modules-interest — Modules discussions — NEW HOME: https://lists.hpsf.io/g/modules-interest

Re: [Modules] Environment Modules at HPSF conference 2026

[Xavier D. v. lists.hpsf.i. <xav...@li...>]

On Wednesday we had our very first Environment Modules Meeting at #HPSFcon.
Many thanks to all the attendees for the very good discussions we had. A
particular interest has been shown in variants, lua modulefile support and
integration with Spack.

Presentations are available at
https://hpsf2026.sched.com/overview/type/Environment+Modules+Project+Meeting

Regards,
Xavier

Le jeu. 19 févr. 2026 à 00:26, Xavier Delaruelle <
xav...@gm...> a écrit :

> 📣 The Environment Modules Project is hosting a *half-day session* at the *HPSF
> Conference* this March!
>
> Join us to connect, share best practices, and explore the future of
> software environment management in HPC.
>
> 🔗 Schedule & registration:
> https://events.linuxfoundation.org/hpsf-conference/
>
>


-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.
View/Reply Online (#24): https://lists.hpsf.io/g/modules-interest/message/24
Mute This Topic: https://lists.hpsf.io/mt/117888589/10150481
Group Owner: mod...@li...
Unsubscribe: https://lists.hpsf.io/g/modules-interest/unsub [mod...@li...]
-=-=-=-=-=-=-=-=-=-=-=-

Re: [Modules] Picking the right module after change in MODULEPATH

[Paul M. v. lists.hpsf.i. <pau...@li...>]

	
Did you try?:

module use "$GIT_ROOT/setup/localmodules"

Also, is your custom_verif specific to each .git repo, or could it be a single module in your standard location (in which case you wouldn't need to change MODULEPATH)?


Or, you could also create a module that does all this for you (and use module switch on that module).

So it could:
Determine the GIT_ROOT
If it has changed:
	in the unload stage:
		unload any modules that are from the old GIT_ROOT
		unuse the old GIT_ROOT
	in the load stage:
		use the new GIT_ROOT
		load the modules for the new GIT_ROOT


You could set up sections for load and unload that will only activate if GIT_ROOT has changed.



On 2026-03-10 3:45 AM, Laurent BESSON wrote:
> Hello Paul,
> 
> 
> 
> Thanks for your valueable feedback! I used to use "module switch" before but I found more elegant to use a simple "module load" (along with "conflict_unload") command. I'll revert to "module switch" then.
> 
> 
> 
> I found also "module reload" would do the trick too: If I add a "module --silent reload" right after changing the MODULEPATH, it works too
> 
> 
> 
> And I had another idea after reading the documentation about MODULEPATH. The doc says:
> 
> 
> 
> Path elements registered in the MODULEPATH environment variable may contain reference to environment variables which are converted to their corresponding value by module command each time it looks at the MODULEPATH value. If an environment variable referred in a path element is not defined, its reference is converted to an empty string.
> 
> 
> 
> So I thought I could place into a variable, say GIT_ROOT, what is returned by `git rev-parse --show-toplevel` and then « module use \$GIT_ROOT/setup/localmodules » but the backslash is not sufficient and I have not found a proper syntax under tcsh to use this trick. Note : GIT_ROOT variable is updated , in my env, when I changed my working directory.
> 
> 
> 
> 
> 
> Last but not least, I'm wondering if a "reload on change" feature would be useful ?
> 
> In this mode, "module load" would analyze the signature of module file (md5sum or something like that) and reload it if signature is different, meaning file has changed, from currently loaded module. The signature could also take into account the path where the module file is found. In that case, it would be another way to solve my issue where the module is found in another directory.
> 
> 
> 
> Of course, such a feature would have an impact on "load" mechanism and should not be "on" by default.
> 
> 
> 
> What do you think ?
> 
> 
> 
> Best regards,
> 
> Laurent.
> 
> 
> 
> -----Message d'origine-----
> De : mod...@li... <mod...@li...> De la part de Paul Markfort via lists.hpsf.io
> Envoyé : lundi 9 mars 2026 17:42
> À : mod...@li...
> Objet : Re: [Modules] Picking the right module after change in MODULEPATH
> 
> 
> 
> Try:
> 
> 
> 
> module switch custom_verif custom_verif
> 
> 
> 
> Make sure you specify custom_verif twice (or it might unload some other module).
> 
> Also module switch will act as a module load if the module to unload is not loaded.
> 
> 
> 
> The issue, is the module command does not load a module that is already loaded (it doesn't know it is a different module).
> 
> 
> 
> 
> 
> A better way to handle this would be (since the two custom_verif modules likely make different changes):
> 
> module unload custom_verif
> 
> cd <new_folder>
> 
> module use --prepend `git rev-parse --show-toplevel`/setup/localmodules
> 
> module load custom_verif
> 
> 
> 
> Note: module unload won't complain if custom_verif is not loaded.
> 
> 
> 
> 
> 
> On 2026-03-09 11:22 AM, Laurent BESSON via lists.hpsf.io wrote:
> 
>> Hello dear Modules community,
> 
>>
> 
>> I have an issue with Modules not loading the « expected » modules files in my unix env. When I say « expected », I mean « the way I thought or would like to behav » 😊.
> 
>> Let me explain the context :
> 
>>
> 
>>
> 
>>     *   Modules version 5.5.0
> 
>>     *   Modules special configuration :  module config conflict_unload 1
> 
>>
> 
>> My unix setup script is under Git and I have Modules files under this Git repo. Here is the beginning of this script :
> 
>>
> 
>> Cd <main branch>
> 
>> module use --prepend `git rev-parse
> 
>> --show-toplevel`/setup/localmodules
> 
>> module load custom_verif
> 
>>
> 
>> The « custom_verif » Modules is under Git and is located in <git
> 
>> root>/setup/localmodules
> 
>>
> 
>> Everything works fine when I start from a fresh terminal/shell or as long as I stay in the same directory.
> 
>>
> 
>> Now, the trouble comes when I change directory. Let’s say I have cloned another branch of the same project in another directory. I setup my shell to work in this new directory :
> 
>>
> 
>> Cd <some branch directory>
> 
>> module use --prepend `git rev-parse
> 
>> --show-toplevel`/setup/localmodules
> 
>> module load custom_verif
> 
>>
> 
>> In that case, « custom_verif » module is not unloaded/reloaded and I
> 
>> keep the content of « custom_verif » as it has been defined in the «
> 
>> main » directory but not in the « some branch » directory. Said
> 
>> another way, I see « custom_verif » in the $LOADEDMODULES variable
> 
>> (this is normal and expected)  and I see the old  « <main
> 
>> branch>/setup/localmodules/custom_verif » in $_LMFILES_. I was
> 
>> expecting to see « <some branch>/setup/localmodules/custom_verif » in
> 
>> this $_LMFILES_
> 
>>
> 
>> I was wondering if there is a trick/setup/configuration in Modules to make it understand the « custom_verif » module MUST BE reloaded in the new context (after the MODULEPATH have changed through « module use ») and must be used from the « branch » directory ?
> 
>>
> 
>> Thanks for your help,
> 
>> Laurent.
> 
>>
> 
>>
> 
>>
> 
>>
> 
>>
> 
>>
> 
>>
> 
> 
> 
> --
> 
> --------------------------------------------------------
> 
> The views and opinions expressed above are strictly those of the author(s).  The content of this message has not been reviewed nor approved by any entity whatsoever.
> 
> --------------------------------------------------------
> 
> Paul FM         Info: http://paulfm.com/~paulfm/
> 
> --------------------------------------------------------
> 
> 
> 
> 
> 
> 
> 
> 
> 
> 

-- 
--------------------------------------------------------
The views and opinions expressed above are strictly
those of the author(s).  The content of this message has
not been reviewed nor approved by any entity whatsoever.
--------------------------------------------------------
Paul FM         Info: http://paulfm.com/~paulfm/
--------------------------------------------------------


-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.
View/Reply Online (#23): https://lists.hpsf.io/g/modules-interest/message/23
Mute This Topic: https://lists.hpsf.io/mt/118224577/10150481
Group Owner: mod...@li...
Unsubscribe: https://lists.hpsf.io/g/modules-interest/unsub [mod...@li...]
-=-=-=-=-=-=-=-=-=-=-=-

Re: [Modules] Picking the right module after change in MODULEPATH

[Laurent B. v. lists.hpsf.i. <lau...@li...>]

Hello Paul,



Thanks for your valueable feedback! I used to use "module switch" before but I found more elegant to use a simple "module load" (along with "conflict_unload") command. I'll revert to "module switch" then.



I found also "module reload" would do the trick too: If I add a "module --silent reload" right after changing the MODULEPATH, it works too



And I had another idea after reading the documentation about MODULEPATH. The doc says:



Path elements registered in the MODULEPATH environment variable may contain reference to environment variables which are converted to their corresponding value by module command each time it looks at the MODULEPATH value. If an environment variable referred in a path element is not defined, its reference is converted to an empty string.



So I thought I could place into a variable, say GIT_ROOT, what is returned by `git rev-parse --show-toplevel` and then « module use \$GIT_ROOT/setup/localmodules » but the backslash is not sufficient and I have not found a proper syntax under tcsh to use this trick. Note : GIT_ROOT variable is updated , in my env, when I changed my working directory.





Last but not least, I'm wondering if a "reload on change" feature would be useful ?

In this mode, "module load" would analyze the signature of module file (md5sum or something like that) and reload it if signature is different, meaning file has changed, from currently loaded module. The signature could also take into account the path where the module file is found. In that case, it would be another way to solve my issue where the module is found in another directory.



Of course, such a feature would have an impact on "load" mechanism and should not be "on" by default.



What do you think ?



Best regards,

Laurent.



-----Message d'origine-----
De : mod...@li... <mod...@li...> De la part de Paul Markfort via lists.hpsf.io
Envoyé : lundi 9 mars 2026 17:42
À : mod...@li...
Objet : Re: [Modules] Picking the right module after change in MODULEPATH



Try:



module switch custom_verif custom_verif



Make sure you specify custom_verif twice (or it might unload some other module).

Also module switch will act as a module load if the module to unload is not loaded.



The issue, is the module command does not load a module that is already loaded (it doesn't know it is a different module).





A better way to handle this would be (since the two custom_verif modules likely make different changes):

module unload custom_verif

cd <new_folder>

module use --prepend `git rev-parse --show-toplevel`/setup/localmodules

module load custom_verif



Note: module unload won't complain if custom_verif is not loaded.





On 2026-03-09 11:22 AM, Laurent BESSON via lists.hpsf.io wrote:

> Hello dear Modules community,

>

> I have an issue with Modules not loading the « expected » modules files in my unix env. When I say « expected », I mean « the way I thought or would like to behav » 😊.

> Let me explain the context :

>

>

>    *   Modules version 5.5.0

>    *   Modules special configuration :  module config conflict_unload 1

>

> My unix setup script is under Git and I have Modules files under this Git repo. Here is the beginning of this script :

>

> Cd <main branch>

> module use --prepend `git rev-parse

> --show-toplevel`/setup/localmodules

> module load custom_verif

>

> The « custom_verif » Modules is under Git and is located in <git

> root>/setup/localmodules

>

> Everything works fine when I start from a fresh terminal/shell or as long as I stay in the same directory.

>

> Now, the trouble comes when I change directory. Let’s say I have cloned another branch of the same project in another directory. I setup my shell to work in this new directory :

>

> Cd <some branch directory>

> module use --prepend `git rev-parse

> --show-toplevel`/setup/localmodules

> module load custom_verif

>

> In that case, « custom_verif » module is not unloaded/reloaded and I

> keep the content of « custom_verif » as it has been defined in the «

> main » directory but not in the « some branch » directory. Said

> another way, I see « custom_verif » in the $LOADEDMODULES variable

> (this is normal and expected)  and I see the old  « <main

> branch>/setup/localmodules/custom_verif » in $_LMFILES_. I was

> expecting to see « <some branch>/setup/localmodules/custom_verif » in

> this $_LMFILES_

>

> I was wondering if there is a trick/setup/configuration in Modules to make it understand the « custom_verif » module MUST BE reloaded in the new context (after the MODULEPATH have changed through « module use ») and must be used from the « branch » directory ?

>

> Thanks for your help,

> Laurent.

>

>

>

>

>

>

>



--

--------------------------------------------------------

The views and opinions expressed above are strictly those of the author(s).  The content of this message has not been reviewed nor approved by any entity whatsoever.

--------------------------------------------------------

Paul FM         Info: http://paulfm.com/~paulfm/

--------------------------------------------------------












-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.
View/Reply Online (#22): https://lists.hpsf.io/g/modules-interest/message/22
Mute This Topic: https://lists.hpsf.io/mt/118224577/10150481
Group Owner: mod...@li...
Unsubscribe: https://lists.hpsf.io/g/modules-interest/unsub [mod...@li...]
-=-=-=-=-=-=-=-=-=-=-=-

Re: [Modules] Picking the right module after change in MODULEPATH

[Paul M. v. lists.hpsf.i. <pau...@li...>]

Try:

module switch custom_verif custom_verif

Make sure you specify custom_verif twice (or it might unload some other module).
Also module switch will act as a module load if the module to unload is not loaded.

The issue, is the module command does not load a module that is already loaded (it doesn't know it is a different module).


A better way to handle this would be (since the two custom_verif modules likely make different changes):
module unload custom_verif
cd <new_folder>
module use --prepend `git rev-parse --show-toplevel`/setup/localmodules
module load custom_verif

Note: module unload won't complain if custom_verif is not loaded.


On 2026-03-09 11:22 AM, Laurent BESSON via lists.hpsf.io wrote:
> Hello dear Modules community,
> 
> I have an issue with Modules not loading the « expected » modules files in my unix env. When I say « expected », I mean « the way I thought or would like to behav » 😊.
> Let me explain the context :
> 
> 
>    *   Modules version 5.5.0
>    *   Modules special configuration :  module config conflict_unload 1
> 
> My unix setup script is under Git and I have Modules files under this Git repo. Here is the beginning of this script :
> 
> Cd <main branch>
> module use --prepend `git rev-parse --show-toplevel`/setup/localmodules
> module load custom_verif
> 
> The « custom_verif » Modules is under Git and is located in <git root>/setup/localmodules
> 
> Everything works fine when I start from a fresh terminal/shell or as long as I stay in the same directory.
> 
> Now, the trouble comes when I change directory. Let’s say I have cloned another branch of the same project in another directory. I setup my shell to work in this new directory :
> 
> Cd <some branch directory>
> module use --prepend `git rev-parse --show-toplevel`/setup/localmodules
> module load custom_verif
> 
> In that case, « custom_verif » module is not unloaded/reloaded and I keep the content of « custom_verif » as it has been defined in the « main » directory but not in the « some branch » directory. Said another way, I see « custom_verif » in the $LOADEDMODULES variable (this is normal and expected)  and I see the old  « <main branch>/setup/localmodules/custom_verif » in $_LMFILES_. I was expecting to see « <some branch>/setup/localmodules/custom_verif » in this $_LMFILES_
> 
> I was wondering if there is a trick/setup/configuration in Modules to make it understand the « custom_verif » module MUST BE reloaded in the new context (after the MODULEPATH have changed through « module use ») and must be used from the « branch » directory ?
> 
> Thanks for your help,
> Laurent.
> 
> 
> 
> 
> 
> 
> 

-- 
--------------------------------------------------------
The views and opinions expressed above are strictly
those of the author(s).  The content of this message has
not been reviewed nor approved by any entity whatsoever.
--------------------------------------------------------
Paul FM         Info: http://paulfm.com/~paulfm/
--------------------------------------------------------


-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.
View/Reply Online (#21): https://lists.hpsf.io/g/modules-interest/message/21
Mute This Topic: https://lists.hpsf.io/mt/118224577/10150481
Group Owner: mod...@li...
Unsubscribe: https://lists.hpsf.io/g/modules-interest/unsub [mod...@li...]
-=-=-=-=-=-=-=-=-=-=-=-

[Modules] Picking the right module after change in MODULEPATH

[Laurent B. v. lists.hpsf.i. <lau...@li...>]

Hello dear Modules community,

I have an issue with Modules not loading the « expected » modules files in my unix env. When I say « expected », I mean « the way I thought or would like to behav » 😊.
Let me explain the context :


  *   Modules version 5.5.0
  *   Modules special configuration :  module config conflict_unload 1

My unix setup script is under Git and I have Modules files under this Git repo. Here is the beginning of this script :

Cd <main branch>
module use --prepend `git rev-parse --show-toplevel`/setup/localmodules
module load custom_verif

The « custom_verif » Modules is under Git and is located in <git root>/setup/localmodules

Everything works fine when I start from a fresh terminal/shell or as long as I stay in the same directory.

Now, the trouble comes when I change directory. Let’s say I have cloned another branch of the same project in another directory. I setup my shell to work in this new directory :

Cd <some branch directory>
module use --prepend `git rev-parse --show-toplevel`/setup/localmodules
module load custom_verif

In that case, « custom_verif » module is not unloaded/reloaded and I keep the content of « custom_verif » as it has been defined in the « main » directory but not in the « some branch » directory. Said another way, I see « custom_verif » in the $LOADEDMODULES variable (this is normal and expected)  and I see the old  « <main branch>/setup/localmodules/custom_verif » in $_LMFILES_. I was expecting to see « <some branch>/setup/localmodules/custom_verif » in this $_LMFILES_

I was wondering if there is a trick/setup/configuration in Modules to make it understand the « custom_verif » module MUST BE reloaded in the new context (after the MODULEPATH have changed through « module use ») and must be used from the « branch » directory ?

Thanks for your help,
Laurent.




-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.
View/Reply Online (#20): https://lists.hpsf.io/g/modules-interest/message/20
Mute This Topic: https://lists.hpsf.io/mt/118224577/10150481
Group Owner: mod...@li...
Unsubscribe: https://lists.hpsf.io/g/modules-interest/unsub [mod...@li...]
-=-=-=-=-=-=-=-=-=-=-=-

[Modules] Environment Modules at HPSF conference 2026

[Xavier D. v. lists.hpsf.i. <xav...@li...>]

📣 The Environment Modules Project is hosting a *half-day session* at the *HPSF
Conference* this March!

Join us to connect, share best practices, and explore the future of
software environment management in HPC.

🔗 Schedule & registration:
https://events.linuxfoundation.org/hpsf-conference/


-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.
View/Reply Online (#19): https://lists.hpsf.io/g/modules-interest/message/19
Mute This Topic: https://lists.hpsf.io/mt/117888589/10150481
Group Owner: mod...@li...
Unsubscribe: https://lists.hpsf.io/g/modules-interest/unsub [mod...@li...]
-=-=-=-=-=-=-=-=-=-=-=-

Re: [Modules] First Technical Steering Committee (TSC) meeting: February 18 at 5:00 PM CET

[Xavier D. v. lists.hpsf.i. <xav...@li...>]

Dear All,

Meeting notes are available at
https://github.com/envmodules/modules/wiki/TSC_meeting_notes_20260218
Next TSC meeting is planned on May 20 5:00 PM CEST

Regards,
Xavier

Le mer. 18 févr. 2026 à 15:55, Xavier Delaruelle via lists.hpsf.io
<xav...@li...> a écrit :

> Dear All,
>
> Reminder: our first TSC meeting will start in about 1 hour (5:00 PM CET).
> See https://github.com/envmodules/modules/wiki for connection details
>
> Regards,
> Xavier
>
> Le jeu. 5 févr. 2026 à 11:25, Xavier Delaruelle <
> xav...@gm...> a écrit :
>
>> Dear All,
>>
>> The first Technical Steering Committee (TSC) meeting for the Modules
>> project will take place on February 18 at 5:00 PM CET.
>>
>> This virtual session is open to anyone interested in the project —
>> contributors, users, and newcomers are all welcome!
>>
>> 📌 On the agenda: discussion of the roadmap for the upcoming version 5.7.
>>
>> 🔎 Learn more and join the conversation:
>> https://github.com/envmodules/modules/wiki
>>
>> Regards,
>> Modules maintainers
>>
> 
>
>


-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.
View/Reply Online (#18): https://lists.hpsf.io/g/modules-interest/message/18
Mute This Topic: https://lists.hpsf.io/mt/117651170/10150481
Group Owner: mod...@li...
Unsubscribe: https://lists.hpsf.io/g/modules-interest/unsub [mod...@li...]
-=-=-=-=-=-=-=-=-=-=-=-

Re: [Modules] First Technical Steering Committee (TSC) meeting: February 18 at 5:00 PM CET

[Xavier D. v. lists.hpsf.i. <xav...@li...>]

Dear All,

Reminder: our first TSC meeting will start in about 1 hour (5:00 PM CET).
See https://github.com/envmodules/modules/wiki for connection details

Regards,
Xavier

Le jeu. 5 févr. 2026 à 11:25, Xavier Delaruelle <xav...@gm...>
a écrit :

> Dear All,
>
> The first Technical Steering Committee (TSC) meeting for the Modules
> project will take place on February 18 at 5:00 PM CET.
>
> This virtual session is open to anyone interested in the project —
> contributors, users, and newcomers are all welcome!
>
> 📌 On the agenda: discussion of the roadmap for the upcoming version 5.7.
>
> 🔎 Learn more and join the conversation:
> https://github.com/envmodules/modules/wiki
>
> Regards,
> Modules maintainers
>


-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.
View/Reply Online (#17): https://lists.hpsf.io/g/modules-interest/message/17
Mute This Topic: https://lists.hpsf.io/mt/117651170/10150481
Group Owner: mod...@li...
Unsubscribe: https://lists.hpsf.io/g/modules-interest/unsub [mod...@li...]
-=-=-=-=-=-=-=-=-=-=-=-

[Modules] [lf-hpsf] 1 Month Until HPSF Conference --> 16-20 March in Chicago 🏙️

[Jennifer B. (L. F. v. lists.hpsf.i. <jbl...@li...>]

We’re only a month away from *HPSF Conference
<https://events.linuxfoundation.org/hpsf-conference/>, 16-20 March** in
Chicago. *⏳ View the schedule
<https://events.linuxfoundation.org/hpsf-conference/program/schedule/>
featuring keynotes and breakout sessions you won’t want to miss. HPSFCon
2026 will also host project meetings including:


   - AMRex
   - Apptainer
   - Chapel
   - Charliecloud
   - Environment Modules
   - Flux Framework
   - HPX
   - Kokkos
   - OpenCHAMI User Summit
   - Spack
   - Trilinos User Group (TUG)

Read our blog on Top 5 Reasons to Attend HPSFCon 2026
<https://hpsf.io/blog/2026/top-5-reasons-to-attend-hpsfcon-2026/>.


Also, we’d love your help in sharing about the event. Feel free to utilize
our Attendee Marketing Kit
<https://docs.google.com/document/d/1vib6zVUyY_u3e2UcMOrtAx_2TbMMT60afKHzAcTiLsY/edit?usp=sharing>
to help get out the word about HPSFCon with your networks.



Don’t forget to *register*
<https://events.linuxfoundation.org/hpsf-conference/register/>* by **February
27* for the Standard Rate before the price increases.



See you in Chicago!



Best,

Jennifer

-- 
Jennifer Bly
Director of Marketing Communications
*HPSF * | *The Linux Foundation*
*https://hpsf.io/ <https://hpsf.io/>*


-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.
View/Reply Online (#17): https://lists.hpsf.io/g/main/message/17
Mute This Topic: https://lists.hpsf.io/mt/117859721/10150481
Group Owner: mai...@li...
Unsubscribe: https://lists.hpsf.io/g/main/unsub [mod...@li...]
-=-=-=-=-=-=-=-=-=-=-=-

Re: [Modules] Default pagination mode

[Xavier D. v. lists.hpsf.i. <xav...@li...>]

Hi All,

New update: -p and -P abbreviated options have been implemented, as well as
the "paginate" configuration option.

This is merged in main branch.

Best regards;
Xavier

Le mar. 3 févr. 2026 à 18:53, Xavier Delaruelle via lists.hpsf.io
<xav...@li...> a écrit :

> Hi All,
>
> Small update on this pagination topic. Recent git releases provides an
> abbreviated option for --no-pager: -P.
>
> So I have just created an issue to track this feature request:
> https://github.com/envmodules/modules/issues/606
>
> Best regards,
> Xavier
>
> Le sam. 11 oct. 2025 à 17:00, Xavier Delaruelle <
> xav...@gm...> a écrit :
>
>> Hi Marc,
>>
>> Many thanks for your message.
>>
>> I have added a feature request to add a new configure option to define
>> if paginate should be done or not, without changing the pager
>> definition.
>>
>> I have also added another feature request to add an abbreviated
>> version for --paginate option. Maybe -p should be ok, as it seems to
>> be the standard abbreviation used by commands (like git).
>>
>> I do not see any command having a abbreviation for --no-pager. So I
>> would prefer not to have an abbreviation for this one.
>>
>> Best regards,
>> Xavier
>>
>> Le jeu. 9 oct. 2025 à 19:07, Paul Markfort <pau...@gm...> a écrit
>> :
>> >
>> > For now, you could define an alias of, module-no-pager which is just
>> "module --no-pager" (autocomplete would fill it out if you use
>> module-n<TAB>).
>> >
>> > If you set MODULES_PAGER="" (or use --no-pager).
>> > Then you can still pipe the output of things like "module avail"
>> through more, etc (which is what I would expect most users to do if it
>> didn't page by default - in the old days that was not easy as the output of
>> module avail went to stderr).
>> >
>> > If you currently use:  "module avail | cat"  this will also stop paging
>> (any redirection of stdout will turn off paging).
>> >
>> >
>> > I do think it could be helpful to add the ability to change the default
>> to --no-pager, then have a --pager  option to reverse it.
>> > Maybe even a --pager=   (so you can override the default without
>> setting a variable)
>> > That is just --pager  Means use default
>> > or you could use --pager=more
>> >
>> >
>> >
>> > I think pager and paginate do different things (I am not sure what,
>> probably needs more detail in the man page).
>> >
>> >         --no-pager
>> >                Do not pipe message output into a pager.
>> >
>> >         --paginate
>> >                Pipe  all  message  output  into  less  (or  if  set,
>> to  the  command  referred  in
>> >                MODULES_PAGER  variable) if error output stream is a
>> terminal. See also MODULES_PAGER
>> >                section.
>> >
>> >         --redirect
>> >                Send message output to stdout instead of stderr. Only
>> supported on sh, bash, ksh, zsh
>> >                and fish shells.
>> >                 (I think this is the default for information commands
>> like module avail).
>> >
>> >         --no-redirect
>> >                Do not send message output to stdout. Keep it on stderr.
>> >
>> >
>> >
>> >
>> >
>> >
>> >
>> >
>> >
>> >
>> >
>> > On 2025-10-09 4:36 AM, Marc Castells wrote:
>> > > Hi Xavier, All,
>> > >
>> > > Is there a way for a user to configure the default pager mode in
>> Modules? As far as I can see, the options are:
>> > >    a) Disable the pager feature altogether by clearing MODULES_PAGER
>> or assigning it the value "cat". No pagination is available.
>> > >    b) Accept the default pagination with "less -eFKRX" and make the
>> user disable it manually with "-no-pager"
>> > >
>> > > I am looking for a third option:
>> > >    c) Keep "less -eFKRX"  in MODULES_PAGER but allow the default
>> pagination to be no-pager. If the user wanted to paginate the output, they
>> would need to explicitly use "-paginate"
>> > >
>> > > This c) option to me is the most natural behaviour and allows the
>> user to configure a default pagination mode whilst having the option to
>> override it with an explicit command line switch. The current functionality
>> is not neutral as It's skewed towards displaying the output with pagination
>> by default.
>> > >
>> > > While on the subject, would it be possible to consider shortcuts for
>> the "-paginate" and "-no-pager" command line options?
>> > >
>> > > Many thanks,
>> > >
>> > > Marc.
>>
>>


-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.
View/Reply Online (#16): https://lists.hpsf.io/g/modules-interest/message/16
Mute This Topic: https://lists.hpsf.io/mt/117620044/10150481
Group Owner: mod...@li...
Unsubscribe: https://lists.hpsf.io/g/modules-interest/unsub [mod...@li...]
-=-=-=-=-=-=-=-=-=-=-=-

[Modules] First Technical Steering Committee (TSC) meeting: February 18 at 5:00 PM CET

[Xavier D. v. lists.hpsf.i. <xav...@li...>]

Dear All,

The first Technical Steering Committee (TSC) meeting for the Modules
project will take place on February 18 at 5:00 PM CET.

This virtual session is open to anyone interested in the project —
contributors, users, and newcomers are all welcome!

📌 On the agenda: discussion of the roadmap for the upcoming version 5.7.

🔎 Learn more and join the conversation:
https://github.com/envmodules/modules/wiki

Regards,
Modules maintainers


-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.
View/Reply Online (#15): https://lists.hpsf.io/g/modules-interest/message/15
Mute This Topic: https://lists.hpsf.io/mt/117651170/10150481
Group Owner: mod...@li...
Unsubscribe: https://lists.hpsf.io/g/modules-interest/unsub [mod...@li...]
-=-=-=-=-=-=-=-=-=-=-=-

[Modules] Invitation: Modules TSC

[Modules - M. <mee...@lf...>]

<html>
    <head>
        <style>
        .join-meeting-btn {
            background: #0099CC 0% 0% no-repeat padding-box;
            border: 1px solid #0099CC;
            border-radius: 30px;
            opacity: 1;
        }
        .join-meeting-text {
            text-align: left;
            font: normal normal 600 14px/19px Open Sans;
            letter-spacing: 0px;
            color: white;
            text-decoration: none;
            text-transform: capitalize;
            opacity: 1;
        }
        .tight-spacing {
            letter-spacing: 0px;
            line-height: 10px;
            padding-bottom: 3px;
        }
        </style>
    </head>
    <body>
        <div>
            
                <p>Hello modules-interest modules-interest,</p>
                <p>You have been invited to a meeting for Modules</p>
            
            <div>
                <h2>Modules TSC</h2>
                <div>
                    <p>When: Wednesday, February 18th, 17:00 Europe/Paris</p>
                </div>
                
                <div>
                    <h3>Meeting Description:</h3>
                    <p>Modules Technical Steering Committee meeting
Agenda:
* Project news
* Recent developments
* Discuss feature set for next version (5.7)
* Q&A</p>
                </div>
                
            </div>
        </div>
        <div>
            <h3>Ways to join meeting:</h3>
            <div>
                <div>
                    <div>
                        <h2>1. Join from PC, Mac, iPad, or Android</h2>
                        <button class="join-meeting-btn"><a class="join-meeting-text" style="color: white;" href="https://zoom-lfx.platform.linuxfoundation.org/meeting/99574628270?password=a922b10d-f715-4de7-bcc3-bfc6ae82435c">Join Meeting</a></button>
                        <p>If the button above does not work, paste this into your browser: <a href="https://zoom-lfx.platform.linuxfoundation.org/meeting/99574628270?password=a922b10d-f715-4de7-bcc3-bfc6ae82435c">https://zoom-lfx.platform.linuxfoundation.org/meeting/99574628270?password=a922b10d-f715-4de7-bcc3-bfc6ae82435c</a></p>
                    </div>
                </div>
                <div>
                    <h2>2. Join via audio</h2>
                    <div class="tight-spacing">
                        <p><b>One tap mobile:</b></p>
                        <p><b>US:</b> <a href="tel:+12532158782,,99574628270%23,,,,*654951%23">+12532158782,,99574628270#,,,,*654951#</a> or <a href="tel:+13462487799,,99574628270%23,,,,*654951%23">+13462487799,,99574628270#,,,,*654951#</a></p>
                    </div>
                    <div class="tight-spacing">
                        <p><b>Or dial: </b></p>
                        <p><b>US:</b> +1 253 215 8782 or +1 346 248 7799 or +1 669 900 6833 or +1 301 715 8592 or +1 312 626 6799 or +1 646 374 8656 or 877 369 0926 (Toll Free) or 855 880 1246 (Toll Free)</p>
                        <p><b>Canada:</b> +1 647 374 4685 or +1 647 558 0588 or +1 778 907 2071 or +1 204 272 7920 or +1 438 809 7799 or +1 587 328 1099 or 855 703 8985 (Toll Free)</p>
                    </div>
                    <p>Meeting ID: 99574628270</p>
                    
                    <p>Meeting Passcode: 654951</p>
                    
                    <a href='https://zoom.us/u/alwnPIaVT'>International numbers</a>
                </div>
            </div>
        </div>
    </body>
</html>

Re: [Modules] Default pagination mode

[Xavier D. v. lists.hpsf.i. <xav...@li...>]

Hi All,

Small update on this pagination topic. Recent git releases provides an
abbreviated option for --no-pager: -P.

So I have just created an issue to track this feature request:
https://github.com/envmodules/modules/issues/606

Best regards,
Xavier

Le sam. 11 oct. 2025 à 17:00, Xavier Delaruelle <xav...@gm...>
a écrit :

> Hi Marc,
>
> Many thanks for your message.
>
> I have added a feature request to add a new configure option to define
> if paginate should be done or not, without changing the pager
> definition.
>
> I have also added another feature request to add an abbreviated
> version for --paginate option. Maybe -p should be ok, as it seems to
> be the standard abbreviation used by commands (like git).
>
> I do not see any command having a abbreviation for --no-pager. So I
> would prefer not to have an abbreviation for this one.
>
> Best regards,
> Xavier
>
> Le jeu. 9 oct. 2025 à 19:07, Paul Markfort <pau...@gm...> a écrit :
> >
> > For now, you could define an alias of, module-no-pager which is just
> "module --no-pager" (autocomplete would fill it out if you use
> module-n<TAB>).
> >
> > If you set MODULES_PAGER="" (or use --no-pager).
> > Then you can still pipe the output of things like "module avail" through
> more, etc (which is what I would expect most users to do if it didn't page
> by default - in the old days that was not easy as the output of module
> avail went to stderr).
> >
> > If you currently use:  "module avail | cat"  this will also stop paging
> (any redirection of stdout will turn off paging).
> >
> >
> > I do think it could be helpful to add the ability to change the default
> to --no-pager, then have a --pager  option to reverse it.
> > Maybe even a --pager=   (so you can override the default without setting
> a variable)
> > That is just --pager  Means use default
> > or you could use --pager=more
> >
> >
> >
> > I think pager and paginate do different things (I am not sure what,
> probably needs more detail in the man page).
> >
> >         --no-pager
> >                Do not pipe message output into a pager.
> >
> >         --paginate
> >                Pipe  all  message  output  into  less  (or  if  set,
> to  the  command  referred  in
> >                MODULES_PAGER  variable) if error output stream is a
> terminal. See also MODULES_PAGER
> >                section.
> >
> >         --redirect
> >                Send message output to stdout instead of stderr. Only
> supported on sh, bash, ksh, zsh
> >                and fish shells.
> >                 (I think this is the default for information commands
> like module avail).
> >
> >         --no-redirect
> >                Do not send message output to stdout. Keep it on stderr.
> >
> >
> >
> >
> >
> >
> >
> >
> >
> >
> >
> > On 2025-10-09 4:36 AM, Marc Castells wrote:
> > > Hi Xavier, All,
> > >
> > > Is there a way for a user to configure the default pager mode in
> Modules? As far as I can see, the options are:
> > >    a) Disable the pager feature altogether by clearing MODULES_PAGER
> or assigning it the value "cat". No pagination is available.
> > >    b) Accept the default pagination with "less -eFKRX" and make the
> user disable it manually with "-no-pager"
> > >
> > > I am looking for a third option:
> > >    c) Keep "less -eFKRX"  in MODULES_PAGER but allow the default
> pagination to be no-pager. If the user wanted to paginate the output, they
> would need to explicitly use "-paginate"
> > >
> > > This c) option to me is the most natural behaviour and allows the user
> to configure a default pagination mode whilst having the option to override
> it with an explicit command line switch. The current functionality is not
> neutral as It's skewed towards displaying the output with pagination by
> default.
> > >
> > > While on the subject, would it be possible to consider shortcuts for
> the "-paginate" and "-no-pager" command line options?
> > >
> > > Many thanks,
> > >
> > > Marc.
> > >
> > >
> > > _______________________________________________
> > > Modules-interest mailing list
> > > Mod...@li...
> > > https://lists.sourceforge.net/lists/listinfo/modules-interest
> > >
> >
> > --
> > --------------------------------------------------------
> > The views and opinions expressed above are strictly
> > those of the author(s).  The content of this message has
> > not been reviewed nor approved by any entity whatsoever.
> > --------------------------------------------------------
> > Paul FM         Info: http://paulfm.com/~paulfm/
> > --------------------------------------------------------
> >
> >
> > _______________________________________________
> > Modules-interest mailing list
> > Mod...@li...
> > https://lists.sourceforge.net/lists/listinfo/modules-interest
>


-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.
View/Reply Online (#14): https://lists.hpsf.io/g/modules-interest/message/14
Mute This Topic: https://lists.hpsf.io/mt/117620044/10150481
Group Owner: mod...@li...
Unsubscribe: https://lists.hpsf.io/g/modules-interest/unsub [mod...@li...]
-=-=-=-=-=-=-=-=-=-=-=-

[Modules] [lf-hpsf] CFP: HPSF Community Summit in Germany, last miles (Deadline: February 1st)

[Julien B. <jul...@ce...>]

This is a friendly reminder to
submit your contributions to HPSF Community Summit.
The deadline is at the end of the week, Sunday February the 1st, 23:59PM.


With the schedule out for the HPSF Conference, in Chicago; it's now time to 
focus on this other event, in Germany.


https://events.academiccloud.de/e/hpsfcs26


HPSF Community Summit 2026 in Braunschweig


Following in the tradition of the European Trilinos & Kokkos User Group 
(EuroTUG) meeting series, the HPSF Community Summit offers a platform in 
Europe for users, developers and members of the High Performance Software 
Foundation (HPSF) and its projects to:
learn about recent developments,
report on their use cases and experiences,
interact with core developers,
form a network of HPSF project users and developers in Europe.


Build 🛠️ the future of high performance software with the High Performance 
Software Foundation. Join 🔗 the the leading minds that drive the advancement 
of community-driven software solutions for performance, portability, and 
productivity in HPC: engineers, researchers, developers, and users, from 
academia, industry, and government laboratories alike. The 2026 HPSF Community 
Summit in Europe covers all aspects of building software for HPC, from the 
users and contributors point of view, spanning small systems to the largest 
supercomputers, in link with AI, and accelerated computing, with a special 
focus on HPSF software. Join us to connect, learn, and lead the charge in 
transforming how we build and use software.


Learn 💡 what’s new with HPSF projects, give us feedback 📣 on your use of HPSF 
software, meet 🤝 with project communities, and tell us 💬 how to grow and 
improve them. 
🎟️ Submit an abstract 🎟️for a presentation, or a poster and present your unique 
point of view and insight to explore together how open collaboration and a 
foundation for high performance software can unlock new possibilities for high 
performance computing.


Submission guidelines


Any subject related to HPSF project is welcome, either from a user or 
contributor point of view, whether it covers technical or community and 
societal aspects.


Submission formats include:
20 minutes talks,
posters.
Depending on the number of submissions, we might offer some submitters to 
switch submission type in order to be accepted, including 7 minutes lightning 
talks.


Submissions will undergo a single-blind review process.
Talk abstracts should target around 500 words and be no more than 1500 words.
Poster submissions should describe the content of the poster in around 500 
words and no more than 1500 words; attaching a draft of the poster is strongly 
recommended.
Posters must be printed by the attendee and should fit in a standard poster 
stand intended for A0 portrait format.
Generative AI may be used to improve grammar and style, and this should 
acknowledged. It must not be used to generate the content.


Suggested topics


Suggested topic for presentations and posters include (but are not limited to)


HPSF Software from the user point of view:
Successes and challenges in adopting HPSF software in your domain.
Lessons learned setting up or scaling HPSF software for your use-case.
Your needs as a user, in engineering, research, academia, industry, or any 
other field.
Impact of a project being part of HPSF on its adoption andimage in your 
community.
Showcase of HPSF software use in unconventional fields.
HPSF Software from the contributor point of view:
Updates and technical deep dives from projects.
Factors that encouraged or discouraged you from contributing to a project vs. 
starting a new one. What makes a welcoming and inclusive contributor 
experience?
Successes and challenges, lessons learned setting up a development model for 
open-source HPC software.
HPC DevOps tools, practices, and automation strategies.
Case studies of adoption or impact of HPSF software in unusual or emerging 
domains.
HPSF Software from a community leader point of view:
Strategies for attracting, onboarding, and supporting new contributors.
Approaches to increase diversity and inclusion within technical communities.
Lessons learned from other foundations and collaborations.
Best strategies to encourage collaboration across different HPC communities 
and organizations.
Handling permissions, testing, and code contributions from external 
collaborators; policies and tools that enable safe and open collaboration.
HPSF in a larger ecosystem:
Collaborations with external contributors, project teams, or institutions.
Stories from outside communities interested in partnering with HPSF.
Projects interested in joining HPSF – motivations, benefits, and expectations.
Adaptation of HPC software and infrastructure to an AI-driven ecosystem and 
new fields such as quantum computing.
Evolving the software of today to handle the hardware of tomorrow.


--
Dr. Julien Bigot
Permanent CEA Research scientist
MdlS Science of Computing Team-leader
web: http://work.julien-bigot.fr
work address:
 Maison de la Simulation
 bat. 565, CEA Saclay
 91191 Gif-sur-Yvette CEDEX
 FRANCE
phone: +33 (0)1 69 08 01 75


--
Dr. Julien Bigot
Permanent CEA Research scientist
MdlS Science of Computing Team-leader
web: http://work.julien-bigot.fr
work address:
 Maison de la Simulation
 bat. 565, CEA Saclay
 91191 Gif-sur-Yvette CEDEX
 FRANCE
phone: +33 (0)1 69 08 01 75



-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.
View/Reply Online (#16): https://lists.hpsf.io/g/main/message/16
Mute This Topic: https://lists.hpsf.io/mt/117504370/10150481
Group Owner: mai...@li...
Unsubscribe: https://lists.hpsf.io/g/main/unsub [mod...@li...]
-=-=-=-=-=-=-=-=-=-=-=-

[Modules] [lf-hpsf] HPSFCon 2026 schedule is live 🎉

[Jennifer B. (L. F. v. lists.hpsf.i. <jbl...@li...>]

The HPSF Conference 2026
<https://events.linuxfoundation.org/hpsf-conference/>* schedule* is now
live! 🎉

HPSFCon returns to Chicago
<https://events.linuxfoundation.org/hpsf-conference/attend/venue-travel/>
from March 16-20, bringing together developers, researchers, architects,
and community leaders working at the forefront of high performance software.

A Few Session Highlights:

   -

   Building Open Source Communities through Effective Tutorials
   <https://hpsf2026.sched.com/event/2Ei7w/building-open-source-communities-through-effective-tutorials-alec-scott-lawrence-livermore-national-laboratory-greg-becker-llnl>
-
   Alec Scott & Greg Becker, Lawrence Livermore National Laboratory
   -

   Navigating the Shift to Heterogeneous Containers
   <https://hpsf2026.sched.com/event/2Ei8x/navigating-the-shift-to-heterogeneous-containers-marina-moore-edera>
   - Marina Moore, Edera
   -

   Paying Down Tech Debt: The Challenges of Maintaining Legacy Code in HPC
   <https://hpsf2026.sched.com/event/2Ei7z/paying-down-tech-debt-the-challenges-of-maintaining-legacy-code-in-hpc-jade-abraham-hpe>
   - Jade Abraham, HPE


EXPLORE THE SCHEDULE
<https://events.linuxfoundation.org/hpsf-conference/program/schedule/>

------------------------------


🎟️ Early Bird Registration Rates End in 1 Week!

Register <https://events.linuxfoundation.org/hpsf-conference/register/>
by January
30 to lock in Early Bird pricing at $299.

Are you a student? Learn more about our $99 student rate
<https://events.linuxfoundation.org/hpsf-conference/register/#registration-types>
.

Need financial assistance? Check out our scholarship and travel funding
options
<https://events.linuxfoundation.org/hpsf-conference/attend/scholarships-travel-funding/>
.

BUY YOUR PASS <https://events.linuxfoundation.org/hpsf-conference/register/>
------------------------------


📣 Become a Sponsor.

Influence the future of high performance computing by becoming a sponsor!
This event is at the forefront of innovation by advancing community-driven
solutions for performance, portability, and productivity in a rapidly
evolving HPC landscape.

As a sponsor, you'll be in the heart of the community, connecting directly
with 300+ expert attendees, researchers, engineers, and decision-makers who
are actively shaping the conversations driving the next generation of HPC
advancements.



SPONSOR THE EVENT
<https://events.linuxfoundation.org/hpsf-conference/sponsor/>

Regards,
The HPSF Team


-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.
View/Reply Online (#15): https://lists.hpsf.io/g/main/message/15
Mute This Topic: https://lists.hpsf.io/mt/117406191/10150481
Group Owner: mai...@li...
Unsubscribe: https://lists.hpsf.io/g/main/unsub [mod...@li...]
-=-=-=-=-=-=-=-=-=-=-=-

[Modules] Welcome Adrien Cotte as a new Modules maintainer

[Xavier D. v. lists.hpsf.i. <xav...@li...>]

Hello Everyone,

I am pleased to announce that *Adrien Cotte* has joined the maintainer team
of *Modules*.

Adrien works at *AS+* as an *HPC Project Manager* and has already been
involved with the project. I am excited to have him step into this role and
help move Modules forward.

Please join me in welcoming Adrien to the maintainer team, and thank you,
Adrien, for your commitment and contributions to the project.

Best regards,
Xavier


-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.
View/Reply Online (#13): https://lists.hpsf.io/g/modules-interest/message/13
Mute This Topic: https://lists.hpsf.io/mt/117396178/10150481
Group Owner: mod...@li...
Unsubscribe: https://lists.hpsf.io/g/modules-interest/unsub [mod...@li...]
-=-=-=-=-=-=-=-=-=-=-=-

[Modules] [lf-hpsf] HPSF @ SCA/HPCAsia

[Chris C. v. lists.hpsf.i. <ccl...@li...>]

Hello all,

Coming to Osaka next week for SCA/HPCAsia <https://www.sca-hpcasia2026.jp/>?
HPSF will be a Silver Sponsor of the conference; be sure to visit us at the
HPSF booth (S-19) on the Exhibition Floor. If you'd like to coordinate with
other folks in the HPSF community who will be on site, come say hello in
the #sca-hpcasia channel <https://hpsf.slack.com/archives/C0A8G26NQTV> in
the HPSF Slack workspace.

And please take note of the following HPSF projects tutorials and talks.
Hope to see many of you there!

   - Intro to Kokkos (all day Monday)
   - Managing HPC Software Complexity with Spack (all day Monday)
   - HPSF @ the Exhibitor Forum  - Wednesday 16:20 - 16:35
   - Spack Community BOF - Wednesday 17:00 - 18:00
   - HPSF BOF - Thursday 11:30 - 12:30


-- 
Chris Clark
Senior Program Manager
The Linux Foundation


-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.
View/Reply Online (#14): https://lists.hpsf.io/g/main/message/14
Mute This Topic: https://lists.hpsf.io/mt/117381294/10150481
Group Owner: mai...@li...
Unsubscribe: https://lists.hpsf.io/g/main/unsub [mod...@li...]
-=-=-=-=-=-=-=-=-=-=-=-

Re: [Modules] Reorder existing entries in path-like variables

[Paul M. v. lists.hpsf.i. <pau...@li...>]

Nice.

I think really old versions of modules (the binary versions) used to do this (by default - no setting to turn on/off).



On 2026-01-18 10:42 PM, Xavier Delaruelle via lists.hpsf.io wrote:
> Dear All,
> 
> Next feature release (v5.7) will provide a new configuration option,
> path_entry_reorder, that lets you control how existing entries are handled
> in path-like environment variables.
> 
> When enabled, prepend-path, append-path, and module use will *reorder*
> existing entries instead of leaving them untouched.
> 
> See details and example :
> https://modules.readthedocs.io/en/latest/MIGRATING.html#reorder-existing-entries-in-path-like-variables
> 
> Many thanks to Achim Gsell for contributing this new feature to the project.
> 
> Best regards,
> Xavier
> 
> 
> 
> 
> 
> 

-- 
--------------------------------------------------------
The views and opinions expressed above are strictly
those of the author(s).  The content of this message has
not been reviewed nor approved by any entity whatsoever.
--------------------------------------------------------
Paul FM         Info: http://paulfm.com/~paulfm/
--------------------------------------------------------


-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.
View/Reply Online (#12): https://lists.hpsf.io/g/modules-interest/message/12
Mute This Topic: https://lists.hpsf.io/mt/117341080/10150481
Group Owner: mod...@li...
Unsubscribe: https://lists.hpsf.io/g/modules-interest/unsub [mod...@li...]
-=-=-=-=-=-=-=-=-=-=-=-

[Modules] Reorder existing entries in path-like variables

[Xavier D. v. lists.hpsf.i. <xav...@li...>]

Dear All,

Next feature release (v5.7) will provide a new configuration option,
path_entry_reorder, that lets you control how existing entries are handled
in path-like environment variables.

When enabled, prepend-path, append-path, and module use will *reorder*
existing entries instead of leaving them untouched.

See details and example :
https://modules.readthedocs.io/en/latest/MIGRATING.html#reorder-existing-entries-in-path-like-variables

Many thanks to Achim Gsell for contributing this new feature to the project.

Best regards,
Xavier


-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.
View/Reply Online (#11): https://lists.hpsf.io/g/modules-interest/message/11
Mute This Topic: https://lists.hpsf.io/mt/117341080/10150481
Group Owner: mod...@li...
Unsubscribe: https://lists.hpsf.io/g/modules-interest/unsub [mod...@li...]
-=-=-=-=-=-=-=-=-=-=-=-

Re: [Modules] HPSF Conference 2026

[Xavier D. v. lists.hpsf.i. <xav...@li...>]

Hi All,

HPSFCon 2026 CfP is closing this sunday. Don't forget to submit your talk
proposals 😀

Cheers,
Xavier

Le mar. 16 déc. 2025 à 20:56, Xavier Delaruelle <xav...@gm...>
a écrit :

> Hi All,
>
> HPSFCon 2026 Call for Proposals are open and this new edition will feature
> an Environment Modules Project Meeting track.
>
> We’re looking for talks, discussions, lightning sessions, and posters that
> push high performance computing and open source software forward.
>
> For the Environment Modules track, we’re inviting submissions that
> highlight:
>
> * Novel “module” developments
> * New or creative use cases
> * Interesting site-specific customizations
> * Lessons learned, best practices, or anything else of broad interest to
> the Environment Modules community
>
> This track is not limited to the Modules project — talks about Lmod or any
> other module implementation are very welcome. If your work touches
> environment modules and you think others can learn from it, we’d love to
> hear from you!
>
> 📅 Submit by January 11, 2026
> 👉 Submit your proposal here:
> https://events.linuxfoundation.org/hpsf-conference/program/cfp/
> 💭 Learn more about what we are looking for:
> https://hpsf.io/blog/2025/hpsfcon-2026-call-for-proposals-now-open/
>
> Sponsorships are also live + early bird registration is open through
> January 30.
>
> See you in Chicago this March! 🍀
>


-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.
View/Reply Online (#10): https://lists.hpsf.io/g/modules-interest/message/10
Mute This Topic: https://lists.hpsf.io/mt/116815556/10150481
Group Owner: mod...@li...
Unsubscribe: https://lists.hpsf.io/g/modules-interest/unsub [mod...@li...]
-=-=-=-=-=-=-=-=-=-=-=-

Re: [Modules] tcl module load slowness debug

[Xavier D. v. lists.hpsf.i. <xav...@li...>]

The minimum required version of tclsh is described in the installation
notes (section "Requirements"):
  https://modules.readthedocs.io/en/latest/INSTALL.html#requirements

Tcl >= 8.5 is required on Modules v5. It is currently not planned to change
this requirement until Modules v6.

Best regards,
Xavier

Le mar. 6 janv. 2026 à 20:10, Paul Markfort via lists.hpsf.io <paulfm.mn=
gma...@li...> a écrit :

> That is a ways out (July, August, or September).  But, it is good to have
> an idea.
>
> Thanks.
>
> It might be helpful to include the minimum required version of tclsh in
> the release notes:
>         https://modules.readthedocs.io/en/latest/NEWS.html
> Since that is the binary it depends on.
>
>
> I just went live with 5.6.1 (installed it a month ago, I like to test for
> a week or so before making it the default version for everyone).
> Note: We have our modules (including the commands and init scripts)
> installed on an NFS V4 share for all our various Desktop Linux machines
> (tclsh is local to the machines).
>
>
> On 2026-01-06 12:43 PM, Xavier Delaruelle via lists.hpsf.io wrote:
> > 5.7.0 is not precisely planned yet, but it should be available I think
> > during Q3 2026.
> >
> > Regards,
> > Xavier
> >
> > Le mar. 6 janv. 2026 à 09:51, Singh, Puneet-2 (Ext) <
> > pun...@no...> a écrit :
> >
> >> Thank you for the update @Xavier Delaruelle <
> xav...@gm...>
> >> and happy new year.
> >> Could you please let us know the tentative release date for module
> version
> >> 5.7.0?
> >>
> >> Regards,
> >> Puneet
> >>
> >>
> >>
> >> *From:* Xavier Delaruelle <xav...@gm...>
> >> *Sent:* 06 January 2026 12:54
> >> *To:* mod...@li...
> >> *Subject:* Re: [Modules] tcl module load slowness debug
> >>
> >>
> >>
> >> Dear All, I wish you a very happy new year 🎉 To kick off the year, I am
> >> pleased to share that recent work on Modules has significantly improved
> >> performance when evaluating large numbers of modulefiles. You can find
> all
> >> the details at the following
> >>
> >> ZjQcmQRYFpfptBannerStart
> >>
> >> *This Message Is From an Untrusted Sender *
> >>
> >> You have not previously corresponded with this sender.
> >>
> >> ZjQcmQRYFpfptBannerEnd
> >>
> >> Dear All,
> >>
> >> I wish you a very happy new year 🎉
> >>
> >> To kick off the year, I am pleased to share that recent work on Modules
> >> has significantly improved performance when evaluating large numbers of
> >> modulefiles.
> >>
> >> You can find all the details at the following link:
> >>
> https://modules.readthedocs.io/en/latest/MIGRATING.html#improving-evaluation-performances
> >>
> >> This work will be part of next version (5.7.0) and is already available
> in
> >> the "main" git branch.
> >>
> >> We encourage you to review this update, especially if you’ve experienced
> >> slower module evaluations in the past or are planning upgrades.
> >>
> >> Best regards,
> >> Xavier
> >>
> >>
> >>
> >> Le lun. 8 déc. 2025 à 06:17, Xavier Delaruelle via lists.hpsf.io
> >> <
> https://urldefense.com/v3/__http:/lists.hpsf.io__;!!N3hqHg43uw!sZc6bjVVBmiDCOwdjqRoLts70ywqqLzfu67vV23yN0HIphd9qYcsk59x08nD0qVY4tgj6Vc4HaSb3arUEqpwEjme9keIEltNiFewbA$
> >
> >> <xav...@li...> a écrit :
> >>
> >> I have made some tests with a module tree generated by EasyBuild.
> Loading
> >> `R-bundle-Bioconductor/3.19-foss-2023b-R-4.4.1` module, that requires
> 136
> >> additional modules, takes 1000ms on my local laptop (modulefiles stored
> on
> >> SSD).
> >>
> >>
> >>
> >> I have added this test setup into the "mb" command (Modules make bench
> >> utility) to track the performance of such module load over time.
> >>
> >>
> >>
> >> I have created an issue to work on this topic and improve the
> performance.
> >> Will try to get a 2x but I foresee that it will not be easy.
> >>
> >> https://github.com/envmodules/modules/issues/603
> >> <
> https://urldefense.com/v3/__https:/github.com/envmodules/modules/issues/603__;!!N3hqHg43uw!sZc6bjVVBmiDCOwdjqRoLts70ywqqLzfu67vV23yN0HIphd9qYcsk59x08nD0qVY4tgj6Vc4HaSb3arUEqpwEjme9keIElta6fShaQ$
> >
> >>
> >>
> >>
> >> When playing with my test setup, I have found that if the Tcl
> modulefiles
> >> are not generated for "EnvironmentModules" module tool, or generated by
> a
> >> version of EasyBuild prior 5.0, it takes 3000ms to load
> >> `R-bundle-Bioconductor` and its dependencies. This is due to the
> extensive
> >> use of "is-loaded" tests in these modulefiles. So if your modulefiles
> >> contains an "is-loaded" test prior each "module load" command, I
> strongly
> >> suggest to regenerate all these modulefiles to benefit from an
> immediate 3x
> >> improvement.
> >>
> >>
> >>
> >> Cheers,
> >>
> >> Xavier
> >>
> >>
> >>
> >>
> >
> >
> >
> >
> >
> >
>
> --
> --------------------------------------------------------
> The views and opinions expressed above are strictly
> those of the author(s).  The content of this message has
> not been reviewed nor approved by any entity whatsoever.
> --------------------------------------------------------
> Paul FM         Info: http://paulfm.com/~paulfm/
> --------------------------------------------------------
>
>
> 
>
>
>


-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.
View/Reply Online (#9): https://lists.hpsf.io/g/modules-interest/message/9
Mute This Topic: https://lists.hpsf.io/mt/116608462/10150481
Group Owner: mod...@li...
Unsubscribe: https://lists.hpsf.io/g/modules-interest/unsub [mod...@li...]
-=-=-=-=-=-=-=-=-=-=-=-

Re: [Modules] tcl module load slowness debug

[Paul M. v. lists.hpsf.i. <pau...@li...>]

That is a ways out (July, August, or September).  But, it is good to have an idea.

Thanks.

It might be helpful to include the minimum required version of tclsh in the release notes:
	https://modules.readthedocs.io/en/latest/NEWS.html
Since that is the binary it depends on.


I just went live with 5.6.1 (installed it a month ago, I like to test for a week or so before making it the default version for everyone).
Note: We have our modules (including the commands and init scripts) installed on an NFS V4 share for all our various Desktop Linux machines (tclsh is local to the machines).


On 2026-01-06 12:43 PM, Xavier Delaruelle via lists.hpsf.io wrote:
> 5.7.0 is not precisely planned yet, but it should be available I think
> during Q3 2026.
> 
> Regards,
> Xavier
> 
> Le mar. 6 janv. 2026 à 09:51, Singh, Puneet-2 (Ext) <
> pun...@no...> a écrit :
> 
>> Thank you for the update @Xavier Delaruelle <xav...@gm...>
>> and happy new year.
>> Could you please let us know the tentative release date for module version
>> 5.7.0?
>>
>> Regards,
>> Puneet
>>
>>
>>
>> *From:* Xavier Delaruelle <xav...@gm...>
>> *Sent:* 06 January 2026 12:54
>> *To:* mod...@li...
>> *Subject:* Re: [Modules] tcl module load slowness debug
>>
>>
>>
>> Dear All, I wish you a very happy new year 🎉 To kick off the year, I am
>> pleased to share that recent work on Modules has significantly improved
>> performance when evaluating large numbers of modulefiles. You can find all
>> the details at the following
>>
>> ZjQcmQRYFpfptBannerStart
>>
>> *This Message Is From an Untrusted Sender *
>>
>> You have not previously corresponded with this sender.
>>
>> ZjQcmQRYFpfptBannerEnd
>>
>> Dear All,
>>
>> I wish you a very happy new year 🎉
>>
>> To kick off the year, I am pleased to share that recent work on Modules
>> has significantly improved performance when evaluating large numbers of
>> modulefiles.
>>
>> You can find all the details at the following link:
>> https://modules.readthedocs.io/en/latest/MIGRATING.html#improving-evaluation-performances
>>
>> This work will be part of next version (5.7.0) and is already available in
>> the "main" git branch.
>>
>> We encourage you to review this update, especially if you’ve experienced
>> slower module evaluations in the past or are planning upgrades.
>>
>> Best regards,
>> Xavier
>>
>>
>>
>> Le lun. 8 déc. 2025 à 06:17, Xavier Delaruelle via lists.hpsf.io
>> <https://urldefense.com/v3/__http:/lists.hpsf.io__;!!N3hqHg43uw!sZc6bjVVBmiDCOwdjqRoLts70ywqqLzfu67vV23yN0HIphd9qYcsk59x08nD0qVY4tgj6Vc4HaSb3arUEqpwEjme9keIEltNiFewbA$>
>> <xav...@li...> a écrit :
>>
>> I have made some tests with a module tree generated by EasyBuild. Loading
>> `R-bundle-Bioconductor/3.19-foss-2023b-R-4.4.1` module, that requires 136
>> additional modules, takes 1000ms on my local laptop (modulefiles stored on
>> SSD).
>>
>>
>>
>> I have added this test setup into the "mb" command (Modules make bench
>> utility) to track the performance of such module load over time.
>>
>>
>>
>> I have created an issue to work on this topic and improve the performance.
>> Will try to get a 2x but I foresee that it will not be easy.
>>
>> https://github.com/envmodules/modules/issues/603
>> <https://urldefense.com/v3/__https:/github.com/envmodules/modules/issues/603__;!!N3hqHg43uw!sZc6bjVVBmiDCOwdjqRoLts70ywqqLzfu67vV23yN0HIphd9qYcsk59x08nD0qVY4tgj6Vc4HaSb3arUEqpwEjme9keIElta6fShaQ$>
>>
>>
>>
>> When playing with my test setup, I have found that if the Tcl modulefiles
>> are not generated for "EnvironmentModules" module tool, or generated by a
>> version of EasyBuild prior 5.0, it takes 3000ms to load
>> `R-bundle-Bioconductor` and its dependencies. This is due to the extensive
>> use of "is-loaded" tests in these modulefiles. So if your modulefiles
>> contains an "is-loaded" test prior each "module load" command, I strongly
>> suggest to regenerate all these modulefiles to benefit from an immediate 3x
>> improvement.
>>
>>
>>
>> Cheers,
>>
>> Xavier
>>
>>
>>
>>
> 
> 
> 
> 
> 
> 

-- 
--------------------------------------------------------
The views and opinions expressed above are strictly
those of the author(s).  The content of this message has
not been reviewed nor approved by any entity whatsoever.
--------------------------------------------------------
Paul FM         Info: http://paulfm.com/~paulfm/
--------------------------------------------------------


-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.
View/Reply Online (#8): https://lists.hpsf.io/g/modules-interest/message/8
Mute This Topic: https://lists.hpsf.io/mt/116608462/10150481
Group Owner: mod...@li...
Unsubscribe: https://lists.hpsf.io/g/modules-interest/unsub [mod...@li...]
-=-=-=-=-=-=-=-=-=-=-=-

Re: [Modules] tcl module load slowness debug

[Xavier D. v. lists.hpsf.i. <xav...@li...>]

5.7.0 is not precisely planned yet, but it should be available I think
during Q3 2026.

Regards,
Xavier

Le mar. 6 janv. 2026 à 09:51, Singh, Puneet-2 (Ext) <
pun...@no...> a écrit :

> Thank you for the update @Xavier Delaruelle <xav...@gm...>
> and happy new year.
> Could you please let us know the tentative release date for module version
> 5.7.0?
>
> Regards,
> Puneet
>
>
>
> *From:* Xavier Delaruelle <xav...@gm...>
> *Sent:* 06 January 2026 12:54
> *To:* mod...@li...
> *Subject:* Re: [Modules] tcl module load slowness debug
>
>
>
> Dear All, I wish you a very happy new year 🎉 To kick off the year, I am
> pleased to share that recent work on Modules has significantly improved
> performance when evaluating large numbers of modulefiles. You can find all
> the details at the following
>
> ZjQcmQRYFpfptBannerStart
>
> *This Message Is From an Untrusted Sender *
>
> You have not previously corresponded with this sender.
>
> ZjQcmQRYFpfptBannerEnd
>
> Dear All,
>
> I wish you a very happy new year 🎉
>
> To kick off the year, I am pleased to share that recent work on Modules
> has significantly improved performance when evaluating large numbers of
> modulefiles.
>
> You can find all the details at the following link:
> https://modules.readthedocs.io/en/latest/MIGRATING.html#improving-evaluation-performances
>
> This work will be part of next version (5.7.0) and is already available in
> the "main" git branch.
>
> We encourage you to review this update, especially if you’ve experienced
> slower module evaluations in the past or are planning upgrades.
>
> Best regards,
> Xavier
>
>
>
> Le lun. 8 déc. 2025 à 06:17, Xavier Delaruelle via lists.hpsf.io
> <https://urldefense.com/v3/__http:/lists.hpsf.io__;!!N3hqHg43uw!sZc6bjVVBmiDCOwdjqRoLts70ywqqLzfu67vV23yN0HIphd9qYcsk59x08nD0qVY4tgj6Vc4HaSb3arUEqpwEjme9keIEltNiFewbA$>
> <xav...@li...> a écrit :
>
> I have made some tests with a module tree generated by EasyBuild. Loading
> `R-bundle-Bioconductor/3.19-foss-2023b-R-4.4.1` module, that requires 136
> additional modules, takes 1000ms on my local laptop (modulefiles stored on
> SSD).
>
>
>
> I have added this test setup into the "mb" command (Modules make bench
> utility) to track the performance of such module load over time.
>
>
>
> I have created an issue to work on this topic and improve the performance.
> Will try to get a 2x but I foresee that it will not be easy.
>
> https://github.com/envmodules/modules/issues/603
> <https://urldefense.com/v3/__https:/github.com/envmodules/modules/issues/603__;!!N3hqHg43uw!sZc6bjVVBmiDCOwdjqRoLts70ywqqLzfu67vV23yN0HIphd9qYcsk59x08nD0qVY4tgj6Vc4HaSb3arUEqpwEjme9keIElta6fShaQ$>
>
>
>
> When playing with my test setup, I have found that if the Tcl modulefiles
> are not generated for "EnvironmentModules" module tool, or generated by a
> version of EasyBuild prior 5.0, it takes 3000ms to load
> `R-bundle-Bioconductor` and its dependencies. This is due to the extensive
> use of "is-loaded" tests in these modulefiles. So if your modulefiles
> contains an "is-loaded" test prior each "module load" command, I strongly
> suggest to regenerate all these modulefiles to benefit from an immediate 3x
> improvement.
>
>
>
> Cheers,
>
> Xavier
>
> 
>
>


-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.
View/Reply Online (#7): https://lists.hpsf.io/g/modules-interest/message/7
Mute This Topic: https://lists.hpsf.io/mt/116608462/10150481
Group Owner: mod...@li...
Unsubscribe: https://lists.hpsf.io/g/modules-interest/unsub [mod...@li...]
-=-=-=-=-=-=-=-=-=-=-=-

[Modules] [lf-hpsf] Submit Your HPSFCon 2026 Proposal by Jan. 11

[Jennifer B. (L. F. v. lists.hpsf.i. <jbl...@li...>]

Hi everyone,

This is your reminder to submit to speak at HPSF Conference 2026
<https://events.linuxfoundation.org/hpsf-conference/program/cfp/>. The CFP
closes on January 11 at 11:59 PM CST.

We’re accepting 15 & 25 minute talks, panels, and poster submissions across
these focus areas:

   - Building and Sustaining Community
   - User and Contributor Experiences
   - Project and Ecosystem Growth
   - Productivity, Performance, and the HPC Pipeline
   - Trust, Security, and Open Collaboration
   - Emerging Technologies and the Future of HPC

Details at: https://events.linuxfoundation.org/hpsf-conference/program/cfp/

Best,
Jennifer

-- 
Jennifer Bly
Director of Marketing Communications
*HPSF * | *The Linux Foundation*
*https://hpsf.io/ <https://hpsf.io/>*


-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.
View/Reply Online (#13): https://lists.hpsf.io/g/main/message/13
Mute This Topic: https://lists.hpsf.io/mt/117105968/10150481
Group Owner: mai...@li...
Unsubscribe: https://lists.hpsf.io/g/main/unsub [mod...@li...]
-=-=-=-=-=-=-=-=-=-=-=-

Re: [Modules] tcl module load slowness debug

[Singh, Puneet-2 (E. v. lists.hpsf.i. <pun...@li...>]

Thank you for the update @Xavier Delaruelle<mailto:xav...@gm...> and happy new year.
Could you please let us know the tentative release date for module version 5.7.0?

Regards,
Puneet

From: Xavier Delaruelle <xav...@gm...>
Sent: 06 January 2026 12:54
To: mod...@li...
Subject: Re: [Modules] tcl module load slowness debug

Dear All, I wish you a very happy new year 🎉 To kick off the year, I am pleased to share that recent work on Modules has significantly improved performance when evaluating large numbers of modulefiles. You can find all the details at the following
ZjQcmQRYFpfptBannerStart
This Message Is From an Untrusted Sender
You have not previously corresponded with this sender.
ZjQcmQRYFpfptBannerEnd

Dear All,

I wish you a very happy new year 🎉

To kick off the year, I am pleased to share that recent work on Modules has significantly improved performance when evaluating large numbers of modulefiles.

You can find all the details at the following link: https://modules.readthedocs.io/en/latest/MIGRATING.html#improving-evaluation-performances

This work will be part of next version (5.7.0) and is already available in the "main" git branch.

We encourage you to review this update, especially if you’ve experienced slower module evaluations in the past or are planning upgrades.

Best regards,
Xavier

Le lun. 8 déc. 2025 à 06:17, Xavier Delaruelle via lists.hpsf.io<https://urldefense.com/v3/__http:/lists.hpsf.io__;!!N3hqHg43uw!sZc6bjVVBmiDCOwdjqRoLts70ywqqLzfu67vV23yN0HIphd9qYcsk59x08nD0qVY4tgj6Vc4HaSb3arUEqpwEjme9keIEltNiFewbA$> <xav...@li...<mailto:gma...@li...>> a écrit :
I have made some tests with a module tree generated by EasyBuild. Loading `R-bundle-Bioconductor/3.19-foss-2023b-R-4.4.1` module, that requires 136 additional modules, takes 1000ms on my local laptop (modulefiles stored on SSD).

I have added this test setup into the "mb" command (Modules make bench utility) to track the performance of such module load over time.

I have created an issue to work on this topic and improve the performance. Will try to get a 2x but I foresee that it will not be easy.
https://github.com/envmodules/modules/issues/603<https://urldefense.com/v3/__https:/github.com/envmodules/modules/issues/603__;!!N3hqHg43uw!sZc6bjVVBmiDCOwdjqRoLts70ywqqLzfu67vV23yN0HIphd9qYcsk59x08nD0qVY4tgj6Vc4HaSb3arUEqpwEjme9keIElta6fShaQ$>

When playing with my test setup, I have found that if the Tcl modulefiles are not generated for "EnvironmentModules" module tool, or generated by a version of EasyBuild prior 5.0, it takes 3000ms to load `R-bundle-Bioconductor` and its dependencies. This is due to the extensive use of "is-loaded" tests in these modulefiles. So if your modulefiles contains an "is-loaded" test prior each "module load" command, I strongly suggest to regenerate all these modulefiles to benefit from an immediate 3x improvement.

Cheers,
Xavier



-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.
View/Reply Online (#6): https://lists.hpsf.io/g/modules-interest/message/6
Mute This Topic: https://lists.hpsf.io/mt/116608462/10150481
Group Owner: mod...@li...
Unsubscribe: https://lists.hpsf.io/g/modules-interest/unsub [mod...@li...]
-=-=-=-=-=-=-=-=-=-=-=-