From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S932161AbVLNPX2 (ORCPT ); Wed, 14 Dec 2005 10:23:28 -0500 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S932430AbVLNPX2 (ORCPT ); Wed, 14 Dec 2005 10:23:28 -0500 Received: from clock-tower.bc.nu ([81.2.110.250]:28390 "EHLO lxorguk.ukuu.org.uk") by vger.kernel.org with ESMTP id S932161AbVLNPX2 (ORCPT ); Wed, 14 Dec 2005 10:23:28 -0500 Subject: Serial: bug in 8250.c when handling PCI or other level triggers From: Alan Cox To: rmk@arm.linux.org.uk, linux-kernel@vger.kernel.org Content-Type: text/plain Content-Transfer-Encoding: 7bit Date: Wed, 14 Dec 2005 15:23:23 +0000 Message-Id: <1134573803.25663.35.camel@localhost.localdomain> Mime-Version: 1.0 X-Mailer: Evolution 2.2.3 (2.2.3-2.fc4) Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org The receive_chars function is designed to handle the case where the port is jammed full on by aborting after 256 characters in one IRQ. Unfortunately the author of this code forgot that some systems are level triggered. On these systems the IRQ simply gets invoked again and the count loop just makes the problem take longer to clear. In the non level case it appears that a jam could leave the IRQ jammed forever and ignored - the remaining bytes never appearing. As far as I can see the code would actually need to clear the interrupt enable and then restore it on a timer event or similar in order to get the intended effect. Alan