aboutsummaryrefslogtreecommitdiff
path: root/fs/fuse/file.c
diff options
context:
space:
mode:
Diffstat (limited to 'fs/fuse/file.c')
-rw-r--r--fs/fuse/file.c506
1 files changed, 80 insertions, 426 deletions
diff --git a/fs/fuse/file.c b/fs/fuse/file.c
index 8cccecb55fb8..09ef2a4d25ed 100644
--- a/fs/fuse/file.c
+++ b/fs/fuse/file.c
@@ -14,32 +14,20 @@
#include <linux/sched.h>
#include <linux/sched/signal.h>
#include <linux/module.h>
-#include <linux/compat.h>
#include <linux/swap.h>
#include <linux/falloc.h>
#include <linux/uio.h>
#include <linux/fs.h>
-static struct page **fuse_pages_alloc(unsigned int npages, gfp_t flags,
- struct fuse_page_desc **desc)
-{
- struct page **pages;
-
- pages = kzalloc(npages * (sizeof(struct page *) +
- sizeof(struct fuse_page_desc)), flags);
- *desc = (void *) (pages + npages);
-
- return pages;
-}
-
-static int fuse_send_open(struct fuse_mount *fm, u64 nodeid, struct file *file,
- int opcode, struct fuse_open_out *outargp)
+static int fuse_send_open(struct fuse_mount *fm, u64 nodeid,
+ unsigned int open_flags, int opcode,
+ struct fuse_open_out *outargp)
{
struct fuse_open_in inarg;
FUSE_ARGS(args);
memset(&inarg, 0, sizeof(inarg));
- inarg.flags = file->f_flags & ~(O_CREAT | O_EXCL | O_NOCTTY);
+ inarg.flags = open_flags & ~(O_CREAT | O_EXCL | O_NOCTTY);
if (!fm->fc->atomic_o_trunc)
inarg.flags &= ~O_TRUNC;
@@ -136,8 +124,8 @@ static void fuse_file_put(struct fuse_file *ff, bool sync, bool isdir)
}
}
-int fuse_do_open(struct fuse_mount *fm, u64 nodeid, struct file *file,
- bool isdir)
+struct fuse_file *fuse_file_open(struct fuse_mount *fm, u64 nodeid,
+ unsigned int open_flags, bool isdir)
{
struct fuse_conn *fc = fm->fc;
struct fuse_file *ff;
@@ -145,7 +133,7 @@ int fuse_do_open(struct fuse_mount *fm, u64 nodeid, struct file *file,
ff = fuse_file_alloc(fm);
if (!ff)
- return -ENOMEM;
+ return ERR_PTR(-ENOMEM);
ff->fh = 0;
/* Default for no-open */
@@ -154,14 +142,14 @@ int fuse_do_open(struct fuse_mount *fm, u64 nodeid, struct file *file,
struct fuse_open_out outarg;
int err;
- err = fuse_send_open(fm, nodeid, file, opcode, &outarg);
+ err = fuse_send_open(fm, nodeid, open_flags, opcode, &outarg);
if (!err) {
ff->fh = outarg.fh;
ff->open_flags = outarg.open_flags;
} else if (err != -ENOSYS) {
fuse_file_free(ff);
- return err;
+ return ERR_PTR(err);
} else {
if (isdir)
fc->no_opendir = 1;
@@ -174,9 +162,19 @@ int fuse_do_open(struct fuse_mount *fm, u64 nodeid, struct file *file,
ff->open_flags &= ~FOPEN_DIRECT_IO;
ff->nodeid = nodeid;
- file->private_data = ff;
- return 0;
+ return ff;
+}
+
+int fuse_do_open(struct fuse_mount *fm, u64 nodeid, struct file *file,
+ bool isdir)
+{
+ struct fuse_file *ff = fuse_file_open(fm, nodeid, file->f_flags, isdir);
+
+ if (!IS_ERR(ff))
+ file->private_data = ff;
+
+ return PTR_ERR_OR_ZERO(ff);
}
EXPORT_SYMBOL_GPL(fuse_do_open);
@@ -268,7 +266,7 @@ out:
}
static void fuse_prepare_release(struct fuse_inode *fi, struct fuse_file *ff,
- int flags, int opcode)
+ unsigned int flags, int opcode)
{
struct fuse_conn *fc = ff->fm->fc;
struct fuse_release_args *ra = ff->release_args;
@@ -297,22 +295,21 @@ static void fuse_prepare_release(struct fuse_inode *fi, struct fuse_file *ff,
ra->args.nocreds = true;
}
-void fuse_release_common(struct file *file, bool isdir)
+void fuse_file_release(struct inode *inode, struct fuse_file *ff,
+ unsigned int open_flags, fl_owner_t id, bool isdir)
{
- struct fuse_inode *fi = get_fuse_inode(file_inode(file));
- struct fuse_file *ff = file->private_data;
+ struct fuse_inode *fi = get_fuse_inode(inode);
struct fuse_release_args *ra = ff->release_args;
int opcode = isdir ? FUSE_RELEASEDIR : FUSE_RELEASE;
- fuse_prepare_release(fi, ff, file->f_flags, opcode);
+ fuse_prepare_release(fi, ff, open_flags, opcode);
if (ff->flock) {
ra->inarg.release_flags |= FUSE_RELEASE_FLOCK_UNLOCK;
- ra->inarg.lock_owner = fuse_lock_owner_id(ff->fm->fc,
- (fl_owner_t) file);
+ ra->inarg.lock_owner = fuse_lock_owner_id(ff->fm->fc, id);
}
/* Hold inode until release is finished */
- ra->inode = igrab(file_inode(file));
+ ra->inode = igrab(inode);
/*
* Normally this will send the RELEASE request, however if
@@ -326,6 +323,12 @@ void fuse_release_common(struct file *file, bool isdir)
fuse_file_put(ff, ff->fm->fc->destroy, isdir);
}
+void fuse_release_common(struct file *file, bool isdir)
+{
+ fuse_file_release(file_inode(file), file->private_data, file->f_flags,
+ (fl_owner_t) file, isdir);
+}
+
static int fuse_open(struct inode *inode, struct file *file)
{
return fuse_open_common(inode, file, false);
@@ -345,7 +348,8 @@ static int fuse_release(struct inode *inode, struct file *file)
return 0;
}
-void fuse_sync_release(struct fuse_inode *fi, struct fuse_file *ff, int flags)
+void fuse_sync_release(struct fuse_inode *fi, struct fuse_file *ff,
+ unsigned int flags)
{
WARN_ON(refcount_read(&ff->count) > 1);
fuse_prepare_release(fi, ff, flags, FUSE_RELEASE);
@@ -798,21 +802,12 @@ static void fuse_short_read(struct inode *inode, u64 attr_ver, size_t num_read,
{
struct fuse_conn *fc = get_fuse_conn(inode);
- if (fc->writeback_cache) {
- /*
- * A hole in a file. Some data after the hole are in page cache,
- * but have not reached the client fs yet. So, the hole is not
- * present there.
- */
- int i;
- int start_idx = num_read >> PAGE_SHIFT;
- size_t off = num_read & (PAGE_SIZE - 1);
-
- for (i = start_idx; i < ap->num_pages; i++) {
- zero_user_segment(ap->pages[i], off, PAGE_SIZE);
- off = 0;
- }
- } else {
+ /*
+ * If writeback_cache is enabled, a short read means there's a hole in
+ * the file. Some data after the hole is in page cache, but has not
+ * reached the client fs yet. So the hole is not present there.
+ */
+ if (!fc->writeback_cache) {
loff_t pos = page_offset(ap->pages[0]) + num_read;
fuse_read_update_size(inode, pos, attr_ver);
}
@@ -1099,6 +1094,7 @@ static ssize_t fuse_send_write_pages(struct fuse_io_args *ia,
struct fuse_file *ff = file->private_data;
struct fuse_mount *fm = ff->fm;
unsigned int offset, i;
+ bool short_write;
int err;
for (i = 0; i < ap->num_pages; i++)
@@ -1113,32 +1109,38 @@ static ssize_t fuse_send_write_pages(struct fuse_io_args *ia,
if (!err && ia->write.out.size > count)
err = -EIO;
+ short_write = ia->write.out.size < count;
offset = ap->descs[0].offset;
count = ia->write.out.size;
for (i = 0; i < ap->num_pages; i++) {
struct page *page = ap->pages[i];
- if (!err && !offset && count >= PAGE_SIZE)
- SetPageUptodate(page);
-
- if (count > PAGE_SIZE - offset)
- count -= PAGE_SIZE - offset;
- else
- count = 0;
- offset = 0;
-
- unlock_page(page);
+ if (err) {
+ ClearPageUptodate(page);
+ } else {
+ if (count >= PAGE_SIZE - offset)
+ count -= PAGE_SIZE - offset;
+ else {
+ if (short_write)
+ ClearPageUptodate(page);
+ count = 0;
+ }
+ offset = 0;
+ }
+ if (ia->write.page_locked && (i == ap->num_pages - 1))
+ unlock_page(page);
put_page(page);
}
return err;
}
-static ssize_t fuse_fill_write_pages(struct fuse_args_pages *ap,
+static ssize_t fuse_fill_write_pages(struct fuse_io_args *ia,
struct address_space *mapping,
struct iov_iter *ii, loff_t pos,
unsigned int max_pages)
{
+ struct fuse_args_pages *ap = &ia->ap;
struct fuse_conn *fc = get_fuse_conn(mapping->host);
unsigned offset = pos & (PAGE_SIZE - 1);
size_t count = 0;
@@ -1191,6 +1193,16 @@ static ssize_t fuse_fill_write_pages(struct fuse_args_pages *ap,
if (offset == PAGE_SIZE)
offset = 0;
+ /* If we copied full page, mark it uptodate */
+ if (tmp == PAGE_SIZE)
+ SetPageUptodate(page);
+
+ if (PageUptodate(page)) {
+ unlock_page(page);
+ } else {
+ ia->write.page_locked = true;
+ break;
+ }
if (!fc->big_writes)
break;
} while (iov_iter_count(ii) && count < fc->max_write &&
@@ -1234,7 +1246,7 @@ static ssize_t fuse_perform_write(struct kiocb *iocb,
break;
}
- count = fuse_fill_write_pages(ap, mapping, ii, pos, nr_pages);
+ count = fuse_fill_write_pages(&ia, mapping, ii, pos, nr_pages);
if (count <= 0) {
err = count;
} else {
@@ -1346,16 +1358,6 @@ out:
return written ? written : err;
}
-static inline void fuse_page_descs_length_init(struct fuse_page_desc *descs,
- unsigned int index,
- unsigned int nr_pages)
-{
- int i;
-
- for (i = index; i < index + nr_pages; i++)
- descs[i].length = PAGE_SIZE - descs[i].offset;
-}
-
static inline unsigned long fuse_get_user_addr(const struct iov_iter *ii)
{
return (unsigned long)ii->iov->iov_base + ii->iov_offset;
@@ -1759,8 +1761,17 @@ static void fuse_writepage_end(struct fuse_mount *fm, struct fuse_args *args,
container_of(args, typeof(*wpa), ia.ap.args);
struct inode *inode = wpa->inode;
struct fuse_inode *fi = get_fuse_inode(inode);
+ struct fuse_conn *fc = get_fuse_conn(inode);
mapping_set_error(inode->i_mapping, error);
+ /*
+ * A writeback finished and this might have updated mtime/ctime on
+ * server making local mtime/ctime stale. Hence invalidate attrs.
+ * Do this only if writeback_cache is not enabled. If writeback_cache
+ * is enabled, we trust local ctime/mtime.
+ */
+ if (!fc->writeback_cache)
+ fuse_invalidate_attr(inode);
spin_lock(&fi->lock);
rb_erase(&wpa->writepages_entry, &fi->writepages);
while (wpa->next) {
@@ -2637,363 +2648,6 @@ static loff_t fuse_file_llseek(struct file *file, loff_t offset, int whence)
}
/*
- * CUSE servers compiled on 32bit broke on 64bit kernels because the
- * ABI was defined to be 'struct iovec' which is different on 32bit
- * and 64bit. Fortunately we can determine which structure the server
- * used from the size of the reply.
- */
-static int fuse_copy_ioctl_iovec_old(struct iovec *dst, void *src,
- size_t transferred, unsigned count,
- bool is_compat)
-{
-#ifdef CONFIG_COMPAT
- if (count * sizeof(struct compat_iovec) == transferred) {
- struct compat_iovec *ciov = src;
- unsigned i;
-
- /*
- * With this interface a 32bit server cannot support
- * non-compat (i.e. ones coming from 64bit apps) ioctl
- * requests
- */
- if (!is_compat)
- return -EINVAL;
-
- for (i = 0; i < count; i++) {
- dst[i].iov_base = compat_ptr(ciov[i].iov_base);
- dst[i].iov_len = ciov[i].iov_len;
- }
- return 0;
- }
-#endif
-
- if (count * sizeof(struct iovec) != transferred)
- return -EIO;
-
- memcpy(dst, src, transferred);
- return 0;
-}
-
-/* Make sure iov_length() won't overflow */
-static int fuse_verify_ioctl_iov(struct fuse_conn *fc, struct iovec *iov,
- size_t count)
-{
- size_t n;
- u32 max = fc->max_pages << PAGE_SHIFT;
-
- for (n = 0; n < count; n++, iov++) {
- if (iov->iov_len > (size_t) max)
- return -ENOMEM;
- max -= iov->iov_len;
- }
- return 0;
-}
-
-static int fuse_copy_ioctl_iovec(struct fuse_conn *fc, struct iovec *dst,
- void *src, size_t transferred, unsigned count,
- bool is_compat)
-{
- unsigned i;
- struct fuse_ioctl_iovec *fiov = src;
-
- if (fc->minor < 16) {
- return fuse_copy_ioctl_iovec_old(dst, src, transferred,
- count, is_compat);
- }
-
- if (count * sizeof(struct fuse_ioctl_iovec) != transferred)
- return -EIO;
-
- for (i = 0; i < count; i++) {
- /* Did the server supply an inappropriate value? */
- if (fiov[i].base != (unsigned long) fiov[i].base ||
- fiov[i].len != (unsigned long) fiov[i].len)
- return -EIO;
-
- dst[i].iov_base = (void __user *) (unsigned long) fiov[i].base;
- dst[i].iov_len = (size_t) fiov[i].len;
-
-#ifdef CONFIG_COMPAT
- if (is_compat &&
- (ptr_to_compat(dst[i].iov_base) != fiov[i].base ||
- (compat_size_t) dst[i].iov_len != fiov[i].len))
- return -EIO;
-#endif
- }
-
- return 0;
-}
-
-
-/*
- * For ioctls, there is no generic way to determine how much memory
- * needs to be read and/or written. Furthermore, ioctls are allowed
- * to dereference the passed pointer, so the parameter requires deep
- * copying but FUSE has no idea whatsoever about what to copy in or
- * out.
- *
- * This is solved by allowing FUSE server to retry ioctl with
- * necessary in/out iovecs. Let's assume the ioctl implementation
- * needs to read in the following structure.
- *
- * struct a {
- * char *buf;
- * size_t buflen;
- * }
- *
- * On the first callout to FUSE server, inarg->in_size and
- * inarg->out_size will be NULL; then, the server completes the ioctl
- * with FUSE_IOCTL_RETRY set in out->flags, out->in_iovs set to 1 and
- * the actual iov array to
- *
- * { { .iov_base = inarg.arg, .iov_len = sizeof(struct a) } }
- *
- * which tells FUSE to copy in the requested area and retry the ioctl.
- * On the second round, the server has access to the structure and
- * from that it can tell what to look for next, so on the invocation,
- * it sets FUSE_IOCTL_RETRY, out->in_iovs to 2 and iov array to
- *
- * { { .iov_base = inarg.arg, .iov_len = sizeof(struct a) },
- * { .iov_base = a.buf, .iov_len = a.buflen } }
- *
- * FUSE will copy both struct a and the pointed buffer from the
- * process doing the ioctl and retry ioctl with both struct a and the
- * buffer.
- *
- * This time, FUSE server has everything it needs and completes ioctl
- * without FUSE_IOCTL_RETRY which finishes the ioctl call.
- *
- * Copying data out works the same way.
- *
- * Note that if FUSE_IOCTL_UNRESTRICTED is clear, the kernel
- * automatically initializes in and out iovs by decoding @cmd with
- * _IOC_* macros and the server is not allowed to request RETRY. This
- * limits ioctl data transfers to well-formed ioctls and is the forced
- * behavior for all FUSE servers.
- */
-long fuse_do_ioctl(struct file *file, unsigned int cmd, unsigned long arg,
- unsigned int flags)
-{
- struct fuse_file *ff = file->private_data;
- struct fuse_mount *fm = ff->fm;
- struct fuse_ioctl_in inarg = {
- .fh = ff->fh,
- .cmd = cmd,
- .arg = arg,
- .flags = flags
- };
- struct fuse_ioctl_out outarg;
- struct iovec *iov_page = NULL;
- struct iovec *in_iov = NULL, *out_iov = NULL;
- unsigned int in_iovs = 0, out_iovs = 0, max_pages;
- size_t in_size, out_size, c;
- ssize_t transferred;
- int err, i;
- struct iov_iter ii;
- struct fuse_args_pages ap = {};
-
-#if BITS_PER_LONG == 32
- inarg.flags |= FUSE_IOCTL_32BIT;
-#else
- if (flags & FUSE_IOCTL_COMPAT) {
- inarg.flags |= FUSE_IOCTL_32BIT;
-#ifdef CONFIG_X86_X32
- if (in_x32_syscall())
- inarg.flags |= FUSE_IOCTL_COMPAT_X32;
-#endif
- }
-#endif
-
- /* assume all the iovs returned by client always fits in a page */
- BUILD_BUG_ON(sizeof(struct fuse_ioctl_iovec) * FUSE_IOCTL_MAX_IOV > PAGE_SIZE);
-
- err = -ENOMEM;
- ap.pages = fuse_pages_alloc(fm->fc->max_pages, GFP_KERNEL, &ap.descs);
- iov_page = (struct iovec *) __get_free_page(GFP_KERNEL);
- if (!ap.pages || !iov_page)
- goto out;
-
- fuse_page_descs_length_init(ap.descs, 0, fm->fc->max_pages);
-
- /*
- * If restricted, initialize IO parameters as encoded in @cmd.
- * RETRY from server is not allowed.
- */
- if (!(flags & FUSE_IOCTL_UNRESTRICTED)) {
- struct iovec *iov = iov_page;
-
- iov->iov_base = (void __user *)arg;
-
- switch (cmd) {
- case FS_IOC_GETFLAGS:
- case FS_IOC_SETFLAGS:
- iov->iov_len = sizeof(int);
- break;
- default:
- iov->iov_len = _IOC_SIZE(cmd);
- break;
- }
-
- if (_IOC_DIR(cmd) & _IOC_WRITE) {
- in_iov = iov;
- in_iovs = 1;
- }
-
- if (_IOC_DIR(cmd) & _IOC_READ) {
- out_iov = iov;
- out_iovs = 1;
- }
- }
-
- retry:
- inarg.in_size = in_size = iov_length(in_iov, in_iovs);
- inarg.out_size = out_size = iov_length(out_iov, out_iovs);
-
- /*
- * Out data can be used either for actual out data or iovs,
- * make sure there always is at least one page.
- */
- out_size = max_t(size_t, out_size, PAGE_SIZE);
- max_pages = DIV_ROUND_UP(max(in_size, out_size), PAGE_SIZE);
-
- /* make sure there are enough buffer pages and init request with them */
- err = -ENOMEM;
- if (max_pages > fm->fc->max_pages)
- goto out;
- while (ap.num_pages < max_pages) {
- ap.pages[ap.num_pages] = alloc_page(GFP_KERNEL | __GFP_HIGHMEM);
- if (!ap.pages[ap.num_pages])
- goto out;
- ap.num_pages++;
- }
-
-
- /* okay, let's send it to the client */
- ap.args.opcode = FUSE_IOCTL;
- ap.args.nodeid = ff->nodeid;
- ap.args.in_numargs = 1;
- ap.args.in_args[0].size = sizeof(inarg);
- ap.args.in_args[0].value = &inarg;
- if (in_size) {
- ap.args.in_numargs++;
- ap.args.in_args[1].size = in_size;
- ap.args.in_pages = true;
-
- err = -EFAULT;
- iov_iter_init(&ii, WRITE, in_iov, in_iovs, in_size);
- for (i = 0; iov_iter_count(&ii) && !WARN_ON(i >= ap.num_pages); i++) {
- c = copy_page_from_iter(ap.pages[i], 0, PAGE_SIZE, &ii);
- if (c != PAGE_SIZE && iov_iter_count(&ii))
- goto out;
- }
- }
-
- ap.args.out_numargs = 2;
- ap.args.out_args[0].size = sizeof(outarg);
- ap.args.out_args[0].value = &outarg;
- ap.args.out_args[1].size = out_size;
- ap.args.out_pages = true;
- ap.args.out_argvar = true;
-
- transferred = fuse_simple_request(fm, &ap.args);
- err = transferred;
- if (transferred < 0)
- goto out;
-
- /* did it ask for retry? */
- if (outarg.flags & FUSE_IOCTL_RETRY) {
- void *vaddr;
-
- /* no retry if in restricted mode */
- err = -EIO;
- if (!(flags & FUSE_IOCTL_UNRESTRICTED))
- goto out;
-
- in_iovs = outarg.in_iovs;
- out_iovs = outarg.out_iovs;
-
- /*
- * Make sure things are in boundary, separate checks
- * are to protect against overflow.
- */
- err = -ENOMEM;
- if (in_iovs > FUSE_IOCTL_MAX_IOV ||
- out_iovs > FUSE_IOCTL_MAX_IOV ||
- in_iovs + out_iovs > FUSE_IOCTL_MAX_IOV)
- goto out;
-
- vaddr = kmap_atomic(ap.pages[0]);
- err = fuse_copy_ioctl_iovec(fm->fc, iov_page, vaddr,
- transferred, in_iovs + out_iovs,
- (flags & FUSE_IOCTL_COMPAT) != 0);
- kunmap_atomic(vaddr);
- if (err)
- goto out;
-
- in_iov = iov_page;
- out_iov = in_iov + in_iovs;
-
- err = fuse_verify_ioctl_iov(fm->fc, in_iov, in_iovs);
- if (err)
- goto out;
-
- err = fuse_verify_ioctl_iov(fm->fc, out_iov, out_iovs);
- if (err)
- goto out;
-
- goto retry;
- }
-
- err = -EIO;
- if (transferred > inarg.out_size)
- goto out;
-
- err = -EFAULT;
- iov_iter_init(&ii, READ, out_iov, out_iovs, transferred);
- for (i = 0; iov_iter_count(&ii) && !WARN_ON(i >= ap.num_pages); i++) {
- c = copy_page_to_iter(ap.pages[i], 0, PAGE_SIZE, &ii);
- if (c != PAGE_SIZE && iov_iter_count(&ii))
- goto out;
- }
- err = 0;
- out:
- free_page((unsigned long) iov_page);
- while (ap.num_pages)
- __free_page(ap.pages[--ap.num_pages]);
- kfree(ap.pages);
-
- return err ? err : outarg.result;
-}
-EXPORT_SYMBOL_GPL(fuse_do_ioctl);
-
-long fuse_ioctl_common(struct file *file, unsigned int cmd,
- unsigned long arg, unsigned int flags)
-{
- struct inode *inode = file_inode(file);
- struct fuse_conn *fc = get_fuse_conn(inode);
-
- if (!fuse_allow_current_process(fc))
- return -EACCES;
-
- if (fuse_is_bad(inode))
- return -EIO;
-
- return fuse_do_ioctl(file, cmd, arg, flags);
-}
-
-static long fuse_file_ioctl(struct file *file, unsigned int cmd,
- unsigned long arg)
-{
- return fuse_ioctl_common(file, cmd, arg, 0);
-}
-
-static long fuse_file_compat_ioctl(struct file *file, unsigned int cmd,
- unsigned long arg)
-{
- return fuse_ioctl_common(file, cmd, arg, FUSE_IOCTL_COMPAT);
-}
-
-/*
* All files which have been polled are linked to RB tree
* fuse_conn->polled_files which is indexed by kh. Walk the tree and
* find the matching one.