From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from eggs.gnu.org ([2001:4830:134:3::10]:38593) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1as5H7-0002TM-GF for qemu-devel@nongnu.org; Mon, 18 Apr 2016 05:15:26 -0400 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1as5H3-0004u8-D1 for qemu-devel@nongnu.org; Mon, 18 Apr 2016 05:15:21 -0400 Received: from mail-wm0-x229.google.com ([2a00:1450:400c:c09::229]:38245) by eggs.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1as5H3-0004u2-03 for qemu-devel@nongnu.org; Mon, 18 Apr 2016 05:15:17 -0400 Received: by mail-wm0-x229.google.com with SMTP id u206so113962101wme.1 for ; Mon, 18 Apr 2016 02:15:16 -0700 (PDT) References: <1460735821-12775-1-git-send-email-alex.bennee@linaro.org> <1460735821-12775-3-git-send-email-alex.bennee@linaro.org> From: Alex =?utf-8?Q?Benn=C3=A9e?= In-reply-to: Date: Mon, 18 Apr 2016 10:15:33 +0100 Message-ID: <87lh4bwa4q.fsf@linaro.org> MIME-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: 8bit Subject: Re: [Qemu-devel] [PATCH 2/4] .travis.yml: reduce target list on core configure tweaks List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: Peter Maydell Cc: QEMU Developers Peter Maydell writes: > On 15 April 2016 at 16:56, Alex Bennée wrote: >> A number of configure options only really affect the core code and any >> arch specific stuff should be flushed out by other builds: >> >> - trace-backends, log build is all targets, others can be less >> - --disable-build, ensuring disabling stuff doesn't break host >> - co-routine, default already built with all, common API >> - tcg interpreter is front end agnostic >> >> I've used the ./configure --target-list stem support to do all of one >> arch type in each reduced build. >> >> Signed-off-by: Alex Bennée >> --- >> .travis.yml | 18 +++++++++--------- >> 1 file changed, 9 insertions(+), 9 deletions(-) >> >> diff --git a/.travis.yml b/.travis.yml >> index 50ac17f..d13bffc 100644 >> --- a/.travis.yml >> +++ b/.travis.yml >> @@ -45,10 +45,10 @@ env: >> matrix: >> - CONFIG="" >> - CONFIG="--enable-debug --enable-debug-tcg --enable-trace-backends=log" >> - - CONFIG="--disable-linux-aio --disable-cap-ng --disable-attr --disable-brlapi --disable-uuid --disable-libusb" >> - - CONFIG="--enable-modules" >> - - CONFIG="--with-coroutine=ucontext" >> - - CONFIG="--with-coroutine=sigaltstack" >> + - CONFIG="--disable-linux-aio --disable-cap-ng --disable-attr --disable-brlapi --disable-uuid --disable-libusb --target-list=86" >> + - CONFIG="--enable-modules --target-list=86" > > I'm not really convinced by the wildcarding, and this seems like > a good demonstration of why it's overbroad. "86" ?? Blame the x86 naming scheme for that. It works nicely for building all arm, mips and ppc stuff quite nicely. > The wildcarding gives you a configure command that will > potentially change behaviour as we introduce new targets, > too; that makes me a bit nervous. Well it doesn't stop you using the full names but I take your point. What about a different selection flag for selecting by backends? > > thanks > -- PMM -- Alex Bennée