Re: [dpdk-dev] [PATCH] rte_ring: fix racy dequeue/enqueue in ppc64

-Original Message- > Date: Tue, 17 Jul 2018 11:54:18 +0900 > From: Takeshi Yoshimura > To: Jerin Jacob > Cc: dev@dpdk.org, sta...@dpdk.org > Subject: Re: [dpdk-dev] [PATCH] rte_ring: fix racy dequeue/enqueue in ppc64 > Cc: olivier.m...@6wind.com Cc: chao...@lin

Re: [dpdk-dev] [PATCH] rte_ring: fix racy dequeue/enqueue in ppc64

ring as another patch. 2018-07-13 2:08 GMT+09:00 Jerin Jacob : > -Original Message- >> Date: Thu, 12 Jul 2018 11:44:14 +0900 >> From: Takeshi Yoshimura >> To: dev@dpdk.org >> Cc: Takeshi Yoshimura , sta...@dpdk.org, Takeshi >> Yoshimura >> Subject: [d

Re: [dpdk-dev] [PATCH] rte_ring: fix racy dequeue/enqueue in ppc64

-Original Message- > Date: Thu, 12 Jul 2018 11:44:14 +0900 > From: Takeshi Yoshimura > To: dev@dpdk.org > Cc: Takeshi Yoshimura , sta...@dpdk.org, Takeshi > Yoshimura > Subject: [dpdk-dev] [PATCH] rte_ring: fix racy dequeue/enqueue in ppc64 > X-Mailer:

[dpdk-dev] [PATCH] rte_ring: fix racy dequeue/enqueue in ppc64

SPDK blobfs encountered a crash around rte_ring dequeues in ppc64. It uses a single consumer and multiple producers for a rte_ring. The problem was a load-load reorder in rte_ring_sc_dequeue_bulk(). The reordered loads happened on r->prod.tail in __rte_ring_move_cons_head() (rte_ring_generic.h) an