961183">

Debian Bug report logs - #961183
metis: providing a 64-bit build

Package: src:metis; Maintainer for src:metis is Debian Science Maintainers <[email protected]>;

Affects: src:parmetis, src:gmsh, src:netgen, src:syrthes, src:deal.ii, src:superlu-dist, src:freefem , libscotchmetis-dev, libdeal.ii-dev, libparmetis-dev

Reported by: Drew Parsons <[email protected]>

Date: Thu, 21 May 2020 03:18:02 UTC

Severity: normal

Blocking fix for 989497: superlu-dist: provide 64-bit build

Reply or subscribe to this bug.

View this report as an mbox folder, status mbox, maintainer mbox


Report forwarded to [email protected], Debian Science Maintainers <[email protected]>:
Bug#961183; Package src:metis. (Thu, 21 May 2020 03:18:04 GMT) (full text, mbox, link).


Acknowledgement sent to Drew Parsons <[email protected]>:
New Bug report received and forwarded. Copy sent to Debian Science Maintainers <[email protected]>. (Thu, 21 May 2020 03:18:04 GMT) (full text, mbox, link).


Message #5 received at [email protected] (full text, mbox, reply):

From: Drew Parsons <[email protected]>
To: Debian Bug Tracking System <[email protected]>
Subject: metis: providing a 64-bit build
Date: Thu, 21 May 2020 11:13:59 0800
Source: metis
Severity: normal

We're discussing introducing a 64 bit-build for the computational
stack. This refers to 64-bit addressing to cells in meshes, etc.

c.f. Bug#953116 (petsc)
Bug#961108 (openmpi)
https://lists.debian.org/debian-science/2020/05/msg00051.html

BLAS is already 64-bit enabled.

For the 64-bit environment to work, it needs to be carried through all
along the library stack, from BLAS to PETSc.

MUMPS is in the middle. Its docs indicate that its 64-bit support
requires METIS to be 64-bit enabled.

Probably we want 2 separate builds, 32-bit (the current libmetis-dev)
and a separate 64-bit libmetis64-dev.

Opening this bug to track the chain of 64-bit package dependencies.



Information forwarded to [email protected], Debian Science Maintainers <[email protected]>:
Bug#961183; Package src:metis. (Thu, 21 May 2020 04:36:03 GMT) (full text, mbox, link).


Acknowledgement sent to [email protected]:
Extra info received and forwarded to list. Copy sent to Debian Science Maintainers <[email protected]>. (Thu, 21 May 2020 04:36:03 GMT) (full text, mbox, link).


Message #10 received at [email protected] (full text, mbox, reply):

From: Drew Parsons <[email protected]>
To: [email protected]
Subject: Re: Bug#961183: metis: providing a 64-bit build
Date: Thu, 21 May 2020 12:24:31 0800
Should add a caveat on the urgency: this bug report is Wishlist, really.

That's because mumps uses scotch rather than metis. This is because 
parmetis is not available (not-free).

If the licence for parmetis became free, then we would use it with mumps 
(and would be looking for a 64-bit build).

Drew



Added indication that bug 961183 blocks 989497 Request was from Drew Parsons <[email protected]> to [email protected]. (Sat, 05 Jun 2021 11:33:04 GMT) (full text, mbox, link).


Information forwarded to [email protected], Debian Science Maintainers <[email protected]>:
Bug#961183; Package src:metis. (Sat, 05 Jun 2021 11:36:03 GMT) (full text, mbox, link).


Acknowledgement sent to Drew Parsons <[email protected]>:
Extra info received and forwarded to list. Copy sent to Debian Science Maintainers <[email protected]>. (Sat, 05 Jun 2021 11:36:03 GMT) (full text, mbox, link).


Message #17 received at [email protected] (full text, mbox, applications by moving the standard header from /usr/include/metis.h > to /usr/include/metis/metis.h. Then client applications will need to > be updated to add -I/usr/include/metis when using the standard build. > > &In-Reply-To=<162289275513.995340.15366191859034122303.reportbug@sandy>">reply):

From: Drew Parsons <[email protected]>
To: Debian Bug Tracking System <[email protected]>
Subject: Re: metis: providing a 64-bit build
Date: Sat, 05 Jun 2021 13:32:35 0200
Package: metis
Followup-For: Bug #961183

64-bit integers are now available for several libraries.

64-bit PETSc is able to use 64-bit SuperLU-Dist, which we don't
currently have.

64-bit SuperLU-Dist requires 64-bit metis.

In principle 64-bit metis is simple to configure: set IDXTYPEWIDTH to
64 in metis.h.

The challenge that we need to address is the location of metis.h.
Currently it's located in /usr/include/metis.h (for the standard 32-bit build)

To distinguish between standard and 64-bit builds, we'd probably want to place
the 64-bit header in /usr/include/metis64/metis.h

-I/usr/include/metis64 can then be used by client applications. The
remaining challenge is that /usr/include/metis.h might be found
instead of /usr/include/metis64/metis.h, since -I/usr/include will
also often be used in build scripts.

I think the best solution is to "break" the builds of client
applications by moving the standard header from /usr/include/metis.h
to /usr/include/metis/metis.h.  Then client applications will need to
be updated to add -I/usr/include/metis when using the standard build.



Information forwarded to [email protected], Debian Science Maintainers <[email protected]>:
Bug#961183; Package src:metis. (Sat, 05 Jun 2021 15:09:02 GMT) (full text, mbox, link).


Acknowledgement sent to Anton Gladky <[email protected]>:
Extra info received and forwarded to list. Copy sent to Debian Science Maintainers <[email protected]>. (Sat, 05 Jun 2021 15:09:02 GMT) (full text, mbox, link).


Message #22 received at [email protected] (full text, mbox, > applications by moving the standard header from /usr/include/metis.h > > to /usr/include/metis/metis.h. Then client applications will need to > > be updated to add -I/usr/include/metis when using the standard build. > > ">reply):

From: Anton Gladky <[email protected]>
To: Drew Parsons <[email protected]>, [email protected]
Subject: Re: Bug#961183: metis: providing a 64-bit build
Date: Sat, 5 Jun 2021 17:07:20 0200
[Message part 1 (text/plain, inline)]
Well, theoretically we can do it. But we need to do some kind
of a transition for all dependencies.

Regards

Anton


Am Sa., 5. Juni 2021 um 13:36 Uhr schrieb Drew Parsons <[email protected]
>:

> ...<skip>
> I think the best solution is to "break" the builds of client
> applications by moving the standard header from /usr/include/metis.h
> to /usr/include/metis/metis.h.  Then client applications will need to
> be updated to add -I/usr/include/metis when using the standard build.
>
[Message part 2 (text/html, inline)]

Information forwarded to [email protected], Debian Science Maintainers <[email protected]>:
Bug#961183; Package src:metis. (Sat, 05 Jun 2021 15:27:02 GMT) (full text, mbox, link).


Acknowledgement sent to [email protected]:
Extra info received and forwarded to list. Copy sent to Debian Science Maintainers <[email protected]>. (Sat, 05 Jun 2021 15:27:02 GMT) (full text, mbox, link).


Message #27 received at [email protected] (full text, mbox, >> applications by moving the standard header from /usr/include/metis.h > >> to /usr/include/metis/metis.h. Then client applications will need > >> to > >> be updated to add -I/usr/include/metis when using the standard > >> build. > > &In-Reply-To=&subject=Re: Bug#961183: metis: providing a 64-bit build&References=<159003083998.19768.1398407870629260980.reportbug@monte.emerall.com> <162289275513.995340.15366191859034122303.reportbug@sandy> ">reply):

From: Drew Parsons <[email protected]>
To: Anton Gladky <[email protected]>
Cc: [email protected]
Subject: Re: Bug#961183: metis: providing a 64-bit build
Date: Sat, 05 Jun 2021 17:21:18 0200
Doing it through a formal transition request is sensible. There's a 
backlog of numerical library upgrades (some still in the NEW queue) 
waiting for bullseye to get released. Perhaps we can run this transition 
at the same time as the others, process them all as a bulk transition.

We'll set it up in experimental first, of course.

I'll prepare the list of affected packages.

Drew



On 2021-06-05 17:07, Anton Gladky wrote:
> Well, theoretically we can do it. But we need to do some kind
> of a transition for all dependencies.
> 
> Regards
> 
> Anton
> 
> Am Sa., 5. Juni 2021 um 13:36 Uhr schrieb Drew Parsons
> <[email protected]>:
> 
>> ...<skip>
>> I think the best solution is to "break" the builds of client
>> applications by moving the standard header from /usr/include/metis.h
>> to /usr/include/metis/metis.h.  Then client applications will need
>> to
>> be updated to add -I/usr/include/metis when using the standard
>> build.



Information forwarded to [email protected], Debian Science Maintainers <[email protected]>:
Bug#961183; Package src:metis. (Sat, 05 Jun 2021 16:03:03 GMT) (full text, mbox, link).


Acknowledgement sent to Drew Parsons <[email protected]>:
Extra info received and forwarded to list. Copy sent to Debian Science Maintainers <[email protected]>. (Sat, 05 Jun 2021 16:03:03 GMT) (full text, mbox, link).


Message #32 received at [email protected] (full text, mbox, reply):

From: Drew Parsons <[email protected]>
To: Debian Bug Tracking System <[email protected]>
Subject: Re: metis: providing a 64-bit build
Date: Sat, 05 Jun 2021 17:59:41 0200
Package: metis
Followup-For: Bug #961183
Control: affects -1 libdeal.ii-dev libparmetis-dev src:suitesparse src:deal.ii src:freefem   src:gmsh src:netgen src:parmetis src:superlu-dist src:syrthes

Here is the list of affected packages identified by
apt-rdepends -r libmetis-dev libmetis5
    
libmetis-dev
  Reverse Depends: libdeal.ii-dev (9.2.0-3 b2)
  Reverse Depends: libparmetis-dev (4.0.3-5 b1)

libmetis5
  Reverse Depends: libcholmod3 (>= 1:5.8.1 dfsg-2)
  Reverse Depends: libdeal.ii-9.2.0 (>= 9.2.0-3 b2)
  Reverse Depends: libfreefem   (>= 3.61.1 dfsg1-6)
  Reverse Depends: libgmsh4 (>= 4.7.1 ds1-5)
  Reverse Depends: libmetis-dev (= 5.1.0.dfsg-7)
  Reverse Depends: libnglib-6.2 (>= 6.2.2006 really6.2.1905 dfsg-2.1)
  Reverse Depends: libparmetis4.0 (4.0.3-5 b1)
  Reverse Depends: libsuperlu-dist7 (>= 7.0.0 dfsg1-1exp2)
  Reverse Depends: metis (= 5.1.0.dfsg-7)
  Reverse Depends: parmetis-test (4.0.3-5 b1)
  Reverse Depends: pyfr (1.5.0-3)
  Reverse Depends: syrthes-tools (>= 4.3.5 20200129-dfsg1-1 b1)


I'm applying affects tags to this bug to let the maintainers know what
we're planning.

Drew



Added indication that 961183 affects libdeal.ii-dev, libparmetis-dev, src:suitesparse, src:deal.ii, src:freefem , src:gmsh, src:netgen, src:parmetis, src:superlu-dist, and src:syrthes Request was from Drew Parsons <[email protected]> to [email protected]. (Sat, 05 Jun 2021 16:03:03 GMT) (full text, mbox, link).


Information forwarded to [email protected], Debian Science Maintainers <[email protected]>:
Bug#961183; Package src:metis. (Sat, 05 Jun 2021 16:06:03 GMT) (full text, mbox, link).


Acknowledgement sent to Drew Parsons <[email protected]>:
Extra info received and forwarded to list. Copy sent to Debian Science Maintainers <[email protected]>. (Sat, 05 Jun 2021 16:06:03 GMT) (full text, mbox, link).


Message #39 received at [email protected] (full text, mbox, reply):

From: Drew Parsons <[email protected]>
To: Debian Bug Tracking System <[email protected]>
Subject: Re: metis: providing a 64-bit build
Date: Sat, 05 Jun 2021 18:03:35 0200
Package: metis
Followup-For: Bug #961183
Control: affects -1 libscotchmetis-dev

Also scotch-metis will need to be updated to match.



Added indication that 961183 affects libscotchmetis-dev Request was from Drew Parsons <[email protected]> to [email protected]. (Sat, 05 Jun 2021 16:06:03 GMT) (full text, mbox, link).


Added indication that bug 961183 blocks 989550 Request was from Drew Parsons <[email protected]> to [email protected]. (Mon, 07 Jun 2021 11:00:04 GMT) (full text, mbox, link).


Information forwarded to [email protected], Debian Science Maintainers <[email protected]>:
Bug#961183; Package src:metis. (Sat, 12 Jun 2021 23:24:03 GMT) (full text, mbox, link).


Acknowledgement sent to Drew Parsons <[email protected]>:
Extra info received and forwarded to list. Copy sent to Debian Science Maintainers <[email protected]>. (Sat, 12 Jun 2021 23:24:03 GMT) (full text, mbox, link).


Message #48 received at [email protected] (full text, mbox, reply):

From: Drew Parsons <[email protected]>
To: Debian Bug Tracking System <[email protected]>
Subject: Re: metis: providing a 64-bit build
Date: Sun, 13 Jun 2021 01:20:31 0200
Package: metis
Followup-For: Bug #961183

Hi Anton, I've prepared packaging for a 64-bit that I'm happy to push
(and upload).

Would you like me to push to salsa (experimental branch) for inspection?

Drew



Information forwarded to [email protected], Debian Science Maintainers <[email protected]>:
Bug#961183; Package src:metis. (Sun, 13 Jun 2021 05:06:02 GMT) (full text, mbox, link).


Acknowledgement sent to Anton Gladky <[email protected]>:
Extra info received and forwarded to list. Copy sent to Debian Science Maintainers <[email protected]>. (Sun, 13 Jun 2021 05:06:02 GMT) (full text, mbox, link).


Message #53 received at [email protected] (full text, mbox, reply):

From: Anton Gladky <[email protected]>
To: Drew Parsons <[email protected]>, [email protected]
Subject: Re: Bug#961183: metis: providing a 64-bit build
Date: Sun, 13 Jun 2021 07:02:31 0200
[Message part 1 (text/plain, inline)]
Hi Drew,

> Would you like me to push to salsa (experimental branch) for inspection?

Yes, please.

Anton


Am So., 13. Juni 2021 um 01:24 Uhr schrieb Drew Parsons <[email protected]
>:

> Package: metis
> Followup-For: Bug #961183
>
> Hi Anton, I've prepared packaging for a 64-bit that I'm happy to push
> (and upload).
>
> Would you like me to push to salsa (experimental branch) for inspection?
>
> Drew
>
> --
> debian-science-maintainers mailing list
> [email protected]
>
> https://alioth-lists.debian.net/cgi-bin/mailman/listinfo/debian-science-maintainers
>
[Message part 2 (text/html, inline)]

Information forwarded to [email protected], Debian Science Maintainers <[email protected]>:
Bug#961183; Package src:metis. (Sun, 13 Jun 2021 08:48:02 GMT) (full text, mbox, link).


Acknowledgement sent to [email protected]:
Extra info received and forwarded to list. Copy sent to Debian Science Maintainers <[email protected]>. (Sun, 13 Jun 2021 08:48:02 GMT) (full text, mbox, link).


Message #58 received at [email protected] (full text, mbox, integers (indexing) not floating point numbers, so I figure we don't > need to change REALTYPEWIDTH as well. Or if we do want to change > REALTYPEWIDTH, then we might want to change it in the standard build > too. > > For comparison, SuperLU-Dist just uses double for values. Hypre uses > HYPRE_Real, which can be configured to single or double or longdouble > (default double). SCOTCH is mainly working with indices (integers > defined as SCOTCH_Num) but internally uses double more often than float > when it needs floating point numbers (it does use both). PETSc uses > PetscReal which can be configured single or double, float128 or fp16 > (default double). > > I figure it's simplest not to change REALTYPEWIDTH unless there's a > specific known need to make the change. We need to set IDXTYPEWIDTH to > 64 to support extremely large systems with billions of degrees of > freedom, but that doesn't necessarily mean the floating point value at > each index also requires higher precision. On the other hand changing to > REALTYPEWIDTH=64 (even in standard libmetis) might better match the > default of other packages. > > Apart from that question, I'm happy with the patch. As far as I'm > concerned it's ready to upload to experimental. > > > Drew > > &In-Reply-To=<[email protected]>">reply):

From: Drew Parsons <[email protected]>
To: Anton Gladky <[email protected]>
Cc: [email protected]
Subject: Re: Bug#961183: metis: providing a 64-bit build
Date: Sun, 13 Jun 2021 10:35:12 0200
On 2021-06-13 07:02, Anton Gladky wrote:
> Hi Drew,
> 
>> Would you like me to push to salsa (experimental branch) for
> inspection?
> 
> Yes, please.


OK, pushed to experimental branch.

Bring the question of REALTYPEWIDTH to mind (the size of floating point 
numbers). The "64-bit support" in other packages essentially concerns 
integers (indexing) not floating point numbers, so I figure we don't 
need to change REALTYPEWIDTH as well. Or if we do want to change 
REALTYPEWIDTH, then we might want to change it in the standard build 
too.

For comparison, SuperLU-Dist just uses double for values.  Hypre uses 
HYPRE_Real, which can be configured to single or double or longdouble 
(default double). SCOTCH is mainly working with indices (integers 
defined as SCOTCH_Num) but internally uses double more often than float 
when it needs floating point numbers (it does use both).  PETSc uses 
PetscReal which can be configured single or double, float128 or fp16 
(default double).

I figure it's simplest not to change REALTYPEWIDTH unless there's a 
specific known need to make the change. We need to set IDXTYPEWIDTH to 
64 to support extremely large systems with billions of degrees of 
freedom, but that doesn't necessarily mean the floating point value at 
each index also requires higher precision. On the other hand changing to 
REALTYPEWIDTH=64 (even in standard libmetis) might better match the 
default of other packages.

Apart from that question, I'm happy with the patch. As far as I'm 
concerned it's ready to upload to experimental.


Drew



Information forwarded to [email protected], Debian Science Maintainers <[email protected]>:
Bug#961183; Package src:metis. (Sun, 13 Jun 2021 09:09:02 GMT) (full text, mbox, link).


Acknowledgement sent to PICCA Frederic-Emmanuel <[email protected]>:
Extra info received and forwarded to list. Copy sent to Debian Science Maintainers <[email protected]>. (Sun, 13 Jun 2021 09:09:02 GMT) (full text, mbox, link).


Message #63 received at [email protected] (full text, mbox, reply):

From: PICCA Frederic-Emmanuel <[email protected]>
To: "[email protected]" <[email protected]>, "[email protected]" <[email protected]>, Anton Gladky <[email protected]>
Subject: RE:Bug#961183: metis: providing a 64-bit build
Date: Sun, 13 Jun 2021 08:58:38 0000
Hello,

why  there is no pkgconfig files provided with metis and metis64.
this simplify the configuration of packages depending on these libraries.

Cheers

Fred



Information forwarded to [email protected], Debian Science Maintainers <[email protected]>:
Bug#961183; Package src:metis. (Tue, 15 Jun 2021 20:48:03 GMT) (full text, mbox, link).


Acknowledgement sent to Drew Parsons <[email protected]>:
Extra info received and forwarded to list. Copy sent to Debian Science Maintainers <[email protected]>. (Tue, 15 Jun 2021 20:48:03 GMT) (full text, mbox, link).


Message #68 received at [email protected] (full text, mbox, reply):

From: Drew Parsons <[email protected]>
To: Debian Bug Tracking System <[email protected]>
Subject: Re: metis: providing a 64-bit build
Date: Tue, 15 Jun 2021 22:45:28 0200
Package: metis
Followup-For: Bug #961183

Hi Fred, I guess the simple reason why there's no pkgconfig for Metis
is that upstream does not provide one.

We could add one ourselves easily enough.  But SCOTCH-Metis
compatibility would want to be taken into consideration.  Historically
SCOTCH-Metis never had full compatibility (#506033, #653647), but
they're about to make a new release expected to address that.



Removed indication that bug 961183 blocks 989550 Request was from Sébastien Villemot <[email protected]> to [email protected]. (Sun, 28 May 2023 10:12:04 GMT) (full text, mbox, link).


Removed indication that 961183 affects src:suitesparse Request was from Sébastien Villemot <[email protected]> to [email protected]. (Sun, 28 May 2023 10:12:04 GMT) (full text, mbox, link).


Send a report that this bug log contains spam.


Debian bug tracking system administrator <[email protected]>. Last modified: Wed Nov 20 16:20:40 2024; Machine Name: bembo

Debian Bug tracking system

Debbugs is free software and licensed under the terms of the GNU Public License version 2. The current version can be obtained from https://bugs.debian.org/debbugs-source/.

Copyright © 1999 Darren O. Benham, 1997,2003 nCipher Corporation Ltd, 1994-97 Ian Jackson, 2005-2017 Don Armstrong, and many other contributors.