From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Date: Fri, 13 Dec 2013 08:19:21 +1100 From: Paul Mackerras To: Tom Musta Subject: Re: [V2 PATCH 3/3] powerpc: Fix Unaligned LE Floating Point Loads and Stores Message-ID: <20131212211921.GA22557@iris.ozlabs.ibm.com> References: <1383244738-5986-1-git-send-email-tommusta@gmail.com> <1383244738-5986-4-git-send-email-tommusta@gmail.com> <20131211035440.GA9399@drongo> <20131211045728.GB9399@drongo> <52A9D17E.3000809@gmail.com> <52AA1DA0.80003@gmail.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii In-Reply-To: <52AA1DA0.80003@gmail.com> Cc: linuxppc-dev@lists.ozlabs.org List-Id: Linux on PowerPC Developers Mail List List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , On Thu, Dec 12, 2013 at 02:33:36PM -0600, Tom Musta wrote: > On 12/12/2013 9:08 AM, Tom Musta wrote: > > On 12/10/2013 10:57 PM, Paul Mackerras wrote: > >> On Wed, Dec 11, 2013 at 02:54:40PM +1100, Paul Mackerras wrote: > > > >>> This breaks 32-bit big-endian (as well as making the code longer and > >>> more complex). > >> > >> And in fact none of this code will get executed in little-endian mode > >> anyway, since we still have this in the middle of emulate_step(): > >> > >> /* > >> * Following cases are for loads and stores, so bail out > >> * if we're in little-endian mode. > >> */ > >> if (regs->msr & MSR_LE) > >> return 0; > >> > >> Paul. > >> > > > > See patch 1/3 to explain how it becomes relevant in LE. > > > > I will take another look at the change. > > > > It appears that patch 1/3 never got picked up, even though I thought Ben & I > had worked through that. > > And I agree that the code could be simpler. I will work up a patch to address > these two issues. The other thing that's important for us to know is how you are testing these changes. For something like this I'd like to see a description of the tests you have done in the commit message. I have been hacking on sstep.c pretty heavily myself recently, so we will need to coordinate on the changes. Paul.