From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mail.kernel.org ([198.145.29.99]:37318 "EHLO mail.kernel.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1732864AbeLNMRD (ORCPT ); Fri, 14 Dec 2018 07:17:03 -0500 From: Greg Kroah-Hartman To: linux-kernel@vger.kernel.org Cc: Greg Kroah-Hartman , stable@vger.kernel.org, Linus Torvalds , Ben Hutchings Subject: [PATCH 4.4 82/88] proc: dont use FOLL_FORCE for reading cmdline and environment Date: Fri, 14 Dec 2018 13:00:56 +0100 Message-Id: <20181214115708.851388434@linuxfoundation.org> In-Reply-To: <20181214115702.151309521@linuxfoundation.org> References: <20181214115702.151309521@linuxfoundation.org> MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Sender: stable-owner@vger.kernel.org List-ID: 4.4-stable review patch. If anyone has any objections, please let me know. ------------------ From: Linus Torvalds commit 272ddc8b37354c3fe111ab26d25e792629148eee upstream. Now that Lorenzo cleaned things up and made the FOLL_FORCE users explicit, it becomes obvious how some of them don't really need FOLL_FORCE at all. So remove FOLL_FORCE from the proc code that reads the command line and arguments from user space. The mem_rw() function actually does want FOLL_FORCE, because gdd (and possibly many other debuggers) use it as a much more convenient version of PTRACE_PEEKDATA, but we should consider making the FOLL_FORCE part conditional on actually being a ptracer. This does not actually do that, just moves adds a comment to that effect and moves the gup_flags settings next to each other. Signed-off-by: Linus Torvalds Signed-off-by: Ben Hutchings Signed-off-by: Greg Kroah-Hartman --- fs/proc/base.c | 18 ++++++++---------- 1 file changed, 8 insertions(+), 10 deletions(-) --- a/fs/proc/base.c +++ b/fs/proc/base.c @@ -254,7 +254,7 @@ static ssize_t proc_pid_cmdline_read(str * Inherently racy -- command line shares address space * with code and data. */ - rv = access_remote_vm(mm, arg_end - 1, &c, 1, FOLL_FORCE); + rv = access_remote_vm(mm, arg_end - 1, &c, 1, 0); if (rv <= 0) goto out_free_page; @@ -272,8 +272,7 @@ static ssize_t proc_pid_cmdline_read(str int nr_read; _count = min3(count, len, PAGE_SIZE); - nr_read = access_remote_vm(mm, p, page, _count, - FOLL_FORCE); + nr_read = access_remote_vm(mm, p, page, _count, 0); if (nr_read < 0) rv = nr_read; if (nr_read <= 0) @@ -308,8 +307,7 @@ static ssize_t proc_pid_cmdline_read(str bool final; _count = min3(count, len, PAGE_SIZE); - nr_read = access_remote_vm(mm, p, page, _count, - FOLL_FORCE); + nr_read = access_remote_vm(mm, p, page, _count, 0); if (nr_read < 0) rv = nr_read; if (nr_read <= 0) @@ -358,8 +356,7 @@ skip_argv: bool final; _count = min3(count, len, PAGE_SIZE); - nr_read = access_remote_vm(mm, p, page, _count, - FOLL_FORCE); + nr_read = access_remote_vm(mm, p, page, _count, 0); if (nr_read < 0) rv = nr_read; if (nr_read <= 0) @@ -871,7 +868,7 @@ static ssize_t mem_rw(struct file *file, unsigned long addr = *ppos; ssize_t copied; char *page; - unsigned int flags = FOLL_FORCE; + unsigned int flags; if (!mm) return 0; @@ -884,6 +881,8 @@ static ssize_t mem_rw(struct file *file, if (!atomic_inc_not_zero(&mm->mm_users)) goto free; + /* Maybe we should limit FOLL_FORCE to actual ptrace users? */ + flags = FOLL_FORCE; if (write) flags |= FOLL_WRITE; @@ -1007,8 +1006,7 @@ static ssize_t environ_read(struct file max_len = min_t(size_t, PAGE_SIZE, count); this_len = min(max_len, this_len); - retval = access_remote_vm(mm, (env_start + src), - page, this_len, FOLL_FORCE); + retval = access_remote_vm(mm, (env_start + src), page, this_len, 0); if (retval <= 0) { ret = retval;