* that transfers data buffers to or from a pipe buffer.
*
* Named by Larry McVoy, original implementation from Linus, extended by
- * Jens to support splicing to files and fixing the initial implementation
- * bugs.
+ * Jens to support splicing to files, network, direct splicing, etc and
+ * fixing lots of bugs.
*
- * Copyright (C) 2005 Jens Axboe <axboe@suse.de>
- * Copyright (C) 2005 Linus Torvalds <torvalds@osdl.org>
+ * Copyright (C) 2005-2006 Jens Axboe <axboe@suse.de>
+ * Copyright (C) 2005-2006 Linus Torvalds <torvalds@osdl.org>
+ * Copyright (C) 2006 Ingo Molnar <mingo@elte.hu>
*
*/
#include <linux/fs.h>
#include <linux/buffer_head.h>
#include <linux/module.h>
#include <linux/syscalls.h>
+#include <linux/uio.h>
+
+struct partial_page {
+ unsigned int offset;
+ unsigned int len;
+};
/*
- * Passed to the actors
+ * Passed to splice_to_pipe
*/
-struct splice_desc {
- unsigned int len, total_len; /* current and remaining length */
+struct splice_pipe_desc {
+ struct page **pages; /* page map */
+ struct partial_page *partial; /* pages[] may not be contig */
+ int nr_pages; /* number of pages in map */
unsigned int flags; /* splice flags */
- struct file *file; /* file to read/write */
- loff_t pos; /* file position */
+ struct pipe_buf_operations *ops;/* ops associated with output pipe */
};
/*
* addition of remove_mapping(). If success is returned, the caller may
* attempt to reuse this page for another destination.
*/
-static int page_cache_pipe_buf_steal(struct pipe_inode_info *info,
+static int page_cache_pipe_buf_steal(struct pipe_inode_info *pipe,
struct pipe_buffer *buf)
{
struct page *page = buf->page;
- struct address_space *mapping = page_mapping(page);
+ struct address_space *mapping;
- WARN_ON(!PageLocked(page));
- WARN_ON(!PageUptodate(page));
+ lock_page(page);
- /*
- * At least for ext2 with nobh option, we need to wait on writeback
- * completing on this page, since we'll remove it from the pagecache.
- * Otherwise truncate wont wait on the page, allowing the disk
- * blocks to be reused by someone else before we actually wrote our
- * data to them. fs corruption ensues.
- */
- wait_on_page_writeback(page);
+ mapping = page_mapping(page);
+ if (mapping) {
+ WARN_ON(!PageUptodate(page));
- if (PagePrivate(page))
- try_to_release_page(page, mapping_gfp_mask(mapping));
+ /*
+ * At least for ext2 with nobh option, we need to wait on
+ * writeback completing on this page, since we'll remove it
+ * from the pagecache. Otherwise truncate wont wait on the
+ * page, allowing the disk blocks to be reused by someone else
+ * before we actually wrote our data to them. fs corruption
+ * ensues.
+ */
+ wait_on_page_writeback(page);
- if (!remove_mapping(mapping, page))
- return 1;
+ if (PagePrivate(page))
+ try_to_release_page(page, mapping_gfp_mask(mapping));
- buf->flags |= PIPE_BUF_FLAG_STOLEN | PIPE_BUF_FLAG_LRU;
- return 0;
+ /*
+ * If we succeeded in removing the mapping, set LRU flag
+ * and return good.
+ */
+ if (remove_mapping(mapping, page)) {
+ buf->flags |= PIPE_BUF_FLAG_LRU;
+ return 0;
+ }
+ }
+
+ /*
+ * Raced with truncate or failed to remove page from current
+ * address space, unlock and return failure.
+ */
+ unlock_page(page);
+ return 1;
}
-static void page_cache_pipe_buf_release(struct pipe_inode_info *info,
+static void page_cache_pipe_buf_release(struct pipe_inode_info *pipe,
struct pipe_buffer *buf)
{
page_cache_release(buf->page);
- buf->page = NULL;
- buf->flags &= ~(PIPE_BUF_FLAG_STOLEN | PIPE_BUF_FLAG_LRU);
+ buf->flags &= ~PIPE_BUF_FLAG_LRU;
}
-static void *page_cache_pipe_buf_map(struct file *file,
- struct pipe_inode_info *info,
- struct pipe_buffer *buf)
+static int page_cache_pipe_buf_pin(struct pipe_inode_info *pipe,
+ struct pipe_buffer *buf)
{
struct page *page = buf->page;
int err;
/*
* Page got truncated/unhashed. This will cause a 0-byte
- * splice, if this is the first page
+ * splice, if this is the first page.
*/
if (!page->mapping) {
err = -ENODATA;
}
/*
- * uh oh, read-error from disk
+ * Uh oh, read-error from disk.
*/
if (!PageUptodate(page)) {
err = -EIO;
}
/*
- * page is ok afterall, fall through to mapping
+ * Page is ok afterall, we are done.
*/
unlock_page(page);
}
- return kmap(page);
+ return 0;
error:
unlock_page(page);
- return ERR_PTR(err);
+ return err;
}
-static void page_cache_pipe_buf_unmap(struct pipe_inode_info *info,
- struct pipe_buffer *buf)
+static struct pipe_buf_operations page_cache_pipe_buf_ops = {
+ .can_merge = 0,
+ .map = generic_pipe_buf_map,
+ .unmap = generic_pipe_buf_unmap,
+ .pin = page_cache_pipe_buf_pin,
+ .release = page_cache_pipe_buf_release,
+ .steal = page_cache_pipe_buf_steal,
+ .get = generic_pipe_buf_get,
+};
+
+static int user_page_pipe_buf_steal(struct pipe_inode_info *pipe,
+ struct pipe_buffer *buf)
{
- kunmap(buf->page);
+ if (!(buf->flags & PIPE_BUF_FLAG_GIFT))
+ return 1;
+
+ buf->flags |= PIPE_BUF_FLAG_LRU;
+ return generic_pipe_buf_steal(pipe, buf);
}
-static struct pipe_buf_operations page_cache_pipe_buf_ops = {
+static struct pipe_buf_operations user_page_pipe_buf_ops = {
.can_merge = 0,
- .map = page_cache_pipe_buf_map,
- .unmap = page_cache_pipe_buf_unmap,
+ .map = generic_pipe_buf_map,
+ .unmap = generic_pipe_buf_unmap,
+ .pin = generic_pipe_buf_pin,
.release = page_cache_pipe_buf_release,
- .steal = page_cache_pipe_buf_steal,
+ .steal = user_page_pipe_buf_steal,
+ .get = generic_pipe_buf_get,
};
/*
* Pipe output worker. This sets up our pipe format with the page cache
* pipe buffer operations. Otherwise very similar to the regular pipe_writev().
*/
-static ssize_t move_to_pipe(struct pipe_inode_info *pipe, struct page **pages,
- int nr_pages, unsigned long offset,
- unsigned long len, unsigned int flags)
+static ssize_t splice_to_pipe(struct pipe_inode_info *pipe,
+ struct splice_pipe_desc *spd)
{
- int ret, do_wakeup, i;
+ int ret, do_wakeup, page_nr;
ret = 0;
do_wakeup = 0;
- i = 0;
+ page_nr = 0;
if (pipe->inode)
mutex_lock(&pipe->inode->i_mutex);
for (;;) {
- int bufs;
-
if (!pipe->readers) {
send_sig(SIGPIPE, current, 0);
if (!ret)
break;
}
- bufs = pipe->nrbufs;
- if (bufs < PIPE_BUFFERS) {
- int newbuf = (pipe->curbuf + bufs) & (PIPE_BUFFERS - 1);
+ if (pipe->nrbufs < PIPE_BUFFERS) {
+ int newbuf = (pipe->curbuf + pipe->nrbufs) & (PIPE_BUFFERS - 1);
struct pipe_buffer *buf = pipe->bufs + newbuf;
- struct page *page = pages[i++];
- unsigned long this_len;
-
- this_len = PAGE_CACHE_SIZE - offset;
- if (this_len > len)
- this_len = len;
-
- buf->page = page;
- buf->offset = offset;
- buf->len = this_len;
- buf->ops = &page_cache_pipe_buf_ops;
- pipe->nrbufs = ++bufs;
- do_wakeup = 1;
-
- ret += this_len;
- len -= this_len;
- offset = 0;
- if (!--nr_pages)
- break;
- if (!len)
+
+ buf->page = spd->pages[page_nr];
+ buf->offset = spd->partial[page_nr].offset;
+ buf->len = spd->partial[page_nr].len;
+ buf->ops = spd->ops;
+ if (spd->flags & SPLICE_F_GIFT)
+ buf->flags |= PIPE_BUF_FLAG_GIFT;
+
+ pipe->nrbufs++;
+ page_nr++;
+ ret += buf->len;
+
+ if (pipe->inode)
+ do_wakeup = 1;
+
+ if (!--spd->nr_pages)
break;
- if (bufs < PIPE_BUFFERS)
+ if (pipe->nrbufs < PIPE_BUFFERS)
continue;
break;
}
- if (flags & SPLICE_F_NONBLOCK) {
+ if (spd->flags & SPLICE_F_NONBLOCK) {
if (!ret)
ret = -EAGAIN;
break;
kill_fasync(&pipe->fasync_readers, SIGIO, POLL_IN);
}
- while (i < nr_pages)
- page_cache_release(pages[i++]);
+ while (page_nr < spd->nr_pages)
+ page_cache_release(spd->pages[page_nr++]);
return ret;
}
static int
-__generic_file_splice_read(struct file *in, struct pipe_inode_info *pipe,
- size_t len, unsigned int flags)
+__generic_file_splice_read(struct file *in, loff_t *ppos,
+ struct pipe_inode_info *pipe, size_t len,
+ unsigned int flags)
{
struct address_space *mapping = in->f_mapping;
- unsigned int offset, nr_pages;
+ unsigned int loff, nr_pages;
struct page *pages[PIPE_BUFFERS];
+ struct partial_page partial[PIPE_BUFFERS];
struct page *page;
- pgoff_t index;
- int i;
-
- index = in->f_pos >> PAGE_CACHE_SHIFT;
- offset = in->f_pos & ~PAGE_CACHE_MASK;
- nr_pages = (len + offset + PAGE_CACHE_SIZE - 1) >> PAGE_CACHE_SHIFT;
+ pgoff_t index, end_index;
+ loff_t isize;
+ size_t total_len;
+ int error, page_nr;
+ struct splice_pipe_desc spd = {
+ .pages = pages,
+ .partial = partial,
+ .flags = flags,
+ .ops = &page_cache_pipe_buf_ops,
+ };
+
+ index = *ppos >> PAGE_CACHE_SHIFT;
+ loff = *ppos & ~PAGE_CACHE_MASK;
+ nr_pages = (len + loff + PAGE_CACHE_SIZE - 1) >> PAGE_CACHE_SHIFT;
if (nr_pages > PIPE_BUFFERS)
nr_pages = PIPE_BUFFERS;
/*
- * initiate read-ahead on this page range. however, don't call into
+ * Initiate read-ahead on this page range. however, don't call into
* read-ahead if this is a non-zero offset (we are likely doing small
* chunk splice and the page is already there) for a single page.
*/
- if (!offset || nr_pages > 1)
- do_page_cache_readahead(mapping, in, index, nr_pages);
+ if (!loff || nr_pages > 1)
+ page_cache_readahead(mapping, &in->f_ra, in, index, nr_pages);
+
+ /*
+ * Now fill in the holes:
+ */
+ error = 0;
+ total_len = 0;
+
+ /*
+ * Lookup the (hopefully) full range of pages we need.
+ */
+ spd.nr_pages = find_get_pages_contig(mapping, index, nr_pages, pages);
/*
- * now fill in the holes
+ * If find_get_pages_contig() returned fewer pages than we needed,
+ * allocate the rest.
*/
- for (i = 0; i < nr_pages; i++, index++) {
+ index += spd.nr_pages;
+ while (spd.nr_pages < nr_pages) {
/*
- * no page there, look one up / create it
+ * Page could be there, find_get_pages_contig() breaks on
+ * the first hole.
*/
- page = find_or_create_page(mapping, index,
- mapping_gfp_mask(mapping));
- if (!page)
- break;
+ page = find_get_page(mapping, index);
+ if (!page) {
+ /*
+ * Make sure the read-ahead engine is notified
+ * about this failure.
+ */
+ handle_ra_miss(mapping, &in->f_ra, index);
+
+ /*
+ * page didn't exist, allocate one.
+ */
+ page = page_cache_alloc_cold(mapping);
+ if (!page)
+ break;
- if (PageUptodate(page))
+ error = add_to_page_cache_lru(page, mapping, index,
+ mapping_gfp_mask(mapping));
+ if (unlikely(error)) {
+ page_cache_release(page);
+ if (error == -EEXIST)
+ continue;
+ break;
+ }
+ /*
+ * add_to_page_cache() locks the page, unlock it
+ * to avoid convoluting the logic below even more.
+ */
unlock_page(page);
- else {
- int error = mapping->a_ops->readpage(in, page);
+ }
+
+ pages[spd.nr_pages++] = page;
+ index++;
+ }
+
+ /*
+ * Now loop over the map and see if we need to start IO on any
+ * pages, fill in the partial map, etc.
+ */
+ index = *ppos >> PAGE_CACHE_SHIFT;
+ nr_pages = spd.nr_pages;
+ spd.nr_pages = 0;
+ for (page_nr = 0; page_nr < nr_pages; page_nr++) {
+ unsigned int this_len;
+
+ if (!len)
+ break;
+
+ /*
+ * this_len is the max we'll use from this page
+ */
+ this_len = min_t(unsigned long, len, PAGE_CACHE_SIZE - loff);
+ page = pages[page_nr];
+
+ /*
+ * If the page isn't uptodate, we may need to start io on it
+ */
+ if (!PageUptodate(page)) {
+ /*
+ * If in nonblock mode then dont block on waiting
+ * for an in-flight io page
+ */
+ if (flags & SPLICE_F_NONBLOCK)
+ break;
+
+ lock_page(page);
+ /*
+ * page was truncated, stop here. if this isn't the
+ * first page, we'll just complete what we already
+ * added
+ */
+ if (!page->mapping) {
+ unlock_page(page);
+ break;
+ }
+ /*
+ * page was already under io and is now done, great
+ */
+ if (PageUptodate(page)) {
+ unlock_page(page);
+ goto fill_it;
+ }
+
+ /*
+ * need to read in the page
+ */
+ error = mapping->a_ops->readpage(in, page);
if (unlikely(error)) {
- page_cache_release(page);
+ /*
+ * We really should re-lookup the page here,
+ * but it complicates things a lot. Instead
+ * lets just do what we already stored, and
+ * we'll get it the next time we are called.
+ */
+ if (error == AOP_TRUNCATED_PAGE)
+ error = 0;
+
break;
}
+
+ /*
+ * i_size must be checked after ->readpage().
+ */
+ isize = i_size_read(mapping->host);
+ end_index = (isize - 1) >> PAGE_CACHE_SHIFT;
+ if (unlikely(!isize || index > end_index))
+ break;
+
+ /*
+ * if this is the last page, see if we need to shrink
+ * the length and stop
+ */
+ if (end_index == index) {
+ loff = PAGE_CACHE_SIZE - (isize & ~PAGE_CACHE_MASK);
+ if (total_len + loff > isize)
+ break;
+ /*
+ * force quit after adding this page
+ */
+ len = this_len;
+ this_len = min(this_len, loff);
+ loff = 0;
+ }
}
- pages[i] = page;
+fill_it:
+ partial[page_nr].offset = loff;
+ partial[page_nr].len = this_len;
+ len -= this_len;
+ total_len += this_len;
+ loff = 0;
+ spd.nr_pages++;
+ index++;
}
- if (i)
- return move_to_pipe(pipe, pages, i, offset, len, flags);
+ /*
+ * Release any pages at the end, if we quit early. 'i' is how far
+ * we got, 'nr_pages' is how many pages are in the map.
+ */
+ while (page_nr < nr_pages)
+ page_cache_release(pages[page_nr++]);
- return 0;
+ if (spd.nr_pages)
+ return splice_to_pipe(pipe, &spd);
+
+ return error;
}
/**
* @flags: splice modifier flags
*
* Will read pages from given file and fill them into a pipe.
- *
*/
-ssize_t generic_file_splice_read(struct file *in, struct pipe_inode_info *pipe,
- size_t len, unsigned int flags)
+ssize_t generic_file_splice_read(struct file *in, loff_t *ppos,
+ struct pipe_inode_info *pipe, size_t len,
+ unsigned int flags)
{
ssize_t spliced;
int ret;
spliced = 0;
while (len) {
- ret = __generic_file_splice_read(in, pipe, len, flags);
+ ret = __generic_file_splice_read(in, ppos, pipe, len, flags);
- if (ret <= 0)
+ if (ret < 0)
break;
+ else if (!ret) {
+ if (spliced)
+ break;
+ if (flags & SPLICE_F_NONBLOCK) {
+ ret = -EAGAIN;
+ break;
+ }
+ }
- in->f_pos += ret;
+ *ppos += ret;
len -= ret;
spliced += ret;
-
- if (!(flags & SPLICE_F_NONBLOCK))
- continue;
- ret = -EAGAIN;
- break;
}
if (spliced)
/*
* Send 'sd->len' bytes to socket from 'sd->file' at position 'sd->pos'
- * using sendpage().
+ * using sendpage(). Return the number of bytes sent.
*/
-static int pipe_to_sendpage(struct pipe_inode_info *info,
+static int pipe_to_sendpage(struct pipe_inode_info *pipe,
struct pipe_buffer *buf, struct splice_desc *sd)
{
struct file *file = sd->file;
loff_t pos = sd->pos;
- unsigned int offset;
- ssize_t ret;
- void *ptr;
- int more;
+ int ret, more;
- /*
- * sub-optimal, but we are limited by the pipe ->map. we don't
- * need a kmap'ed buffer here, we just want to make sure we
- * have the page pinned if the pipe page originates from the
- * page cache
- */
- ptr = buf->ops->map(file, info, buf);
- if (IS_ERR(ptr))
- return PTR_ERR(ptr);
-
- offset = pos & ~PAGE_CACHE_MASK;
- more = (sd->flags & SPLICE_F_MORE) || sd->len < sd->total_len;
+ ret = buf->ops->pin(pipe, buf);
+ if (!ret) {
+ more = (sd->flags & SPLICE_F_MORE) || sd->len < sd->total_len;
- ret = file->f_op->sendpage(file, buf->page, offset, sd->len, &pos,more);
-
- buf->ops->unmap(info, buf);
- if (ret == sd->len)
- return 0;
+ ret = file->f_op->sendpage(file, buf->page, buf->offset,
+ sd->len, &pos, more);
+ }
- return -EIO;
+ return ret;
}
/*
* SPLICE_F_MOVE isn't set, or we cannot move the page, we simply create
* a new page in the output file page cache and fill/dirty that.
*/
-static int pipe_to_file(struct pipe_inode_info *info, struct pipe_buffer *buf,
+static int pipe_to_file(struct pipe_inode_info *pipe, struct pipe_buffer *buf,
struct splice_desc *sd)
{
struct file *file = sd->file;
struct address_space *mapping = file->f_mapping;
gfp_t gfp_mask = mapping_gfp_mask(mapping);
- unsigned int offset;
+ unsigned int offset, this_len;
struct page *page;
pgoff_t index;
- char *src;
int ret;
/*
* make sure the data in this buffer is uptodate
*/
- src = buf->ops->map(file, info, buf);
- if (IS_ERR(src))
- return PTR_ERR(src);
+ ret = buf->ops->pin(pipe, buf);
+ if (unlikely(ret))
+ return ret;
index = sd->pos >> PAGE_CACHE_SHIFT;
offset = sd->pos & ~PAGE_CACHE_MASK;
+ this_len = sd->len;
+ if (this_len + offset > PAGE_CACHE_SIZE)
+ this_len = PAGE_CACHE_SIZE - offset;
+
/*
- * reuse buf page, if SPLICE_F_MOVE is set
+ * Reuse buf page, if SPLICE_F_MOVE is set and we are doing a full
+ * page.
*/
- if (sd->flags & SPLICE_F_MOVE) {
+ if ((sd->flags & SPLICE_F_MOVE) && this_len == PAGE_CACHE_SIZE) {
/*
- * If steal succeeds, buf->page is now pruned from the vm
- * side (LRU and page cache) and we can reuse it.
+ * If steal succeeds, buf->page is now pruned from the
+ * pagecache and we can reuse it. The page will also be
+ * locked on successful return.
*/
- if (buf->ops->steal(info, buf))
+ if (buf->ops->steal(pipe, buf))
goto find_page;
- /*
- * this will also set the page locked
- */
page = buf->page;
- if (add_to_page_cache(page, mapping, index, gfp_mask))
+ if (add_to_page_cache(page, mapping, index, gfp_mask)) {
+ unlock_page(page);
goto find_page;
+ }
+
+ page_cache_get(page);
if (!(buf->flags & PIPE_BUF_FLAG_LRU))
lru_cache_add(page);
} else {
find_page:
- ret = -ENOMEM;
- page = find_or_create_page(mapping, index, gfp_mask);
- if (!page)
- goto out_nomem;
+ page = find_lock_page(mapping, index);
+ if (!page) {
+ ret = -ENOMEM;
+ page = page_cache_alloc_cold(mapping);
+ if (unlikely(!page))
+ goto out_nomem;
+
+ /*
+ * This will also lock the page
+ */
+ ret = add_to_page_cache_lru(page, mapping, index,
+ gfp_mask);
+ if (unlikely(ret))
+ goto out;
+ }
/*
- * If the page is uptodate, it is also locked. If it isn't
- * uptodate, we can mark it uptodate if we are filling the
- * full page. Otherwise we need to read it in first...
+ * We get here with the page locked. If the page is also
+ * uptodate, we don't need to do more. If it isn't, we
+ * may need to bring it in if we are not going to overwrite
+ * the full page.
*/
if (!PageUptodate(page)) {
- if (sd->len < PAGE_CACHE_SIZE) {
+ if (this_len < PAGE_CACHE_SIZE) {
ret = mapping->a_ops->readpage(file, page);
if (unlikely(ret))
goto out;
if (!PageUptodate(page)) {
/*
- * page got invalidated, repeat
+ * Page got invalidated, repeat.
*/
if (!page->mapping) {
unlock_page(page);
ret = -EIO;
goto out;
}
- } else {
- WARN_ON(!PageLocked(page));
+ } else
SetPageUptodate(page);
- }
}
}
- ret = mapping->a_ops->prepare_write(file, page, 0, sd->len);
- if (ret == AOP_TRUNCATED_PAGE) {
+ ret = mapping->a_ops->prepare_write(file, page, offset, offset+this_len);
+ if (unlikely(ret)) {
+ loff_t isize = i_size_read(mapping->host);
+
+ if (ret != AOP_TRUNCATED_PAGE)
+ unlock_page(page);
page_cache_release(page);
- goto find_page;
- } else if (ret)
+ if (ret == AOP_TRUNCATED_PAGE)
+ goto find_page;
+
+ /*
+ * prepare_write() may have instantiated a few blocks
+ * outside i_size. Trim these off again.
+ */
+ if (sd->pos + this_len > isize)
+ vmtruncate(mapping->host, isize);
+
goto out;
+ }
- if (!(buf->flags & PIPE_BUF_FLAG_STOLEN)) {
- char *dst = kmap_atomic(page, KM_USER0);
+ if (buf->page != page) {
+ /*
+ * Careful, ->map() uses KM_USER0!
+ */
+ char *src = buf->ops->map(pipe, buf, 1);
+ char *dst = kmap_atomic(page, KM_USER1);
- memcpy(dst + offset, src + buf->offset, sd->len);
+ memcpy(dst + offset, src + buf->offset, this_len);
flush_dcache_page(page);
- kunmap_atomic(dst, KM_USER0);
+ kunmap_atomic(dst, KM_USER1);
+ buf->ops->unmap(pipe, buf, src);
}
- ret = mapping->a_ops->commit_write(file, page, 0, sd->len);
- if (ret == AOP_TRUNCATED_PAGE) {
+ ret = mapping->a_ops->commit_write(file, page, offset, offset+this_len);
+ if (!ret) {
+ /*
+ * Return the number of bytes written and mark page as
+ * accessed, we are now done!
+ */
+ ret = this_len;
+ mark_page_accessed(page);
+ balance_dirty_pages_ratelimited(mapping);
+ } else if (ret == AOP_TRUNCATED_PAGE) {
page_cache_release(page);
goto find_page;
- } else if (ret)
- goto out;
-
- mark_page_accessed(page);
- balance_dirty_pages_ratelimited(mapping);
-out:
- if (!(buf->flags & PIPE_BUF_FLAG_STOLEN)) {
- page_cache_release(page);
- unlock_page(page);
}
+out:
+ page_cache_release(page);
+ unlock_page(page);
out_nomem:
- buf->ops->unmap(info, buf);
return ret;
}
-typedef int (splice_actor)(struct pipe_inode_info *, struct pipe_buffer *,
- struct splice_desc *);
-
/*
* Pipe input worker. Most of this logic works like a regular pipe, the
* key here is the 'actor' worker passed in that actually moves the data
* to the wanted destination. See pipe_to_file/pipe_to_sendpage above.
*/
-static ssize_t move_from_pipe(struct pipe_inode_info *pipe, struct file *out,
- size_t len, unsigned int flags,
- splice_actor *actor)
+ssize_t splice_from_pipe(struct pipe_inode_info *pipe, struct file *out,
+ loff_t *ppos, size_t len, unsigned int flags,
+ splice_actor *actor)
{
int ret, do_wakeup, err;
struct splice_desc sd;
sd.total_len = len;
sd.flags = flags;
sd.file = out;
- sd.pos = out->f_pos;
+ sd.pos = *ppos;
if (pipe->inode)
mutex_lock(&pipe->inode->i_mutex);
for (;;) {
- int bufs = pipe->nrbufs;
-
- if (bufs) {
- int curbuf = pipe->curbuf;
- struct pipe_buffer *buf = pipe->bufs + curbuf;
+ if (pipe->nrbufs) {
+ struct pipe_buffer *buf = pipe->bufs + pipe->curbuf;
struct pipe_buf_operations *ops = buf->ops;
sd.len = buf->len;
sd.len = sd.total_len;
err = actor(pipe, buf, &sd);
- if (err) {
+ if (err <= 0) {
if (!ret && err != -ENODATA)
ret = err;
break;
}
- ret += sd.len;
- buf->offset += sd.len;
- buf->len -= sd.len;
+ ret += err;
+ buf->offset += err;
+ buf->len -= err;
+
+ sd.len -= err;
+ sd.pos += err;
+ sd.total_len -= err;
+ if (sd.len)
+ continue;
+
if (!buf->len) {
buf->ops = NULL;
ops->release(pipe, buf);
- curbuf = (curbuf + 1) & (PIPE_BUFFERS - 1);
- pipe->curbuf = curbuf;
- pipe->nrbufs = --bufs;
- do_wakeup = 1;
+ pipe->curbuf = (pipe->curbuf + 1) & (PIPE_BUFFERS - 1);
+ pipe->nrbufs--;
+ if (pipe->inode)
+ do_wakeup = 1;
}
- sd.pos += sd.len;
- sd.total_len -= sd.len;
if (!sd.total_len)
break;
}
- if (bufs)
+ if (pipe->nrbufs)
continue;
if (!pipe->writers)
break;
kill_fasync(&pipe->fasync_writers, SIGIO, POLL_OUT);
}
- mutex_lock(&out->f_mapping->host->i_mutex);
- out->f_pos = sd.pos;
- mutex_unlock(&out->f_mapping->host->i_mutex);
return ret;
-
}
/**
*/
ssize_t
generic_file_splice_write(struct pipe_inode_info *pipe, struct file *out,
- size_t len, unsigned int flags)
+ loff_t *ppos, size_t len, unsigned int flags)
{
struct address_space *mapping = out->f_mapping;
ssize_t ret;
- ret = move_from_pipe(pipe, out, len, flags, pipe_to_file);
-
- /*
- * if file or inode is SYNC and we actually wrote some data, sync it
- */
- if (unlikely((out->f_flags & O_SYNC) || IS_SYNC(mapping->host))
- && ret > 0) {
+ ret = splice_from_pipe(pipe, out, ppos, len, flags, pipe_to_file);
+ if (ret > 0) {
struct inode *inode = mapping->host;
- int err;
- mutex_lock(&inode->i_mutex);
- err = generic_osync_inode(mapping->host, mapping,
- OSYNC_METADATA|OSYNC_DATA);
- mutex_unlock(&inode->i_mutex);
+ *ppos += ret;
+
+ /*
+ * If file or inode is SYNC and we actually wrote some data,
+ * sync it.
+ */
+ if (unlikely((out->f_flags & O_SYNC) || IS_SYNC(inode))) {
+ int err;
+
+ mutex_lock(&inode->i_mutex);
+ err = generic_osync_inode(inode, mapping,
+ OSYNC_METADATA|OSYNC_DATA);
+ mutex_unlock(&inode->i_mutex);
- if (err)
- ret = err;
+ if (err)
+ ret = err;
+ }
}
return ret;
*
*/
ssize_t generic_splice_sendpage(struct pipe_inode_info *pipe, struct file *out,
- size_t len, unsigned int flags)
+ loff_t *ppos, size_t len, unsigned int flags)
{
- return move_from_pipe(pipe, out, len, flags, pipe_to_sendpage);
+ return splice_from_pipe(pipe, out, ppos, len, flags, pipe_to_sendpage);
}
EXPORT_SYMBOL(generic_splice_sendpage);
* Attempt to initiate a splice from pipe to file.
*/
static long do_splice_from(struct pipe_inode_info *pipe, struct file *out,
- size_t len, unsigned int flags)
+ loff_t *ppos, size_t len, unsigned int flags)
{
- loff_t pos;
int ret;
- if (!out->f_op || !out->f_op->splice_write)
+ if (unlikely(!out->f_op || !out->f_op->splice_write))
return -EINVAL;
- if (!(out->f_mode & FMODE_WRITE))
+ if (unlikely(!(out->f_mode & FMODE_WRITE)))
return -EBADF;
- pos = out->f_pos;
-
- ret = rw_verify_area(WRITE, out, &pos, len);
+ ret = rw_verify_area(WRITE, out, ppos, len);
if (unlikely(ret < 0))
return ret;
- return out->f_op->splice_write(pipe, out, len, flags);
+ return out->f_op->splice_write(pipe, out, ppos, len, flags);
}
/*
* Attempt to initiate a splice from a file to a pipe.
*/
-static long do_splice_to(struct file *in, struct pipe_inode_info *pipe,
- size_t len, unsigned int flags)
+static long do_splice_to(struct file *in, loff_t *ppos,
+ struct pipe_inode_info *pipe, size_t len,
+ unsigned int flags)
{
- loff_t pos, isize, left;
+ loff_t isize, left;
int ret;
- if (!in->f_op || !in->f_op->splice_read)
+ if (unlikely(!in->f_op || !in->f_op->splice_read))
return -EINVAL;
- if (!(in->f_mode & FMODE_READ))
+ if (unlikely(!(in->f_mode & FMODE_READ)))
return -EBADF;
- pos = in->f_pos;
-
- ret = rw_verify_area(READ, in, &pos, len);
+ ret = rw_verify_area(READ, in, ppos, len);
if (unlikely(ret < 0))
return ret;
isize = i_size_read(in->f_mapping->host);
- if (unlikely(in->f_pos >= isize))
+ if (unlikely(*ppos >= isize))
return 0;
- left = isize - in->f_pos;
- if (left < len)
+ left = isize - *ppos;
+ if (unlikely(left < len))
len = left;
- return in->f_op->splice_read(in, pipe, len, flags);
+ return in->f_op->splice_read(in, ppos, pipe, len, flags);
}
-long do_splice_direct(struct file *in, struct file *out, size_t len,
- unsigned int flags)
+long do_splice_direct(struct file *in, loff_t *ppos, struct file *out,
+ size_t len, unsigned int flags)
{
struct pipe_inode_info *pipe;
long ret, bytes;
+ loff_t out_off;
umode_t i_mode;
int i;
* 'out' and transfer the wanted data from 'in' to 'out' through that
*/
pipe = current->splice_pipe;
- if (!pipe) {
+ if (unlikely(!pipe)) {
pipe = alloc_pipe_info(NULL);
if (!pipe)
return -ENOMEM;
/*
* We don't have an immediate reader, but we'll read the stuff
- * out of the pipe right after the move_to_pipe(). So set
+ * out of the pipe right after the splice_to_pipe(). So set
* PIPE_READERS appropriately.
*/
pipe->readers = 1;
}
/*
- * do the splice
+ * Do the splice.
*/
ret = 0;
bytes = 0;
+ out_off = 0;
while (len) {
size_t read_len, max_read_len;
*/
max_read_len = min(len, (size_t)(PIPE_BUFFERS*PAGE_SIZE));
- ret = do_splice_to(in, pipe, max_read_len, flags);
+ ret = do_splice_to(in, ppos, pipe, max_read_len, flags);
if (unlikely(ret < 0))
goto out_release;
* must not do the output in nonblocking mode as then we
* could get stuck data in the internal pipe:
*/
- ret = do_splice_from(pipe, out, read_len,
+ ret = do_splice_from(pipe, out, &out_off, read_len,
flags & ~SPLICE_F_NONBLOCK);
if (unlikely(ret < 0))
goto out_release;
size_t len, unsigned int flags)
{
struct pipe_inode_info *pipe;
+ loff_t offset, *off;
+ long ret;
pipe = in->f_dentry->d_inode->i_pipe;
if (pipe) {
if (off_out) {
if (out->f_op->llseek == no_llseek)
return -EINVAL;
- if (copy_from_user(&out->f_pos, off_out,
- sizeof(loff_t)))
+ if (copy_from_user(&offset, off_out, sizeof(loff_t)))
return -EFAULT;
- }
+ off = &offset;
+ } else
+ off = &out->f_pos;
+
+ ret = do_splice_from(pipe, out, off, len, flags);
- return do_splice_from(pipe, out, len, flags);
+ if (off_out && copy_to_user(off_out, off, sizeof(loff_t)))
+ ret = -EFAULT;
+
+ return ret;
}
pipe = out->f_dentry->d_inode->i_pipe;
if (off_in) {
if (in->f_op->llseek == no_llseek)
return -EINVAL;
- if (copy_from_user(&in->f_pos, off_in, sizeof(loff_t)))
+ if (copy_from_user(&offset, off_in, sizeof(loff_t)))
return -EFAULT;
- }
+ off = &offset;
+ } else
+ off = &in->f_pos;
+
+ ret = do_splice_to(in, off, pipe, len, flags);
+
+ if (off_in && copy_to_user(off_in, off, sizeof(loff_t)))
+ ret = -EFAULT;
- return do_splice_to(in, pipe, len, flags);
+ return ret;
}
return -EINVAL;
}
+/*
+ * Map an iov into an array of pages and offset/length tupples. With the
+ * partial_page structure, we can map several non-contiguous ranges into
+ * our ones pages[] map instead of splitting that operation into pieces.
+ * Could easily be exported as a generic helper for other users, in which
+ * case one would probably want to add a 'max_nr_pages' parameter as well.
+ */
+static int get_iovec_page_array(const struct iovec __user *iov,
+ unsigned int nr_vecs, struct page **pages,
+ struct partial_page *partial, int aligned)
+{
+ int buffers = 0, error = 0;
+
+ /*
+ * It's ok to take the mmap_sem for reading, even
+ * across a "get_user()".
+ */
+ down_read(¤t->mm->mmap_sem);
+
+ while (nr_vecs) {
+ unsigned long off, npages;
+ void __user *base;
+ size_t len;
+ int i;
+
+ /*
+ * Get user address base and length for this iovec.
+ */
+ error = get_user(base, &iov->iov_base);
+ if (unlikely(error))
+ break;
+ error = get_user(len, &iov->iov_len);
+ if (unlikely(error))
+ break;
+
+ /*
+ * Sanity check this iovec. 0 read succeeds.
+ */
+ if (unlikely(!len))
+ break;
+ error = -EFAULT;
+ if (unlikely(!base))
+ break;
+
+ /*
+ * Get this base offset and number of pages, then map
+ * in the user pages.
+ */
+ off = (unsigned long) base & ~PAGE_MASK;
+
+ /*
+ * If asked for alignment, the offset must be zero and the
+ * length a multiple of the PAGE_SIZE.
+ */
+ error = -EINVAL;
+ if (aligned && (off || len & ~PAGE_MASK))
+ break;
+
+ npages = (off + len + PAGE_SIZE - 1) >> PAGE_SHIFT;
+ if (npages > PIPE_BUFFERS - buffers)
+ npages = PIPE_BUFFERS - buffers;
+
+ error = get_user_pages(current, current->mm,
+ (unsigned long) base, npages, 0, 0,
+ &pages[buffers], NULL);
+
+ if (unlikely(error <= 0))
+ break;
+
+ /*
+ * Fill this contiguous range into the partial page map.
+ */
+ for (i = 0; i < error; i++) {
+ const int plen = min_t(size_t, len, PAGE_SIZE - off);
+
+ partial[buffers].offset = off;
+ partial[buffers].len = plen;
+
+ off = 0;
+ len -= plen;
+ buffers++;
+ }
+
+ /*
+ * We didn't complete this iov, stop here since it probably
+ * means we have to move some of this into a pipe to
+ * be able to continue.
+ */
+ if (len)
+ break;
+
+ /*
+ * Don't continue if we mapped fewer pages than we asked for,
+ * or if we mapped the max number of pages that we have
+ * room for.
+ */
+ if (error < npages || buffers == PIPE_BUFFERS)
+ break;
+
+ nr_vecs--;
+ iov++;
+ }
+
+ up_read(¤t->mm->mmap_sem);
+
+ if (buffers)
+ return buffers;
+
+ return error;
+}
+
+/*
+ * vmsplice splices a user address range into a pipe. It can be thought of
+ * as splice-from-memory, where the regular splice is splice-from-file (or
+ * to file). In both cases the output is a pipe, naturally.
+ *
+ * Note that vmsplice only supports splicing _from_ user memory to a pipe,
+ * not the other way around. Splicing from user memory is a simple operation
+ * that can be supported without any funky alignment restrictions or nasty
+ * vm tricks. We simply map in the user memory and fill them into a pipe.
+ * The reverse isn't quite as easy, though. There are two possible solutions
+ * for that:
+ *
+ * - memcpy() the data internally, at which point we might as well just
+ * do a regular read() on the buffer anyway.
+ * - Lots of nasty vm tricks, that are neither fast nor flexible (it
+ * has restriction limitations on both ends of the pipe).
+ *
+ * Alas, it isn't here.
+ *
+ */
+static long do_vmsplice(struct file *file, const struct iovec __user *iov,
+ unsigned long nr_segs, unsigned int flags)
+{
+ struct pipe_inode_info *pipe = file->f_dentry->d_inode->i_pipe;
+ struct page *pages[PIPE_BUFFERS];
+ struct partial_page partial[PIPE_BUFFERS];
+ struct splice_pipe_desc spd = {
+ .pages = pages,
+ .partial = partial,
+ .flags = flags,
+ .ops = &user_page_pipe_buf_ops,
+ };
+
+ if (unlikely(!pipe))
+ return -EBADF;
+ if (unlikely(nr_segs > UIO_MAXIOV))
+ return -EINVAL;
+ else if (unlikely(!nr_segs))
+ return 0;
+
+ spd.nr_pages = get_iovec_page_array(iov, nr_segs, pages, partial,
+ flags & SPLICE_F_GIFT);
+ if (spd.nr_pages <= 0)
+ return spd.nr_pages;
+
+ return splice_to_pipe(pipe, &spd);
+}
+
+asmlinkage long sys_vmsplice(int fd, const struct iovec __user *iov,
+ unsigned long nr_segs, unsigned int flags)
+{
+ struct file *file;
+ long error;
+ int fput;
+
+ error = -EBADF;
+ file = fget_light(fd, &fput);
+ if (file) {
+ if (file->f_mode & FMODE_WRITE)
+ error = do_vmsplice(file, iov, nr_segs, flags);
+
+ fput_light(file, fput);
+ }
+
+ return error;
+}
+
asmlinkage long sys_splice(int fd_in, loff_t __user *off_in,
int fd_out, loff_t __user *off_out,
size_t len, unsigned int flags)
return error;
}
+
+/*
+ * Link contents of ipipe to opipe.
+ */
+static int link_pipe(struct pipe_inode_info *ipipe,
+ struct pipe_inode_info *opipe,
+ size_t len, unsigned int flags)
+{
+ struct pipe_buffer *ibuf, *obuf;
+ int ret, do_wakeup, i, ipipe_first;
+
+ ret = do_wakeup = ipipe_first = 0;
+
+ /*
+ * Potential ABBA deadlock, work around it by ordering lock
+ * grabbing by inode address. Otherwise two different processes
+ * could deadlock (one doing tee from A -> B, the other from B -> A).
+ */
+ if (ipipe->inode < opipe->inode) {
+ ipipe_first = 1;
+ mutex_lock(&ipipe->inode->i_mutex);
+ mutex_lock(&opipe->inode->i_mutex);
+ } else {
+ mutex_lock(&opipe->inode->i_mutex);
+ mutex_lock(&ipipe->inode->i_mutex);
+ }
+
+ for (i = 0;; i++) {
+ if (!opipe->readers) {
+ send_sig(SIGPIPE, current, 0);
+ if (!ret)
+ ret = -EPIPE;
+ break;
+ }
+ if (ipipe->nrbufs - i) {
+ ibuf = ipipe->bufs + ((ipipe->curbuf + i) & (PIPE_BUFFERS - 1));
+
+ /*
+ * If we have room, fill this buffer
+ */
+ if (opipe->nrbufs < PIPE_BUFFERS) {
+ int nbuf = (opipe->curbuf + opipe->nrbufs) & (PIPE_BUFFERS - 1);
+
+ /*
+ * Get a reference to this pipe buffer,
+ * so we can copy the contents over.
+ */
+ ibuf->ops->get(ipipe, ibuf);
+
+ obuf = opipe->bufs + nbuf;
+ *obuf = *ibuf;
+
+ /*
+ * Don't inherit the gift flag, we need to
+ * prevent multiple steals of this page.
+ */
+ obuf->flags &= ~PIPE_BUF_FLAG_GIFT;
+
+ if (obuf->len > len)
+ obuf->len = len;
+
+ opipe->nrbufs++;
+ do_wakeup = 1;
+ ret += obuf->len;
+ len -= obuf->len;
+
+ if (!len)
+ break;
+ if (opipe->nrbufs < PIPE_BUFFERS)
+ continue;
+ }
+
+ /*
+ * We have input available, but no output room.
+ * If we already copied data, return that. If we
+ * need to drop the opipe lock, it must be ordered
+ * last to avoid deadlocks.
+ */
+ if ((flags & SPLICE_F_NONBLOCK) || !ipipe_first) {
+ if (!ret)
+ ret = -EAGAIN;
+ break;
+ }
+ if (signal_pending(current)) {
+ if (!ret)
+ ret = -ERESTARTSYS;
+ break;
+ }
+ if (do_wakeup) {
+ smp_mb();
+ if (waitqueue_active(&opipe->wait))
+ wake_up_interruptible(&opipe->wait);
+ kill_fasync(&opipe->fasync_readers, SIGIO, POLL_IN);
+ do_wakeup = 0;
+ }
+
+ opipe->waiting_writers++;
+ pipe_wait(opipe);
+ opipe->waiting_writers--;
+ continue;
+ }
+
+ /*
+ * No input buffers, do the usual checks for available
+ * writers and blocking and wait if necessary
+ */
+ if (!ipipe->writers)
+ break;
+ if (!ipipe->waiting_writers) {
+ if (ret)
+ break;
+ }
+ /*
+ * pipe_wait() drops the ipipe mutex. To avoid deadlocks
+ * with another process, we can only safely do that if
+ * the ipipe lock is ordered last.
+ */
+ if ((flags & SPLICE_F_NONBLOCK) || ipipe_first) {
+ if (!ret)
+ ret = -EAGAIN;
+ break;
+ }
+ if (signal_pending(current)) {
+ if (!ret)
+ ret = -ERESTARTSYS;
+ break;
+ }
+
+ if (waitqueue_active(&ipipe->wait))
+ wake_up_interruptible_sync(&ipipe->wait);
+ kill_fasync(&ipipe->fasync_writers, SIGIO, POLL_OUT);
+
+ pipe_wait(ipipe);
+ }
+
+ mutex_unlock(&ipipe->inode->i_mutex);
+ mutex_unlock(&opipe->inode->i_mutex);
+
+ if (do_wakeup) {
+ smp_mb();
+ if (waitqueue_active(&opipe->wait))
+ wake_up_interruptible(&opipe->wait);
+ kill_fasync(&opipe->fasync_readers, SIGIO, POLL_IN);
+ }
+
+ return ret;
+}
+
+/*
+ * This is a tee(1) implementation that works on pipes. It doesn't copy
+ * any data, it simply references the 'in' pages on the 'out' pipe.
+ * The 'flags' used are the SPLICE_F_* variants, currently the only
+ * applicable one is SPLICE_F_NONBLOCK.
+ */
+static long do_tee(struct file *in, struct file *out, size_t len,
+ unsigned int flags)
+{
+ struct pipe_inode_info *ipipe = in->f_dentry->d_inode->i_pipe;
+ struct pipe_inode_info *opipe = out->f_dentry->d_inode->i_pipe;
+
+ /*
+ * Link ipipe to the two output pipes, consuming as we go along.
+ */
+ if (ipipe && opipe)
+ return link_pipe(ipipe, opipe, len, flags);
+
+ return -EINVAL;
+}
+
+asmlinkage long sys_tee(int fdin, int fdout, size_t len, unsigned int flags)
+{
+ struct file *in;
+ int error, fput_in;
+
+ if (unlikely(!len))
+ return 0;
+
+ error = -EBADF;
+ in = fget_light(fdin, &fput_in);
+ if (in) {
+ if (in->f_mode & FMODE_READ) {
+ int fput_out;
+ struct file *out = fget_light(fdout, &fput_out);
+
+ if (out) {
+ if (out->f_mode & FMODE_WRITE)
+ error = do_tee(in, out, len, flags);
+ fput_light(out, fput_out);
+ }
+ }
+ fput_light(in, fput_in);
+ }
+
+ return error;
+}