mm: make do_mmap_pgoff return populate as a size in bytes, not as a bool
do_mmap_pgoff() rounds up the desired size to the next PAGE_SIZE
multiple, however there was no equivalent code in mm_populate(), which
caused issues.
This could be fixed by introduced the same rounding in mm_populate(),
however I think it's preferable to make do_mmap_pgoff() return populate
as a size rather than as a boolean, so we don't have to duplicate the
size rounding logic in mm_populate().
Signed-off-by: Michel Lespinasse <walken@google.com>
Acked-by: Rik van Riel <riel@redhat.com>
Tested-by: Andy Lutomirski <luto@amacapital.net>
Cc: Greg Ungerer <gregungerer@westnet.com.au>
Cc: David Howells <dhowells@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
diff --git a/fs/aio.c b/fs/aio.c
index 82eec7c..064bfbe 100644
--- a/fs/aio.c
+++ b/fs/aio.c
@@ -101,9 +101,8 @@
struct aio_ring *ring;
struct aio_ring_info *info = &ctx->ring_info;
unsigned nr_events = ctx->max_reqs;
- unsigned long size;
+ unsigned long size, populate;
int nr_pages;
- bool populate;
/* Compensate for the ring buffer's head/tail overlap entry */
nr_events += 2; /* 1 is required, 2 for good luck */
@@ -150,7 +149,7 @@
return -EAGAIN;
}
if (populate)
- mm_populate(info->mmap_base, info->mmap_size);
+ mm_populate(info->mmap_base, populate);
ctx->user_id = info->mmap_base;