Re: [PATCH v7 2/2] tests/intel/xe_oa: Wa_14026633728

["Dixit, Ashutosh" <ashutosh.dixit@intel.com>]

On Thu, 07 May 2026 19:13:29 -0700, Shekhar Chauhan wrote:
>
> For MERTOA in CRI, oa buffer can be in device memory. Because of slower
> device mem access, OA exponent values lower than 8 can result in buffer
> overflows. Bump the OA exponent value.
>
> Signed-off-by: Shekhar Chauhan <shekhar.chauhan@intel.com>
> ---
>  tests/intel/xe_oa.c | 13 +++++++++++++
>  1 file changed, 13 insertions(+)
>
> diff --git a/tests/intel/xe_oa.c b/tests/intel/xe_oa.c
> index 988c46df6..2f73b5210 100644
> --- a/tests/intel/xe_oa.c
> +++ b/tests/intel/xe_oa.c
> @@ -24,6 +24,7 @@
>  #include "igt_device.h"
>  #include "igt_syncobj.h"
>  #include "igt_sysfs.h"
> +#include "intel_wa.h"
>  #include "xe/xe_ioctl.h"
>  #include "xe/xe_query.h"
>  #include "xe/xe_oa.h"
> @@ -2671,6 +2672,18 @@ test_non_zero_reason(const struct drm_xe_oa_unit *oau, size_t oa_buffer_size)
>	int len, check_idx;
>	u32 oa_status;
>
> +	/*
> +	 * Wa_14026633728: For MERTOA in CRI, oa buffer can be in device memory.
> +	 * Because of slower device mem access, OA exponent values lower than 8
> +	 * can result in buffer overflows. Also, update the value of oa_exponent
> +	 * in properties[].

This last part in the comment, "Also, update the value of oa_exponent in
properties[]" is obvious from reading the code. It should not be documented
in comments.

Anyway I have fixed this comment up this time and merged this
series. Thanks.

> +	 */
> +	if (oau->oa_unit_type == DRM_XE_OA_UNIT_TYPE_MERT &&
> +		igt_has_intel_wa(drm_fd, "14026633728")) {
> +		oa_exponent = max(oa_exponent, 8);
> +		properties[9] = oa_exponent;
> +	}
> +
>	igt_assert(buf);
>
>	igt_debug("Ready to read about %u bytes\n", buf_size);
> --
> 2.53.0
>