#
✕ nick:axboe
#asahi · 2022-02-19 21:02
axboebase tdp 125W, turbo 241W
#asahi · 2022-02-19 21:00
axboeI'd say it's pretty damn fast
#asahi · 2022-02-19 21:00
axboemy qemu kernel takes ~85s, and the one I run on the m1 takes ~3 min
#asahi · 2022-02-19 20:59
axboenot this is allmodconfig x86, so it's basically all of it
#asahi · 2022-02-19 20:59
axboeI'd say 20.8m is very respectable
#asahi · 2022-02-19 20:59
axboetpw_rules: 12900k does it in 12.45 min
#asahi · 2022-02-19 20:59
axboetpw_rules: I ran on a 5950x and 12900k for comparison, and x1 is completing it too (but that'll take a while)
#asahi · 2022-02-19 20:59
axboetpw_rules: yep, small one though, 6 p cores
#asahi · 2022-02-19 20:57
axboetpw_rules: did allmodconfig and matches your results - 20.8m for 16k ps, 25.9m for 4k ps
#asahi · 2022-02-19 16:45
axboetiming allmodconfig, let's see how big of a difference it is here with all the crap enabled
#asahi · 2022-02-19 16:37
axboemps: massive difference indeed
#asahi · 2022-02-19 16:34
axboemine is small enough that it's around 85s for the whole thing
#asahi · 2022-02-19 16:34
axboemps: what's the baseline time for the kernel build you are doing?
#asahi · 2022-02-19 16:33
axboeI'll try a bigger compile, but that just makes the 16k page size even more compelling
#asahi · 2022-02-19 16:29
axboetpw_rules: that's huge - since my compile is not that big, linking is probably skewing the numbers on my side in terms of raw compile speed differences
#asahi · 2022-02-19 16:27
axboetpw_rules: yep pro with 6 p cores, just compiling my for-next release kernel that I use in qemu for local testing
#asahi · 2022-02-19 16:21
axboe(not your fault, looks like an alsa oddity)
#asahi · 2022-02-19 16:20
axboewould also clean up those weird <= loops
#asahi · 2022-02-19 16:20
axboeidiomatic way is usually to have FOO_NR, where FOO_NR is indeed last-valid + 1
#asahi · 2022-02-19 16:20
axboepovik: man, that's a counter intuitive way to define LAST
#asahi · 2022-02-19 15:50
axboethough I probably personally save more by just using FF instead of chromium than I waste in the page cache ;)
#asahi · 2022-02-19 15:49
axboewould be nifty with some page cache stats that show you how much space you're wasting there with 16k pages
#asahi · 2022-02-19 15:48
axboeand 4k is about 7% slower for me
#asahi · 2022-02-19 15:48
axboefwiw, I tested 16k vs 4k page size for my compilation test
#asahi · 2022-02-19 15:47
axboevery different scenario from asahi, thankfully
#asahi · 2022-02-19 15:47
axboeyou need an eye wash station closeby for a lot of those, if you look at the changes/code
#asahi · 2022-02-19 15:47
axboeand they often just abandon the tree once the product is shipped, it's not a long term maintained thing :/
#asahi · 2022-02-19 15:46
axboethose vendors never target upstream, so they don't really care how they drift
#asahi · 2022-02-19 15:46
axboe18k isn't much, still doesn't mean it won't take some time to flush it out
#asahi · 2022-02-19 15:46
axboeindeed
#asahi · 2022-02-19 15:44
axboelittle victories
#asahi · 2022-02-19 15:43
axboej`ey: yeah I know, just talking about the feature patches and fixes I stacked on top of 'asahi'
#asahi · 2022-02-19 15:41
axboeand drastically shrunk my on-top patch pile
#asahi · 2022-02-19 15:41
axboedone, works for me
#asahi · 2022-02-19 15:38
axboebrb, time to reboot into new kernel
#asahi · 2022-02-19 15:37
axboeI never do meetings on the laptop, need a bigger screen
#asahi · 2022-02-19 15:37
axboej`ey: hah
#asahi · 2022-02-19 15:36
axboeok
#asahi · 2022-02-19 15:35
axboepovik: t600x support for audio yet, or is it still just t8x?
#asahi · 2022-02-19 15:35
axboetime to update this sucker!
#asahi · 2022-02-19 15:35
axboefixes and marches on
#asahi · 2022-02-19 15:33
axboejust ran into that too when rebasing
#asahi · 2022-02-18 17:47
axboe15 second press always does it for me, the two minor blinks at 10 and 11 seconds help my synchronize it :)
#asahi · 2022-02-18 17:32
axboej`ey: will try and remove the efi bios argument, boots fine without it
#asahi · 2022-02-18 17:31
axboeI get it sporadically here
#asahi · 2022-02-18 17:31
axboej`ey: good point, I was a bit puzzled in figuring out where it's coming from
#asahi · 2022-02-18 17:28
axboecurious if anyone have experienced: "Synchronous Exception at 0x664C" when running qemu?
#asahi · 2022-02-18 17:07
axboeshould not show up in smartlog from the drive
#asahi · 2022-02-18 17:06
axboethough of course who knows what'll happen with a marginal setup
#asahi · 2022-02-18 17:06
axboecabling would usually show transport issues, not uncorrectable media errors

plus ›