All of lore.kernel.org
 help / color / mirror / Atom feed
From: Eric Engestrom <eric@engestrom.ch>
To: robert.foss@collabora.com
Cc: Daniel Stone <daniels@collabora.com>,
	intel-gfx@lists.freedesktop.org,
	Gustavo Padovan <gustavo.padovan@collabora.com>,
	mesa-dev@lists.freedesktop.org
Subject: Re: [PATCH v1 11/13] tests/sw_sync: Add subtest test_sync_random_merge
Date: Tue, 23 Aug 2016 23:38:43 +0100	[thread overview]
Message-ID: <20160823223842.GC5972@engestrom.ch> (raw)
In-Reply-To: <1471974975-11531-12-git-send-email-robert.foss@collabora.com>

On Tue, Aug 23, 2016 at 01:56:13PM -0400, robert.foss@collabora.com wrote:
> From: Robert Foss <robert.foss@collabora.com>
> 
> This subtest verifies that creating many timelines and merging random fences
> from each timeline with eachother results in merged fences that are fully
> functional.
> 
> Signed-off-by: Robert Foss <robert.foss@collabora.com>
> ---
>  tests/sw_sync.c | 73 +++++++++++++++++++++++++++++++++++++++++++++++++++++++++
>  1 file changed, 73 insertions(+)
> 
> diff --git a/tests/sw_sync.c b/tests/sw_sync.c
> index 0e67ad5..8e5a7c9 100644
> --- a/tests/sw_sync.c
> +++ b/tests/sw_sync.c
> @@ -383,6 +383,76 @@ static void test_sync_multi_consumer_producer(void)
>  	igt_assert_f(thread_ret == 0, "A sync thread reported failure.\n");
>  }
>  
> +static void test_sync_random_merge(void)
> +{
> +	int i, size, ret;
> +	const int nbr_timeline = 32;
> +	const int nbr_merge = 1024;
> +	int fence_map[nbr_timeline];
> +	int timeline_arr[nbr_timeline];
> +	int fence, tmpfence, merged;
> +	int timeline, timeline_offset, sync_pt;
> +
> +	srand(time(NULL));
> +
> +	for (i = 0; i < nbr_timeline; i++)
> +		timeline_arr[i] = sw_sync_timeline_create();
> +
> +	memset(fence_map, -1, sizeof(fence_map));

This sets each byte to -1, which happens to be the same as an int -1,
but I don't really like it: this is only true for two special values, -1
and 0, and would become a bug if it was ever changed.

I'd much prefer if you simply set:
	fence_map[i] = -1;
in the loop you already have just above.

BTW, seeing how many times you set and test for -1 as the "invalid fd
sentinel", how about a:
	#define INVALID_FD (-1)
and use that token throughout the code? I think it would improve
readability as well

> +
> +	sync_pt = rand();
> +	fence = sw_sync_fence_create(timeline_arr[0], sync_pt);
> +
> +	fence_map[0] = sync_pt;
> +
> +	/* Randomly create syncpoints out of a fixed set of timelines,
> +	 * and merge them together.
> +	 */
> +	for (i = 0; i < nbr_merge; i++) {
> +		/* Generate syncpoint. */
> +		timeline_offset = rand() % nbr_timeline;
> +		timeline = timeline_arr[timeline_offset];
> +		sync_pt = rand();
> +
> +		/* Keep track of the latest sync_pt in each timeline. */
> +		if (fence_map[timeline_offset] == -1)
> +			fence_map[timeline_offset] = sync_pt;
> +		else if (fence_map[timeline_offset] < sync_pt)
> +			fence_map[timeline_offset] = sync_pt;
> +
> +		/* Merge. */
> +		tmpfence = sw_sync_fence_create(timeline, sync_pt);
> +		merged = sw_sync_merge(tmpfence, fence);
> +		sw_sync_fence_destroy(tmpfence);
> +		sw_sync_fence_destroy(fence);
> +		fence = merged;
> +	}
> +
> +	size = 0;
> +	for (i = 0; i < nbr_timeline; i++)
> +		if (fence_map[i] != -1)
> +			size++;
> +
> +	/* Trigger the merged fence. */
> +	for (i = 0; i < nbr_timeline; i++) {
> +		if (fence_map[i] != -1) {
> +			ret = sw_sync_wait(fence, 0);
> +			igt_assert_f(ret == 0,
> +				    "Failure waiting on fence until timeout\n");
> +			/* Increment the timeline to the last sync_pt */
> +			sw_sync_timeline_inc(timeline_arr[i], fence_map[i]);
> +		}
> +	}
> +
> +	/* Check that the fence is triggered. */
> +	ret = sw_sync_wait(fence, 0);
> +	igt_assert_f(ret > 0, "Failure triggering fence\n");
> +
> +	sw_sync_fence_destroy(fence);
> +	for (i = 0; i < nbr_timeline; i++)
> +		sw_sync_timeline_destroy(timeline_arr[i]);
> +}
> +
>  igt_main
>  {
>  	igt_subtest("alloc_timeline")
> @@ -411,5 +481,8 @@ igt_main
>  
>  	igt_subtest("sync_multi_consumer_producer")
>  		test_sync_multi_consumer_producer();
> +
> +	igt_subtest("sync_random_merge")
> +		test_sync_random_merge();
>  }
>  
> -- 
> 2.7.4
_______________________________________________
mesa-dev mailing list
mesa-dev@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/mesa-dev

  reply	other threads:[~2016-08-23 22:38 UTC|newest]

Thread overview: 24+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2016-08-23 17:56 [PATCH v1 00/13] Implement sw_sync test robert.foss
2016-08-23 17:56 ` [PATCH v1 01/13] lib/sw_sync: Add helper functions for managing synchronization primitives robert.foss
2016-08-23 22:05   ` [Mesa-dev] " Eric Engestrom
2016-08-23 17:56 ` [PATCH v1 02/13] tests/sw_sync: Add sw_sync test robert.foss
2016-08-23 22:12   ` [Mesa-dev] " Eric Engestrom
2016-08-23 17:56 ` [PATCH v1 03/13] tests/sw_sync: Add subtest test_alloc_fence robert.foss
2016-08-23 17:56 ` [PATCH v1 04/13] tests/sw_sync: Add subtest test_alloc_fence_invalid_timeline robert.foss
2016-08-23 17:56 ` [PATCH v1 05/13] tests/sw_sync: Add subtest test_alloc_merge_fence robert.foss
2016-08-23 17:56 ` [PATCH v1 06/13] tests/sw_sync: Add subtest test_sync_wait robert.foss
2016-08-23 17:56 ` [PATCH v1 07/13] tests/sw_sync: Add subtest test_sync_merge robert.foss
2016-08-23 17:56 ` [PATCH v1 08/13] tests/sw_sync: Add subtest test_sync_merge_same robert.foss
2016-08-23 17:56 ` [PATCH v1 09/13] tests/sw_sync: Add subtest test_sync_multi_consumer robert.foss
2016-08-23 17:56 ` [PATCH v1 10/13] tests/sw_sync: Add subtest test_sync_multi_consumer_producer robert.foss
2016-08-23 17:56 ` [PATCH v1 11/13] tests/sw_sync: Add subtest test_sync_random_merge robert.foss
2016-08-23 22:38   ` Eric Engestrom [this message]
2016-08-23 17:56 ` [PATCH v1 12/13] tests/sw_sync: Add subtest test_sync_multi_timeline_wait robert.foss
2016-08-23 17:56 ` [PATCH v1 13/13] tests/sw_sync: Add subtest test_sync_multi_producer_single_consumer robert.foss
2016-08-23 22:56 ` [Mesa-dev] [PATCH v1 00/13] Implement sw_sync test Eric Engestrom
2016-08-24 20:27   ` Robert Foss
2016-08-24  6:01 ` Daniel Vetter
2016-08-24 13:51   ` Robert Foss
2016-08-24 20:41   ` Gustavo Padovan
2016-08-25  6:56     ` Daniel Vetter
  -- strict thread matches above, loose matches on Subject: below --
2016-08-23 17:53 robert.foss
2016-08-23 17:53 ` [PATCH v1 11/13] tests/sw_sync: Add subtest test_sync_random_merge robert.foss

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20160823223842.GC5972@engestrom.ch \
    --to=eric@engestrom.ch \
    --cc=daniels@collabora.com \
    --cc=gustavo.padovan@collabora.com \
    --cc=intel-gfx@lists.freedesktop.org \
    --cc=mesa-dev@lists.freedesktop.org \
    --cc=robert.foss@collabora.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.