GNU bug report logs - #55901
[PATCH WIP] gnu: Add lis.

Previous Next

To add a comment to this bug, you must first unarchive it, by sending
a message to control AT debbugs.gnu.org, with unarchive 55901 in the body.
You can then email your comments to 55901 AT debbugs.gnu.org in the normal way.

Toggle the display of automated, internal messages from the tracker.

Message #5 received at submit <at> debbugs.gnu.org (full text, mbox):

From: Liliana Marie Prikler <liliana.prikler <at> gmail.com>
To: guix-patches <at> gnu.org
Subject: [PATCH WIP] gnu: Add lis.
Date: Sat, 11 Jun 2022 09:49:23 +0200

* gnu/packages/maths.scm (lis): New variable.
---
 gnu/packages/maths.scm | 40 ++++++++++++++++++++++++++++++++++++++++
 1 file changed, 40 insertions(+)

diff --git a/gnu/packages/maths.scm b/gnu/packages/maths.scm
index 83f0a1810b..551b58e3f6 100644
--- a/gnu/packages/maths.scm
+++ b/gnu/packages/maths.scm
@@ -934,6 +934,46 @@ (define-public lapack
     (license (license:non-copyleft "file://LICENSE"
                                 "See LICENSE in the distribution."))))
 
+(define-public lis
+  (package
+   (name "lis")
+   (version "2.0.34")
+   (source (origin
+            (method url-fetch)
+            (uri (string-append "https://www.ssisc.org/lis/dl/lis-"
+                                version ".zip"))
+            (sha256
+             (base32 "0wnsvy75rcxgy0afckyr9bi4hfrk0wsqczw5az95afd9j8ymj5pj"))))
+   (build-system gnu-build-system)
+   (arguments
+    (list
+     #:configure-flags #~(list "--enable-fortran"
+                               "--enable-f90"
+                               "--enable-openmp"
+                               "--enable-complex"
+                               "--disable-sse2" ;; XXX: tuning
+                               "--enable-shared")
+     #:phases
+     #~(modify-phases %standard-phases
+         (add-after 'install 'install-doc
+           (lambda* (#:key outputs #:allow-other-keys)
+             (let* ((out (assoc-ref outputs "out"))
+                    (docdir (string-append out "/share/doc/lis-" #$version))
+                    (mandir (string-append out "/share/man")))
+
+               (copy-recursively "doc/man" mandir)
+               ;; TODO: Build the manuals ourselves
+               (install-file "doc/lis-ug-en.pdf" docdir)
+               (install-file "doc/lis-ug-ja.pdf" docdir)))))))
+   (inputs (list openmpi))
+   (native-inputs (list gfortran unzip))
+   (home-page "https://www.ssisc.org/lis")
+   (synopsis "Solve discretized linear equations and eigenvalue problems")
+   (description "Lis is a parallel software library for solving discretized
+linear equations and eigenvalue problems that arise in the numerical solution
+of partial differential equations using iterative methods.")
+   (license license:bsd-3)))
+
 (define-public clapack
   (package
     (name "clapack")
-- 
2.36.1

Message #8 received at 55901 <at> debbugs.gnu.org (full text, mbox):

From: Andreas Enge <andreas <at> enge.fr>
To: Liliana Marie Prikler <liliana.prikler <at> gmail.com>
Cc: 55901 <at> debbugs.gnu.org, Sharlatan Hellseher <sharlatanus <at> gmail.com>,
 Eric Bavier <bavier <at> posteo.net>
Subject: Re: [PATCH WIP] gnu: Add lis.
Date: Wed, 26 Mar 2025 10:21:53 +0100

Hello Liliana,

going through open science team issues, I come upon this one.
Is it still of interest? If yes, could you propose an updated patch
for the latest version?
Otherwise I would suggest to close this issue.

Andreas

Message #11 received at 55901 <at> debbugs.gnu.org (full text, mbox):

From: Liliana Marie Prikler <liliana.prikler <at> gmail.com>
To: 55901 <at> debbugs.gnu.org
Cc: Andreas Enge <andreas <at> enge.fr>, Sharlatan Hellseher <sharlatanus <at> gmail.com>,
 Eric Bavier <bavier <at> posteo.net>
Subject: [PATCH v2] gnu: Add lis.
Date: Sat, 11 Jun 2022 09:49:23 +0200

* gnu/packages/maths.scm (lis): New variable.
---
Hi Andreas,

thanks for looking into this.  It's not been in my focus for a while,
because I'm not dealing with linear algebra daily, but given that the
package still sees development, I think it's worthwile to be included
in Guix.

I've updated the version and kept everything else about the package the
same.  Tests still run fine on my machine.

Cheers

 gnu/packages/maths.scm | 40 ++++++++++++++++++++++++++++++++++++++++
 1 file changed, 40 insertions(+)

diff --git a/gnu/packages/maths.scm b/gnu/packages/maths.scm
index 8baf3a0a7f..27c30f223e 100644
--- a/gnu/packages/maths.scm
+++ b/gnu/packages/maths.scm
@@ -1198,6 +1198,46 @@ (define-public lapack
                                 "See LICENSE in the distribution."))
     (properties '((tunable? . #t)))))
 
+(define-public lis
+  (package
+   (name "lis")
+   (version "2.1.8")
+   (source (origin
+            (method url-fetch)
+            (uri (string-append "https://www.ssisc.org/lis/dl/lis-"
+                                version ".zip"))
+            (sha256
+             (base32 "0nh2593xkcdv1c3gmj7i64ca393nn0ngqfl522yiwbidh9dvd1nl"))))
+   (build-system gnu-build-system)
+   (arguments
+    (list
+     #:configure-flags #~(list "--enable-fortran"
+                               "--enable-f90"
+                               "--enable-openmp"
+                               "--enable-complex"
+                               "--disable-sse2" ;; XXX: tuning
+                               "--enable-shared")
+     #:phases
+     #~(modify-phases %standard-phases
+         (add-after 'install 'install-doc
+           (lambda* (#:key outputs #:allow-other-keys)
+             (let* ((out (assoc-ref outputs "out"))
+                    (docdir (string-append out "/share/doc/lis-" #$version))
+                    (mandir (string-append out "/share/man")))
+
+               (copy-recursively "doc/man" mandir)
+               ;; TODO: Build the manuals ourselves
+               (install-file "doc/lis-ug-en.pdf" docdir)
+               (install-file "doc/lis-ug-ja.pdf" docdir)))))))
+   (inputs (list openmpi))
+   (native-inputs (list gfortran unzip))
+   (home-page "https://www.ssisc.org/lis")
+   (synopsis "Solve discretized linear equations and eigenvalue problems")
+   (description "Lis is a parallel software library for solving discretized
+linear equations and eigenvalue problems that arise in the numerical solution
+of partial differential equations using iterative methods.")
+   (license license:bsd-3)))
+
 (define-public clapack
   (package
     (name "clapack")

base-commit: dbef60edb356246855ad6749936ee511fc1a9b4b
-- 
2.49.0

Message #14 received at 55901 <at> debbugs.gnu.org (full text, mbox):

From: Andreas Enge <andreas <at> enge.fr>
To: Liliana Marie Prikler <liliana.prikler <at> gmail.com>
Cc: 55901 <at> debbugs.gnu.org, Sharlatan Hellseher <sharlatanus <at> gmail.com>,
 Eric Bavier <bavier <at> posteo.net>
Subject: Re: [PATCH v2] gnu: Add lis.
Date: Wed, 26 Mar 2025 13:41:34 +0100

[Message part 1 (text/plain, inline)]

Hi again,

your new message was dated to 2022, which is a bit surprising :)

The patch looks good, I have made a small modification to avoid the
assoc-ref, see attached. Please feel free to push.

The project itself has a strange organisation: Their git repo is on
github, but there the tags stop at version 2.0.30, and latest releases
exist only as zip files (while they had tar.gz files in the past).

Andreas

[0001-gnu-Add-lis.patch (text/plain, attachment)]

Message #17 received at 55901 <at> debbugs.gnu.org (full text, mbox):

From: Liliana Marie Prikler <liliana.prikler <at> gmail.com>
To: Andreas Enge <andreas <at> enge.fr>
Cc: 55901 <at> debbugs.gnu.org, Sharlatan Hellseher <sharlatanus <at> gmail.com>,
 Eric Bavier <bavier <at> posteo.net>
Subject: Re: [PATCH v2] gnu: Add lis.
Date: Wed, 26 Mar 2025 14:08:38 +0100

Am Mittwoch, dem 26.03.2025 um 13:41 +0100 schrieb Andreas Enge:
> your new message was dated to 2022, which is a bit surprising :)
Yeah, I did not reset the author date.

> The patch looks good, I have made a small modification to avoid the
> assoc-ref, see attached. Please feel free to push.
Did you send the right file?  It lools like the assoc-ref is still
there.  I know we could use #$output, but I don't feel too strongly
about it.  WDYT?

Cheers

Message #20 received at 55901 <at> debbugs.gnu.org (full text, mbox):

From: Andreas Enge <andreas <at> enge.fr>
To: Liliana Marie Prikler <liliana.prikler <at> gmail.com>
Cc: 55901 <at> debbugs.gnu.org, Sharlatan Hellseher <sharlatanus <at> gmail.com>,
 Eric Bavier <bavier <at> posteo.net>
Subject: Re: [PATCH v2] gnu: Add lis.
Date: Wed, 26 Mar 2025 14:44:32 +0100

[Message part 1 (text/plain, inline)]

Am Wed, Mar 26, 2025 at 02:08:38PM +0100 schrieb Liliana Marie Prikler:
> > The patch looks good, I have made a small modification to avoid the
> > assoc-ref, see attached. Please feel free to push.
> Did you send the right file?  It lools like the assoc-ref is still
> there.  I know we could use #$output, but I don't feel too strongly
> about it.  WDYT?

Ah, sorry, I had probably forgotten to commit the local changes. Here it is.
I would say that #$output goes together with all the "modernising" changes
to packaging such as gexps, not using assoc-ref, not ending a phase with #t,
so I would prefer it.

Andreas

[0001-gnu-Add-lis.patch (text/plain, attachment)]

Message #25 received at 55901-done <at> debbugs.gnu.org (full text, mbox):

From: Liliana Marie Prikler <liliana.prikler <at> gmail.com>
To: Andreas Enge <andreas <at> enge.fr>
Cc: 55901-done <at> debbugs.gnu.org, Sharlatan Hellseher <sharlatanus <at> gmail.com>,
 Eric Bavier <bavier <at> posteo.net>
Subject: Re: [PATCH v2] gnu: Add lis.
Date: Wed, 26 Mar 2025 22:59:50 +0100

Am Mittwoch, dem 26.03.2025 um 14:44 +0100 schrieb Andreas Enge:
> Am Wed, Mar 26, 2025 at 02:08:38PM +0100 schrieb Liliana Marie
> Prikler:
> > > The patch looks good, I have made a small modification to avoid
> > > the
> > > assoc-ref, see attached. Please feel free to push.
> > Did you send the right file?  It lools like the assoc-ref is still
> > there.  I know we could use #$output, but I don't feel too strongly
> > about it.  WDYT?
> 
> Ah, sorry, I had probably forgotten to commit the local changes. Here
> it is.  I would say that #$output goes together with all the
> "modernising" changes to packaging such as gexps, not using assoc-
> ref, not ending a phase with #t, so I would prefer it.
Fair enough.

Pushed with these changes.

Previous Next

GNU bug tracking system
Copyright (C) 1999 Darren O. Benham, 1997,2003 nCipher Corporation Ltd, 1994-97 Ian Jackson.