On Wed, Oct 16, 2024 at 12:51:23PM +0100, Matthew Wilcox wrote: > > It's obviously a try_upgrade_read, right? > > Well, that's confusing. "try" usually means "don't sleep", and this > sleeps. Maybe it shouldn't sleep; ie we make this fail if there's any > other reader? It'll succeed less often, but it'll be easier to > understand. To me try primarily implies that it can fail and the return value needs to be checked. But I guess it has different implications to different people.
