[Ask for help]: Met a problem with strtof

[Joseph S. Myers]

On Sat, 26 Apr 2014, Wang Weidong wrote:

> The syntax: Within FE_UPWARD mode, the operands is 32bits , do 
> `+0x1.4EF009P-21/-0x1.000000P-5`, the expect result is '-0x1.4EF009P-16'.

But +0x1.4EF009P-21 and -0x1.4EF009P-16 aren't representable in binary32 
(their mantissas span 25 bits), so such a test makes no sense at all.  
Your problem seems to be with this third-party testsuite, not with glibc.

-- 
Joseph S. Myers
joseph@codesourcery.com