Re: [RP] Solaris/CDE problem solved

[Brian Lo Bue]

Thanks for the heads up.

I'm pretty busy and won't be able to test soon.  Also I'll soon lose
access to any Solaris machines.  So maybe someone else can test this?

Thanks,
Brian

shawn writes:
 > >Date: Fri,  5 Jan 2001 15:51:52 -0800 (PST)
 > >
 > >
 > >Hi All,
 > >
 > >If you've been looking at the list the past few days you've seen my
 > >attempt to get ratpoison to play nice with Solaris/CDE.  I've figured
 > >out what is going on.  
 > >
 > >Way back in the mist of time when a process called signal() the
 > >handler was only set for one invocation.  So you'd have to put another 
 > >signal() call in your handler so that the next signal would be
 > >handled.  As it turns out SVR4 systems, like Solaris, use the default
 > >old behavior for signal and BSD type systems reinstall the handler
 > >automatically.  I'm gonna infer from this that Linux reinstalls the
 > >handler since I'm pretty sure that this works on other people's
 > >machines.  Long story short, I changed the calls to a POSIX version
 > >rebuilt and now it behaves as expected.  The diffs follow:
 > 
 > 
 > >Feel free to include it in the tree if you'd like.
 > 
 > I've integrated your diff into CVS. It still works fine for me, but it
 > always did. Could you or someone who has access to such a machine test
 > it?