From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Date: Mon, 18 Sep 2006 17:45:03 -0700 From: "Mark A. Greer" To: Paul Mackerras Subject: Re: [PATCH 1/6] bootwrapper: arch/powerpc/boot code reorg Message-ID: <20060919004503.GA920@mag.az.mvista.com> References: <20060719230014.GB3887@mag.az.mvista.com> <20060908033508.GA5203@mag.az.mvista.com> <17674.32333.958250.868794@cargo.ozlabs.ibm.com> <20060915180129.GE1884@mag.az.mvista.com> <17675.30508.252375.59449@cargo.ozlabs.ibm.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii In-Reply-To: <17675.30508.252375.59449@cargo.ozlabs.ibm.com> Cc: linuxppc-dev List-Id: Linux on PowerPC Developers Mail List List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , On Sat, Sep 16, 2006 at 02:01:48PM +1000, Paul Mackerras wrote: Hi Paul, > Mark A. Greer writes: > > > Yes. Whether you pass the NULL or not hinges on whether its flat tree > > or not so I made it a dt_op. What are you thinking? > > There's basically only two cases: if we have a builtin flat tree, we > pass that and NULL in r5; if we don't, and we have OF, we pass NULL in > r3 and the OF entry point in r5. I think that could be handled in > main.c. Okay. > We also lost the "enforce minimum address only for 64-bit" thing in > the translation, too. You mean the "if (claim_base < PROG_START)" if stmt, right? Would adding an "#if defined(__powerpc64__) && defined(PROG_START)" be sufficient? I should change simple_alloc to _ALIGN_UP only 4K while I'm at it. > I believe that just patch 1/6 on its own won't give us a working > zImage, so we'll have to split up the changes differently when it > comes time to check this stuff in to git. You're right. I'll fix that up. Mark