mm: change return values of finish_mkwrite_fault()
authorJan Kara <jack@suse.cz>
Wed, 14 Dec 2016 23:07:42 +0000 (15:07 -0800)
committerLinus Torvalds <torvalds@linux-foundation.org>
Thu, 15 Dec 2016 00:04:09 +0000 (16:04 -0800)
commita19e25536ed3a20845f642ce531e10c27fb2add5
tree7bcc1c969a3138ef3c31c2fb8ae815c96afe4382
parent66a6197c118540d454913eef24d68d7491ab5d5f
mm: change return values of finish_mkwrite_fault()

Currently finish_mkwrite_fault() returns 0 when PTE got changed before
we acquired PTE lock and VM_FAULT_WRITE when we succeeded in modifying
the PTE.  This is somewhat confusing since 0 generally means success, it
is also inconsistent with finish_fault() which returns 0 on success.
Change finish_mkwrite_fault() to return 0 on success and VM_FAULT_NOPAGE
when PTE changed.  Practically, there should be no behavioral difference
since we bail out from the fault the same way regardless whether we
return 0, VM_FAULT_NOPAGE, or VM_FAULT_WRITE.  Also note that
VM_FAULT_WRITE has no effect for shared mappings since the only two
places that check it - KSM and GUP - care about private mappings only.
Generally the meaning of VM_FAULT_WRITE for shared mappings is not well
defined and we should probably clean that up.

Link: http://lkml.kernel.org/r/1479460644-25076-17-git-send-email-jack@suse.cz
Signed-off-by: Jan Kara <jack@suse.cz>
Acked-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>
Cc: Ross Zwisler <ross.zwisler@linux.intel.com>
Cc: Dan Williams <dan.j.williams@intel.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
mm/memory.c