glibc 2.1.94 testing on alpha

[Andreas Jaeger]

>>>>> Christian Iseli writes:

 > aj@suse.de said:
>> Let's fix __printf_fphex () first - and then send the ulps file. 

 > Ok, I've recompiled with "-O2 -g -mcpu=ev56".  It still doesn't SEGFAULT, but 
 > the output is "interesting"...  For example, from test-float:

 > testing float (without inline functions)
 > Failure: Test: frexp (12.8, &x) == 0.8
 > Result:
 >  is:          1.28000001907348632812e+01   0x1.99999a0000000Äùp^B^@^@^@p+3
 >  should be:   8.00000011920928955078e-01   0x1.99999a0000000Äùp^B^@^@^@p-1
 >  difference:  1.20000000000000000000e+01   0x1.8000000000000Äùp^B^@^@^@p+3
 >  ulp       :  201326592.0000
 >  max.ulp   :  0.0000
 > ... (more similar failures...)

 > What do you think ?

Sorry, I don't have time to check the printf_fphex routine - and don't
know enough about the alpha to make a good guess.

Alpha developers, please have a look at this!

Andreas
-- 
 Andreas Jaeger
  SuSE Labs aj@suse.de
   private aj@arthur.inka.de
    http://www.suse.de/~aj