From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org X-Spam-Level: X-Spam-Status: No, score=-5.4 required=3.0 tests=BAYES_00,DKIM_SIGNED, DKIM_VALID,DKIM_VALID_AU,HEADER_FROM_DIFFERENT_DOMAINS,MAILING_LIST_MULTI, SPF_HELO_NONE,SPF_PASS,USER_AGENT_SANE_2 autolearn=no autolearn_force=no version=3.4.0 Received: from mail.kernel.org (mail.kernel.org [198.145.29.99]) by smtp.lore.kernel.org (Postfix) with ESMTP id 31E84C4363A for ; Tue, 27 Oct 2020 10:15:55 +0000 (UTC) Received: from vger.kernel.org (vger.kernel.org [23.128.96.18]) by mail.kernel.org (Postfix) with ESMTP id 8BD4A22264 for ; Tue, 27 Oct 2020 10:15:52 +0000 (UTC) Authentication-Results: mail.kernel.org; dkim=pass (2048-bit key) header.d=infradead.org header.i=@infradead.org header.b="i9NPGKG5" Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S2897669AbgJ0KPv (ORCPT ); Tue, 27 Oct 2020 06:15:51 -0400 Received: from merlin.infradead.org ([205.233.59.134]:44160 "EHLO merlin.infradead.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S2897649AbgJ0KPt (ORCPT ); Tue, 27 Oct 2020 06:15:49 -0400 DKIM-Signature: v=1; a=rsa-sha256; q=dns/txt; c=relaxed/relaxed; d=infradead.org; s=merlin.20170209; h=Mime-Version:Content-Type:References: In-Reply-To:Date:To:From:Subject:Message-ID:Sender:Reply-To:Cc: Content-Transfer-Encoding:Content-ID:Content-Description; bh=qb563W1vqZmMkjF7LlJizia7QGOlYAmbTQbfPRbAxSs=; b=i9NPGKG5pPdHLsQU4MAZ7TBomx zn9hcnR5Rr1q1txsQT4Ugp7ejTgciSBAYv3MFaJskLHbAA05Hc677KecFv85F7KbFCbaqX+K3x6mx fNMkElSIB22meYAu72I69x7FT1glNyufY6U8nVwCSuaWs0TL5SOvcJ5WvpFbI2CfD4Uwiw+gV/2QH eQk/XFnYY3lak4oAs9GSQr4AdD1pgdKEsjpt2Ti7n63FdP3jhpZonA8V12OA6d1TX636puTImQvgw tejYun1P8VBL9vsSLsNSOE/6ig+YmQ+Qm/9MvRfOla+oZ+D/s6Trj3vNK2M3VYfWKPtQDkL6d3iWd e+QCZMiQ==; Received: from 54-240-197-239.amazon.com ([54.240.197.239] helo=edge-m3-r3-171.e-iad51.amazon.com) by merlin.infradead.org with esmtpsa (Exim 4.92.3 #3 (Red Hat Linux)) id 1kXM0l-0003DU-Cl; Tue, 27 Oct 2020 10:15:27 +0000 Message-ID: Subject: Re: [RFC PATCH 2/2] kvm/eventfd: Use priority waitqueue to catch events before userspace From: David Woodhouse To: Paolo Bonzini , Ingo Molnar , Peter Zijlstra , Juri Lelli , Vincent Guittot , Dietmar Eggemann , Steven Rostedt , Ben Segall , Mel Gorman , Daniel Bristot de Oliveira , linux-kernel@vger.kernel.org, kvm@vger.kernel.org Date: Tue, 27 Oct 2020 10:15:24 +0000 In-Reply-To: <1faa5405-3640-f4ad-5cd9-89a9e5e834e9@redhat.com> References: <20201026175325.585623-1-dwmw2@infradead.org> <20201026175325.585623-2-dwmw2@infradead.org> <1faa5405-3640-f4ad-5cd9-89a9e5e834e9@redhat.com> Content-Type: multipart/signed; micalg="sha-256"; protocol="application/x-pkcs7-signature"; boundary="=-Jgd4iQZKFZLBFQGP2AoD" X-Mailer: Evolution 3.28.5-0ubuntu0.18.04.2 Mime-Version: 1.0 X-SRS-Rewrite: SMTP reverse-path rewritten from by merlin.infradead.org. See http://www.infradead.org/rpr.html Precedence: bulk List-ID: X-Mailing-List: linux-kernel@vger.kernel.org --=-Jgd4iQZKFZLBFQGP2AoD Content-Type: text/plain; charset="UTF-8" Content-Transfer-Encoding: quoted-printable On Tue, 2020-10-27 at 09:01 +0100, Paolo Bonzini wrote: > On 26/10/20 18:53, David Woodhouse wrote: > > From: David Woodhouse > >=20 > > As far as I can tell, when we use posted interrupts we silently cut off > > the events from userspace, if it's listening on the same eventfd that > > feeds the irqfd. > >=20 > > I like that behaviour. Let's do it all the time, even without posted > > interrupts. It makes it much easier to handle IRQ remapping invalidatio= n > > without having to constantly add/remove the fd from the userspace poll > > set. We can just leave userspace polling on it, and the bypass will... > > well... bypass it. >=20 > This looks good, though of course it depends on the somewhat hackish > patch 1. I thought it was quite neat :) > However don't you need to read the eventfd as well, since > userspace will never be able to do so? Yes. Although that's a separate cleanup as it was already true before my patch. Right now, userspace needs to explicitly stop polling on the VFIO eventfd while it's assigned as KVM IRQFD (to avoid injecting duplicate interrupts when the kernel isn't using PI and allows events to leak). So it isn't going to consume the events in that case either. Nothing's really changed. The VFIO virqfd is just the same. The count just builds up when the kernel handles the events, and is eventually cleared by eventfd_ctx_remove_wait_queue(). In both cases, that actually works fine because in practice the events are raised by eventfd_signal() in the kernel, and that works even if the count reaches ULLONG_MAX. It's just that sending further events from *userspace* would block in that case. Both of them theoretically want fixing =E2=80=94 regardless of the priority patch. Since the wq lock is held while the wakeup function (virqfd_wakeup or=20 irqfd_wakeup for VFIO/KVM respectively) run, all they really need to do is call eventfd_ctx_do_read() to consume the events. I'll look at whether I can find a nicer option than just exporting that. --=-Jgd4iQZKFZLBFQGP2AoD Content-Type: application/x-pkcs7-signature; name="smime.p7s" Content-Disposition: attachment; filename="smime.p7s" Content-Transfer-Encoding: base64 MIAGCSqGSIb3DQEHAqCAMIACAQExDzANBglghkgBZQMEAgEFADCABgkqhkiG9w0BBwEAAKCCECow ggUcMIIEBKADAgECAhEA4rtJSHkq7AnpxKUY8ZlYZjANBgkqhkiG9w0BAQsFADCBlzELMAkGA1UE BhMCR0IxGzAZBgNVBAgTEkdyZWF0ZXIgTWFuY2hlc3RlcjEQMA4GA1UEBxMHU2FsZm9yZDEaMBgG A1UEChMRQ09NT0RPIENBIExpbWl0ZWQxPTA7BgNVBAMTNENPTU9ETyBSU0EgQ2xpZW50IEF1dGhl bnRpY2F0aW9uIGFuZCBTZWN1cmUgRW1haWwgQ0EwHhcNMTkwMTAyMDAwMDAwWhcNMjIwMTAxMjM1 OTU5WjAkMSIwIAYJKoZIhvcNAQkBFhNkd213MkBpbmZyYWRlYWQub3JnMIIBIjANBgkqhkiG9w0B AQEFAAOCAQ8AMIIBCgKCAQEAsv3wObLTCbUA7GJqKj9vHGf+Fa+tpkO+ZRVve9EpNsMsfXhvFpb8 RgL8vD+L133wK6csYoDU7zKiAo92FMUWaY1Hy6HqvVr9oevfTV3xhB5rQO1RHJoAfkvhy+wpjo7Q cXuzkOpibq2YurVStHAiGqAOMGMXhcVGqPuGhcVcVzVUjsvEzAV9Po9K2rpZ52FE4rDkpDK1pBK+ uOAyOkgIg/cD8Kugav5tyapydeWMZRJQH1vMQ6OVT24CyAn2yXm2NgTQMS1mpzStP2ioPtTnszIQ Ih7ASVzhV6csHb8Yrkx8mgllOyrt9Y2kWRRJFm/FPRNEurOeNV6lnYAXOymVJwIDAQABo4IB0zCC Ac8wHwYDVR0jBBgwFoAUgq9sjPjF/pZhfOgfPStxSF7Ei8AwHQYDVR0OBBYEFLfuNf820LvaT4AK xrGK3EKx1DE7MA4GA1UdDwEB/wQEAwIFoDAMBgNVHRMBAf8EAjAAMB0GA1UdJQQWMBQGCCsGAQUF BwMEBggrBgEFBQcDAjBGBgNVHSAEPzA9MDsGDCsGAQQBsjEBAgEDBTArMCkGCCsGAQUFBwIBFh1o dHRwczovL3NlY3VyZS5jb21vZG8ubmV0L0NQUzBaBgNVHR8EUzBRME+gTaBLhklodHRwOi8vY3Js LmNvbW9kb2NhLmNvbS9DT01PRE9SU0FDbGllbnRBdXRoZW50aWNhdGlvbmFuZFNlY3VyZUVtYWls Q0EuY3JsMIGLBggrBgEFBQcBAQR/MH0wVQYIKwYBBQUHMAKGSWh0dHA6Ly9jcnQuY29tb2RvY2Eu Y29tL0NPTU9ET1JTQUNsaWVudEF1dGhlbnRpY2F0aW9uYW5kU2VjdXJlRW1haWxDQS5jcnQwJAYI KwYBBQUHMAGGGGh0dHA6Ly9vY3NwLmNvbW9kb2NhLmNvbTAeBgNVHREEFzAVgRNkd213MkBpbmZy YWRlYWQub3JnMA0GCSqGSIb3DQEBCwUAA4IBAQALbSykFusvvVkSIWttcEeifOGGKs7Wx2f5f45b nv2ghcxK5URjUvCnJhg+soxOMoQLG6+nbhzzb2rLTdRVGbvjZH0fOOzq0LShq0EXsqnJbbuwJhK+ PnBtqX5O23PMHutP1l88AtVN+Rb72oSvnD+dK6708JqqUx2MAFLMevrhJRXLjKb2Mm+/8XBpEw+B 7DisN4TMlLB/d55WnT9UPNHmQ+3KFL7QrTO8hYExkU849g58Dn3Nw3oCbMUgny81ocrLlB2Z5fFG Qu1AdNiBA+kg/UxzyJZpFbKfCITd5yX49bOriL692aMVDyqUvh8fP+T99PqorH4cIJP6OxSTdxKM MIIFHDCCBASgAwIBAgIRAOK7SUh5KuwJ6cSlGPGZWGYwDQYJKoZIhvcNAQELBQAwgZcxCzAJBgNV BAYTAkdCMRswGQYDVQQIExJHcmVhdGVyIE1hbmNoZXN0ZXIxEDAOBgNVBAcTB1NhbGZvcmQxGjAY BgNVBAoTEUNPTU9ETyBDQSBMaW1pdGVkMT0wOwYDVQQDEzRDT01PRE8gUlNBIENsaWVudCBBdXRo ZW50aWNhdGlvbiBhbmQgU2VjdXJlIEVtYWlsIENBMB4XDTE5MDEwMjAwMDAwMFoXDTIyMDEwMTIz NTk1OVowJDEiMCAGCSqGSIb3DQEJARYTZHdtdzJAaW5mcmFkZWFkLm9yZzCCASIwDQYJKoZIhvcN AQEBBQADggEPADCCAQoCggEBALL98Dmy0wm1AOxiaio/bxxn/hWvraZDvmUVb3vRKTbDLH14bxaW /EYC/Lw/i9d98CunLGKA1O8yogKPdhTFFmmNR8uh6r1a/aHr301d8YQea0DtURyaAH5L4cvsKY6O 0HF7s5DqYm6tmLq1UrRwIhqgDjBjF4XFRqj7hoXFXFc1VI7LxMwFfT6PStq6WedhROKw5KQytaQS vrjgMjpICIP3A/CroGr+bcmqcnXljGUSUB9bzEOjlU9uAsgJ9sl5tjYE0DEtZqc0rT9oqD7U57My ECIewElc4VenLB2/GK5MfJoJZTsq7fWNpFkUSRZvxT0TRLqznjVepZ2AFzsplScCAwEAAaOCAdMw ggHPMB8GA1UdIwQYMBaAFIKvbIz4xf6WYXzoHz0rcUhexIvAMB0GA1UdDgQWBBS37jX/NtC72k+A CsaxitxCsdQxOzAOBgNVHQ8BAf8EBAMCBaAwDAYDVR0TAQH/BAIwADAdBgNVHSUEFjAUBggrBgEF BQcDBAYIKwYBBQUHAwIwRgYDVR0gBD8wPTA7BgwrBgEEAbIxAQIBAwUwKzApBggrBgEFBQcCARYd aHR0cHM6Ly9zZWN1cmUuY29tb2RvLm5ldC9DUFMwWgYDVR0fBFMwUTBPoE2gS4ZJaHR0cDovL2Ny bC5jb21vZG9jYS5jb20vQ09NT0RPUlNBQ2xpZW50QXV0aGVudGljYXRpb25hbmRTZWN1cmVFbWFp bENBLmNybDCBiwYIKwYBBQUHAQEEfzB9MFUGCCsGAQUFBzAChklodHRwOi8vY3J0LmNvbW9kb2Nh LmNvbS9DT01PRE9SU0FDbGllbnRBdXRoZW50aWNhdGlvbmFuZFNlY3VyZUVtYWlsQ0EuY3J0MCQG CCsGAQUFBzABhhhodHRwOi8vb2NzcC5jb21vZG9jYS5jb20wHgYDVR0RBBcwFYETZHdtdzJAaW5m cmFkZWFkLm9yZzANBgkqhkiG9w0BAQsFAAOCAQEAC20spBbrL71ZEiFrbXBHonzhhirO1sdn+X+O W579oIXMSuVEY1LwpyYYPrKMTjKECxuvp24c829qy03UVRm742R9Hzjs6tC0oatBF7KpyW27sCYS vj5wbal+TttzzB7rT9ZfPALVTfkW+9qEr5w/nSuu9PCaqlMdjABSzHr64SUVy4ym9jJvv/FwaRMP gew4rDeEzJSwf3eeVp0/VDzR5kPtyhS+0K0zvIWBMZFPOPYOfA59zcN6AmzFIJ8vNaHKy5QdmeXx RkLtQHTYgQPpIP1Mc8iWaRWynwiE3ecl+PWzq4i+vdmjFQ8qlL4fHz/k/fT6qKx+HCCT+jsUk3cS jDCCBeYwggPOoAMCAQICEGqb4Tg7/ytrnwHV2binUlYwDQYJKoZIhvcNAQEMBQAwgYUxCzAJBgNV BAYTAkdCMRswGQYDVQQIExJHcmVhdGVyIE1hbmNoZXN0ZXIxEDAOBgNVBAcTB1NhbGZvcmQxGjAY BgNVBAoTEUNPTU9ETyBDQSBMaW1pdGVkMSswKQYDVQQDEyJDT01PRE8gUlNBIENlcnRpZmljYXRp b24gQXV0aG9yaXR5MB4XDTEzMDExMDAwMDAwMFoXDTI4MDEwOTIzNTk1OVowgZcxCzAJBgNVBAYT AkdCMRswGQYDVQQIExJHcmVhdGVyIE1hbmNoZXN0ZXIxEDAOBgNVBAcTB1NhbGZvcmQxGjAYBgNV BAoTEUNPTU9ETyBDQSBMaW1pdGVkMT0wOwYDVQQDEzRDT01PRE8gUlNBIENsaWVudCBBdXRoZW50 aWNhdGlvbiBhbmQgU2VjdXJlIEVtYWlsIENBMIIBIjANBgkqhkiG9w0BAQEFAAOCAQ8AMIIBCgKC AQEAvrOeV6wodnVAFsc4A5jTxhh2IVDzJXkLTLWg0X06WD6cpzEup/Y0dtmEatrQPTRI5Or1u6zf +bGBSyD9aH95dDSmeny1nxdlYCeXIoymMv6pQHJGNcIDpFDIMypVpVSRsivlJTRENf+RKwrB6vcf WlP8dSsE3Rfywq09N0ZfxcBa39V0wsGtkGWC+eQKiz4pBZYKjrc5NOpG9qrxpZxyb4o4yNNwTqza aPpGRqXB7IMjtf7tTmU2jqPMLxFNe1VXj9XB1rHvbRikw8lBoNoSWY66nJN/VCJv5ym6Q0mdCbDK CMPybTjoNCQuelc0IAaO4nLUXk0BOSxSxt8kCvsUtQIDAQABo4IBPDCCATgwHwYDVR0jBBgwFoAU u69+Aj36pvE8hI6t7jiY7NkyMtQwHQYDVR0OBBYEFIKvbIz4xf6WYXzoHz0rcUhexIvAMA4GA1Ud DwEB/wQEAwIBhjASBgNVHRMBAf8ECDAGAQH/AgEAMBEGA1UdIAQKMAgwBgYEVR0gADBMBgNVHR8E RTBDMEGgP6A9hjtodHRwOi8vY3JsLmNvbW9kb2NhLmNvbS9DT01PRE9SU0FDZXJ0aWZpY2F0aW9u QXV0aG9yaXR5LmNybDBxBggrBgEFBQcBAQRlMGMwOwYIKwYBBQUHMAKGL2h0dHA6Ly9jcnQuY29t b2RvY2EuY29tL0NPTU9ET1JTQUFkZFRydXN0Q0EuY3J0MCQGCCsGAQUFBzABhhhodHRwOi8vb2Nz cC5jb21vZG9jYS5jb20wDQYJKoZIhvcNAQEMBQADggIBAHhcsoEoNE887l9Wzp+XVuyPomsX9vP2 SQgG1NgvNc3fQP7TcePo7EIMERoh42awGGsma65u/ITse2hKZHzT0CBxhuhb6txM1n/y78e/4ZOs 0j8CGpfb+SJA3GaBQ+394k+z3ZByWPQedXLL1OdK8aRINTsjk/H5Ns77zwbjOKkDamxlpZ4TKSDM KVmU/PUWNMKSTvtlenlxBhh7ETrN543j/Q6qqgCWgWuMAXijnRglp9fyadqGOncjZjaaSOGTTFB+ E2pvOUtY+hPebuPtTbq7vODqzCM6ryEhNhzf+enm0zlpXK7q332nXttNtjv7VFNYG+I31gnMrwfH M5tdhYF/8v5UY5g2xANPECTQdu9vWPoqNSGDt87b3gXb1AiGGaI06vzgkejL580ul+9hz9D0S0U4 jkhJiA7EuTecP/CFtR72uYRBcunwwH3fciPjviDDAI9SnC/2aPY8ydehzuZutLbZdRJ5PDEJM/1t yZR2niOYihZ+FCbtf3D9mB12D4ln9icgc7CwaxpNSCPt8i/GqK2HsOgkL3VYnwtx7cJUmpvVdZ4o gnzgXtgtdk3ShrtOS1iAN2ZBXFiRmjVzmehoMof06r1xub+85hFQzVxZx5/bRaTKTlL8YXLI8nAb R9HWdFqzcOoB/hxfEyIQpx9/s81rgzdEZOofSlZHynoSMYIDyjCCA8YCAQEwga0wgZcxCzAJBgNV BAYTAkdCMRswGQYDVQQIExJHcmVhdGVyIE1hbmNoZXN0ZXIxEDAOBgNVBAcTB1NhbGZvcmQxGjAY BgNVBAoTEUNPTU9ETyBDQSBMaW1pdGVkMT0wOwYDVQQDEzRDT01PRE8gUlNBIENsaWVudCBBdXRo ZW50aWNhdGlvbiBhbmQgU2VjdXJlIEVtYWlsIENBAhEA4rtJSHkq7AnpxKUY8ZlYZjANBglghkgB ZQMEAgEFAKCCAe0wGAYJKoZIhvcNAQkDMQsGCSqGSIb3DQEHATAcBgkqhkiG9w0BCQUxDxcNMjAx MDI3MTAxNTI0WjAvBgkqhkiG9w0BCQQxIgQgDiHqMBSBX+iiwDRC7w+bRZd+ooVjiUBvFzLVy6jt fF0wgb4GCSsGAQQBgjcQBDGBsDCBrTCBlzELMAkGA1UEBhMCR0IxGzAZBgNVBAgTEkdyZWF0ZXIg TWFuY2hlc3RlcjEQMA4GA1UEBxMHU2FsZm9yZDEaMBgGA1UEChMRQ09NT0RPIENBIExpbWl0ZWQx PTA7BgNVBAMTNENPTU9ETyBSU0EgQ2xpZW50IEF1dGhlbnRpY2F0aW9uIGFuZCBTZWN1cmUgRW1h aWwgQ0ECEQDiu0lIeSrsCenEpRjxmVhmMIHABgsqhkiG9w0BCRACCzGBsKCBrTCBlzELMAkGA1UE BhMCR0IxGzAZBgNVBAgTEkdyZWF0ZXIgTWFuY2hlc3RlcjEQMA4GA1UEBxMHU2FsZm9yZDEaMBgG A1UEChMRQ09NT0RPIENBIExpbWl0ZWQxPTA7BgNVBAMTNENPTU9ETyBSU0EgQ2xpZW50IEF1dGhl bnRpY2F0aW9uIGFuZCBTZWN1cmUgRW1haWwgQ0ECEQDiu0lIeSrsCenEpRjxmVhmMA0GCSqGSIb3 DQEBAQUABIIBACPCSgdhjA4BGNbxcQL0I7TARsN53cd4ZKMCBuvRsB7HIpWarClRHaWJqnaS/5rP 3/iSLErSn7S4/PdQcCtCE3Cu+4XdjhjKq0CdF1nSpBpdlksydGey0eLz60fdwl6WmJEGgCZqyhXX 67uYL8rfLHaN8wubYABJWOy6eSvinRoQAt24tuxn77PJ+4ONPFyiqFfeOkj2FkjjDY7VoWgQWbMI VEwDvbGIK/+zuPHP2Mkej8H4ojhh5XK6MYtrmfegLZTC1wmLCnfy5HrRDtEgmF+Du9jrLwrDJttI 5ycqiWnSY0ZFJmylONKO+6BQotV/8+amlFSH/e0CeLPrFk0UdF0AAAAAAAA= --=-Jgd4iQZKFZLBFQGP2AoD--