Re: Cannot Build svn 0.21.0 server on Solaris 2.8

phLi wrote:

> Perhaps it is a mistake to set APR_PATH_MAX this
> way? Is there something else I should try to correct the "no decision has been
> made on APR_PATH_MAX" error?

This is likely unrelated, or an symptom of the same cause that somehow
makes sigset_t invisible.

> If setting APR_PATH_MAX in this way is okay, then I guess I will continue
> trying to determine what is going on with sigset_t by looking at preprocessor
> output.

It is not ok; APR_PATH_MAX ought to be defined based on PATH_MAX. If
limits.h was included (check with trace-includes), yet did not define
PATH_MAX, you should be seriously concerned. Step through the
preprocessor output, line by line, and see how things from the header
files on disk do or don't get included into the preprocessor output.

The procedure you will follow is the same whether you analyse the
sigset_t issue, or the PATH_MAX issue, both are likely preprocessor
problems. For the sigset_t issue, it might be easier to trace things,
though, since gcc normally does not copy #define lines into the
preprocessor output (so you can't tell whether or not PATH_MAX got defined).

Depending on your gcc version, you might be able to use the -dN or -dD
options, which will cause gcc to indicate, in the preprocessor output,
what defines it has processed.

Don't try random changes in the hope that the problem goes away
suddenly. Instead, try to study and understand the problem in detail,
and you will find its cause, and a cure.

Regards,
Martin

---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@subversion.tigris.org
For additional commands, e-mail: dev-help@subversion.tigris.org
Received on Sun May 4 18:50:15 2003