Glibc testsuite errors

Ken Moffat ken at linuxfromscratch.org
Tue Feb 6 03:07:25 PST 2007


On Mon, Feb 05, 2007 at 06:38:24PM -0600, Randy McMurchy wrote:
> Hi all,
> 
> Just an FYI, but I had errors building Glibc in chapter 6 of the dev
> book, which I've not seen before. Doesn't seem to be a great big deal,
> but they are unusual for me on this system.
> 
> rml at rmlinux: /mnt/rml3/build/glibc-build > grep Error check.log
> 
> make[2]: *** [/build/glibc-build/dlfcn/bug-atexit3.out] Error 1
> make[1]: *** [dlfcn/tests] Error 2

 I've not seen that.
> make[2]: [/build/glibc-build/posix/annexc.out] Error 1 (ignored)
> make[2]: *** [/build/glibc-build/nptl/tst-cancel1.out] Error 1

 Common, at least for me.
> make[2]: *** [/build/glibc-build/nptl/tst-cancel24.out] Error 127
> make[1]: *** [nptl/tests] Error 2
> make: *** [check] Error 2
> 
 I assume that the config from the kernel running at that time had
CONFIGCOMPAT_VDSO not set (seems a fair bet, setting it normally
gives me several more errors).  Except in that context, I've not seen
this, but my experience on other architectures suggests some of the
nptl tests are fragile (easily upset by minor kernel changes).  If
you have the full logs, can you take a look at what it says just
before this, please - I think Dan is right that a program is missing.

 On last night's build against 2.6.20 (while running 2.6.20-rc6) I
only had errors from nptl/tst-cancel1.out and rt/tst-cpuclock2.out
(disregarding the ignored posix/annexc.out).  I think my rt/ error
is new with the 2.6..20 series.

ĸen
-- 
das eine Mal als Tragödie, das andere Mal als Farce



More information about the lfs-dev mailing list