[PATCH] Reduce benchtests time

classic Classic list List threaded Threaded
2 messages Options
Reply | Threaded
Open this post in threaded view
|

[PATCH] Reduce benchtests time

Wilco Dijkstra-2
Reduce the total time taken by benchtests.  The malloc thread test takes 4
minutes to run which is significantly more than most other tests. Reduce
this to a more reasonable 40 seconds.  The math tests take 10 seconds each,
eventhough all they do is loop on the same input.  Anything more than 1
second runtime is way overkill, so set the limit to 1 second.

ChangeLog:
2019-04-15  Wilco Dijkstra  <[hidden email]>

        * benchtests/Makefile (BENCH_DURATION): Set to 1 second.
        * benchtests/bench-malloc-thread.c (BENCH_DURATION): Set to 10 seconds.

--

diff --git a/benchtests/Makefile b/benchtests/Makefile
index 09f7cb8e475a312268eebb4d346edde70d22bb3d..e8a8887b5a402ba7daaf5ec03f46eabe8fba8c21 100644
--- a/benchtests/Makefile
+++ b/benchtests/Makefile
@@ -120,9 +120,9 @@ binaries-bench := $(addprefix $(objpfx)bench-,$(bench))
 binaries-benchset := $(addprefix $(objpfx)bench-,$(benchset))
 binaries-bench-malloc := $(addprefix $(objpfx)bench-,$(bench-malloc))
 
-# The default duration: 10 seconds.
+# The default duration: 1 seconds.
 ifndef BENCH_DURATION
-BENCH_DURATION := 10
+BENCH_DURATION := 1
 endif
 
 CPPFLAGS-nonlib += -DDURATION=$(BENCH_DURATION) -D_ISOMAC
diff --git a/benchtests/bench-malloc-thread.c b/benchtests/bench-malloc-thread.c
index 52261425b0f1af32c17328ea5e0a5bb6f230df47..9d7dcf26b7ba64addd946b3766cef73d9cf25ec3 100644
--- a/benchtests/bench-malloc-thread.c
+++ b/benchtests/bench-malloc-thread.c
@@ -31,7 +31,7 @@
 #include "json-lib.h"
 
 /* Benchmark duration in seconds.  */
-#define BENCHMARK_DURATION 60
+#define BENCHMARK_DURATION 10
 #define RAND_SEED 88
 
 #ifndef NUM_THREADS

Reply | Threaded
Open this post in threaded view
|

Re: [PATCH] Reduce benchtests time

Siddhesh Poyarekar-8
On 15/04/19 7:59 PM, Wilco Dijkstra wrote:
> Reduce the total time taken by benchtests.  The malloc thread test takes 4
> minutes to run which is significantly more than most other tests. Reduce
> this to a more reasonable 40 seconds.  The math tests take 10 seconds each,
> eventhough all they do is loop on the same input.  Anything more than 1
> second runtime is way overkill, so set the limit to 1 second.

This is fine.

Siddhesh

> ChangeLog:
> 2019-04-15  Wilco Dijkstra  <[hidden email]>
>
>         * benchtests/Makefile (BENCH_DURATION): Set to 1 second.
>         * benchtests/bench-malloc-thread.c (BENCH_DURATION): Set to 10 seconds.
>
> --
>
> diff --git a/benchtests/Makefile b/benchtests/Makefile
> index 09f7cb8e475a312268eebb4d346edde70d22bb3d..e8a8887b5a402ba7daaf5ec03f46eabe8fba8c21 100644
> --- a/benchtests/Makefile
> +++ b/benchtests/Makefile
> @@ -120,9 +120,9 @@ binaries-bench := $(addprefix $(objpfx)bench-,$(bench))
>  binaries-benchset := $(addprefix $(objpfx)bench-,$(benchset))
>  binaries-bench-malloc := $(addprefix $(objpfx)bench-,$(bench-malloc))
>  
> -# The default duration: 10 seconds.
> +# The default duration: 1 seconds.
>  ifndef BENCH_DURATION
> -BENCH_DURATION := 10
> +BENCH_DURATION := 1
>  endif
>  
>  CPPFLAGS-nonlib += -DDURATION=$(BENCH_DURATION) -D_ISOMAC
> diff --git a/benchtests/bench-malloc-thread.c b/benchtests/bench-malloc-thread.c
> index 52261425b0f1af32c17328ea5e0a5bb6f230df47..9d7dcf26b7ba64addd946b3766cef73d9cf25ec3 100644
> --- a/benchtests/bench-malloc-thread.c
> +++ b/benchtests/bench-malloc-thread.c
> @@ -31,7 +31,7 @@
>  #include "json-lib.h"
>  
>  /* Benchmark duration in seconds.  */
> -#define BENCHMARK_DURATION 60
> +#define BENCHMARK_DURATION 10
>  #define RAND_SEED 88
>  
>  #ifndef NUM_THREADS
>