#asahi-dev · 2024-05-09 10:23
leioyeah, that's missing all the cool new features to make spotting trouble easier; but looks like you already spotted something anyways then#asahi-dev · 2024-05-09 10:22
chadmedi made it on sysprof 44 which is packaged in ::gentoo#asahi-dev · 2024-05-09 10:21
leioanyhow, more on topic I wish I had a newer sysprof to look at the capture :)#asahi-dev · 2024-05-09 10:21
leiohttps://bpa.st/WDNA#asahi-dev · 2024-05-09 10:19
chadmedof course#asahi-dev · 2024-05-09 10:19
leioI hope said thing actually forces IPv6#asahi-dev · 2024-05-09 10:19
chadmedso dunno what route your isp is taking#asahi-dev · 2024-05-09 10:19
chadmedusing telstra's looking glass i get a clean trace from their frankfurt pop to the aussiebb segment im on#asahi-dev · 2024-05-09 10:11
leiobut with IPv6 I gather something goes bad between singapore and AU#asahi-dev · 2024-05-09 10:11
leioI could just -4 it for the file#asahi-dev · 2024-05-09 10:11
chadmedi guarantee my website is not worth your time :p#asahi-dev · 2024-05-09 10:10
leioit's literally the only address I've ever had problems with, I'm not going to go bother at that level#asahi-dev · 2024-05-09 10:09
PaulFertserleio: guess you need to talk to your ISP about that, hope they're sane and can figure it out and ping the right people.#asahi-dev · 2024-05-09 10:09
leionow once it showed GW-AussieBroadband.retn.net too, in red with mtr#asahi-dev · 2024-05-09 10:08
leiolocal traceroute goes to question marks and bad after rt.eqx.sin.sg.retn.net#asahi-dev · 2024-05-09 10:08
PaulFertserMy traceroutes for the reference https://paste.debian.net/1316420/#asahi-dev · 2024-05-09 10:06
leiono idea then, but here that curl command has been trying for 1m30 now#asahi-dev · 2024-05-09 10:05
PaulFertserBut ICMPv6 pings are indeed ignored.#asahi-dev · 2024-05-09 10:04
PaulFertserOne can suspect HE<->Cogent peering dispute but in this case apparently there're alternative routes so it works from HE tunnel too.#asahi-dev · 2024-05-09 10:04
jannau`curl -6 -v -LO https://chadmed.au/res/kwin.syscap` works for me, startup seems a little slow but it finishes in a reasonable amount of time overall#asahi-dev · 2024-05-09 10:03
PaulFertserTested from two different locations, one native one HE.net, both work.#asahi-dev · 2024-05-09 10:03
leioI'm not sure I can trust this online ipv6 traceroute (didn't have anything installed here yet locally), but it shows it's stuck after 2403:5800:1701:3::2#asahi-dev · 2024-05-09 10:00
chadmedhmm thats correct and it works from my mobile over 5G...#asahi-dev · 2024-05-09 10:00
leioit does not ACK pings, 2403:580a:80ed:0:cb9d:978f:8d50:57fe#asahi-dev · 2024-05-09 09:57
chadmedit should ACK pings#asahi-dev · 2024-05-09 09:57
chadmedwhat ipv6 address do you get#asahi-dev · 2024-05-09 09:56
leiochadmed: that link connects forever over IPv6 but works over IPv4 btw#asahi-dev · 2024-05-09 08:55
chadmedhttps://chadmed.au/res/kwin.syscap#asahi-dev · 2024-05-09 08:54
chadmedif im reading this sysprof capture correctly, the stutters coincide with bunched up calls to the gpu driver#asahi-dev · 2024-05-08 20:42
nicolas17damn spicy headers#asahi-dev · 2024-05-08 20:41
jannaunicolas17: https://mastodon.social/@bshanks/112401605018159567#asahi-dev · 2024-05-08 20:40
nicolas17jannau: got a source? o.o#asahi-dev · 2024-05-08 19:58
j`eyyay for standard hw#asahi-dev · 2024-05-08 19:56
jannau"enhanced, next-generation ML accelerators" seems to be indeed SME/SVE2#asahi-dev · 2024-05-08 05:54
jannauof course they will, early point releases usually still have dcp firmware API changes#asahi-dev · 2024-05-08 05:50
nicolas17yeah so a matter of dealing with macOS 15 changing the firmware interface because of course they will :P#asahi-dev · 2024-05-08 05:46
jannaudisplay changes are hopefully mostly abstracted away by the firmware interface. the changes could be necessary because each panel is driven separately#asahi-dev · 2024-05-07 22:49
nicolas17M4 is T8132#asahi-dev · 2024-05-07 18:33
sawyerCould potentially mean calibrated timings/power levels per-unit which would make it a bigger headache--apple doesn't seem to like putting that stuff in "hidden" storage. All still speculation at the moment, though#asahi-dev · 2024-05-07 18:28
flokliI'm getting flashbacks to eink world#asahi-dev · 2024-05-07 18:28
flokliMhhh, it's probably still per-display, right?#asahi-dev · 2024-05-07 18:26
sawyerflokli: I have a feeling they added some hardware acceleration for doing brightness compensation for the VRR oled panels, since if you pulse an image less often you need to maintain the same power/duty cycle ratio--seems like they may have "proper" VRR support on those panels instead of the "you have 5 display modes the panel can switch between with predefined timings" that phones have had for a while#asahi-dev · 2024-05-07 18:18
floklimmmh#asahi-dev · 2024-05-07 18:13
nicolas17unfortunately no firmware to look at yet#asahi-dev · 2024-05-07 18:08
floklicurious if this is just marketing blabla or there's actually some significant change#asahi-dev · 2024-05-07 18:07
flokli> It also features an entirely new display engine to drive the stunning precision, color, and brightness of the breakthrough Ultra Retina XDR display on iPad Pro.#asahi-dev · 2024-05-07 15:15
j`eyjannau: that would be nice#asahi-dev · 2024-05-07 14:46
jannauI hope "enhanced, next-generation ML accelerators" for the CPU cores is SME/SVE2#asahi-dev · 2024-05-07 14:46
jannaucpu cores, neural engine and dcp might be updated#asahi-dev · 2024-05-07 14:46
svenpossibly just the cpu cores from the last iPhone I guess