Glibc make check errors

Dan Nicholson dbn.lists at gmail.com
Tue Jun 19 07:15:16 PDT 2007


On 6/19/07, M Sandell <mikes at comp.leeds.ac.uk> wrote:
>
> I'm building LFS 6.2 on an AMD-K6 (500MHz, 56MB ram) using the LiveCD as
> host.  Just wanted to ask the group's opinion on some make check errors
> from Section 6.9, Glibc.  (Haven't found anything useful in
> archives/google.)
>
> Given the old hardware, the math failures (float, ifloat, ildoubl, ldoubl)
> are expected(?), and of course the annexc failure.
>
> However, I have an nptl/ failure different to the ones mentioned in the
> book (mutex5).  I also have two posix/ failures other than annexc (chmod
> and nanosleep), and I also have two failures (timer4 and timer5) from rt/.
>
> Given the hardware, are these failures acceptable and I'm ok to proceed?
> If not, any tips about resolving these issues?

Disclaimer: I don't actually understand any of the glibc tests, so I
don't really know what's failing besides looking at the name. I would
probably just move on. Thing's will be functional.

One thing you can do is look at the output files from the tests to see
if there's anything that really stands out. For instance, there should
be a nptl/mutex5.out file that has the log from the test.

If you're interested in resolving some test failures, you may want to
try using different compiler flags, especially -march. That can also
lead to more failures, though.

--
Dan



More information about the lfs-support mailing list