This is the mail archive of the gcc-patches@gcc.gnu.org mailing list for the GCC 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: Unreviewed math builtins patch (3x)


 > From: Jim Wilson <wilson@tuliptree.org>
 > 
 > Kaveh R. Ghazi wrote:
 > > Patch for gamma, bessel and error function math builtins 
 > > http://gcc.gnu.org/ml/gcc-patches/2003-08/msg00398.html
 > 
 > Ah.  This adds some of the functions I noticed were missing.  The 
 > complex arithmetic ones are still missing though.
 > 
 > I have some of the same reservations I have with the last patch.  Do we 
 > really want to give warnings by default for any program that uses 
 > function names like j0 and y0?

It's not that bad.  See Roger's followup.
http://gcc.gnu.org/ml/gcc-patches/2003-08/msg01819.html


 > You have erf{,f,l} as ATTR_MATHFN_FPROUNDING_ERRNO, but the description 
 > doesn't say anything about domain errors.

Thanks for catching this, fixed erf* (and also erfc*).


 > The man page for gamma discourages use of it.  Do we really want to 
 > implement it?

Yes.  The gamma function vs tgamma/lgamma is analogous to bcopy vs
memcpy/memmove.  And we do implement a bcopy builtin, so IMHO we
should also do gamma.  If the user ignores the man page and uses
gamma, we may as well perform whatever opts we would have done for
lgamma.


 > No man page for significand, but I found it in the info documentation. 
 > Now that I thought to check the docs, I found pow10 and exp10 info.
 > 
 > Otherwise this looks OK.

Thanks for the review, installed.

		--Kaveh
--
Kaveh R. Ghazi			ghazi@caip.rutgers.edu


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