From mboxrd@z Thu Jan 1 00:00:00 1970 From: Zumeng Chen Subject: Re: [PATCH 1/1] tg3: fix meaningless hw_stats reading after tg3_halt memset 0 hw_stats Date: Sun, 29 Apr 2018 14:43:19 +0800 Message-ID: References: <20180428031544.210054-1-zumeng.chen@gmail.com> Mime-Version: 1.0 Content-Type: text/plain; charset=utf-8; format=flowed Content-Transfer-Encoding: 8bit Cc: Netdev , open list , Siva Reddy Kallam , "michael.chan@broadcom.com" , "prashant.sreedharan@broadcom.com" , David Miller To: Michael Chan Return-path: In-Reply-To: Sender: linux-kernel-owner@vger.kernel.org List-Id: netdev.vger.kernel.org On 2018年04月29日 02:36, Michael Chan wrote: > On Fri, Apr 27, 2018 at 8:15 PM, Zumeng Chen wrote: > >> diff --git a/drivers/net/ethernet/broadcom/tg3.h b/drivers/net/ethernet/broadcom/tg3.h >> index 3b5e98e..6727d93 100644 >> --- a/drivers/net/ethernet/broadcom/tg3.h >> +++ b/drivers/net/ethernet/broadcom/tg3.h >> @@ -3352,6 +3352,7 @@ struct tg3 { >> struct pci_dev *pdev_peer; >> >> struct tg3_hw_stats *hw_stats; >> + bool hw_stats_flag; > You can just add another bit to enum TG3_FLAGS for this purpose. Right, it's a good idea, I didn't notice it, I'll send V2 with that later. > > While this scheme will probably work, I think a better and more > elegant way to fix this is to use RCU. IMHO, RCU is not necessary for this simple two consumers, and no frequent ops on tg3_halt, plus no new locker involved either. Cheers, Zumeng > >> dma_addr_t stats_mapping; >> struct work_struct reset_task; >> >> -- >> 2.9.3 >>