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=-13.8 required=3.0 tests=BAYES_00, HEADER_FROM_DIFFERENT_DOMAINS,INCLUDES_CR_TRAILER,INCLUDES_PATCH, MAILING_LIST_MULTI,SPF_HELO_NONE,SPF_PASS,URIBL_BLOCKED autolearn=ham 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 B6827C432C3 for ; Thu, 18 Feb 2021 19:24:06 +0000 (UTC) Received: from vger.kernel.org (vger.kernel.org [23.128.96.18]) by mail.kernel.org (Postfix) with ESMTP id 6ED6764ECD for ; Thu, 18 Feb 2021 19:24:06 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S233157AbhBRTXh (ORCPT ); Thu, 18 Feb 2021 14:23:37 -0500 Received: from out02.mta.xmission.com ([166.70.13.232]:58554 "EHLO out02.mta.xmission.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S230074AbhBRS0W (ORCPT ); Thu, 18 Feb 2021 13:26:22 -0500 Received: from in02.mta.xmission.com ([166.70.13.52]) by out02.mta.xmission.com with esmtps (TLS1.3) tls TLS_ECDHE_RSA_WITH_AES_256_GCM_SHA384 (Exim 4.93) (envelope-from ) id 1lCnzd-00AhAQ-G9; Thu, 18 Feb 2021 11:25:37 -0700 Received: from ip68-227-160-95.om.om.cox.net ([68.227.160.95] helo=fess.xmission.com) by in02.mta.xmission.com with esmtpsa (TLS1.3) tls TLS_ECDHE_RSA_WITH_AES_256_GCM_SHA384 (Exim 4.93) (envelope-from ) id 1lCnzW-008PRY-LE; Thu, 18 Feb 2021 11:25:33 -0700 From: ebiederm@xmission.com (Eric W. Biederman) To: "Dmitry V. Levin" Cc: "David S. Miller" , Alexey Gladkov , Gleb Fotengauer-Malinovskiy , Anatoly Pugachev , sparclinux@vger.kernel.org, linux-kernel@vger.kernel.org References: <20210217080000.GA25861@altlinux.org> Date: Thu, 18 Feb 2021 12:21:40 -0600 In-Reply-To: <20210217080000.GA25861@altlinux.org> (Dmitry V. Levin's message of "Wed, 17 Feb 2021 08:00:00 +0000") Message-ID: User-Agent: Gnus/5.13 (Gnus v5.13) Emacs/26.1 (gnu/linux) MIME-Version: 1.0 Content-Type: text/plain X-XM-SPF: eid=1lCnzW-008PRY-LE;;;mid=;;;hst=in02.mta.xmission.com;;;ip=68.227.160.95;;;frm=ebiederm@xmission.com;;;spf=neutral X-XM-AID: U2FsdGVkX1+0aLczMW3Vl2GWSnFg6d1KsNSFnL/apdU= X-SA-Exim-Connect-IP: 68.227.160.95 X-SA-Exim-Mail-From: ebiederm@xmission.com Subject: Re: [PATCH] sparc: make copy_thread honor pid namespaces X-SA-Exim-Version: 4.2.1 (built Sat, 08 Feb 2020 21:53:50 +0000) X-SA-Exim-Scanned: Yes (on in02.mta.xmission.com) Precedence: bulk List-ID: X-Mailing-List: linux-kernel@vger.kernel.org "Dmitry V. Levin" writes: > On sparc, fork and clone syscalls have an unusual semantics of > returning the pid of the parent process to the child process. > > Apparently, the implementation did not honor pid namespaces at all, > so the child used to get the pid of its parent in the init namespace. > > This bug was found by strace test suite. > > Reproducer: > > $ gcc -Wall -O2 -xc - <<'EOF' > #define _GNU_SOURCE > #include > #include > #include > #include > #include > #include > #include > #include > #include > int main(void) > { > if (unshare(CLONE_NEWPID | CLONE_NEWUSER) < 0) > err(1, "unshare"); > int pid = syscall(__NR_fork); > if (pid < 0) > err(1, "fork"); > fprintf(stderr, "current: %d, parent: %d, fork returned: %d\n", > getpid(), getppid(), pid); > int status; > if (wait(&status) < 0) { > if (errno == ECHILD) > _exit(0); > err(1, "wait"); > } > return !WIFEXITED(status) || WEXITSTATUS(status) != 0; > } > EOF > $ sh -c ./a.out > current: 10001, parent: 10000, fork returned: 10002 > current: 1, parent: 0, fork returned: 10001 >From glibc's sysdeps/unix/sparc/fork.S: > SYSCALL__ (fork, 0) > /* %o1 is now 0 for the parent and 1 for the child. Decrement it to > make it -1 (all bits set) for the parent, and 0 (no bits set) > for the child. Then AND it with %o0, so the parent gets > %o0&-1==pid, and the child gets %o0&0==0. */ > sub %o1, 1, %o1 > retl > and %o0, %o1, %o0 > libc_hidden_def (__fork) > > weak_alias (__fork, fork) This needs to be shared to make the test case make sense. The change below looks reasonable. Unfortunately because copy_thread comes before init_task_pid task_active_pid_ns(p) will not work correctly. The code below needs to use current->nsproxy->pid_ns_for_children instead of task_active_pid_ns(p). For sparc people. Do we know of anyone who actually uses the parent pid returned from fork to the child process? If not the code can simply return 0 and we don't have to do this. Eric > Cc: Eric W. Biederman > Cc: stable@vger.kernel.org > Signed-off-by: Dmitry V. Levin > --- > > Although the fix seems to be obvious, I have no means to test it myself, > so any help with the testing is much appreciated. > > arch/sparc/kernel/process_32.c | 2 +- > arch/sparc/kernel/process_64.c | 2 +- > 2 files changed, 2 insertions(+), 2 deletions(-) > > diff --git a/arch/sparc/kernel/process_32.c b/arch/sparc/kernel/process_32.c > index a02363735915..7a89969befa8 100644 > --- a/arch/sparc/kernel/process_32.c > +++ b/arch/sparc/kernel/process_32.c > @@ -368,7 +368,7 @@ int copy_thread(unsigned long clone_flags, unsigned long sp, unsigned long arg, > #endif > > /* Set the return value for the child. */ > - childregs->u_regs[UREG_I0] = current->pid; > + childregs->u_regs[UREG_I0] = task_pid_nr_ns(current, task_active_pid_ns(p)); > childregs->u_regs[UREG_I1] = 1; > > /* Set the return value for the parent. */ > diff --git a/arch/sparc/kernel/process_64.c b/arch/sparc/kernel/process_64.c > index 6f8c7822fc06..ec97217ab970 100644 > --- a/arch/sparc/kernel/process_64.c > +++ b/arch/sparc/kernel/process_64.c > @@ -629,7 +629,7 @@ int copy_thread(unsigned long clone_flags, unsigned long sp, unsigned long arg, > t->utraps[0]++; > > /* Set the return value for the child. */ > - t->kregs->u_regs[UREG_I0] = current->pid; > + t->kregs->u_regs[UREG_I0] = task_pid_nr_ns(current, task_active_pid_ns(p)); > t->kregs->u_regs[UREG_I1] = 1; > > /* Set the second return value for the parent. */