X-Git-Url: https://git.camperquake.de/gitweb.cgi?a=blobdiff_plain;f=module%2Fzfs%2Fzpl_file.c;h=db6a72c982fa9ed55202dfdd8260b915c32c63e2;hb=5853fe790d1df58c5dd85ea52c5e165b6d43013c;hp=c2e3a6bdcf489811b03035a1fc7f39bb060967c2;hpb=cfc9a5c88f91f7b4d606fce89505e1f404691ea5;p=zfs.git

diff --git a/module/zfs/zpl_file.c b/module/zfs/zpl_file.c
index c2e3a6b..db6a72c 100644
--- a/module/zfs/zpl_file.c
+++ b/module/zfs/zpl_file.c
@@ -76,37 +76,78 @@ zpl_readdir(struct file *filp, void *dirent, filldir_t filldir)
 	return (error);
 }
 
+#if defined(HAVE_FSYNC_WITH_DENTRY)
 /*
- * 2.6.35 API change,
- * As of 2.6.35 the dentry argument to the .fsync() vfs hook was deemed
+ * Linux 2.6.x - 2.6.34 API,
+ * Through 2.6.34 the nfsd kernel server would pass a NULL 'file struct *'
+ * to the fops->fsync() hook.  For this reason, we must be careful not to
+ * use filp unconditionally.
+ */
+static int
+zpl_fsync(struct file *filp, struct dentry *dentry, int datasync)
+{
+	cred_t *cr = CRED();
+	int error;
+
+	crhold(cr);
+	error = -zfs_fsync(dentry->d_inode, datasync, cr);
+	crfree(cr);
+	ASSERT3S(error, <=, 0);
+
+	return (error);
+}
+
+#elif defined(HAVE_FSYNC_WITHOUT_DENTRY)
+/*
+ * Linux 2.6.35 - 3.0 API,
+ * As of 2.6.35 the dentry argument to the fops->fsync() hook was deemed
  * redundant.  The dentry is still accessible via filp->f_path.dentry,
  * and we are guaranteed that filp will never be NULL.
- *
- * 2.6.34 API change,
- * Prior to 2.6.34 the nfsd kernel server would pass a NULL file struct *
- * to the .fsync() hook.  For this reason, we must be careful not to use
- * filp unconditionally in the 3 argument case.
  */
-#ifdef HAVE_2ARGS_FSYNC
 static int
 zpl_fsync(struct file *filp, int datasync)
 {
-	struct dentry *dentry = filp->f_path.dentry;
-#else
+	struct inode *inode = filp->f_mapping->host;
+	cred_t *cr = CRED();
+	int error;
+
+	crhold(cr);
+	error = -zfs_fsync(inode, datasync, cr);
+	crfree(cr);
+	ASSERT3S(error, <=, 0);
+
+	return (error);
+}
+
+#elif defined(HAVE_FSYNC_RANGE)
+/*
+ * Linux 3.1 - 3.x API,
+ * As of 3.1 the responsibility to call filemap_write_and_wait_range() has
+ * been pushed down in to the .fsync() vfs hook.  Additionally, the i_mutex
+ * lock is no longer held by the caller, for zfs we don't require the lock
+ * to be held so we don't acquire it.
+ */
 static int
-zpl_fsync(struct file *filp, struct dentry *dentry, int datasync)
+zpl_fsync(struct file *filp, loff_t start, loff_t end, int datasync)
 {
-#endif /* HAVE_2ARGS_FSYNC */
+	struct inode *inode = filp->f_mapping->host;
 	cred_t *cr = CRED();
 	int error;
 
+	error = filemap_write_and_wait_range(inode->i_mapping, start, end);
+	if (error)
+		return (error);
+
 	crhold(cr);
-	error = -zfs_fsync(dentry->d_inode, datasync, cr);
+	error = -zfs_fsync(inode, datasync, cr);
 	crfree(cr);
 	ASSERT3S(error, <=, 0);
 
 	return (error);
 }
+#else
+#error "Unsupported fops->fsync() implementation"
+#endif
 
 ssize_t
 zpl_read_common(struct inode *ip, const char *buf, size_t len, loff_t pos,
@@ -260,60 +301,6 @@ zpl_mmap(struct file *filp, struct vm_area_struct *vma)
 	return (error);
 }
 
-static struct page **
-pages_vector_from_list(struct list_head *pages, unsigned nr_pages)
-{
-	struct page **pl;
-	struct page *t;
-	unsigned page_idx;
-
-	pl = kmalloc(sizeof(*pl) * nr_pages, GFP_NOFS);
-	if (!pl)
-		return ERR_PTR(-ENOMEM);
-
-	page_idx = 0;
-	list_for_each_entry_reverse(t, pages, lru) {
-		pl[page_idx] = t;
-		page_idx++;
-	}
-
-	return pl;
-}
-
-static int
-zpl_readpages(struct file *file, struct address_space *mapping,
-	struct list_head *pages, unsigned nr_pages)
-{
-	struct inode *ip;
-	struct page  **pl;
-	struct page  *p, *n;
-	int          error;
-
-	ip = mapping->host;
-
-	pl = pages_vector_from_list(pages, nr_pages);
-	if (IS_ERR(pl))
-		return PTR_ERR(pl);
-
-	error = -zfs_getpage(ip, pl, nr_pages);
-	if (error)
-		goto error;
-
-	list_for_each_entry_safe_reverse(p, n, pages, lru) {
-
-		list_del(&p->lru);
-
-		flush_dcache_page(p);
-		SetPageUptodate(p);
-		unlock_page(p);
-		page_cache_release(p);
-	}
-
-error:
-	kfree(pl);
-	return error;
-}
-
 /*
  * Populate a page with data for the Linux page cache.  This function is
  * only used to support mmap(2).  There will be an identical copy of the
@@ -349,33 +336,39 @@ zpl_readpage(struct file *filp, struct page *pp)
 	return error;
 }
 
+/*
+ * Populate a set of pages with data for the Linux page cache.  This
+ * function will only be called for read ahead and never for demand
+ * paging.  For simplicity, the code relies on read_cache_pages() to
+ * correctly lock each page for IO and call zpl_readpage().
+ */
+static int
+zpl_readpages(struct file *filp, struct address_space *mapping,
+	struct list_head *pages, unsigned nr_pages)
+{
+	return (read_cache_pages(mapping, pages,
+	    (filler_t *)zpl_readpage, filp));
+}
+
 int
 zpl_putpage(struct page *pp, struct writeback_control *wbc, void *data)
 {
-	int error;
+	struct address_space *mapping = data;
+
+	ASSERT(PageLocked(pp));
+	ASSERT(!PageWriteback(pp));
+	ASSERT(!(current->flags & PF_NOFS));
 
 	/*
-	 * Disable the normal reclaim path for zpl_putpage().  This
-	 * ensures that all memory allocations under this call path
-	 * will never enter direct reclaim.  If this were to happen
-	 * the VM might try to write out additional pages by calling
-	 * zpl_putpage() again resulting in a deadlock.
+	 * Annotate this call path with a flag that indicates that it is
+	 * unsafe to use KM_SLEEP during memory allocations due to the
+	 * potential for a deadlock.  KM_PUSHPAGE should be used instead.
 	 */
-	current->flags |= PF_MEMALLOC;
-	error = -zfs_putpage(pp, wbc, data);
-	current->flags &= ~PF_MEMALLOC;
-
-	if (error) {
-		SetPageError(pp);
-		ClearPageUptodate(pp);
-	} else {
-		ClearPageError(pp);
-		SetPageUptodate(pp);
-		flush_dcache_page(pp);
-	}
+	current->flags |= PF_NOFS;
+	(void) zfs_putpage(mapping->host, pp, wbc);
+	current->flags &= ~PF_NOFS;
 
-	unlock_page(pp);
-	return error;
+	return (0);
 }
 
 static int
@@ -396,6 +389,50 @@ zpl_writepage(struct page *pp, struct writeback_control *wbc)
 	return zpl_putpage(pp, wbc, pp->mapping);
 }
 
+/*
+ * The only flag combination which matches the behavior of zfs_space()
+ * is FALLOC_FL_PUNCH_HOLE.  This flag was introduced in the 2.6.38 kernel.
+ */
+long
+zpl_fallocate_common(struct inode *ip, int mode, loff_t offset, loff_t len)
+{
+	cred_t *cr = CRED();
+	int error = -EOPNOTSUPP;
+
+	if (mode & FALLOC_FL_KEEP_SIZE)
+		return (-EOPNOTSUPP);
+
+	crhold(cr);
+
+#ifdef FALLOC_FL_PUNCH_HOLE
+	if (mode & FALLOC_FL_PUNCH_HOLE) {
+		flock64_t bf;
+
+		bf.l_type = F_WRLCK;
+		bf.l_whence = 0;
+		bf.l_start = offset;
+		bf.l_len = len;
+		bf.l_pid = 0;
+
+		error = -zfs_space(ip, F_FREESP, &bf, FWRITE, offset, cr);
+	}
+#endif /* FALLOC_FL_PUNCH_HOLE */
+
+	crfree(cr);
+
+	ASSERT3S(error, <=, 0);
+	return (error);
+}
+
+#ifdef HAVE_FILE_FALLOCATE
+static long
+zpl_fallocate(struct file *filp, int mode, loff_t offset, loff_t len)
+{
+	return zpl_fallocate_common(filp->f_path.dentry->d_inode,
+	    mode, offset, len);
+}
+#endif /* HAVE_FILE_FALLOCATE */
+
 const struct address_space_operations zpl_address_space_operations = {
 	.readpages	= zpl_readpages,
 	.readpage	= zpl_readpage,
@@ -409,9 +446,11 @@ const struct file_operations zpl_file_operations = {
 	.llseek		= generic_file_llseek,
 	.read		= zpl_read,
 	.write		= zpl_write,
-	.readdir	= zpl_readdir,
 	.mmap		= zpl_mmap,
 	.fsync		= zpl_fsync,
+#ifdef HAVE_FILE_FALLOCATE
+	.fallocate      = zpl_fallocate,
+#endif /* HAVE_FILE_FALLOCATE */
 };
 
 const struct file_operations zpl_dir_file_operations = {