From mboxrd@z Thu Jan 1 00:00:00 1970 From: Ben Warren Date: Thu, 27 Dec 2007 11:27:50 -0500 Subject: [U-Boot-Users] Pull request: u-boot-freebsd In-Reply-To: <4773930B.9070903@semihalf.com> References: <20071227000534.69AA024889@gemini.denx.de> <4773930B.9070903@semihalf.com> Message-ID: <4773D286.8080405@qstreams.com> List-Id: MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit To: u-boot@lists.denx.de Hi Rafal, Rafal Jaworowski wrote: > Dear Wolfgang, > you wrote: > >> I'm afraid I have to reject this pull request. >> >> First, I think it breaks Linux booting support for a couple of boards >> (all 4xx ones, that is). >> >> >>> The following changes since commit 41be969f4957115ed7b1fe8b890bfaee99d7a7a2: >>> Wolfgang Denk (1): >>> Release v1.3.1 >>> >>> are available in the git repository at: >>> >> ... >> >>> [POWERPC] Simplify bd_info struct >>> >> This is the culprit. >> >> With your patch, you *always* include an entry "unsigned long >> bi_bar;" in the bd_infor structure - even for processors which didn't >> have one before (like 4xx boards). Thus you change the layout of the >> bd_info structure for such systems, which causes an incompatibility >> with the respective structure used by the Linux kernel. >> >> Also, I have to admit that I dislike this type of #ifdef based >> "simplification". I don't think the resulting code becomes more >> readable. >> >> > > Mhm, right. Maybe it was too much of an improvement :) Given all the legacy > dependencies between bd_info and Linux it's probably better to just leave it > as is and deal with those #idef'ed fields locally (where they are accessed). > > >> Please fix and resubmit. >> >> > > Will do. > > >> Then, we have this part: >> >> >>> [Net] Introduce standalone eth_receive() routine >>> >> This adds a lot of code to the networking code which is not neede dby >> most of the boards. Please make this an optional feature that get's >> only compiled in for boards that explicitely request it. Then run this >> patch separately through the network custodian. >> >> > > I already sent this patch to Ben and the list some time ago and got an initial > ACK, but I'll re-spin with. > > When you send the re-spun patch I'll be sure to pull it in. regards, Ben