| lists.openwall.net | lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC | |
|
Open Source and information security mailing list archives
| ||
|
Message-ID: <20240130055414.2143959-2-willy@infradead.org>
Date: Tue, 30 Jan 2024 05:54:11 +0000
From: "Matthew Wilcox (Oracle)" <willy@...radead.org>
To: Christoph Hellwig <hch@....de>
Cc: "Matthew Wilcox (Oracle)" <willy@...radead.org>,
linux-fsdevel@...r.kernel.org,
linux-ext4@...r.kernel.org
Subject: [PATCH 1/3] fs: Introduce buffered_write_operations
Start the process of moving write_begin and write_end out from the
address_space_operations to their own struct. Make them take a folio
instead of a page, and pass a pointer to the fsdata to write_end
(so that we don't need to initialise it).
Pass an optional buffered_write_operations pointer to various functions
in filemap.c. The old names are available as macros for now, except
for generic_file_write_iter() which is used as a function pointer by
many filesystems.
Signed-off-by: Matthew Wilcox (Oracle) <willy@...radead.org>
---
fs/jfs/file.c | 3 +-
fs/ramfs/file-mmu.c | 3 +-
fs/ufs/file.c | 2 +-
include/linux/fs.h | 3 --
include/linux/pagemap.h | 22 +++++++++++++
mm/filemap.c | 70 +++++++++++++++++++++++++++--------------
6 files changed, 71 insertions(+), 32 deletions(-)
diff --git a/fs/jfs/file.c b/fs/jfs/file.c
index 01b6912e60f8..9c62445ea6be 100644
--- a/fs/jfs/file.c
+++ b/fs/jfs/file.c
@@ -4,8 +4,7 @@
* Portions Copyright (C) Christoph Hellwig, 2001-2002
*/
-#include <linux/mm.h>
-#include <linux/fs.h>
+#include <linux/pagemap.h>
#include <linux/posix_acl.h>
#include <linux/quotaops.h>
#include "jfs_incore.h"
diff --git a/fs/ramfs/file-mmu.c b/fs/ramfs/file-mmu.c
index c7a1aa3c882b..a621b08b0235 100644
--- a/fs/ramfs/file-mmu.c
+++ b/fs/ramfs/file-mmu.c
@@ -24,8 +24,7 @@
* caches is sufficient.
*/
-#include <linux/fs.h>
-#include <linux/mm.h>
+#include <linux/pagemap.h>
#include <linux/ramfs.h>
#include <linux/sched.h>
diff --git a/fs/ufs/file.c b/fs/ufs/file.c
index 6558882a89ef..b557d4a14143 100644
--- a/fs/ufs/file.c
+++ b/fs/ufs/file.c
@@ -24,7 +24,7 @@
* ext2 fs regular file handling primitives
*/
-#include <linux/fs.h>
+#include <linux/pagemap.h>
#include "ufs_fs.h"
#include "ufs.h"
diff --git a/include/linux/fs.h b/include/linux/fs.h
index ed5966a70495..92dc0cf08b1f 100644
--- a/include/linux/fs.h
+++ b/include/linux/fs.h
@@ -3048,10 +3048,7 @@ extern int generic_file_rw_checks(struct file *file_in, struct file *file_out);
ssize_t filemap_read(struct kiocb *iocb, struct iov_iter *to,
ssize_t already_read);
extern ssize_t generic_file_read_iter(struct kiocb *, struct iov_iter *);
-extern ssize_t __generic_file_write_iter(struct kiocb *, struct iov_iter *);
-extern ssize_t generic_file_write_iter(struct kiocb *, struct iov_iter *);
extern ssize_t generic_file_direct_write(struct kiocb *, struct iov_iter *);
-ssize_t generic_perform_write(struct kiocb *, struct iov_iter *);
ssize_t direct_write_fallback(struct kiocb *iocb, struct iov_iter *iter,
ssize_t direct_written, ssize_t buffered_written);
diff --git a/include/linux/pagemap.h b/include/linux/pagemap.h
index 2df35e65557d..a5c474ad230e 100644
--- a/include/linux/pagemap.h
+++ b/include/linux/pagemap.h
@@ -18,6 +18,28 @@
struct folio_batch;
+struct buffered_write_operations {
+ int (*write_begin)(struct file *, struct address_space *mapping,
+ loff_t pos, size_t len, struct folio **foliop,
+ void **fsdata);
+ int (*write_end)(struct file *, struct address_space *mapping,
+ loff_t pos, size_t len, size_t copied,
+ struct folio *folio, void **fsdata);
+};
+
+ssize_t filemap_write_iter(struct kiocb *, struct iov_iter *,
+ const struct buffered_write_operations *);
+ssize_t __filemap_write_iter(struct kiocb *, struct iov_iter *,
+ const struct buffered_write_operations *);
+ssize_t filemap_perform_write(struct kiocb *, struct iov_iter *,
+ const struct buffered_write_operations *);
+
+ssize_t generic_file_write_iter(struct kiocb *, struct iov_iter *);
+#define generic_perform_write(kiocb, iter) \
+ filemap_perform_write(kiocb, iter, NULL)
+#define __generic_file_write_iter(kiocb, iter) \
+ __filemap_write_iter(kiocb, iter, NULL)
+
unsigned long invalidate_mapping_pages(struct address_space *mapping,
pgoff_t start, pgoff_t end);
diff --git a/mm/filemap.c b/mm/filemap.c
index 750e779c23db..66d779b787c8 100644
--- a/mm/filemap.c
+++ b/mm/filemap.c
@@ -95,7 +95,7 @@
* ->invalidate_lock (filemap_fault)
* ->lock_page (filemap_fault, access_process_vm)
*
- * ->i_rwsem (generic_perform_write)
+ * ->i_rwsem (filemap_perform_write)
* ->mmap_lock (fault_in_readable->do_page_fault)
*
* bdi->wb.list_lock
@@ -3890,7 +3890,8 @@ generic_file_direct_write(struct kiocb *iocb, struct iov_iter *from)
}
EXPORT_SYMBOL(generic_file_direct_write);
-ssize_t generic_perform_write(struct kiocb *iocb, struct iov_iter *i)
+ssize_t filemap_perform_write(struct kiocb *iocb, struct iov_iter *i,
+ const struct buffered_write_operations *ops)
{
struct file *file = iocb->ki_filp;
loff_t pos = iocb->ki_pos;
@@ -3900,9 +3901,9 @@ ssize_t generic_perform_write(struct kiocb *iocb, struct iov_iter *i)
ssize_t written = 0;
do {
- struct page *page;
- unsigned long offset; /* Offset into pagecache page */
- unsigned long bytes; /* Bytes to write to page */
+ struct folio *folio;
+ size_t offset; /* Offset into pagecache folio */
+ size_t bytes; /* Bytes to write to page */
size_t copied; /* Bytes copied from user */
void *fsdata = NULL;
@@ -3927,19 +3928,31 @@ ssize_t generic_perform_write(struct kiocb *iocb, struct iov_iter *i)
break;
}
- status = a_ops->write_begin(file, mapping, pos, bytes,
+ if (ops)
+ status = ops->write_begin(file, mapping, pos, bytes,
+ &folio, &fsdata);
+ else {
+ struct page *page;
+ status = a_ops->write_begin(file, mapping, pos, bytes,
&page, &fsdata);
+ if (status >= 0)
+ folio = page_folio(page);
+ }
if (unlikely(status < 0))
break;
if (mapping_writably_mapped(mapping))
- flush_dcache_page(page);
+ flush_dcache_folio(folio);
- copied = copy_page_from_iter_atomic(page, offset, bytes, i);
- flush_dcache_page(page);
+ copied = copy_folio_from_iter_atomic(folio, offset, bytes, i);
+ flush_dcache_folio(folio);
- status = a_ops->write_end(file, mapping, pos, bytes, copied,
- page, fsdata);
+ if (ops)
+ status = ops->write_end(file, mapping, pos, bytes,
+ copied, folio, &fsdata);
+ else
+ status = a_ops->write_end(file, mapping, pos, bytes,
+ copied, &folio->page, fsdata);
if (unlikely(status != copied)) {
iov_iter_revert(i, copied - max(status, 0L));
if (unlikely(status < 0))
@@ -3969,12 +3982,13 @@ ssize_t generic_perform_write(struct kiocb *iocb, struct iov_iter *i)
iocb->ki_pos += written;
return written;
}
-EXPORT_SYMBOL(generic_perform_write);
+EXPORT_SYMBOL(filemap_perform_write);
/**
- * __generic_file_write_iter - write data to a file
- * @iocb: IO state structure (file, offset, etc.)
- * @from: iov_iter with data to write
+ * __filemap_write_iter - write data to a file
+ * @iocb: IO state structure (file, offset, etc.)
+ * @from: iov_iter with data to write
+ * @ops: How to inform the filesystem that a write is starting/finishing.
*
* This function does all the work needed for actually writing data to a
* file. It does all basic checks, removes SUID from the file, updates
@@ -3992,7 +4006,8 @@ EXPORT_SYMBOL(generic_perform_write);
* * number of bytes written, even for truncated writes
* * negative error code if no data has been written at all
*/
-ssize_t __generic_file_write_iter(struct kiocb *iocb, struct iov_iter *from)
+ssize_t __filemap_write_iter(struct kiocb *iocb, struct iov_iter *from,
+ const struct buffered_write_operations *ops)
{
struct file *file = iocb->ki_filp;
struct address_space *mapping = file->f_mapping;
@@ -4019,27 +4034,29 @@ ssize_t __generic_file_write_iter(struct kiocb *iocb, struct iov_iter *from)
if (ret < 0 || !iov_iter_count(from) || IS_DAX(inode))
return ret;
return direct_write_fallback(iocb, from, ret,
- generic_perform_write(iocb, from));
+ filemap_perform_write(iocb, from, ops));
}
- return generic_perform_write(iocb, from);
+ return filemap_perform_write(iocb, from, ops);
}
-EXPORT_SYMBOL(__generic_file_write_iter);
+EXPORT_SYMBOL(__filemap_write_iter);
/**
- * generic_file_write_iter - write data to a file
+ * filemap_write_iter - write data to a file
* @iocb: IO state structure
* @from: iov_iter with data to write
*
- * This is a wrapper around __generic_file_write_iter() to be used by most
+ * This is a wrapper around __filemap_write_iter() to be used by most
* filesystems. It takes care of syncing the file in case of O_SYNC file
* and acquires i_rwsem as needed.
+ *
* Return:
- * * negative error code if no data has been written at all of
+ * * negative error code if no data has been written at all or if
* vfs_fsync_range() failed for a synchronous write
* * number of bytes written, even for truncated writes
*/
-ssize_t generic_file_write_iter(struct kiocb *iocb, struct iov_iter *from)
+ssize_t filemap_write_iter(struct kiocb *iocb, struct iov_iter *from,
+ const struct buffered_write_operations *ops)
{
struct file *file = iocb->ki_filp;
struct inode *inode = file->f_mapping->host;
@@ -4048,13 +4065,18 @@ ssize_t generic_file_write_iter(struct kiocb *iocb, struct iov_iter *from)
inode_lock(inode);
ret = generic_write_checks(iocb, from);
if (ret > 0)
- ret = __generic_file_write_iter(iocb, from);
+ ret = __filemap_write_iter(iocb, from, ops);
inode_unlock(inode);
if (ret > 0)
ret = generic_write_sync(iocb, ret);
return ret;
}
+
+ssize_t generic_file_write_iter(struct kiocb *iocb, struct iov_iter *from)
+{
+ return filemap_write_iter(iocb, from, NULL);
+}
EXPORT_SYMBOL(generic_file_write_iter);
/**
--
2.43.0
Powered by blists - more mailing lists