Date: Fri, 06 Mar 2009 05:05:27 +0530 From: wahjava.ml@gmail.com (Ashish SHUKLA) To: John Baldwin <jhb@freebsd.org> Cc: freebsd-stable@freebsd.org, Mark Atkinson <atkin901@yahoo.com> Subject: Re: GCC segfaulting while trying to compile latest Qt4 code Message-ID: <87r61bbl1c.fsf@chateau.d.lf> In-Reply-To: <200903051449.59782.jhb@freebsd.org> (John Baldwin's message of "Thu, 5 Mar 2009 14:49:59 -0500") References: <87wsb5clk9.fsf@chateau.d.lf> <87iqmohcl5.fsf@chateau.d.lf> <goovod$bba$1@ger.gmane.org> <200903051449.59782.jhb@freebsd.org>
next in thread | previous in thread | raw e-mail | index | archive | help
--=-=-= Content-Transfer-Encoding: quoted-printable John Baldwin writes: > On Thursday 05 March 2009 11:48:11 am Mark Atkinson wrote: >> Ashish SHUKLA wrote: >>=20 >> > Daniel O'Connor writes: >> >> On Thursday 05 March 2009 02:44:14 Ashish SHUKLA wrote: >> >>> Hi everone, >> >>>=20 >> >>> While trying to update my 'qt4-gui' port, gcc segfaults during >> >>> compilation. I'm running FreeBSD 7.1-STABLE (amd64 architecture) upd= ated >> >>> on 3rd March, 2009. >> >=20 >> >> I would not rule out a hardware problem - seg faults while compiling >> >> heavy code (ie big C++ projects :) can be indicative of >> >> memory/CPU/PSU/motherboard problems. >> >=20 >> > Okay, but I even tried compiling that file manually and that also >> > resulted in segfault. >> >=20 >> >> I'd suggest running memtest86 and prime95 for a few hours if you can. >> >=20 >> > I'll try running memtest86. I've tried memtest86+ with two passes and no memory related errors are recieved. >> >=20 >> > Thanks >>=20 >> I believe superpages were mfc'd to -stable as well. I have one machine= =20 >> that faults in gcc with this turned on. You can try adding=20 >>=20 >> vm.pmap.pg_ps_enabled=3D"0" >>=20 >> to /boot/loader.conf, rebooting and see if that fixes the problem. If s= o,=20=20 >> you might provide the list with your hardware configuration. I booted with above line but nothing changed. Then I booted in to my old kernel (dated January 29, 2009) with new userland, but there also I received that segfault, so I guess this is something userspace related. BtW, I also tried compiling deskutils/google-gadgets 0.10.5[1] port and there too I've received similar segfault. I'm attaching the output of 'dmesg' booted in verbose logging mode in both the kernels. The boot.new.log contains output from booting in new kernel, whereas boot.old.log contains output from booting in old kernel. References: [1] http://www.freebsd.org/cgi/query-pr.cgi?pr=3D132291 HTH =2D-=20 Ashish SHUKLA --=-=-= Content-Type: application/pgp-signature -----BEGIN PGP SIGNATURE----- Version: GnuPG v2.0.10 (GNU/Linux) iEYEARECAAYFAkmwYcUACgkQHy+EEHYuXnQ+hwCfdHvE9suFl6MOKFhmxd9quIGz lNQAoOTyY1OR2FuiYhNskIIj6nzzo+9r =pVAL -----END PGP SIGNATURE----- --=-=-=--
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?87r61bbl1c.fsf>