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 Received: from kanga.kvack.org (kanga.kvack.org [205.233.56.17]) by smtp.lore.kernel.org (Postfix) with ESMTP id DCCC8C001B0 for ; Wed, 19 Jul 2023 21:55:01 +0000 (UTC) Received: by kanga.kvack.org (Postfix) id 68AF5280094; Wed, 19 Jul 2023 17:55:01 -0400 (EDT) Received: by kanga.kvack.org (Postfix, from userid 40) id 63B9F28004C; Wed, 19 Jul 2023 17:55:01 -0400 (EDT) X-Delivered-To: int-list-linux-mm@kvack.org Received: by kanga.kvack.org (Postfix, from userid 63042) id 5043E280094; Wed, 19 Jul 2023 17:55:01 -0400 (EDT) X-Delivered-To: linux-mm@kvack.org Received: from relay.hostedemail.com (smtprelay0016.hostedemail.com [216.40.44.16]) by kanga.kvack.org (Postfix) with ESMTP id 40B9C28004C for ; Wed, 19 Jul 2023 17:55:01 -0400 (EDT) Received: from smtpin21.hostedemail.com (a10.router.float.18 [10.200.18.1]) by unirelay06.hostedemail.com (Postfix) with ESMTP id 0B08BB12A0 for ; Wed, 19 Jul 2023 21:55:01 +0000 (UTC) X-FDA: 81029717202.21.543FD64 Received: from mail-ej1-f41.google.com (mail-ej1-f41.google.com [209.85.218.41]) by imf06.hostedemail.com (Postfix) with ESMTP id 27A2618000A for ; Wed, 19 Jul 2023 21:54:58 +0000 (UTC) Authentication-Results: imf06.hostedemail.com; dkim=pass header.d=google.com header.s=20221208 header.b=NB8jnaIL; dmarc=pass (policy=reject) header.from=google.com; spf=pass (imf06.hostedemail.com: domain of axelrasmussen@google.com designates 209.85.218.41 as permitted sender) smtp.mailfrom=axelrasmussen@google.com ARC-Message-Signature: i=1; a=rsa-sha256; c=relaxed/relaxed; d=hostedemail.com; s=arc-20220608; t=1689803699; h=from:from:sender:reply-to:subject:subject:date:date: message-id:message-id:to:to:cc:cc:mime-version:mime-version: content-type:content-type: content-transfer-encoding:content-transfer-encoding: in-reply-to:in-reply-to:references:references:dkim-signature; bh=RpNnWnPyxzC0NSwOwUxwFXML73N2PXfa+cgDdImdoSQ=; b=lBS4iEQ83f83D14kiB4WLicbTqxhnlnwpjA0Pur3cWbOOAWNXEsr0ZJuzPV3hlffjDbd2s oY5H83sjQ9rTswCppv/f3G97gNAllzLup9SC52lgJrHSXmqtTwg94p5jX00IJoBglV7lzT K0+pWxVMFGJWuuAwvz0UMImfD2YXiZA= ARC-Authentication-Results: i=1; imf06.hostedemail.com; dkim=pass header.d=google.com header.s=20221208 header.b=NB8jnaIL; dmarc=pass (policy=reject) header.from=google.com; spf=pass (imf06.hostedemail.com: domain of axelrasmussen@google.com designates 209.85.218.41 as permitted sender) smtp.mailfrom=axelrasmussen@google.com ARC-Seal: i=1; s=arc-20220608; d=hostedemail.com; t=1689803699; a=rsa-sha256; cv=none; b=NEoO57O+vxceFJHqm25m+ULiuRBBigP65ee4ef0URUMnU+3u8cm06aj+6kQyWDdGEtmMvT xlOw4VMjsgCXJu1mMhLFM8/4wG7N3jUlxvAwi0wXqCGRhSeC/DDYnnOw7OsewYdPG0kMII Gff/d1HThmjvz8/Z+xkMe0CVViM6VpI= Received: by mail-ej1-f41.google.com with SMTP id a640c23a62f3a-992f6d7c7fbso28695066b.3 for ; Wed, 19 Jul 2023 14:54:58 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=20221208; t=1689803697; x=1690408497; h=content-transfer-encoding:cc:to:subject:message-id:date:from :in-reply-to:references:mime-version:from:to:cc:subject:date :message-id:reply-to; bh=RpNnWnPyxzC0NSwOwUxwFXML73N2PXfa+cgDdImdoSQ=; b=NB8jnaILqWlUSjoqa2DuwIlr5sIiVBlH5ll0GB7VJ94jTNRcxdq+q2az4u6lN7VXg4 o6F5xWDsnFWsZDWizA9OIwjzMHSlrRUaoLo029vmZSQpdr1QnED5t4IJxVLOx3N5Chj5 vJ6nW9JX+Q34odPv8TCKxME8ZsorQXxh6hnJ29YYOL3G8jGAk9ugu3PK/op2Xo3Txdly /8lQPm/wBgWwivXi0h7eex59f2qSQk5GouUACFh/Dvh6ukKzlYRku86zlOqDmc5ynWMO 12bMXDyHEdfjz7deE400RiH/PICVgW1X3e4/Nctw9ZRy8leGuoN1jIgsesu/ws9lKslt yseQ== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20221208; t=1689803697; x=1690408497; h=content-transfer-encoding:cc:to:subject:message-id:date:from :in-reply-to:references:mime-version:x-gm-message-state:from:to:cc :subject:date:message-id:reply-to; bh=RpNnWnPyxzC0NSwOwUxwFXML73N2PXfa+cgDdImdoSQ=; b=MOB1iqSbdKEP0qCNSW87zWNfzbL6y4lO+kENpsbqopEe8hhY5H3Bzj0b/5p4eNRIZL /t6Wsn6Kr9+5kWpfbcRbt+FA44EvgDybLvXs3UP9tIWkcm1KbFoaKeICkNIqcucLq7K4 +ypX9C8SHJVlUs/uMi7nKqgxS+dlAVccTNTAmB3if3PhLphH0eEN74bBPOE+hzfNT6yI /Pxl7SLkw0A4o/hXyxkPE68GiW5DIDo6RRN4UiWDvMAMjua0qniyjkDQihgk8BlG7XUp v8uzK/c5K0+wzOEr1+A73WWM2o9xBEBgqfzqYTEnSZqAtWACXE1jPMpojbzEpquqV3a2 AWVw== X-Gm-Message-State: ABy/qLZjwNLDzig52TAuCiiBX3g0CxxivaOS+kTMR7gJeV9rVVd4Mvn7 EhT/xvG6Db/aV2FJPkQHMJ95DZiqQeq68ed88w5Gpg== X-Google-Smtp-Source: APBJJlFNP/u4PLWdbUpDUO/EDcj+Vvn+ADWg7TLotRFNwc67uCHj7496GlfZJDuLIKnSUNN95ya3n1jNcy6cG+EYYmI= X-Received: by 2002:a17:906:7a16:b0:982:2278:bcef with SMTP id d22-20020a1709067a1600b009822278bcefmr3469134ejo.60.1689803697537; Wed, 19 Jul 2023 14:54:57 -0700 (PDT) MIME-Version: 1.0 References: <79375b71-db2e-3e66-346b-254c90d915e2@cslab.ece.ntua.gr> <20230719211631.890995-1-axelrasmussen@google.com> In-Reply-To: <20230719211631.890995-1-axelrasmussen@google.com> From: Axel Rasmussen Date: Wed, 19 Jul 2023 14:54:21 -0700 Message-ID: Subject: Re: Using userfaultfd with KVM's async page fault handling causes processes to hung waiting for mmap_lock to be released To: Dimitris Siakavaras Cc: viro@zeniv.linux.org.uk, linux-fsdevel@vger.kernel.org, linux-kernel@vger.kernel.org, Peter Xu , linux-mm@kvack.org Content-Type: text/plain; charset="UTF-8" Content-Transfer-Encoding: quoted-printable X-Rspamd-Server: rspam09 X-Rspamd-Queue-Id: 27A2618000A X-Stat-Signature: fz7561tknmotcnkbhq8pmhokhtq6xspk X-Rspam-User: X-HE-Tag: 1689803698-432208 X-HE-Meta: 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 sfEnh85V 8Pj2Ht86DPS8SKsylYO++RH5L/PpVmCfjAz8F7SHo+Kobk8DAsnZH3Lo4fk/brQhXQ6z8AZFXXD0/hpTgGlU9+xSwICSVmhPgHXbSx3vuN2s3zVjnxLkcquZuMma3QM83t+U4LUtjsLPHleATWftJvC6f119gsbd+vaj2mqQsfsqpxiyAri3xqefegRj2dfEt1R5NlfUSIrUbBRdEGYWWto+YXIFHLu9ZuXF9d3Mnl9tlbjqV+Zkl0zln1HdKpfwdZIj6KFApRyL0Iy/8FlR3FJ5yckA7qmwb6B/mmweo9+82r9nNlF988SlQJxdGVcUn8ZfEmvsU5l+TivljjL08kMp+MpPSp9SRpqpw X-Bogosity: Ham, tests=bogofilter, spamicity=0.191438, version=1.2.4 Sender: owner-linux-mm@kvack.org Precedence: bulk X-Loop: owner-majordomo@kvack.org List-ID: On Wed, Jul 19, 2023 at 2:16=E2=80=AFPM Axel Rasmussen wrote: > > Thanks for the detailed report Dimitris! I've CCed the MM mailing list an= d some > folks who work on userfaultfd. Apologies, I should have quoted the original message for the others I added to CC: https://lore.kernel.org/lkml/79375b71-db2e-3e66-346b-254c90d91= 5e2@cslab.ece.ntua.gr/T/#u > > I took a look at this today, but I haven't quite come up with a solution. > > I thought it might be as easy as changing userfaultfd_release() to set re= leased > *after* taking the lock. But no such luck, the ordering is what it is to = deal > with another subtle case: > > > WRITE_ONCE(ctx->released, true); > > if (!mmget_not_zero(mm)) > goto wakeup; > > /* > * Flush page faults out of all CPUs. NOTE: all page faults > * must be retried without returning VM_FAULT_SIGBUS if > * userfaultfd_ctx_get() succeeds but vma->vma_userfault_ctx > * changes while handle_userfault released the mmap_lock. So > * it's critical that released is set to true (above), before > * taking the mmap_lock for writing. > */ > mmap_write_lock(mm); > > I think perhaps the right thing to do is to have handle_userfault() relea= se > mmap_lock when it returns VM_FAULT_NOPAGE, and to have GUP deal with that > appropriately? But, some investigation is required to be sure that's okay= to do > in the other non-GUP ways we can end up in handle_userfault().