| 1 |
Hi all, |
| 2 |
|
| 3 |
For several years now I have keeping pari up to date in the |
| 4 |
sage-on-gentoo overlay, even getting upstream snapshots as |
| 5 |
needed for sage. For a while my stable ebuilds have been picked |
| 6 |
up for the main tree by a kind soul. |
| 7 |
|
| 8 |
For several years now I haven't been able to successfully run |
| 9 |
any of pari testsuite from portage. And it seems to be something |
| 10 |
in the portage environment. I have now decided to seek some help |
| 11 |
on the issue as I have run out of things I can think of a long |
| 12 |
time ago. |
| 13 |
|
| 14 |
So if I run "ebuild pari-2.11.2.ebuild test" (valid with all older |
| 15 |
ebuilds available) I get |
| 16 |
>>> Source compiled. |
| 17 |
>>> Test phase: sci-mathematics/pari-2.11.2 |
| 18 |
make -j8 dobench |
| 19 |
Making dobench in Olinux-x86_64 |
| 20 |
make[1]: Entering directory |
| 21 |
'/dev/shm/portage/sci-mathematics/pari-2.11.2/work/pari-2.11.2/Olinux-x86_64' |
| 22 |
* Testing objets gp-dyn..TIME= 1 |
| 23 |
* Testing analyz gp-dyn..TIME= 3 |
| 24 |
* Testing number gp-dyn..BUG [10] |
| 25 |
* Testing polyser gp-dyn..BUG [3] |
| 26 |
* Testing linear gp-dyn..BUG [0] |
| 27 |
* Testing elliptic gp-dyn..BUG [14] |
| 28 |
* Testing sumiter gp-dyn..TIME= 5 |
| 29 |
* Testing graph gp-dyn..TIME= 3 |
| 30 |
* Testing program gp-dyn..TIME= 0 |
| 31 |
* Testing trans gp-dyn..TIME= 11 |
| 32 |
* Testing nfields gp-dyn..BUG [38] |
| 33 |
+++ [BUG] Total bench for gp-dyn is 57 |
| 34 |
|
| 35 |
PROBLEMS WERE NOTED. The following files list them in diff format: |
| 36 |
Directory: |
| 37 |
/dev/shm/portage/sci-mathematics/pari-2.11.2/work/pari-2.11.2/Olinux-x86_64 |
| 38 |
number-dyn.dif |
| 39 |
polyser-dyn.dif |
| 40 |
linear-dyn.dif |
| 41 |
elliptic-dyn.dif |
| 42 |
nfields-dyn.dif |
| 43 |
make[1]: *** [Makefile:344: dobench] Error 1 |
| 44 |
make[1]: Leaving directory |
| 45 |
'/dev/shm/portage/sci-mathematics/pari-2.11.2/work/pari-2.11.2/Olinux-x86_64' |
| 46 |
make: *** [Makefile:34: dobench] Error 2 |
| 47 |
* ERROR: sci-mathematics/pari-2.11.2::sage-on-gentoo failed (test phase): |
| 48 |
* emake failed |
| 49 |
* |
| 50 |
* If you need support, post the output of `emerge --info |
| 51 |
'=sci-mathematics/pari-2.11.2::sage-on-gentoo'`, |
| 52 |
* the complete build log and the output of `emerge -pqv |
| 53 |
'=sci-mathematics/pari-2.11.2::sage-on-gentoo'`. |
| 54 |
* The complete build log is located at |
| 55 |
'/var/log/portage/build/sci-mathematics/pari-2.11.2:20190718-215929.log'. |
| 56 |
* For convenience, a symlink to the build log is located at |
| 57 |
'/dev/shm/portage/sci-mathematics/pari-2.11.2/temp/build.log'. |
| 58 |
* The ebuild environment file is located at |
| 59 |
'/dev/shm/portage/sci-mathematics/pari-2.11.2/temp/environment'. |
| 60 |
* Working directory: |
| 61 |
'/dev/shm/portage/sci-mathematics/pari-2.11.2/work/pari-2.11.2' |
| 62 |
* S: '/dev/shm/portage/sci-mathematics/pari-2.11.2/work/pari-2.11.2' |
| 63 |
fbissey@moonloop ~/Work/Overlays/sage-on-gentoo/sci-mathematics/pari $ |
| 64 |
cd /dev/shm/portage/sci-mathematics/pari-2.11.2/work/pari-2.11.2 |
| 65 |
fbissey@moonloop |
| 66 |
/dev/shm/portage/sci-mathematics/pari-2.11.2/work/pari-2.11.2 $ make -j8 |
| 67 |
dobench |
| 68 |
Making dobench in Olinux-x86_64 |
| 69 |
make[1]: Entering directory |
| 70 |
'/dev/shm/portage/sci-mathematics/pari-2.11.2/work/pari-2.11.2/Olinux-x86_64' |
| 71 |
* Testing objets gp-dyn..TIME= 0 |
| 72 |
* Testing analyz gp-dyn..TIME= 0 |
| 73 |
* Testing number gp-dyn..TIME= 1 |
| 74 |
* Testing polyser gp-dyn..TIME= 3 |
| 75 |
* Testing linear gp-dyn..TIME= 0 |
| 76 |
* Testing elliptic gp-dyn..TIME= 14 |
| 77 |
* Testing sumiter gp-dyn..TIME= 5 |
| 78 |
* Testing graph gp-dyn..TIME= 3 |
| 79 |
* Testing program gp-dyn..TIME= 3 |
| 80 |
* Testing trans gp-dyn..TIME= 11 |
| 81 |
* Testing nfields gp-dyn..TIME= 42 |
| 82 |
+++ Total bench for gp-dyn is 48 |
| 83 |
make[1]: Leaving directory |
| 84 |
'/dev/shm/portage/sci-mathematics/pari-2.11.2/work/pari-2.11.2/Olinux-x86_64' |
| 85 |
|
| 86 |
That's right the tests fails in the ebuild but pass when I run the same |
| 87 |
command manually/ It is not a fluke due to running the tests twice. |
| 88 |
If I do "ebuild ... compile" and then run the tests manually, they will |
| 89 |
succeed. |
| 90 |
|
| 91 |
When I look at the diff produced by the test suite I gat stuff like |
| 92 |
*** ../src/test/32/number 2018-07-26 06:00:21.000000000 +1200 |
| 93 |
--- gp.out 2019-07-19 10:11:20.505605285 +1200 |
| 94 |
*************** |
| 95 |
*** 60,70 **** |
| 96 |
|
| 97 |
? factor(100!+1,0) |
| 98 |
|
| 99 |
! [101 1] |
| 100 |
|
| 101 |
! [14303 1] |
| 102 |
|
| 103 |
! [149239 1] |
| 104 |
|
| 105 |
|
| 106 |
[432885273849892962613071800918658949059679308685024481795740765527568493010 |
| 107 |
|
| 108 |
727023757461397498800981521440877813288657839195622497225621499427628453 1] |
| 109 |
--- 60,73 ---- |
| 110 |
|
| 111 |
? factor(100!+1,0) |
| 112 |
|
| 113 |
! [ |
| 114 |
|
| 115 |
! |
| 116 |
101 1] |
| 117 |
|
| 118 |
! [ |
| 119 |
|
| 120 |
! |
| 121 |
14303 1] |
| 122 |
|
| 123 |
! [ |
| 124 |
|
| 125 |
! |
| 126 |
149239 1] |
| 127 |
|
| 128 |
|
| 129 |
[432885273849892962613071800918658949059679308685024481795740765527568493010 |
| 130 |
|
| 131 |
727023757461397498800981521440877813288657839195622497225621499427628453 1] |
| 132 |
|
| 133 |
The only difference is a widely different formatting of the output when |
| 134 |
run under the portage environment. |
| 135 |
|
| 136 |
Does anyone have a clue as to how this is possible? What could cause it |
| 137 |
and more importantly, how to fix it. |
| 138 |
|
| 139 |
Cheers, |
| 140 |
François |
Archives