Re: Some progress, Guix rumpdisk still crashes...

[Damien Zammit]

Try '-M q35' passed to qemu as this will use a newer chipset emulation that includes AHCI controller by default.  Im not sure if IDE is working at this point.

Damien


Sent from ProtonMail mobile



-------- Original Message --------
On 18 May 2023, 4:24 am, Janneke Nieuwenhuizen < janneke@gnu.org> wrote:

Hi!

So, Guix's glibc-2.35 has patches for centiseconds and monotonic (and
others) that were forward ported since glibc 2.31, instead of refreshed
from upstream Debian Salsa:

https://git.savannah.gnu.org/cgit/guix.git/tree/gnu/packages/patches/glibc-hurd-clock_gettime_monotonic.patch
https://git.savannah.gnu.org/cgit/guix.git/tree/gnu/packages/patches/glibc-hurd-clock_t_centiseconds.patch
https://git.savannah.gnu.org/cgit/guix.git/tree/gnu/packages/patches/glibc-hurd-gettyent.patch
https://git.savannah.gnu.org/cgit/guix.git/tree/gnu/packages/patches/glibc-hurd-mach-print.patch
https://git.savannah.gnu.org/cgit/guix.git/tree/gnu/packages/patches/glibc-hurd-signal-sa-siginfo.patch

As discussed on IRC, I reverted

glibc-hurd-clock_gettime_monotonic.patch
glibc-hurd-clock_t_centiseconds.patch

and applied

https://salsa.debian.org/glibc-team/glibc/-/blob/sid/debian/patches/hurd-i386/local-clock_gettime_MONOTONIC.diff
https://salsa.debian.org/glibc-team/glibc/-/blob/sid/debian/patches/hurd-i386/unsubmitted-clock_t_centiseconds.diff

I disregarded/kept the others as they were (gettyent, mach-print,
signal-sa-siginfo).

With this newly patched glibc

https://gitlab.com/janneke/guix/-/tree/wip-hurd12

rumpdisk still crashes, but the good news (I guess) is that it seems to
get somewhat further, or at least it crashes differently. Here are the
last 24 (WTF, 1980 wants their screensize back!?) lines (I don't know
how to get the full log from QEMU):

--8<---------------cut here---------------start------------->8---
[..] ??? how to capture these? using --display curses
[ 1.0000040] timecounter: Timecounter "clockinterrupt" frequency 100 Hz qualit
y 0
[ 1.0000050] cpu0 at thinair0: rump virtual cpu
[ 1.0000050] entropy: WARNING: extracting entropy too early
[ 1.0200050] root file system type: rumpfs
[ 1.0200050] kern.module.path=/stand/i386/9.99.88/modules
[ 1.0200050] mainbus0 (root)
[ 1.0200050] pci0 at mainbus0 bus 0
[ 1.0200050] pci0: i/o space, memory space enabled, rd/line, rd/mult, wr/inv o
k
[ 1.0200050] vendor 8086 product 1237 (host bridge, revision 0x02) at pci0 dev
0 function 0 not configured
[ 1.0200050] vendor 8086 product 7000 (ISA bridge) at pci0 dev 1 function 0 no
t configured
[ 1.0200050] vendor 8086 product 7010 (IDE mass storage, interface 0x80) at pc
i0 dev 1 function 1 not configured
[ 1.0200050] vendor 8086 product 7113 (miscellaneous bridge, revision 0x03) at
pci0 dev 1 function 3 not configured
[ 1.0200050] vendor 1234 product 1111 (VGA display, revision 0x02) at pci0 dev
2 function 0 not configured
[ 1.0200050] vendor 10ec product 8139 (ethernet network, revision 0x20) at pci
0 dev 3 function 0 not configured
[ 1.3000050] blake2s: self-test passed
[ 1.3000050] chacha: Portable C ChaCha
--8<---------------cut here---------------end--------------->8---

I'm aware that

https://salsa.debian.org/glibc-team/glibc/-/blob/sid/debian/patches/hurd-i386

has an odd 50 patches that we are not using in Guix...Also, we're using
a much newer rumpkernel source (latest), so it could be anything...

Again, any help or insights higly appreciated!

Greetings,
Janneke

--
Janneke Nieuwenhuizen <janneke@gnu.org> | GNU LilyPond https://LilyPond.org
Freelance IT https://www.JoyOfSource.com | Avatar® https://AvatarAcademy.com