-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA256

On Fri, Jan 25, 2019 at 01:58:59PM +0100, Patrik Hagara wrote:
> On 1/24/19 5:18 PM, Patrik Hagara wrote:
> > On 1/20/19 1:57 AM, Marek Marczykowski-Górecki wrote:
> >> Hi all,
> >>
> >> There is updated "kernel" package available in current-testing
> >> repository - it's a Linux long term support 4.19.x series, as an update
> >> over 4.14.x before. Since the upgrade switches to the next major LTS
> >> branch, I'll keep it in current-testing repository longer than usual 1-2
> >> weeks. This also applies to kernel package for VMs: kernel-qubes-vm.
> >> Please report new issues the usual way, at qubes-issues[1], or
> >> simply by replying here. In either case, please mark it clearly it
> >> happens after updating to 4.19, preferably including a link to the
> >> update:
> >> https://github.com/QubesOS/updates-status/issues/850
> >>
> >> 4.19.x kernel was already available as kernel-latest package for some
> >> time. Users of kernel-latest will see the update to 4.19.15 too, but
> >> kernel-latest soon will carry 4.20.x kernel version.
> >>
> >> [1] https://github.com/QubesOS/qubes-issues/issues
> >>
> >>
> > 
> > I get weird graphical artifacts with the new kernel after ~an hour of
> > usage. Windows from AppVMs turn all white sometimes when switching
> > workspaces in i3wm. Events like mousing over an interactive table rows
> > in a browser (when the current row gets highlighted) return that
> > particular section of the window back to normal (but not the whole
> > window, for that I need to trigger a repaint of the whole window by eg.
> > making it full-screen and immediately switching back to non-full-screen).
> 
> The only error message I've been able to find so far is in dom0 Xorg log:
> 
> > (EE) intel(0): Failed to submit rendering commands (Bad address),
> disabling acceleration.

This is very likely related. Normally I'd say "Bad address" indicate
user-space issue, but the only thing changed is the kernel version... It
may be also that some kernel API have changed and the driver is using
parts that weren't there before.

Anyway, I've looked into 'intel' X driver sources and the version we
currently have (2.99.917) is the latest one. On the other hand, there
was over 800 commits since that release and some of them may be related.
For example maybe this: https://bugs.freedesktop.org/show_bug.cgi?id=105886

This suggests you may want to try enabling or disabling composition, if
i3wm supports it.

> Duckduckgo-ing the error message yielded a few [1][2] Arch Linux bug
> reports describing the same symptoms. The first bug report also has a
> kernel patch [3] linked, which supposedly fixes the issue (haven't tried
> it).

That patch is from 2014, already included in 3.19+

> [1] https://bugs.archlinux.org/task/43143
> [2] https://bugs.archlinux.org/task/55732
> [3]
> https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?id=d472fcc8379c062bd56a3876fc6ef22258f14a91
> 
> Cheers,
> Patrik

- -- 
Best Regards,
Marek Marczykowski-Górecki
Invisible Things Lab
A: Because it messes up the order in which people normally read text.
Q: Why is top-posting such a bad thing?
-----BEGIN PGP SIGNATURE-----

iQEzBAEBCAAdFiEEhrpukzGPukRmQqkK24/THMrX1ywFAlxLGisACgkQ24/THMrX
1yyRKwf9G9TX89Bh2aePabdq7k40zDEHK68sKmsbL7xcm0JpfsdXHK/MuM+B4AyJ
BT7PrEIr8n1wXc++EArbtwapIPldICAhnBRK4fFdazHmtgAeW5S1GztAFisa4EaD
w0AWDoLLVg4DR7AcwFi1EXse4jgT0/CSYkHIENM0QRl4uevEV6lKlpN4lS8Rgjm8
cUCXajC5RCLT3RVDUTzufUOxLLt/syRzGVtBsgJqCwvVdnOxArZqlEJgSI7wq9lN
HR6OSv1ETGdlubxegn2LsAtqLvHXD+vnV11hgT4EvSZhHTfcbOI8FJdsnU8YxNY1
vsHV3L772QpTm3+jZ05X8AxJLEYrHA==
=XaRm
-----END PGP SIGNATURE-----

-- 
You received this message because you are subscribed to the Google Groups 
"qubes-users" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to qubes-users+unsubscr...@googlegroups.com.
To post to this group, send email to qubes-users@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/qubes-users/20190125141611.GH1429%40mail-itl.
For more options, visit https://groups.google.com/d/optout.

Reply via email to