[Top][All Lists]

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

Re: The %.o: rule

From: Clark Rawlins
Subject: Re: The %.o: rule
Date: Mon, 4 Jun 2001 17:53:25 -0500
User-agent: Mutt/1.3.18i

On Sun, Jun 03, 2001 at 11:03:23PM -0400, Harlan Stenn wrote:
> > Besides the reason that it's not nessasary are there any reasons why
> > the automake inference rules for C and C++ don't use -o ?
> > 
> > I seem to remember that some compilers don't accept -o is this correct?
> Yes.  Some older compilers will not accept both -c and -o .
> > If so what is the typical way of dealing with the lack of -o when
> > you want it?
> either mv the produced .o name to what you want, or find a way to compile
> the source from a faked .c file.

Is there any kind of concensus on what would be the 'correct' way to do 
this in automake?  Is there a consensus that automake should continue to
support these old compilers?   I am attempting to modify automake to 
support paths and want the output files to end up in the same directory 
structure as the original files.

I am not an expert in autoconf or automake, so I ask those who are, would 
a macro to detect whether the compiler supports -o with -c and some 
additional code to handle the case when it dosn't be a better way to 
handle this condition?

> You could also mv an existing .o out of the way, compile, and rename back,
> but that gets messy and doesn't do well with parallel makes.
> H

Attachment: pgpQm_aJr9Cjp.pgp
Description: PGP signature

reply via email to

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