[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20200117180324.GA2623847@rani.riverdale.lan>
Date: Fri, 17 Jan 2020 13:03:26 -0500
From: Arvind Sankar <nivedita@...m.mit.edu>
To: Eric Dumazet <eric.dumazet@...il.com>
Cc: Peter Zijlstra <peterz@...radead.org>,
Shaokun Zhang <zhangshaokun@...ilicon.com>,
David Miller <davem@...emloft.net>, netdev@...r.kernel.org,
linux-kernel@...r.kernel.org, jinyuqi@...wei.com,
kuznet@....inr.ac.ru, yoshfuji@...ux-ipv6.org, edumazet@...gle.com,
guoyang2@...wei.com, Will Deacon <will@...nel.org>
Subject: Re: [PATCH] net: optimize cmpxchg in ip_idents_reserve
On Fri, Jan 17, 2020 at 08:35:07AM -0800, Eric Dumazet wrote:
>
>
> On 1/17/20 4:32 AM, Peter Zijlstra wrote:
>
> >
> > That's crazy, just accept that UBSAN is taking bonghits and ignore it.
> > Use atomic_add_return() unconditionally.
> >
>
> Yes, we might simply add a comment so that people do not bug us if
> their compiler is too old.
>
> /* If UBSAN reports an error there, please make sure your compiler
> * supports -fno-strict-overflow before reporting it.
> */
> return atomic_add_return(segs + delta, p_id) - segs;
>
Do we need that comment any more? The flag was apparently introduced in
gcc-4.2 and we only support 4.6+ anyway?
Powered by blists - more mailing lists