On Thu, 31 Oct 2024 at 12:02, Linus Torvalds
<torvalds@xxxxxxxxxxxxxxxxxxxx> wrote:
>
> I still get a fair number of calls to make_vfsuid() even with my patch
> - I guess I'll have to look into why. This is my regular "full build
> of an already built kernel", which I *expected* to be mainly just a
> lot of fstat() calls by 'make' which I would have thought almost
> always hit the fast case.
>
> I might have messed something up.
Added some stats, and on my load (reading email in the web browser,
some xterms and running an allmodconfig kernel build), I get about a
45% hit-rate for the fast-case: out of 44M calls to
generic_permission(), about 20M hit the fast-case path.
So it's noticeable, but not *quite* as noticeable as I would have
hoped for. I suspect there are a fair number of owner calls for write,
and then because permissions are 0644, the code actually has to check
the owner.
Linus
