The 4.14.4 stable kernel rebase contains new features, new hardware support, and a number of fixes across the tree.
sudo dnf upgrade --refresh --advisory=FEDORA-2017-63b75cfe71Please login to add feedback.
| 0 | 1 | Test Case kernel regression |
This update has been submitted for testing by jcline.
works for me on several x86_64 pyhsical machines and on a dozen of VMware guests as well as VMware Workstation 14 running on top
Seems to work just fine on x86_64 workstation.
karma: +1
everything but apache trafficserver works - https://github.com/apache/trafficserver/issues/2908 - while this is sure a bug in trafficserver the kernel "don't break userspace" rule shouldn't lead to such a issue given that ATS 7.1.1 is out now for a longer time - for now excluded 4.14 from the proxy machine, vereything else works fine
still having sound distortions and scratches in sound output with this kernel.
This update has been pushed to testing.
System frequently (but not 100%) fails to reboot. Attempting to do so results in a kernel panic as shown here https://imgur.com/h1S903b . The freeze is such that the reset button is non-operative, requiring a hard ACPI shutdown. This being a Ryzen system I have seen this hard freeze before, but that was idle death, whereas this is triggered, not consistently, but >50%. The only two times a reboot was successful was when the reboot was initiated very soon after startup. This might be related to the time for the network bridge I have set up to finish its learning phase.
Is this problem present in the 4.13 kernels, or is it new in 4.14?
It looks like that's a null pointer dereference in cgroups, but it's difficult to make everything out in that picture. Could you file a bug report with a clearer picture?
Thanks for the feedback!
work fine on x86_64
no regression noted on my x86_64 machine.
This update has been submitted for batched by jcline.
This update has been submitted for stable by jcline.
this kernel release caused two machines (different motherboards) with NViDIA graphics cards to hang during boot when trying to start Gnome manager. Had to revert ot 4.13
karma: -1
"with NViDIA graphics cards" -> closed source blob driver? if yes - don't matter at all for Fedora - saif from a Fedor auser for over a decade which don't by known problemartic hardware - you can't have a blleding edge distribution and closed source drivers at the same time without problems - other hardware or LTS distribution is the solution
This update has been pushed to stable.
Works, but this new extra error printing in drm/amdkfd
https://patchwork.freedesktop.org/patch/171967/
brought up: kfd kfd: kgd2kfd_probe failed
Any hints about this? AMD E-450 APU with Radeon HD 6320 GPU.
Reboot bug reported to https://bugzilla.redhat.com/show_bug.cgi?id=1525279
I'm wondering what happened to the packaging guidelines, though. 4.14.4 pushed to F26 stable when F27 has 4.14.3, and 4.14.4 not even in testing for F27.
I'm fully aware that those are in fact "different packages" for the different branches, but it breaks (or at least complicates) any F26->F27 upgrades, which is why packaging guidelines forbid it.
besides that you can't sync karma based updates nor control mirrors are in sync and hence -1 is not deserved:
maybe it's time to fix that bunch of recommended crap for dist-upgrades when "dnf/yum --releasever=27 distro-sync" works for ages without rely on the package needs to have a newer releasever and in doubt you boot with the F26 kernel without any issues
that would also solve the ton of stupid updates within a release cycle doing nothing other than sync the release tag of a package without any real change
mjg: You're right, and I shouldn't have pushed it. I'm sorry.
For what it's worth, 4.14.5 is queued for F27 stable today which should fix the upgrade path.
on i686 java 8 breaks it. observed when for e.g I was installing bunch of plugins on netbeans.
kernel 1.14 breaks my onboard audio system (alsa.name = "AD1989B Analog", alsa.card_name = "HDA Intel") I hear an ugly noise in my speakers.
Hi raveit65,
You're very likely hitting https://bugzilla.redhat.com/show_bug.cgi?id=1525104. Hans would probably like to know what brand and model of the hardware (motherboard if not a laptop) you are using, so if you could let us know on that bug (or here) that'd be great.
The configuration has been changed back in v4.14.6 which should be on its way to testing for F26 (and already is in testing for F27), and you can also turn power saving back off by passing snd_hda_intel.power_save=0 on the kernel command line which should fix the issue.
Thank you for the feedback!
@jcline i answered here https://bugzilla.redhat.com/show_bug.cgi?id=1525104#c23