1
1

opal/fifo: use atomics to set fifo head in opal_fifo_push

This commit changes the opal_fifo_push code to use
opal_update_counted_pointer to set the head. This fixes a data race
that occurs because the read of the fifo head in opal_fifo_pop
requires two instructions. This combined with the non-atomic update in
opal_fifo_push can lead to an ABA issue that puts the fifo in an
inconsistant state.

There are other ways this problem could be fixed. One way would be to
introduce an opal_atomic_read_128 implementation. On x86_64 this would
have to use the cmpxchg16b instruction. Since this instruction would
have to be in the pop path (and always executed) it would be slower
than the fix in this commit.

Closes open-mpi/ompi#1460.

Signed-off-by: Nathan Hjelm <hjelmn@lanl.gov>
Этот коммит содержится в:
Nathan Hjelm 2016-03-17 13:12:29 -06:00
родитель f0e374096a
Коммит dc000213ea

Просмотреть файл

@ -12,7 +12,7 @@
* All rights reserved.
* Copyright (c) 2007 Voltaire All rights reserved.
* Copyright (c) 2010 IBM Corporation. All rights reserved.
* Copyright (c) 2014-2015 Los Alamos National Security, LLC. All rights
* Copyright (c) 2014-2016 Los Alamos National Security, LLC. All rights
* reseved.
* $COPYRIGHT$
*
@ -101,7 +101,8 @@ static inline opal_list_item_t *opal_fifo_push_atomic (opal_fifo_t *fifo,
if (&fifo->opal_fifo_ghost == tail.data.item) {
/* update the head */
fifo->opal_fifo_head.data.item = item;
opal_counted_pointer_t head = {.value = fifo->opal_fifo_head.value};
opal_update_counted_pointer (&fifo->opal_fifo_head, head, item);
} else {
/* update previous item */
tail.data.item->opal_list_next = item;