Re: [PATCH] drm/xe/lrc: Fix uninitialized new_ts when capturing context timestamp

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

On Wed, 11 Mar 2026 03:49:02 -0700, Umesh Nerlige Ramappa wrote:
>
> Getting engine specific CTX TIMESTAMP register can fail. In that case,
> if the context is active, new_ts is uninitialized. Fix that case by
> initializing new_ts to lrc_ts.
>
> Flagged by static analysis.
>
> Signed-off-by: Umesh Nerlige Ramappa <umesh.nerlige.ramappa@intel.com>
> ---
>  drivers/gpu/drm/xe/xe_lrc.c | 4 +++-
>  1 file changed, 3 insertions(+), 1 deletion(-)
>
> diff --git a/drivers/gpu/drm/xe/xe_lrc.c b/drivers/gpu/drm/xe/xe_lrc.c
> index aa26c71ae34f..9d766ef414ed 100644
> --- a/drivers/gpu/drm/xe/xe_lrc.c
> +++ b/drivers/gpu/drm/xe/xe_lrc.c
> @@ -2563,7 +2563,7 @@ static int get_ctx_timestamp(struct xe_lrc *lrc, u32 engine_id, u64 *reg_ctx_ts)
>   * @lrc: Pointer to the lrc.
>   *
>   * Return latest ctx timestamp. With support for active contexts, the
> - * calculation may bb slightly racy, so follow a read-again logic to ensure that
> + * calculation may be slightly racy, so follow a read-again logic to ensure that
>   * the context is still active before returning the right timestamp.
>   *
>   * Returns: New ctx timestamp value
> @@ -2584,6 +2584,8 @@ u64 xe_lrc_timestamp(struct xe_lrc *lrc)
>		engine_id = xe_lrc_engine_id(lrc);
>		if (!get_ctx_timestamp(lrc, engine_id, &reg_ts))
>			new_ts = reg_ts;
> +		else
> +			new_ts = lrc_ts;

Sorry, don't understand this too well. But looks like here 'lrc_ts ==
CONTEXT_ACTIVE' which is not a real timestamp value (it's value is 1). So
can we return that? See also the final 'if' statement in the function.

>
>		/* read lrc again to ensure context is still active */
>		lrc_ts = xe_lrc_ctx_timestamp(lrc);
> --
> 2.51.0
>