Retire .write/.read file operations

The .write/.read file operations callbacks can be retired since
support for .read_iter/.write_iter and .aio_read/.aio_write has
been added.  The vfs_write()/vfs_read() entry functions will
select the correct interface for the kernel.  This is desirable
because all VFS write/read operations now rely on common code.

This change also add the generic write checks to make sure that
ulimits are enforced correctly on write.

Signed-off-by: Brian Behlendorf <[email protected]>
Signed-off-by: Chunwei Chen <[email protected]>
Closes #5587 
Closes #5673 

6 files changed, 132 insertions, 37 deletions

diff --git a/config/kernel-vfs-rw-iterate.m4 b/config/kernel-vfs-rw-iterate.m4
index f8dc42229..af44beb7b 100644
--- a/config/kernel-vfs-rw-iterate.m4
+++ b/config/kernel-vfs-rw-iterate.m4
@@ -25,3 +25,24 @@ AC_DEFUN([ZFS_AC_KERNEL_VFS_RW_ITERATE],
 		AC_MSG_RESULT(no)
 	])
 ])
+
+dnl #
+dnl # Linux 4.1.x API
+dnl #
+AC_DEFUN([ZFS_AC_KERNEL_GENERIC_WRITE_CHECKS],
+	[AC_MSG_CHECKING([whether generic_write_checks() takes kiocb])
+	ZFS_LINUX_TRY_COMPILE([
+		#include <linux/fs.h>
+
+	],[
+		struct kiocb *iocb = NULL;
+		struct iov_iter *iov = NULL;
+		generic_write_checks(iocb, iov);
+	],[
+		AC_MSG_RESULT(yes)
+		AC_DEFINE(HAVE_GENERIC_WRITE_CHECKS_KIOCB, 1,
+			[generic_write_checks() takes kiocb])
+	],[
+		AC_MSG_RESULT(no)
+	])
+])
diff --git a/config/kernel.m4 b/config/kernel.m4
index 980d2801f..afb0cc7fa 100644
--- a/config/kernel.m4
+++ b/config/kernel.m4
@@ -104,6 +104,7 @@ AC_DEFUN([ZFS_AC_CONFIG_KERNEL], [
 	ZFS_AC_KERNEL_LSEEK_EXECUTE
 	ZFS_AC_KERNEL_VFS_ITERATE
 	ZFS_AC_KERNEL_VFS_RW_ITERATE
+	ZFS_AC_KERNEL_GENERIC_WRITE_CHECKS
 	ZFS_AC_KERNEL_KMAP_ATOMIC_ARGS
 	ZFS_AC_KERNEL_FOLLOW_DOWN_ONE
 	ZFS_AC_KERNEL_MAKE_REQUEST_FN
diff --git a/module/zfs/zpl_file.c b/module/zfs/zpl_file.c
index 356fa78f8..75a3b55d1 100644
--- a/module/zfs/zpl_file.c
+++ b/module/zfs/zpl_file.c
@@ -259,21 +259,6 @@ zpl_read_common(struct inode *ip, const char *buf, size_t len, loff_t *ppos,
 }
 
 static ssize_t
-zpl_read(struct file *filp, char __user *buf, size_t len, loff_t *ppos)
-{
-	cred_t *cr = CRED();
-	ssize_t read;
-
-	crhold(cr);
-	read = zpl_read_common(filp->f_mapping->host, buf, len, ppos,
-	    UIO_USERSPACE, filp->f_flags, cr);
-	crfree(cr);
-
-	file_accessed(filp);
-	return (read);
-}
-
-static ssize_t
 zpl_iter_read_common(struct kiocb *kiocb, const struct iovec *iovp,
     unsigned long nr_segs, size_t count, uio_seg_t seg, size_t skip)
 {
@@ -311,7 +296,14 @@ static ssize_t
 zpl_aio_read(struct kiocb *kiocb, const struct iovec *iovp,
     unsigned long nr_segs, loff_t pos)
 {
-	return (zpl_iter_read_common(kiocb, iovp, nr_segs, kiocb->ki_nbytes,
+	ssize_t ret;
+	size_t count;
+
+	ret = generic_segment_checks(iovp, &nr_segs, &count, VERIFY_WRITE);
+	if (ret)
+		return (ret);
+
+	return (zpl_iter_read_common(kiocb, iovp, nr_segs, count,
 	    UIO_USERSPACE, 0));
 }
 #endif /* HAVE_VFS_RW_ITERATE */
@@ -349,6 +341,7 @@ zpl_write_common_iovec(struct inode *ip, const struct iovec *iovp, size_t count,
 
 	return (wrote);
 }
+
 inline ssize_t
 zpl_write_common(struct inode *ip, const char *buf, size_t len, loff_t *ppos,
     uio_seg_t segment, int flags, cred_t *cr)
@@ -363,20 +356,6 @@ zpl_write_common(struct inode *ip, const char *buf, size_t len, loff_t *ppos,
 }
 
 static ssize_t
-zpl_write(struct file *filp, const char __user *buf, size_t len, loff_t *ppos)
-{
-	cred_t *cr = CRED();
-	ssize_t wrote;
-
-	crhold(cr);
-	wrote = zpl_write_common(filp->f_mapping->host, buf, len, ppos,
-	    UIO_USERSPACE, filp->f_flags, cr);
-	crfree(cr);
-
-	return (wrote);
-}
-
-static ssize_t
 zpl_iter_write_common(struct kiocb *kiocb, const struct iovec *iovp,
     unsigned long nr_segs, size_t count, uio_seg_t seg, size_t skip)
 {
@@ -396,16 +375,40 @@ zpl_iter_write_common(struct kiocb *kiocb, const struct iovec *iovp,
 static ssize_t
 zpl_iter_write(struct kiocb *kiocb, struct iov_iter *from)
 {
+	size_t count;
 	ssize_t ret;
 	uio_seg_t seg = UIO_USERSPACE;
+
+#ifndef HAVE_GENERIC_WRITE_CHECKS_KIOCB
+	struct file *file = kiocb->ki_filp;
+	struct address_space *mapping = file->f_mapping;
+	struct inode *ip = mapping->host;
+	int isblk = S_ISBLK(ip->i_mode);
+
+	count = iov_iter_count(from);
+	ret = generic_write_checks(file, &kiocb->ki_pos, &count, isblk);
+#else
+	/*
+	 * XXX - ideally this check should be in the same lock region with
+	 * write operations, so that there's no TOCTTOU race when doing
+	 * append and someone else grow the file.
+	 */
+	ret = generic_write_checks(kiocb, from);
+	count = ret;
+#endif
+	if (ret <= 0)
+		return (ret);
+
 	if (from->type & ITER_KVEC)
 		seg = UIO_SYSSPACE;
 	if (from->type & ITER_BVEC)
 		seg = UIO_BVEC;
+
 	ret = zpl_iter_write_common(kiocb, from->iov, from->nr_segs,
-	    iov_iter_count(from), seg, from->iov_offset);
+	    count, seg, from->iov_offset);
 	if (ret > 0)
 		iov_iter_advance(from, ret);
+
 	return (ret);
 }
 #else
@@ -413,7 +416,22 @@ static ssize_t
 zpl_aio_write(struct kiocb *kiocb, const struct iovec *iovp,
     unsigned long nr_segs, loff_t pos)
 {
-	return (zpl_iter_write_common(kiocb, iovp, nr_segs, kiocb->ki_nbytes,
+	struct file *file = kiocb->ki_filp;
+	struct address_space *mapping = file->f_mapping;
+	struct inode *ip = mapping->host;
+	int isblk = S_ISBLK(ip->i_mode);
+	size_t count;
+	ssize_t ret;
+
+	ret = generic_segment_checks(iovp, &nr_segs, &count, VERIFY_READ);
+	if (ret)
+		return (ret);
+
+	ret = generic_write_checks(file, &pos, &count, isblk);
+	if (ret)
+		return (ret);
+
+	return (zpl_iter_write_common(kiocb, iovp, nr_segs, count,
 	    UIO_USERSPACE, 0));
 }
 #endif /* HAVE_VFS_RW_ITERATE */
@@ -835,8 +853,6 @@ const struct file_operations zpl_file_operations = {
 	.open		= zpl_open,
 	.release	= zpl_release,
 	.llseek		= zpl_llseek,
-	.read		= zpl_read,
-	.write		= zpl_write,
 #ifdef HAVE_VFS_RW_ITERATE
 	.read_iter	= zpl_iter_read,
 	.write_iter	= zpl_iter_write,
diff --git a/tests/runfiles/linux.run b/tests/runfiles/linux.run
index f09001d24..0a8198236 100644
--- a/tests/runfiles/linux.run
+++ b/tests/runfiles/linux.run
@@ -458,8 +458,9 @@ tests = ['inuse_004_pos']
 post =
 
 # DISABLED: needs investigation
-#[tests/functional/large_files]
-#tests = ['large_files_001_pos']
+# large_files_001_pos
+[tests/functional/large_files]
+tests = ['large_files_002_pos']
 
 # DISABLED: needs investigation
 #[tests/functional/largest_pool]
diff --git a/tests/zfs-tests/tests/functional/large_files/Makefile.am b/tests/zfs-tests/tests/functional/large_files/Makefile.am
index d78ea4722..0e4715330 100644
--- a/tests/zfs-tests/tests/functional/large_files/Makefile.am
+++ b/tests/zfs-tests/tests/functional/large_files/Makefile.am
@@ -2,4 +2,5 @@ pkgdatadir = $(datadir)/@PACKAGE@/zfs-tests/tests/functional/large_files
 dist_pkgdata_SCRIPTS = \
 	setup.ksh \
 	cleanup.ksh \
-	large_files_001_pos.ksh
+	large_files_001_pos.ksh \
+	large_files_002_pos.ksh
diff --git a/tests/zfs-tests/tests/functional/large_files/large_files_002_pos.ksh b/tests/zfs-tests/tests/functional/large_files/large_files_002_pos.ksh
new file mode 100755
index 000000000..152fe2e08
--- /dev/null
+++ b/tests/zfs-tests/tests/functional/large_files/large_files_002_pos.ksh
@@ -0,0 +1,55 @@
+#!/bin/ksh -p
+#
+# CDDL HEADER START
+#
+# The contents of this file are subject to the terms of the
+# Common Development and Distribution License (the "License").
+# You may not use this file except in compliance with the License.
+#
+# You can obtain a copy of the license at usr/src/OPENSOLARIS.LICENSE
+# or http://www.opensolaris.org/os/licensing.
+# See the License for the specific language governing permissions
+# and limitations under the License.
+#
+# When distributing Covered Code, include this CDDL HEADER in each
+# file and include the License file at usr/src/OPENSOLARIS.LICENSE.
+# If applicable, add the following below this CDDL HEADER, with the
+# fields enclosed by brackets "[]" replaced with your own identifying
+# information: Portions Copyright [yyyy] [name of copyright owner]
+#
+# CDDL HEADER END
+#
+
+#
+# Copyright (c) 2015 by Lawrence Livermore National Security, LLC.
+# All rights reserved.
+#
+
+. $STF_SUITE/include/libtest.shlib
+
+#
+# DESCRIPTION:
+# Verify 'ulimit -f' file size restrictions are enforced.
+#
+# STRATEGY:
+# 1. Set ulimit file size to unlimited, verify files can be created.
+# 2. Reduce ulimit file size, verify the expected error is returned.
+#
+
+verify_runnable "both"
+
+log_assert "Verify 'ulimit -f' maximum file size is enforced"
+
+# Verify 'ulimit -f unlimited' works
+log_must ulimit -f unlimited
+log_must sh -c 'dd if=/dev/zero of=$TESTDIR/ulimit_write_file bs=1M count=2'
+log_must sh -c '$TRUNCATE -s2M $TESTDIR/ulimit_trunc_file'
+log_must $RM $TESTDIR/ulimit_write_file $TESTDIR/ulimit_trunc_file
+
+# Verify 'ulimit -f <size>' works
+log_must ulimit -f 1024
+log_mustnot sh -c 'dd if=/dev/zero of=$TESTDIR/ulimit_write_file bs=1M count=2'
+log_mustnot sh -c '$TRUNCATE -s2M $TESTDIR/ulimit_trunc_file'
+log_must $RM $TESTDIR/ulimit_write_file $TESTDIR/ulimit_trunc_file
+
+log_pass "Successfully enforced 'ulimit -f' maximum file size"