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=-7.5 required=3.0 tests=BAYES_00,DKIM_SIGNED, DKIM_VALID,DKIM_VALID_AU,HEADER_FROM_DIFFERENT_DOMAINS,MAILING_LIST_MULTI, NICE_REPLY_A,SPF_HELO_NONE,SPF_PASS,USER_AGENT_SANE_1 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 3729BC433E0 for ; Mon, 8 Feb 2021 09:48:14 +0000 (UTC) Received: from vger.kernel.org (vger.kernel.org [23.128.96.18]) by mail.kernel.org (Postfix) with ESMTP id D5A6F64E27 for ; Mon, 8 Feb 2021 09:48:13 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S231944AbhBHJrc (ORCPT ); Mon, 8 Feb 2021 04:47:32 -0500 Received: from mx2.suse.de ([195.135.220.15]:55460 "EHLO mx2.suse.de" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S231734AbhBHJl7 (ORCPT ); Mon, 8 Feb 2021 04:41:59 -0500 X-Virus-Scanned: by amavisd-new at test-mx.suse.de DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=suse.com; s=susede1; t=1612777262; h=from:from:reply-to:date:date:message-id:message-id:to:to:cc:cc: mime-version:mime-version:content-type:content-type: in-reply-to:in-reply-to:references:references; bh=fF169tx56y6mupjIvrPQt9dqN5xV/kHLsueKWM8xvUg=; b=MoSRipFUt6ufW8TtS8lzbiX5/FVZJzmez4YJCZRHeMBFobJx4wUaMpOEY0Ksoxlj08E0EB BMga+pAmqKxvJzr+h92HO0x1eYAOFojiRnJ+LvHFYMgWcOQM2yxISYYR1UJ41qM4vIXkKZ 9dFMw0ot34fjhu/eiJLf+M+wr0MVcMQ= Received: from relay2.suse.de (unknown [195.135.221.27]) by mx2.suse.de (Postfix) with ESMTP id EAB7EAE53; Mon, 8 Feb 2021 09:41:01 +0000 (UTC) To: Julien Grall , xen-devel@lists.xenproject.org, linux-kernel@vger.kernel.org, linux-block@vger.kernel.org, netdev@vger.kernel.org, linux-scsi@vger.kernel.org Cc: Boris Ostrovsky , Stefano Stabellini , stable@vger.kernel.org, Konrad Rzeszutek Wilk , =?UTF-8?Q?Roger_Pau_Monn=c3=a9?= , Jens Axboe , Wei Liu , Paul Durrant , "David S. Miller" , Jakub Kicinski References: <20210206104932.29064-1-jgross@suse.com> From: =?UTF-8?B?SsO8cmdlbiBHcm/Dnw==?= Subject: Re: [PATCH 0/7] xen/events: bug fixes and some diagnostic aids Message-ID: <7aaeeb3d-1e1b-6166-84e9-481153811b62@suse.com> Date: Mon, 8 Feb 2021 10:41:00 +0100 User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:78.0) Gecko/20100101 Thunderbird/78.7.0 MIME-Version: 1.0 In-Reply-To: Content-Type: multipart/signed; micalg=pgp-sha256; protocol="application/pgp-signature"; boundary="GMBB9RCSmiMsxNr0VEDv5TlmxZ6GktJjs" Precedence: bulk List-ID: X-Mailing-List: linux-kernel@vger.kernel.org This is an OpenPGP/MIME signed message (RFC 4880 and 3156) --GMBB9RCSmiMsxNr0VEDv5TlmxZ6GktJjs Content-Type: multipart/mixed; boundary="k4gyNUY4aI5Lw9phW38uxpSLf7g9cDz0D"; protected-headers="v1" From: =?UTF-8?B?SsO8cmdlbiBHcm/Dnw==?= To: Julien Grall , xen-devel@lists.xenproject.org, linux-kernel@vger.kernel.org, linux-block@vger.kernel.org, netdev@vger.kernel.org, linux-scsi@vger.kernel.org Cc: Boris Ostrovsky , Stefano Stabellini , stable@vger.kernel.org, Konrad Rzeszutek Wilk , =?UTF-8?Q?Roger_Pau_Monn=c3=a9?= , Jens Axboe , Wei Liu , Paul Durrant , "David S. Miller" , Jakub Kicinski Message-ID: <7aaeeb3d-1e1b-6166-84e9-481153811b62@suse.com> Subject: Re: [PATCH 0/7] xen/events: bug fixes and some diagnostic aids References: <20210206104932.29064-1-jgross@suse.com> In-Reply-To: --k4gyNUY4aI5Lw9phW38uxpSLf7g9cDz0D Content-Type: multipart/mixed; boundary="------------8545B5C2DC00D006FC21850E" Content-Language: en-US This is a multi-part message in MIME format. --------------8545B5C2DC00D006FC21850E Content-Type: text/plain; charset=utf-8; format=flowed Content-Transfer-Encoding: quoted-printable On 08.02.21 10:11, Julien Grall wrote: > Hi Juergen, >=20 > On 07/02/2021 12:58, J=C3=BCrgen Gro=C3=9F wrote: >> On 06.02.21 19:46, Julien Grall wrote: >>> Hi Juergen, >>> >>> On 06/02/2021 10:49, Juergen Gross wrote: >>>> The first three patches are fixes for XSA-332. The avoid WARN splats= >>>> and a performance issue with interdomain events. >>> >>> Thanks for helping to figure out the problem. Unfortunately, I still = >>> see reliably the WARN splat with the latest Linux master=20 >>> (1e0d27fce010) + your first 3 patches. >>> >>> I am using Xen 4.11 (1c7d984645f9) and dom0 is forced to use the 2L=20 >>> events ABI. >>> >>> After some debugging, I think I have an idea what's went wrong. The=20 >>> problem happens when the event is initially bound from vCPU0 to a=20 >>> different vCPU. >>> >>> =C2=A0From the comment in xen_rebind_evtchn_to_cpu(), we are masking = the=20 >>> event to prevent it being delivered on an unexpected vCPU. However, I= =20 >>> believe the following can happen: >>> >>> vCPU0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0= =C2=A0=C2=A0=C2=A0=C2=A0 | vCPU1 >>> =C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2= =A0=C2=A0=C2=A0=C2=A0=C2=A0 | >>> =C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2= =A0=C2=A0=C2=A0=C2=A0=C2=A0 | Call xen_rebind_evtchn_to_cpu() >>> receive event X=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0= =C2=A0=C2=A0 | >>> =C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2= =A0=C2=A0=C2=A0=C2=A0=C2=A0 | mask event X >>> =C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2= =A0=C2=A0=C2=A0=C2=A0=C2=A0 | bind to vCPU1 >>> =C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 | unmask= event X >>> =C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2= =A0=C2=A0=C2=A0=C2=A0=C2=A0 | >>> =C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2= =A0=C2=A0=C2=A0=C2=A0=C2=A0 | receive event X >>> =C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2= =A0=C2=A0=C2=A0=C2=A0=C2=A0 | >>> =C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2= =A0=C2=A0=C2=A0=C2=A0=C2=A0 | handle_edge_irq(X) >>> handle_edge_irq(X)=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 |=C2=A0 = -> handle_irq_event() >>> =C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2= =A0=C2=A0=C2=A0=C2=A0=C2=A0 |=C2=A0=C2=A0 -> set IRQD_IN_PROGRESS >>> =C2=A0=C2=A0-> set IRQS_PENDING=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2= =A0 | >>> =C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2= =A0=C2=A0=C2=A0=C2=A0=C2=A0 |=C2=A0=C2=A0 -> evtchn_interrupt() >>> =C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2= =A0=C2=A0=C2=A0=C2=A0=C2=A0 |=C2=A0=C2=A0 -> clear IRQD_IN_PROGRESS >>> =C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2= =A0=C2=A0=C2=A0=C2=A0=C2=A0 |=C2=A0 -> IRQS_PENDING is set >>> =C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2= =A0=C2=A0=C2=A0=C2=A0=C2=A0 |=C2=A0 -> handle_irq_event() >>> =C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2= =A0=C2=A0=C2=A0=C2=A0=C2=A0 |=C2=A0=C2=A0 -> evtchn_interrupt() >>> =C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2= =A0=C2=A0=C2=A0=C2=A0=C2=A0 |=C2=A0=C2=A0=C2=A0=C2=A0 -> WARN() >>> =C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2= =A0=C2=A0=C2=A0=C2=A0=C2=A0 | >>> >>> All the lateeoi handlers expect a ONESHOT semantic and=20 >>> evtchn_interrupt() is doesn't tolerate any deviation. >>> >>> I think the problem was introduced by 7f874a0447a9 ("xen/events: fix = >>> lateeoi irq acknowledgment") because the interrupt was disabled=20 >>> previously. Therefore we wouldn't do another iteration in=20 >>> handle_edge_irq(). >> >> I think you picked the wrong commit for blaming, as this is just >> the last patch of the three patches you were testing. >=20 > I actually found the right commit for blaming but I copied the=20 > information from the wrong shell :/. The bug was introduced by: >=20 > c44b849cee8c ("xen/events: switch user event channels to lateeoi model"= ) >=20 >> >>> Aside the handlers, I think it may impact the defer EOI mitigation=20 >>> because in theory if a 3rd vCPU is joining the party (let say vCPU A = >>> migrate the event from vCPU B to vCPU C). So info->{eoi_cpu,=20 >>> irq_epoch, eoi_time} could possibly get mangled? >>> >>> For a fix, we may want to consider to hold evtchn_rwlock with the=20 >>> write permission. Although, I am not 100% sure this is going to=20 >>> prevent everything. >> >> It will make things worse, as it would violate the locking hierarchy >> (xen_rebind_evtchn_to_cpu() is called with the IRQ-desc lock held). >=20 > Ah, right. >=20 >> >> On a first glance I think we'll need a 3rd masking state ("temporarily= >> masked") in the second patch in order to avoid a race with lateeoi. >> >> In order to avoid the race you outlined above we need an "event is bei= ng >> handled" indicator checked via test_and_set() semantics in >> handle_irq_for_port() and reset only when calling clear_evtchn(). >=20 > It feels like we are trying to workaround the IRQ flow we are using=20 > (i.e. handle_edge_irq()). I'm not really sure this is the main problem here. According to your analysis the main problem is occurring when handling the event, not when handling the IRQ: the event is being received on two vcpus. Our problem isn't due to the IRQ still being pending, but due it being raised again, which should happen for a one shot IRQ the same way. But maybe I'm misunderstanding your idea. Juergen --------------8545B5C2DC00D006FC21850E Content-Type: application/pgp-keys; name="OpenPGP_0xB0DE9DD628BF132F.asc" Content-Transfer-Encoding: quoted-printable Content-Disposition: attachment; filename="OpenPGP_0xB0DE9DD628BF132F.asc" -----BEGIN PGP PUBLIC KEY BLOCK----- xsBNBFOMcBYBCACgGjqjoGvbEouQZw/ToiBg9W98AlM2QHV+iNHsEs7kxWhKMjrioyspZKOBy= cWx w3ie3j9uvg9EOB3aN4xiTv4qbnGiTr3oJhkB1gsb6ToJQZ8uxGq2kaV2KL9650I1SJvedYm8O= f8Z d621lSmoKOwlNClALZNew72NjJLEzTalU1OdT7/i1TXkH09XSSI8mEQ/ouNcMvIJNwQpd369y= 9bf IhWUiVXEK7MlRgUG6MvIj6Y3Am/BBLUVbDa4+gmzDC9ezlZkTZG2t14zWPvxXP3FAp2pkW0xq= G7/ 377qptDmrk42GlSKN4z76ELnLxussxc7I2hx18NUcbP8+uty4bMxABEBAAHNHEp1ZXJnZW4gR= 3Jv c3MgPGpnQHBmdXBmLm5ldD7CwHkEEwECACMFAlOMcBYCGwMHCwkIBwMCAQYVCAIJCgsEFgIDA= QIe AQIXgAAKCRCw3p3WKL8TL0KdB/93FcIZ3GCNwFU0u3EjNbNjmXBKDY4FUGNQH2lvWAUy+dnyT= hpw dtF/jQ6j9RwE8VP0+NXcYpGJDWlNb9/JmYqLiX2Q3TyevpB0CA3dbBQp0OW0fgCetToGIQrg0= MbD 1C/sEOv8Mr4NAfbauXjZlvTj30H2jO0u+6WGM6nHwbh2l5O8ZiHkH32iaSTfN7Eu5RnNVUJbv= oPH Z8SlM4KWm8rG+lIkGurqqu5gu8q8ZMKdsdGC4bBxdQKDKHEFExLJK/nRPFmAuGlId1E3fe10v= 5QL +qHI3EIPtyfE7i9Hz6rVwi7lWKgh7pe0ZvatAudZ+JNIlBKptb64FaiIOAWDCx1SzR9KdWVyZ= 2Vu IEdyb3NzIDxqZ3Jvc3NAc3VzZS5jb20+wsB5BBMBAgAjBQJTjHCvAhsDBwsJCAcDAgEGFQgCC= QoL BBYCAwECHgECF4AACgkQsN6d1ii/Ey/HmQf/RtI7kv5A2PS4RF7HoZhPVPogNVbC4YA6lW7Dr= Wf0 teC0RR3MzXfy6pJ+7KLgkqMlrAbN/8Dvjoz78X+5vhH/rDLa9BuZQlhFmvcGtCF8eR0T1v0nC= /nu AFVGy+67q2DH8As3KPu0344TBDpAvr2uYM4tSqxK4DURx5INz4ZZ0WNFHcqsfvlGJALDeE0Lh= ITT d9jLzdDad1pQSToCnLl6SBJZjDOX9QQcyUigZFtCXFst4dlsvddrxyqT1f17+2cFSdu7+ynLm= XBK 7abQ3rwJY8SbRO2iRulogc5vr/RLMMlscDAiDkaFQWLoqHHOdfO9rURssHNN8WkMnQfvUewRz= 80h SnVlcmdlbiBHcm9zcyA8amdyb3NzQG5vdmVsbC5jb20+wsB5BBMBAgAjBQJTjHDXAhsDBwsJC= AcD AgEGFQgCCQoLBBYCAwECHgECF4AACgkQsN6d1ii/Ey8PUQf/ehmgCI9jB9hlgexLvgOtf7PJn= FOX gMLdBQgBlVPO3/D9R8LtF9DBAFPNhlrsfIG/SqICoRCqUcJ96Pn3P7UUinFG/I0ECGF4EvTE1= jnD kfJZr6jrbjgyoZHiw/4BNwSTL9rWASyLgqlA8u1mf+c2yUwcGhgkRAd1gOwungxcwzwqgljf0= N51 N5JfVRHRtyfwq/ge+YEkDGcTU6Y0sPOuj4Dyfm8fJzdfHNQsWq3PnczLVELStJNdapwPOoE+l= otu fe3AM2vAEYJ9rTz3Cki4JFUsgLkHFqGZarrPGi1eyQcXeluldO3m91NK/1xMI3/+8jbO0tsn1= tqS EUGIJi7ox80eSnVlcmdlbiBHcm9zcyA8amdyb3NzQHN1c2UuZGU+wsB5BBMBAgAjBQJTjHDrA= hsD BwsJCAcDAgEGFQgCCQoLBBYCAwECHgECF4AACgkQsN6d1ii/Ey+LhQf9GL45eU5vOowA2u5N3= g3O ZUEBmDHVVbqMtzwlmNC4k9Kx39r5s2vcFl4tXqW7g9/ViXYuiDXb0RfUpZiIUW89siKrkzmQ5= dM7 wRqzgJpJwK8Bn2MIxAKArekWpiCKvBOB/Cc+3EXE78XdlxLyOi/NrmSGRIov0karw2RzMNOu5= D+j LRZQd1Sv27AR+IP3I8U4aqnhLpwhK7MEy9oCILlgZ1QZe49kpcumcZKORmzBTNh30FVKK1Evm= V2x AKDoaEOgQB4iFQLhJCdP1I5aSgM5IVFdn7v5YgEYuJYx37IoN1EblHI//x/e2AaIHpzK5h88N= Eaw QsaNRpNSrcfbFmAg987ATQRTjHAWAQgAyzH6AOODMBjgfWE9VeCgsrwH3exNAU32gLq2xvjpW= nHI s98ndPUDpnoxWQugJ6MpMncr0xSwFmHEgnSEjK/PAjppgmyc57BwKII3sV4on+gDVFJR6Y8ZR= wgn BC5mVM6JjQ5xDk8WRXljExRfUX9pNhdE5eBOZJrDRoLUmmjDtKzWaDhIg/+1Hzz93X4fCQkNV= bVF LELU9bMaLPBG/x5q4iYZ2k2ex6d47YE1ZFdMm6YBYMOljGkZKwYde5ldM9mo45mmwe0icXKLk= pEd IXKTZeKDO+Hdv1aqFuAcccTg9RXDQjmwhC3yEmrmcfl0+rPghO0Iv3OOImwTEe4co3c1mwARA= QAB wsBfBBgBAgAJBQJTjHAWAhsMAAoJELDendYovxMvQ/gH/1ha96vm4P/L+bQpJwrZ/dneZcmEw= Tbe 8YFsw2V/Buv6Z4Mysln3nQK5ZadD534CF7TDVft7fC4tU4PONxF5D+/tvgkPfDAfF77zy2AH1= vJz Q1fOU8lYFpZXTXIHb+559UqvIB8AdgR3SAJGHHt4RKA0F7f5ipYBBrC6cyXJyyoprT10EMvU8= VGi wXvTyJz3fjoYsdFzpWPlJEBRMedCot60g5dmbdrZ5DWClAr0yau47zpWj3enf1tLWaqcsuylW= svi uGjKGw7KHQd3bxALOknAp4dN3QwBYCKuZ7AddY9yjynVaD5X7nF9nO5BjR/i1DG86lem3iBDX= zXs ZDn8R38=3D =3D2wuH -----END PGP PUBLIC KEY BLOCK----- --------------8545B5C2DC00D006FC21850E-- --k4gyNUY4aI5Lw9phW38uxpSLf7g9cDz0D-- --GMBB9RCSmiMsxNr0VEDv5TlmxZ6GktJjs Content-Type: application/pgp-signature; name="OpenPGP_signature.asc" Content-Description: OpenPGP digital signature Content-Disposition: attachment; filename="OpenPGP_signature" -----BEGIN PGP SIGNATURE----- wsB5BAABCAAjFiEEhRJncuj2BJSl0Jf3sN6d1ii/Ey8FAmAhBywFAwAAAAAACgkQsN6d1ii/Ey+X 7Qf+OnrlhZHoEZMFLJ8YuxArgwY/dc0hT9fYNejZag9l2vEAND4Z36Oqpy4lMt9l1wsocKvxEhhF pSpzfvQWh9qFvJbNZtU3gSfVM6z8XQ+FXWlr2vfGOqMmYTZKKcZrMtTmL0spDeb4EDwM1G86mtZ4 88bV0RVAoliWwy5CUk5MyZyHwYOC3YeH3VP7K3kBLy4F1gopdOrTWHvI50FFrTBavw1ZnCRTzlgj qYWYHD2ztYmPtDEtlSsfcyxj2+xur511a16/B/RfP0f+g1maWYySmaxhVFNQa+icHKzu8VgfenHR sXi1Zu4uQNMf/zQXInigQg8IPawMiI2GTHjN+tovog== =fwXt -----END PGP SIGNATURE----- --GMBB9RCSmiMsxNr0VEDv5TlmxZ6GktJjs--