Hello *,
I've just committed the maxima-5.28.0 ebuild to Gentoo. On my x86 box,
testsuite times (full, in seconds) and failed tests are
sbcl-1.0.58 180
gcl-2.6.8_pre 187
cmucl-20c 209
ccl-1.8 294 rtest15 (37, 193, 196)
ecl-12.7.1 380
clisp-2.49 671
I've noticed one inconvenient thing about 5.28.0. If I configure maxima
for one specific lisp from the above list, compile, and run tests,
everything's OK. If I configure for all 6 lisps, compilation fails in gcl,
after
; - Loading source file
;
"/var/tmp/portage/sci-mathematics/maxima-5.28.0/work/maxima-5.28.0/src/init-cl.lisp"
Loading
/var/tmp/portage/sci-mathematics/maxima-5.28.0/work/maxima-5.28.0/src/init-cl.lisp
Finished loading
/var/tmp/portage/sci-mathematics/maxima-5.28.0/work/maxima-5.28.0/src/init-cl.lisp
; - Providing system maxima
make[1]: Leaving directory
`/var/tmp/portage/sci-mathematics/maxima-5.28.0/work/maxima-5.28.0/src'
make: *** [all-recursive] Error 1
If I remove gcl and configure for 5 remaining lisps, compilation fails in
sbcl, after
; - Loading source file
;
"/var/tmp/portage/sci-mathematics/maxima-5.28.0/work/maxima-5.28.0/src/init-cl.lisp"
; loading
#P"/var/tmp/portage/sci-mathematics/maxima-5.28.0/work/maxima-5.28.0/src/init-cl.lisp"
; - Providing system maxima
[undoing binding stack and other enclosing state... done]
[saving current Lisp image into binary-sbcl/maxima.core:
writing 2760 bytes from the read-only space at 0x01000000
writing 1688 bytes from the static space at 0x01100000
writing 55083008 bytes from the dynamic space at 0x09000000
done]
make[1]: Leaving directory
`/var/tmp/portage/sci-mathematics/maxima-5.28.0/work/maxima-5.28.0/src'
make: *** [all-recursive] Error 1
Why builds for different lisps now interfere? This was not the case in
previous releases. It was much more convenient. Can we find what change in
the build system causes this interference?
Andrey