We watched a rte panic of mbuf_autotest in our qualcomm arm64 server
(Amberwing).
Root cause:
In __rte_ring_move_cons_head()
...
do {
/* Restore n as it may change every loop */
n = max;
*old_head = r->cons.head; //1st load
const uint32_t prod_tail = r->prod.tail; //2nd load
In weak memory order architectures (powerpc,arm), the 2nd load might be
reodered before the 1st load, that makes *entries is bigger than we wanted.
This nasty reording messed enque/deque up.
cpu1(producer) cpu2(consumer) cpu3(consumer)
load r->prod.tail
in enqueue:
load r->cons.tail
load r->prod.head
store r->prod.tail
load r->cons.head
load r->prod.tail
...
store r->cons.{head,tail}
load r->cons.head
Then, r->cons.head will be bigger than prod_tail, then make *entries very
big and the consumer will go forward incorrectly.
After this patch, the old cons.head will be recaculated after failure of
rte_atomic32_cmpset
There is no such issue on X86, because X86 is strong memory order model.
But rte_smp_rmb() doesn't have impact on runtime performance on X86, so
keep the same code without architectures specific concerns.
Fixes: 50d7690548 ("ring: add burst API")
Cc: stable@dpdk.org
Signed-off-by: Jia He <jia.he@hxt-semitech.com>
Signed-off-by: Jie Liu <jie2.liu@hxt-semitech.com>
Signed-off-by: Bing Zhao <bing.zhao@hxt-semitech.com>
Acked-by: Jerin Jacob <jerin.jacob@caviumnetworks.com>
Acked-by: Jianbo Liu <jianbo.liu@arm.com>
9bc2cbb007