From mboxrd@z Thu Jan 1 00:00:00 1970 From: Stephen Hemminger Subject: Fw: [Bug 83131] New: can: ti_hecc.c: closing device while receiving can messages, effectively disables all socket transfer Date: Mon, 25 Aug 2014 14:45:07 -0700 Message-ID: <20140825144507.09c952cf@urahara> Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit To: netdev@vger.kernel.org Return-path: Received: from mail-pd0-f180.google.com ([209.85.192.180]:63184 "EHLO mail-pd0-f180.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753453AbaHYVpL (ORCPT ); Mon, 25 Aug 2014 17:45:11 -0400 Received: by mail-pd0-f180.google.com with SMTP id v10so20791714pde.25 for ; Mon, 25 Aug 2014 14:45:11 -0700 (PDT) Received: from urahara (static-50-53-65-80.bvtn.or.frontiernet.net. [50.53.65.80]) by mx.google.com with ESMTPSA id yw4sm763645pbc.69.2014.08.25.14.45.10 for (version=TLSv1.2 cipher=ECDHE-RSA-AES128-GCM-SHA256 bits=128/128); Mon, 25 Aug 2014 14:45:10 -0700 (PDT) Sender: netdev-owner@vger.kernel.org List-ID: Begin forwarded message: Date: Sun, 24 Aug 2014 06:07:49 -0700 From: "bugzilla-daemon@bugzilla.kernel.org" To: "stephen@networkplumber.org" Subject: [Bug 83131] New: can: ti_hecc.c: closing device while receiving can messages, effectively disables all socket transfer https://bugzilla.kernel.org/show_bug.cgi?id=83131 Bug ID: 83131 Summary: can: ti_hecc.c: closing device while receiving can messages, effectively disables all socket transfer Product: Networking Version: 2.5 Kernel Version: 3.16.1 Hardware: All OS: Linux Tree: PREEMPT_RT Status: NEW Severity: normal Priority: P1 Component: Other Assignee: shemminger@linux-foundation.org Reporter: jan.sondhauss@gmail.com Regression: No drivers/net/can/ti_hecc.c: When closing the can device, its not ensured that it gets removed from the socket-poll list. The problem here lies in the ti_hecc_rx_poll() function which returns early when the device is closed already. This leads to napi_complete() never being called, thus the device stays in the napi's poll list forever and every task reading from any socket is polling this device. The system is now basically unreachable. The patch attached seams to fix this problem by removing the early return in the ti_hecc_rx_poll function. But I'm not sure whether this is the correct solution. -- You are receiving this mail because: You are the assignee for the bug.