On 24.02.20 г. 19:18 ч., Al Viro wrote: > On Mon, Feb 24, 2020 at 07:01:42PM +0200, Nikolay Borisov wrote: >> Current is_local_mountpoint is a simple wrapper with added d_mountpoint >> check. However, the same check is the first thing which >> __is_local_mountpoint performs. So remove the wrapper and promote the >> private helper to is_local_mountpoint. No semantics changes. > > NAK. "No semantics changes" does not cut it - inline helper that checks > some unlikely condition and calls an out-of-line version is a fairly > common pattern, with legitimate uses. It *may* be unwarranted here, > but you need more serious analysis than that. I'm not saying that > the patch is wrong, but you'll also need to explain why removing the > check from __is_local_mountpoint() (and marking the condition unlikely > in the wrapper) would be worse than what you propose. > My main motivation was to collapse the number of functions so it's easier to see what's going on and removing duplicate check. If you are going to accept the alternative version you proposed I'm fine sending it.
