From: Kim Phillips Subject: Re: talitos_probe hangs Date: Tue, 5 May 2009 12:49:13 -0500 Message-ID: <20090505124913.b7f75e55.kim.phillips@freescale.com> References: <728619.83165.qm@web50504.mail.re2.yahoo.com> Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Cc: linux-crypto@vger.kernel.org, Herbert Xu , "David S. Miller" To: kishor_ravula@yahoo.com Return-path: Received: from az33egw02.freescale.net ([192.88.158.103]:37221 "EHLO az33egw02.freescale.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751521AbZEERty (ORCPT ); Tue, 5 May 2009 13:49:54 -0400 In-Reply-To: <728619.83165.qm@web50504.mail.re2.yahoo.com> Sender: linux-crypto-owner@vger.kernel.org List-ID: On Mon, 4 May 2009 23:01:23 -0700 (PDT) Ravula Kishor wrote: > I am using ltib-20080811 (2.6.25 kernel) . I am porting it to custom board based on mcf5474. I observed that talitos_probe hangs during the kernel initialization. > > Is it a known issue. so the the mcf5474 utilizes SEC h/w version 1.1 (or a variant thereof), with which the talitos driver is currently incompatible. Even if you've managed to fake the device tree node properties, the driver has the master reset bit in a different position than the one on the mcf547x, so it doesn't even know how to reset the device properly. you might try entering a ticket with the freescale coldfire support web - who knows, they might be willing to assist you in porting the driver to handle sec h/w versions < 2.0. AFAIK, various registers, various bits within registers, and the descriptor submission method differ in the h/w, and the driver needs to double as a platform driver for coldfire arch. Kim