From mboxrd@z Thu Jan 1 00:00:00 1970 From: Heikki Krogerus Subject: Re: Serial port initialization broken on Armada 370/XP due to "serial: 8250_dw: Don't use UPF_FIXED_TYPE" Date: Thu, 28 Feb 2013 14:34:21 +0200 Message-ID: <20130228123421.GC5688@xps8300> References: <512E1AD9.4090403@free-electrons.com> <512E2F64.2010706@free-electrons.com> <20130228092602.GA5688@xps8300> <512F428E.3080400@free-electrons.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Return-path: Received: from mga03.intel.com ([143.182.124.21]:61048 "EHLO mga03.intel.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751937Ab3B1Me0 (ORCPT ); Thu, 28 Feb 2013 07:34:26 -0500 Content-Disposition: inline In-Reply-To: <512F428E.3080400@free-electrons.com> Sender: linux-serial-owner@vger.kernel.org List-Id: linux-serial@vger.kernel.org To: Gregory CLEMENT Cc: Jamie Iles , Greg Kroah-Hartman , Jason Cooper , Andrew Lunn , Thomas Petazzoni , Ezequiel Garcia , "linux-kernel@vger.kernel.org" , linux-serial@vger.kernel.org Hi, On Thu, Feb 28, 2013 at 12:42:06PM +0100, Gregory CLEMENT wrote: > >> Would you agree with this kind of patch to fix the issue? > >> > >> diff --git a/drivers/tty/serial/8250/8250.c b/drivers/tty/serial/8250/8250.c > >> index e2ac25a..0b284c6 100644 > >> --- a/drivers/tty/serial/8250/8250.c > >> +++ b/drivers/tty/serial/8250/8250.c > >> @@ -1119,7 +1119,7 @@ static void autoconfig(struct uart_8250_port *up, unsigned int probeflags) > >> serial_out(up, UART_LCR, 0); > >> > >> serial_out(up, UART_FCR, UART_FCR_ENABLE_FIFO); > >> - scratch = serial_in(up, UART_IIR) >> 6; > >> + scratch = (serial_in(up, UART_IIR) & 0xFF) >> 6; > >> > >> switch (scratch) { > >> case 0: > > > > Instead, can you test if it's enough for you to set the reg-io-width > > to 1 instead of 4: > > Yes indeed it worked and it seems to be the correct description of my > hardware. So I will fix the dtsi file. > > However isn't buggy to use a function as it returned a char whereas > it returns an int? Yes, the driver should probable be cleaned. It seems to be happening in quite a few places in 8250.c. autoconfig_16550a() has pretty much identical code in it, where UART_IIR is read to unsigned char and shifted without a mask. Br, -- heikki