[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-Id: <1662374882.xll36f5qp9.naveen@linux.ibm.com>
Date: Mon, 05 Sep 2022 16:25:42 +0530
From: "Naveen N. Rao" <naveen.n.rao@...ux.vnet.ibm.com>
To: Peter Zijlstra <peterz@...radead.org>,
Sathvika Vasireddy <sv@...ux.ibm.com>
Cc: aik@...abs.ru, chenzhongjin@...wei.com,
christophe.leroy@...roup.eu, jpoimboe@...hat.com,
linux-kernel@...r.kernel.org, linuxppc-dev@...ts.ozlabs.org,
mbenes@...e.cz, mingo@...hat.com, mpe@...erman.id.au,
npiggin@...il.com, rostedt@...dmis.org
Subject: Re: [PATCH v2 11/16] objtool: Add --mnop as an option to --mcount
Peter Zijlstra wrote:
> On Mon, Aug 29, 2022 at 11:22:18AM +0530, Sathvika Vasireddy wrote:
>> Architectures can select HAVE_NOP_MCOUNT if they choose
>> to nop out mcount call sites. If that config option is
>> selected, then --mnop is passed as an option to objtool,
>> along with --mcount.
>>
>> Also, make sure that --mnop can be passed as an option
>> to objtool only when --mcount is passed.
>>
>> Signed-off-by: Sathvika Vasireddy <sv@...ux.ibm.com>
>
> IIRC you want to nop the things yourself because PLT entries or
> something?
For two reasons:
1. We want to maintain a pointer to 'struct module' as part of the
ftrace rec so that we can patch ftrace sites in modules to go to module
stubs.
2. We depend on compiler generated long branches to support large
kernels.
>
> Anyway, patch looks fine, even though the Changelog doesn't relaly
> justify the change.
Sure. We should add that powerpc kernel does not support nop'ed out
ftrace locations.
>
> Acked-by: Peter Zijlstra (Intel) <peterz@...radead.org>
>
Thanks,
Naveen
Powered by blists - more mailing lists