From mboxrd@z Thu Jan 1 00:00:00 1970 From: Linus Torvalds Subject: Re: brcm80211 breakage.. Date: Wed, 11 Jan 2012 17:06:15 -0800 Message-ID: References: <4F0D6806.4080201@broadcom.com> <4F0DC030.6090500@lwfinger.net> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: QUOTED-PRINTABLE Cc: Arend van Spriel , "John W. Linville" , Network Development , "Franky (Zhenhui) Lin" , =?ISO-8859-2?Q?Rafa=B3_Mi=B3ecki?= To: Larry Finger Return-path: Received: from mail-wi0-f174.google.com ([209.85.212.174]:34475 "EHLO mail-wi0-f174.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751835Ab2ALBGh convert rfc822-to-8bit (ORCPT ); Wed, 11 Jan 2012 20:06:37 -0500 Received: by wibhm14 with SMTP id hm14so838560wib.19 for ; Wed, 11 Jan 2012 17:06:36 -0800 (PST) In-Reply-To: <4F0DC030.6090500@lwfinger.net> Sender: netdev-owner@vger.kernel.org List-ID: On Wed, Jan 11, 2012 at 9:00 AM, Larry Finger wrote: > --- wireless-testing-new.orig/drivers/bcma/sprom.c > +++ wireless-testing-new/drivers/bcma/sprom.c > @@ -230,6 +230,7 @@ int bcma_sprom_get(struct bcma_bus *bus) > =A0 =A0 =A0 =A0 * TODO: understand this condition and use it */ > =A0 =A0 =A0 =A0offset =3D (bus->chipinfo.id =3D=3D 0x4331) ? BCMA_CC_= SPROM : > =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0BCMA_CC_SPROM_PCIE6; > + =A0 =A0 =A0 pr_debug("SPROM offset 0x%x\n", offset); > =A0 =A0 =A0 =A0bcma_sprom_read(bus, offset, sprom); > > =A0 =A0 =A0 =A0if (bus->chipinfo.id =3D=3D 0x4331) So I did that - the offset is printed out as 0x0830 (BCMA_CC_SPROM_PCIE6). And when I change the offset to 0x0800 (BCMA_CC_SPROM), the complaint about SPROM version goes away. IOW, these messages no longer exist: [ 0.815750] bcma: Unsupported SPROM revision: 255 [ 0.815804] bcma: No SPROM available but that doesn't actually make anything *work*. The brcms errors still remain the same. So the SPROM issue seems to be real, but irrelevant. I also noticed that the CONFIG_BRCMDBG onyl enables pr_debug(), but doesn't actually #define DEBUG, so pr_debug() gets compiled out anyway. I'll try with -DDEBUG in the brcm80211 subdirectory and see if I get more interesting output that way. Linus