From patchwork Thu Oct 24 16:57:32 2019 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: David Howells X-Patchwork-Id: 11210345 Return-Path: Received: from mail.kernel.org (pdx-korg-mail-1.web.codeaurora.org [172.30.200.123]) by pdx-korg-patchwork-2.web.codeaurora.org (Postfix) with ESMTP id F135814ED for ; Thu, 24 Oct 2019 16:57:47 +0000 (UTC) Received: from vger.kernel.org (vger.kernel.org [209.132.180.67]) by mail.kernel.org (Postfix) with ESMTP id C1A7420679 for ; Thu, 24 Oct 2019 16:57:47 +0000 (UTC) Authentication-Results: mail.kernel.org; dkim=pass (1024-bit key) header.d=redhat.com header.i=@redhat.com header.b="hpbu6utc" Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S2410003AbfJXQ5r (ORCPT ); Thu, 24 Oct 2019 12:57:47 -0400 Received: from us-smtp-delivery-1.mimecast.com ([207.211.31.120]:27526 "EHLO us-smtp-1.mimecast.com" rhost-flags-OK-OK-OK-FAIL) by vger.kernel.org with ESMTP id S2410011AbfJXQ5q (ORCPT ); Thu, 24 Oct 2019 12:57:46 -0400 DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=redhat.com; s=mimecast20190719; t=1571936265; h=from:from: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; bh=NYl4FrLbBWaBG1625TSNrJR25CyJ2Un0Gw1Isq0q23I=; b=hpbu6utcg3EbBFzZbS/dRnVGuV40kL4RX7b7lNhVvS3o+ouXRb0WoWrD5XC/n7Cnn1n/Ph Q3PDdXpbuEejwMaxWhtvA4lhyC8KlTwwgNhiotYo8UoMx6I4ltR7gbqb2TPZMn96N9pOw8 r6gnAoGQvyyzBg05lk1PlcewOXOtkZk= Received: from mimecast-mx01.redhat.com (mimecast-mx01.redhat.com [209.132.183.4]) (Using TLS) by relay.mimecast.com with ESMTP id us-mta-377-1QOKCWTXMqehQKb2ajkHSg-1; Thu, 24 Oct 2019 12:57:38 -0400 Received: from smtp.corp.redhat.com (int-mx08.intmail.prod.int.phx2.redhat.com [10.5.11.23]) (using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits)) (No client certificate requested) by mimecast-mx01.redhat.com (Postfix) with ESMTPS id 14C19801E66; Thu, 24 Oct 2019 16:57:36 +0000 (UTC) Received: from warthog.procyon.org.uk (ovpn-121-40.rdu2.redhat.com [10.10.121.40]) by smtp.corp.redhat.com (Postfix) with ESMTP id 2966B5888; Thu, 24 Oct 2019 16:57:33 +0000 (UTC) Organization: Red Hat UK Ltd. Registered Address: Red Hat UK Ltd, Amberley Place, 107-111 Peascod Street, Windsor, Berkshire, SI4 1TE, United Kingdom. Registered in England and Wales under Company Registration No. 3798903 From: David Howells In-Reply-To: <157186182463.3995.13922458878706311997.stgit@warthog.procyon.org.uk> References: <157186182463.3995.13922458878706311997.stgit@warthog.procyon.org.uk> To: torvalds@linux-foundation.org Cc: dhowells@redhat.com, Rasmus Villemoes , Greg Kroah-Hartman , Peter Zijlstra , nicolas.dichtel@6wind.com, raven@themaw.net, Christian Brauner , keyrings@vger.kernel.org, linux-usb@vger.kernel.org, linux-block@vger.kernel.org, linux-security-module@vger.kernel.org, linux-fsdevel@vger.kernel.org, linux-api@vger.kernel.org, linux-kernel@vger.kernel.org Subject: [RFC PATCH 11/10] pipe: Add fsync() support [ver #2] MIME-Version: 1.0 Content-ID: <30392.1571936252.1@warthog.procyon.org.uk> Date: Thu, 24 Oct 2019 17:57:32 +0100 Message-ID: <30394.1571936252@warthog.procyon.org.uk> X-Scanned-By: MIMEDefang 2.84 on 10.5.11.23 X-MC-Unique: 1QOKCWTXMqehQKb2ajkHSg-1 X-Mimecast-Spam-Score: 0 Sender: owner-linux-security-module@vger.kernel.org Precedence: bulk List-ID: pipe: Add fsync() support The keyrings testsuite needs the ability to wait for all the outstanding notifications in the queue to have been processed so that it can then go through them to find out whether the notifications it expected have been emitted. Implement fsync() support for pipes to provide this. The tailmost buffer at the point of calling is marked and fsync adds itself to the list of waiters, noting the tail position to be waited for and marking the buffer as no longer mergeable. Then when the buffer is consumed, if the flag is set, any matching waiters are woken up. Signed-off-by: David Howells --- fs/fuse/dev.c | 1 fs/pipe.c | 61 ++++++++++++++++++++++++++++++++++++++++++++++ fs/splice.c | 3 ++ include/linux/pipe_fs_i.h | 22 ++++++++++++++++ lib/iov_iter.c | 2 - 5 files changed, 88 insertions(+), 1 deletion(-) diff --git a/fs/fuse/dev.c b/fs/fuse/dev.c index 5ef57a322cb8..9617a35579cb 100644 --- a/fs/fuse/dev.c +++ b/fs/fuse/dev.c @@ -1983,6 +1983,7 @@ static ssize_t fuse_dev_splice_write(struct pipe_inode_info *pipe, if (rem >= ibuf->len) { *obuf = *ibuf; ibuf->ops = NULL; + pipe_wake_fsync(pipe, ibuf, tail); tail++; pipe_commit_read(pipe, tail); } else { diff --git a/fs/pipe.c b/fs/pipe.c index 6a982a88f658..8e5fd7314be1 100644 --- a/fs/pipe.c +++ b/fs/pipe.c @@ -30,6 +30,12 @@ #include "internal.h" +struct pipe_fsync { + struct list_head link; /* Link in pipe->fsync */ + struct completion done; + unsigned int tail; /* The buffer being waited for */ +}; + /* * The max size that a non-root user is allowed to grow the pipe. Can * be set by root in /proc/sys/fs/pipe-max-size @@ -269,6 +275,58 @@ static bool pipe_buf_can_merge(struct pipe_buffer *buf) return buf->ops == &anon_pipe_buf_ops; } +/* + * Wait for all the data currently in the pipe to be consumed. + */ +static int pipe_fsync(struct file *file, loff_t a, loff_t b, int datasync) +{ + struct pipe_inode_info *pipe = file->private_data; + struct pipe_buffer *buf; + struct pipe_fsync fsync; + unsigned int head, tail, mask; + + pipe_lock(pipe); + + head = pipe->head; + tail = pipe->tail; + mask = pipe->ring_size - 1; + + if (pipe_empty(head, tail)) { + pipe_unlock(pipe); + return 0; + } + + init_completion(&fsync.done); + fsync.tail = tail; + buf = &pipe->bufs[tail & mask]; + buf->flags |= PIPE_BUF_FLAG_FSYNC; + pipe_buf_mark_unmergeable(buf); + list_add_tail(&fsync.link, &pipe->fsync); + pipe_unlock(pipe); + + if (wait_for_completion_interruptible(&fsync.done) < 0) { + pipe_lock(pipe); + list_del(&fsync.link); + pipe_unlock(pipe); + return -EINTR; + } + + return 0; +} + +void __pipe_wake_fsync(struct pipe_inode_info *pipe, unsigned int tail) +{ + struct pipe_fsync *fsync, *p; + + list_for_each_entry_safe(fsync, p, &pipe->fsync, link) { + if (fsync->tail == tail) { + list_del_init(&fsync->link); + complete(&fsync->done); + } + } +} +EXPORT_SYMBOL(__pipe_wake_fsync); + static ssize_t pipe_read(struct kiocb *iocb, struct iov_iter *to) { @@ -325,6 +383,7 @@ pipe_read(struct kiocb *iocb, struct iov_iter *to) if (!buf->len) { pipe_buf_release(pipe, buf); spin_lock_irq(&pipe->wait.lock); + pipe_wake_fsync(pipe, buf, tail); tail++; pipe_commit_read(pipe, tail); do_wakeup = 1; @@ -717,6 +776,7 @@ struct pipe_inode_info *alloc_pipe_info(void) pipe->ring_size = pipe_bufs; pipe->user = user; mutex_init(&pipe->mutex); + INIT_LIST_HEAD(&pipe->fsync); return pipe; } @@ -1060,6 +1120,7 @@ const struct file_operations pipefifo_fops = { .llseek = no_llseek, .read_iter = pipe_read, .write_iter = pipe_write, + .fsync = pipe_fsync, .poll = pipe_poll, .unlocked_ioctl = pipe_ioctl, .release = pipe_release, diff --git a/fs/splice.c b/fs/splice.c index 3f72bc31b6ec..e106367e1be6 100644 --- a/fs/splice.c +++ b/fs/splice.c @@ -523,6 +523,7 @@ static int splice_from_pipe_feed(struct pipe_inode_info *pipe, struct splice_des if (!buf->len) { pipe_buf_release(pipe, buf); + pipe_wake_fsync(pipe, buf, tail); tail++; pipe_commit_read(pipe, tail); if (pipe->files) @@ -771,6 +772,7 @@ iter_file_splice_write(struct pipe_inode_info *pipe, struct file *out, ret -= buf->len; buf->len = 0; pipe_buf_release(pipe, buf); + pipe_wake_fsync(pipe, buf, tail); tail++; pipe_commit_read(pipe, tail); if (pipe->files) @@ -1613,6 +1615,7 @@ static int splice_pipe_to_pipe(struct pipe_inode_info *ipipe, */ *obuf = *ibuf; ibuf->ops = NULL; + pipe_wake_fsync(ipipe, ibuf, i_tail); i_tail++; pipe_commit_read(ipipe, i_tail); input_wakeup = true; diff --git a/include/linux/pipe_fs_i.h b/include/linux/pipe_fs_i.h index 90055ff16550..1a3027089558 100644 --- a/include/linux/pipe_fs_i.h +++ b/include/linux/pipe_fs_i.h @@ -8,6 +8,7 @@ #define PIPE_BUF_FLAG_ATOMIC 0x02 /* was atomically mapped */ #define PIPE_BUF_FLAG_GIFT 0x04 /* page is a gift */ #define PIPE_BUF_FLAG_PACKET 0x08 /* read() as a packet */ +#define PIPE_BUF_FLAG_FSYNC 0x10 /* fsync() is waiting for this buffer to die */ /** * struct pipe_buffer - a linux kernel pipe buffer @@ -43,6 +44,7 @@ struct pipe_buffer { * @w_counter: writer counter * @fasync_readers: reader side fasync * @fasync_writers: writer side fasync + * @fsync: Waiting fsyncs * @bufs: the circular array of pipe buffers * @user: the user who created this pipe **/ @@ -62,6 +64,7 @@ struct pipe_inode_info { struct page *tmp_page; struct fasync_struct *fasync_readers; struct fasync_struct *fasync_writers; + struct list_head fsync; struct pipe_buffer *bufs; struct user_struct *user; }; @@ -268,6 +271,25 @@ extern const struct pipe_buf_operations nosteal_pipe_buf_ops; long pipe_fcntl(struct file *, unsigned int, unsigned long arg); struct pipe_inode_info *get_pipe_info(struct file *file); +void __pipe_wake_fsync(struct pipe_inode_info *pipe, unsigned int tail); + +/** + * pipe_wake_fsync - Wake up anyone waiting with fsync for this point + * @pipe: The pipe that owns the buffer + * @buf: The pipe buffer in question + * @tail: The index in the ring of the buffer + * + * Check to see if anyone is waiting for the pipe ring to clear up to and + * including this buffer, and, if they are, wake them up. + */ +static inline void pipe_wake_fsync(struct pipe_inode_info *pipe, + struct pipe_buffer *buf, + unsigned int tail) +{ + if (unlikely(buf->flags & PIPE_BUF_FLAG_FSYNC)) + __pipe_wake_fsync(pipe, tail); +} + int create_pipe_files(struct file **, int); unsigned int round_pipe_size(unsigned long size); diff --git a/lib/iov_iter.c b/lib/iov_iter.c index e22f4e283f6d..38d52524cd21 100644 --- a/lib/iov_iter.c +++ b/lib/iov_iter.c @@ -404,7 +404,7 @@ static size_t copy_page_to_iter_pipe(struct page *page, size_t offset, size_t by buf->offset = offset; buf->len = bytes; - pipe_commit_read(pipe, i_head); + pipe_commit_write(pipe, i_head); i->iov_offset = offset + bytes; i->head = i_head; out: