Dan Sugalski wrote:
At 1:05 PM +0100 10/27/02, Leopold Toetsch wrote:
Though, /me thinks, that we should always have a valid interpreter.
I tried very hard to make sure that there was always a valid interpreter.
Now we have one ;-)
I did change the startup sequence, so that _all_ memory allocated during early start up (except of course the managing structures themselves) gets managed in our pools. These are mainly vtable->name()s and the Parrot_base_classname_hash.
I did change PIO_eprintf slightly: It does now an immediate vfprintf(stderr,..), when there is no interpreter. This means, that w/o interpreter only arguments according to printf(3) are valid.
There a still function's like PackFile_new w/o interpreter, but they don't allocate any unmanaged memory except the packfile itself.
You might need a "make realclean && perl Configure.pl && make" to get things running again.
leo
