[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20190308213343.GE2482@worktop.programming.kicks-ass.net>
Date: Fri, 8 Mar 2019 22:33:43 +0100
From: Peter Zijlstra <peterz@...radead.org>
To: Josh Poimboeuf <jpoimboe@...hat.com>
Cc: torvalds@...ux-foundation.org, tglx@...utronix.de, hpa@...or.com,
julien.thierry@....com, will.deacon@....com, luto@...capital.net,
mingo@...nel.org, catalin.marinas@....com, james.morse@....com,
valentin.schneider@....com, brgerst@...il.com, luto@...nel.org,
bp@...en8.de, dvlasenk@...hat.com, linux-kernel@...r.kernel.org,
dvyukov@...gle.com, rostedt@...dmis.org
Subject: Re: [PATCH 20/20] objtool: Add Direction Flag validation
On Fri, Mar 08, 2019 at 03:16:03PM -0600, Josh Poimboeuf wrote:
> On Thu, Mar 07, 2019 at 12:45:31PM +0100, Peter Zijlstra wrote:
> > Having DF escape is BAD(tm).
> >
> > Suggested-by: Linus Torvalds <torvalds@...ux-foundation.org>
> > Signed-off-by: Peter Zijlstra (Intel) <peterz@...radead.org>
>
> Can you elaborate on why (in the patch description)? Did this actually
> find any occurrences?
Nope, didn't find anything. Also, all DF users are in asm so I didn't
really expect any. Having it escape would probably result in fairly
instant wreckage though.
DF=1 results in things like rep mov going _backwards_.
Powered by blists - more mailing lists