linux-kernel - Re: [PATCH] hrtimer: Update softirq_expires_next correctly after __hrtimer_get_next

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Date:   Wed, 21 Apr 2021 23:40:54 +0900
From:   Lorenzo Colitti <lorenzo@...gle.com>
To:     Thomas Gleixner <tglx@...utronix.de>
Cc:     Greg KH <gregkh@...uxfoundation.org>,
        Maciej Żenczykowski <zenczykowski@...il.com>,
        Ingo Molnar <mingo@...nel.org>,
        Anna-Maria Behnsen <anna-maria@...utronix.de>,
        lkml <linux-kernel@...r.kernel.org>,
        mikael.beckius@...driver.com,
        Maciej Żenczykowski <maze@...gle.com>,
        Will Deacon <will@...nel.org>
Subject: Re: [PATCH] hrtimer: Update softirq_expires_next correctly after __hrtimer_get_next_event()

On Wed, Apr 21, 2021 at 11:08 PM Lorenzo Colitti <lorenzo@...gle.com> wrote:
> load the timer will almost never fire since there will always be
> another packet coming. So the speed of adding/updating the timer is
> much more important than the accuracy. We will try to move it to
> timer_list.

... but actually, that's not acceptable, because AFAICS timer_list
counts in jiffies so its resolution might be as low as 10ms. It's not
acceptable to delay network traffic by up to 10ms under load. Not sure
if there is another timer API that can be used here.