I'm glad to invide Shuai Xue <xueshuai(a)linux.alibaba.com> to be a
maintainer of ANCK perf subsystem.
Shuai is working on linux perf subsystem for 2 years, and show great
interest in kernel technologies. Here are some highlight contributions
from Shuai:
1. Design and implement CMN, DDR and PCIe PMU driver for performance
monitor, and is Alibaba PMU maintainer in Linux community now.
2. Design and verify synthetic events with tools/perf for functional
debugging to facilitate system maintenance and support the wide
application of these events.
3. Backport and enable SPE features for perf-mem, perf-c2c and
perf-spe-c2c and apply SPE to optimise application performance, e.g.
SPECjbb2015.
4. Proficient in problem analysis, and skilled at using ftrace, kprobe,
tracepoint,PMU, etc.
Working with Shuai is efficient and makes a sense of achievement. Let us
hope it to happen soon.
Hi, All
I've been working on linux kernel memory over the past few years,
making some
contributions in the field and alway helping review the related code.
Here are some highlight contributions from myself:
1、Implement the cold slab design and development. It aims to reclaim
the kernel slab memory to
reduce the unexpected stability issues and saving some memory
space. The feautre has widely
applied to alibaba group.
2、Implement the swap high feature in cgroup v1 and make use of oomd to
proactive kill in the process
rather than trigger the Oom killer in the kernel, which aims to
decrease the risk of unexpected hang
or Oom. hence the user can control when to kill and what process
to kill.
The ability has been applied to ASI control components.
3、 Design and implement the performance optimization to reduce the cost
for redis snapshot.
The performance benifits is that the whole cost from second to
millisecond.
4、Design and implement the dup page in the early stage. It's proved
that has a great improvments when
the app will access the remote binary program frequently.
5、Backport mglru feature to make the page to identify the cold and hot
and implement proactiving the anonymous page with
zswap and file page to reclaim within the reasonable refault value.
Flink has take use of the alibity in the product environment to
achieve the goals of saving the memory about at least 5% only
in page cache.
With the above key contributions, I think I am qualified with the job.
Thanks
I've checked the code, yes, now anck 4.19 supports pidfd_open but not
pidfd_send_signal.
When cherry-pick features from upstream, it depends on requirements.
So if you want anck 4.19 supports pidfd_send_signal, please submit a
requirement, thanks.
Joseph
On 7/25/22 5:20 PM, lywjk(a)outlook.com wrote:
> 内核：Linux g2-alive-hz-03 4.19.91-26.an8.x86_64 #1 SMP Tue May 24 13:10:09 CST 2022 x86_64 x86_64 x86_64 GNU/Linux
> AnolisOS 8.4 ANCK内核，8.6也一样。
>
> crun开发者的回复：
> I am fine to add a workaround, but I'd like to understand what is going on.
> What system are you running?
> It seems your kernel has support for pidfd_open but not for pidfd_send_signal.
> pidfd_open was added to Linux v5.3 while pidfd_send_signal to v5.1
>
> 是否是ANCK内核添加了5.3的特性又没有添加5.1的特性呢？
> _______________________________________________
> Anolis OS mailing list -- os(a)lists.openanolis.cn
> To unsubscribe send an email to os-leave(a)lists.openanolis.cn
Hi, all
I'm glad to invide Yihao Wu <wuyihao(a)linux.alibaba.com> as maintainer
of ANCK scheduler.
Yihao is working on linux kernel scheduler for 5 years, and show great
interest in kernel technologies.
Here are some highlight contributions from Yihao:
1. Implement the lastest version of CPU SLI feature (Sevice Level Index)
for rich container;
this new version show great advantages in performance and accuracy.
2. Design and implement cpu isolation solution for function computing
running on Daishu, which
enable 2000 microvm running on one machine which acceptable overhead
(< 1%).
3. The core develper in plugsched project, which aims to upgrade linux
kernel scheduler subsystem online.
more details please visit project repo:
https://gitee.com/anolis/plugsched
4. Backport cpuidle-haltpoll feature to ANCK and optimize performance
for virtual machine.
5. Proficient in problem analysis, be skilled using gdb, crash, pyGDB,
qemu ...
https://lore.kernel.org/all/20220428103726.108597-1-wuyihao@linux.alibaba.c…
Working with Yihao is efficient and have a sense of achievement. Let us
hope it happens soon.
-- csp
> Hi, all
>
>
> I'm glad to invide Yihao Wu <wuyihao(a)linux.alibaba.com> as maintainer
> of ANCK scheduler.
> Yihao is working on linux kernel scheduler for 5 years, and show great
> interest in kernel technologies.
>
> Here are some highlight contributions from Yihao:
>
> 1. Implement the lastest version of CPU SLI feature (Sevice Level Index)
> for rich container;
> this new version show great advantages in performance and accuracy.
>
> 2. Design and implement cpu isolation solution for function computing
> running on Daishu, which
> enable 2000 microvm running on one machine which acceptable overhead
> (< 1%).
>
> 3. The core develper in plugsched project, which aims to upgrade linux
> kernel scheduler subsystem online.
> more details please visit project repo:
> https://gitee.com/anolis/plugsched
>
> 4. Backport cpuidle-haltpoll feature to ANCK and optimize performance
> for virtual machine.
>
> 5. Proficient in problem analysis, be skilled using gdb, crash, pyGDB,
> qemu ...
> https://lore.kernel.org/all/20220428103726.108597-1-wuyihao@linux.alibaba.c…
>
>
>
> Working with Yihao is efficient and have a sense of achievement. Let us
> hope it happens soon.
Looks great to me.
Acked-by: Baolin Wang <baolin.wang(a)linux.alibaba.com>
Hi,
The ANCK-5.10 next release has been code frozen yesterday（4.20), the
develop branch "devel-5.10" has been rebased to the latest upstream
stable branch "5.10.112" as planned.
Please note that your forked branch need pull the latest devel-5.10,
especially for the PR in progress.
Thanks for your attention!
BS.
-- Shiloong
