FEDORA-2019-7aecfe1c4b

security update in Fedora 30 for kernel, kernel-headers, & 1 more

Status: stable 2 months ago

The 5.2.5 stable kernel update contains a number of important fixes across the tree. This should also fix the black screen issue that several i915 users were hitting with 5.2.4.


The 5.2.4 kernel rebase contains new features and hardware support, and a number of important fixes across the tree.

Reboot Required

After installing this update it is required that you reboot your system to ensure the changes supplied by this update are applied properly.

How to install

sudo dnf upgrade --advisory=FEDORA-2019-7aecfe1c4b

Comments 39

This update has been submitted for testing by jforbes.

This update's test gating status has been changed to 'waiting'.

This update has obsoleted kernel-tools-5.2.4-200.fc30, and has inherited its bugs and notes.

This update's test gating status has been changed to 'ignored'.

works for me: laptop 8GB Intel i5-2520M CPU Lenovo T420 - using the Mate Desktop Environment (surprisingly, the elapsed time for the performance test was under ten minutes, unsusally it is over a 100 minutes!)

karma: +1 critpath: +1 kernel regression: +1

works for me: laptop 8GB Intel i5-2520M CPU Lenovo T420 - using the Mate Desktop Environment (surprisingly, the elapsed time for the performance test was under ten minutes, most often it is over a 100 minutes!) -- had to correct an annoying typo in previous comment

karma: +1 critpath: +1 kernel regression: +1

Black screen issue on Skylake is indeed fixed for me, thanks. I'll be back with karma after I've used this kernel a bit more.

karma: +1 critpath: +1 kernel regression: +1

Sorry no. Still suspend and restart issues. desktop Ryzen & Nvida (Negativo17 driver) - last good kernel 5.1.19

karma: -1 critpath: -1

This update has been pushed to testing.

bare metal x86_64, i5-7200U w/embedded GPU, HW model HP 850 G4, KDE, VPN IPsec, VirtualBox 6.0.10, podman 1.4.4.

karma: +1 critpath: +1 kernel regression: +1

This update can be pushed to stable now if the maintainer wishes

Bridge inteface configured with systemd-networkd in off state, when booted kernel 5.1.20 this interface is configured.

karma: -1 critpath: -1

Nucleo: networkd issue is caused by systemd's issue: https://bugzilla.redhat.com/show_bug.cgi?id=1718192

T450s and XS35GTv2.

karma: +1 critpath: +1

LGTM as always. NVIDIA GTX1060. :)

karma: +1 critpath: +1 kernel regression: +1

Unfortunately without #1718192 fixed in systemd for Fedora 30, this update is a show stopper.

critpath: -1

jforbes edited this update.

All good on a i5-6200U.

karma: +1 critpath: +1

test on baremetal and in a VM

1.) T440s, all intel hardware: This kernel update works fine for me, no regression noted. Passed default regression test.

2.) VirtualBox guest (host is Win 10 on Dell Vostro): This update works fine, no regression noted.

karma: +1 kernel regression: +1

I can confirm the blackscreen issue on my T450s is fixed with this version. Works fine now :)

karma: +1 critpath: +1

works on baremetal Intel i5-3550 with prop. nvidia driver from rpmfusion (GTX980 card), VirtualBox 6.0.10 from Oracle and Gnome-Desktop. Regression test passed (paxtest skipped).

karma: +1 critpath: +1 kernel regression: +1

All Good.

karma: +1 critpath: +1 kernel regression: +1

Not a big issue but I still have the problem of logging back in to my Plasma DE after logging out. Dropping to a shell I can get around this. Never had this with 5.1.x FWIW nVidia card negativo17 RPM's and I still have not found anything of use in the logs.
Apart from that all looks good and the systems shuts down OK now.

karma: -1

i7-9700K(UHD Graphics 630): Kernel regression test case passed. Works fine for daily use. Raspberry Pi 3 B+: Kernel regression test case passed.

karma: +1 critpath: +1 kernel regression: +1
karma: +1 critpath: +1

suspend to RAM fails and spams the log with pcieconf receiving Spurious native PME interrupts (last working in 5.1.19-300.fc30; 5.1.20-300.fc30 also regressed, but in a different way)

critpath: -1

suspend to RAM fails and spams the log with pcieconf receiving Spurious native PME interrupts (last working in 5.1.19-300.fc30; 5.1.20-300.fc30 also regressed, but in a different way)

I added in a bug report to Red hat on this. If you have anything new to add, you are welcome to (and it looks like you might because I have only tested against what I know repro's the problem) https://bugzilla.redhat.com/show_bug.cgi?id=1735786

Works on a set of different machines here, most of them with the Nvidia prop. stuff.

karma: +1 critpath: +1

Obtaining a vanilla v5.2.5 from stable/linux-5.2.y upstream and running git revert 5817d78eba34f6c86f5462ae2c5212f80a013357 then installing things including nvidia 430.40 driver (fedora-multimedia) through dkms gives me a working kernel that survives 2 suspend/resume cycles with S3 (STR) sleep.

commit 5817d78eba34f6c86f5462ae2c5212f80a013357
Author: Mika Westerberg <mika.westerberg@linux.intel.com>
Date:   Wed Jun 12 13:57:38 2019 +0300

    PCI: Add missing link delays required by the PCIe spec

    [ Upstream commit c2bf1fc212f7e6f25ace1af8f0b3ac061ea48ba5 ]

    Currently Linux does not follow PCIe spec regarding the required delays
    after reset. A concrete example is a Thunderbolt add-in-card that
    consists of a PCIe switch and two PCIe endpoints:

      +-1b.0-[01-6b]----00.0-[02-6b]--+-00.0-[03]----00.0 TBT controller
                                      +-01.0-[04-36]-- DS hotplug port
                                      +-02.0-[37]----00.0 xHCI controller
                                      \-04.0-[38-6b]-- DS hotplug port

    The root port (1b.0) and the PCIe switch downstream ports are all PCIe
    gen3 so they support 8GT/s link speeds.

[...]

This kernel breaks nftables rules loading. With kernel 5.1.20-300.fc30.x86_64 I can just add rule "nft add rule ip nat prerouting tcp dport 2222 counter redirect to :22" but with 5.2.5 this don't work

Error: Could not process rule: No such file or directory

add rule ip nat prerouting tcp dport 2222 counter redirect to :22

Only after manual loading kernel module nft_redir this rule can be added.

karma: -1 critpath: -1

VLC (from rpmfusion), no longer plays video (x264 and x265) on Wayland through Intel integrated graphics with the error: libva error: va_getDriverName() failed with unknown libva error,driver_name=(null).

As this works on Xorg and nvidia's proprietary driver, as well as because of the recent work to i915, I suspect that this is related to the kernel.

karma: +1 critpath: -1

Works ok on T480s, i5, iGPU

karma: +1 critpath: +1 kernel regression: +1

Tested on the following devices:

ARMv7: Cubietruck, Hummingboard2, Panda-ES, RPi2, RPi3B

aarch64: Mustang, Pine64, Rock960, RPi3B+, RPi3A+

x86_64: UpSquared

karma: +1 critpath: +1 #1733874: +1 #1734242: +1 #1735630: +1 #1735631: +1 kernel regression: +1

Working on Toshiba P50 laptop (Intel i7-4725HQ) with Intel 8265 wifi.

karma: +1 critpath: +1 kernel regression: +1

Works well on Lenovo T460p

karma: +1 critpath: +1

This update has been submitted for stable by jforbes.

This update has been pushed to stable.

Add Comment & Feedback

Please login to add feedback.

Content Type
RPM
Status
stable
Test Gating
Submitted by
Update Type
security
Update Severity
medium
Karma
+15
stable threshold: 3
unstable threshold: -3
Autopush (karma)
Disabled
Autopush (time)
Disabled
Dates
submitted 3 months ago
in testing 2 months ago
in stable 2 months ago
modified 2 months ago

Related Bugs 4

0+1 #1733874 CVE-2019-10207 kernel: null-pointer dereference in hci_uart_set_flow_control
0+1 #1734242 CVE-2019-10207 kernel: null-pointer dereference in hci_uart_set_flow_control [fedora-all]
0+1 #1735630 CVE-2019-13648 kernel: denial of service in arch/powerpc/kernel/signal_32.c and arch/powerpc/kernel/signal_64.c via sigreturn() system call
0+1 #1735631 CVE-2019-13648 kernel: denial of service in arch/powerpc/kernel/signal_32.c and arch/powerpc/kernel/signal_64.c via sigreturn() system call [fedora-all]

Automated Test Results

Test Cases

0+11 Test Case kernel regression