mersenneforum.org  

Go Back   mersenneforum.org > Factoring Projects > CADO-NFS

Reply
 
Thread Tools
Old 2021-10-21, 19:41   #1
EdH
 
EdH's Avatar
 
"Ed Hall"
Dec 2009
Adirondack Mtns

24×257 Posts
Default CADO-NFS git pull doesn't appear to be sufficient

Note: The server in this case is remote.

I'm running into an interesting situation - I might have to send a note to the CADO-NFS Team!

I'm finding that "git pull" is not sufficient (even with "git reset --hard") for my installs. (I did include "make clean.")

(I think) all, but one, of my machines I updated so far with "git pull" have failed "make check" with:
Code:
445/581 Test #445: test_matpoly_u64k1 ...........................................***Exception: SegFault  0.51 sec

99% tests passed, 1 tests failed out of 581

Total Test time (real) = 963.03 sec

The following tests FAILED:
    445 - test_matpoly_u64k1 (SEGFAULT)
The one that passed "make check" wouldn't communicate properly with the server. I tried twice with a couple of the machines, before moving to clean cloning.

I have been having to start with a new clone to get a working install that passes "make check" and works with the server.

All comments welcome. . .
EdH is offline   Reply With Quote
Old 2021-10-22, 06:24   #2
ccorn
 
ccorn's Avatar
 
Apr 2010

15110 Posts
Default

If you have -march specified in your CFLAGS, CXXFLAGS, CPPFLAGS, e.g. -march=haswell, and your build machine does not support that (e.g. because it is -march=sandybridge), then some autoconfiguration tests go wrong. There is a related thread on the Arch User Repository.

Edit: Scrap the above, that would not explain why the clean clone works.
I do remember that they have moved to a Gitlab instance a while ago.
That should not affect the clones (an update like git remote set-url origin NEW_URL should suffice), but perhaps they did some non-straightforward export/import instead, changing the root revision's checksum, so the clones would not accept the new remote repo. I remember having had to re-clone some moved projects on such occasions, but I am not sure whether CADO-NFS was among those.

Last fiddled with by ccorn on 2021-10-22 at 06:43
ccorn is offline   Reply With Quote
Old 2021-10-22, 12:50   #3
EdH
 
EdH's Avatar
 
"Ed Hall"
Dec 2009
Adirondack Mtns

24·257 Posts
Default

I think these were replicated from a single clone since the newer repository move, but git pull was working last year. However, I never ran a "make check" in the past, so maybe this error was always there, way back when I did my other builds and I didn't know it. I've saved some of the previous folders. Maybe I should try a "make check" and see what I get.
EdH is offline   Reply With Quote
Old 2021-10-22, 13:06   #4
EdH
 
EdH's Avatar
 
"Ed Hall"
Dec 2009
Adirondack Mtns

24×257 Posts
Default

Quote:
Originally Posted by EdH View Post
I think these were replicated from a single clone since the newer repository move, but git pull was working last year. However, I never ran a "make check" in the past, so maybe this error was always there, way back when I did my other builds and I didn't know it. I've saved some of the previous folders. Maybe I should try a "make check" and see what I get.
I tried "make check" in an original folder and it doesn't appear to exist. The last time I updated the revision on all my machines was a year ago. I'm sure much has changed and "make check" must have been added.

I guess at least for now, I've got "brand new, shiny" builds. . .

Thanks for the reply.
EdH is offline   Reply With Quote
Reply

Thread Tools


Similar Threads
Thread Thread Starter Forum Replies Last Post
Just because I'm Paranoid doesn't mean They're not out to Get Me. chalsall chalsall 5 2020-08-09 15:27
necessary but not sufficient condition for primes Alberico Lepore Alberico Lepore 22 2018-01-03 16:17
Devaraj numbers- necessary and sufficient condition devarajkandadai Number Theory Discussion Group 7 2017-09-23 02:58
Questions on security and resource pull MiniGeek Information & Answers 5 2017-01-29 03:54
Is a 18M-FFT sufficient to test a 100M number? __HRB__ Information & Answers 13 2010-05-23 13:37

All times are UTC. The time now is 04:57.


Sun Dec 5 04:57:41 UTC 2021 up 134 days, 23:26, 0 users, load averages: 1.12, 1.42, 1.41

Powered by vBulletin® Version 3.8.11
Copyright ©2000 - 2021, Jelsoft Enterprises Ltd.

This forum has received and complied with 0 (zero) government requests for information.

Permission is granted to copy, distribute and/or modify this document under the terms of the GNU Free Documentation License, Version 1.2 or any later version published by the Free Software Foundation.
A copy of the license is included in the FAQ.