extra error from gcc-2.3.1 tests (typo ... gcc-3.3.1)

Allard allard at nospam.nl
Mon Nov 3 15:48:38 PST 2003


On Tuesday 04 November 2003 00:27, Greg Schafer wrote:
> On Mon, Nov 03, 2003 at 10:40:40PM +0100, Allard wrote:
> > The failed gcc.c-torture/unsorted/set386.c, -O0 wasn't there after pass2
> > of chapter 5. Do I need to worry? Since I never really compile with -O0
> > I'd like to ignore this, on the other hand ... the test in question
> > caused a kernel oops pasted below.
>
> I wouldn't worry about the failed test. But I would worry about the kernel
> oops, but that is completely orthagonal to the test failure. Any time your
> kernel oop's, you should be concerned. If it's repeatable and you care
> enough, consider sending a bug report to the makers of the host distro (if
> it's a distro kernel) or to LKML. Hopefully, once you boot into your
> new LFS with an LFS compiled kernel, yuo won't have any problems.

My host distro is LFS-4.1 with some modifications I made which I wasn't happy 
with afterwards (linux-2.4.22, gcc-3.2.3 and glibc-2.3.2), hence my move to 
LFS-5. I had problems getting the glibc to compile. The problem will more 
than likely be there somewhere. Once the base LFS is finished and rebooted I 
will redo the gcc compile. I happened to be watching the compile output 
(eating dinner) - the oops happened exactly when the offending test failed.

Appreciate the advice,
Allard

-- 
allard at quicknet.nl



More information about the lfs-support mailing list