RE: [Gcl-devel] 2.6.2.....

[Mike Thomas]

Hi Camm.

| OK, I'll put SGC as an unimplemented feature.

Please do.

|  Should we say the same
| for dlopen and bfd?

dlopen unimplemented.

bfd - I think unimplemented as the problem seems to be that the BFD library
does not support the method of relocation we are using.

|  I.e. would these work in principle were we to
| code it,

BFD should in principle be workable assuming someone has the time and
ability to make whatever additions to the library are needed, (and then,
very likely, have to support it forever more as well) but now that the
custom-reloc linking seems stable, I'm not inclined to worry about it,
although your recent discussions about BLAS etc did leave me with doubts.

SGC I don't know, as I'm not familiar with the semantics of the Unix
functions/signals involved and I don't understand what triggers the garbage
collection.  I have a feeling you explained this to me recently but have
misplaced the message.

My next GCL targets as 2.7.0 proceeds will be the Axiom seg-faults and
testing Michael Ks work on paths.

Cheers

Mike Thomas