#
✕ nick:axboe
#asahi · 2022-02-10 17:02
axboe~100s to build the current kernel for qemu testing
#asahi · 2022-02-10 16:47
axboekettenis: as long as it works in grub so I can select a given kernel, that's all I really need for now. and it does
#asahi · 2022-02-10 16:45
axboethis is exciting though, being self hosted will make testing kernel changes that much easier
#asahi · 2022-02-10 16:45
axboekettenis: sorry ;)
#asahi · 2022-02-10 16:44
axboe_jannau_: yep that was it
#asahi · 2022-02-10 16:41
axboe_jannau_: oh probably grabbed the dtb before that commit
#asahi · 2022-02-10 16:41
axboe_jannau_: I have that one
#asahi · 2022-02-10 16:40
axboeI used t6000-j314s.dtb - just that one
#asahi · 2022-02-10 16:38
axboethough kbd doesn't work in grub
#asahi · 2022-02-10 16:38
axboeu-boot -> grub seems to work
#asahi · 2022-02-10 16:28
axboeGlanzmann: here?
#asahi · 2022-02-10 16:09
axboethe hardship of being a minority platform still :)
#asahi · 2022-02-10 16:08
axboelooks like that commit in mainline added that, as it was triggering on s390 too
#asahi · 2022-02-10 16:08
axboeAuthor: Christian Borntraeger <borntraeger@linux.ibm.com>
#asahi · 2022-02-10 16:08
axboeKVM: avoid warning on s390 in mark_page_dirty
#asahi · 2022-02-10 16:08
axboeDate: Thu Jan 13 13:29:24 2022 +0100
#asahi · 2022-02-10 16:08
axboecommit e09fccb5435d7b9ab3fd5dfeada8ae40cfa56e08
#asahi · 2022-02-10 16:07
axboemaz: ah, the smc/work branch does not have that ifdef around it
#asahi · 2022-02-10 16:06
axboeoh, guess I should read it more carefully
#asahi · 2022-02-10 16:04
axboeGlanzmann sent me instructions, very helpful
#asahi · 2022-02-10 16:04
axboeyep
#asahi · 2022-02-10 16:01
axboewill try u-boot now...
#asahi · 2022-02-10 16:01
axboemaz: I'll send out an email
#asahi · 2022-02-10 15:59
axboemaz: are you interested in the mark_page_dirty_in_slot()? or is that a known thing
#asahi · 2022-02-10 15:59
axboemaz: you bet, will do
#asahi · 2022-02-10 15:59
axboemaz: so don't put much thought into it
#asahi · 2022-02-10 15:58
axboemaz: as I said, most likely my screwup. I'll give it a go later again, when I have u-boot on it for easier kernel testing
#asahi · 2022-02-10 15:57
axboehttps://kernel.dk/dmesg-m1.txt
#asahi · 2022-02-10 15:56
axboesure, 2 sec
#asahi · 2022-02-10 15:56
axboeright :)
#asahi · 2022-02-10 15:56
axboeor just the current one
#asahi · 2022-02-10 15:56
axboemaz: on the kernel where it hung?
#asahi · 2022-02-10 15:55
axboethis one
#asahi · 2022-02-10 15:54
axboeif (WARN_ON_ONCE(!vcpu) || WARN_ON_ONCE(vcpu->kvm != kvm))
#asahi · 2022-02-10 15:54
axboebut doesn't seem to cause any issues
#asahi · 2022-02-10 15:54
axboeI do consistently hit the WARN_ON_ONCE() in virt/kvm/kvm_main.c:mark_page_dirty_in_slot() on my current kernel
#asahi · 2022-02-10 15:53
axboemaz: I wouldn't put too much into it, was running a merged kernel. it's either a 5.17-rc issue, or it's a faulty merge on my end
#asahi · 2022-02-10 15:53
axboeyep
#asahi · 2022-02-10 15:53
axboemaz: and yeah, this is qemu+kvm
#asahi · 2022-02-10 15:53
axboemaz: since I don't have serial console on the laptop, no idea where it actually hung. but had to be power cycled
#asahi · 2022-02-10 15:52
axboemaz: on the kernel I ran, it would hang the machine after 2-4 invocations
#asahi · 2022-02-10 15:44
axboegot a debian installed, and it boots in a few seconds, so testing is now covered for me
#asahi · 2022-02-10 15:44
axboeto follow up on the qemu situation, things appear solid on the smc/work + various patches base
#asahi · 2022-02-10 15:43
axboeGlanzmann: thanks for the u-boot info! I'm going to give it a go right now
#asahi · 2022-02-10 02:35
axboeit was my janky kernel, works fine now
#asahi · 2022-02-09 20:52
axboecould be my janky kernel
#asahi · 2022-02-09 20:50
axboemachine
#asahi · 2022-02-09 20:49
axboeand again, hmm
#asahi · 2022-02-09 20:45
axboewell I think I crashed it :)
#asahi · 2022-02-09 20:39
axboethanks

plus ›