[prev in list] [next in list] [prev in thread] [next in thread] 

List:       freebsd-alpha
Subject:    Re: strtod cleanup (patch)
From:       Peter Wemm <peter () wemm ! org>
Date:       2001-11-30 11:32:47
[Download RAW message or body]

Alexey Zelkin wrote:
> hi,
> 
> On Thu, Nov 22, 2001 at 09:42:41PM -0800, Peter Wemm wrote:
> 
> > > Most of this patch is mechanical merge, but I not sure
> > > on types usage correctness (long/int/int32_t) in this chunk:
> > > 
> > > #if BYTE_ORDER == BIG_ENDIAN
> > > #define IEEE_BIG_ENDIAN
> > > #define Long   int32_t
> > > #define ULong  u_int32_t
> > > #else
> > > #define IEEE_LITTLE_ENDIAN
> > > #define Long   long
> > > #define ULong  unsigned long
> > > #endif
> > 
> > This isn't quite right..  The #define Long int32_t etc should be used in
> > all cases, including i386.   The ia64 and alpha ports are little endian,
> > and work best with int32_t.
> > 
> > Also, maybe use:
> > typedef int32_t   Long;
> > typedef u_int32_t ULong;
> > .. that way there is no chance of preprocessor side effects with the space
> > in ULong.
> 
> Ok. Noted.
> 
> Updated patch included. Can someone who has alpha/ia64 hardware test this
> patch ? It should not break anything, but want to be sure.

I've been using this on my Alpha.  So far it works fine.  I used the
standard strtod() for a while and discovered it broke my perl build.n (!).
If I apply your patches, the perl problem goes away.

I am currently having big problems with ia64, I think it is the same thing,
but I haven't got a chance to test it out yet.  Can you please commit this?
It works on Alpha, and I think it solves a nightmare I'm having on ia64.

Cheers,
-Peter
--
Peter Wemm - peter@FreeBSD.org; peter@yahoo-inc.com; peter@netplex.com.au
"All of this is for nothing if we don't go to the stars" - JMS/B5


To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-alpha" in the body of the message


[prev in list] [next in list] [prev in thread] [next in thread]