X-Authentication-Warning: delorie.com: mail set sender to djgpp-bounces using -f
From: philippe <philippe DOT meynard AT gmail DOT com>
Newsgroups: comp.os.msdos.djgpp
Subject: Re: PB on CPU Atom
Date: Tue, 4 Jan 2011 08:25:20 -0800 (PST)
Organization: http://groups.google.com
Lines: 53
Message-ID: <e0a39940-61ac-4916-90f4-5a7b06c05a9d@i25g2000prd.googlegroups.com>
References: <4e67d035-8f32-4e9e-a78c-2370909f9ffb AT k11g2000vbf DOT googlegroups DOT com>
 <c85fca63-338e-4f0d-987d-c78972bd9561 AT 12g2000prq DOT googlegroups DOT com>
 <beb4d0eb-5be6-4a05-97ed-7f9377d01d79 AT c17g2000prm DOT googlegroups DOT com> <e9683889-a26e-42e9-81b5-31b3c40091e0 AT i41g2000vbn DOT googlegroups DOT com>
NNTP-Posting-Host: 80.70.37.40
Mime-Version: 1.0
Content-Type: text/plain; charset=ISO-8859-1
X-Trace: posting.google.com 1294158320 11655 127.0.0.1 (4 Jan 2011 16:25:20 GMT)
X-Complaints-To: groups-abuse AT google DOT com
NNTP-Posting-Date: Tue, 4 Jan 2011 16:25:20 +0000 (UTC)
Complaints-To: groups-abuse AT google DOT com
Injection-Info: i25g2000prd.googlegroups.com; posting-host=80.70.37.40; posting-account=aoZx2woAAADtfEzA-h1Fpu5B96M9fx2b
User-Agent: G2/1.0
X-HTTP-UserAgent: Mozilla/4.0 (compatible; MSIE 8.0; Windows NT 5.1;
 Trident/4.0; .NET CLR 2.0.50727; InfoPath.1; .NET CLR 1.1.4322),gzip(gfe)
Bytes: 2901
To: djgpp AT delorie DOT com
DJ-Gateway: from newsgroup comp.os.msdos.djgpp
Reply-To: djgpp AT delorie DOT com

> I discovered that: When I used a call who is in another source file
> that my source file test then the program is very slow but when I put
> the call function in the same file that the source test then the
> program is very fast. Why?


Hi, everybody and happy new year.
I come back on my problem and in fact I discovered this.
my test program use 2 source file. see :

test1.c

vois test1(void)
{
                char a;
	uint64 debclock1,finclock1;
	uint64 tpscalc;
	long tpscalc2;
	char chaine[80];
	int i;

	init_cpu_speed();
                printf("start\n");
                debclock1=get_rdtsc();
	for(i=0;i<3200000;i++){
		affcarinchtest2(a,chaine);
		a++;
		if(a>'Z') a=65;
	}
	finclock1=get_rdtsc();
	tpscalc=finclock1-debclock1;
	tpscalc2=(long)(((tpscalc*MULT_COUNT_US)/COUNT_CPU_BY_SECOND));
	printf("time 1: %ld\n",((long)tpscalc2));
}

test2.c

void affcarinchtest(char chaine,char *ptrch)
{
	*ptrch=chaine;
}

when I compile this only 2 files I have a executed time correct, but
When I compile this file with another source files and who create a
big program (> 3Mo) then I have a executed time very long (10 time
longer!)  I suppose that the problem is the call function is different
if the call is short or long ?
I test with many different option(like Rugxulo suggest : mtune,
without O3, etc), without succes. With different version of gcc,
without succces
I haven't test with -Atom option of gcc because it's only implemented
on gcc 4.5 and not on djgpp.
Have you a suggestion?