[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <CAL+tcoDyCYjdW1E7gs0EsKGAN1Esv524EeA_=ORt+Chyd5Sj5w@mail.gmail.com>
Date: Tue, 9 May 2023 21:25:19 +0800
From: Jason Xing <kerneljasonxing@...il.com>
To: Thomas Gleixner <tglx@...utronix.de>
Cc: paulmck@...nel.org, peterz@...radead.org, bigeasy@...utronix.de,
frederic@...nel.org, linux-kernel@...r.kernel.org,
netdev@...r.kernel.org, Jason Xing <kernelxing@...cent.com>
Subject: Re: [PATCH] softirq: let the userside tune the SOFTIRQ_NOW_MASK with sysctl
On Tue, May 9, 2023 at 9:05 PM Thomas Gleixner <tglx@...utronix.de> wrote:
>
> On Mon, Apr 10 2023 at 10:30, Jason Xing wrote:
> > From: Jason Xing <kernelxing@...cent.com>
> >
> > Currently we have two exceptions which could avoid ksoftirqd when
> > invoking softirqs: HI_SOFTIRQ and TASKLET_SOFTIRQ. They were introduced
> > in the commit 3c53776e29f8 ("Mark HI and TASKLET softirq synchronous")
> > which says if we don't mask them, it will cause excessive latencies in
> > some cases.
>
> As we are ripping this out, I'll ignore this patch.
Sure, please ignore this heuristic patch. Paolo and I have already
tested that revert patch in the production environment before and
verified its usefulness, please take that one if you could.
Thanks,
Jason
Powered by blists - more mailing lists