#
…

2022-12-15

00:05Gaspare has quit [Quit: Gaspare]
00:11cylm_ has quit [Ping timeout: 480 seconds]
01:41derzahl has joined #asahi-dev
02:00derzahl has quit [Ping timeout: 480 seconds]
02:03riker77_ has joined #asahi-dev
02:05derzahl has joined #asahi-dev
02:08riker77 has quit [Ping timeout: 480 seconds]
02:08riker77_ is now known as riker77
02:37swaggie has quit [Remote host closed the connection]
02:40___nick___ has joined #asahi-dev
02:40___nick___ has quit []
02:42___nick___ has joined #asahi-dev
03:06c10l has quit [Read error: Connection reset by peer]
03:06c10l has joined #asahi-dev
03:08pjakobsson has joined #asahi-dev
03:11pjakobsson_ has quit [Ping timeout: 480 seconds]
03:17amarioguy has joined #asahi-dev
03:17amarioguy has quit [Remote host closed the connection]
03:18amarioguy has joined #asahi-dev
03:35derzahl has quit [Ping timeout: 480 seconds]
03:52user982492 has joined #asahi-dev
04:20swaggie has joined #asahi-dev
04:53unixgeek has quit [Remote host closed the connection]
05:09pjakobsson_ has joined #asahi-dev
05:10WindowPain_ is now known as WindowPain
05:12pjakobsson has quit [Ping timeout: 480 seconds]
05:41user982492 has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
06:26___nick___ has quit [Ping timeout: 480 seconds]
06:59swaggie has quit [Remote host closed the connection]
07:27c10l has quit [Quit: Ping timeout (120 seconds)]
07:39swaggie has joined #asahi-dev
07:42c10l has joined #asahi-dev
07:47swaggie has quit [Ping timeout: 480 seconds]
07:53swaggie has joined #asahi-dev
08:02marcanzzywysm: looks sane, and that reminds me I should send out the t600x bluetooth changes too (those are in the intersection of two things now already merged in)
08:03marcan(this is why I prefer to do DT changes via our tree, because then I don't have to wait for the upstream merge to get dependencies in and it avoids the merge conflicts)
08:19c10l has quit [Quit: Ping timeout (120 seconds)]
08:19svenI think Bluetooth had to go through the bt tree because I introduced that generic BT controller binding and they was already at least one follow up commit based on that
08:20sven*there was
08:28SSJ_GZ has joined #asahi-dev
08:34c10l has joined #asahi-dev
09:19gladiac has joined #asahi-dev
09:20c10l has quit [Ping timeout: 480 seconds]
09:24c10l has joined #asahi-dev
09:59swaggie has quit [Remote host closed the connection]
09:59swaggie has joined #asahi-dev
10:07swaggie has quit [Ping timeout: 480 seconds]
10:13MajorBiscuit has joined #asahi-dev
10:13c10l has quit [Ping timeout: 480 seconds]
10:22c10l has joined #asahi-dev
10:31swaggie has joined #asahi-dev
10:36karpouzi has joined #asahi-dev
10:39swaggie has quit [Ping timeout: 480 seconds]
10:40karpouzithe mac mini finally arrived
10:41karpouziis it still true that the latest DP/TB3 branch is atcphc-20221130?
10:42karpouzihmm looks like it
10:47pjakobsson has joined #asahi-dev
10:50pjakobsson_ has quit [Ping timeout: 480 seconds]
10:54svenif you want to work on something relatively independent: we'll probably need something like https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/tree/drivers/usb/typec/altmodes/displayport.c for thunderbolt.c
11:00MajorBiscuit has quit [Quit: WeeChat 3.6]
11:01MajorBiscuit has joined #asahi-dev
11:02r0ni has quit [Quit: Textual IRC Client: www.textualapp.com]
11:05swaggie has joined #asahi-dev
11:07c10l has quit [Ping timeout: 480 seconds]
11:08mps_ has joined #asahi-dev
11:08karpouziwill check it out, sven, thanks!
11:09svenfair warning, that subsystem is a bit strange
11:09svenbut then again, so is everything USB related
11:09maz(understatement of 2022)
11:09karpouziwarning appreciated
11:09mps has quit [Ping timeout: 480 seconds]
11:10svenmaz: I didn't want to scare karpouzi away before they even started ;)
11:10c10l has joined #asahi-dev
11:10svenkarpouzi: heikki usually replies pretty quickly to emails and feel free to ask me here as well. I've already spent quite some time banging my head against usb and typec
11:11karpouzity! i am still in the 'setting up' phase with this kit, will ping when ready
11:15bcrumb has joined #asahi-dev
11:16marcanmaz: re TSO, how much pushback do you think I'll get for trying to get it added to prctl? :p
11:18mazmarcan: gut feeling: at best, you'll get ignored. at worse, this will go down into a flame war around an effective fork of the whole userspace ABI... and all the variants in between.
11:19marcanuh... effective fork of the whole userspace ABI?
11:19mazthe main issue is that SW that fails without TSO and works with it will simply always enable it instead of debugging the stuff.
11:20marcanthis is intended for x86 emulators
11:20maz... and will be used by everyone because it will hide bugs.
11:20maz(see, we're getting there! ;-)
11:21marcanyeah okay, any better ideas besides crippling performance of x86 emulation? :p
11:21karpouziquick question... if you're working with different versions of the kernel, ie different branches of the same linux git repo, is the best way to save build time... to have duplicate linux repos? or is there another technique
11:21marcanI get your point but I don't see how you could possibly do both
11:21ChaosPrincessif process.name == 'fex': enable_tso() :P
11:21karpouzitso = timesharing option?
11:21svenkarpouzi: ccache + fast-ish VM for building works for me
11:22karpouziah i guess i don't technically have to build it on the m1 oops
11:22karpouziplenty of buildserver around, thanks for the reminder
11:22svenit's a good idea to build it on the M1 though because those machine make for a fast build machine ;)
11:22ChaosPrincessprobably argue that tso support is coming to new arm64 anyway, just that apple uses a non-standard switch
11:23mazChaosPrincess: you obviously have better information on the architecture than I do...
11:23mazthe direction of travel so far is the exact opposite.
11:25marcanmaz: so what you're saying is we should just enable it downstream and let the gamer voices eventually become loud enough to annoy upstream into finding a solution ;)
11:25mazmarcan: I wonder if we can somehow make the TSO stuff a VM property instead of a per-process thing.
11:25marcanhm, what's the distinction exactly?
11:26swaggie has quit [Ping timeout: 480 seconds]
11:26mazthe distinction is that Linux itself doesn't need to know. the whole VM is TSO, end of story.
11:26nullEnabling the IMP-DEF bit *within* the kernel for this goes against the very strong upstream policy that we do not do so, doesn't work in VMs and so on (because it'll trap to the hypervisor)
11:26nullLikewise, so does advertising it
11:26marcanoh like kvm VMs?
11:26karpouzisven: actually, the m1 wasn't bad - built in ~15min
11:27ChaosPrincessmaz: ok, i might have misremembered since i cant find it, but there is definitely that fujitsu arm that already has tso
11:27marcannull: IIRC you can actually control whether VMs trap to EL2 or not for this
11:27nullor any hypervisor, since when running at EL1 we don't know the hypervisor
11:27mazChaosPrincess: that's owing to its SPARC heritage. nobody forced them to build shit HW... ;-)
11:27nullmarcan: sure, I'm just trying to say that in general it doesn't work, and to make it work requires a lot more finesse than may be initially obvious, since you need to reliably inform both ends, etc
11:28ChaosPrincesshey, if the 3 sparc-lovers would have been itc, they would be very upset
11:28null... and the policy in general is simply "Do not expose IMP-DEF stuff to lower ELs"
11:28bcrumb has quit [Quit: WeeChat 3.7.1]
11:28null(and likewise, lower ELs should not assume the presence of IMP-DEF stuff unless explicilty advertised (e.g. in DT for PMUs and such)
11:29marcansure, but that seems like a somewhat orthogonal problem to support at EL2 for userspace purposes?
11:29nullWIth neted virt it's not ;)
11:29marcanOTOH there is an argument that x86 emu folks should use VMs anyway to get 4K pages, which could also involve a VM-global TSO switch
11:29marcanbut then with nested virt you have the same problem
11:30nullI'm very much not keen on relaxing the policy, because it implies a very long tail of consequences, and we have no guarantee tthat a vendor won't completely change the thing in arbitrary ways in a future release, so we can't really commit to an ABI
11:30null... and the architecture is clearly going in a different direction, with special instructions rather than a mode
11:32marcanI don't have a particularly big horse in this race, but if FEX ends up being like twice as fast with TSO mode the end result is just going to be us supporting it downstream and ending up with the resulting API fork, because it's too much benefit to just leave on the table
11:32marcanand even if M4 or whatever ends up doing it a more "standard" way, I'm not keen on leaving earlier users screwed
11:32nullI appreciate that argument, but that's "twice as fast, on specific hardware, using an IMP-DEF feature', and crap everywhere else
11:33marcanwell, this hardware is also twice as fast as the competition already :)
11:33marcandon't blame me for every other vendor failing to deliver
11:33marcanI'm not a fan of IMP-DEF nonsense but I see why Apple did what they did here
11:33nullI don't think it's leaving users screwed as such, and while I appreciate people want every ounce of performance they can get, I think this is setting ourselves up for a much larger set of problems going forwards
11:34nullNot blaming you, I just want ot be clear here so it's not a surprise on the list
11:34bcrumb has joined #asahi-dev
11:34marcanyeah, I'm just saying, if upstream doesn't want this to happen whatsoever, I don't see ourselves not doing it downstream if the benefits are significant
11:35marcanto be clear, my policy here isn't to implement all the Apple nonsense, kernel be damned
11:35marcanI have no intention of implementing AMX because that sounds like a giant pain, with more EL0 state and undocumented *instructions* and just no
11:35marcanbut if there's a bit I can flip to make FEX not suck the calculus is different
11:36marcanthe reality is we have a huge amount of hardware out there that could potentially benefit from this, and if that conflicts with kernel policy, that's kind of a problem
11:37marcan(mind you, this is all conditional on TSO actually mattering - it's entirely possible FEX could just use atomics everywhere and not lose too much, we'll find out once I put a knob for systemwide TSO in m1n1 so they can try both)
11:37marcan(if it's like 5% benefit for normal load/stores and TSO over normal mode and atomics, I don't care)
11:37nullI do appreciate that argument, but on the flip side there are a bunch of such things that over vendors do that could arguably be similar, and the calulus for upstream isn't just "enablnig this one thing for this one vendor's HW"
11:38nullI must really learn to type better
11:38maz(still using that funky split keyboard?)
11:38nullNah, I'm actually on an M1 macbook pro today; on the ergodox I'm a bit better
11:38marcanright, I think the issue here is how much benefit we get out of it
11:38marcanas I said, if it's 5%, whatever
11:39marcanbut if it's "you can actually play x86 games at good performance on fanless laptops with better graphics than on macOS with three graphics translation layers, but only if Linux supports this one thing" then...
11:39marcanobviously we're some time away from that point
11:39marcanbut we'll get there eventually
11:40marcanand so there might come a day, if things work out like that, where gaming on Macs ends up being conditional on Linux supporting that TSO bit
11:40cylm_ has joined #asahi-dev
11:40svendid anyone test how much slower everything gets if we just enable TSO globally?
11:41marcanmind you, yeah I was just about to say that
11:41marcanbecause the short-term solution is almost certainly going to be a m1n1 option to enable TSO globally, and a /chosen/asahi,tso flag so userspace can know it's available
11:41marcanand then we'll find out exactly what we can do with it
11:48bcrumb has quit [Read error: Connection reset by peer]
11:50c10l has quit [Quit: Ping timeout (120 seconds)]
11:51swaggie has joined #asahi-dev
12:01swaggie has quit [Ping timeout: 480 seconds]
12:05c10l has joined #asahi-dev
12:32bcrumb has joined #asahi-dev
12:32bcrumbi noticed someone posted an additional PR in asahi-scripts
12:44karpouzi_ has joined #asahi-dev
12:46Gaspare has joined #asahi-dev
12:47c10l has quit [Quit: Ping timeout (120 seconds)]
12:50karpouzi has quit [Ping timeout: 480 seconds]
12:54karpouzi_ is now known as karpouzi
12:54karpouzi has quit [Quit: leaving]
12:59karpouzi has joined #asahi-dev
12:59karpouzidid anyone hear an update about the RNDIS kerfuffle
13:00karpouzi /msg NickServ SENDPASS karpouzi
13:00karpouziuh whoops
13:06marcankarpouzi: that patch is not upstream, I think it was never merged
13:06karpouzi... weird!
13:07marcanI imagine some discussion happened offlist after the NACKs he got, or it just got stuck in limbo
13:09jluthra has quit [Remote host closed the connection]
13:09svensmells like some "unfixable" 0-day that either isn't unfixable or actually isn't a bug
13:09karpouzihe's not, like, off his rocker or something
13:09jluthra has joined #asahi-dev
13:09karpouzimakes sense to me.
13:11bcrumb has quit [Ping timeout: 480 seconds]
13:18Gaspare has quit [Quit: Gaspare]
13:36karpouzi has left #asahi-dev [#asahi-dev]
14:02iaguis has joined #asahi-dev
14:02gladiac has quit [Quit: k thx bye]
14:03gladiac has joined #asahi-dev
14:23gladiac has quit [Quit: k thx bye]
14:35kettenishey, how dare you to say bad words about SPARC when I'm not watching! ;)
14:54mazkettenis: I love SPARC. one of the first architectures I actually enjoyed playing with (well, sun4m -- I really didn't like the SW TLB on sun4c).
14:54mazkettenis: do you want to take this v490 from me? I also have a spare T1000 if you want!
14:55kettenisI certainly don't want the v490 and I have a t1k ;)
14:55mazcrap. I need a victim.
15:15MajorBiscuit has quit [Read error: Connection reset by peer]
15:34swaggie has joined #asahi-dev
15:42swaggie has quit [Ping timeout: 480 seconds]
16:06swaggie has joined #asahi-dev
16:14swaggie has quit [Ping timeout: 480 seconds]
16:37swaggie has joined #asahi-dev
16:45swaggie has quit [Ping timeout: 480 seconds]
17:19iaguis has quit [Quit: leaving]
17:26swaggie has joined #asahi-dev
17:38swaggie has quit [Ping timeout: 480 seconds]
18:55nullmaz: I forget, did you end up with my T1000 in the end? IIRC I still have the rails in the cupboard under my stairs
19:02bcrumb has joined #asahi-dev
19:05bcrumb has quit []
19:31maznull: yeah, I have two now. no need for rails though, the machines are piled 1.5m high. high density computing! ;-)
19:54bcrumb has joined #asahi-dev
19:58renatorabelo has joined #asahi-dev
19:58bluetail has quit [Read error: Connection reset by peer]
19:58bluetail has joined #asahi-dev
20:05bluetail has quit [Read error: Connection reset by peer]
20:05bluetail has joined #asahi-dev
20:10Gaspare has joined #asahi-dev
20:10bcrumb has quit [Quit: WeeChat 3.7.1]
20:11bcrumb has joined #asahi-dev
20:17bcrumb has quit [Quit: WeeChat 3.7.1]
20:19Gaspare has quit [Quit: Gaspare]
20:25bcrumb has joined #asahi-dev
20:29bluetail has quit [Ping timeout: 480 seconds]
20:39bcrumb has quit [Quit: WeeChat 3.7.1]
21:06amarioguy has quit [Remote host closed the connection]
21:06renato has joined #asahi-dev
21:06amarioguy has joined #asahi-dev
21:13renatorabelo has quit [Ping timeout: 480 seconds]
21:21swaggie has joined #asahi-dev
21:35bcrumb has joined #asahi-dev
21:36bcrumbit would be cool if someone could look @ those asahi-scripts PRs for me, they would make life much easier when upgrading to new kernel versions :D
21:36bcrumbobviously, when you guys have time, just mentioning
21:41bcrumb has quit [Quit: WeeChat 3.7.1]
21:49bcrumb has joined #asahi-dev
21:51bcrumbalso i think the installer should notice whether the presets in mkinitcpio.d are x-ed out and then upgrade those, instead of writing new .presets
21:51bcrumbso, the whole Make process
21:52bcrumbalso sorry, i don't know if this exactly only kernel talk or also stuff like this fits here
21:52bcrumb has quit []
21:59swaggie has quit [Remote host closed the connection]
22:01swaggie has joined #asahi-dev
22:12wCPO624365 has quit []
22:25renatorabelo has joined #asahi-dev
22:25javier_varez__ has joined #asahi-dev
22:26zzywysm_ has joined #asahi-dev
22:27nathanchance_ has joined #asahi-dev
22:27jbowen_ has joined #asahi-dev
22:28wCPO624365 has joined #asahi-dev
22:28Method_ has joined #asahi-dev
22:28jbowen has quit [Ping timeout: 480 seconds]
22:28zzywysm has quit [Read error: Connection reset by peer]
22:28digicyc has quit [Remote host closed the connection]
22:28javier_varez_ has quit [Ping timeout: 480 seconds]
22:28javier_varez__ is now known as javier_varez_
22:28digicyc has joined #asahi-dev
22:28jbowen_ has quit []
22:29jbowen_ has joined #asahi-dev
22:29renato has quit [resistance.oftc.net larich.oftc.net]
22:29Method has quit [resistance.oftc.net larich.oftc.net]
22:29nathanchance has quit [resistance.oftc.net larich.oftc.net]
22:29ids1024 has quit [resistance.oftc.net larich.oftc.net]
22:29jbowen_ has quit [resistance.oftc.net larich.oftc.net]
22:30jbowen_ has joined #asahi-dev
22:30ids1024 has joined #asahi-dev
22:30Method has joined #asahi-dev
22:30renato has joined #asahi-dev
22:30Method has quit [Ping timeout: 482 seconds]
22:31renato has quit [Ping timeout: 482 seconds]
22:42SSJ_GZ has quit [Ping timeout: 480 seconds]
23:10ajxu2 has joined #asahi-dev
23:11ajxu2 has quit []
23:31manawyrm has quit [Quit: Read error: 2.99792458 x 10^8 meters/second (Excessive speed of light)]
23:31manawyrm has joined #asahi-dev
23:56nathanchance_ has quit []
23:57nathanchance has joined #asahi-dev
…