public inbox for gcc-help@gcc.gnu.org
 help / color / mirror / Atom feed
From: "Jean-Paul de Vooght" <jean-paul.devooght@smallrivers.com>
To: <gcc-help@gcc.gnu.org>
Subject: RE: use of fabsl and logbl builtins
Date: Sat, 16 Aug 2008 11:43:00 -0000	[thread overview]
Message-ID: <000101c8ff94$6d943010$48bc9030$@devooght@smallrivers.com> (raw)
In-Reply-To: <003001c8fec6$b217c0e0$164742a0$@de>

Just to close this post, I eventually figured out that gcc built-ins are
really a library matter and that cygwin does not provide an implementation
for the ones below. MinGW does.

Eventually I found my salvation through the use of GNU MP BugNum library
which has been ported to cygwin.
- JP

> -----Original Message-----
> From: JP de Vooght [mailto:jp@vooght.de]
> Sent: 15 August 2008 13:05
> To: gcc-help@gcc.gnu.org
> Subject: use of fabsl and logbl builtins
> 
> Hello,
> I recently discovered the world of GNU R and its optional packages.
> While
> trying to install igraph, gcc 3.4.4 on cygwin generated the following
> output:
> 
> g++ -I/usr/local/lib/R/include  -I/usr/local/include   -DUSING_R   -g -
> O2 -c
> bliss.cc -o bliss.o
> In file included from bliss_graph.hh:32,
>                  from bliss.cc:19:
> bliss_bignum.hh: In member function `int
> igraph::BigNum::tostring(char**)':
> bliss_bignum.hh:76: error: `fabsl' undeclared (first use this function)
> bliss_bignum.hh:76: error: (Each undeclared identifier is reported only
> once
> for each function it appears in.)
> bliss_bignum.hh:76: error: `logbl' undeclared (first use this function)
> make: *** [bliss.o] Error 1
> chmod: cannot access `/usr/local/lib/R/library/igraph/libs/*': No such
> file
> or directory
> ERROR: compilation failed for package 'igraph'
> ** Removing '/usr/local/lib/R/library/igraph'
> 
> The error is related to the use of two built-ins fabsl and logbl.
> Replacing
> these functions with their __builtin_ homologues just defers the
> problem to
> ld which fails with undefined _logbl.
> 
> Any suggestion on how to work around this? The header file imports
> math.h
> and uses fabsl and logbl as follows:
> 
> <snip>
>  67 class BigNum
>  68 {
>  69   long double v;
>  70 public:
>  71   BigNum(): v(0.0) {}
>  72   void assign(const int n) {v = (long double)n; }
>  73   void multiply(const int n) {v *= (long double)n; }
>  74   int print(FILE *fp) {return fprintf(fp, "%Lg", v); }
>  75   int tostring(char **str) {
>  76     int size=static_cast<int>( (logbl(fabsl(v))/log(10.0))+4 );
>  77     *str=igraph_Calloc(size, char );
>  78     if (! *str) {
>  79       IGRAPH_ERROR("Cannot convert big number to string",
> IGRAPH_ENOMEM);
>  80     }
>  81     snprintf(*str, size, "%.0Lf", v);
>  82     return 0;
>  83   }
>  84 };
> </snip>
> 
> TIA
> - JP
> 
> 
> 
> 
> 


      reply	other threads:[~2008-08-16 11:38 UTC|newest]

Thread overview: 2+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2008-08-15 14:25 JP de Vooght
2008-08-16 11:43 ` Jean-Paul de Vooght [this message]

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to='000101c8ff94$6d943010$48bc9030$@devooght@smallrivers.com' \
    --to=jean-paul.devooght@smallrivers.com \
    --cc=gcc-help@gcc.gnu.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).