From mboxrd@z Thu Jan 1 00:00:00 1970 From: Tony Lindgren Subject: Re: [PATCH 1/4] usb: musb: musb on omap3 has 32 endpoints Date: Thu, 26 Jun 2008 15:15:58 +0300 Message-ID: <20080626121557.GD12992@atomide.com> References: <1213351443-3336-1-git-send-email-felipe.balbi@nokia.com> <1213351443-3336-2-git-send-email-felipe.balbi@nokia.com> <20080623111009.GM7741@atomide.com> <20080623111455.GH4414@gandalf.research.nokia.com> <5A47E75E594F054BAF48C5E4FC4B92AB022BE46969@dbde02.ent.ti.com> <48625971.8060800@gmail.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Return-path: Received: from mho-01-bos.mailhop.org ([63.208.196.178]:50731 "EHLO mho-01-bos.mailhop.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1750765AbYFZMQK (ORCPT ); Thu, 26 Jun 2008 08:16:10 -0400 Content-Disposition: inline In-Reply-To: <48625971.8060800@gmail.com> Sender: linux-omap-owner@vger.kernel.org List-Id: linux-omap@vger.kernel.org To: Dmitry Krivoschekov Cc: "Gadiyar, Anand" , "felipe.balbi@nokia.com" , "linux-omap@vger.kernel.org" * Dmitry Krivoschekov [080625 17:46]: > apologies for being late with my comment, and it might be not directly > related to the initial patch but, > > Gadiyar, Anand wrote: > >>>> Fix hdrc_cnf.h for omap3 configuration. > >>> Does this break the configuration for omap2? > >>> > >>> We should set these dynamically if different for omap2 and omap3 > >>> so we can compile both into the same kernel. > >> Should check omap2430 trm to be sure, maybe someone in TI > >> could help us with it ? > > > > It won't break the 2430. The 2430 has 32 endpoints as well. > to be more precise, OMAP2430 has 1 control + 15 TX + 15 RX endpoints, > i.e.total of 31 eps. > > To me it's not clear if these TX/RX endpoints is really independent. In > other words, is it > really possible to schedule for example 1st TX and 1st RX endpoint at > the same time (in the same USB frame)? > > If you look at Felipe's patch, it changes a total number of eps from 8 > to 16. In my understanding the driver is now able to schedule only this > number of endpoints. That is, the driver consider both 1TX and 1RX > endpoints as the same endpoint that is able to operate for Tx and Rx. > And, the driver can use only 16 endpoints at the same time, not 31. > > But if all those 15Tx and 15Rx eps are mutually independent then it > makes sense to treat them in another way, MUSB_C_NUM_EPS should be 31 > plus appropriate modifications to the driver are needed. > > > Of course I may be missing something, please correct me if I'm wrong > with my conclusions. Let's wait on this until Felipe has the patch for getting the endpoint configuration from board-*.c files. Tony > > > Thanks, > Dmitry > > > > > >> Maybe we could bring it via platform_data as well? > > > > Yep. That would be good. > > > > - Anand > > -- > > To unsubscribe from this list: send the line "unsubscribe linux-omap" in > > the body of a message to majordomo@vger.kernel.org > > More majordomo info at http://vger.kernel.org/majordomo-info.html > > >