Public bug reported:
After upgrading to kernel image version linux-image-5.15.0-75-generic my
WiFi adaptor stops working after 5 or 10 minutes. I can get it to work
again temporarily (another 5 or 10 mins) by stop Turning Off the WiFi on
the desktop UI and Turning it On again. After booting back i
The automatically attached dmesg would be from my previous kernel which
I booted into from grub. Here's the dmesg output from boot where bug
manifested, i.e. journalctl -o short-precise -k -b -1 > /tmp/dmesg.txt
** Attachment added: "dmesg log from problematic kernel"
https://bugs.launchpad.ne
This might be fixed in a future kernel release. I see the above patches
in github (18 days old).
https://github.com/torvalds/linux/commit/8d3a6c37d50d5a0504c126c932cc749e6dd9c78f
I can see aq_vec_stop in your stack trace and I can see the above patch
addresses that frame.
That said, are you sure
Just a couple of other suggestions.
You can generate a patch file from the upstream kernel repo. eg
cd linux-jammy
curl
https://github.com/torvalds/linux/commit/8d3a6c37d50d5a0504c126c932cc749e6dd9c78f.patch
-o ./atlantic.patch
git diff
git apply ./atlantic.patch
git diff
I was wrong about pri
Actually I took a look at the aq_nic.c and it looks like there are still
places in the code that need patching to avoid UB. For example, from
your dmesg I see it pointing to this section of code
(https://github.com/torvalds/linux/blob/8d3a6c37d50d5a0504c126c932cc749e6dd9c78f/drivers/net/ethernet/aq
I think the last/3rd one might be rewritten like:
for (tc = 0U; tc < self->aq_nic_cfg.tcs; tc++) {
for (i = 0U; self->aq_vecs > i; ++i) {
aq_vec = self->aq_vec[i];
if (aq_vec) {
data += count;
Public bug reported:
The undefined behaviour sanitiser picks up an array-index-out-of-bounds
in the aquantia atlantic driver. The NIC is (I think) built into my
gigabyte motherboard
(https://www.gigabyte.com/uk/Motherboard/X399-AORUS-XTREME-
rev-10/sp#sp). This wasn't an issue before I upgrading f
My system seems to be working now. My dmesg shows the driver as loading:
[1.439880] atlantic :07:00.0 enp7s0: renamed from eth0
I had some problems with my distribution upgrade and had to re-run it,
ie "apt dist-upgrade" along with other commands.
--
You received this bug notification b
I followed instructions under the wiki
(https://wiki.ubuntu.com/Testing/EnableProposed) using the "Developer
Options" to enable pre-released updates (jammy-proposed); I then also
followed section "Selective upgrading from -proposed" to set a pin-
priority of 400. I then installed proposed kernel as
For me, when I ran journalctl (see above) the following message was
present: "kernel: PM: Image not found (code -16)". The minus sign is
prefixed to the error code and 16 is a standard linux kernel error code,
i.e.
$ grep 16 /usr/include/asm-generic/errno-base.h
#define EBUSY 16 /*
This bug returned on my machine. Not sure why it disappeared &
reappeared. Also now on 5.15.0-18-generic.
--
You received this bug notification because you are a member of Kernel
Packages, which is subscribed to linux in Ubuntu.
https://bugs.launchpad.net/bugs/1958770
Title:
Aquantia GbE LAN d
Public bug reported:
Hopefully, the automatically collected information is useful. The issue
manifests when I try to hibernate the device. I see the following in the
logs:
~$ journalctl --no-hostname -b -1 | tail -6
Feb 25 08:23:54 systemd[1]: nvidia-hibernate.service: Deactivated successfully.
F
> Is the swap space big enough?
A good point. This worked on previous versions on Ubuntu with swap
(18GB) only 50% of RAM (32GB). I recreated my /swapfile with 38GB and
the same issue persists. ie.
$ free -m
total ...
Mem: 31964
Swap: 38911
> Furthermore, the h
This doesn't seem to resolve it for me. It's been a long time since I
recompiled kernel modules, these are the steps I took:
# module compile instructions from
https://wiki.ubuntu.com/Kernel/SourceCode
$ uname -r
5.15.0-18-generic
$ git clone
git://git.launchpad.net/~ubuntu-kernel/ubuntu/+source
Oh I forgot to add that i also applied the patch before compiling.
--
You received this bug notification because you are a member of Kernel
Packages, which is subscribed to linux in Ubuntu.
https://bugs.launchpad.net/bugs/1958770
Title:
Aquantia GbE LAN driver causes UBSAN error during kernel
Patch as file.
** Patch added: "aq_nic.patch"
https://bugs.launchpad.net/ubuntu/+source/linux/+bug/1958770/+attachment/5565904/+files/aq_nic.patch
--
You received this bug notification because you are a member of Kernel
Packages, which is subscribed to linux in Ubuntu.
https://bugs.launchpad
This new patch doesn't appear to resolve issue either. However, I did
get it to work if I extended your technique to aq_nic.c. See following
comment with patch.
Then I see the following in my dmesg:
[ 2991.604548] atlantic :07:00.0 enp7s0: renamed from eth0
--
You received this bug notifica
A potential (temporary) workaround is to remove the offending symlinks
from package file as follows:
mkdir linux-firmware
cd linux-firmware
cp /var/cache/apt/archives/linux-firmware_20220314.gitcd01f857-0ubuntu1_all.deb
./
ar xf linux-firmware_20220314.gitcd01f857-0ubuntu1_all.deb
zstd -d data.t
I have the same problem. Curiously, the patch previously submitted
should fix those code lines reported in the UBSAN report
https://launchpadlibrarian.net/601133815/aq_nic.c.patch
--
You received this bug notification because you are a member of Kernel
Packages, which is subscribed to linux-oem-
I had another look at my hibernation settings and realised my resume
offset was wrong in my /etc/default/grub. After updating this with the
value from command (found on
https://wiki.archlinux.org/title/Power_management/Suspend_and_hibernate):
~$ sudo filefrag -v /swapfile | awk '$1=="0:" {print su
20 matches
Mail list logo
