Opened 9 years ago

Last modified 9 years ago

#9999 new enhancement

Status of AIX port of Sage.

Reported by: drkirkby Owned by: drkirkby
Priority: minor Milestone: sage-wishlist
Component: porting: AIX or HP-UX Keywords:
Cc: Merged in:
Authors: Reviewers:
Report Upstream: N/A Work issues:
Branch: Commit:
Dependencies: Stopgaps:

Description (last modified by drkirkby)

This ticket lists those parts of Sage that have built in Sage, along with whether they have passed the tests. Unless otherwise stated, they results are from the following hardware.

  • IBM RS/6000 7025 F50
  • 4 x 332 MHz 32-bit PowerPC CPUs
  • 3 GB RAM
  • A fairly wide mixture of disks sizes (3 x 9 GB, 1 x 18 GB, 2 x 36 GB and 1 x 73 GB)
  • DDS-4 tape drive
  • AIX 5.3 (A POSIX certified operating system)
Package Build TestNotes
atlas-3.8.3.p14
blas-20070724
boehm_gc-7.1.p6
boost-cropped-1.34.1 OK
cddlib-094f.p7 OK
cephes-2.8 OK
cliquer-1.2.p7 No, #9979
conway_polynomials-0.2OK
cvxopt-0.9.p9
cython-0.12.1
docutils-0.5.p0
ecl-10.2.1.p2
eclib-20100711
ecm-6.2.1.p2
elliptic_curves-0.1 OK
examples-4.6.alpha1 OK
extcode-4.6.alpha1
f2c-20070816.p2
flint-1.5.0.p5#9996 Builds, but uses wrong library extension
flintqs-20070817.p5OK
fortran-20100629
freetype-2.3.5.p2#9985
gap-4.4.12.p4
gd-2.0.35.p5
gdmodule-0.56.p7
genus2reduction-0.3.p8
gfan-0.4plus.p1OK
givaro-3.2.13rc2.p2 OK
glpk-4.44OK
gnutls-2.2.1.p5 No, #9975
graphs-20070722.p1 OK
gsl-1.14 No, #10000 A likely fix is given
iconv-1.13.1.p3 OK
iml-1.0.1.p13
ipython-0.9.1.p0
jinja2-2.1.1.p0
lapack-20071123.p1
lcalc-20100428-1.23.p2
libfplll-3.0.12.p1 OK
libgcrypt-1.4.4.p3 OK
libgpg_error-1.6.p3
libm4ri-20100701.p1 OK
libpng-1.2.35.p2OK
linbox-1.1.6.p3
matplotlib-0.99.3
maxima-5.20.1.p0
mercurial-1.3.1.p2
moin-1.9.1.p1
mpfi-1.3.4-cvs20071125.p8
mpfr-2.4.2OK
mpir-1.2.2.p1OK
mpmath-0.15
networkx-1.2.p1
ntl-5.4.2.p12 #9998 Builds, but uses wrong library extension
numpy-1.3.0.p4
opencdk-0.6.6.p5 OK
palp-1.1.p3OK
patchNo, #10979 Is this the result of an AIX patch?
pari-2.4.3.svn-12577.p5No, #9990
pexpect-2.0.p4
pil-1.1.6.p2
polybori-0.6.4.p6
polytopes_db-20100210 OK
pycrypto-2.1.0
pygments-0.11.1.p0
pynac-0.2.0.p5
python_gnutls-1.1.4.p7
python-2.6.4.p9
r-2.10.1.p3
ratpoints-2.1.3.p1OK
readline-6.0.p2 #9987 Builds, but uses wrong library extension
rubiks-20070912.p12 #9991 Builds, but IBM s install will not install
sage_scripts-4.6.alpha1
sage-4.6.alpha1
sagenb-0.8.2
sagetex-2.2.5
scipy_sandbox-20071020.p5
scipy-0.7.p5
scons-1.2.0
setuptools-0.6c9.p0
singular-3-1-1-4.p2No, #9993
sphinx-0.6.3.p4
sqlalchemy-0.5.8
sqlite-3.6.22 OK
symmetrica-2.0.p5
sympow-1.018.1.p8OK
sympy-0.6.4.p0
tachyon-0.98beta.p11No, #9997 Does not try on AIX
termcap-1.3.1.p1 OK
twisted-9.0.p2
weave-0.4.9.p0
zlib-1.2.5#9988 Builds, but uses wrong library extension
zn_poly-0.9.p5
zodb3-3.7.0.p4

Change History (8)

comment:1 Changed 9 years ago by drkirkby

  • Description modified (diff)

comment:2 Changed 9 years ago by drkirkby

  • Description modified (diff)

comment:3 Changed 9 years ago by robertwb

  • Type changed from defect to enhancement

comment:4 follow-up: Changed 9 years ago by fbissey

Since I have the hardware and the software, I'll add myself here.

I should have a aix-7.1 test box soon and some power 7 box on which we may install anything from aix-5.3 to 7.1 by July/August?.

comment:5 in reply to: ↑ 4 Changed 9 years ago by drkirkby

Replying to fbissey:

Since I have the hardware and the software, I'll add myself here.

I should have a aix-7.1 test box soon and some power 7 box on which we may install anything from aix-5.3 to 7.1 by July/August?.

I'm limited to AIX 5.3 due to the rather old hardware - you can see what I have listed above.

There was some interest from IBM to semi-fund an AIX port of Sage. They contacted William and were considering giving one or two people access to a quick machine. But due to security issues about where the machine was hosted, it could only be one or two developers and from a couple of IP addresses. As such, both William and I said it was not worth bothering with.

Then it became apparent IBM wanted us to concentrate on one specific package - it seemed to me IBM were hoping to get a specific tool working on AIX for zero cost to them. I said I'd do that bit if they paid me, but nothing came of that.

Dave

comment:6 follow-up: Changed 9 years ago by fbissey

Given the kind of relationship we have with IBM here I should make some enquiries. Can you name the tool in question?

comment:7 in reply to: ↑ 6 Changed 9 years ago by drkirkby

Replying to fbissey:

Given the kind of relationship we have with IBM here I should make some enquiries. Can you name the tool in question?

Yes, I just checked my emails. It was Numpy and Scipy that someone at IBM wanted on AIX - I assume for his personal work. I've no idea of the work in doing this, but I'd be willing to at least consider it on a contract basis.

I don't think there would be much appetite for an AIX port of Sage by sage-developers. In any case, for a port to take place, some decent hardware would be needed. My machine is too old, so even permitting others to use it (and I've done that several times), it would not be suitable for Sage development. Neither is a machine which has access restricted to a couple of people.

The offers made by IBM before, while I'm sure were made with good intentions, were not acceptable to either William or I.

Another issue, is that even if IBM gave William a server, he has no AIX administrator. I don't know AIX that well - though perhaps just about to set up a server. I've set my own up OK.

IBM were supposed to be giving me an account on this 4 GHz AIX box, but I never got it, so I'm stuck using my own RS/6000, which is too old.

Dave

comment:8 Changed 9 years ago by drkirkby

  • Description modified (diff)
Note: See TracTickets for help on using tickets.