[PATCH] Some sparc{,v9,64} glibc fixes

[Andreas Jaeger]

Jakub Jelinek <jakub@redhat.com> writes:

> On Thu, Dec 16, 2004 at 07:47:59PM +0100, Andreas Jaeger wrote:
>> > BTW, Andreas, can you please recompute:
>> > Failure: Test: atan2 (-0.00756827042671106339, -.001792735857538728036) == -1.80338464113663849327153994380
>> > Result:
>> >  is:         -1.80338464113663849327e+00  -0x1.cdaa9da7ca620c5349dcp+0
>> >  should be:  -1.80338464113663849327e+00  -0x1.cdaa9da7ca620c5349dcp+0
>> >  difference:  3.60900012278724126504e-30   0x1.24cc0000000000000000p-98
>> >  ulp       :  18739.0000
>> >  max.ulp   :  0.0000
>> > with additional precision?  This is something I haven't included in
>> > libm-test-ulps patch, as that is not atan2 ldbl-128 bug, but testcase.
>> > 30 digits aren't enough, all other numbers have at least 36.  Thanks.
>> 
>> the result with 36 digits is:
>> -1.80338464113663849327153994379639112
>> 
>> Ok to commit?
>> 
>> Andreas
>> 
>> 2004-12-16  Andreas Jaeger  <aj@suse.de>
>> 
>> 	* math/libm-test.inc (atan2_test): Compute value with 36 digits.
>
> Thanks.  If you also do
> find sysdeps -name libm-test-ulps | xargs sed -i s/1.80338464113663849327153994380/1.80338464113663849327153994379639112/
> then I'd appreciate that (verified this number only appears in this exact
> atan2 test).

Will do now - thanks for the handy command,

Andreas
-- 
 Andreas Jaeger, aj@suse.de, http://www.suse.de/~aj
  SUSE Linux Products GmbH, Maxfeldstr. 5, 90409 Nürnberg, Germany
   GPG fingerprint = 93A3 365E CE47 B889 DF7F  FED1 389A 563C C272 A126
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 188 bytes
Desc: not available
URL: <http://sourceware.org/pipermail/libc-hacker/attachments/20041217/795a77bf/attachment.sig>