RAOF | Sarvatt: It's past FF; we'll just ship r300c and not do r300g at all. | 00:22 |
---|---|---|
RAOF | Sarvatt: Also, git commit 6a154a5208613ec is incorrect - those symbols that got dropped from libOpenVG were public; we need to re-expose them (they got dropped in the mapi move). | 01:03 |
RAOF | Let the crazy mesa building begin! | 07:19 |
bryceh | :-) | 07:28 |
bryceh | I hope my reply to hugh's email wasn't too pushy | 07:29 |
RAOF | I don't recall it being pushy at all? | 07:30 |
bryceh | ok good, I just hadn't seen a reply | 07:30 |
bryceh | perhaps it's just that I'm thinking too far down the line for it to be relevant | 07:31 |
RAOF | I'm not exactly thinking hard about UDS at this point, no :) | 07:32 |
RAOF | How many times do we _really_ need to build mesa? :/ | 07:35 |
RAOF | Wait. Is mesa building i965_dri.so for OSMesa32? | 07:56 |
=== yofel_ is now known as yofel | ||
RAOF | I wish I wish I wish mesa didn't take forever to build :/ | 10:41 |
tseliot | RAOF: the new mesa? | 10:44 |
jcristau | every mesa ever. | 10:45 |
tseliot | yes, I know, I was just curious to know whether we were getting the new mesa or now | 10:45 |
tseliot | *not | 10:45 |
tseliot | it's still better than rebuilding a kernel... | 10:46 |
RAOF | tseliot: Yeah, the new mesa. | 11:14 |
RAOF | Oh, dear lord it still hasn't finished. | 11:14 |
tseliot | nice | 11:21 |
tseliot | we may get unity to work with radeon | 11:21 |
tseliot | (it did with an old snapshot of mesa) | 11:21 |
RAOF | Yeah, that's one of the motivations. | 11:22 |
tjaalton | sounds like you're not parallelizing the build | 11:22 |
RAOF | Yeah, I'm not. | 11:22 |
RAOF | I should work out why the last parallelised build didn't work properly. | 11:23 |
tjaalton | it made a difference? | 11:23 |
jcristau | isn't "because mesa's build system" good enough? :/ | 11:23 |
RAOF | To the build time? Dunno, I didn't count. To the build CPU usage? Yes. Did the result work properly? No. | 11:23 |
RAOF | It takes quite a while to lzma compress ~100Mb of binaries | 11:24 |
tjaalton | yep | 11:26 |
RAOF | Now, is egl working this timeā¦ | 11:26 |
chrisccoulson | RAOF, shouldn't you be asleep now? | 11:27 |
RAOF | Not at 8:30 pm | 11:27 |
chrisccoulson | oh, i thought you were further ahead than that ;) | 11:27 |
jcristau | only in the other half of the year | 11:27 |
RAOF | chrisccoulson: Shouldn't you still be in bed? It's surely still the morning for you :) | 11:27 |
RAOF | :P | 11:28 |
chrisccoulson | lol | 11:28 |
chrisccoulson | i actually got to bed early last night - 1am! | 11:28 |
RAOF | Sweet! | 11:28 |
tjaalton | but... friday :) | 11:28 |
RAOF | ARGH. Why is libEGL looking in /usr/local/lib/egl? | 11:29 |
Sarvatt | RAOF: hmm yeah, I'm not sure what's up with vgu symbols being dropped from libOpenVG | 14:22 |
Sarvatt | looking into it, I mean vgu.h and vgu.c are still there. according to khronos VGU is supposed to be shipped in another lib, libOpenVGU | 14:29 |
Dr_Jakob | Sarvatt: really? Can you point to where it says so? | 14:41 |
Sarvatt | one sec, trying to dig it up again, here's the nokia page though - http://library.forum.nokia.com/index.jsp?topic=/Nokia_Symbian3_Developers_Library/GUID-7EFBEEAD-3E74-5165-B305-313F7DE4BEB4.html | 14:43 |
Sarvatt | it doesn't say anything in openvg-1.1.pdf from khronos about separate libraries, hmm | 14:45 |
Dr_Jakob | hmm... | 14:45 |
Dr_Jakob | I can ask on the OpenVG working group. | 14:46 |
Sarvatt | ok the sample implementation on the khronos website is all in libOpenVG but apparently there are a bunch of things actually using it expecting the separate libOpenVGU that the proprietary implementations ship.. | 14:48 |
Dr_Jakob | Ugh... | 14:49 |
* Sarvatt kicks self for not using tests/quick.tests in piglit that time | 15:46 | |
Sarvatt | vish: regarding unity, can you try the mesa in this PPA? https://edge.launchpad.net/~sarvatt/+archive/mesa | 15:51 |
vish | ooh! yay! | 15:52 |
vish | Sarvatt: will try that in a bit :) | 15:52 |
Sarvatt | thanks :) | 15:52 |
vish | np.. | 15:53 |
Sarvatt | tormod says its fixed but I believe he's using r300g not r300c that we'll ship :( | 15:54 |
Sarvatt | RAOF: swrast is looking good - http://sarvatt.com/downloads/mesa-swrast/ | 16:09 |
penguin42 | anyone else seeing a ~30second boot pause just after loading the Radeon driver? | 16:16 |
jcristau | missing firmware? | 16:16 |
penguin42 | jcristau: Hmm - but X+3d seems to be working | 16:17 |
jcristau | k. | 16:17 |
penguin42 | jcristau: Actually, looking more closely - I get a burst of drm/radeon messages and it loads the microcode, does some other stuff and then the hang is after drm] Initialized radeon 2.5.0 20080528 for 0000:07:00.0 on minor 0 | 16:19 |
vish | Sarvatt: hi, nope dint help , all i get is a blank white screen. one change i see is that unity does not keep reloading , which might be due to changes in unity that mikkel mentioned | 16:46 |
vish | xsession-errore > http://paste.ubuntu.com/484542/ | 16:46 |
vish | errors* | 16:46 |
vish | Sarvatt: any other debugging you might need? | 16:47 |
ohmy | rebonsoit | 19:00 |
Sarvatt | heh nouveau is not a happy camper in mesa 7.9, dont think any piglit test has passed yet :) going to compare it to 7.8.2 after this is done | 19:16 |
=== simar__mohaar is now known as simar | ||
Sarvatt | oh darn, it just warned on all of these because of the unknown PIPE_CAP's on nouveau - http://sarvatt.com/downloads/piglit/nouveau/ | 19:52 |
Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!