00:00
cphealy has joined #panfrost
00:20
WoC has quit [Remote host closed the connection]
00:27
WoC has joined #panfrost
00:33
Green has quit [Quit: ...]
00:35
Green has joined #panfrost
01:11
WoC has quit [Remote host closed the connection]
01:11
WoC has joined #panfrost
01:17
Green has quit [Quit: ...]
01:19
Green has joined #panfrost
01:27
raster has quit [Quit: Gettin' stinky!]
01:40
sphalerite has quit [Ping timeout: 258 seconds]
01:55
Green has quit [Quit: ...]
01:56
Green has joined #panfrost
02:02
kaspter has joined #panfrost
02:03
vstehle has quit [Ping timeout: 246 seconds]
02:11
WoC has quit [Ping timeout: 260 seconds]
02:27
stikonas has quit [Ping timeout: 260 seconds]
02:30
Green has quit [Quit: ...]
02:31
Green has joined #panfrost
02:32
<
macc24 >
devfreq broke
02:33
Green has quit [Client Quit]
02:34
Green has joined #panfrost
02:56
camus has joined #panfrost
02:56
kaspter has quit [Ping timeout: 260 seconds]
02:56
camus is now known as kaspter
03:52
archetech has joined #panfrost
04:28
Green has quit [Quit: ...]
04:31
Green has joined #panfrost
04:31
warpme_ has quit [Quit: Connection closed for inactivity]
04:40
Green has quit [Quit: ...]
04:41
Green has joined #panfrost
04:59
Green has quit [Quit: ...]
05:04
Green has joined #panfrost
05:05
felipealmeida has quit [Ping timeout: 268 seconds]
05:05
felipealmeida has joined #panfrost
05:09
Green has quit [Quit: ...]
05:10
Green has joined #panfrost
06:00
vstehle has joined #panfrost
06:14
davidlt has joined #panfrost
07:08
kaspter has quit [Ping timeout: 256 seconds]
07:08
kaspter has joined #panfrost
07:35
kaspter has quit [Ping timeout: 240 seconds]
07:36
camus has joined #panfrost
07:38
camus is now known as kaspter
07:46
yawniek has quit [Ping timeout: 272 seconds]
07:50
kaspter has quit [Ping timeout: 264 seconds]
07:50
kaspter has joined #panfrost
08:09
raster has joined #panfrost
08:39
yawniek has joined #panfrost
09:11
icecream95 has quit [Ping timeout: 240 seconds]
10:08
yann has joined #panfrost
10:10
yawniek has quit [Quit: WeeChat 2.9]
10:15
yann has quit [Ping timeout: 260 seconds]
10:37
kaspter has quit [Remote host closed the connection]
10:37
kaspter has joined #panfrost
10:40
sphalerite has joined #panfrost
11:01
yawniek has joined #panfrost
11:09
chewitt has quit [Ping timeout: 256 seconds]
11:26
camus has joined #panfrost
11:28
kaspter has quit [Ping timeout: 272 seconds]
11:28
camus is now known as kaspter
11:30
chewitt has joined #panfrost
11:36
raster has quit [Quit: Gettin' stinky!]
11:37
Green has quit [Read error: Connection reset by peer]
12:18
stikonas has joined #panfrost
12:22
raster has joined #panfrost
12:47
warpme_ has joined #panfrost
13:42
gcl has quit [Ping timeout: 240 seconds]
13:43
kaspter has quit [Quit: kaspter]
13:44
gcl has joined #panfrost
13:59
gcl_ has joined #panfrost
14:02
gcl has quit [Ping timeout: 246 seconds]
14:15
robmur01 has joined #panfrost
14:18
* robmur01
chuckles at HdkR "I completely expect Nvidia to kill the ARM display IP"
14:19
<
robmur01 >
Arm already killed Arm's display IP
14:19
<
robmur01 >
or from another perspective, Arm's customers did. By not buying it :P
14:23
<
HdkR >
The truth is crueler than my imagination
14:23
<
macc24 >
at least they triedf
14:32
<
cphealy >
NXP uses the ARM display IP in the LS1028A (Oddly, it's one of the few IP cores in the SoC design that has zero detail in the ~7000 page reference manual...)
14:33
<
alyssa >
Fascinating
14:35
<
cphealy >
The Vivante GPU in the LS1028A has at least 20 pages dedicated to it. The ARM display IP = "This information is under NDA. Contact local NXP sales representative for more details." ;-)
14:44
<
macc24 >
now that i used linux on kevin for some time
14:44
<
macc24 >
i never realized that the bar was so low for usability xD
14:50
yann has joined #panfrost
15:03
<
robmur01 >
cphealy: yeah, there doesn't appear to be a Non-Confidential version of the DP500 TRM (or indeed any other DPU TRMs) - TBH it's probably mostly that there was just never any justification for the effort of releasing one
15:05
<
robmur01 >
from an end-user (i.e. not SoC integrator) perspective there's probably not much in there that you couldn't figure out from the Linux driver anyway; I don't think they're massively complex beasts
15:10
yann has quit [Ping timeout: 260 seconds]
15:30
<
alyssa >
I have to write unit tests.
15:36
warpme_ has quit [Quit: Connection closed for inactivity]
15:36
<
kinkinkijkin >
pcsxr (not rearmed) works fine full-speed on duet with panfrost
15:37
<
kinkinkijkin >
despite being interpreter only
15:39
<
macc24 >
i wonder if ps2 emulation would work
15:40
<
kinkinkijkin >
probably yes, but also probably very slow
15:43
<
robmur01 >
but then you could use the PS2 emulator to run Linux :D
15:51
archetech has joined #panfrost
15:59
<
macc24 >
kinkinkijkin: i don'
15:59
<
macc24 >
i don't think that second computer is needed to recover from bad cadmium update
16:08
<
alyssa >
DEVLEOPERS
16:09
<
macc24 >
alyssa: ???
16:11
WoC has joined #panfrost
16:13
<
alyssa >
Developers developers developers
16:14
* robmur01
nervously glances around for throwable chairs...
16:17
<
macc24 >
hypothesis: duet has no usb on pogo pins on bottom
16:19
<
macc24 >
kinkinkijkin: what's the output of 'lsusb -v | grep iConfiguration' on your duet?
17:04
yann has joined #panfrost
17:06
<
alyssa >
nothing like bifrost's famous 78-bit instructions :<
17:09
yann has quit [Ping timeout: 260 seconds]
17:20
<
WoC >
anyone who knows if there is a OpenCL that can be used in combination with panfrost on mali-t860 ?
17:30
<
macc24 >
hypothesis 2: duet keyboard is i2c device, and pads on bottom of board are connected to pins on bottom of device
17:42
<
alyssa >
macc24: means we have to implement int128 operations to handle packing :V
17:42
<
macc24 >
alyssa: oh my god
17:43
<
alyssa >
macc24: oh! and instead of having things be 16-bit aligned....
17:44
<
alyssa >
everything is 15-bit aligned
17:44
<
alyssa >
yes. really
17:44
<
macc24 >
this is beyond cursed
17:44
<
alyssa >
(* except for an 8-bit shift at the start because of course)
17:44
<
alyssa >
128-bit quadword = 8-bit tag + 7 * 15-bit subwords
17:45
<
alyssa >
so a 78-bit instructions fits snugly in 5 15-subwords
17:45
<
alyssa >
with 3-bits thrown in some other field where they can fit it
17:49
<
macc24 >
i'm happy that i don't have to deal with this ;D
18:39
robmur01 has quit [Quit: Leaving]
18:42
* alyssa
is down a debug hellhole
19:10
<
macc24 >
i sure do love solving screens of merge conflicts
19:14
yann has joined #panfrost
19:19
yann has quit [Ping timeout: 256 seconds]
19:20
<
alyssa >
So I burned enough time trying to get
*good* test cases and...
19:21
<
alyssa >
TL;DR whipped up a Python script to hackily parse disassemblies from clauses in the blob and generate unit tests :p
19:32
<
alyssa >
Lyude: Guess who's working on the packing code ^^
19:44
<
alyssa >
Code invoked subtle undefined behaviour causing "impossible test failures"
19:44
<
alyssa >
Since gcc's constant folding differed from arm64's implementation. ughh
19:45
tomboy64 has quit [Ping timeout: 240 seconds]
19:46
davidlt has quit [Ping timeout: 240 seconds]
19:52
warpme_ has joined #panfrost
19:57
tomboy64 has joined #panfrost
20:00
WoC has quit [Remote host closed the connection]
20:03
stikonas has quit [Remote host closed the connection]
20:04
stikonas has joined #panfrost
20:13
WoC has joined #panfrost
20:18
archetech has quit [Quit: Konversation terminated!]
20:42
<
Lyude >
alyssa: cool!
20:43
<
alyssa >
Lyude: I have it
_almost_ working
20:43
<
alyssa >
A minute ago i'd've said I had it
20:43
<
alyssa >
added another test case, and we're failing this one
20:46
<
macc24 >
i may have working 5.11 kernel for duet
20:46
<
alyssa >
or not, looks like the test case is wrong
20:46
<
alyssa >
so I do have it working
21:24
dstzd has quit [Read error: Connection reset by peer]
21:24
dstzd_ has joined #panfrost
21:24
dstzd_ is now known as dstzd
21:25
raster has quit [Quit: Gettin' stinky!]
21:26
<
alyssa >
still need to deal with constants
21:27
<
alyssa >
including a genuinely -awful -- mchanism that never got r/e'd
21:28
lvrp16 has quit [Ping timeout: 260 seconds]
21:31
lvrp16 has joined #panfrost
21:36
icecream95 has joined #panfrost
21:41
<
icecream95 >
So
*of course* the faults don't happen when I set PAN_MESA_DEBUG=trace...
21:42
yann has joined #panfrost
21:53
<
alyssa >
anything I can help with?
21:56
* alyssa
has exhausted her ability to do scheduling stuff today
21:57
yann has quit [Ping timeout: 256 seconds]
22:24
icecream95 has quit [Ping timeout: 264 seconds]
22:29
icecream95 has joined #panfrost
22:34
<
icecream95 >
alyssa: Shouldn't the lane0 modifier of LOAD.i16 be for the instruction destination, not src[0] as it currently is?
22:38
<
alyssa >
Yes, it should... that's a really annoying edge case :|
22:39
popolon has joined #panfrost
22:39
<
alyssa >
Same goes for LOAD.i8...
22:40
<
alyssa >
(and LOAD.i32 but it's not clear how widening 32 to 64 makes any sense)
22:44
<
alyssa >
Bifrost is just full of surprises.
23:07
<
macc24 >
icecream95: got usb-c ccd thing?
23:07
<
icecream95 >
macc24: no
23:36
* macc24
got pinout of duet's keyboard
23:36
<
macc24 >
it's usb after all
23:46
<
icecream95 >
macc24: What's the fifth pin for, then?
23:49
<
icecream95 >
Ahh, I always wondered why the keyboard sometimes acted drunk :P
23:49
<
macc24 >
my guess is "id":
23:50
<
macc24 >
from left to right: vbus, d+, gnd, d-, d- and pint
23:54
<
macc24 >
it still boots after reassembling
23:54
* macc24
lets out sigh of relief
23:56
<
macc24 >
and it looks like there is no mt8183 uart on testpads