Re: [PATCH i-g-t] lib/igt_kms: adapt to boolean debugfs for i915_bigjoiner_force_enable

["Ville Syrjälä" <ville.syrjala@linux.intel.com>]

On Tue, Apr 02, 2024 at 11:12:31AM +0530, Kunal Joshi wrote:
> adapt to boolean debugfs for i915_bigjoiner_force_enable
> 
> Cc: Ankit Nautiyal <ankit.k.nautiyal@intel.com>
> Signed-off-by: Kunal Joshi <kunal1.joshi@intel.com>
> ---
>  lib/igt_kms.c | 4 ++--
>  1 file changed, 2 insertions(+), 2 deletions(-)
> 
> diff --git a/lib/igt_kms.c b/lib/igt_kms.c
> index c38021dd7..2a518eb8d 100644
> --- a/lib/igt_kms.c
> +++ b/lib/igt_kms.c
> @@ -6218,8 +6218,8 @@ bool igt_force_and_check_bigjoiner_status(int drmfd, char *connector_name, bool
>  	close(debugfs_fd);
>  	igt_assert_f(ret > 0, "Could not read i915_bigjoiner_force_enable for connector %s\n", connector_name);
>  
> -	return enable ? strstr(buf, "Bigjoiner enable: 1") :
> -		    strstr(buf, "Bigjoiner enable: 0");
> +	return enable ? strstr(buf, "Y") :
> +			strstr(buf, "N");
>  }

It looks like IGT is forgetting to reset this after running the
bigjoiner tests. Random tests are failing due to the bigjoiner
being left on. Some are just modesets failing due to not getting 
the slave pipe, cursor tests are failing due to missing legacy
cursor fastpath with bigjoiner, and some PSR tests are failing
because the PSR code apparently doesn't work with bigjoiner.

-- 
Ville Syrjälä
Intel