From owner-freebsd-arm@freebsd.org Fri Feb 5 20:49:40 2016 Return-Path: Delivered-To: freebsd-arm@mailman.ysv.freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:1900:2254:206a::19:1]) by mailman.ysv.freebsd.org (Postfix) with ESMTP id BA819A9C7DC; Fri, 5 Feb 2016 20:49:40 +0000 (UTC) (envelope-from jhb@freebsd.org) Received: from bigwig.baldwin.cx (bigwig.baldwin.cx [IPv6:2001:470:1f11:75::1]) (using TLSv1 with cipher DHE-RSA-CAMELLIA256-SHA (256/256 bits)) (Client did not present a certificate) by mx1.freebsd.org (Postfix) with ESMTPS id 99DBE7F7; Fri, 5 Feb 2016 20:49:40 +0000 (UTC) (envelope-from jhb@freebsd.org) Received: from ralph.baldwin.cx (c-73-231-226-104.hsd1.ca.comcast.net [73.231.226.104]) by bigwig.baldwin.cx (Postfix) with ESMTPSA id 937ECB948; Fri, 5 Feb 2016 15:49:39 -0500 (EST) From: John Baldwin To: freebsd-ppc@freebsd.org Cc: Marcin Mazurek , freebsd-current@freebsd.org, freebsd-arm@freebsd.org, freebsd-mips@freebsd.org Subject: Re: Request for testing bus_get_bus_tag() nexus method Date: Fri, 05 Feb 2016 12:49:17 -0800 Message-ID: <1784059.mPpRYc4e0u@ralph.baldwin.cx> User-Agent: KMail/4.14.3 (FreeBSD/10.2-STABLE; KDE/4.14.3; amd64; ; ) In-Reply-To: References: MIME-Version: 1.0 Content-Transfer-Encoding: 7Bit Content-Type: text/plain; charset="us-ascii" X-Greylist: Sender succeeded SMTP AUTH, not delayed by milter-greylist-4.2.7 (bigwig.baldwin.cx); Fri, 05 Feb 2016 15:49:39 -0500 (EST) X-BeenThere: freebsd-arm@freebsd.org X-Mailman-Version: 2.1.20 Precedence: list List-Id: "Porting FreeBSD to ARM processors." List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 05 Feb 2016 20:49:40 -0000 On Friday, February 05, 2016 05:11:19 PM Marcin Mazurek wrote: > Hello, > > I am looking for testers for a patch to add BUS_GET_BUS_TAG method to some > platforms nexus that return per platform specific default tag. > > It works fine on arm, but I do not have any powerpc or mips hardware to > test it on, > so I would like it if this could be tested on these platforms using this > code to check > if it does not break them. > > Any comments and feedback are welcome. I'm guess you mean this for memory access? Have you thought about accepting the resource type so you can do: bus_get_bus_tag(dev, SYS_RES_MEMORY); vs bus_get_bus_tag(dev, SYS_RES_IOPORT); I would also be inclined to call it 'bus_get_bus_space_tag()' since it is returning a bus_space tag. "bus tag" might too generic of a name. In general though it seems to be a workaround for not wanting to allocate an actual resource with bus_alloc_resource() and then using bus_activate_resource() to obtain a valid (tag, handle) tuple? -- John Baldwin