If you have not seen it, it is now possible to build Sage packages in parallel.
See http://trac.sagemath.org/sage_trac/ticket/8306
This is in my mind one of the most impressive improvements I've ever seen to aid
building Sage.
On my Sun Ultra 27, running OpenSolaris, I can build 52 packages in just 8
minutes! (I've not even bothered using ccache, which apparently would speed it
up even more).
You can probably do better with some of the big servers on the sage network.
It bombs out at that point, as Maxima is not building on OpenSolaris - one of
the very few packages which remain to be fixed for a full 64-bit build on
OpenSolaris.
drkir...@hawk:~/sage-4.5.alpha0$ ls -lrt spkg/installed
total 99
-rw-r--r-- 1 drkirkby staff 0 Jun 27 19:07 dir-0.1
-rw-r--r-- 1 drkirkby staff 146 Jun 27 19:07 fortran-20100626
-rw-r--r-- 1 drkirkby staff 140 Jun 27 19:07 cephes-2.8
-rw-r--r-- 1 drkirkby staff 143 Jun 27 19:07 blas-20070724
-rw-r--r-- 1 drkirkby staff 148 Jun 27 19:07 lapack-20071123.p1
-rw-r--r-- 1 drkirkby staff 0 Jun 27 19:07 prereq-0.7
-rw-r--r-- 1 drkirkby staff 0 Jun 27 19:07 bzip2-1.0.5
-rw-r--r-- 1 drkirkby staff 223 Jun 27 19:07 sage_scripts-4.5.alpha0
-rw-r--r-- 1 drkirkby staff 219 Jun 27 19:07 examples-4.5.alpha0
-rw-r--r-- 1 drkirkby staff 222 Jun 27 19:07 conway_polynomials-0.2
-rw-r--r-- 1 drkirkby staff 220 Jun 27 19:07 boost-cropped-1.34.1
-rw-r--r-- 1 drkirkby staff 218 Jun 27 19:07 graphs-20070722.p1
-rw-r--r-- 1 drkirkby staff 216 Jun 27 19:07 termcap-1.3.1.p1
-rw-r--r-- 1 drkirkby staff 219 Jun 27 19:07 elliptic_curves-0.1
-rw-r--r-- 1 drkirkby staff 221 Jun 27 19:07 polytopes_db-20100210
-rw-r--r-- 1 drkirkby staff 215 Jun 27 19:07 f2c-20070816.p2
-rw-r--r-- 1 drkirkby staff 210 Jun 27 19:07 zlib-1.2.5
-rw-r--r-- 1 drkirkby staff 217 Jun 27 19:07 sympow-1.018.1.p7
-rw-r--r-- 1 drkirkby staff 219 Jun 27 19:08 rubiks-20070912.p11
-rw-r--r-- 1 drkirkby staff 211 Jun 27 19:08 palp-1.1.p3
-rw-r--r-- 1 drkirkby staff 216 Jun 27 19:08 libpng-1.2.35.p2
-rw-r--r-- 1 drkirkby staff 220 Jun 27 19:08 tachyon-0.98beta.p11
-rw-r--r-- 1 drkirkby staff 215 Jun 27 19:08 readline-6.0.p2
-rw-r--r-- 1 drkirkby staff 217 Jun 27 19:08 freetype-2.3.5.p2
-rw-r--r-- 1 drkirkby staff 217 Jun 27 19:08 symmetrica-2.0.p5
-rw-r--r-- 1 drkirkby staff 215 Jun 27 19:08 boehm_gc-7.1.p6
-rw-r--r-- 1 drkirkby staff 216 Jun 27 19:08 libm4ri-20100221
-rw-r--r-- 1 drkirkby staff 215 Jun 27 19:08 iconv-1.13.1.p2
-rw-r--r-- 1 drkirkby staff 219 Jun 27 19:09 libgpg_error-1.6.p3
-rw-r--r-- 1 drkirkby staff 212 Jun 27 19:09 gd-2.0.35.p5
-rw-r--r-- 1 drkirkby staff 218 Jun 27 19:09 libgcrypt-1.4.4.p3
-rw-r--r-- 1 drkirkby staff 213 Jun 27 19:09 sqlite-3.6.22
-rw-r--r-- 1 drkirkby staff 216 Jun 27 19:09 opencdk-0.6.6.p4
-rw-r--r-- 1 drkirkby staff 213 Jun 27 19:09 mpir-1.2.2.p1
-rw-r--r-- 1 drkirkby staff 219 Jun 27 19:09 flintqs-20070817.p5
-rw-r--r-- 1 drkirkby staff 218 Jun 27 19:09 ratpoints-2.1.3.p1
-rw-r--r-- 1 drkirkby staff 213 Jun 27 19:10 pari-2.3.5.p1
-rw-r--r-- 1 drkirkby staff 222 Jun 27 19:10 genus2reduction-0.3.p6
-rw-r--r-- 1 drkirkby staff 214 Jun 27 19:10 cddlib-094f.p7
-rw-r--r-- 1 drkirkby staff 215 Jun 27 19:10 gfan-0.4plus.p1
-rw-r--r-- 1 drkirkby staff 212 Jun 27 19:11 ecm-6.2.1.p2
-rw-r--r-- 1 drkirkby staff 219 Jun 27 19:11 givaro-3.2.13rc2.p2
-rw-r--r-- 1 drkirkby staff 215 Jun 27 19:11 gnutls-2.2.1.p5
-rw-r--r-- 1 drkirkby staff 214 Jun 27 19:11 zn_poly-0.9.p4
-rw-r--r-- 1 drkirkby staff 210 Jun 27 19:12 mpfr-2.4.2
-rw-r--r-- 1 drkirkby staff 225 Jun 27 19:12 mpfi-1.3.4-cvs20071125.p8
-rw-r--r-- 1 drkirkby staff 218 Jun 27 19:12 libfplll-3.0.12.p1
-rw-r--r-- 1 drkirkby staff 222 Jun 27 19:12 lcalc-20100428-1.23.p0
-rw-r--r-- 1 drkirkby staff 210 Jun 27 19:12 ecl-10.4.1
-rw-r--r-- 1 drkirkby staff 215 Jun 27 19:13 python-2.6.4.p9
-rw-r--r-- 1 drkirkby staff 213 Jun 27 19:15 ntl-5.4.2.p12
I think we need to ensure 'prereq' is built earlier in the cycle, as it seems
silly that lapack, which needs a fortran compiler, is built before we have even
checked if the Fortran compiler works or not.
But pretty impressive. Well done Mitesh Patel
Dave
--
To post to this group, send an email to sage-devel@googlegroups.com
To unsubscribe from this group, send an email to
sage-devel+unsubscr...@googlegroups.com
For more options, visit this group at http://groups.google.com/group/sage-devel
URL: http://www.sagemath.org
