On Wed, Aug 21, 2024 at 07:28:08PM +0100, Mark Brown wrote: > part of an otherwise invalid write. The validation is checking for > unknown features and enables. With clone3() we could relax the enable > check, but I've just pulled that out of the series for the time being. Actually thinking about it some more I'll just remove the check for enable, the support for threads with GCS enabled and no kernel allocated GCS is already there and I didn't pull that bit out.
Attachment:
signature.asc
Description: PGP signature