From patchwork Thu Sep 2 21:58:59 2021 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Andrew Morton X-Patchwork-Id: 12473183 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=-15.8 required=3.0 tests=BAYES_00,DKIM_SIGNED, DKIM_VALID,DKIM_VALID_AU,HEADER_FROM_DIFFERENT_DOMAINS,INCLUDES_CR_TRAILER, INCLUDES_PATCH,MAILING_LIST_MULTI,SPF_HELO_NONE,SPF_PASS autolearn=unavailable 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 A5A05C433EF for ; Thu, 2 Sep 2021 21:59:02 +0000 (UTC) Received: from kanga.kvack.org (kanga.kvack.org [205.233.56.17]) by mail.kernel.org (Postfix) with ESMTP id 5A3126056B for ; Thu, 2 Sep 2021 21:59:02 +0000 (UTC) DMARC-Filter: OpenDMARC Filter v1.4.1 mail.kernel.org 5A3126056B Authentication-Results: mail.kernel.org; dmarc=none (p=none dis=none) header.from=linux-foundation.org Authentication-Results: mail.kernel.org; spf=pass smtp.mailfrom=kvack.org Received: by kanga.kvack.org (Postfix) id 036FD6B017B; Thu, 2 Sep 2021 17:59:02 -0400 (EDT) Received: by kanga.kvack.org (Postfix, from userid 40) id F03E66B017C; Thu, 2 Sep 2021 17:59:01 -0400 (EDT) X-Delivered-To: int-list-linux-mm@kvack.org Received: by kanga.kvack.org (Postfix, from userid 63042) id D7CA56B017D; Thu, 2 Sep 2021 17:59:01 -0400 (EDT) X-Delivered-To: linux-mm@kvack.org Received: from forelay.hostedemail.com (smtprelay0007.hostedemail.com [216.40.44.7]) by kanga.kvack.org (Postfix) with ESMTP id BFBCB6B017B for ; Thu, 2 Sep 2021 17:59:01 -0400 (EDT) Received: from smtpin08.hostedemail.com (10.5.19.251.rfc1918.com [10.5.19.251]) by forelay03.hostedemail.com (Postfix) with ESMTP id 84E4F82E097F for ; Thu, 2 Sep 2021 21:59:01 +0000 (UTC) X-FDA: 78543999282.08.9B411F6 Received: from mail.kernel.org (mail.kernel.org [198.145.29.99]) by imf15.hostedemail.com (Postfix) with ESMTP id 2B188D000099 for ; Thu, 2 Sep 2021 21:59:01 +0000 (UTC) Received: by mail.kernel.org (Postfix) with ESMTPSA id 1A63860F21; Thu, 2 Sep 2021 21:59:00 +0000 (UTC) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=linux-foundation.org; s=korg; t=1630619940; bh=iZniyZtzjNO34lmKOESr+S0lbGZOEzi7ac/pqRqjgCE=; h=Date:From:To:Subject:In-Reply-To:From; b=rpEN/9CmnxjU4WEybr3/mm5Vmp8RpPFVZwe1BdDQVmCXQ/ybFi4juNhL4L9h+CiUy pZyhLkxg6eVjnhiCMWnnNkHSSOrd4LqcBzA/eQhDpq2pFdY41r9snplrJM0PrgWP0K aBrldBPPHFiLuzK3I+onIpfE9SDB0KMVC601DolI= Date: Thu, 02 Sep 2021 14:58:59 -0700 From: Andrew Morton To: aarcange@redhat.com, akpm@linux-foundation.org, axboe@kernel.dk, axelrasmussen@google.com, linux-mm@kvack.org, mm-commits@vger.kernel.org, namit@vmware.com, peterx@redhat.com, rppt@linux.vnet.ibm.com, torvalds@linux-foundation.org, viro@zeniv.linux.org.uk Subject: [patch 171/212] userfaultfd: prevent concurrent API initialization Message-ID: <20210902215859.yE9smvbXR%akpm@linux-foundation.org> In-Reply-To: <20210902144820.78957dff93d7bea620d55a89@linux-foundation.org> User-Agent: s-nail v14.8.16 Authentication-Results: imf15.hostedemail.com; dkim=pass header.d=linux-foundation.org header.s=korg header.b="rpEN/9Cm"; spf=pass (imf15.hostedemail.com: domain of akpm@linux-foundation.org designates 198.145.29.99 as permitted sender) smtp.mailfrom=akpm@linux-foundation.org; dmarc=none X-Rspamd-Server: rspam05 X-Rspamd-Queue-Id: 2B188D000099 X-Stat-Signature: gd1fxij6gyuhifgqqxzxfw4ds1k8wndq X-HE-Tag: 1630619941-143358 X-Bogosity: Ham, tests=bogofilter, spamicity=0.000000, version=1.2.4 Sender: owner-linux-mm@kvack.org Precedence: bulk X-Loop: owner-majordomo@kvack.org List-ID: From: Nadav Amit Subject: userfaultfd: prevent concurrent API initialization userfaultfd assumes that the enabled features are set once and never changed after UFFDIO_API ioctl succeeded. However, currently, UFFDIO_API can be called concurrently from two different threads, succeed on both threads and leave userfaultfd's features in non-deterministic state. Theoretically, other uffd operations (ioctl's and page-faults) can be dispatched while adversely affected by such changes of features. Moreover, the writes to ctx->state and ctx->features are not ordered, which can - theoretically, again - let userfaultfd_ioctl() think that userfaultfd API completed, while the features are still not initialized. To avoid races, it is arguably best to get rid of ctx->state. Since there are only 2 states, record the API initialization in ctx->features as the uppermost bit and remove ctx->state. Link: https://lkml.kernel.org/r/20210808020724.1022515-3-namit@vmware.com Fixes: 9cd75c3cd4c3d ("userfaultfd: non-cooperative: add ability to report non-PF events from uffd descriptor") Signed-off-by: Nadav Amit Cc: Alexander Viro Cc: Andrea Arcangeli Cc: Axel Rasmussen Cc: Jens Axboe Cc: Mike Rapoport Cc: Peter Xu Signed-off-by: Andrew Morton --- fs/userfaultfd.c | 91 +++++++++++++++++++++------------------------ 1 file changed, 44 insertions(+), 47 deletions(-) --- a/fs/userfaultfd.c~userfaultfd-prevent-concurrent-api-initialization +++ a/fs/userfaultfd.c @@ -33,11 +33,6 @@ int sysctl_unprivileged_userfaultfd __re static struct kmem_cache *userfaultfd_ctx_cachep __read_mostly; -enum userfaultfd_state { - UFFD_STATE_WAIT_API, - UFFD_STATE_RUNNING, -}; - /* * Start with fault_pending_wqh and fault_wqh so they're more likely * to be in the same cacheline. @@ -69,8 +64,6 @@ struct userfaultfd_ctx { unsigned int flags; /* features requested from the userspace */ unsigned int features; - /* state machine */ - enum userfaultfd_state state; /* released */ bool released; /* memory mappings are changing because of non-cooperative event */ @@ -104,6 +97,14 @@ struct userfaultfd_wake_range { unsigned long len; }; +/* internal indication that UFFD_API ioctl was successfully executed */ +#define UFFD_FEATURE_INITIALIZED (1u << 31) + +static bool userfaultfd_is_initialized(struct userfaultfd_ctx *ctx) +{ + return ctx->features & UFFD_FEATURE_INITIALIZED; +} + static int userfaultfd_wake_function(wait_queue_entry_t *wq, unsigned mode, int wake_flags, void *key) { @@ -667,7 +668,6 @@ int dup_userfaultfd(struct vm_area_struc refcount_set(&ctx->refcount, 1); ctx->flags = octx->flags; - ctx->state = UFFD_STATE_RUNNING; ctx->features = octx->features; ctx->released = false; atomic_set(&ctx->mmap_changing, 0); @@ -944,38 +944,33 @@ static __poll_t userfaultfd_poll(struct poll_wait(file, &ctx->fd_wqh, wait); - switch (ctx->state) { - case UFFD_STATE_WAIT_API: + if (!userfaultfd_is_initialized(ctx)) return EPOLLERR; - case UFFD_STATE_RUNNING: - /* - * poll() never guarantees that read won't block. - * userfaults can be waken before they're read(). - */ - if (unlikely(!(file->f_flags & O_NONBLOCK))) - return EPOLLERR; - /* - * lockless access to see if there are pending faults - * __pollwait last action is the add_wait_queue but - * the spin_unlock would allow the waitqueue_active to - * pass above the actual list_add inside - * add_wait_queue critical section. So use a full - * memory barrier to serialize the list_add write of - * add_wait_queue() with the waitqueue_active read - * below. - */ - ret = 0; - smp_mb(); - if (waitqueue_active(&ctx->fault_pending_wqh)) - ret = EPOLLIN; - else if (waitqueue_active(&ctx->event_wqh)) - ret = EPOLLIN; - return ret; - default: - WARN_ON_ONCE(1); + /* + * poll() never guarantees that read won't block. + * userfaults can be waken before they're read(). + */ + if (unlikely(!(file->f_flags & O_NONBLOCK))) return EPOLLERR; - } + /* + * lockless access to see if there are pending faults + * __pollwait last action is the add_wait_queue but + * the spin_unlock would allow the waitqueue_active to + * pass above the actual list_add inside + * add_wait_queue critical section. So use a full + * memory barrier to serialize the list_add write of + * add_wait_queue() with the waitqueue_active read + * below. + */ + ret = 0; + smp_mb(); + if (waitqueue_active(&ctx->fault_pending_wqh)) + ret = EPOLLIN; + else if (waitqueue_active(&ctx->event_wqh)) + ret = EPOLLIN; + + return ret; } static const struct file_operations userfaultfd_fops; @@ -1170,7 +1165,7 @@ static ssize_t userfaultfd_read(struct f int no_wait = file->f_flags & O_NONBLOCK; struct inode *inode = file_inode(file); - if (ctx->state == UFFD_STATE_WAIT_API) + if (!userfaultfd_is_initialized(ctx)) return -EINVAL; for (;;) { @@ -1909,9 +1904,10 @@ out: static inline unsigned int uffd_ctx_features(__u64 user_features) { /* - * For the current set of features the bits just coincide + * For the current set of features the bits just coincide. Set + * UFFD_FEATURE_INITIALIZED to mark the features as enabled. */ - return (unsigned int)user_features; + return (unsigned int)user_features | UFFD_FEATURE_INITIALIZED; } /* @@ -1924,12 +1920,10 @@ static int userfaultfd_api(struct userfa { struct uffdio_api uffdio_api; void __user *buf = (void __user *)arg; + unsigned int ctx_features; int ret; __u64 features; - ret = -EINVAL; - if (ctx->state != UFFD_STATE_WAIT_API) - goto out; ret = -EFAULT; if (copy_from_user(&uffdio_api, buf, sizeof(uffdio_api))) goto out; @@ -1953,9 +1947,13 @@ static int userfaultfd_api(struct userfa ret = -EFAULT; if (copy_to_user(buf, &uffdio_api, sizeof(uffdio_api))) goto out; - ctx->state = UFFD_STATE_RUNNING; + /* only enable the requested features for this uffd context */ - ctx->features = uffd_ctx_features(features); + ctx_features = uffd_ctx_features(features); + ret = -EINVAL; + if (cmpxchg(&ctx->features, 0, ctx_features) != 0) + goto err_out; + ret = 0; out: return ret; @@ -1972,7 +1970,7 @@ static long userfaultfd_ioctl(struct fil int ret = -EINVAL; struct userfaultfd_ctx *ctx = file->private_data; - if (cmd != UFFDIO_API && ctx->state == UFFD_STATE_WAIT_API) + if (cmd != UFFDIO_API && !userfaultfd_is_initialized(ctx)) return -EINVAL; switch(cmd) { @@ -2086,7 +2084,6 @@ SYSCALL_DEFINE1(userfaultfd, int, flags) refcount_set(&ctx->refcount, 1); ctx->flags = flags; ctx->features = 0; - ctx->state = UFFD_STATE_WAIT_API; ctx->released = false; atomic_set(&ctx->mmap_changing, 0); ctx->mm = current->mm;