| 1 |
On Sat, 3 Dec 2022, Haelwenn (lanodan) Monnier wrote: |
| 2 |
> Well Alpine is using the ecl route: |
| 3 |
> https://git.alpinelinux.org/aports/tree/community/sbcl/APKBUILD |
| 4 |
> And GNU GuixSD is using the clisp route but per the comments ecl would be |
| 5 |
> better: |
| 6 |
> https://git.savannah.gnu.org/cgit/guix.git/tree/gnu/packages/lisp.scm#n424 |
| 7 |
In principle, we can do something similar: |
| 8 |
|
| 9 |
1. If elibc_musl, sbcl should BDEPEND on ecls (or maybe clisp) |
| 10 |
|
| 11 |
2. In the same case, the build system should be patched to use another |
| 12 |
bootstrap lisp |
| 13 |
|
| 14 |
3. For elibc_glibc nothing should change |
| 15 |
|
| 16 |
4. All this should be tested on a computer running a musl profile |
| 17 |
|
| 18 |
Personally I can do nothing of this list. I have no computer with musl |
| 19 |
profile. Does any developer running such a system volunteer to do this |
| 20 |
work - to be a co-maintained of sbcl on musl? |
| 21 |
|
| 22 |
If not, I'll go forward and pmask all packages which unconditionally |
| 23 |
depend on sbcl (maybe via some intermediates) in |
| 24 |
profiles/features/musl/use.mask. |
| 25 |
|
| 26 |
Andrey |
Archives