| 1 |
It should be libmpfr.so.3 not .so.4 which are the new ones. |
| 2 |
|
| 3 |
> On 29/01/2018, at 15:55, Steven Trogdon <strogdon@×××××.edu> wrote: |
| 4 |
> |
| 5 |
> Hello, |
| 6 |
> |
| 7 |
> emerge of the latest sage-9999 from sage-on-gentoo requires |
| 8 |
> |
| 9 |
> emerge -1 sage |
| 10 |
> Calculating dependencies... done! |
| 11 |
> [ebuild U #] dev-libs/mpfr-4.0.0-r1 [3.1.6] |
| 12 |
> [ebuild N #] sci-mathematics/sage-9999 USE="debug doc-html doc-pdf sagenb |
| 13 |
> testsuite -X -bliss -latex -libbraiding -libhomfly -modular_decomposition" |
| 14 |
> L10N="en -ca -de -es -fr -hu -it -ja -pt -ru -tr" PYTHON_TARGETS="python2_7 |
| 15 |
> -python3_6" |
| 16 |
> |
| 17 |
> Presently, mpfr-4.0.0-r1 is package masked. |
| 18 |
> |
| 19 |
> # Matthias Maier <tamiko@g.o> (26 Dec 2017) |
| 20 |
> # gcc depends on mpfr and this version changes soname. Spare users with |
| 21 |
> # FEATURES=-preserve-libs from completely frying their system. |
| 22 |
> =dev-libs/mpfr-4.0.0-r1 |
| 23 |
> |
| 24 |
> So, since I did not have FEATURES=-preserve-libs I proceeded to unmask |
| 25 |
> mpfr-4.0.0-r1 and continued with the emerge of sage with the hope that libs |
| 26 |
> were preserved. The short story is that I ended up with a broken gcc since the |
| 27 |
> subject libraries were not preserved. This is more elaborately documented at |
| 28 |
> https://github.com/cschwan/sage-on-gentoo/issues/497 |
| 29 |
> |
| 30 |
> Fortunately, the gcc was recovered by linking to the host (debian) libmpfr and |
| 31 |
> rebuilding the prefix mpfr-3.1.6. So, is there any way to check before one |
| 32 |
> attempts to do this sort of thing that libs in question are being preserved? I |
| 33 |
> can certainly try this again - it may have been gremlins - since I believe I can |
| 34 |
> recover if it fails, but it would be nice to know whether things will work or |
| 35 |
> whether portage is broken. Here I'm using portage-2.3.20 |
| 36 |
> |
| 37 |
> Thanks everyone, |
| 38 |
> Steven |
| 39 |
|
| 40 |
|
| 41 |
This email may be confidential and subject to legal privilege, it may |
| 42 |
not reflect the views of the University of Canterbury, and it is not |
| 43 |
guaranteed to be virus free. If you are not an intended recipient, |
| 44 |
please notify the sender immediately and erase all copies of the message |
| 45 |
and any attachments. |
| 46 |
|
| 47 |
Please refer to http://www.canterbury.ac.nz/emaildisclaimer for more |
| 48 |
information. |
Archives