| 1 |
-----BEGIN PGP SIGNED MESSAGE----- |
| 2 |
Hash: SHA1 |
| 3 |
|
| 4 |
On 2004-07-21 19:02:06 -0400 Travis Tilley <lv@g.o> wrote: |
| 5 |
> i'd like for libffi to be the default if possible, since it's |
| 6 |
> required for |
| 7 |
> gnustep to work with hardened at all, or on archs like the one i |
| 8 |
> happen to |
| 9 |
> use. also, isnt ffcall unmainained? |
| 10 |
|
| 11 |
There's a good chance that ffcall is, for all intents and purposes, |
| 12 |
unmaintained, although people do still regularly use it. Libffi is a |
| 13 |
little less "popular", but for our ends, it is really the only choice. |
| 14 |
Sadly, the gcc build process doesn't compile it unless java and c++ |
| 15 |
are enabled, but if the target is only "install-target-libffi", it |
| 16 |
only increased the build time from 1 minute to 7, and not to ~35 |
| 17 |
minutes for the entire gcc w/ java. |
| 18 |
|
| 19 |
Does anyone know if the different arch testing machines available can |
| 20 |
host remote X sessions? If so, I really should get in touch with the |
| 21 |
maintainer of the amd64 machine, and get to the root of the issues |
| 22 |
with the GNUstep core library and amd64. So far, gdb backtraces on |
| 23 |
that arch have died in the same function, so I'm crossing my fingers |
| 24 |
it remains only that one. |
| 25 |
|
| 26 |
__Armando Di Cianno |
| 27 |
-----BEGIN PGP SIGNATURE----- |
| 28 |
Version: GnuPG v1.2.4 (GNU/Linux) |
| 29 |
Comment: Using the GPG bundle for GNUMail.app |
| 30 |
|
| 31 |
iD8DBQFA/8CFwgiTPLI9xhcRAjxFAJ9q3l3KinQMuqEHFw+RrAxg/z3Q6QCfdY2L |
| 32 |
8GaUVIi+GqjxKbDSmMCjes4= |
| 33 |
=yCbM |
| 34 |
-----END PGP SIGNATURE----- |
| 35 |
|
| 36 |
|
| 37 |
-- |
| 38 |
gentoo-dev@g.o mailing list |
Archives