Re: [libav-devel] [PATCH] checkasm: vp9dsp: Benchmark the dc-only version of idct_idct separately

[Ronald S. Bultje]

Hi,

On Thu, Nov 17, 2016 at 7:37 AM, Martin Storsjö  wrote:

> On Thu, 17 Nov 2016, Ronald S. Bultje wrote:
>
> Hi,
>>
>> On Mon, Nov 14, 2016 at 4:46 PM, Martin Storsjö  wrote:
>>
>> The dc-only mode is already checked to work correctly above, but this
>>> allows benchmarking this mode for performance tuning, and allows making
>>> sure that it actually is correctly hooked up.
>>> ---
>>>  tests/checkasm/vp9dsp.c | 6 ++
>>>  1 file changed, 6 insertions(+)
>>>
>>> diff --git a/tests/checkasm/vp9dsp.c b/tests/checkasm/vp9dsp.c
>>> index 690e0cf..b9d1c73 100644
>>> --- a/tests/checkasm/vp9dsp.c
>>> +++ b/tests/checkasm/vp9dsp.c
>>> @@ -297,6 +297,12 @@ static void check_itxfm(void)
>>>  }
>>>  bench_new(dst, sz * SIZEOF_PIXEL, coef, sz * sz);
>>>  }
>>> +if (txtp == 0 && tx != 4) {
>>> +if (check_func(dsp.itxfm_add[tx][txtp],
>>> "vp9_inv_%s_%dx%d_dc_add",
>>> +   txtp_types[txtp], sz, sz)) {
>>> +bench_new(dst, sz * SIZEOF_PIXEL, coef, 1);
>>> +}
>>> +}
>>>  }
>>>  }
>>>  report("itxfm");
>>> --
>>> 2.7.4
>>>
>>
>>
>> I had a different local modification that allows tuning all the relevant
>> sub-IDCTs, basically re-arranging the loops so check_func is inside the
>> sub-IDCT loop and we bench each sub-IDCT separately. That's more generic
>> and probably more useful.
>>
>
> Right, that's probably more useful. Would you care to finish that
> modification to get it upstreamed in either project?


Sure, no problem.

Ronald
___
libav-devel mailing list
libav-devel@libav.org
https://lists.libav.org/mailman/listinfo/libav-devel

Re: [libav-devel] [PATCH] checkasm: vp9dsp: Benchmark the dc-only version of idct_idct separately

[Martin Storsjö]

On Thu, 17 Nov 2016, Ronald S. Bultje wrote:


Hi,

On Mon, Nov 14, 2016 at 4:46 PM, Martin Storsjö  wrote:


The dc-only mode is already checked to work correctly above, but this
allows benchmarking this mode for performance tuning, and allows making
sure that it actually is correctly hooked up.
---
 tests/checkasm/vp9dsp.c | 6 ++
 1 file changed, 6 insertions(+)

diff --git a/tests/checkasm/vp9dsp.c b/tests/checkasm/vp9dsp.c
index 690e0cf..b9d1c73 100644
--- a/tests/checkasm/vp9dsp.c
+++ b/tests/checkasm/vp9dsp.c
@@ -297,6 +297,12 @@ static void check_itxfm(void)
 }
 bench_new(dst, sz * SIZEOF_PIXEL, coef, sz * sz);
 }
+if (txtp == 0 && tx != 4) {
+if (check_func(dsp.itxfm_add[tx][txtp],
"vp9_inv_%s_%dx%d_dc_add",
+   txtp_types[txtp], sz, sz)) {
+bench_new(dst, sz * SIZEOF_PIXEL, coef, 1);
+}
+}
 }
 }
 report("itxfm");
--
2.7.4



I had a different local modification that allows tuning all the relevant
sub-IDCTs, basically re-arranging the loops so check_func is inside the
sub-IDCT loop and we bench each sub-IDCT separately. That's more generic
and probably more useful.


Right, that's probably more useful. Would you care to finish that 
modification to get it upstreamed in either project?


// Martin
___
libav-devel mailing list
libav-devel@libav.org
https://lists.libav.org/mailman/listinfo/libav-devel

Re: [libav-devel] [PATCH] checkasm: vp9dsp: Benchmark the dc-only version of idct_idct separately

[Ronald S. Bultje]

Hi,

On Mon, Nov 14, 2016 at 4:46 PM, Martin Storsjö  wrote:

> The dc-only mode is already checked to work correctly above, but this
> allows benchmarking this mode for performance tuning, and allows making
> sure that it actually is correctly hooked up.
> ---
>  tests/checkasm/vp9dsp.c | 6 ++
>  1 file changed, 6 insertions(+)
>
> diff --git a/tests/checkasm/vp9dsp.c b/tests/checkasm/vp9dsp.c
> index 690e0cf..b9d1c73 100644
> --- a/tests/checkasm/vp9dsp.c
> +++ b/tests/checkasm/vp9dsp.c
> @@ -297,6 +297,12 @@ static void check_itxfm(void)
>  }
>  bench_new(dst, sz * SIZEOF_PIXEL, coef, sz * sz);
>  }
> +if (txtp == 0 && tx != 4) {
> +if (check_func(dsp.itxfm_add[tx][txtp],
> "vp9_inv_%s_%dx%d_dc_add",
> +   txtp_types[txtp], sz, sz)) {
> +bench_new(dst, sz * SIZEOF_PIXEL, coef, 1);
> +}
> +}
>  }
>  }
>  report("itxfm");
> --
> 2.7.4


I had a different local modification that allows tuning all the relevant
sub-IDCTs, basically re-arranging the loops so check_func is inside the
sub-IDCT loop and we bench each sub-IDCT separately. That's more generic
and probably more useful.

Ronald
___
libav-devel mailing list
libav-devel@libav.org
https://lists.libav.org/mailman/listinfo/libav-devel

Re: [libav-devel] [PATCH] checkasm: vp9dsp: Benchmark the dc-only version of idct_idct separately

[Janne Grunau]

On 2016-11-14 23:46:14 +0200, Martin Storsjö wrote:
> The dc-only mode is already checked to work correctly above, but this
> allows benchmarking this mode for performance tuning, and allows making
> sure that it actually is correctly hooked up.
> ---
>  tests/checkasm/vp9dsp.c | 6 ++
>  1 file changed, 6 insertions(+)
> 
> diff --git a/tests/checkasm/vp9dsp.c b/tests/checkasm/vp9dsp.c
> index 690e0cf..b9d1c73 100644
> --- a/tests/checkasm/vp9dsp.c
> +++ b/tests/checkasm/vp9dsp.c
> @@ -297,6 +297,12 @@ static void check_itxfm(void)
>  }
>  bench_new(dst, sz * SIZEOF_PIXEL, coef, sz * sz);
>  }
> +if (txtp == 0 && tx != 4) {
> +if (check_func(dsp.itxfm_add[tx][txtp], 
> "vp9_inv_%s_%dx%d_dc_add",
> +   txtp_types[txtp], sz, sz)) {
> +bench_new(dst, sz * SIZEOF_PIXEL, coef, 1);
> +}
> +}
>  }
>  }
>  report("itxfm");

ok

Janne
___
libav-devel mailing list
libav-devel@libav.org
https://lists.libav.org/mailman/listinfo/libav-devel

Re: [libav-devel] [PATCH] checkasm: vp9dsp: Benchmark the dc-only version of idct_idct separately

[Luca Barbato]

On 14/11/2016 22:46, Martin Storsjö wrote:
> The dc-only mode is already checked to work correctly above, but this
> allows benchmarking this mode for performance tuning, and allows making
> sure that it actually is correctly hooked up.
> ---
>  tests/checkasm/vp9dsp.c | 6 ++
>  1 file changed, 6 insertions(+)
> 
> diff --git a/tests/checkasm/vp9dsp.c b/tests/checkasm/vp9dsp.c
> index 690e0cf..b9d1c73 100644
> --- a/tests/checkasm/vp9dsp.c
> +++ b/tests/checkasm/vp9dsp.c
> @@ -297,6 +297,12 @@ static void check_itxfm(void)
>  }
>  bench_new(dst, sz * SIZEOF_PIXEL, coef, sz * sz);
>  }
> +if (txtp == 0 && tx != 4) {
> +if (check_func(dsp.itxfm_add[tx][txtp], 
> "vp9_inv_%s_%dx%d_dc_add",
> +   txtp_types[txtp], sz, sz)) {
> +bench_new(dst, sz * SIZEOF_PIXEL, coef, 1);
> +}
> +}
>  }
>  }
>  report("itxfm");
> 

Probably Ok.
___
libav-devel mailing list
libav-devel@libav.org
https://lists.libav.org/mailman/listinfo/libav-devel

[libav-devel] [PATCH] checkasm: vp9dsp: Benchmark the dc-only version of idct_idct separately

[Martin Storsjö]

The dc-only mode is already checked to work correctly above, but this
allows benchmarking this mode for performance tuning, and allows making
sure that it actually is correctly hooked up.
---
 tests/checkasm/vp9dsp.c | 6 ++
 1 file changed, 6 insertions(+)

diff --git a/tests/checkasm/vp9dsp.c b/tests/checkasm/vp9dsp.c
index 690e0cf..b9d1c73 100644
--- a/tests/checkasm/vp9dsp.c
+++ b/tests/checkasm/vp9dsp.c
@@ -297,6 +297,12 @@ static void check_itxfm(void)
 }
 bench_new(dst, sz * SIZEOF_PIXEL, coef, sz * sz);
 }
+if (txtp == 0 && tx != 4) {
+if (check_func(dsp.itxfm_add[tx][txtp], 
"vp9_inv_%s_%dx%d_dc_add",
+   txtp_types[txtp], sz, sz)) {
+bench_new(dst, sz * SIZEOF_PIXEL, coef, 1);
+}
+}
 }
 }
 report("itxfm");
-- 
2.7.4

___
libav-devel mailing list
libav-devel@libav.org
https://lists.libav.org/mailman/listinfo/libav-devel