ddd
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: ddd-3.3.9-test2 build feedback HPUX once again...


From: Arnaud Desitter
Subject: Re: ddd-3.3.9-test2 build feedback HPUX once again...
Date: Fri, 28 May 2004 10:08:14 +0100

Hi,

----- Original Message ----- 
From: "Tuomo Tikkanen" <address@hidden>
To: "ext Arnaud Desitter" <address@hidden>
Cc: <address@hidden>
Sent: Thursday, May 27, 2004 7:47 PM
Subject: Re: ddd-3.3.9-test2 build feedback HPUX once again...


> ext Arnaud Desitter wrote:
> 
> >
> >I posted you a patch to hopefully fix the configure
> >problem.
> >  
> >
> I don't know if you have got the email I wrote yesterday. I have not yet 
> got it through address@hidden mailing list and I suspect that there is some 
> problems on listserver as my earlier letter has now been on the way more 
> than 24 hours. Thus I send this reply to you directly too.
> 
> I have now applied the patch you provided to clean ddd-3.3.9-test2 
> directory (one that was just untarred from tarball). This version seems 
> to do much better job, but there are two things I noticed:
> 
> 1) configure seems to select now ONLY shared version of the 
> Motif-library (libXm) on both HPUX 10.20 and HPUX 11.00. So it is not 
> using static libary given in configure script even though file exists in 
> the system where compilation is made.

Another typo of mine in configure.ac.
for file in "/usr/lib/libXm.a /usr/lib/Motif1.2_R6/libXm.a";
should have been
for file in "/usr/lib/libXm.a" "/usr/lib/Motif1.2_R6/libXm.a";
Now fixed.

> 2) when HP aCC compiler is used (on HPUX 11.00) configure does a bit bad 
> job as it claims that the compiler has builtin expect i.e. defines  
> HAVE_BUILTIN_EXPECT in ddd/config.h. This is due a bug in aclocal.m4:
> 
> -----8<-----8<-----8<-----8<-----8<-----8<-----
> <hepux1100:/share/wrk/tikkanen/maketus/a/ddd-3.3.9-test2> diff 
> aclocal.m4 aclocal.m4.afterpatch
> 648c648
> < if test "$ice_cv_have_builtin_expect" = yes; then
> ---
>  > if test "$ice_cv_have_attribute_noreturn" = yes; then
> -----8<-----8<-----8<-----8<-----8<-----8<-----
> 

Well spotted ! Applied to acinclude.m4. Thanks for your testing.

Regards,





reply via email to

[Prev in Thread] Current Thread [Next in Thread]