From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from metis.ext.4.pengutronix.de ([92.198.50.35]:44100 "EHLO metis.ext.pengutronix.de" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752161AbbJLNei (ORCPT ); Mon, 12 Oct 2015 09:34:38 -0400 Message-ID: <1444656873.3364.14.camel@pengutronix.de> Subject: Re: [PATCH v2 1/2] PCI: designware: Move LTSSM state definitions to pcie-designware.h From: Lucas Stach To: Fabio Estevam Cc: Bjorn Helgaas , Pratyush Anand Thakur , m-karicheri2 , "linux-pci@vger.kernel.org" , Fabio Estevam Date: Mon, 12 Oct 2015 15:34:33 +0200 In-Reply-To: References: <1444355306-21486-1-git-send-email-festevam@gmail.com> <1444645188.3364.5.camel@pengutronix.de> <1444645685.3364.8.camel@pengutronix.de> Content-Type: text/plain; charset="UTF-8" Mime-Version: 1.0 Sender: linux-pci-owner@vger.kernel.org List-ID: Am Montag, den 12.10.2015, 09:31 -0300 schrieb Fabio Estevam: > On Mon, Oct 12, 2015 at 7:28 AM, Lucas Stach wrote: > > > Actually the TRM states that i.MX6 is using the same narrower mask of > > 0x1f for the LTSSM state, so another reason to just use that. > > The manual I have seems to tell a different value :-) > > Please check: http://cache.freescale.com/files/32bit/doc/ref_manual/IMX6DQRM.pdf > > "48.12.11 > Debug Register 0 (PCIE_PL_DEBUG0)" > > says "[5:0]: xmlh_ltssm_state LTSSM current state. See source for encodings" > Urgh. Seems we got confused by all the different defines. In fact [5:0] is 0x1f, so i.MX6 uses the same mask as Keystone, yet the common and shared header now states a mask of 0x3f with Keystone being the exception. That doesn't reflect reality which isn't nice IMHO. I think it would be better to just use 0x1f as the LTSSM mask in the shared header and use this for all drivers until a need arises for using the extra bit on SPEAR. Regards, Lucas -- Pengutronix e.K. | Lucas Stach | Industrial Linux Solutions | http://www.pengutronix.de/ |