RE: [Gcl-devel] STABLE, WINDOWS: read_fasd1() and alloc_relblock()

[Mike Thomas]

Hi Vadim.

Thanks very much for confirming.

| > Vadim, are you able to reproduce the results below using
| current CVS stable?
| > Note the combinations of gcc/binutils versions I am using,
| which I believe
| > are different to yours.
| >
|
| I have
|
| binutils 2.15.90 20040222
| gcc 3.3.1 20030804-1
|
| So called "Current" gcc (3.3.3 is "Candidate")
| and "Candidate" binutils.

OK, that probably rules out binutils 2.15.90 by itself as the culprit which
leaves either gcc 3.3.3 by itself or a combination with that version of
binutils.

| > 1. GCL/gcc 3.3.1/binutils 2.14.90 builds Maxima and passes
| tests both with
| > and without ignore-errors - I haven't been able to replicate the
| > "ignore-errors" path mangling bug so that looks good.
| >
|
| I confirm - latest Maxima CVS (2002.04.17) builds and passes all tests
| both with and without "ignore-errors".

Thanks.

If we choose not to pursue the cause of the gcc 3.3.3 Maxima crash (see my
other email) then I think we may have reached another stable point provided
that ACL2 and the random tester are OK.  I'll leave a random tester running
overnight hopefully to close off that potential point of failure.

Cheers

Mike Thomas