2021-10-21
00:37amw_ is now known as amw
01:00DragoonAethis has quit [Quit: hej-hej!]
01:00DragoonAethis has joined #asahi
01:03Gues__________________________ has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
01:08eric_engestrom has quit [Ping timeout: 480 seconds]
01:08steev_ has joined #asahi
01:09steev has quit [Ping timeout: 480 seconds]
01:10jkkm has quit [Ping timeout: 480 seconds]
01:10jkkm has joined #asahi
01:10eichin has quit [Ping timeout: 480 seconds]
01:11eric_engestrom has joined #asahi
01:12eichin has joined #asahi
01:20yuyichao has joined #asahi
01:32malvo has quit [Read error: Connection reset by peer]
01:32malvo has joined #asahi
02:02steev_ has quit []
02:03steev has joined #asahi
02:15phiologe has joined #asahi
02:18Gues__________________________ has joined #asahi
02:19PhilippvK has quit [Ping timeout: 480 seconds]
03:17marvin24_ has joined #asahi
03:21marvin24 has quit [Ping timeout: 480 seconds]
03:55chadmedhttps://browser.geekbench.com/v5/cpu/10476727 im sure most have probably seen this already but this puts the M1 Max at approx. a stock ryzen 5800x in terms of raw performance. quite impressive, imo
04:05kov has quit [Quit: Coyote finally caught me]
04:05kov has joined #asahi
04:49riker77_ has joined #asahi
04:51riker77 has quit [Ping timeout: 480 seconds]
04:51riker77_ is now known as riker77
05:30jannauhttps://browser.geekbench.com/v5/cpu/search?q=MacBookPro18 has more results with newer macos builds which seems to report the CPU frequency correctly
06:03crabbedhaloablut has quit [Ping timeout: 480 seconds]
06:13crabbedhaloablut has joined #asahi
06:21nobodynada has quit [Quit: Lost terminal]
07:05bingoChecker has joined #asahi
07:32tophevich[m]jannau: afaik for cpu benchmarks the only thing that could make a difference is the memory bandwidth difference between the 32 pro and the 64 max, where the benchmark would realy make use of the extra bandwidth. when talking about the rest of the soc you'd need specific benchmarks as well like one for ProRes, graphics ... and so on
07:38_jannau_yes, only memory badnwidth and probably larger system level caches could make a difference but either they don't or the difference is too small within geekbench
07:42X-Scale` has joined #asahi
07:47X-Scale has quit [Ping timeout: 480 seconds]
08:03minithat's ridiculous really... 5800X level performance in a third or less of the power? mental
08:07tophevich[m]yea, I'd say the have a bit of a thermal headroom for the desktop pro models :D
08:07tophevich[m]s/the/they
08:08miniI actually happen to have a 5800X, so I'l enjoy comparing the two when my new shiny arrives :)
08:14tophevich[m]But I'm currious what they'll do about the die/soc size, and the scaling of it. If they keep all the gpu bits in there (which they will want to scale up even further) and scale the cpu cores + caches at the same time ...
08:17miniwell, by all accounts the M1 Max die is absolutely massive, I'm just curious as to how they do the desktop ones
08:17frytaped has quit [Quit: went bye bye]
08:17frytaped has joined #asahi
08:17tophevich[m]that's what I'm saying :D
08:23pwg has quit [Ping timeout: 480 seconds]
08:26chadmedi wouldnt be surprised at all if they are haemorrhaging money on the chips, but it doesnt matter if theyre turning a profit on the system as a whole
08:27chadmedtheyll get to MCMs eventually too, and probably next gen just have two dies that they bin for each product segment. it cannot be cheap supplying 3 dies with so few bins on each
08:33pwg has joined #asahi
08:35maennich has quit [Read error: Connection reset by peer]
08:35maennich has joined #asahi
08:38chadmed has quit [Remote host closed the connection]
08:38chadmed has joined #asahi
09:09marcanI wonder how the GPUs work in that, do they actually schedule work cross-die or is it all software-managed?
09:10chadmedif theres no hidden hardware for that kind of thing on die right now, id imagine theyre probably just hooked together with AMBA or some other fabric and its all handled in software
09:13chadmedim hoping before christmas to save me the hassle of getting a maxxed-out mbp, but its probably unlikely :(
09:17kettenismarcan: any chance for you (or someone else on here) to dump the ADT in .adt.txt format like you did for the earlier ones?
09:19kettenisneed to figure out how to do that myself at some point, but I'm really curious to see what changed and I don't really want to download that full ipsw right now
09:21chadmedyouve already got one of the new machines? theyre not shipping for at least another monthe here lol
09:24kettenisso it lacks the details filled in by iBoot, but it gives a good idea how the hardware changed
09:45marcannow what... I guess I comment out the early bring-up debug in m1n1 and make the real uart code just parse the ADT
09:49marcanthe dramcfg thing definitely changed; I wonder what they do now and how it interacts with the two p-clusters
09:50marcanthey also have PLL_GFX01 and PLL_GFX23 now, which looks like the GPU is segmented to some extent
09:51marcanstill just a single sgx node though, so I'm guessing at least in single-chip cases that's all still magically hidden
09:55maor26 has joined #asahi
10:08marcanthese GPUs are tiled, and were already split into cores anyway, so there's always been some magic dispatch
10:11dottedmagAre there several command streams per GPU (1 per core?) in M1 already, or is doing magical dispatch by parsing a single command stream?
10:18svenright. and i wouldn't have seen it on the mac mini when i was tracing the SEP because it obviously doesn't have touch id
10:23svenyeah, i guess i just assumed it was all hidden inside the SEP because i never saw any IPC messages
10:37latosca[m] has joined #asahi
10:38X-Scale has joined #asahi
10:44X-Scale` has quit [Ping timeout: 480 seconds]
10:45yorgos has joined #asahi
10:49yorgos_ has joined #asahi
10:49yorgos has quit []
10:51yorgos_ has quit []
10:54bgb_marcan: I'm confused about the iodev thing. if i boot linux with run_guest.py, USB1(ttyACM1) is used for vuart redirection, USB0 is used for uartproxy(default M1N1DEVICE=ttyACM0), is that right?
11:07chadmed has quit [Quit: Konversation terminated!]
11:45chadmed has joined #asahi
12:53bgb_marcan: I see "Initializing hypervisor over iodev IODEV.USB1", and for HV, self.iodev = self.p.iodev_whoami() = uartproxy_iodev(uartproxy.c).
13:05Gues__________________________ has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
13:11brentr123[m]Is this on the new Macs?
13:14_jannau_bgb_: the iodev numbers on the m1 depend on the port used for uartproxy. that has nothing to do with the ttyACM devices on the connected linux host
14:06Gues__________________________ has joined #asahi
14:06yuyichao has quit [Ping timeout: 480 seconds]
14:26yuyichao has joined #asahi
14:33aleasto has joined #asahi
15:32Gues__________________________ has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
15:57nobodynada has joined #asahi
16:11bakedpotatoez[m] has joined #asahi
16:12bingoChecker has quit [Quit: Leaving]
16:12bakedpotatoez[m]Hey I'm new here. I just heard of Asahi and I was wondering if it's ready for use or if it's still in development?
16:27tertu has quit [Quit: so long...]
16:29tertu has joined #asahi
16:35gabuscus has joined #asahi
16:40gabuscus_ has quit [Ping timeout: 480 seconds]
17:07gabuscus_ has joined #asahi
17:11gabuscus has quit [Ping timeout: 480 seconds]
17:52rohin has joined #asahi
17:58nobodynada has quit [Ping timeout: 480 seconds]
18:19nobodynada has joined #asahi
18:20bakedpotatoez[m]<marcan> "still in development; adventurou..." <- Aw okay. Thanks anyways!
18:21bakedpotatoez[m]There isn't an ETA is there?
18:27nobodynada has quit [Ping timeout: 480 seconds]
18:39i509vcb[m]<bakedpotatoez[m]> "There isn't an ETA is there?" <- What's an ETA lol
18:40i509vcb[m](anyways welcome to open source)
18:43nobodynada has joined #asahi
18:43jnsome stuff works already, and you can arguably run a functional mac mini desktop with asahilinux now (after finding the right kernel branches and spending some time on installing)
18:44jn(side note: https://github.com/AsahiLinux/docs/wiki/Project-status seems a bit outdated, now that i look at it)
18:46bakedpotatoez[m]<i509vcb[m]> "What's an ETA lol" <- Estimated Time of Arrival. In case I'm referring to when Asahi will be ready for use.
18:46i509vcb[m](my message was in jest)
18:46bakedpotatoez[m]ah lol
18:46bakedpotatoez[m]idk how i missed that
18:47jnhmm, i guess an interesting and significant milestone would be the completion of the automatic installer
18:51i509vcb[m]Is asahi-gpu the place to be if I want to volunteer on implementing vulkan on the gpu or is that the reverse engineering channel
18:55i509vcb[m]Yeah zink is OpenGL on Vulkan
19:04aleasto has quit [Remote host closed the connection]
19:09aleasto has joined #asahi
19:15kenjigashu has joined #asahi
19:18riker77 has quit [Quit: Quitting IRC - gone for good...]
19:20Glanzmann has joined #asahi
19:20suricato has quit [Ping timeout: 480 seconds]
19:23jacoxon has joined #asahi
19:24suricato has joined #asahi
19:24Glanzmannkettenis: OpenBSD beats Macos in SSD 4k random write performance. Question: What size is your NVMe? I updated https://ab34.de/u/m1-nvme-performance-linux-vs-macos-2021-10-18.txt
19:28tertuthe modern minis are passively cooled right, maybe they'll just go to fans like on the G4 mini
19:29riker77 has joined #asahi
19:35Glanzmannmini: I normally use an intel nuc as workstation which is really noisy. If the nuc is off, I don't hear anything. For me the mini appears to be silent.
19:35miniGlanzmann: yeah, it's not even close. I have a nuc8i5beh here too and that is very very noisy
19:40Glanzmannmini: I have the very same model (NUC8i5BEH): [ 0.000000] DMI: Intel(R) Client Systems NUC8i5INH/NUC8i5INB, BIOS INWHL357.0043.2021.0817.1957 08/17/2021
19:43Gues__________________________ has joined #asahi
19:48kenjigashu has quit []
19:56aleasto has quit [Remote host closed the connection]
19:57aleasto has joined #asahi
19:57aleasto has quit [Remote host closed the connection]
19:58aleasto has joined #asahi
19:59miniGlanzmann: no, I haven't. to be honest I use it as little as I can get away with, it's just too loud
20:03aleasto has quit [Remote host closed the connection]
20:09nobodynada has quit [Ping timeout: 480 seconds]
20:35aleasto has joined #asahi
20:35alexstore06 has joined #asahi
20:36alexstore06 has quit [Quit: Leaving...]
20:38alexstore06 has joined #asahi
20:43Glanzmann has quit [Quit: EOF]
21:03yuyichao_ has joined #asahi
21:07yuyichao has quit [Ping timeout: 480 seconds]
21:09nobodynada has joined #asahi
21:10jeffmiw has joined #asahi
21:12nskl has joined #asahi
21:16nsklaus_ has quit [Ping timeout: 480 seconds]
21:23yuyichao has joined #asahi
21:27yuyichao_ has quit [Ping timeout: 480 seconds]
21:38jacoxon has quit []
21:40Gues__________________________ has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
21:47citizen1[m]how linux will work with a macbook with a notch?
21:50j_eythe first attempt will be to probably just try make the screen only go up to the bottom of the notch
21:54mort_it would be cool to let application content only use the part below the notch, but then put some permanent status bar type thing in the space with the notch in such a way that it's not even covered by true fullscreen apps
21:54citizen1[m]this is not exactly asahi question but m1 . So m1 is powerful enough thats its actually just as powerful as discreet gpus from nvidia and amd?
21:55mort_apple's benchmarks showed that the m1 max is very nearly as fast as the mobile edition of the RTX 3080
21:55citizen1[m]mort_: maybe a widget to make virtual bezel on the sides of the notch to hide it
21:57citizen1[m]mort_: mind boggling , rtx has its own fans
21:57citizen1[m]that risc is some black magic
21:58mort_but the machine which Apple compared the M1 Max to is apparently this one: https://www.msi.com/Laptop/GE76-Raider-11UX
22:00mort_the memory is insane though, DDR4 has a throughput of 25.6 GiB/s, the new M1s have 200 and 400 GiB/s
22:01citizen1[m]my understanding that ddr4 is ram memory, how this affects speed of computation?
22:04mort_there are a bunch of levels of cache, typically 3 (l1, l2, l3), where l1 is basically instant to access but small, l2 is a bit slower to access but bigger, l3 is slower still to access but even bigger, and then there's the RAM which is huge but very slow to access
22:04mort_if your entire dataset fits within the CPU's cache, then it doesn't have to touch RAM and memory bandwidth doesn't matter at all
22:05mort_if your dataset doesn't fit within the cache, you have to read it from RAM, at which point the speed of accessing the memory becomes a potential issue
22:06mort_if you have to read from RAM, but you're doing random access to RAM, throughput doesn't matter a lot, latency is much more important
22:06mort_but if you have to read from RAM, and you're doing sequential (or otherwise predictable) access to RAM, *then* throughput becomes very important
22:07mort_example: if you're applying some effect to an image, you'll be iterating through every pixel in the image; you'll probably be doing that with all 10 cores at the same time, going over different sections; all 10 cores will be doing sequential access, and if the amount of work per pixel is relatively small, you may end up being limited by RAM
22:08aleasto has quit [Remote host closed the connection]
22:10mort_sorry for the wall of text, but how RAM affects the speed of computation doesn't have a very simple explanation
22:13citizen1[m]thanks for explaining
22:32alexstore06 has quit [Ping timeout: 480 seconds]
22:33Gue___________________________ has joined #asahi
22:56Gue___________________________ has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
23:02Gue___________________________ has joined #asahi
23:08Jakob[m]123 has joined #asahi
23:26maor26 has quit [Ping timeout: 480 seconds]
23:32yuyichao has quit [Ping timeout: 480 seconds]
23:50FireFox317 has quit [Ping timeout: 480 seconds]
23:51gladiac is now known as Guest3722
23:51gladiac has joined #asahi
23:57Guest3722 has quit [Ping timeout: 480 seconds]