From mboxrd@z Thu Jan 1 00:00:00 1970 From: Ricardo Wurmus Subject: Re: Status update+patches:Re: I managed to build guix natively on Debian GNU/Hurd , what's next? Date: Tue, 03 Sep 2019 14:58:30 +0200 Message-ID: <875zm9h6yx.fsf@elephly.net> References: <8736hjh7c7.fsf@elephly.net> <5c5233873e372138e82f5d52225e287ee1b86a38.camel@gmail.com> <53512769-C494-42BC-91C2-83B3F1586512@lepiller.eu> <9888a399c79968627aeeebe05419304eca155fd2.camel@gmail.com> <87k1aufiie.fsf@elephly.net> <18bb81c74e24323fb974bc5126a1b94094b3be71.camel@gmail.com> <87ftlifgcm.fsf@elephly.net> <32d67ecf3f8e43cd8d4607287d4c4b57380f47ef.camel@gmail.com> <87ef11fs9i.fsf@elephly.net> <30bdaa745f0a41617e0e81fc6bc63b1228dab729.camel@gmail.com> <60498ff6153c81ffa586c9088e35f2e403270827.camel@gmail.com> <45800d03ca1fab9f80bae932a9354f60f6c12a2d.camel@gmail.com> <5f5d2b063290517ff9619cf59398b700998b5a2b.camel@gmail.com> Mime-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: quoted-printable Return-path: Received: from eggs.gnu.org ([2001:470:142:3::10]:45520) by lists.gnu.org with esmtp (Exim 4.90_1) (envelope-from ) id 1i58OO-00056y-Am for guix-devel@gnu.org; Tue, 03 Sep 2019 08:58:41 -0400 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1i58ON-00071u-5u for guix-devel@gnu.org; Tue, 03 Sep 2019 08:58:40 -0400 Received: from sender-of-o51.zoho.com ([135.84.80.216]:21223) by eggs.gnu.org with esmtps (TLS1.0:RSA_AES_256_CBC_SHA1:32) (Exim 4.71) (envelope-from ) id 1i58OM-00071S-Rk for guix-devel@gnu.org; Tue, 03 Sep 2019 08:58:39 -0400 In-reply-to: <5f5d2b063290517ff9619cf59398b700998b5a2b.camel@gmail.com> List-Id: "Development of GNU Guix and the GNU System distribution." List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: guix-devel-bounces+gcggd-guix-devel=m.gmane.org@gnu.org Sender: "Guix-devel" To: svante.signell@gmail.com Cc: guix-devel Hi Svante, > Then I continued with as Richardo wrote: > ./pre-inst-env guix build -e '(@@ (gnu packages commencement) gnu-make-b= oot0))'=20 > 2>&1 | tee ../pre-inst-env_guix-build-e-no-chroot_new-hash.log > The build went a lot further, but failed here: >=20=20 > successfully built /gnu/store/6sybpxd8hy07h4fqf4wyblxrsi17r4zn-gcc-bootst= rap- > 0.drv > building /gnu/store/h0y1izzm1nxr3gqwagg84dnssa723sby-make-4.2.1.tar.xz.dr= v... > builder for `/gnu/store/h0y1izzm1nxr3gqwagg84dnssa723sby-make-4.2.1.tar.x= z.drv' > failed with exit code 1 > build of /gnu/store/h0y1izzm1nxr3gqwagg84dnssa723sby-make-4.2.1.tar.xz.drv > failed > View build log at '/usr/var/log/guix/drvs/h0/y1izzm1nxr3gqwagg84dnssa723s= by- > make-4.2.1.tar.xz.drv.gz'. > cannot build derivation `/gnu/store/j18911vbapx9kx0gqcyn6nkd3bwz52lj-make= -boot0- > 4.2.1.drv': 1 dependencies couldn't be built > guix build: error: build of `/gnu/store/j18911vbapx9kx0gqcyn6nkd3bwz52lj-= make- > boot0-4.2.1.drv' failed This looks very much like the tar segfault I encountered when I last played with the Hurd port. My notes say this: --8<---------------cut here---------------start------------->8--- - tar fails to run when building gnu-make-boot0 ./pre-inst-env guix build -e '(@@ (gnu packages commencement) gnu-make-boo= t0))' here=E2=80=99s the error: building /gnu/store/ybi3cyxqwm55gvwn9i1nczg2ns5ha34q-make-4.2.1.tar.xz.drv.= ..=20=20=20=20=20=20=20=20=20=20=20=20=20=20=20=20=20=20=20 /hurd/crash: /gnu/store/dqlhjyvg0n8v1kdvwfpliqy46n7kpjqb-bootstrap-binaries= -0/bin/tar cvfa /gnu/store/p278mqb3aa0xrkrrw4rg1fxbb19hbdyh-make-4.2.1.tar.= xz --mtime=3D@0 --owner=3Droot --group=3Droot --sort=3Dname make-4.2.1(2316= ) crashed, signal {no:1 1, code:8176, error:2}, exception {1, code:2, subco= de:8176}, PCs: {0x8089ce7, 0x 815091c}, writing core file.=20=20=20=20=20= =20=20=20=20=20=20=20=20=20=20=20=20=20=20=20=20=20 This works: /gnu/store/dqlhjyvg0n8v1kdvwfpliqy46n7kpjqb-bootstrap-binaries-0/bin/tar cv= fa hello.tar.xz /tmp But it segfaults with =E2=80=9C--mtime=3D@0=E2=80=9D --8<---------------cut here---------------end--------------->8--- > The cross-built xz is definitely buggy, I don't know about the other file= s at: > gnu/packages/bootstrap/i586-gnu: bash mkdir tar xz > > Maybe there are other buggy cross-built binaries. Quite likely! > A question: The gcc used is very old: 5.5.0, is i possible that guix could > upgrade to a newer version, preferrably later than gcc-7. How would that help? Or is it just a guess that a later GCC would be better here? On the =E2=80=9Ccore-updates=E2=80=9D branch we are using a later GCC alrea= dy. --=20 Ricardo