On Thu, Sep 9, 2021 at 6:56 AM Michael S. Tsirkin <mst at redhat.com>
wrote:>
> NB: when merging this with
> b542e383d8c0 ("eventfd: Make signal recursion protection a task
bit")
> from Linus' tree, replace eventfd_signal_count with
> eventfd_signal_allowed, and drop the export of eventfd_wake_count from
> ("eventfd: Export eventfd_wake_count to modules").
What? No. That can't be right.
Do you mean "replace eventfd_signal_count with
!eventfd_signal_allowed()"?
Because if I read the logic correctly, the issue is that
'vduse_vq_kick()' will call eventfd_signal().
Which it must not do it eventfd_signal_allowed() returns false.
So if eventfd_signal_allowed() is _not_ set, the code needs to defer
it to the workqueue.
No?
Side note: I was _this_ close to just not pulling this. The commits
were all from after the merge window opened, and I got this pull
request in the latter half of the second week of the merge window.
Your "explanations" for this pull are also not any language I
recognize, or even remotely human-readable.
WTF does "vduse driver supporting blk" mean, and how is that supposed
to explain anything at all?
That is NOT how these things are supposed to work. AT ALL.
So you are hereby put on notice: next time I get this kind of
half-arsed garbage pull request, I won't spend the energy on trying to
figure out what is actually going on. I will just throw it in the
trash.
Because honestly, that's where this pull request belonged.
Linus