Re: [x86] d55564cfc2: will-it-scale.per_thread_ops -5.8% regression

[Linus Torvalds]

On Thu, Jan 7, 2021 at 10:34 AM Al Viro <viro@xxxxxxxxxxxxxxxxxx> wrote:
>
> I'm not sure it's the best approach, TBH.  How about simply
>         for (walk = head; walk; ufds += walk->len, walk = walk->next) {
>                 if (copy_to_user(ufds, walk->entries,
>                                  walk->len * sizeof(struct pollfd))
>                         goto out_fds;
>         }
> in there?  It's both simpler (obviously matches the copyin side) and
> might very well be faster...

I started doing that, but ..  Nope.

It's not copying the whole entry. It's literally just modifying one
16-bit word in each entry.

Now, the "whole entry" is just 8 bytes, so it's possible that it would
actually be faster to do a copy of the whole thing rather than write
just the 16 bits. But I got very nervous about it, because I could
easily see some threaded app actually changing the 'fd' (or the
'event' field) in place (ie writing -1 to it as they close and re-use
it)

The man-pages even document that only the 'revent' field is an output parameter.

So I think my patch is a _lot_ safer than your arguably simpler one,
because mine keeps the original semantics.

             Linus