When upgrading to libvirt 10.4-0 I can confirm that if /usr/sbin is in
the PATH , the command : " virsh -c qemu:///session connect" works ;
virt-manager can connect only after launching the virsh command.
I can confirm that downgrading libvirt and almost all related packages
to version 10.3.0-3 restores the usual behaviour.
> After upgrade to 10.4.0 I have to start my "local session" using
> virsh -c qemu:///session connect
This does not work for me ; it works only if a prepend sudo ; root is
the only user that has access to qemu:///session which is a paradox
because users like me are using qemu:///session (for
> Personally I don't see value in pw support in qemu for now.
I didn't even know that qemu has introduced pipewire support in the
latest version.
I discovered it while trying to find an alternative for the audiodev
driver because of the issue I've experienced :
- pulse+vga , pulse+virgl , pulse
Package: qemu-system
Version: 1:8.1.2+ds-1
Severity: normal
Dear Maintainer,
Qemu mainline version 8.1.2 claims supporting pipewire audio as host
audio driver , but debian packages is not compiled with that feature :
> qemu-system-amd64 -audiodev help
Available audio drivers:
none
alsa
dbus
>(Completely unrelated, but browsing the issues it popped up that encryption is
>very broken above 6.3,
>and it has been somewhat broken on older kernels too; not one people lost their
>data due to upgrading
> beyond 6.3, so beware.
"Encryption is very broken" seems too generic : it seems (to me)
A google search of 'crypto_tfm_ctx' will show that the declaration of
this function is located in 'include/crypto/algapi.h' in the linux
6.3.x series.
So this is not complicated to solve : it's enough to add '#include
' into the magma_generic.c file , after the '#include
' line.
Hi
After updating linux-image-686 with the latest update (5.17.0-2-686) ,
finally my machine successfully boots. The problem should be solved
also for other people affected by this bug.
P.S. : For other people interested , starting from linux-5.18-rc5-rt4
and beyond ( latest I've built is
Hi
I've done more tests (basically recompiled and installed 2 times the
same kernel) with linux-5.17.5 but I am still having no luck : I can't
figure out why it does not detect the controller ... I've checked the
config and the drivers for the controller (ata_piix , ata_generic )
are built but
Hi ,
reverting d6b88ce2eb9d ("ACPI: processor idle: Allow playing dead in
C3 state") made kernel version 5.15.35 booting , but still not kernel
version 5.17.5 in my case.
The last one has changed behaviour : now the kernel seems to boot - at
least it is not freezing and key combination works to
Hi ,
5.16 series is EOL so I've just continued to do tests with vanilla
linux-5.17.4 and linux-5.17.5 : both do not boot and stop at the same
point as indicated in the messages above _but_ ... something strange
is happening also with longterm 5.15 series since version 5.15.35 and
5.15.36 do not
Damien Le Moal wrote :
> I posted another patch which disables the read log command for the adapter
> instead of just for your disk. Can you try that one too ?
It did not work in my case ... and neither blacklisting the entire
ICH4 class of pci adapters worked.
Blacklisting components and/or
I've also tested linux-5.17.1 from experimental but it does not boot
as expected :)
It could be interesting to know if there is any 32 bit pentium-m based
platform with ICH4 controller that boots with any kernel in the range
from 5.16.0 to 5.17.1
Anyway I think I will remain with my custom
Hello everyone
I've tested also a patch based on
https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=1006149#260 and
modified simply by adding the disk drive model installed in my machine
to the backlist, but the final result is exactly the same ... the
computer stops booting always at the same
Hi people !
Big news here :
- I didn't apply the first patch since the disk drive installed in my
machine is exactly an "old" Western Digital :)
> hdparm -I /dev/sda
!
! /dev/sda:
!
!ATA device, with non-removable media
!Model Number: WDC WD1200BEVE-00WZT0
.
Axel Beckert wrote:
> ... but I actually have no idea how to cross-compile a
> kernel package from amd64 to i386.
The 64 bits compiler supports native build of 32 bits binaries through
the -m32 flag so you have just to install kernel build dependencies ,
multilibs support and point to the kernel
> Did you try the patch that Damien posted here:
> https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=1006149#131
Exactly ... when I write "same behaviour" I mean exactly the same :)
It stops at the same point as shown in the videos , no further
messages ... with a vanilla kernel 5.16.18 with
> As you appear to be familiar with kernel building, could you look at the
> patches that Damien Le Moal posted in this bug report?
Done ... same behaviour : as before the only way to restart the
machine is by pressing the power button ; Ctl+Alt+Del and Alt+SysReq+b
do not work.
> AFAICT, we're
linux-image-5.16.0-6-686 (5.16.18-1) still does not boot on a
different machine with the same controller.
Vanilla kernel 5.16.18 built with custom .config for the specific
machine fails to boot at the same point too.
I'm also trying to track stable patches in the stable tree queue
The same here with linux-image-5.16.0-4-686 : different machine but
the same ata controller (ata_piix).
linux-image-5.15.x were all fine ; the same with custom kernels of the
5.15.x series : custom kernels of the 5.16 series all fail to boot and
stop loading at the same point.
20 matches
Mail list logo