From mboxrd@z Thu Jan 1 00:00:00 1970 From: Nicholas Piggin Subject: Re: [PATCH] kbuild: provide THIN_ARCHIVES option for all architectures Date: Mon, 5 Jun 2017 17:00:45 +1000 Message-ID: <20170605170045.11760813@roar.ozlabs.ibm.com> References: <20170529081103.29999-1-npiggin@gmail.com> Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Return-path: In-Reply-To: Sender: linux-kbuild-owner@vger.kernel.org To: Masahiro Yamada Cc: Arnd Bergmann , linux-arch , Linux Kbuild mailing list , Linus Torvalds , Stephen Rothwell List-Id: linux-arch.vger.kernel.org On Mon, 5 Jun 2017 08:17:26 +0900 Masahiro Yamada wrote: > Hi. > > 2017-06-01 6:13 GMT+09:00 Arnd Bergmann : > > On Mon, May 29, 2017 at 10:11 AM, Nicholas Piggin wrote: > >> Supporting two different intermediate-artifact packaging schemes > >> was only ever intended as a temporary transition. > >> > >> This has so far caused no problems for powerpc, after a small fix > >> for how the arch invoked ar. So now allow any arch to select the > >> option, continue defaulting to N. > >> > >> Cc: Stephen Rothwell > >> Signed-off-by: Nicholas Piggin > >> --- > >> The next step will be to have archs always select THIN_ARCHIVES > >> when they are known to work. Then remove the option entirely. > >> > >> x86 has always just worked for me, so that should be easy. > > > > I have build-tested many thousand randconfig kernels on arm32 with > > this option enabled, and did not run into build-time regressions > > besides some initial problems from a broken binutils snapshot > > (all released binutils versions should be fine). > > > > > Please let me mention two advantages of using THIN_ARCHIVES. [snip] Thank you for the post. I ran it through 0day and found some issues with a few archs that I'm working through. Some changes are needed to the generic thin archives build scripts, and some arch tweaks required. So it's not 100% trivial. 4.13 might still be a realistic target if I make some progress. Thanks, Nick From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mail-pf0-f194.google.com ([209.85.192.194]:35574 "EHLO mail-pf0-f194.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751274AbdFEHA5 (ORCPT ); Mon, 5 Jun 2017 03:00:57 -0400 Date: Mon, 5 Jun 2017 17:00:45 +1000 From: Nicholas Piggin Subject: Re: [PATCH] kbuild: provide THIN_ARCHIVES option for all architectures Message-ID: <20170605170045.11760813@roar.ozlabs.ibm.com> In-Reply-To: References: <20170529081103.29999-1-npiggin@gmail.com> MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: linux-arch-owner@vger.kernel.org List-ID: To: Masahiro Yamada Cc: Arnd Bergmann , linux-arch , Linux Kbuild mailing list , Linus Torvalds , Stephen Rothwell Message-ID: <20170605070045.wZ8oi-2tjceNfwdC_q29aqXaCmW5szMpQ_pcRPsrM_8@z> On Mon, 5 Jun 2017 08:17:26 +0900 Masahiro Yamada wrote: > Hi. > > 2017-06-01 6:13 GMT+09:00 Arnd Bergmann : > > On Mon, May 29, 2017 at 10:11 AM, Nicholas Piggin wrote: > >> Supporting two different intermediate-artifact packaging schemes > >> was only ever intended as a temporary transition. > >> > >> This has so far caused no problems for powerpc, after a small fix > >> for how the arch invoked ar. So now allow any arch to select the > >> option, continue defaulting to N. > >> > >> Cc: Stephen Rothwell > >> Signed-off-by: Nicholas Piggin > >> --- > >> The next step will be to have archs always select THIN_ARCHIVES > >> when they are known to work. Then remove the option entirely. > >> > >> x86 has always just worked for me, so that should be easy. > > > > I have build-tested many thousand randconfig kernels on arm32 with > > this option enabled, and did not run into build-time regressions > > besides some initial problems from a broken binutils snapshot > > (all released binutils versions should be fine). > > > > > Please let me mention two advantages of using THIN_ARCHIVES. [snip] Thank you for the post. I ran it through 0day and found some issues with a few archs that I'm working through. Some changes are needed to the generic thin archives build scripts, and some arch tweaks required. So it's not 100% trivial. 4.13 might still be a realistic target if I make some progress. Thanks, Nick