From mboxrd@z Thu Jan  1 00:00:00 1970
From: mulyadi.santosa@gmail.com (Mulyadi Santosa)
Date: Sat, 26 Feb 2011 11:31:08 +0700
Subject: syscalls performance
In-Reply-To: <C8CB91A9-6C65-4556-95B4-9F881A3CE724@cs.ucsd.edu>
References: <AANLkTimZac++X4jUsz2-bTTtEJwOR45NHAQ=GR3xCVnG@mail.gmail.com>
	<AANLkTiniirL=jN=yOmsev6MF3a1mpwGUkm-_Qu7zh6AR@mail.gmail.com>
	<AANLkTina4Yrukqti5oeA7Yt-qcRph2Qr9Q99GkuHz=oh@mail.gmail.com>
	<AANLkTi=S0just=K_+_OJspgceX2T7-xtiRos4AQU6xA+@mail.gmail.com>
	<AANLkTimdHuw7SDuLCbsLHcLvVhv9XSNWwyFOOzzdfX9v@mail.gmail.com>
	<C8CB91A9-6C65-4556-95B4-9F881A3CE724@cs.ucsd.edu>
Message-ID: <AANLkTinrn43VRRrC0-FGdrWWuOsAdk_Zo5of7HbrgvON@mail.gmail.com>
To: kernelnewbies@lists.kernelnewbies.org
List-Id: kernelnewbies.lists.kernelnewbies.org

Hi Enrico...

On Sat, Feb 26, 2011 at 03:42, Enrico Granata <egranata@ucsd.edu> wrote:
>
>
> I modified the source code to show exactly how many clock ticks it is taking
> for each call. It seems that the behavior hinted by Mauro Romano Trajber is
> actually there:
> [enrico at espresso ~]$ ./syscallperf 15
> 4925
> 1190
> 942

watching strace, objdump etc etc, I roughly conclude the high initial
tsc number might be due to initial symbol resolve to mygetpid() and/or
syscall() (IIRC this is called PLT/GOT symbol resolve).

After the GOT/PLT table had the syscall/mygetpid address, the rest went fast.

PS: if using gcc -O2, seems like mygetpid() is really inlined....

-- 
regards,

Mulyadi Santosa
Freelance Linux trainer and consultant

blog: the-hydra.blogspot.com
training: mulyaditraining.blogspot.com