[PATCH] seccomp: Improve performance by optimizing memory barrier · wanghongzhe <hidden> · 2021-02-01
Re: [PATCH] seccomp: Improve performance by optimizing memory barrier · Andy Lutomirski <luto@amacapital.net> · 2021-02-01
RE: [PATCH] seccomp: Improve performance by optimizing memory barrier · Wanghongzhe (Hongzhe, EulerOS) <hidden> · 2021-02-02
[PATCH v1 1/1] Firstly, as Andy mentioned, this should be smp_rmb() instead of rmb(). considering that TSYNC is a cross-thread situation, and rmb() is a mandatory barrier which should not be used to control SMP effects, since mandatory barriers impose unnecessary overhead on both SMP and UP systems, as kernel Documentation said. · wanghongzhe <hidden> · 2021-02-02
Re: [PATCH v1 1/1] Firstly, as Andy mentioned, this should be smp_rmb() instead of rmb(). considering that TSYNC is a cross-thread situation, and rmb() is a mandatory barrier which should not be used to control SMP effects, since mandatory barriers impose unnecessary overhead on both SMP and UP systems, as kernel Documentation said. · Greg KH <gregkh@linuxfoundation.org> · 2021-02-02
Re: [PATCH v1 1/1] Firstly, as Andy mentioned, this should be smp_rmb() instead of rmb(). considering that TSYNC is a cross-thread situation, and rmb() is a mandatory barrier which should not be used to control SMP effects, since mandatory barriers impose unnecessary overhead on both SMP and UP systems, as kernel Documentation said. · Kees Cook <hidden> · 2021-02-02
RE: [PATCH v1 1/1] Firstly, as Andy mentioned, this should be smp_rmb() instead of rmb(). considering that TSYNC is a cross-thread situation, and rmb() is a mandatory barrier which should not be used to control SMP effects, since mandatory barriers imp... · Wanghongzhe (Hongzhe, EulerOS) <hidden> · 2021-02-04

RE: [PATCH v1 1/1] Firstly, as Andy mentioned, this should be smp_rmb() instead of rmb(). considering that TSYNC is a cross-thread situation, and rmb() is a mandatory barrier which should not be used to control SMP effects, since mandatory barriers imp...

From: Wanghongzhe (Hongzhe, EulerOS) <hidden>
Date: 2021-02-04 08:46:28
Also in: bpf, lkml

Let's start with a patch that just replaces rmb() with smp_rmb() and then work
on optimizing. Can you provide performance numbers that show
rmb() (and soon smp_rmb()) is causing actual problems here?

Ok, I will send a patch that just replaces rmb() with smp_rmb() and give performance numbers.

BUG() should never be used[1]. This is a recoverable situation, I think, and
should be handled as such.

I just follow the default case behind. Let's discuss this issue in next patches. 

-- 
wanghongzhe

`h`	back out one level
`j`	next message in thread
`k`	previous message in thread
`l`	drill in
`Esc`	close help / fold thread tree
`?`	toggle this help