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 1/3] New generic sincosf


On Fri, Dec 15, 2017 at 6:18 AM, Joseph Myers <joseph@codesourcery.com> wrote:
> On Fri, 15 Dec 2017, H.J. Lu wrote:
>
>> On Thu, Dec 14, 2017 at 4:29 PM, Joseph Myers <joseph@codesourcery.com> wrote:
>> > On Wed, 13 Dec 2017, Rajalakshmi Srinivasaraghavan wrote:
>> >
>> >> This implementation is based on generic s_sinf.c and s_cosf.c.
>> >> Tested on s390x, powerpc64le and powerpc32.
>> >>
>> >> 2017-12-12  Rajalakshmi Srinivasaraghavan  <raji@linux.vnet.ibm.com>
>> >>
>> >>       * sysdeps/ieee754/flt-32/s_cosf.c: Move reduced() and
>> >>       constants to s_sincos.h file.
>> >>       * sysdeps/ieee754/flt-32/s_sinf.c: Likewise.
>> >>       * sysdeps/ieee754/flt-32/s_sincosf.c: New implementation.
>> >>       * sysdeps/ieee754/flt-32/s_sincos.h: New file.
>> >
>> > OK.
>> >
>>
>> Please put those static constants into a different file and mark them
>> hidden so that they can be shared when they are compiled twice
>> for FMA.
>
> glibc is build with -fmerge-all-constants, so constants should already be
> shared at link time; making constants hidden would be relevant only for
> arrays, not for individual floating-point numbers (and if the individual
> constants aren't put in appropriate sections with link-time merging, it's
> probably because not doing so is more efficient on a particular
> architecture).
>

What do you suggest for x86-64 to avoid array duplication?


-- 
H.J.


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