From mboxrd@z Thu Jan 1 00:00:00 1970 From: Fabrice Fontaine Date: Mon, 19 Dec 2016 23:25:41 +0100 Subject: [Buildroot] [PATCH v3,3/3] soletta: new package In-Reply-To: <20161219231208.46c66492@free-electrons.com> References: <1482101300-9180-1-git-send-email-fabrice.fontaine@orange.com> <1482101300-9180-3-git-send-email-fabrice.fontaine@orange.com> <20161219231208.46c66492@free-electrons.com> Message-ID: List-Id: MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit To: buildroot@busybox.net Hello 2016-12-19 23:12 GMT+01:00 Thomas Petazzoni < thomas.petazzoni@free-electrons.com>: > Hello, > > On Sun, 18 Dec 2016 23:48:20 +0100, Fabrice Fontaine wrote: > > Signed-off-by: Fabrice Fontaine > > I'm sorry, but this still doesn't work with old make versions. I get: > > make[1]: Entering directory `/home/test/buildroot/output/build/soletta-v1' > Makefile.smallos:35: "LDFLAGS not set. This is probably an error" > Makefile.smallos:57: *** multiple target patterns. Stop. > make[1]: Leaving directory `/home/test/buildroot/output/build/soletta-v1' > > This is due to the := assignments in the target definitions. If I change: > > $(soletta_target): private export BOARD_NAME:=$(BOARD_NAME) > $(soletta_config): private export BOARD_NAME:=$(BOARD_NAME) > > to > > $(soletta_target): private export BOARD_NAME=$(BOARD_NAME) > $(soletta_config): private export BOARD_NAME=$(BOARD_NAME) > > then a different problem happens: the "private" keyword apparently > didn't exist in make 3.81, and therefore is interpreted as being a > target, and therefore: > > make[1]: Entering directory `/home/test/buildroot/output/build/soletta-v1' > Makefile.smallos:35: "LDFLAGS not set. This is probably an error" > make[1]: *** No rule to make target `private', needed by > `/home/test/buildroot/output/build/soletta-v1/.config'. Stop. > make[1]: Leaving directory `/home/test/buildroot/output/build/soletta-v1' > > So, really, this package that requires a brand new version of make, and > python3, is a bit annoying in terms of build process... Maybe this is > something you can work out with upstream? > > In the mean time, we cannot apply this, as it will badly fail on our > autobuilders, and for many of our users who run fairly old systems. > I perfectly understand your point, thanks a lot for the time you passed on testing these patches. > > Thanks, > > Thomas > -- > Thomas Petazzoni, CTO, Free Electrons > Embedded Linux and Kernel engineering > http://free-electrons.com > Best Regards, Fabrice -------------- next part -------------- An HTML attachment was scrubbed... URL: