** Also affects: linux-lowlatency (Ubuntu Lunar)
Importance: Undecided
Status: New
** Also affects: linux-lowlatency (Ubuntu Jammy)
Importance: Undecided
Status: New
** Changed in: linux-lowlatency (Ubuntu Jammy)
Assignee: (unassigned) => gerald.yang (gerald-yang
Thanks Jay for pointing this out!
I just read the man page of vdso, it says gettimeofday is not a real system
call and just read the shared memory exports by kernel.
It shouldn't be used to measure the user-kernel context switch overhead caused
by NO_HZ_FULL.
>From what the tests does, I think
Some observations from test results between NO_HZ_FULL built-in but not enable
and default kernel
Tests are from LTP scheduling related under "realtime" category
And there is "no" taskset when running the tests
- Gettimeofday latency (ns basis)
For no_hz_full built-in:
The average is almost the
Another note is that NO_HZ_FULL is already built-in on 6.5 "lowlatency" kernel:
https://bugs.launchpad.net/ubuntu/+source/linux-lowlatency/+bug/2023007
But currently it's only available on Mantic, I think we should also
consider if it's more proper for lowlatency kernel instead of generic
--
The attached is ltp test results with NO_HZ_FULL built-in and activate in
kernel cmdline, e.g.
isolcpus=2-15,18-31 nohz_full=2-15,18-31 rcu_nocbs=2-15,18-31
but tests were run on non-activate CPU 16
** Attachment added: "nohz-log-noisolcpu.tgz"
The attached is ltp test results with NO_HZ_FULL built-in and activate on
kernel cmdline, e.g.
isolcpus=2-15,18-31 nohz_full=2-15,18-31 rcu_nocbs=2-15,18-31
tests were run on cpu 4
** Attachment added: "nohz-log-isolcpu.tgz"
The attached is ltp test results with NO_HZ_FULL built-in but not
actiavte
** Attachment added: "nohz-log.tgz"
https://bugs.launchpad.net/ubuntu/+source/linux/+bug/1919154/+attachment/5713512/+files/nohz-log.tgz
--
You received this bug notification because you are a member of Kernel
The attached is ltp test results with default ubuntu kernel
** Attachment added: "log.tgz"
https://bugs.launchpad.net/ubuntu/+source/linux/+bug/1919154/+attachment/5713511/+files/log.tgz
--
You received this bug notification because you are a member of Kernel
Packages, which is subscribed
The above tests is based on getpid() system call, which doesn't have
much workload except context switch, so we evaluate the additional
overhead caused by NO_HZ_FULL built-in on AMD EPYC machine
I also used LTP to run scheduler related tests, will attach the test
data later
--
You received this
on 6.2 kernel
With NO_HZ_FULL built-in, the context switch performance is ~4% worse
than default config
--
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/1919154
Title:
Enable
Run some tests with Jammy hwe kernel (6.2.0) on AMD EPYC 7252
Test case 1, without NO_HZ_FULL built-in (default ubuntu kernel config):
Run test program 4 times without taskset
tail -n 2 log/notaskset.*
==> log/notaskset.1 <==
total 23703299350 nsec
avg 237 nsec
==> log/notaskset.2 <==
total
Spend some time to get access to AMD EPYC machine
For 5.15 kernel
If we build NO_HZ_FULL into kernel and compare with default kernel (NO_HZ_IDLE)
- on Intel machine, there is no much different
- on Arm64 machine, interestingly, the context switch is a bit faster with
NO_HZ_FULL built-in
- on
On AMD EPYC 7252
Hardware configs:
AMD64
32 CPUs
128G RAM, numa nodes: 2
Software configs:
OS: ubuntu 20.04
Official kernel: 5.15 hwe (5.15.0-86.96~20.04.1)
Test kernel: 5.15 hwe (5.15.0-86.96~20.04.1+test20231013b0)
https://launchpad.net/~gerald-yang-tw/+archive/ubuntu/focal-no-hz-full
Test
On arm64 machine
Hardware configs:
Aarch64
128 CPUs
502G RAM, numa nodes: 4
Software configs:
OS: ubuntu 20.04
Official kernel: 5.15 hwe (5.15.0-86.96~20.04.1)
Test kernel: 5.15 hwe (5.15.0-86.96~20.04.1+test20231013b0)
https://launchpad.net/~gerald-yang-tw/+archive/ubuntu/focal-no-hz-full
Test
Results on Intel machine
Hardware configs:
Dell PowerEdge R730xd
Intel(R) Xeon(R) CPU E5-2650 v3 @ 2.30GHz
40 CPUs
188G RAM, numa nodes: 2
Software configs:
OS: ubuntu 20.04
Official kernel: 5.15 hwe (5.15.0-86.96~20.04.1)
Test kernel: 5.15 hwe (5.15.0-86.96~20.04.1+test20231013b0)
The attached test code is borrowed from Jay, it measures the average
time for running getpid() 1 times to see if NO_HZ_FULL will
cause any performance degradation for context switch when it’s built
into kernel
There are 4 test scopes:
1. Without NO_HZ_FULL built-in
the default value of
Test program from Jay
** Attachment added: "syscall-time.c"
https://bugs.launchpad.net/ubuntu/+source/linux/+bug/1919154/+attachment/5710594/+files/syscall-time.c
--
You received this bug notification because you are a member of Kernel
Packages, which is subscribed to linux in Ubuntu.
jammy hwe kernel (6.2) with NO_HZ_FULL:
https://launchpad.net/~gerald-yang-tw/+archive/ubuntu/no-hz-full-6.2
--
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/1919154
Title:
Enable
Since we have some customers need NO_HZ_FULL, I'd like to provide some
updates and progress:
I've borrowed some machines including intel, AMD EPYC, arm64 servers
and now running some tests on a test kernel with
1. CONFIG_NO_HZ_FULL=y
2. not enable nohz_full in kernel cmdline
to evaluate if there
** Changed in: linux (Ubuntu Jammy)
Status: New => In Progress
** Changed in: linux (Ubuntu Lunar)
Status: New => In Progress
** Changed in: linux (Ubuntu Jammy)
Assignee: (unassigned) => gerald.yang (gerald-yang-tw)
** Changed in: linux (Ubuntu Lunar)
Assignee: (u
** Description changed:
[ Impact ]
When running fio on a NVME on an AWS test instance with 5.19 kernel, IOs
get stuck and fio never ends
fio command:
sudo fio --name=read_iops_test --filename=/dev/nvme1n1 --filesize=50G
--time_based --ramp_time=2s --runtime=1m
** Tags added: sts sts-sru-needed
--
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/2022318
Title:
[SRU] IO gets stuck on 5.19 kernel
Status in linux package in Ubuntu:
In Progress
** Description changed:
[ Impact ]
When running fio on a NVME on an AWS test instance with 5.19 kernel, IOs
get stuck and fio never ends
fio command:
sudo fio --name=read_iops_test --filename=/dev/nvme1n1 --filesize=50G
--time_based --ramp_time=2s --runtime=1m
** Changed in: linux (Ubuntu Kinetic)
Status: New => In Progress
** Changed in: linux (Ubuntu Kinetic)
Importance: Undecided => High
** Changed in: linux (Ubuntu Kinetic)
Assignee: (unassigned) => gerald.yang (gerald-yang-tw)
** Description changed:
[ Impact ]
** Description changed:
[ Impact ]
When running fio on a NVME on an AWS test instance with 5.19 kernel, IOs
get stuck and fio never ends
fio command:
sudo fio --name=read_iops_test --filename=/dev/nvme1n1 --filesize=50G
--time_based --ramp_time=2s --runtime=1m
used in blk-mq in block layer, scsi drivers and fungible
ethernet driver
If there is any issue happens in sbitmap, the symptom should be IO hung, or
packets get stuck in fungible driver
** Affects: linux (Ubuntu)
Importance: High
Assignee: gerald.yang (gerald-yang-tw)
Status:
This has been verified on 5.4.0-135.152 generic kernel and also released to
focal-updates
see comment #28
Set to verification done
** Tags removed: verification-needed-focal
** Tags added: verification-done-focal
--
You received this bug notification because you are a member of Kernel
I've tested linux-azure/5.15.0-1030.37 in jammy-proposed and confirmed
it has fixed the issue
gerald@myvm-j1:~$ uname -a
Linux myvm-j1 5.15.0-1030-azure #37-Ubuntu SMP Mon Dec 12 19:17:45 UTC 2022
aarch64 aarch64 aarch64 GNU/Linux
all VFs can be detected and VM boot successfully
** Tags
Hi Tim,
I've tested kinetic azure kernel and it works perfectly
but when I check jammy azure kernel, this patch seems not in the git tree
has it been merged to the jammy azure kernel too?
Thanks,
Gerald
--
You received this bug notification because you are a member of Kernel
Packages, which is
I've tested linux-azure/5.19.0-1014.15 in kinetic-proposed and confirmed it
fixes the issue
uname -a
Linux myvm-k 5.19.0-1014-azure #15-Ubuntu SMP Mon Nov 28 19:54:09 UTC 2022
aarch64 aarch64 aarch64 GNU/Linux
steps:
create an ARM64 VM on azure
create 8 VFs and attach to the VM
with
** Changed in: linux (Ubuntu)
Status: In Progress => Invalid
** Changed in: linux (Ubuntu Focal)
Status: In Progress => Invalid
** Changed in: linux (Ubuntu Jammy)
Status: In Progress => Invalid
** Changed in: linux (Ubuntu Kinetic)
Status: In Progress => Invalid
--
could occur]
The patch is to use the correct type for the variables and reviewed by upstream
There shouldn't be any problem
** Affects: linux (Ubuntu)
Importance: High
Assignee: gerald.yang (gerald-yang-tw)
Status: In Progress
** Affects: linux (Ubuntu Focal)
Importance
Add one more case caused by this issue:
When SEV is enabled on the host, the guest also enables SEV and running 5.15
kernel
with this commit:
https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?id=e998879d4fb7991856916972168cf27c0d86ed12
The SWIOTLB driver in the guest
/linux/+bug/1842320
** Affects: grub2-unsigned (Ubuntu)
Importance: High
Assignee: gerald.yang (gerald-yang-tw)
Status: In Progress
** Package changed: linux (Ubuntu) => grub2-unsigned (Ubuntu)
** Changed in: grub2-unsigned (Ubuntu)
Assignee: (unassigned) => geral
This should be the minimal set of patches needed for hibernation on 5.15
aws kernel
270756abd4ee x86: Disable KASLR when Xen is detected
7fea4363d9ed xen: Restore xen-pirqs on resume from hibernation
90c68a4dc544 xen-netfront: call netif_device_attach on resume
145e43c7bcb1 xen: Only restore the
We are hitting the hibernation issue on Jammy aws kernel (5.15), it gets
stuck when hibernating
Turn out it needs some patches from 5.15 amazonlinux to make hibernation work
https://github.com/amazonlinux/linux/tree/amazon-5.15.y/mainline
--
You received this bug notification because you are a
** Changed in: linux (Ubuntu)
Status: Invalid => Fix Committed
--
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/1965723
Title:
audit: improve audit queue handling when "audit=1" on
Thanks a lot for the help Kleber
--
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/1965723
Title:
audit: improve audit queue handling when "audit=1" on cmdline
Status in linux package in
the fix in Impish is still in progress here LP: #1966021
--
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/1965723
Title:
audit: improve audit queue handling when "audit=1" on cmdline
This one has been merged in the last upstream stable patch set
** No longer affects: linux (Ubuntu Impish)
** No longer affects: linux (Ubuntu Focal)
** No longer affects: linux (Ubuntu Bionic)
** Changed in: linux (Ubuntu)
Status: In Progress => Invalid
--
You received this bug
** No longer affects: linux (Ubuntu Jammy)
--
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/1965723
Title:
audit: improve audit queue handling when "audit=1" on cmdline
Status in linux
** Also affects: linux (Ubuntu Jammy)
Importance: Critical
Assignee: gerald.yang (gerald-yang-tw)
Status: In Progress
** Also affects: linux (Ubuntu Bionic)
Importance: Undecided
Status: New
** Also affects: linux (Ubuntu Impish)
Importance: Undecided
Status
Importance: Critical
Assignee: gerald.yang (gerald-yang-tw)
Status: In Progress
** Affects: linux (Ubuntu Bionic)
Importance: Critical
Assignee: gerald.yang (gerald-yang-tw)
Status: In Progress
** Affects: linux (Ubuntu Focal)
Importance: Critical
Assignee: gera
w => In Progress
** Changed in: linux (Ubuntu Bionic)
Assignee: (unassigned) => gerald.yang (gerald-yang-tw)
** Changed in: linux (Ubuntu Groovy)
Assignee: (unassigned) => gerald.yang (gerald-yang-tw)
** Changed in: linux (Ubuntu Focal)
Assignee: (unassigned) => gerald.ya
** Description changed:
+ === SRU Justification ===
+ [Impact]
+ To get better performance for dm-crypt in some cases, bypass kcryptd
workqueue can reduce the overhead in context switch between workers,
cherry-pick commit 39d42fa96ba1b7d2544db3f8ed5da8fb0d5cb877 from mainline
kernel, and this
** Changed in: linux (Ubuntu)
Assignee: (unassigned) => gerald.yang (gerald-yang-tw)
** Changed in: linux (Ubuntu)
Status: Confirmed => In Progress
--
You received this bug notification because you are a member of Kernel
Packages, which is subscribed to linux in Ubuntu.
The new kernel in xenial-proposed fixes this issue, i40evf is included in
linux-modules
kernel version: 4.4.0-171
old in xenial-update:
dpkg -L linux-modules-4.4.0-170-generic | grep 'i40e'
/lib/modules/4.4.0-170-generic/kernel/drivers/net/ethernet/intel/i40e
** Also affects: linux (Ubuntu Xenial)
Importance: Undecided
Status: New
** Changed in: linux (Ubuntu Xenial)
Assignee: (unassigned) => gerald.yang (gerald-yang-tw)
** Changed in: linux (Ubuntu Xenial)
Status: New => In Progress
--
You received this bug notifi
** Tags added: sts
--
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/1852663
Title:
i40e: general protection fault in i40e_config_vf_promiscuous_mode
Status in linux package in Ubuntu:
Undecided
Assignee: gerald.yang (gerald-yang-tw)
Status: In Progress
** Affects: linux (Ubuntu Eoan)
Importance: Undecided
Assignee: gerald.yang (gerald-yang-tw)
Status: In Progress
** Also affects: linux (Ubuntu Eoan)
Importance: Undecided
Status: New
** Changed
** Changed in: linux (Ubuntu Eoan)
Assignee: (unassigned) => gerald.yang (gerald-yang-tw)
** Changed in: linux (Ubuntu Disco)
Assignee: (unassigned) => gerald.yang (gerald-yang-tw)
** Changed in: linux (Ubuntu Disco)
Status: New => In Progress
** Changed in: linux (Ub
** Changed in: linux (Ubuntu Bionic)
Status: New => In Progress
--
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/1848481
Title:
cloudimg: no iavf/i40evf module so no network
** Description changed:
- For example with bionic, i40evf module is in linux-modules-
- extra-4.15*-generic. However, cloudimg doesn't have linux-modules-extra
- seeded:
+ === SRU Justification ===
+ [Impact]
+ Intel's VF drivers such as igbvf and ixgbevf are already available in
+ generic. It
** Changed in: linux (Ubuntu Bionic)
Assignee: (unassigned) => gerald.yang (gerald-yang-tw)
--
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/1848481
Title:
cloudimg: no iavf/i40
** Also affects: livecd-rootfs (Ubuntu Bionic)
Importance: Undecided
Status: New
** Also affects: linux (Ubuntu Bionic)
Importance: Undecided
Status: New
** Also affects: livecd-rootfs (Ubuntu Eoan)
Importance: Undecided
Status: New
** Also affects: linux (Ubuntu
Verified linux-image-4.4.0-100-generic on AS7712, it works find,
all broadcom modules are loaded and bcm0 interface is created successfully.
--
You received this bug notification because you are a member of Kernel
Packages, which is subscribed to linux in Ubuntu.
@Zygmunt
So does it mean even I add a configure hook in meta/hooks/ to copy those
binaries to boot partition (/boot/uboot/),
there is no such interface in snapd to allow this?
Thanks
--
You received this bug notification because you are a member of Kernel
Packages, which is subscribed to linux
I am seeing this issue too, but in my case, it seems there is something
wrong in bbswitch-dkms, if I remove it manually by 'sudo rm
/lib/modules/4.4.0-xx-generic/updates/dkms/bbswitch.ko' I can boot into
system with both intel and nvidia mode (prime-select intel or prime-
select nvidia).
Since
58 matches
Mail list logo
