[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-Id: <1653564857.f06fbbl3vg.naveen@linux.ibm.com>
Date: Thu, 26 May 2022 17:21:02 +0530
From: "Naveen N. Rao" <naveen.n.rao@...ux.vnet.ibm.com>
To: Peter Zijlstra <peterz@...radead.org>
Cc: "aik@...abs.ru" <aik@...abs.ru>,
Christophe Leroy <christophe.leroy@...roup.eu>,
"jpoimboe@...hat.com" <jpoimboe@...hat.com>,
"linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>,
"linuxppc-dev@...ts.ozlabs.org" <linuxppc-dev@...ts.ozlabs.org>,
"mbenes@...e.cz" <mbenes@...e.cz>,
"mpe@...erman.id.au" <mpe@...erman.id.au>,
"rostedt@...dmis.org" <rostedt@...dmis.org>,
Sathvika Vasireddy <sv@...ux.ibm.com>
Subject: Re: [RFC PATCH 1/4] objtool: Add --mnop as an option to --mcount
Peter Zijlstra wrote:
> On Tue, May 24, 2022 at 04:01:48PM +0530, Naveen N. Rao wrote:
>
>> We need to know for sure either way. Nop'ing out the _mcount locations at
>> boot allows us to discover existing long branch trampolines. If we want to
>> avoid it, we need to note down those locations during build time.
>>
>> Do you have a different approach in mind?
>
> If you put _mcount in a separate section then the compiler cannot tell
> where it is and is forced to always emit a long branch trampoline.
>
> Does that help?
That's an interesting thought. Depending on the type of trampoline the
compiler emits, I might be able to use this approach. We will still need
objtool on powerpc so that we can note down those trampoline locations.
Thanks,
Naveen
Powered by blists - more mailing lists