mmap: arch_get_unmapped_area(): use proper mmap base for bottom up direction
Heiko Carstens [Tue, 12 Nov 2013 23:07:54 +0000 (15:07 -0800)]
This is more or less the generic variant of commit 41aacc1eea64 ("x86
get_unmapped_area: Access mmap_legacy_base through mm_struct member").

So effectively architectures which use an own arch_pick_mmap_layout()
implementation but call the generic arch_get_unmapped_area() now can
also randomize their mmap_base.

All architectures which have an own arch_pick_mmap_layout() and call the
generic arch_get_unmapped_area() (arm64, s390, tile) currently set
mmap_base to TASK_UNMAPPED_BASE.  This is also true for the generic
arch_pick_mmap_layout() function.  So this change is a no-op currently.

Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com>
Cc: Radu Caragea <sinaelgl@gmail.com>
Cc: Michel Lespinasse <walken@google.com>
Cc: Catalin Marinas <catalin.marinas@arm.com>
Cc: Will Deacon <will.deacon@arm.com>
Cc: Chris Metcalf <cmetcalf@tilera.com>
Cc: Martin Schwidefsky <schwidefsky@de.ibm.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

Bug 1486057

Cherry-pick upstream patch 4e99b02131b280b064d30a5926ef1c4763f3097b

Change-Id: I027ce31f1b9e87c2953bfa0a9d82288d92ba11fd
Signed-off-by: Li Li <lli5@nvidia.com>
Reviewed-on: http://git-master/r/387939
Reviewed-by: Eric Miao <emiao@nvidia.com>
Reviewed-by: Peter Zu <pzu@nvidia.com>
Reviewed-by: Bharat Nihalani <bnihalani@nvidia.com>

mm/mmap.c

index 93cbba8..e0a152e 100644 (file)
--- a/mm/mmap.c
+++ b/mm/mmap.c
@@ -1878,7 +1878,7 @@ arch_get_unmapped_area(struct file *filp, unsigned long addr,
 
        info.flags = 0;
        info.length = len;
-       info.low_limit = TASK_UNMAPPED_BASE;
+       info.low_limit = mm->mmap_base;
        info.high_limit = TASK_SIZE;
        info.align_mask = 0;
        return vm_unmapped_area(&info);