[prev in list] [next in list] [prev in thread] [next in thread] 

List:       ltp-list
Subject:    Re: [LTP] [PATCH 2/4][RT][take#5] matrix_mult.c: dynamic iterations
From:       Darren Hart <dvhltc () us ! ibm ! com>
Date:       2008-10-31 21:12:42
Message-ID: 490B74CA.2040403 () us ! ibm ! com
[Download RAW message or body]

Gilles Carry wrote:
> Dimension (128) of statistics arrays must be a multiple of number of cpu.
> This is a hassle as you have to recompile every time your arch changes
> or if you want to disable cpus.
> Failing to do so causes segfault when requiring statistics if
> 128 modulo nr_cpus != 0
> 
> This patch adds option -i to specify how many iterations are wanted
> (still defaulting to 128).
> It rounds up iterations to the nearest multiple of online cpus.
> 
> Signed-off-by: Gilles Carry <gilles.carry@bull.net>

Acked-by: Darren Hart <dvhltc@us.ibm.com>

> ---
> testcases/realtime/func/matrix_mult/matrix_mult.c |   42 ++++++++++++++++-----
> 1 files changed, 32 insertions(+), 10 deletions(-)
> 
> diff --git a/testcases/realtime/func/matrix_mult/matrix_mult.c \
> b/testcases/realtime/func/matrix_mult/matrix_mult.c index afaca06..1b33154 100644
> --- a/testcases/realtime/func/matrix_mult/matrix_mult.c
> +++ b/testcases/realtime/func/matrix_mult/matrix_mult.c
> @@ -51,7 +51,7 @@
> 					/* (and therefore SMP performance goes up) */
> #define PASS_CRITERIA	0.75		/* Avg concurrent time * pass criteria < avg seq time - \
> */  /* for every addition of a cpu */
> -#define ITERATIONS	128		/* HAS to be a multiple of 'numcpus' */
> +#define ITERATIONS	128
> #define HIST_BUCKETS	100
> 
> #define THREAD_WAIT	1
> @@ -68,6 +68,7 @@ static int *mult_index;
> static int *tids;
> static int *flags;
> static int online_cpu_id = -1;
> +static int iterations = ITERATIONS;
> 
> stats_container_t sdat, cdat, *curdat;
> stats_container_t shist, chist;
> @@ -81,6 +82,7 @@ void usage(void)
> 	printf("matrix_mult specific options:\n");
> 	printf("  -j            enable jvmsim\n");
> 	printf("  -l#           #: number of multiplications per iteration (load)\n");
> +	printf("  -i#           #: number of iterations\n");
> }
> 
> int parse_args(int c, char *v)
> @@ -90,6 +92,9 @@ int parse_args(int c, char *v)
> 	case 'j':
> 		run_jvmsim = 1;
> 		break;
> +	case 'i':
> +		iterations = atoi(v);
> +		break;
> 	case 'l':
> 		ops = atoi(v);
> 		break;
> @@ -231,9 +236,9 @@ void main_thread(void)
> 	float savg, cavg;
> 	int cpuid;
> 
> -	if (	stats_container_init(&sdat, ITERATIONS) ||
> +	if (	stats_container_init(&sdat, iterations) ||
> 		stats_container_init(&shist, HIST_BUCKETS) ||
> -		stats_container_init(&cdat, ITERATIONS/numcpus) ||
> +		stats_container_init(&cdat, iterations/numcpus) ||
> 		stats_container_init(&chist, HIST_BUCKETS)
> 	)
> 	{
> @@ -273,12 +278,12 @@ void main_thread(void)
> 	curdat = &sdat;
> 	printf("\nRunning sequential operations\n");
> 	start = rt_gettime();
> -	for (i = 0; i < ITERATIONS; i++)
> +	for (i = 0; i < iterations; i++)
> 		matrix_mult_record(MATRIX_SIZE, i);
> 	end = rt_gettime();
> 	delta = (long)((end - start)/NS_PER_US);
> 
> -	savg = delta/ITERATIONS; /* don't use the stats record, use the total time \
> recorded */ +	savg = delta/iterations; /* don't use the stats record, use the total \
> time recorded */  smin = stats_min(&sdat);
> 	smax = stats_max(&sdat);
> 
> @@ -311,9 +316,9 @@ void main_thread(void)
> 
> 	/* run matrix mult operation concurrently */
> 	curdat = &cdat;
> -	printf("\nRunning concurrent operations (%dx)\n", ITERATIONS);
> +	printf("\nRunning concurrent operations (%dx)\n", iterations);
> 	start = rt_gettime();
> -	for (i = 0; i < ITERATIONS; i++)
> +	for (i = 0; i < iterations; i++)
> 		concurrent_ops();
> 	end = rt_gettime();
> 	delta = (long)((end - start)/NS_PER_US);
> @@ -328,7 +333,7 @@ void main_thread(void)
> 		pthread_mutex_unlock(&t->mutex);
> 	}
> 
> -	cavg = delta/ITERATIONS; /* don't use the stats record, use the total time \
> recorded */ +	cavg = delta/iterations; /* don't use the stats record, use the total \
> time recorded */  cmin = stats_min(&cdat);
> 	cmax = stats_max(&cdat);
> 
> @@ -367,15 +372,32 @@ int main(int argc, char *argv[])
> {
> 	setup();
> 	pass_criteria = PASS_CRITERIA;
> -	rt_init("jl:h", parse_args, argc, argv);
> +	rt_init("jl:i:h", parse_args, argc, argv);
> 	numcpus = sysconf(_SC_NPROCESSORS_ONLN);
> 	/* the minimum avg concurrent multiplier to pass */
> 	criteria = pass_criteria * numcpus;
> +	int new_iterations;
> +
> +	if (iterations <= 0) {
> +		fprintf(stderr, "iterations must be greater than zero\n");
> +		exit(1);
> +	}
> 
> 	printf("\n---------------------------------------\n");
> 	printf("Matrix Multiplication (SMP Performance)\n");
> 	printf("---------------------------------------\n\n");
> -	printf("Running %d iterations\n", ITERATIONS);
> +
> +	/* Line below rounds up iterations to a multiple of numcpus.
> +	 * Without this, having iterations not a mutiple of numcpus causes
> +	 * stats to segfault (overflow stats array).
> +	 */
> +	new_iterations = (int) ( (iterations + numcpus - 1) / numcpus) * numcpus;
> +	if (new_iterations != iterations)
> +		printf("Rounding up iterations value to nearest multiple of total online \
> CPUs\n"); +
> +	iterations = new_iterations;
> +
> +	printf("Running %d iterations\n", iterations);
> 	printf("Matrix Dimensions: %dx%d\n", MATRIX_SIZE, MATRIX_SIZE);
> 	printf("Calculations per iteration: %d\n", ops);
> 	printf("Number of CPUs: %u\n", numcpus);


-- 
Darren Hart
IBM Linux Technology Center
Real-Time Linux Team

-------------------------------------------------------------------------
This SF.Net email is sponsored by the Moblin Your Move Developer's challenge
Build the coolest Linux based applications with Moblin SDK & win great prizes
Grand prize is a trip for two to an Open Source event anywhere in the world
http://moblin-contest.org/redirect.php?banner_id=100&url=/
_______________________________________________
Ltp-list mailing list
Ltp-list@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/ltp-list


[prev in list] [next in list] [prev in thread] [next in thread]