summaryrefslogtreecommitdiff
path: root/mm/mprotect.c
diff options
context:
space:
mode:
authorDavid Howells <dhowells@redhat.com>2006-06-23 13:03:43 +0400
committerLinus Torvalds <torvalds@g5.osdl.org>2006-06-23 18:42:51 +0400
commit9637a5efd4fbe36164c5ce7f6a0ee68b2bf22b7f (patch)
tree38b86e3e2151e78f952076e36bee4fd7d77e3baf /mm/mprotect.c
parentbd96b9eb7cfd6ab24ba244360a09980a720874d2 (diff)
downloadlinux-9637a5efd4fbe36164c5ce7f6a0ee68b2bf22b7f.tar.xz
[PATCH] add page_mkwrite() vm_operations method
Add a new VMA operation to notify a filesystem or other driver about the MMU generating a fault because userspace attempted to write to a page mapped through a read-only PTE. This facility permits the filesystem or driver to: (*) Implement storage allocation/reservation on attempted write, and so to deal with problems such as ENOSPC more gracefully (perhaps by generating SIGBUS). (*) Delay making the page writable until the contents have been written to a backing cache. This is useful for NFS/AFS when using FS-Cache/CacheFS. It permits the filesystem to have some guarantee about the state of the cache. (*) Account and limit number of dirty pages. This is one piece of the puzzle needed to make shared writable mapping work safely in FUSE. Needed by cachefs (Or is it cachefiles? Or fscache? <head spins>). At least four other groups have stated an interest in it or a desire to use the functionality it provides: FUSE, OCFS2, NTFS and JFFS2. Also, things like EXT3 really ought to use it to deal with the case of shared-writable mmap encountering ENOSPC before we permit the page to be dirtied. From: Peter Zijlstra <a.p.zijlstra@chello.nl> get_user_pages(.write=1, .force=1) can generate COW hits on read-only shared mappings, this patch traps those as mkpage_write candidates and fails to handle them the old way. Signed-off-by: David Howells <dhowells@redhat.com> Cc: Miklos Szeredi <miklos@szeredi.hu> Cc: Joel Becker <Joel.Becker@oracle.com> Cc: Mark Fasheh <mark.fasheh@oracle.com> Cc: Anton Altaparmakov <aia21@cantab.net> Cc: David Woodhouse <dwmw2@infradead.org> Cc: Hugh Dickins <hugh@veritas.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Diffstat (limited to 'mm/mprotect.c')
-rw-r--r--mm/mprotect.c11
1 files changed, 9 insertions, 2 deletions
diff --git a/mm/mprotect.c b/mm/mprotect.c
index 14f93e62270f..638edabaff71 100644
--- a/mm/mprotect.c
+++ b/mm/mprotect.c
@@ -123,6 +123,7 @@ mprotect_fixup(struct vm_area_struct *vma, struct vm_area_struct **pprev,
unsigned long oldflags = vma->vm_flags;
long nrpages = (end - start) >> PAGE_SHIFT;
unsigned long charged = 0;
+ unsigned int mask;
pgprot_t newprot;
pgoff_t pgoff;
int error;
@@ -149,8 +150,6 @@ mprotect_fixup(struct vm_area_struct *vma, struct vm_area_struct **pprev,
}
}
- newprot = protection_map[newflags & 0xf];
-
/*
* First try to merge with previous and/or next vma.
*/
@@ -177,6 +176,14 @@ mprotect_fixup(struct vm_area_struct *vma, struct vm_area_struct **pprev,
}
success:
+ /* Don't make the VMA automatically writable if it's shared, but the
+ * backer wishes to know when pages are first written to */
+ mask = VM_READ|VM_WRITE|VM_EXEC|VM_SHARED;
+ if (vma->vm_ops && vma->vm_ops->page_mkwrite)
+ mask &= ~VM_SHARED;
+
+ newprot = protection_map[newflags & mask];
+
/*
* vm_flags and vm_page_prot are protected by the mmap_sem
* held in write mode.