linux-kernel - RE: [PATCH] net : rps : supoort a single flow to use rps

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] [day] [month] [year] [list]

Message-ID: <2ed297c680f24879aba6f15df7630b96@AcuMS.aculab.com>
Date:   Wed, 29 Jun 2022 09:09:35 +0000
From:   David Laight <David.Laight@...LAB.COM>
To:     'Alexander Lobakin' <alexandr.lobakin@...el.com>,
        Albert Huang <huangjie.albert@...edance.com>
CC:     "David S. Miller" <davem@...emloft.net>,
        Eric Dumazet <edumazet@...gle.com>,
        Jakub Kicinski <kuba@...nel.org>,
        Paolo Abeni <pabeni@...hat.com>,
        Sebastian Andrzej Siewior <bigeasy@...utronix.de>,
        Menglong Dong <imagedong@...cent.com>,
        Petr Machata <petrm@...dia.com>,
        "Kumar Kartikeya Dwivedi" <memxor@...il.com>,
        Arnd Bergmann <arnd@...db.de>,
        "Antoine Tenart" <atenart@...nel.org>,
        Phil Auld <pauld@...hat.com>,
        "Frederic Weisbecker" <frederic@...nel.org>,
        Xin Long <lucien.xin@...il.com>,
        "netdev@...r.kernel.org" <netdev@...r.kernel.org>,
        "linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>
Subject: RE: [PATCH] net : rps : supoort a single flow to use rps

....
> I'd suggest to use some flags bitfield instead, so it would be way
> more scalable and at the same time have the pre-determined size. Ar
> per several LKML discussions, `bool` may have different size and
> logics depending on architecture and compiler, so when using it not
> on the stack inside a function, but in a structure, it might be not
> easy then to track cacheline layout etc.
> So, maybe
> 
> 	unsigned long rps_flags;
>  #define RPS_SINGLE_FLOW_ENABLE BIT(0)
> 
> or even
> 
> 	DECLARE_BITMAP(rps_flags);
>  #define RPS_SINGLE_FLOW_ENABLE 0

I don't think BITMAPs are a good idea unless you really
need a lot of bits and (probably) locked accesses.

You can use C bitfields - the compiler doesn't (usually)
make too much of a 'pigs breakfast' of them unless you
needs to set/test multiple ones at the same time.

	David

-
Registered Address Lakeside, Bramley Road, Mount Farm, Milton Keynes, MK1 1PT, UK
Registration No: 1397386 (Wales)