[igt-dev] [PATCH i-g-t] i915/gem_spin_batch: Use __for_each_physical_engine

[Chris Wilson]

Quoting Ashutosh Dixit (2020-05-09 01:21:24)
> Instead of the legacy for_each_physical_engine use
> __for_each_physical_engine to utilize all engines.
> 
> Fixes: 57efd81037 ("i915/gem_spin_batch: Add test to execute in parallel on all engines")
> Signed-off-by: Ashutosh Dixit <ashutosh.dixit at intel.com>
> ---
>  tests/i915/gem_spin_batch.c | 5 +++--
>  1 file changed, 3 insertions(+), 2 deletions(-)
> 
> diff --git a/tests/i915/gem_spin_batch.c b/tests/i915/gem_spin_batch.c
> index e7dd58ec2..6a7753ced 100644
> --- a/tests/i915/gem_spin_batch.c
> +++ b/tests/i915/gem_spin_batch.c
> @@ -137,10 +137,11 @@ static void spin_on_all_engines(int fd, unsigned int timeout_sec)
>  static void spin_all(int i915, unsigned int flags)
>  #define PARALLEL_SPIN_NEW_CTX BIT(0)
>  {
> +       const struct intel_execution_engine2 *e2;
>         struct igt_spin *spin, *n;
>         IGT_LIST_HEAD(list);
>  
> -       for_each_physical_engine(e, i915) {
> +       __for_each_physical_engine(i915, e2) {

This is intentionally using the legacy interface; there's a variant to
use the ctx->engines[] path as well.
-Chris