This is the mail archive of the libc-alpha@sourceware.org mailing list for the glibc project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

Re: [PATCH] Add a new macro to mask a float


On Wed, 29 Jun 2016, Adhemerval Zanella wrote:

> Now that you raised these questioning I do not think this change is safe
> for float values in POWER.  Current patch does:

In that case we get back to the question of whether you can show that each 
individual instance is in fact safe because the code cannot get subnormals 
or because the resulting masking is in fact safe when it does get 
subnormals.  In which case changes in glibc would make sense instead of 
compiler changes (as the compiler probably wouldn't be able to tell the 
code sequence is safe in a particular case), given performance evidence, 
appropriate comments explaining the rules for when the macro can be used, 
coding standards fixes, etc.

-- 
Joseph S. Myers
joseph@codesourcery.com


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]