From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1753741AbbAVUIN (ORCPT ); Thu, 22 Jan 2015 15:08:13 -0500 Received: from exprod5og106.obsmtp.com ([64.18.0.182]:57101 "EHLO mail-pa0-f41.google.com" rhost-flags-OK-OK-OK-FAIL) by vger.kernel.org with ESMTP id S1753141AbbAVUIJ (ORCPT ); Thu, 22 Jan 2015 15:08:09 -0500 X-Greylist: delayed 444 seconds by postgrey-1.27 at vger.kernel.org; Thu, 22 Jan 2015 15:08:08 EST From: Iyappan Subramanian To: davem@davemloft.net, netdev@vger.kernel.org Cc: linux-kernel@vger.kernel.org, linux-arm-kernel@lists.infradead.org, mlangsdo@redhat.com, patches@apm.com, Iyappan Subramanian , Keyur Chudgar Subject: [PATCH] drivers: net: xgene: fix: Out of order descriptor bytes read Date: Thu, 22 Jan 2015 12:03:27 -0800 Message-Id: <1421957007-720-1-git-send-email-isubramanian@apm.com> X-Mailer: git-send-email 1.9.1 Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org This patch fixes the following kernel crash, WARNING: CPU: 2 PID: 0 at net/ipv4/tcp_input.c:3079 tcp_clean_rtx_queue+0x658/0x80c() Call trace: [] dump_backtrace+0x0/0x184 [] show_stack+0x10/0x1c [] dump_stack+0x74/0x98 [] warn_slowpath_common+0x88/0xb0 [] warn_slowpath_null+0x14/0x20 [] tcp_clean_rtx_queue+0x654/0x80c [] tcp_ack+0x454/0x688 [] tcp_rcv_established+0x4a4/0x62c [] tcp_v4_do_rcv+0x16c/0x350 [] tcp_v4_rcv+0x8e8/0x904 [] ip_local_deliver_finish+0x100/0x26c [] ip_local_deliver+0xac/0xc4 [] ip_rcv_finish+0xe8/0x328 [] ip_rcv+0x24c/0x38c [] __netif_receive_skb_core+0x29c/0x7c8 [] __netif_receive_skb+0x28/0x7c [] netif_receive_skb_internal+0x5c/0xe0 [] napi_gro_receive+0xb4/0x110 [] xgene_enet_process_ring+0x144/0x338 [] xgene_enet_napi+0x1c/0x50 [] net_rx_action+0x154/0x228 [] __do_softirq+0x110/0x28c [] irq_exit+0x8c/0xc0 [] handle_IRQ+0x44/0xa8 [] gic_handle_irq+0x38/0x7c [...] Software writes poison data into the descriptor bytes[15:8] and upon receiving the interrupt, if those bytes are overwritten by the hardware with the valid data, software also reads bytes[7:0] and executes receive/tx completion logic. If the CPU executes the above two reads in out of order fashion, then the bytes[7:0] will have older data and causing the kernel panic. We have to force the order of the reads and thus this patch introduces read memory barrier between these reads. Signed-off-by: Iyappan Subramanian Signed-off-by: Keyur Chudgar Tested-by: Mark Langsdorf --- drivers/net/ethernet/apm/xgene/xgene_enet_main.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/drivers/net/ethernet/apm/xgene/xgene_enet_main.c b/drivers/net/ethernet/apm/xgene/xgene_enet_main.c index 83a5028..3622cdb 100644 --- a/drivers/net/ethernet/apm/xgene/xgene_enet_main.c +++ b/drivers/net/ethernet/apm/xgene/xgene_enet_main.c @@ -369,6 +369,8 @@ static int xgene_enet_process_ring(struct xgene_enet_desc_ring *ring, if (unlikely(xgene_enet_is_desc_slot_empty(raw_desc))) break; + /* read fpqnum field after dataaddr field */ + smp_rmb(); if (is_rx_desc(raw_desc)) ret = xgene_enet_rx_frame(ring, raw_desc); else -- 1.9.1