mirror of https://lore.kernel.org/lkml/
 help / color / mirror / Atom feed
* 2.6.0-test{2,4}, gcc, smp problem
@ 2003-09-03 15:37 David Chambers
  0 siblings, 0 replies; only message in thread
From: David Chambers @ 2003-09-03 15:37 UTC (permalink / raw)
  To: linux-kernel

Hi,

I have recently been trying the 2.6.0-test kernels on an up-to-date Red Hat 9 system (Dual Xeon, 2 GB RAM, Tyan 7505 motherboard).  Being lazy :-) I downloaded arjanv's source RPMs and built from those.  These RPMs apply 2 small patches, one framebuffer related and one implements a noninteractive "make oldconfig", otherwise the 2.6.0 source is untouched.

Compilation of the initial 2.6.0-test2 with gcc 3.2 (Red Hat released version) (while running 2.4.21) went perfectly.  I booted the new 2.6.0-test2 and I have to say I never want to go back to 2.4!  Desktop responsiveness is **much** improved - superb job, guys!.

However...  When I tried to compile 2.6.0-test2 again, (running under 2.6.0-test2) I get irreproducible internal compiler errors or segmentation faults.  I have tracked this down to the use of "make -j4" during the build.  If I disable the multiple job compilation, everything works.  If I so much as use "make -j2" I get the errors again.

Same thing happens with 2.6.0-test4.  I tried upgrading gcc to 3.3.1 (Rawhide rpm) and this has made no difference.  I have not yet tried this with anything but the smp kernel, btw.

So... What am I looking at here?  Problem with hardware? kernel? gcc?  Any comments and suggestions gratefully received!!

- David Chambers

^ permalink raw reply	[flat|nested] only message in thread

only message in thread, other threads:[~2003-09-03 15:37 UTC | newest]

Thread overview: (only message) (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2003-09-03 15:37 2.6.0-test{2,4}, gcc, smp problem David Chambers

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox

all inboxes | Powered by JetHome®