1 |
On 12/10/12 06:58, Steven Trogdon wrote: |
2 |
> On Wed, 10 Oct 2012 17:04:11 -0700 |
3 |
> Thomas Kahle <tomka@g.o> wrote: |
4 |
> |
5 |
>> Update: |
6 |
>> |
7 |
>> The maxima-5.26 build failure is only with ecls-12.2.1, downgrading to |
8 |
>> ecls-11 or upgading to ecls-12.2.7 makes maxima compile again and then |
9 |
>> recompiling sage makes my original problem go away. So this is just a |
10 |
>> compile failure of maxima against ecls-12.2.1. Maybe the |
11 |
>> package.keywords file that comes with sage should avoid this particular |
12 |
>> version? |
13 |
>> |
14 |
>> Cheers, |
15 |
>> Thomas |
16 |
>> |
17 |
> |
18 |
> Well something definitely has changed. At one time maxima-5.26.0 and ecls-12.2.1 |
19 |
> could co-exist here as: |
20 |
> |
21 |
> --- |
22 |
> [I] sci-mathematics/maxima |
23 |
> Available versions: 5.18.1 (~)5.20.1-r99[1] (~)5.26.0 [m](~)5.27.0-r1 |
24 |
> [m](~)5.28.0 {{X clisp clozurecl cmucl ecl ecls emacs gcl latex linguas_es |
25 |
> linguas_pt linguas_pt_BR nls openmcl sbcl tk unicode xemacs}} Installed |
26 |
> versions: 5.26.0(05:31:36 PM 04/11/2012)(X clisp ecls emacs nls sbcl tk |
27 |
> unicode -clozurecl -cmucl -gcl -latex -linguas_es -linguas_pt -linguas_pt_BR |
28 |
> -xemacs) |
29 |
> |
30 |
> [I] dev-lisp/ecls |
31 |
> Available versions: 9.12.3 ~10.4.1 (~)11.1.1-r1 **11.1.1-r99[1] (~)12.2.1 |
32 |
> ~12.7.1 {{X debug doc (-)emacs gengc precisegc (+)threads +unicode}} Installed |
33 |
> versions: 12.2.1(11:23:54 PM 02/12/2012)(X -debug -emacs -gengc -precisegc |
34 |
> -threads -unicode) |
35 |
> --- |
36 |
> |
37 |
> But now I'm unable to rebuild maxima-5.26.0, even with ecls-12.7.1 installed. |
38 |
> The build just hangs with |
39 |
> |
40 |
> ; - Loading binary file "binary-ecl/init-cl.fas" |
41 |
> ;;; Loading |
42 |
> "/var/tmp/portage/sci-mathematics/maxima-5.26.0/work/maxima-5.26.0/src/binary-ecl/init-cl.fas" ; |
43 |
> - Providing system maxima |
44 |
> |
45 |
> If I remove the sbcl useflag from maxima then it builds, even with |
46 |
> ecls-12.2.1. I now have |
47 |
> |
48 |
> --- |
49 |
> [I] sci-mathematics/maxima |
50 |
> Available versions: 5.18.1 (~)5.20.1-r99[1] (~)5.26.0 [m](~)5.27.0-r1 |
51 |
> [m](~)5.28.0 {{X clisp clozurecl cmucl ecl ecls emacs gcl latex linguas_es |
52 |
> linguas_pt linguas_pt_BR nls openmcl sbcl tk unicode xemacs}} |
53 |
> Installed versions: 5.26.0(12:40:21 PM 10/11/2012)(X clisp ecls emacs nls |
54 |
> tk unicode -clozurecl -cmucl -gcl -latex -linguas_es -linguas_pt |
55 |
> -linguas_pt_BR -sbcl -xemacs |
56 |
> |
57 |
> [I] dev-lisp/ecls |
58 |
> Available versions: 9.12.3 ~10.4.1 (~)11.1.1-r1 **11.1.1-r99[1] (~)12.2.1 |
59 |
> ~12.7.1 {{X debug doc (-)emacs gengc precisegc (+)threads +unicode}} |
60 |
> Installed versions: 12.2.1(11:13:42 AM 10/11/2012)(X unicode -debug |
61 |
> -emacs -gengc -precisegc -threads) |
62 |
> --- |
63 |
> |
64 |
> But I'm not sure whether this will cure your particular Sage failure. |
65 |
> |
66 |
|
67 |
Building maxima-5.26.0 with sbcl is now broken by the latest version of |
68 |
asdf so the sbcl useflag needs to be masked for that particular version |
69 |
of maxima. I remember getting the message during the compile of the sbcl |
70 |
bits of maxima not the ecls ones but I could be wrong. |
71 |
|
72 |
Francois |