diff options
Diffstat (limited to 'fs/splice.c')
| -rw-r--r-- | fs/splice.c | 442 |
1 files changed, 218 insertions, 224 deletions
diff --git a/fs/splice.c b/fs/splice.c index 6909d89d0da..f5cb9ba8451 100644 --- a/fs/splice.c +++ b/fs/splice.c @@ -31,6 +31,9 @@ #include <linux/security.h> #include <linux/gfp.h> #include <linux/socket.h> +#include <linux/compat.h> +#include <linux/aio.h> +#include "internal.h" /* * Attempt to steal a page from a pipe buffer. This should perhaps go into @@ -134,8 +137,6 @@ error: const struct pipe_buf_operations page_cache_pipe_buf_ops = { .can_merge = 0, - .map = generic_pipe_buf_map, - .unmap = generic_pipe_buf_unmap, .confirm = page_cache_pipe_buf_confirm, .release = page_cache_pipe_buf_release, .steal = page_cache_pipe_buf_steal, @@ -154,8 +155,6 @@ static int user_page_pipe_buf_steal(struct pipe_inode_info *pipe, static const struct pipe_buf_operations user_page_pipe_buf_ops = { .can_merge = 0, - .map = generic_pipe_buf_map, - .unmap = generic_pipe_buf_unmap, .confirm = generic_pipe_buf_confirm, .release = page_cache_pipe_buf_release, .steal = user_page_pipe_buf_steal, @@ -217,7 +216,7 @@ ssize_t splice_to_pipe(struct pipe_inode_info *pipe, page_nr++; ret += buf->len; - if (pipe->inode) + if (pipe->files) do_wakeup = 1; if (!--spd->nr_pages) @@ -545,14 +544,28 @@ EXPORT_SYMBOL(generic_file_splice_read); static const struct pipe_buf_operations default_pipe_buf_ops = { .can_merge = 0, - .map = generic_pipe_buf_map, - .unmap = generic_pipe_buf_unmap, .confirm = generic_pipe_buf_confirm, .release = generic_pipe_buf_release, .steal = generic_pipe_buf_steal, .get = generic_pipe_buf_get, }; +static int generic_pipe_buf_nosteal(struct pipe_inode_info *pipe, + struct pipe_buffer *buf) +{ + return 1; +} + +/* Pipe buffer operations for a socket and similar. */ +const struct pipe_buf_operations nosteal_pipe_buf_ops = { + .can_merge = 0, + .confirm = generic_pipe_buf_confirm, + .release = generic_pipe_buf_release, + .steal = generic_pipe_buf_nosteal, + .get = generic_pipe_buf_get, +}; +EXPORT_SYMBOL(nosteal_pipe_buf_ops); + static ssize_t kernel_readv(struct file *file, const struct iovec *vec, unsigned long vlen, loff_t offset) { @@ -569,7 +582,7 @@ static ssize_t kernel_readv(struct file *file, const struct iovec *vec, return res; } -static ssize_t kernel_write(struct file *file, const char *buf, size_t count, +ssize_t kernel_write(struct file *file, const char *buf, size_t count, loff_t pos) { mm_segment_t old_fs; @@ -578,11 +591,12 @@ static ssize_t kernel_write(struct file *file, const char *buf, size_t count, old_fs = get_fs(); set_fs(get_ds()); /* The cast to a user pointer is valid due to the set_fs() */ - res = vfs_write(file, (const char __user *)buf, count, &pos); + res = vfs_write(file, (__force const char __user *)buf, count, &pos); set_fs(old_fs); return res; } +EXPORT_SYMBOL(kernel_write); ssize_t default_file_splice_read(struct file *in, loff_t *ppos, struct pipe_inode_info *pipe, size_t len, @@ -692,7 +706,7 @@ static int pipe_to_sendpage(struct pipe_inode_info *pipe, loff_t pos = sd->pos; int more; - if (!likely(file->f_op && file->f_op->sendpage)) + if (!likely(file->f_op->sendpage)) return -EINVAL; more = (sd->flags & SPLICE_F_MORE) ? MSG_MORE : 0; @@ -704,63 +718,6 @@ static int pipe_to_sendpage(struct pipe_inode_info *pipe, sd->len, &pos, more); } -/* - * This is a little more tricky than the file -> pipe splicing. There are - * basically three cases: - * - * - Destination page already exists in the address space and there - * are users of it. For that case we have no other option that - * copying the data. Tough luck. - * - Destination page already exists in the address space, but there - * are no users of it. Make sure it's uptodate, then drop it. Fall - * through to last case. - * - Destination page does not exist, we can add the pipe page to - * the page cache and avoid the copy. - * - * If asked to move pages to the output file (SPLICE_F_MOVE is set in - * sd->flags), we attempt to migrate pages from the pipe to the output - * file address space page cache. This is possible if no one else has - * the pipe page referenced outside of the pipe and page cache. If - * 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. - */ -int pipe_to_file(struct pipe_inode_info *pipe, struct pipe_buffer *buf, - struct splice_desc *sd) -{ - struct file *file = sd->u.file; - struct address_space *mapping = file->f_mapping; - unsigned int offset, this_len; - struct page *page; - void *fsdata; - int ret; - - offset = sd->pos & ~PAGE_CACHE_MASK; - - this_len = sd->len; - if (this_len + offset > PAGE_CACHE_SIZE) - this_len = PAGE_CACHE_SIZE - offset; - - ret = pagecache_write_begin(file, mapping, sd->pos, this_len, - AOP_FLAG_UNINTERRUPTIBLE, &page, &fsdata); - if (unlikely(ret)) - goto out; - - if (buf->page != page) { - char *src = buf->ops->map(pipe, buf, 1); - char *dst = kmap_atomic(page); - - memcpy(dst + offset, src + buf->offset, this_len); - flush_dcache_page(page); - kunmap_atomic(dst); - buf->ops->unmap(pipe, buf, src); - } - ret = pagecache_write_end(file, mapping, sd->pos, this_len, this_len, - page, fsdata); -out: - return ret; -} -EXPORT_SYMBOL(pipe_to_file); - static void wakeup_pipe_writers(struct pipe_inode_info *pipe) { smp_mb(); @@ -789,7 +746,7 @@ static void wakeup_pipe_writers(struct pipe_inode_info *pipe) * locking is required around copying the pipe buffers to the * destination. */ -int splice_from_pipe_feed(struct pipe_inode_info *pipe, struct splice_desc *sd, +static int splice_from_pipe_feed(struct pipe_inode_info *pipe, struct splice_desc *sd, splice_actor *actor) { int ret; @@ -826,7 +783,7 @@ int splice_from_pipe_feed(struct pipe_inode_info *pipe, struct splice_desc *sd, ops->release(pipe, buf); pipe->curbuf = (pipe->curbuf + 1) & (pipe->buffers - 1); pipe->nrbufs--; - if (pipe->inode) + if (pipe->files) sd->need_wakeup = true; } @@ -836,7 +793,6 @@ int splice_from_pipe_feed(struct pipe_inode_info *pipe, struct splice_desc *sd, return 1; } -EXPORT_SYMBOL(splice_from_pipe_feed); /** * splice_from_pipe_next - wait for some data to splice from @@ -848,7 +804,7 @@ EXPORT_SYMBOL(splice_from_pipe_feed); * value (one) if pipe buffers are available. It will return zero * or -errno if no more data needs to be spliced. */ -int splice_from_pipe_next(struct pipe_inode_info *pipe, struct splice_desc *sd) +static int splice_from_pipe_next(struct pipe_inode_info *pipe, struct splice_desc *sd) { while (!pipe->nrbufs) { if (!pipe->writers) @@ -873,7 +829,6 @@ int splice_from_pipe_next(struct pipe_inode_info *pipe, struct splice_desc *sd) return 1; } -EXPORT_SYMBOL(splice_from_pipe_next); /** * splice_from_pipe_begin - start splicing from pipe @@ -884,12 +839,11 @@ EXPORT_SYMBOL(splice_from_pipe_next); * splice_from_pipe_next() and splice_from_pipe_feed() to * initialize the necessary fields of @sd. */ -void splice_from_pipe_begin(struct splice_desc *sd) +static void splice_from_pipe_begin(struct splice_desc *sd) { sd->num_spliced = 0; sd->need_wakeup = false; } -EXPORT_SYMBOL(splice_from_pipe_begin); /** * splice_from_pipe_end - finish splicing from pipe @@ -901,12 +855,11 @@ EXPORT_SYMBOL(splice_from_pipe_begin); * be called after a loop containing splice_from_pipe_next() and * splice_from_pipe_feed(). */ -void splice_from_pipe_end(struct pipe_inode_info *pipe, struct splice_desc *sd) +static void splice_from_pipe_end(struct pipe_inode_info *pipe, struct splice_desc *sd) { if (sd->need_wakeup) wakeup_pipe_writers(pipe); } -EXPORT_SYMBOL(splice_from_pipe_end); /** * __splice_from_pipe - splice data from a pipe to given actor @@ -972,7 +925,7 @@ ssize_t splice_from_pipe(struct pipe_inode_info *pipe, struct file *out, } /** - * generic_file_splice_write - splice data from a pipe to a file + * iter_file_splice_write - splice data from a pipe to a file * @pipe: pipe info * @out: file to write to * @ppos: position in @out @@ -982,42 +935,122 @@ ssize_t splice_from_pipe(struct pipe_inode_info *pipe, struct file *out, * Description: * Will either move or copy pages (determined by @flags options) from * the given pipe inode to the given file. + * This one is ->write_iter-based. * */ ssize_t -generic_file_splice_write(struct pipe_inode_info *pipe, struct file *out, +iter_file_splice_write(struct pipe_inode_info *pipe, struct file *out, loff_t *ppos, size_t len, unsigned int flags) { - struct address_space *mapping = out->f_mapping; - struct inode *inode = mapping->host; struct splice_desc sd = { .total_len = len, .flags = flags, .pos = *ppos, .u.file = out, }; + int nbufs = pipe->buffers; + struct bio_vec *array = kcalloc(nbufs, sizeof(struct bio_vec), + GFP_KERNEL); ssize_t ret; - sb_start_write(inode->i_sb); + if (unlikely(!array)) + return -ENOMEM; pipe_lock(pipe); splice_from_pipe_begin(&sd); - do { + while (sd.total_len) { + struct iov_iter from; + struct kiocb kiocb; + size_t left; + int n, idx; + ret = splice_from_pipe_next(pipe, &sd); if (ret <= 0) break; - mutex_lock_nested(&inode->i_mutex, I_MUTEX_CHILD); - ret = file_remove_suid(out); - if (!ret) { - ret = file_update_time(out); - if (!ret) - ret = splice_from_pipe_feed(pipe, &sd, - pipe_to_file); + if (unlikely(nbufs < pipe->buffers)) { + kfree(array); + nbufs = pipe->buffers; + array = kcalloc(nbufs, sizeof(struct bio_vec), + GFP_KERNEL); + if (!array) { + ret = -ENOMEM; + break; + } } - mutex_unlock(&inode->i_mutex); - } while (ret > 0); + + /* build the vector */ + left = sd.total_len; + for (n = 0, idx = pipe->curbuf; left && n < pipe->nrbufs; n++, idx++) { + struct pipe_buffer *buf = pipe->bufs + idx; + size_t this_len = buf->len; + + if (this_len > left) + this_len = left; + + if (idx == pipe->buffers - 1) + idx = -1; + + ret = buf->ops->confirm(pipe, buf); + if (unlikely(ret)) { + if (ret == -ENODATA) + ret = 0; + goto done; + } + + array[n].bv_page = buf->page; + array[n].bv_len = this_len; + array[n].bv_offset = buf->offset; + left -= this_len; + } + + /* ... iov_iter */ + from.type = ITER_BVEC | WRITE; + from.bvec = array; + from.nr_segs = n; + from.count = sd.total_len - left; + from.iov_offset = 0; + + /* ... and iocb */ + init_sync_kiocb(&kiocb, out); + kiocb.ki_pos = sd.pos; + kiocb.ki_nbytes = sd.total_len - left; + + /* now, send it */ + ret = out->f_op->write_iter(&kiocb, &from); + if (-EIOCBQUEUED == ret) + ret = wait_on_sync_kiocb(&kiocb); + + if (ret <= 0) + break; + + sd.num_spliced += ret; + sd.total_len -= ret; + *ppos = sd.pos = kiocb.ki_pos; + + /* dismiss the fully eaten buffers, adjust the partial one */ + while (ret) { + struct pipe_buffer *buf = pipe->bufs + pipe->curbuf; + if (ret >= buf->len) { + const struct pipe_buf_operations *ops = buf->ops; + ret -= buf->len; + buf->len = 0; + buf->ops = NULL; + ops->release(pipe, buf); + pipe->curbuf = (pipe->curbuf + 1) & (pipe->buffers - 1); + pipe->nrbufs--; + if (pipe->files) + sd.need_wakeup = true; + } else { + buf->offset += ret; + buf->len -= ret; + ret = 0; + } + } + } +done: + kfree(array); splice_from_pipe_end(pipe, &sd); pipe_unlock(pipe); @@ -1025,32 +1058,21 @@ generic_file_splice_write(struct pipe_inode_info *pipe, struct file *out, if (sd.num_spliced) ret = sd.num_spliced; - if (ret > 0) { - int err; - - err = generic_write_sync(out, *ppos, ret); - if (err) - ret = err; - else - *ppos += ret; - balance_dirty_pages_ratelimited(mapping); - } - sb_end_write(inode->i_sb); - return ret; } -EXPORT_SYMBOL(generic_file_splice_write); +EXPORT_SYMBOL(iter_file_splice_write); static int write_pipe_buf(struct pipe_inode_info *pipe, struct pipe_buffer *buf, struct splice_desc *sd) { int ret; void *data; + loff_t tmp = sd->pos; - data = buf->ops->map(pipe, buf, 0); - ret = kernel_write(sd->u.file, data + buf->offset, sd->len, sd->pos); - buf->ops->unmap(pipe, buf, data); + data = kmap(buf->page); + ret = __kernel_write(sd->u.file, data + buf->offset, sd->len, &tmp); + kunmap(buf->page); return ret; } @@ -1097,19 +1119,8 @@ static long do_splice_from(struct pipe_inode_info *pipe, struct file *out, { ssize_t (*splice_write)(struct pipe_inode_info *, struct file *, loff_t *, size_t, unsigned int); - int ret; - - if (unlikely(!(out->f_mode & FMODE_WRITE))) - return -EBADF; - if (unlikely(out->f_flags & O_APPEND)) - return -EINVAL; - - ret = rw_verify_area(WRITE, out, ppos, len); - if (unlikely(ret < 0)) - return ret; - - if (out->f_op && out->f_op->splice_write) + if (out->f_op->splice_write) splice_write = out->f_op->splice_write; else splice_write = default_file_splice_write; @@ -1135,7 +1146,7 @@ static long do_splice_to(struct file *in, loff_t *ppos, if (unlikely(ret < 0)) return ret; - if (in->f_op && in->f_op->splice_read) + if (in->f_op->splice_read) splice_read = in->f_op->splice_read; else splice_read = default_file_splice_read; @@ -1170,7 +1181,7 @@ ssize_t splice_direct_to_actor(struct file *in, struct splice_desc *sd, * randomly drop data for eg socket -> socket splicing. Use the * piped splicing for that! */ - i_mode = in->f_path.dentry->d_inode->i_mode; + i_mode = file_inode(in)->i_mode; if (unlikely(!S_ISREG(i_mode) && !S_ISBLK(i_mode))) return -EINVAL; @@ -1180,7 +1191,7 @@ ssize_t splice_direct_to_actor(struct file *in, struct splice_desc *sd, */ pipe = current->splice_pipe; if (unlikely(!pipe)) { - pipe = alloc_pipe_info(NULL); + pipe = alloc_pipe_info(); if (!pipe) return -ENOMEM; @@ -1270,7 +1281,7 @@ static int direct_splice_actor(struct pipe_inode_info *pipe, { struct file *file = sd->u.file; - return do_splice_from(pipe, file, &file->f_pos, sd->total_len, + return do_splice_from(pipe, file, sd->opos, sd->total_len, sd->flags); } @@ -1279,6 +1290,7 @@ static int direct_splice_actor(struct pipe_inode_info *pipe, * @in: file to splice from * @ppos: input file offset * @out: file to splice to + * @opos: output file offset * @len: number of bytes to splice * @flags: splice modifier flags * @@ -1290,7 +1302,7 @@ static int direct_splice_actor(struct pipe_inode_info *pipe, * */ long do_splice_direct(struct file *in, loff_t *ppos, struct file *out, - size_t len, unsigned int flags) + loff_t *opos, size_t len, unsigned int flags) { struct splice_desc sd = { .len = len, @@ -1298,9 +1310,20 @@ long do_splice_direct(struct file *in, loff_t *ppos, struct file *out, .flags = flags, .pos = *ppos, .u.file = out, + .opos = opos, }; long ret; + if (unlikely(!(out->f_mode & FMODE_WRITE))) + return -EBADF; + + if (unlikely(out->f_flags & O_APPEND)) + return -EINVAL; + + ret = rw_verify_area(WRITE, out, opos, len); + if (unlikely(ret < 0)) + return ret; + ret = splice_direct_to_actor(in, &sd, direct_splice_actor); if (ret > 0) *ppos = sd.pos; @@ -1321,7 +1344,7 @@ static long do_splice(struct file *in, loff_t __user *off_in, { struct pipe_inode_info *ipipe; struct pipe_inode_info *opipe; - loff_t offset, *off; + loff_t offset; long ret; ipipe = get_pipe_info(in); @@ -1352,13 +1375,27 @@ static long do_splice(struct file *in, loff_t __user *off_in, return -EINVAL; if (copy_from_user(&offset, off_out, sizeof(loff_t))) return -EFAULT; - off = &offset; - } else - off = &out->f_pos; + } else { + offset = out->f_pos; + } - ret = do_splice_from(ipipe, out, off, len, flags); + if (unlikely(!(out->f_mode & FMODE_WRITE))) + return -EBADF; - if (off_out && copy_to_user(off_out, off, sizeof(loff_t))) + if (unlikely(out->f_flags & O_APPEND)) + return -EINVAL; + + ret = rw_verify_area(WRITE, out, &offset, len); + if (unlikely(ret < 0)) + return ret; + + file_start_write(out); + ret = do_splice_from(ipipe, out, &offset, len, flags); + file_end_write(out); + + if (!off_out) + out->f_pos = offset; + else if (copy_to_user(off_out, &offset, sizeof(loff_t))) ret = -EFAULT; return ret; @@ -1372,13 +1409,15 @@ static long do_splice(struct file *in, loff_t __user *off_in, return -EINVAL; if (copy_from_user(&offset, off_in, sizeof(loff_t))) return -EFAULT; - off = &offset; - } else - off = &in->f_pos; + } else { + offset = in->f_pos; + } - ret = do_splice_to(in, off, opipe, len, flags); + ret = do_splice_to(in, &offset, opipe, len, flags); - if (off_in && copy_to_user(off_in, off, sizeof(loff_t))) + if (!off_in) + in->f_pos = offset; + else if (copy_to_user(off_in, &offset, sizeof(loff_t))) ret = -EFAULT; return ret; @@ -1492,116 +1531,50 @@ static int get_iovec_page_array(const struct iovec __user *iov, static int pipe_to_user(struct pipe_inode_info *pipe, struct pipe_buffer *buf, struct splice_desc *sd) { - char *src; - int ret; - - /* - * See if we can use the atomic maps, by prefaulting in the - * pages and doing an atomic copy - */ - if (!fault_in_pages_writeable(sd->u.userptr, sd->len)) { - src = buf->ops->map(pipe, buf, 1); - ret = __copy_to_user_inatomic(sd->u.userptr, src + buf->offset, - sd->len); - buf->ops->unmap(pipe, buf, src); - if (!ret) { - ret = sd->len; - goto out; - } - } - - /* - * No dice, use slow non-atomic map and copy - */ - src = buf->ops->map(pipe, buf, 0); - - ret = sd->len; - if (copy_to_user(sd->u.userptr, src + buf->offset, sd->len)) - ret = -EFAULT; - - buf->ops->unmap(pipe, buf, src); -out: - if (ret > 0) - sd->u.userptr += ret; - return ret; + int n = copy_page_to_iter(buf->page, buf->offset, sd->len, sd->u.data); + return n == sd->len ? n : -EFAULT; } /* * For lack of a better implementation, implement vmsplice() to userspace * as a simple copy of the pipes pages to the user iov. */ -static long vmsplice_to_user(struct file *file, const struct iovec __user *iov, +static long vmsplice_to_user(struct file *file, const struct iovec __user *uiov, unsigned long nr_segs, unsigned int flags) { struct pipe_inode_info *pipe; struct splice_desc sd; - ssize_t size; - int error; long ret; + struct iovec iovstack[UIO_FASTIOV]; + struct iovec *iov = iovstack; + struct iov_iter iter; + ssize_t count; pipe = get_pipe_info(file); if (!pipe) return -EBADF; - pipe_lock(pipe); - - error = ret = 0; - while (nr_segs) { - void __user *base; - size_t len; - - /* - * 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; - if (unlikely(!base)) { - error = -EFAULT; - break; - } - - if (unlikely(!access_ok(VERIFY_WRITE, base, len))) { - error = -EFAULT; - break; - } - - sd.len = 0; - sd.total_len = len; - sd.flags = flags; - sd.u.userptr = base; - sd.pos = 0; - - size = __splice_from_pipe(pipe, &sd, pipe_to_user); - if (size < 0) { - if (!ret) - ret = size; - - break; - } + ret = rw_copy_check_uvector(READ, uiov, nr_segs, + ARRAY_SIZE(iovstack), iovstack, &iov); + if (ret <= 0) + goto out; - ret += size; + count = ret; + iov_iter_init(&iter, READ, iov, nr_segs, count); - if (size < len) - break; - - nr_segs--; - iov++; - } + sd.len = 0; + sd.total_len = count; + sd.flags = flags; + sd.u.data = &iter; + sd.pos = 0; + pipe_lock(pipe); + ret = __splice_from_pipe(pipe, &sd, pipe_to_user); pipe_unlock(pipe); - if (!ret) - ret = error; +out: + if (iov != iovstack) + kfree(iov); return ret; } @@ -1687,6 +1660,27 @@ SYSCALL_DEFINE4(vmsplice, int, fd, const struct iovec __user *, iov, return error; } +#ifdef CONFIG_COMPAT +COMPAT_SYSCALL_DEFINE4(vmsplice, int, fd, const struct compat_iovec __user *, iov32, + unsigned int, nr_segs, unsigned int, flags) +{ + unsigned i; + struct iovec __user *iov; + if (nr_segs > UIO_MAXIOV) + return -EINVAL; + iov = compat_alloc_user_space(nr_segs * sizeof(struct iovec)); + for (i = 0; i < nr_segs; i++) { + struct compat_iovec v; + if (get_user(v.iov_base, &iov32[i].iov_base) || + get_user(v.iov_len, &iov32[i].iov_len) || + put_user(compat_ptr(v.iov_base), &iov[i].iov_base) || + put_user(v.iov_len, &iov[i].iov_len)) + return -EFAULT; + } + return sys_vmsplice(fd, iov, nr_segs, flags); +} +#endif + SYSCALL_DEFINE6(splice, int, fd_in, loff_t __user *, off_in, int, fd_out, loff_t __user *, off_out, size_t, len, unsigned int, flags) |