Quoting Mika Kuoppala (2020-06-30 17:16:43)
> Chris Wilson <chris@xxxxxxxxxxxxxxxxxx> writes:
>
> > As we close a handle GEM object, we update the drm_file's idr with an
> > error pointer to indicate the in-progress closure, and finally set it to
>
> The error pointer part stage seems to be missing.
Yeah, the ERR_PTR stage seems to be my faulty memory, we just set it to
NULL to indicate in-progress. Ok, I'm not going totally mad:
commit f6cd7daecff558fab2c45d15283d3e52f688342d
Author: Chris Wilson <chris@xxxxxxxxxxxxxxxxxx>
Date: Fri Apr 15 12:55:08 2016 +0100
drm: Release driver references to handle before making it available again
...
v2: Use NULL rather than an ERR_PTR to avoid having to adjust callers.
idr_alloc() tracks existing handles using an internal bitmap, so we are
free to use the NULL object as our stale identifier.
-Chris
