[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <20170824084906.mtmj6rz4igo63dl5@hirez.programming.kicks-ass.net>
Date: Thu, 24 Aug 2017 10:49:06 +0200
From: Peter Zijlstra <peterz@...radead.org>
To: Hoeun Ryu <hoeun.ryu@...il.com>
Cc: Kees Cook <keescook@...omium.org>,
Andrew Morton <akpm@...ux-foundation.org>,
Nicholas Piggin <npiggin@...il.com>,
Ingo Molnar <mingo@...nel.org>, Michal Marek <mmarek@...e.com>,
Laura Abbott <labbott@...hat.com>,
Al Viro <viro@...iv.linux.org.uk>,
Andy Lutomirski <luto@...nel.org>, mic@...ikod.net,
Russell King <linux@...linux.org.uk>,
Vladimir Murzin <vladimir.murzin@....com>,
Heiko Carstens <heiko.carstens@...ibm.com>,
Jonathan Austin <jonathan.austin@....com>,
Masahiro Yamada <yamada.masahiro@...ionext.com>,
Mark Rutland <mark.rutland@....com>,
linux-kernel@...r.kernel.org, linux-arm-kernel@...ts.infradead.org,
linux-arch@...r.kernel.org
Subject: Re: [RFC 3/3] sched: depend on 64BIT_ATOMIC_ACCESS to determine if
to use min_vruntime_copy
On Thu, Aug 24, 2017 at 02:42:57PM +0900, Hoeun Ryu wrote:
> +#ifndef CONFIG_64BIT_ATOMIC_ALIGNED_ACCESS
> u64 min_vruntime;
> -#ifndef CONFIG_64BIT
> +#else
> + u64 min_vruntime __attribute__((aligned(sizeof(u64))));
> +#endif
That's stupid, just make sure your platform defines u64 as naturally
aligned when you have this 64BIT_ATOMIC foo.
Also, please try and dig out more 32bit archs that can use this and make
sure to include performance numbers to justify this extra cruft.
Powered by blists - more mailing lists