00:12
popolon has quit [Quit: WeeChat 3.0]
00:17
icecream95 has quit [Ping timeout: 260 seconds]
00:18
icecream95 has joined #panfrost
00:22
<
kinkinkijkin >
im getting an issue finding vgem_dri.so ive just noticed
00:24
<
alyssa >
"n e v e r do that" Guilty as charged.
00:25
<
alyssa >
...and have had to spin up recovery media at the eleventh hour
00:25
<
alyssa >
or is it the 13th?
00:25
<
macc24 >
alyssa: it's 1am
00:26
<
alyssa >
...So it is. Good riddance, go to sleep! :p
00:26
<
macc24 >
my brain told me that i can't sleepp so i can't sleep
00:26
<
alyssa >
Aaaalright. Go be a good capitalist and do some online shopping for Boxing Day? :P
00:26
<
macc24 >
"Boxing Day"?
00:27
<
icecream95 >
alyssa: load_sample_id generates incorrect code on Bifrost
00:27
<
kinkinkijkin >
ew capitalism, money is only to be used for tinder and buying obscure tech
00:28
<
macc24 >
imagine having money
00:28
<
kinkinkijkin >
wait i just remembered tinder is also an application
00:28
<
kinkinkijkin >
do not use money for tinder the dating app
00:28
<
alyssa >
icecream95: hm?
00:31
<
icecream95 >
alyssa: Looks like it's correct after all, ignore
00:32
<
alyssa >
icecream95: your discard patch has my tested-by btw
00:32
* macc24
learned that tested-by is a thing
00:34
<
icecream95 >
alyssa: What were you testing with?
00:41
_whitelogger has joined #panfrost
00:43
stikonas has quit [Remote host closed the connection]
00:44
<
icecream95 >
I've almost finished with shader images, so I won't be able to avoid atomics for too much longer…
00:45
<
alyssa >
icecream95: Unfortunately you'll have to hold off for a little while on atomics :(
00:45
<
alyssa >
There's some seriously gnarly scheduling requirements that the current compiler can't handle.
00:45
stikonas has joined #panfrost
00:45
<
alyssa >
New scheduler will be able to (and is specifically designed to handle the constraints here) but there's not really a way to hack around it.
00:46
stikonas has quit [Client Quit]
00:47
stikonas has joined #panfrost
01:23
<
alyssa >
Behaviour is changed for 1, 3, 5, 7, 9, which don't need to be padded.
01:23
<
alyssa >
(does 9 need to be padded? Maybe, can't remember that one)
01:23
<
icecream95 >
alyssa: Note the `if (idx < 10) return idx`
01:25
<
alyssa >
-----Yeah, you're right, my bad
01:25
<
alyssa >
Why am I rereading patches on a Friday night? Sorry :p
01:27
<
HdkR >
Friday night is the most productive time of day
01:30
icecream95 has quit [Ping timeout: 272 seconds]
01:31
icecream95 has joined #panfrost
01:32
<
kinkinkijkin >
it's also a majour holiday i think
01:33
<
icecream95 >
... as opposed to a minor holiday, an augmented holiday or an add-9 sus-4 holiday
01:50
<
HdkR >
Augmented reality holiday
02:00
raster has quit [Quit: Gettin' stinky!]
02:02
stikonas has quit [Remote host closed the connection]
02:03
vstehle has quit [Ping timeout: 272 seconds]
02:59
_whitelogger has joined #panfrost
03:17
<
kinkinkijkin >
i still need to figure out whether this is my bug or not
03:17
<
kinkinkijkin >
back to reading docs
03:32
<
kinkinkijkin >
the error is on my end
03:56
_whitelogger has joined #panfrost
04:16
<
kinkinkijkin >
the error on my end is fixed
04:17
<
kinkinkijkin >
i was drawing elements but passing the number of verts, not the number of faces
04:17
<
kinkinkijkin >
very annoying error
04:19
<
kinkinkijkin >
just a word off somewhere in piles of code and i could render
*either* a triangle,
*or* a teapot, by changing a magic number that had no business being there
04:19
<
kinkinkijkin >
fix that and i can render both
04:19
<
HdkR >
Rendering index by number of teapots
04:21
<
kinkinkijkin >
currently my test gives two teapots (can be any obj really) and a triangle in different places, colours, and sizes
05:38
mixfix41 has joined #panfrost
06:00
vstehle has joined #panfrost
06:54
davidlt has joined #panfrost
07:17
yann has quit [Ping timeout: 268 seconds]
07:53
dstzd has joined #panfrost
09:19
mixfix41 has quit [Quit: Lost terminal]
09:29
_whitelogger has joined #panfrost
10:09
icecream95 has quit [Ping timeout: 246 seconds]
10:13
mixfix41 has joined #panfrost
10:23
unoccupied has joined #panfrost
12:00
stikonas has joined #panfrost
12:07
raster has joined #panfrost
12:11
solidhal has quit [Quit: Ping timeout (120 seconds)]
12:23
warpme_ has joined #panfrost
12:26
<
raster >
br_: oooh. a bvit slow though for just a plain wm coming u0 but.... wokring. you have gl.. but is that swrast?
12:28
<
macc24 >
raster: panfrost on xorg has been slow for me too
12:30
<
raster >
well cursors change so xorg is alive a while earlier while things are coming up still. maybe just slow i/o - cruddy micro-sd?
12:30
<
br_ >
raster: I have lot of printfs that slows down things...
12:30
<
raster >
that won't help
12:31
<
br_ >
also hacked mesa to wait a result of each job submitted (sync wait)
12:32
<
br_ >
need to add mutex locking everywhere in my panfrost driver
12:32
<
raster >
that shouldnt affect anything until your gl samples com up
12:32
<
raster >
unless you use glamor?
12:32
<
br_ >
not swrast, my video buffers are not mmaped to userspace at all
12:33
<
raster >
aaaah ok. that explains a bit then.. :)
12:34
<
br_ >
will continue work on that next year.. only spent 6 weeks so far
12:36
<
raster >
take a break.... a whole new year of funtimes is coming :)
13:01
alpernebbi has joined #panfrost
14:32
archetech has joined #panfrost
15:15
zkrx has quit [Quit: I'm done]
15:23
zkrx has joined #panfrost
15:57
rak-zero has quit [Ping timeout: 268 seconds]
16:00
rak-zero has joined #panfrost
17:36
raster has quit [Quit: Gettin' stinky!]
17:58
warpme_ has quit []
17:58
warpme_ has joined #panfrost
18:01
alpernebbi has quit [Quit: alpernebbi]
18:04
sphalerite has quit [Quit: 'tis the season to be rebooting]
18:06
sphalerite has joined #panfrost
18:29
archetech has quit [Quit: Konversation terminated!]
19:39
stikonas_ has joined #panfrost
19:40
stikonas has quit [Read error: Connection reset by peer]
20:03
davidlt has quit [Ping timeout: 240 seconds]
20:40
stikonas_ has quit [Remote host closed the connection]
20:43
stikonas has joined #panfrost
20:44
raster has joined #panfrost
20:51
ezequielg has quit [Ping timeout: 260 seconds]
20:53
ezequielg has joined #panfrost
21:22
<
macc24 >
battery on duet seems to be held with magic pull tab things
21:25
<
macc24 >
if it weren't for the fact that i can't drill hole in chassis i would have drilled a hole in chassis to get easy access to uart
21:29
<
macc24 >
duet is surprisingly easy to disassemble for a tablet
21:30
<
macc24 >
and there's nothing interesting on it aside from what's already documented
21:56
raster has quit [Quit: Gettin' stinky!]
22:31
<
macc24 >
oh it's finally not out of stock
22:32
<
tchebb >
Digi-Key also stocks them and is sometimes in stock when SparkFun isn't, FYI
22:48
raster has joined #panfrost