From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from eggs.gnu.org ([2001:4830:134:3::10]:50892) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1eo4Ws-0006Iy-VD for guix-patches@gnu.org; Tue, 20 Feb 2018 04:48:08 -0500 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1eo4Wo-0003CG-TV for guix-patches@gnu.org; Tue, 20 Feb 2018 04:48:07 -0500 Received: from debbugs.gnu.org ([208.118.235.43]:44233) by eggs.gnu.org with esmtps (TLS1.0:RSA_AES_128_CBC_SHA1:16) (Exim 4.71) (envelope-from ) id 1eo4Wo-0003C3-OR for guix-patches@gnu.org; Tue, 20 Feb 2018 04:48:02 -0500 Received: from Debian-debbugs by debbugs.gnu.org with local (Exim 4.84_2) (envelope-from ) id 1eo4Wo-00083J-Hq for guix-patches@gnu.org; Tue, 20 Feb 2018 04:48:02 -0500 Subject: [bug#30411] [PATCH] gnu: ldc: Update to 1.7.0. Resent-Message-ID: Date: Tue, 20 Feb 2018 10:47:20 +0100 From: Danny Milosavljevic Message-ID: <20180220104720.2cc49871@scratchpost.org> In-Reply-To: <20180218214110.GB9488@thebird.nl> References: <20180210141208.GA13051@thebird.nl> <20180218060944.7608-1-pjotr.public12@thebird.nl> <20180218133449.0273e0b3@scratchpost.org> <20180218214110.GB9488@thebird.nl> MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: guix-patches-bounces+kyle=kyleam.com@gnu.org Sender: "Guix-patches" To: Pjotr Prins , 30411@debbugs.gnu.org Hi Pjotr, On Sun, 18 Feb 2018 22:41:10 +0100 Pjotr Prins wrote: > Danny, do you mind fixing. I am on the road and would like to get it > into the new release. Sure, no problem. But building ldc 0.17.4 I get phobos datetime test failures. Not sure why - because as you've written it, ctest shouldn't have tested phobos. Also, ldc 1.7 does use phobos. I determined this as follows: I removed phobos-src from ldc-bootstrap, then used the resulting ldc-bootstrap to build ldc 1.7. It failed. I tried to excise phobos users from it but gave up. (Even ldc 1.1 used std.datetime during building) So I reverted the "check" phase. Also, ldc 1.7.0 test failure on the X200 laptop: sh: line 1: 16660 Illegal instruction /tmp/guix-build-ldc-1.7.0.drv-0/build/dmd-testsuite-debug/runnable/test_cdvecfill_1 > /tmp/guix-build-ldc-1.7.0.drv-0/build/dmd-testsuite-debug/1d696d6e3478bf73e2aaef0f8b23bba98e05e3cd74f80b02330a43bd3d2595a3 2>&1 ... runnable/test_cdvecfill.d -O (-mcpu=avx -mcpu=avx2) -conf=/tmp/guix-build-ldc-1.7.0.drv-0/build/bin/ldc2.conf -g -link-debuglib Test failed. The logged output: /tmp/guix-build-ldc-1.7.0.drv-0/build/bin/ldmd2 -conf= -m64 -Irunnable -O -od/tmp/guix-build-ldc-1.7.0.drv-0/build/dmd-testsuite-debug/runnable -of/tmp/guix-build-ldc-1.7.0.drv-0/build/dmd-testsuite-debug/runnable/test_cdvecfill_0 runnable/test_cdvecfill.d /tmp/guix-build-ldc-1.7.0.drv-0/build/dmd-testsuite-debug/runnable/test_cdvecfill_0 /tmp/guix-build-ldc-1.7.0.drv-0/build/bin/ldmd2 -conf= -m64 -Irunnable -O -mcpu=avx -od/tmp/guix-build-ldc-1.7.0.drv-0/build/dmd-testsuite-debug/runnable -of/tmp/guix-build-ldc-1.7.0.drv-0/build/dmd-testsuite-debug/runnable/test_cdvecfill_1 runnable/test_cdvecfill.d /tmp/guix-build-ldc-1.7.0.drv-0/build/dmd-testsuite-debug/runnable/test_cdvecfill_1 Which is understandable because the X200 laptop doesn't have AVX. It's building another round now - let's see.