linux-kernel - Re: [PATCH 01/21] fs: Remove unused function in afs/write.c

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

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <19617.1392307576@warthog.procyon.org.uk>
Date:	Thu, 13 Feb 2014 16:06:16 +0000
From:	David Howells <dhowells@...hat.com>
To:	Rashika Kheria <rashika.kheria@...il.com>
Cc:	dhowells@...hat.com, linux-kernel@...r.kernel.org,
	linux-afs@...ts.infradead.org, josh@...htriplett.org
Subject: Re: [PATCH 01/21] fs: Remove unused function in afs/write.c

Rashika Kheria <rashika.kheria@...il.com> wrote:

> Remove unused function in afs/write.c.
> 
> This eliminates the following warning in afs/write.c:
> fs/afs/write.c:749:5: warning: no previous prototype for ‘afs_page_mkwrite’ [-Wmissing-prototypes]
> 
> Signed-off-by: Rashika Kheria <rashika.kheria@...il.com>
> Reviewed-by: Josh Triplett <josh@...htriplett.org>

I would like to counter with the attached patch.  Yes, I know that
afs_page_mkwrite() still won't be called after this patch - yet.  I have
someone working on RxRPC and AFS stuff and I'd like to have him try and fix up
the missing stuff.

David
---
commit bf0e3d6fc98ba1567deea0a4182580d8041361aa
Author: David Howells <dhowells@...hat.com>
Date:   Thu Feb 13 13:40:57 2014 +0000

    AFS: Actually use afs_page_mkwrite()
    
    When afs_page_mkwrite() was added in commit:
    
    	9b3f26c9110dcea62716aca9b8c68ceb482227ef
    	FS-Cache: Make kAFS use FS-Cache
    
    it wasn't actually called.  It should be called, however, so that we don't see
    local changes not associated with the old data version appearing in fscache's
    cache (afs_page_mkwrite() holds up the write to a file's page through an
    mmap'd page until that page has finished being written to the cache).
    
    Signed-off-by: David Howells <dhowells@...hat.com>
    cc: Rashika Kheria <rashika.kheria@...il.com>
    cc: Josh Triplett <josh@...htriplett.org>

diff --git a/fs/afs/file.c b/fs/afs/file.c
index 66d50fe2ee45..014792a8d34e 100644
--- a/fs/afs/file.c
+++ b/fs/afs/file.c
@@ -18,6 +18,7 @@
 #include <linux/gfp.h>
 #include "internal.h"
 
+static int afs_file_mmap(struct file *file, struct vm_area_struct *vma);
 static int afs_readpage(struct file *file, struct page *page);
 static void afs_invalidatepage(struct page *page, unsigned int offset,
 			       unsigned int length);
@@ -35,7 +36,7 @@ const struct file_operations afs_file_operations = {
 	.write		= do_sync_write,
 	.aio_read	= generic_file_aio_read,
 	.aio_write	= afs_file_write,
-	.mmap		= generic_file_readonly_mmap,
+	.mmap		= afs_file_mmap,
 	.splice_read	= generic_file_splice_read,
 	.fsync		= afs_fsync,
 	.lock		= afs_lock,
@@ -378,3 +379,25 @@ static int afs_releasepage(struct page *page, gfp_t gfp_flags)
 	_leave(" = T");
 	return 1;
 }
+
+static const struct vm_operations_struct afs_vm_ops = {
+	.fault		= filemap_fault,
+	.page_mkwrite	= afs_page_mkwrite,
+	.remap_pages	= generic_file_remap_pages,
+};
+
+/*
+ * Handle setting up a memory mapping on an AFS file.
+ *
+ * At the moment we only support read-only mappings as writable mappings must
+ * set up a writeback record.
+ */
+static int afs_file_mmap(struct file *file, struct vm_area_struct *vma)
+{
+	int ret;
+
+	ret = generic_file_readonly_mmap(file, vma);
+	if (ret == 0)
+		vma->vm_ops = &afs_vm_ops;
+	return ret;
+}
diff --git a/fs/afs/internal.h b/fs/afs/internal.h
index 6621f8008122..a074279bdab7 100644
--- a/fs/afs/internal.h
+++ b/fs/afs/internal.h
@@ -750,6 +750,7 @@ extern ssize_t afs_file_write(struct kiocb *, const struct iovec *,
 			      unsigned long, loff_t);
 extern int afs_writeback_all(struct afs_vnode *);
 extern int afs_fsync(struct file *, loff_t, loff_t, int);
+extern int afs_page_mkwrite(struct vm_area_struct *, struct page *);
 
 
 /*****************************************************************************/
diff --git a/fs/afs/write.c b/fs/afs/write.c
index a890db4b9898..a593c2c0970e 100644
--- a/fs/afs/write.c
+++ b/fs/afs/write.c
@@ -759,6 +759,13 @@ int afs_page_mkwrite(struct vm_area_struct *vma, struct page *page)
 	fscache_wait_on_page_write(vnode->cache, page);
 #endif
 
+	/* TODO: Get the key from vma->vm_file, flush any overlapping
+	 * contradictory writeback record, set up new a writeback record if
+	 * needed and remove the R/O check from afs_file_mmap().
+	 *
+	 * The code can probably be common with much of afs_write_begin().
+	 */
+
 	_leave(" = 0");
 	return 0;
 }
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/