This is the mail archive of the
gcc-patches@gcc.gnu.org
mailing list for the GCC project.
Re: [PATCH -v4 9/9] tracing: add function graph tracer support for MIPS
- From: Wu Zhangjin <wuzhangjin at gmail dot com>
- To: pajko <kpajko79 at gmail dot com>
- Cc: linux-kernel at vger dot kernel dot org, GCC Patches <gcc-patches at gcc dot gnu dot org>, wuzhangjin at gmail dot com, Adam Nemet <anemet at caviumnetworks dot com>, rostedt at goodmis dot org, linux-kernel at vger dot kernel dot org, linux-mips at linux-mips dot org, Thomas Gleixner <tglx at linutronix dot de>, Ralf Baechle <ralf at linux-mips dot org>, Nicholas Mc Guire <der dot herr at hofr dot at>, Richard Sandiford <rdsandiford at googlemail dot com>, David Daney <ddaney at caviumnetworks dot com>
- Date: Sun, 25 Oct 2009 18:48:37 +0800
- Subject: Re: [PATCH -v4 9/9] tracing: add function graph tracer support for MIPS
- References: <028867b99ec532b84963a35e7d552becc783cafc.1256135456.git.wuzhangjin@gmail.com> <2f73eae542c47ac5bbb9f7280e6c0271d193e90d.1256135456.git.wuzhangjin@gmail.com> <3f0d3515f74a58f4cfd11e61b62a129fdc21e3a7.1256135456.git.wuzhangjin@gmail.com> <96110ea5dd4d3d54eb97d0bb708a5bd81c7a50b5.1256135456.git.wuzhangjin@gmail.com> <af3ec1b5cd06b6f6a461c9fa7d09a51fabccb08d.1256135456.git.wuzhangjin@gmail.com> <a6f2959a69b6a77dd32cc36a5c8202f97d524f1e.1256135456.git.wuzhangjin@gmail.com> <53bdfdd95ec4fa00d4cc505bb5972cf21243a14d.1256135456.git.wuzhangjin@gmail.com> <26008418.post@talk.nabble.com>
- Reply-to: wuzhangjin at gmail dot com
(Added linux-mips mailing list and the other people to the CC list.)
On Thu, 2009-10-22 at 04:37 -0700, pajko wrote:
[...]
> >
>
> All this stuff can be avoided having PROFILE_BEFORE_PROLOGUE enabled in GCC
> (gcc/config/mips/mips.h), like it is done one several other architectures.
> Some of them even require it to be set for a working _mcount.
> Putting the call of _mcount before the function prologue should make no harm
> (it's working for me), and this way RA can be hooked for function graph
> tracing
> before it is saved to stack in the function prologue. Thus there will be no
> difference between leaf and non-leaf functions.
Good idea! Seems PROFILE_BEFORE_PROLOGUE is commented by default in
gcc/config/mips/mips.h of gcc 4.4:
/* #define PROFILE_BEFORE_PROLOGUE */
if we enable this macro, the handling will be the same to non-leaf and
leaf function, so, David's patch to gcc is not need again.
> This change is no big deal as GCC has to be patched anyway to get dynamic
> ftracing (and with 4.2.1 I had to patch it to get ftracing in modules
> working
> - using -mlong-calls -, but it's possible that this is fixed in the newer
> releases).
>
do you mean if enabling PROFILE_BEFORE_PROLOGUE, there will be some
problems with module support using -mlong-calls?
I have made dynamic ftrace support module with -mlong-calls, I think, if
we move the codes before prologue, it should have no side effect.
lui v1,HI16bit (&_mcount -> v1)
daddiu v1,v1,LOW16bit
move at,ra
jalr v1
Regards,
Wu Zhangjin