From mboxrd@z Thu Jan 1 00:00:00 1970 From: Eric Dumazet Subject: [PATCH net 09/11] sfc: remove ndo_poll_controller Date: Thu, 27 Sep 2018 09:31:59 -0700 Message-ID: <20180927163201.56609-10-edumazet@google.com> References: <20180927163201.56609-1-edumazet@google.com> Mime-Version: 1.0 Content-Transfer-Encoding: 8bit Cc: netdev , Eric Dumazet , Michael Chan , Aviad Krawczyk , Song Liu , Douglas Miller , Yisen Zhuang , "Michael S . Tsirkin" , Jason Wang , Harish Patil , Manish Chopra , Netanel Belgazal , Solarflare linux maintainers , Thomas Falcon , Edward Cree , Bert Kenward To: "David S . Miller" Return-path: Received: from mail-pg1-f195.google.com ([209.85.215.195]:46252 "EHLO mail-pg1-f195.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1728139AbeI0Wvw (ORCPT ); Thu, 27 Sep 2018 18:51:52 -0400 Received: by mail-pg1-f195.google.com with SMTP id b129-v6so2318900pga.13 for ; Thu, 27 Sep 2018 09:32:48 -0700 (PDT) In-Reply-To: <20180927163201.56609-1-edumazet@google.com> Sender: netdev-owner@vger.kernel.org List-ID: As diagnosed by Song Liu, ndo_poll_controller() can be very dangerous on loaded hosts, since the cpu calling ndo_poll_controller() might steal all NAPI contexts (for all RX/TX queues of the NIC). This capture can last for unlimited amount of time, since one cpu is generally not able to drain all the queues under load. sfc uses NAPI for TX completions, so we better let core networking stack call the napi->poll() to avoid the capture. Signed-off-by: Eric Dumazet Cc: Edward Cree Cc: Bert Kenward Cc: Solarflare linux maintainers --- drivers/net/ethernet/sfc/efx.c | 26 -------------------------- 1 file changed, 26 deletions(-) diff --git a/drivers/net/ethernet/sfc/efx.c b/drivers/net/ethernet/sfc/efx.c index 330233286e785254f5f29c87f9557a305974f606..3d0dd39c289e05b8a7a6778363461ef5698dc62b 100644 --- a/drivers/net/ethernet/sfc/efx.c +++ b/drivers/net/ethernet/sfc/efx.c @@ -2206,29 +2206,6 @@ static void efx_fini_napi(struct efx_nic *efx) efx_fini_napi_channel(channel); } -/************************************************************************** - * - * Kernel netpoll interface - * - *************************************************************************/ - -#ifdef CONFIG_NET_POLL_CONTROLLER - -/* Although in the common case interrupts will be disabled, this is not - * guaranteed. However, all our work happens inside the NAPI callback, - * so no locking is required. - */ -static void efx_netpoll(struct net_device *net_dev) -{ - struct efx_nic *efx = netdev_priv(net_dev); - struct efx_channel *channel; - - efx_for_each_channel(channel, efx) - efx_schedule_channel(channel); -} - -#endif - /************************************************************************** * * Kernel net device interface @@ -2509,9 +2486,6 @@ static const struct net_device_ops efx_netdev_ops = { #endif .ndo_get_phys_port_id = efx_get_phys_port_id, .ndo_get_phys_port_name = efx_get_phys_port_name, -#ifdef CONFIG_NET_POLL_CONTROLLER - .ndo_poll_controller = efx_netpoll, -#endif .ndo_setup_tc = efx_setup_tc, #ifdef CONFIG_RFS_ACCEL .ndo_rx_flow_steer = efx_filter_rfs, -- 2.19.0.605.g01d371f741-goog