktama hopes he pointed vmware at the right disk..
23:35 < pmdata> ohw, heavy buzzword usage detected, core dumped
23:44 -!- svu [n=svu@83.167.239.56] has quit ["Ex-Chat"]
23:51 < marcheu> darktama: weirdness ?
23:51 -!- nael [n=nael@ras75-1-81-57-62-96.fbx.proxad.net] has joined #nouveau
23:51 < darktama> the 0-sized commands, with surface pitches mixed in among the nops
23:52 < darktama> http://sh.nu/p/3126
23:52 < marcheu> we had those before, it looks like an exception
23:52 < marcheu> pmdata: pixel readback ?
23:52 < marcheu> pmdata: it's already accelerated on geforces, do the point at a particular difference ?
23:52 < darktama> so you think the 0xbee4 at the start is the command, just in a different format than usual?
23:52 -!- svu [n=svu@83.167.239.56] has joined #nouveau
23:53 < marcheu> darktama: that's how I used to understand it, yes
23:53 < darktama> so how does the GPU distinguish between a "normal" subc=5, size=0, method=0x1ee4?
23:53 -!- svu [n=svu@83.167.239.56] has quit [Client Quit]
23:53 < marcheu> pmdata: hmm, now that you mention it, I have to papers from nvidia about quadros
23:54 < marcheu> darktama: no idea, but it's not nv40-specific
23:54 < darktama> yup, I know.. it's in nv30/g70 too
23:54 < darktama> perhaps the size=0 is the difference
23:54 < marcheu> size=0 usually binds an object..
23:54 < darktama> no, that's method=0 that does that
23:54 < darktama> size should be one in that case
23:54 < marcheu> hmm, right
23:57 < marcheu> yeah, the equivalent on nv20 is bee0
23:59 < marcheu> 0x0000bee0 that is
23:59 < pmdata> something like 0xbeef, with only 12 bits?
23:59 < marcheu> pmdata: yeah, the common part is "ee"
--- Log closed Пнд Сен 11 00:00:24 2006
