Re: [PATCH] igt/test/pm_rps: load GPU to force not-idle to idle transition.

From: Imre Deak <imre.deak@intel.com>
To: Bob Paauwe <bob.j.paauwe@intel.com>,
	intel-gfx <intel-gfx@lists.freedesktop.org>
Subject: Re: [PATCH] igt/test/pm_rps: load GPU to force not-idle to idle transition.
Date: Wed, 09 Dec 2015 17:31:33 +0200	[thread overview]
Message-ID: <1449675093.11719.4.camel@intel.com> (raw)
In-Reply-To: <1449525215-4140-1-git-send-email-bob.j.paauwe@intel.com>

On ma, 2015-12-07 at 13:53 -0800, Bob Paauwe wrote:
> When changing the sysfs GT min frequency, the kernel won't
> automatcilly drop the GT frequency to idle unless the GPU
> transitions from busy to idle.
> 
> Load the GPU after increasing the GT min frequency to force
> a busy to idle transition. This matches the behavior when
> decreasing the GT min frequency.
> 
> Signed-off-by: Bob Paauwe <bob.j.paauwe@intel.com>

Yep, this fixes igt/pm_rps for me on BXT:
Reviewed-by: Imre Deak <imre.deak@intel.com>

I pushed the patch to igt.

> ---
>  tests/pm_rps.c | 4 ++++
>  1 file changed, 4 insertions(+)
> 
> diff --git a/tests/pm_rps.c b/tests/pm_rps.c
> index 9d054fd..9f752f8 100644
> --- a/tests/pm_rps.c
> +++ b/tests/pm_rps.c
> @@ -388,10 +388,14 @@ static void min_max_config(void (*check)(void),
> bool load_gpu)
>  
>  	igt_debug("\nIncrease min to midpoint...\n");
>  	writeval(stuff[MIN].filp, fmid);
> +	if (load_gpu)
> +		do_load_gpu();
>  	check();
>  
>  	igt_debug("\nIncrease min to RP0...\n");
>  	writeval(stuff[MIN].filp, origfreqs[RP0]);
> +	if (load_gpu)
> +		do_load_gpu();
>  	check();
>  
>  	igt_debug("\nIncrease min above RP0 (invalid)...\n");
_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/intel-gfx