Re: make check failures on Alpha

[Andreas Jaeger <aj at suse dot de>]

mru@kth.se (Måns Rullgård) writes:

> [...]
> math/tst-float:
> testing float (without inline functions)
> Failure: Test: atan2 (-0.00756827042671106339, -.001792735857538728036) == -1.80
> 338464113663849327153994380
> Result:
>  is:         -1.80338394641876220703e+00  -0x1.cdaa9200000000000000p+0
>  should be:  -1.80338466167449951172e+00  -0x1.cdaa9e00000000000000p+0
>  difference:  7.15255737304687500000e-07   0x1.80000000000000000000p-21
>  ulp       :  6.0000
>  max.ulp   :  0.0000
> Maximal error of `atan2'
>  is      : 6 ulp
>  accepted: 3 ulp
>
> Test suite completed:
>   2524 test cases plus 2313 tests for exception flags executed.
>   2 errors occurred.
>
> math/tst-ifloat: likewise

Read math/README.libm-test and send a new ULPs file...

>
> nptl/tst-cancel4:
> Timed out: killed the child process
> last line: early cancel test of 'accept' successful
>
> nptl/tst-cancel5: likewise
>
> nptl/tst-cancel19:
> pthread_setcanceltype after close returned invalid value 1
> pthread_setcanceltype after 2nd close returned invalid value 0
> pthread_setcanceltype after 2nd pause returned invalid value 0
> pthread_setcanceltype after 2nd creat returned invalid value 0
> pthread_setcanceltype after 2nd open returned invalid value 0
> pthread_setcanceltype after select returned invalid value 1
>
> nptl/tst-cancel4x:
> Timed out: killed the child process
> last line: early cancel test of 'accept' successful
>
> nptl/tst-cancel5x: likewise
>
> All other tests passed.
>
> Tools used:
> gcc 3.3.3
> binutils 2.15.90

Linux kernel version?  Kernel headers used?

Alpha is currently not in an as good shape as x86 and AMD64, you have
to do more handwork if you use Alpha...

Andreas
-- 
 Andreas Jaeger, aj@suse.de, http://www.suse.de/~aj
  SuSE Linux AG, Maxfeldstr. 5, 90409 Nürnberg, Germany
   GPG fingerprint = 93A3 365E CE47 B889 DF7F  FED1 389A 563C C272 A126

Attachment: pgp00000.pgp
Description: PGP signature