Received-SPF: pass (google.com: best guess record for domain of linux-kernel-owner@vger.kernel.org designates 209.132.180.67 as permitted sender) client-ip=209.132.180.67;
Date:   Wed, 19 Feb 2020 10:37:04 +0000
From:   Stefan Hajnoczi <stefanha@gmail.com>
To:     Avi Kivity <avi@scylladb.com>
Cc:     Paolo Bonzini <pbonzini@redhat.com>,
        Stefan Hajnoczi <stefanha@redhat.com>,
        linux-fsdevel@vger.kernel.org, kvm@vger.kernel.org,
        linux-kernel@vger.kernel.org,
        Davide Libenzi <davidel@xmailserver.org>,
        Alexander Viro <viro@zeniv.linux.org.uk>
Subject: Re: [RFC] eventfd: add EFD_AUTORESET flag
Message-ID: <20200219103704.GA1076032@stefanha-x1.localdomain>
References: <20200129172010.162215-1-stefanha@redhat.com>
 <66566792-58a4-bf65-6723-7d2887c84160@redhat.com>
 <20200212102912.GA464050@stefanha-x1.localdomain>
 <156cb709-282a-ddb6-6f34-82b4bb211f73@redhat.com>
 <cadb4320-4717-1a41-dfb5-bb782fd0a5da@scylladb.com>
MIME-Version: 1.0
Content-Type: multipart/signed; micalg=pgp-sha256;
        protocol="application/pgp-signature"; boundary="bp/iNruPH9dso1Pn"
Content-Disposition: inline
In-Reply-To: <cadb4320-4717-1a41-dfb5-bb782fd0a5da@scylladb.com>
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk


--bp/iNruPH9dso1Pn
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
Content-Transfer-Encoding: quoted-printable

On Wed, Feb 12, 2020 at 12:54:30PM +0200, Avi Kivity wrote:
>=20
> On 12/02/2020 12.47, Paolo Bonzini wrote:
> > On 12/02/20 11:29, Stefan Hajnoczi wrote:
> > > On Wed, Feb 12, 2020 at 09:31:32AM +0100, Paolo Bonzini wrote:
> > > > On 29/01/20 18:20, Stefan Hajnoczi wrote:
> > > > > +	/* Semaphore semantics don't make sense when autoreset is enabl=
ed */
> > > > > +	if ((flags & EFD_SEMAPHORE) && (flags & EFD_AUTORESET))
> > > > > +		return -EINVAL;
> > > > > +
> > > > I think they do, you just want to subtract 1 instead of setting the
> > > > count to 0.  This way, writing 1 would be the post operation on the
> > > > semaphore, while poll() would be the wait operation.
> > > True!  Then EFD_AUTORESET is not a fitting name.  EFD_AUTOREAD or
> > > EFD_POLL_READS?
> > Avi's suggestion also makes sense.  Switching the event loop from poll()
> > to IORING_OP_POLL_ADD would be good on its own, and then you could make
> > it use IORING_OP_READV for eventfds.
> >=20
> > In QEMU parlance, perhaps you need a different abstraction than
> > EventNotifier (let's call it WakeupNotifier) which would also use
> > eventfd but it would provide a smaller API.  Thanks to the smaller API,
> > it would not need EFD_NONBLOCK, unlike the regular EventNotifier, and it
> > could either set up a poll() handler calling read(), or use
> > IORING_OP_READV when io_uring is in use.
> >=20
>=20
> Just to be clear, for best performance don't use IORING_OP_POLL_ADD, just
> IORING_OP_READ. That's what you say in the second paragraph but the first
> can be misleading.

Thanks, that's a nice idea!  I already have experimental io_uring fd
monitoring code written for QEMU and will extend it to use IORING_OP_READ.

Stefan

--bp/iNruPH9dso1Pn
Content-Type: application/pgp-signature; name="signature.asc"

-----BEGIN PGP SIGNATURE-----

iQEzBAEBCAAdFiEEhpWov9P5fNqsNXdanKSrs4Grc8gFAl5ND9AACgkQnKSrs4Gr
c8gMHAf/f9UuwAwjiMV/pwZZbwcYseAmnQ2FPZYE5HhiWScgJAg19otfE4JxYwR5
/NhXUNyYhp7sa0+fB+/aHJF/0mpIY70Mn1PX2umT7kb4ZpP6Qel0ZyVRuNOSp+k5
Wm19LAvKD4AyVu2Kxd4RJpibSu+UYelAT3Nh1d/c8Grrf0zHdcAj4Y4YwfTao1Xo
kPgrIWGOPS7qxvS+fyg9ifT13LfhwFhpP3679AlOhUozQUrxnM77G8ehAOJQV8mS
Ae+//m8fs/RVH2SRNgy7uRIr/v8pWdGt241pW8507oFHJChsRLFhNXJIWHP0dlUU
xHILARbPURoXO+KG1YPtyO2Y+16cZQ==
=5mbR
-----END PGP SIGNATURE-----

--bp/iNruPH9dso1Pn--