--- Log opened Срд Авг 16 00:00:15 2006
00:00 < lumag> I'll commit test in couple of minutes
00:00 < marcheu> ok, I'll look at it then
00:01 < marcheu> I've been using texture_shader before (although not the OFFSET functionality) and it usually works fine
00:02 -!- shenki [n=shenki@ppp144-143.lns3.adl2.internode.on.net] has quit [Read error: 110 (Connection timed out)]
00:02 < lumag> yes, the FIFO it pretty logical for all other ops.
00:03 < lumag> do you want working test or not working one ? :))
00:04 < marcheu> all ? :)
00:04 < lumag> oh...
00:04 < marcheu> what's the difference between them ,
00:04 < marcheu> ?
00:05 < marcheu> anyway, food time, bbiab
00:06 < KoalaBR> Lumag: Thanks alot, will add this tomorrow
00:07 < lumag> KoalaBR: thank you.
00:07 < KoalaBR> Well...I only try to add some docs to the project, thanks for helping me out
00:10 -!- pmdata [i=patrice@ANantes-154-1-62-230.w81-53.abo.wanadoo.fr] has joined #nouveau
00:10 < pmdata> the user list is growing :)
00:10 < lumag> commited a small test. I think it should work, but instead I get op1=NONE :(
00:21 -!- johane_ [n=johan@84-217-5-143.tn.glocalnet.net] has joined #nouveau
00:25 -!- stringfellow [n=stringfe@ip56503c9f.direct-adsl.nl] has quit [Remote closed the connection]
00:29 -!- johane [n=johan@84-217-5-143.tn.glocalnet.net] has quit [Read error: 60 (Operation timed out)]
00:33 -!- johane__ [n=johan@84-217-5-143.tn.glocalnet.net] has joined #nouveau
00:35 < KoalaBR> So, added the Scissor data and as I have rescued enough kittens for today. I'll report back tomorrow, good night :)
00:35 -!- KoalaBR [n=KoalaBR@port-83-236-15-60.dynamic.qsc.de] has quit ["ChatZilla 0.9.61 [Mozilla rv:1.7.13/20060417]"]
00:39 < pmdata> lumag> I tried array-ize some matrix, but could not get it to work
00:39 < pmdata> besides the A_(name) and name__SIZE, is there anything else needed?
00:41 -!- tibbs|h is now known as tibbs
00:42 -!- johane_ [n=johan@84-217-5-143.tn.glocalnet.net] has quit [Read error: 60 (Operation timed out)]
00:49 -!- atcl [n=Universe@L8064.l.pppool.de] has quit [Read error: 104 (Connection reset by peer)]
00:51 < gabrielg> keep up the good work!
00:51 -!- gabrielg [n=gabriel@c-a6df72d5.011-178-73746f44.cust.bredbandsbolaget.se] has quit ["zzz"]
00:52 -!- maxtoo [n=maxtoo@berryx.homedns.org] has quit [Remote closed the connection]
00:55 -!- EdB [n=EdB@ARennes-251-1-83-55.w86-199.abo.wanadoo.fr] has quit ["Konversation terminated!"]
01:08 < lumag> pmdata: the name(i) should be the offset of i-th element of array. If it's a single matrix, the NV_..._MATRIX(n) = (start + n<<2) should do the trick.
01:09 < lumag> If it's an array of matrices, A_() won't work :(
01:10 < pmdata> I should try writing a PRINT_ARRAY(4,4,FP32) then?
01:15 < lumag> no, there should be M_(name) = name, 16, 4, name :)
01:15 < lumag> which would describe exactly one matrix
01:21 -!- You're now known as lumag_offline
01:22 < lumag_offline> night!
01:32 -!- johane__ [n=johan@84-217-5-143.tn.glocalnet.net] has quit ["Leaving"]
01:54 < pmdata> marcheu> I would need test_single_vertex dumps for nv15 and nv28 in softquadro mode
01:55 < pmdata> on non quadro, it always uses vertex array to send vertex position
01:55 < pmdata> (or nv11 or nv18, I don't mind instead of nv15)
01:58 -!- pmdata [i=patrice@ANantes-154-1-62-230.w81-53.abo.wanadoo.fr] has quit ["using sirc version 2.211+KSIRC/1.3.11"]
02:07 -!- swany [n=swany@81-234-181-143-o1108.tbon.telia.com] has quit []
02:23 < Duke`> hey I've found worse than ATI's proprietary driver
02:24 < Duke`> Thomson's firmwares... :/
02:39 < marcheu> hmm, pmdata's right I promised him I'd do that long ago
02:39 < marcheu> Duke`: what's up with those ?
02:40 < Duke`> paranormal activity when editing your records on DVD recorders
02:40 < Duke`> really really weird
02:42 < Duke`> you hide a chapter, then reshow it, an you lose a portion of your movie... you must divide it in two titles (where you want) to recover the whole movie o_O
02:56 -!- Duke` [n=gnu@ANantes-251-1-100-79.w86-203.abo.wanadoo.fr] has quit ["Fatal signal: Segmentation Fault"]
03:35 -!- ag [i=ag@caladan.roxor.cx] has quit ["BRB"]
03:35 -!- ag [i=ag@caladan.roxor.cx] has joined #nouveau
03:49 < marcheu> pmdata: ok, I'm running the nv28 in quadro mode now
03:52 < marcheu> darktama: btw I couldn't test the memory management stuff since my libdri.so seems broken
03:52 < marcheu> so maybe there are bugs... :)
04:08 -!- hiyuh [n=hiyuh@ZL050248.ppp.dion.ne.jp] has joined #nouveau
04:24 < airlied> marcheu: btw which siss/via memory manager you base things on? the drm git tree has a common cleaned version that does proper locking etc..
05:02 -!- hiyuh [n=hiyuh@ZL050248.ppp.dion.ne.jp] has quit ["Leaving"]
05:06 -!- hiyuh [n=hiyuh@ZL050248.ppp.dion.ne.jp] has joined #nouveau
05:08 -!- Kotori [n=Aexoden@207-118-81-252.dyn.centurytel.net] has joined #nouveau
05:10 -!- Shipon [n=Aexoden@207-118-75-166.dyn.centurytel.net] has quit [Read error: 60 (Operation timed out)]
05:17 -!- Aexoden2 [n=Aexoden@p5-66-112-127-99.rb3.ka.centurytel.net] has joined #nouveau
05:17 -!- Aexoden2 is now known as Aexoden
05:20 -!- Kotori [n=Aexoden@207-118-81-252.dyn.centurytel.net] has quit [Read error: 60 (Operation timed out)]
08:11 < darktama> marcheu: ok.  I'll attempt to modify the ddx to alloc the cursor and FB soon (hopefully).. I had a quick look and can't figure out how to make EXA alloc offscreen memory through the ddx however...
08:12 < darktama> s/through the ddx/through the drm/
08:34 -!- Myrizio [n=Myrizio@host107-98.pool80104.interbusiness.it] has quit ["Goodbye Ruby Tuesday (Perl Friday)"]
08:35 -!- tibbs is now known as tibbs|h
09:56 < airlied> darktama: EXA doesn't know drm ...
09:56 < airlied> not sure how to solve that problem yet...
09:57 < darktama> would an optional hook for OffscreenAlloc be feasible?
09:57 < airlied> darktama: I think that would be the way to do it.. but I'm not very EXA knowledgeable..
09:58 < airlied> might be worth asking anholt or MrCooper on dri-devel..
09:58 < darktama> hm, good idea :)
10:55 -!- EdB [n=EdB@ARennes-251-1-36-69.w81-250.abo.wanadoo.fr] has joined #nouveau
11:31 -!- cptn [n=jw@217-162-119-116.dclient.hispeed.ch] has joined #nouveau
11:59 -!- Duke` [n=gnu@ANantes-251-1-101-163.w86-203.abo.wanadoo.fr] has joined #nouveau
12:10 -!- Unavowed [n=silent@host81-151-27-73.range81-151.btcentralplus.com] has joined #nouveau
12:17 -!- maxtoo [n=maxtoo@berryx.homedns.org] has joined #nouveau
13:11 -!- stillunknown [n=madman20@82-168-177-167.dsl.ip.tiscali.nl] has quit [Remote closed the connection]
14:03 < marcheu> airlied: I used the radeon code, which I modified heavily
14:03 < marcheu> darktama: we can't do that, we have to use the allocator for everything at once, or things will start to overlap 
14:17 -!- Aexoden [n=Aexoden@p5-66-112-127-99.rb3.ka.centurytel.net] has quit [Read error: 60 (Operation timed out)]
14:22 -!- K [i=hazel@85-57-132-205.gij1.adsl.uni2.es] has joined #nouveau
16:33 -!- stillunknown [n=madman20@82-168-177-167.dsl.ip.tiscali.nl] has joined #nouveau
16:35 -!- Aexoden [n=Aexoden@p5-66-112-127-205.rb3.ka.centurytel.net] has joined #nouveau
17:09 < darktama> marcheu: yup, I realised that soon after I mentioned it :)
17:30 -!- EdB [n=EdB@ARennes-251-1-36-69.w81-250.abo.wanadoo.fr] has quit [Remote closed the connection]
17:32 -!- EdB [n=EdB@ARennes-251-1-36-69.w81-250.abo.wanadoo.fr] has joined #nouveau
17:51 -!- johane [n=johan@84-217-5-143.tn.glocalnet.net] has joined #nouveau
18:19 -!- Myrizio [n=Myrizio@host205-98.pool80104.interbusiness.it] has joined #nouveau
18:51 -!- hiyuh [n=hiyuh@ZL050248.ppp.dion.ne.jp] has quit ["Leaving"]
19:32 -!- EdB [n=EdB@ARennes-251-1-36-69.w81-250.abo.wanadoo.fr] has quit ["Konversation terminated!"]
19:43 -!- darktama_ [n=darktama@124-168-232-15.dyn.iinet.net.au] has joined #nouveau
19:47 -!- darktama [n=darktama@gentoo/contributor/darktama] has quit [Read error: 60 (Operation timed out)]
19:48 -!- stringfellow [n=stringfe@ip56503c9f.direct-adsl.nl] has joined #nouveau
19:54 -!- swany [n=swany@81-234-181-143-o1108.tbon.telia.com] has joined #nouveau
19:56 -!- Unavowed [n=silent@host81-151-27-73.range81-151.btcentralplus.com] has quit ["leaving"]
20:17 -!- johane_ [n=johan@84-217-5-143.tn.glocalnet.net] has joined #nouveau
20:26 -!- johane [n=johan@84-217-5-143.tn.glocalnet.net] has quit [Read error: 60 (Operation timed out)]
20:46 -!- darktama [n=darktama@gentoo/contributor/darktama] has joined #nouveau
20:49 -!- darktama_ [n=darktama@124-168-232-15.dyn.iinet.net.au] has quit [Read error: 60 (Operation timed out)]
20:50 -!- pmdata [i=patrice@ANantes-154-1-71-121.w86-195.abo.wanadoo.fr] has joined #nouveau
20:59 -!- etzel [n=thisnuke@69-160-140-26.ontrca.adelphia.net] has quit [Read error: 104 (Connection reset by peer)]
21:12 -!- etzel [n=thisnuke@69-160-140-26.ontrca.adelphia.net] has joined #nouveau
21:19 < pmdata> hello
21:19 < pmdata> anyone with nv1x quadro?
21:19 -!- Lumag [n=Lumag@chimpanzee.school.ioffe.ru] has joined #nouveau
21:20 < Lumag> hi all
21:23 < pmdata> hi lumag
21:24 < marcheu> hi, no I've not nv11 ATM
21:24 < marcheu> hi, no I've got no nv11 ATM
21:25 < pmdata> can you put it in softquadro mode? and dump single_vertex from cvs?
21:25 < marcheu> no, because they're not in any computer ATM
21:25 < pmdata> arf, you have it or not? :)
21:25 < pmdata> ok
21:26 < marcheu> it's really weird though, I have 4 nvidia cards that can be softquadroed without issues, and every other people I talked to can't get it to work
21:27 -!- Myrizio [n=Myrizio@host205-98.pool80104.interbusiness.it] has quit [Read error: 104 (Connection reset by peer)]
21:28 < pmdata> could you list the rev of your cards? my nv15 is rev. A4
21:29 < marcheu> I'd have to plug it to know..;
21:33 < pmdata> I thought that not only having a win32 version of renouveau would be nice, but we could maybe also test direct3d features not present in opengl
21:34 < marcheu> what features do you have in mind ?
21:35 < pmdata> I don't know, I never read d3d stuff, just would be nice to compare
21:35 < pmdata> maybe some unknown commands in tcl are related to d3d
21:38 < pmdata> would be fun if win32 users could help :)
21:38 < marcheu> I don't think anyone here has windows, and thus could do a port
21:38 < marcheu> but...
21:38  * marcheu looks at the PPC owners in here...
21:38 < marcheu> yeah, I know who you are
21:45 < Duke`> my father have a Windows licence, but it often crashes
21:46 < Duke`> and its 5600XT seems to be missing some hardware pieces :P
21:47 < Lumag> :)
21:55 -!- cptn [n=jw@217-162-119-116.dclient.hispeed.ch] has quit ["leaving"]
22:05 -!- KoalaBR [n=KoalaBR@port-83-236-14-10.dynamic.qsc.de] has joined #nouveau
22:07 < KoalaBR> Hi all
22:12 < Duke`> hi
22:12 < Lumag> hi
22:13 < KoalaBR> Lumag: Added your suggestions to the doc
22:13 < Lumag> :)
22:14 < marcheu> KoalaBR: btw what's your plan with that doc ? is moving to the website a plan ?
22:15 < KoalaBR> marcheu: It depends on you all: If you like it / find it useful, I can add the annotation to the sourcecode (Beware large!) and commit it along with the doxygen config file and some notes in the readme on how to use it
22:16 < KoalaBR> or I can just add it to the Wiki, if I can figure out how to upload premade pages
22:16 < KoalaBR> Whatever you like
22:17 < marcheu> yeah, I don't think we should embed the documentation in the source code
22:17 < marcheu> it should really be online
22:18 < KoalaBR> Ok, I will maintain it as external patch set (If you want it, just ask) and add it to the Wiki
22:18 < KoalaBR> I mean I add the page to the Wiki 
22:19 < KoalaBR> I'll do it on Saturday or Sunday, ok?
22:20 < KoalaBR> marcheu: Don't know, whether you noticed: There is a link to my personal Webspace on the Wikifrontpage
22:20 < marcheu> KoalaBR: yup, but we should rather make it the official renouveau doc, since it's more complete
22:20 < marcheu> instead of the README
22:21 < marcheu> maybe replace the README contents with an URL to it
22:22 < KoalaBR> Ok, I will add the most current version aduring the weekend, add the patch as tarfile (so it doesn't get lost) and maintain it as long there is a need
22:23 < KoalaBR> s/aduring/during/
22:33 < KoalaBR> marcheu: Regarding the NV_blend_square: Am I right that this is nothing more than glBlendFunc() with 4 new constants? Or did you have something different in mind?
22:33 < marcheu> KoalaBR: yup, that's it. you asked for something easy, didn't you ?
22:34 < KoalaBR> Well, ok :=)
22:34 < marcheu> feel free to do something else, any progress is progress :)
22:36 < KoalaBR> Well, I always ask you, because I'm not sure which functions are still needed. I'm not so confident whether the feature matrix is up to date (and even then, I wouldn't know for which functions to look)
22:38 < pmdata> the feature matrix should better match the nv objects
22:39 < KoalaBR> Yea, I have the suspicion that we are much further along than the matrix suggests :)
22:40 < pmdata> most of tcl for nv10 is done
22:40 < pmdata> marcheu> do you think the view frustum clip planes is somewhere, or is it hardcoded in gpu?
22:42 < marcheu> pmdata: did you play with glFrustum ?
22:44 < pmdata> hum, no
22:48  * pmdata reading *_point_* extension doc
22:50 < pmdata> strange that nv_point_sprite is only hw accelerated on nv2x, according to the doc
22:50 < pmdata> time to write a test for nv28 then
22:55 -!- johane__ [n=johan@84-217-5-143.tn.glocalnet.net] has joined #nouveau
22:56 < KoalaBR> BTW: I slightly added SDL_NOFRAME to the SDL_SetVideoMode as the window decorations may have an influence on the  offsets of some commands. You may set the Window position via  export SDL_VIDEO_WINDOW_POS=x,y as you like
23:03 -!- johane_ [n=johan@84-217-5-143.tn.glocalnet.net] has quit [Read error: 60 (Operation timed out)]
23:05 < pmdata> real programmers don't use a window manager
23:05 < marcheu> real programmers work in console mode anyway, because they broke their Xorg
23:06 < pmdata> real programmers outsource in india
23:06 < Lumag> no real programmers whistle on 2400 bod, because all other input methods are broken :)
23:06 < KoalaBR> pmdata: Sorry, I know I'm a complete failure :)
23:07 < KoalaBR> but as real programmers don't document their code and I do exactly that....
23:07 < KoalaBR> ;)
23:07 < pmdata> the source IS the doc
23:09 < KoalaBR> Yes master, teach me the ways :)
23:11 < KoalaBR> marcheu: Only 2 defines for gl_data were missing, now it works ok and doesn't say "UNKNOWN" anymore
23:12 < pmdata> koala> renouveau is just a tool, so we can do crappy code in it, to force the driver to show us commands :)
23:15 < KoalaBR> pmdata: I know. My code looks nearly the same, if I'm experimenting with a concept....
23:15 < KoalaBR> or try to find something out
23:19 -!- atcl [n=Universe@L822b.l.pppool.de] has joined #nouveau
23:23 -!- Myrizio [n=Myrizio@host139-96.pool80104.interbusiness.it] has joined #nouveau
23:27 < KoalaBR> marcheu: Now what's next?
23:29 -!- cptn [n=jw@217-162-119-116.dclient.hispeed.ch] has joined #nouveau
23:29 < pmdata> woot, nv2x[0xa1c] seems to be point sprite parameter
23:29 < pmdata> time to commit test
23:31 < KoalaBR> Shall I try out something?
23:33 < pmdata> what is your hw?
23:37 < KoalaBR> NV43
23:43 < pmdata> then checkout cvs, and dump nv_point_sprite and single_vertex, hope to find some stuff
23:43 < KoalaBR> Ok
23:46 < KoalaBR> Point_sprite: http://sh.nu/p/2745
23:48 < KoalaBR> http://sh.nu/p/2746 for single vertex
23:49 < KoalaBR> marcheu: Just post, what you need next. I'll pick it up through the logs
23:49 < marcheu> hmmm
23:49 < marcheu> I have to say I'm not sure what we need more
23:49 < KoalaBR> Oh back :)
23:49 < marcheu> plus I'm not the one who does most of the RE now :)
23:49 < marcheu> I suppose pmdata knows better than I do
23:50 < KoalaBR> Ok, I'll pester him :)
23:50 < KoalaBR> So how is your work going?
23:50 < marcheu> well, I commited the memory manager
23:50 < marcheu> next step is actually using it
23:50 < marcheu> but I have to fix my Xorg first... my libdri is broken for some reason
23:51 < marcheu> or maybe it's the drm that's broken... have to try all that anyway
23:51 < KoalaBR> If you need help (regarding the source, not X11), just ask.
23:52 < marcheu> KoalaBR: btw you could put your sumpds on sf.net, you have access there
23:52 < KoalaBR> pmdata: What else do we need for a NV40 card ? Any test to do or any new tests to write?
23:52 < pmdata> we need more b33r
23:52 < marcheu> KoalaBR: through ssh it's at shell.sf.net:/home/groups/n/no/nouveau/htdocs/tests/
23:52 < marcheu> pmdata: we always need more b33r
23:53 < KoalaBR> Ok, will do so marcheu
23:53 < marcheu> pmdata: what strip was that again
23:53 < marcheu> I can't remamber the name
23:54 < pmdata> megatokyo
23:54 < marcheu> ah yes
23:54 < pmdata> I still read it
23:55 < KoalaBR> pmdata: Just post what tests are still needed for the NV40 and which one is most important currently. Will read the logs and start working on that.
23:56 < marcheu> KoalaBR: one idea is to look at the extension string from glxinfo, implement a small test and see what unknown packets come out of it
23:56 < marcheu> s/a small test/a small test for one of the extensions/
23:57 < KoalaBR> Ok. will try that
23:57 < pmdata> nv40[0x1ee8] seems to be the point sprite parameter
23:58 < pmdata> I added it in renouveau for nv20, it has same format
23:58 < pmdata> if someone wants to add it for nv30 tcl
23:59 < KoalaBR> But only tomorrow. I need to sleep - good night
23:59 -!- KoalaBR [n=KoalaBR@port-83-236-14-10.dynamic.qsc.de] has quit ["ChatZilla 0.9.61 [Mozilla rv:1.7.13/20060417]"]
--- Log closed Чтв Авг 17 00:00:15 2006
