summaryrefslogtreecommitdiff
diff options
context:
space:
mode:
authorAl Viro <viro@zeniv.linux.org.uk>2014-02-11 23:37:41 (GMT)
committerPaul Osmialowski <p.osmialowsk@samsung.com>2015-04-15 15:45:09 (GMT)
commit778a43408f732464f684f06af742eacdb9748eb8 (patch)
tree87eb906dd43462d30129a0451ea514e5bbfaf840
parent8d4f4bf0cc146699097ba8417b03e0bef8b0f976 (diff)
downloadlinux-3.10-sandbox/pawelo/kdbus-next-wip.zip
linux-3.10-sandbox/pawelo/kdbus-next-wip.tar.gz
linux-3.10-sandbox/pawelo/kdbus-next-wip.tar.bz2
new methods: ->read_iter() and ->write_iter()sandbox/pawelo/kdbus-next-wip
Beginning to introduce those. Just the callers for now, and it's clumsier than it'll eventually become; once we finish converting aio_read and aio_write instances, the things will get nicer. For now, these guys are in parallel to ->aio_read() and ->aio_write(); they take iocb and iov_iter, with everything in iov_iter already validated. File offset is passed in iocb->ki_pos, iov/nr_segs - in iov_iter. Main concerns in that series are stack footprint and ability to split the damn thing cleanly. [fix from Peter Ujfalusi <peter.ujfalusi@ti.com> folded] Change-Id: Ib9c6cff05b7ecf8f1052a74c4ec603efb21514e5 Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
-rw-r--r--Documentation/filesystems/Locking2
-rw-r--r--Documentation/filesystems/vfs.txt10
-rw-r--r--fs/aio.c17
-rw-r--r--fs/read_write.c90
-rw-r--r--include/linux/fs.h5
5 files changed, 113 insertions, 11 deletions
diff --git a/Documentation/filesystems/Locking b/Documentation/filesystems/Locking
index bdd82b2..130b4f5 100644
--- a/Documentation/filesystems/Locking
+++ b/Documentation/filesystems/Locking
@@ -414,6 +414,8 @@ prototypes:
ssize_t (*write) (struct file *, const char __user *, size_t, loff_t *);
ssize_t (*aio_read) (struct kiocb *, const struct iovec *, unsigned long, loff_t);
ssize_t (*aio_write) (struct kiocb *, const struct iovec *, unsigned long, loff_t);
+ ssize_t (*read_iter) (struct kiocb *, struct iov_iter *);
+ ssize_t (*write_iter) (struct kiocb *, struct iov_iter *);
int (*iterate) (struct file *, struct dir_context *);
unsigned int (*poll) (struct file *, struct poll_table_struct *);
long (*unlocked_ioctl) (struct file *, unsigned int, unsigned long);
diff --git a/Documentation/filesystems/vfs.txt b/Documentation/filesystems/vfs.txt
index 4a35f66..63851b8 100644
--- a/Documentation/filesystems/vfs.txt
+++ b/Documentation/filesystems/vfs.txt
@@ -777,6 +777,8 @@ struct file_operations {
ssize_t (*write) (struct file *, const char __user *, size_t, loff_t *);
ssize_t (*aio_read) (struct kiocb *, const struct iovec *, unsigned long, loff_t);
ssize_t (*aio_write) (struct kiocb *, const struct iovec *, unsigned long, loff_t);
+ ssize_t (*read_iter) (struct kiocb *, struct iov_iter *);
+ ssize_t (*write_iter) (struct kiocb *, struct iov_iter *);
int (*iterate) (struct file *, struct dir_context *);
unsigned int (*poll) (struct file *, struct poll_table_struct *);
long (*unlocked_ioctl) (struct file *, unsigned int, unsigned long);
@@ -809,11 +811,15 @@ otherwise noted.
read: called by read(2) and related system calls
- aio_read: called by io_submit(2) and other asynchronous I/O operations
+ aio_read: vectored, possibly asynchronous read
+
+ read_iter: possibly asynchronous read with iov_iter as destination
write: called by write(2) and related system calls
- aio_write: called by io_submit(2) and other asynchronous I/O operations
+ aio_write: vectored, possibly asynchronous write
+
+ write_iter: possibly asynchronous write with iov_iter as source
iterate: called when the VFS needs to read the directory contents
diff --git a/fs/aio.c b/fs/aio.c
index 9491a00..7147d71 100644
--- a/fs/aio.c
+++ b/fs/aio.c
@@ -882,6 +882,7 @@ SYSCALL_DEFINE1(io_destroy, aio_context_t, ctx)
typedef ssize_t (aio_rw_op)(struct kiocb *, const struct iovec *,
unsigned long, loff_t);
+typedef ssize_t (rw_iter_op)(struct kiocb *, struct iov_iter *);
static ssize_t aio_setup_vectored_rw(int rw, struct kiocb *kiocb, bool compat)
{
@@ -932,7 +933,9 @@ static ssize_t aio_run_iocb(struct kiocb *req, bool compat)
ssize_t ret;
int rw;
fmode_t mode;
+ rw_iter_op *iter_op;
aio_rw_op *rw_op;
+ struct iov_iter iter;
switch (req->ki_opcode) {
case IOCB_CMD_PREAD:
@@ -940,6 +943,7 @@ static ssize_t aio_run_iocb(struct kiocb *req, bool compat)
mode = FMODE_READ;
rw = READ;
rw_op = file->f_op->aio_read;
+ iter_op = file->f_op->read_iter;
goto rw_common;
case IOCB_CMD_PWRITE:
@@ -947,12 +951,13 @@ static ssize_t aio_run_iocb(struct kiocb *req, bool compat)
mode = FMODE_WRITE;
rw = WRITE;
rw_op = file->f_op->aio_write;
+ iter_op = file->f_op->write_iter;
goto rw_common;
rw_common:
if (unlikely(!(file->f_mode & mode)))
return -EBADF;
- if (!rw_op)
+ if (!rw_op && !iter_op)
return -EINVAL;
ret = (req->ki_opcode == IOCB_CMD_PREADV ||
@@ -978,8 +983,14 @@ rw_common:
if (rw == WRITE)
file_start_write(file);
- ret = rw_op(req, req->ki_iovec,
- req->ki_nr_segs, req->ki_pos);
+ if (iter_op) {
+ iov_iter_init(&iter, rw, req->ki_iovec,
+ req->ki_nr_segs, req->ki_nbytes);
+ ret = iter_op(req, &iter);
+ } else {
+ ret = rw_op(req, req->ki_iovec,
+ req->ki_nr_segs, req->ki_pos);
+ }
if (rw == WRITE)
file_end_write(file);
diff --git a/fs/read_write.c b/fs/read_write.c
index 290c551..e49c3e2 100644
--- a/fs/read_write.c
+++ b/fs/read_write.c
@@ -25,6 +25,7 @@
typedef ssize_t (*io_fn_t)(struct file *, char __user *, size_t, loff_t *);
typedef ssize_t (*iov_fn_t)(struct kiocb *, const struct iovec *,
unsigned long, loff_t);
+typedef ssize_t (*iter_fn_t)(struct kiocb *, struct iov_iter *);
const struct file_operations generic_ro_fops = {
.llseek = generic_file_llseek,
@@ -395,6 +396,27 @@ ssize_t do_sync_read(struct file *filp, char __user *buf, size_t len, loff_t *pp
EXPORT_SYMBOL(do_sync_read);
+ssize_t new_sync_read(struct file *filp, char __user *buf, size_t len, loff_t *ppos)
+{
+ struct iovec iov = { .iov_base = buf, .iov_len = len };
+ struct kiocb kiocb;
+ struct iov_iter iter;
+ ssize_t ret;
+
+ init_sync_kiocb(&kiocb, filp);
+ kiocb.ki_pos = *ppos;
+ kiocb.ki_nbytes = len;
+ iov_iter_init(&iter, READ, &iov, 1, len);
+
+ ret = filp->f_op->read_iter(&kiocb, &iter);
+ if (-EIOCBQUEUED == ret)
+ ret = wait_on_sync_kiocb(&kiocb);
+ *ppos = kiocb.ki_pos;
+ return ret;
+}
+
+EXPORT_SYMBOL(new_sync_read);
+
ssize_t vfs_read(struct file *file, char __user *buf, size_t count, loff_t *pos)
{
ssize_t ret;
@@ -411,8 +433,10 @@ ssize_t vfs_read(struct file *file, char __user *buf, size_t count, loff_t *pos)
count = ret;
if (file->f_op->read)
ret = file->f_op->read(file, buf, count, pos);
- else
+ else if (file->f_op->aio_read)
ret = do_sync_read(file, buf, count, pos);
+ else
+ ret = new_sync_read(file, buf, count, pos);
if (ret > 0) {
fsnotify_access(file);
add_rchar(current, ret);
@@ -444,6 +468,27 @@ ssize_t do_sync_write(struct file *filp, const char __user *buf, size_t len, lof
EXPORT_SYMBOL(do_sync_write);
+ssize_t new_sync_write(struct file *filp, const char __user *buf, size_t len, loff_t *ppos)
+{
+ struct iovec iov = { .iov_base = (void __user *)buf, .iov_len = len };
+ struct kiocb kiocb;
+ struct iov_iter iter;
+ ssize_t ret;
+
+ init_sync_kiocb(&kiocb, filp);
+ kiocb.ki_pos = *ppos;
+ kiocb.ki_nbytes = len;
+ iov_iter_init(&iter, WRITE, &iov, 1, len);
+
+ ret = filp->f_op->write_iter(&kiocb, &iter);
+ if (-EIOCBQUEUED == ret)
+ ret = wait_on_sync_kiocb(&kiocb);
+ *ppos = kiocb.ki_pos;
+ return ret;
+}
+
+EXPORT_SYMBOL(new_sync_write);
+
ssize_t __kernel_write(struct file *file, const char *buf, size_t count, loff_t *pos)
{
mm_segment_t old_fs;
@@ -460,8 +505,10 @@ ssize_t __kernel_write(struct file *file, const char *buf, size_t count, loff_t
count = MAX_RW_COUNT;
if (file->f_op->write)
ret = file->f_op->write(file, p, count, pos);
- else
+ else if (file->f_op->aio_write)
ret = do_sync_write(file, p, count, pos);
+ else
+ ret = new_sync_write(file, p, count, pos);
set_fs(old_fs);
if (ret > 0) {
fsnotify_modify(file);
@@ -488,8 +535,10 @@ ssize_t vfs_write(struct file *file, const char __user *buf, size_t count, loff_
file_start_write(file);
if (file->f_op->write)
ret = file->f_op->write(file, buf, count, pos);
- else
+ else if (file->f_op->aio_write)
ret = do_sync_write(file, buf, count, pos);
+ else
+ ret = new_sync_write(file, buf, count, pos);
if (ret > 0) {
fsnotify_modify(file);
add_wchar(current, ret);
@@ -604,6 +653,25 @@ unsigned long iov_shorten(struct iovec *iov, unsigned long nr_segs, size_t to)
}
EXPORT_SYMBOL(iov_shorten);
+static ssize_t do_iter_readv_writev(struct file *filp, int rw, const struct iovec *iov,
+ unsigned long nr_segs, size_t len, loff_t *ppos, iter_fn_t fn)
+{
+ struct kiocb kiocb;
+ struct iov_iter iter;
+ ssize_t ret;
+
+ init_sync_kiocb(&kiocb, filp);
+ kiocb.ki_pos = *ppos;
+ kiocb.ki_nbytes = len;
+
+ iov_iter_init(&iter, rw, iov, nr_segs, len);
+ ret = fn(&kiocb, &iter);
+ if (ret == -EIOCBQUEUED)
+ ret = wait_on_sync_kiocb(&kiocb);
+ *ppos = kiocb.ki_pos;
+ return ret;
+}
+
static ssize_t do_sync_readv_writev(struct file *filp, const struct iovec *iov,
unsigned long nr_segs, size_t len, loff_t *ppos, iov_fn_t fn)
{
@@ -741,6 +809,7 @@ static ssize_t do_readv_writev(int type, struct file *file,
ssize_t ret;
io_fn_t fn;
iov_fn_t fnv;
+ iter_fn_t iter_fn;
if (!file->f_op) {
ret = -EINVAL;
@@ -761,13 +830,18 @@ static ssize_t do_readv_writev(int type, struct file *file,
if (type == READ) {
fn = file->f_op->read;
fnv = file->f_op->aio_read;
+ iter_fn = file->f_op->read_iter;
} else {
fn = (io_fn_t)file->f_op->write;
fnv = file->f_op->aio_write;
+ iter_fn = file->f_op->write_iter;
file_start_write(file);
}
- if (fnv)
+ if (iter_fn)
+ ret = do_iter_readv_writev(file, type, iov, nr_segs, tot_len,
+ pos, iter_fn);
+ else if (fnv)
ret = do_sync_readv_writev(file, iov, nr_segs, tot_len,
pos, fnv);
else
@@ -918,6 +992,7 @@ static ssize_t compat_do_readv_writev(int type, struct file *file,
ssize_t ret;
io_fn_t fn;
iov_fn_t fnv;
+ iter_fn_t iter_fn;
ret = -EINVAL;
if (!file->f_op)
@@ -941,13 +1016,18 @@ static ssize_t compat_do_readv_writev(int type, struct file *file,
if (type == READ) {
fn = file->f_op->read;
fnv = file->f_op->aio_read;
+ iter_fn = file->f_op->read_iter;
} else {
fn = (io_fn_t)file->f_op->write;
fnv = file->f_op->aio_write;
+ iter_fn = file->f_op->write_iter;
file_start_write(file);
}
- if (fnv)
+ if (iter_fn)
+ ret = do_iter_readv_writev(file, type, iov, nr_segs, tot_len,
+ pos, iter_fn);
+ else if (fnv)
ret = do_sync_readv_writev(file, iov, nr_segs, tot_len,
pos, fnv);
else
diff --git a/include/linux/fs.h b/include/linux/fs.h
index bc37b65..37ec2f9 100644
--- a/include/linux/fs.h
+++ b/include/linux/fs.h
@@ -288,7 +288,6 @@ enum positive_aop_returns {
struct page;
struct address_space;
struct writeback_control;
-struct iov_iter;
struct iov_iter {
int type;
@@ -1568,6 +1567,8 @@ struct file_operations {
ssize_t (*write) (struct file *, const char __user *, size_t, loff_t *);
ssize_t (*aio_read) (struct kiocb *, const struct iovec *, unsigned long, loff_t);
ssize_t (*aio_write) (struct kiocb *, const struct iovec *, unsigned long, loff_t);
+ ssize_t (*read_iter) (struct kiocb *, struct iov_iter *);
+ ssize_t (*write_iter) (struct kiocb *, struct iov_iter *);
int (*iterate) (struct file *, struct dir_context *);
unsigned int (*poll) (struct file *, struct poll_table_struct *);
long (*unlocked_ioctl) (struct file *, unsigned int, unsigned long);
@@ -2454,6 +2455,8 @@ extern ssize_t generic_file_buffered_write(struct kiocb *, const struct iovec *,
unsigned long, loff_t, loff_t *, size_t, ssize_t);
extern ssize_t do_sync_read(struct file *filp, char __user *buf, size_t len, loff_t *ppos);
extern ssize_t do_sync_write(struct file *filp, const char __user *buf, size_t len, loff_t *ppos);
+extern ssize_t new_sync_read(struct file *filp, char __user *buf, size_t len, loff_t *ppos);
+extern ssize_t new_sync_write(struct file *filp, const char __user *buf, size_t len, loff_t *ppos);
extern int generic_segment_checks(const struct iovec *iov,
unsigned long *nr_segs, size_t *count, int access_flags);