00:45
mmu_man has quit [Ping timeout: 252 seconds]
00:47
sobkas has quit [Quit: sobkas]
00:55
sobkas has joined #u-boot
01:06
vagrantc has quit [Quit: leaving]
01:10
jclsn7 has quit [Quit: Ping timeout (120 seconds)]
01:10
jclsn7 has joined #u-boot
01:15
jclsn7 has quit [Ping timeout: 240 seconds]
01:21
jclsn7 has joined #u-boot
01:35
jclsn7 has quit [Ping timeout: 240 seconds]
01:41
jclsn7 has joined #u-boot
01:48
jclsn7 has quit [Ping timeout: 240 seconds]
01:53
jclsn7 has joined #u-boot
01:58
jclsn7 has quit [Ping timeout: 272 seconds]
02:03
jclsn7 has joined #u-boot
02:07
<
sjg1 >
milkylainen: OK, not sure if that is the issue though. Does it actually jump to Linux?
02:09
thopiekar_ has joined #u-boot
02:09
thopiekar has quit [Killed (tantalum.libera.chat (Nickname regained by services))]
02:09
thopiekar_ is now known as thopiekar
02:11
jclsn7 has quit [Ping timeout: 240 seconds]
02:17
jclsn7 has joined #u-boot
02:23
jclsn7 has quit [Ping timeout: 256 seconds]
02:29
jclsn7 has joined #u-boot
02:38
jclsn7 has quit [Ping timeout: 256 seconds]
02:44
jclsn7 has joined #u-boot
02:51
jclsn7 has quit [Ping timeout: 272 seconds]
02:57
jclsn7 has joined #u-boot
03:04
jclsn77 has joined #u-boot
03:06
jclsn7 has quit [Ping timeout: 256 seconds]
03:06
jclsn77 is now known as jclsn7
03:06
apalos has joined #u-boot
03:07
thopiekar has quit [Ping timeout: 256 seconds]
03:07
thopiekar has joined #u-boot
03:22
akaWolf has quit [Ping timeout: 240 seconds]
03:28
jclsn7 has quit [Ping timeout: 272 seconds]
03:31
akaWolf has joined #u-boot
03:33
jclsn7 has joined #u-boot
03:39
jclsn7 has quit [Ping timeout: 240 seconds]
03:45
jclsn7 has joined #u-boot
03:57
jclsn7 has quit [Ping timeout: 256 seconds]
04:02
jclsn7 has joined #u-boot
06:09
sbach has quit [Read error: Connection reset by peer]
06:11
sbach has joined #u-boot
08:10
Thorn has quit [Ping timeout: 240 seconds]
08:13
Thorn has joined #u-boot
08:43
mmu_man has joined #u-boot
09:05
mmu_man has quit [Ping timeout: 256 seconds]
09:16
mmu_man has joined #u-boot
09:36
mmu_man has quit [Ping timeout: 256 seconds]
09:47
mmu_man has joined #u-boot
10:17
chrfle has quit [Ping timeout: 240 seconds]
10:20
chrfle has joined #u-boot
11:20
mmu_man has quit [Ping timeout: 256 seconds]
11:40
<
kettenis >
that is now fully in-synch with the DTs on the asahi branch of the linux kernel repo
11:41
redbrain has joined #u-boot
11:43
<
kettenis >
oops, didn't intend that for this channel, but it is still on-topic ;)
11:47
mmu_man has joined #u-boot
12:22
sobkas_ has joined #u-boot
12:25
sobkas has quit [Ping timeout: 256 seconds]
12:25
sobkas_ is now known as sobkas
12:44
<
mps >
kettenis: I confirm that it works
12:45
JoaoSchim has quit [Remote host closed the connection]
12:58
JoaoSchim has joined #u-boot
13:39
___nick___ has joined #u-boot
13:40
___nick___ has quit [Client Quit]
13:42
___nick___ has joined #u-boot
15:00
<
marex >
kettenis: how is the m1 machine performance-wise ? is it build-machine material ?
15:02
<
mps >
marex: 'build-machine material', what that means
15:03
<
kettenis >
performance is comparable with a fast x86 laptop I'd say
15:03
<
marex >
mps: something you can do e.g. kernel builds on
15:03
<
kettenis >
absolutely
15:03
<
mps >
marex: sure, I build kernel in less than 5 minutes usually
15:04
<
kettenis >
my M1 machines do an OpenBSD kernel build in less than 2 minutes and a full build of the base OS in just over an hour (that includes building clang)
15:04
<
kettenis >
it is faster than the 32-core Ampere eMAG machine I have ;)
15:05
<
mps >
hmm, with these Jens's patches I should try again to see is it faster
15:05
<
marex >
I was about to point out 5 minutes isn't very quick
15:05
<
marex >
but with cpufreq, it looks better indeed
15:05
<
marex >
j`ey: that ... looks nice
15:06
<
mps >
marex: well, on my alpine develepment neoverse-m1 it takes about 10-15 minutes
15:06
<
marex >
kettenis: get AB/RB from Bin Meng and I'll pick it, he's doing the usb3 stuff
15:07
<
mps >
ah, and I set conservative governor usually
15:07
<
marex >
kettenis: just CC him if you didn't already, and ask for AB/RB, it should be quick
15:07
<
kettenis >
marex: ok; I made it an RFC because I'm not sure how u-boot wants to pick up the role switch stuff that Linux has
15:10
mmu_man has quit [Ping timeout: 272 seconds]
15:10
<
marex >
kettenis: the usb2 parts already do role switch, so why not ?
15:11
<
marex >
the usb stack is also partly pulled from linux already
15:11
<
marex >
that could make things easier too
15:15
<
kettenis >
well, I'm not really interested in adding full role-switch support; just in making sure that we pick up host mode if that's what the device tree says
15:16
sobkas has quit [Ping timeout: 250 seconds]
15:17
<
mps >
I forgot to tell that I build on kernel with 4K page size, which is slower than default 16K
15:18
<
mps >
and ofc speed of kernel build depends how much modules/options are enabled in config
15:23
<
marex >
kettenis: ah, that, so host/peripheral yes, but not otg
15:55
mmu_man has joined #u-boot
16:03
grgy has joined #u-boot
16:08
<
tg-bridge-bot-ub >
<Clamor_S> why CONFIG_SKIP_LOWLEVEL_INIT gives +10mb of size to image
16:55
mmu_man has quit [Ping timeout: 272 seconds]
17:05
mmu_man has joined #u-boot
18:09
<
milkylainen_ >
marex: It's about as fast as a latest generation x86, but easily wins out on perf/W. 5nm production has advantages. :)
18:38
<
marex >
milkylainen_: so how does it compare e.g. to RPi4 ? :)
18:38
<
marex >
(that might be something more people have access to)
18:39
<
j`ey >
muccccch faster
18:40
<
j`ey >
my kernel build on rpi is 30mins+, 1-2min on m1
18:43
thopiekar has quit [Ping timeout: 256 seconds]
18:45
thopiekar has joined #u-boot
18:48
<
marex >
the m1 has what, 8 cores ? rpi4 has 2 A72 ?
18:48
<
marex >
so that's like 4 times faster per core ?
18:48
<
j`ey >
rpi4 has 4 cores
18:49
<
marex >
4 A72 then ?
18:49
<
marex >
and you build from some pcie ssd in both cases ?
18:49
<
marex >
or NFSroot ?
18:51
<
j`ey >
thats where the comparison breaks down, my rpi has much worse storage
18:53
<
milkylainen_ >
marex: That is pretty much comparing apples to oranges. The bcm2711 suffer so badly compared to a desktop class CPU. :)
18:58
<
marex >
milkylainen_: that's like comparing apples and raspberries ... is what you wanted to say ?
18:59
<
marex >
j`ey: so rpi has memory bandwidth / pcie bandwidth issues ? hah ...
19:06
<
kettenis >
a cluster of rpi4's probably still beats the M1 for performance per peso
19:17
<
mps >
marex: I build sometimes kernel on rk3399 (samsung gru-kevin chromebook) with FS on external usb-c 3.1 speed ssd disk
19:18
<
mps >
it takes about 30-45 minutes
19:19
<
mps >
maybe even more, I didn't built kernel on it for about year
19:56
vagrantc has joined #u-boot
20:39
darkapex_ has joined #u-boot
20:40
darkapex has quit [Ping timeout: 240 seconds]
21:04
___nick___ has quit [Ping timeout: 272 seconds]
21:40
redbrain has quit [Ping timeout: 256 seconds]
21:47
sobkas has joined #u-boot
21:52
macromorgan has quit [Read error: Connection reset by peer]
21:52
macromorgan has joined #u-boot
22:45
thopiekar_ has joined #u-boot
22:45
thopiekar is now known as Guest5415
22:45
thopiekar_ is now known as thopiekar
22:46
Guest5415 has quit [Ping timeout: 256 seconds]