Re: [OMPI users] OpenMPI 2.1.1 bug on Ubuntu 18.04.2 LTS

2019-08-02 Thread Jeff Squyres (jsquyres) via users
Ouch.  :-(

I just followed up on that original bug.  Thanks for the heads up.


> On Aug 2, 2019, at 4:15 PM, Zhang, Junchao  wrote:
> 
> Looks the bug was reported to Ubuntu almost two years ago but still 
> Unassigned, https://bugs.launchpad.net/ubuntu/+source/openmpi/+bug/1731938
> I don't know how many hours of people's time were wasted on re-discovering 
> this issue.
> --Junchao Zhang
> 
> 
> On Fri, Aug 2, 2019 at 2:54 PM Jeff Squyres (jsquyres) via users 
>  wrote:
> Ah, got it.
> 
> Yes, if I compile with --enable-heterogeneous, the I can replicate the 
> problem.
> 
> Looks like we opted not to fix this problem in the 2.1.x series.  I'll update 
> the downstream Ubuntu bug to tell them to disable heterogeneous.
> 
> 
> > On Aug 1, 2019, at 7:40 PM, Gilles Gouaillardet via users 
> >  wrote:
> > 
> > Juanchao,
> > 
> > 
> > Is the issue related to https://github.com/open-mpi/ompi/pull/4501 ?
> > 
> > 
> > Jeff,
> > 
> > 
> > you might have to configure with --enable-heterogeneous to evidence the 
> > issue
> > 
> > 
> > 
> > Cheers,
> > 
> > 
> > Gilles
> > 
> > On 8/2/2019 4:06 AM, Jeff Squyres (jsquyres) via users wrote:
> >> I am able to replicate the issue on a stock Ubuntu 18.04 install with 
> >> their Open MPI package.
> >> 
> >> But if I compile my own Open MPI 2.1.1, it works fine.
> >> Also, if I compile my own Open MPI 2.1.6, it works fine.
> >> 
> >> I filed a bug at Ubuntu about this:
> >> 
> >> https://bugs.launchpad.net/ubuntu/+source/xubuntu-meta/+bug/1838684
> >> 
> >> 
> >> 
> >>> On Aug 1, 2019, at 2:33 PM, Zhang, Junchao  wrote:
> >>> 
> >>> $ aptitude versions libopenmpi-dev
> >>> Package libopenmpi-dev:
> >>> i   2.1.1-8  bionic  500
> >>> Package libopenmpi-dev:i386:
> >>> p   2.1.1-8 bionic  500
> >>> 
> >>> $ sudo apt-get install libopenmpi-dev=2.1.6
> >>> Reading package lists... Done
> >>> Building dependency tree
> >>> Reading state information... Done
> >>> E: Version '2.1.6' for 'libopenmpi-dev' was not found
> >>> 
> >>> --Junchao Zhang
> >>> 
> >>> 
> >>> On Thu, Aug 1, 2019 at 1:15 PM Jeff Squyres (jsquyres) 
> >>>  wrote:
> >>> Does the bug exist in Open MPI v2.1.6?
> >>> 
>  On Jul 31, 2019, at 2:19 PM, Zhang, Junchao via users 
>   wrote:
>  
>  Hello,
>    I met a bug with OpenMPI 2.1.1 distributed in the latest Ubuntu 
>  18.04.2 LTS. It happens with self to self send/recv using MPI_ANY_SOURCE 
>  for message matching.  See the attached test code.  You can reproduce it 
>  even with one process.
>    It is a severe bug. Since this Ubuntu is widely used and has long term 
>  support, could it be somehow fixed?
>    Thanks a lot.
>  
>  --Junchao Zhang
>  ___
>  users mailing list
>  users@lists.open-mpi.org
>  https://lists.open-mpi.org/mailman/listinfo/users
> >>> 
> >>> -- 
> >>> Jeff Squyres
> >>> jsquy...@cisco.com
> >>> 
> >> 
> > ___
> > users mailing list
> > users@lists.open-mpi.org
> > https://lists.open-mpi.org/mailman/listinfo/users
> 
> 
> -- 
> Jeff Squyres
> jsquy...@cisco.com
> 
> ___
> users mailing list
> users@lists.open-mpi.org
> https://lists.open-mpi.org/mailman/listinfo/users


-- 
Jeff Squyres
jsquy...@cisco.com

___
users mailing list
users@lists.open-mpi.org
https://lists.open-mpi.org/mailman/listinfo/users


Re: [OMPI users] OpenMPI 2.1.1 bug on Ubuntu 18.04.2 LTS

2019-08-02 Thread Zhang, Junchao via users
Looks the bug was reported to Ubuntu almost two years ago but still Unassigned, 
https://bugs.launchpad.net/ubuntu/+source/openmpi/+bug/1731938
I don't know how many hours of people's time were wasted on re-discovering this 
issue.
--Junchao Zhang


On Fri, Aug 2, 2019 at 2:54 PM Jeff Squyres (jsquyres) via users 
mailto:users@lists.open-mpi.org>> wrote:
Ah, got it.

Yes, if I compile with --enable-heterogeneous, the I can replicate the problem.

Looks like we opted not to fix this problem in the 2.1.x series.  I'll update 
the downstream Ubuntu bug to tell them to disable heterogeneous.


> On Aug 1, 2019, at 7:40 PM, Gilles Gouaillardet via users 
> mailto:users@lists.open-mpi.org>> wrote:
>
> Juanchao,
>
>
> Is the issue related to https://github.com/open-mpi/ompi/pull/4501 ?
>
>
> Jeff,
>
>
> you might have to configure with --enable-heterogeneous to evidence the issue
>
>
>
> Cheers,
>
>
> Gilles
>
> On 8/2/2019 4:06 AM, Jeff Squyres (jsquyres) via users wrote:
>> I am able to replicate the issue on a stock Ubuntu 18.04 install with their 
>> Open MPI package.
>>
>> But if I compile my own Open MPI 2.1.1, it works fine.
>> Also, if I compile my own Open MPI 2.1.6, it works fine.
>>
>> I filed a bug at Ubuntu about this:
>>
>> https://bugs.launchpad.net/ubuntu/+source/xubuntu-meta/+bug/1838684
>>
>>
>>
>>> On Aug 1, 2019, at 2:33 PM, Zhang, Junchao 
>>> mailto:jczh...@mcs.anl.gov>> wrote:
>>>
>>> $ aptitude versions libopenmpi-dev
>>> Package libopenmpi-dev:
>>> i   2.1.1-8  bionic  500
>>> Package libopenmpi-dev:i386:
>>> p   2.1.1-8 bionic  500
>>>
>>> $ sudo apt-get install libopenmpi-dev=2.1.6
>>> Reading package lists... Done
>>> Building dependency tree
>>> Reading state information... Done
>>> E: Version '2.1.6' for 'libopenmpi-dev' was not found
>>>
>>> --Junchao Zhang
>>>
>>>
>>> On Thu, Aug 1, 2019 at 1:15 PM Jeff Squyres (jsquyres) 
>>> mailto:jsquy...@cisco.com>> wrote:
>>> Does the bug exist in Open MPI v2.1.6?
>>>
 On Jul 31, 2019, at 2:19 PM, Zhang, Junchao via users 
 mailto:users@lists.open-mpi.org>> wrote:

 Hello,
   I met a bug with OpenMPI 2.1.1 distributed in the latest Ubuntu 18.04.2 
 LTS. It happens with self to self send/recv using MPI_ANY_SOURCE for 
 message matching.  See the attached test code.  You can reproduce it even 
 with one process.
   It is a severe bug. Since this Ubuntu is widely used and has long term 
 support, could it be somehow fixed?
   Thanks a lot.

 --Junchao Zhang
 ___
 users mailing list
 users@lists.open-mpi.org
 https://lists.open-mpi.org/mailman/listinfo/users
>>>
>>> --
>>> Jeff Squyres
>>> jsquy...@cisco.com
>>>
>>
> ___
> users mailing list
> users@lists.open-mpi.org
> https://lists.open-mpi.org/mailman/listinfo/users


--
Jeff Squyres
jsquy...@cisco.com

___
users mailing list
users@lists.open-mpi.org
https://lists.open-mpi.org/mailman/listinfo/users
___
users mailing list
users@lists.open-mpi.org
https://lists.open-mpi.org/mailman/listinfo/users

Re: [OMPI users] OpenMPI 2.1.1 bug on Ubuntu 18.04.2 LTS

2019-08-02 Thread Zhang, Junchao via users
I run ompi_info on Ubuntu and it has "Heterogeneous support: yes"
Thanks for the update.

--Junchao Zhang


On Thu, Aug 1, 2019 at 6:49 PM Gilles Gouaillardet via users 
mailto:users@lists.open-mpi.org>> wrote:
Juanchao,


Is the issue related to https://github.com/open-mpi/ompi/pull/4501 ?


Jeff,


you might have to configure with --enable-heterogeneous to evidence the
issue



Cheers,


Gilles

On 8/2/2019 4:06 AM, Jeff Squyres (jsquyres) via users wrote:
> I am able to replicate the issue on a stock Ubuntu 18.04 install with their 
> Open MPI package.
>
> But if I compile my own Open MPI 2.1.1, it works fine.
> Also, if I compile my own Open MPI 2.1.6, it works fine.
>
> I filed a bug at Ubuntu about this:
>
>  https://bugs.launchpad.net/ubuntu/+source/xubuntu-meta/+bug/1838684
>
>
>
>> On Aug 1, 2019, at 2:33 PM, Zhang, Junchao 
>> mailto:jczh...@mcs.anl.gov>> wrote:
>>
>> $ aptitude versions libopenmpi-dev
>> Package libopenmpi-dev:
>> i   2.1.1-8  bionic  500
>> Package libopenmpi-dev:i386:
>> p   2.1.1-8 bionic  500
>>
>> $ sudo apt-get install libopenmpi-dev=2.1.6
>> Reading package lists... Done
>> Building dependency tree
>> Reading state information... Done
>> E: Version '2.1.6' for 'libopenmpi-dev' was not found
>>
>> --Junchao Zhang
>>
>>
>> On Thu, Aug 1, 2019 at 1:15 PM Jeff Squyres (jsquyres) 
>> mailto:jsquy...@cisco.com>> wrote:
>> Does the bug exist in Open MPI v2.1.6?
>>
>>> On Jul 31, 2019, at 2:19 PM, Zhang, Junchao via users 
>>> mailto:users@lists.open-mpi.org>> wrote:
>>>
>>> Hello,
>>>I met a bug with OpenMPI 2.1.1 distributed in the latest Ubuntu 18.04.2 
>>> LTS. It happens with self to self send/recv using MPI_ANY_SOURCE for 
>>> message matching.  See the attached test code.  You can reproduce it even 
>>> with one process.
>>>It is a severe bug. Since this Ubuntu is widely used and has long term 
>>> support, could it be somehow fixed?
>>>Thanks a lot.
>>>
>>> --Junchao Zhang
>>> ___
>>> users mailing list
>>> users@lists.open-mpi.org
>>> https://lists.open-mpi.org/mailman/listinfo/users
>>
>> --
>> Jeff Squyres
>> jsquy...@cisco.com
>>
>
___
users mailing list
users@lists.open-mpi.org
https://lists.open-mpi.org/mailman/listinfo/users
___
users mailing list
users@lists.open-mpi.org
https://lists.open-mpi.org/mailman/listinfo/users

Re: [OMPI users] OpenMPI 2.1.1 bug on Ubuntu 18.04.2 LTS

2019-08-02 Thread Jeff Squyres (jsquyres) via users
Ah, got it.

Yes, if I compile with --enable-heterogeneous, the I can replicate the problem.

Looks like we opted not to fix this problem in the 2.1.x series.  I'll update 
the downstream Ubuntu bug to tell them to disable heterogeneous.


> On Aug 1, 2019, at 7:40 PM, Gilles Gouaillardet via users 
>  wrote:
> 
> Juanchao,
> 
> 
> Is the issue related to https://github.com/open-mpi/ompi/pull/4501 ?
> 
> 
> Jeff,
> 
> 
> you might have to configure with --enable-heterogeneous to evidence the issue
> 
> 
> 
> Cheers,
> 
> 
> Gilles
> 
> On 8/2/2019 4:06 AM, Jeff Squyres (jsquyres) via users wrote:
>> I am able to replicate the issue on a stock Ubuntu 18.04 install with their 
>> Open MPI package.
>> 
>> But if I compile my own Open MPI 2.1.1, it works fine.
>> Also, if I compile my own Open MPI 2.1.6, it works fine.
>> 
>> I filed a bug at Ubuntu about this:
>> 
>> https://bugs.launchpad.net/ubuntu/+source/xubuntu-meta/+bug/1838684
>> 
>> 
>> 
>>> On Aug 1, 2019, at 2:33 PM, Zhang, Junchao  wrote:
>>> 
>>> $ aptitude versions libopenmpi-dev
>>> Package libopenmpi-dev:
>>> i   2.1.1-8  bionic  500
>>> Package libopenmpi-dev:i386:
>>> p   2.1.1-8 bionic  500
>>> 
>>> $ sudo apt-get install libopenmpi-dev=2.1.6
>>> Reading package lists... Done
>>> Building dependency tree
>>> Reading state information... Done
>>> E: Version '2.1.6' for 'libopenmpi-dev' was not found
>>> 
>>> --Junchao Zhang
>>> 
>>> 
>>> On Thu, Aug 1, 2019 at 1:15 PM Jeff Squyres (jsquyres)  
>>> wrote:
>>> Does the bug exist in Open MPI v2.1.6?
>>> 
 On Jul 31, 2019, at 2:19 PM, Zhang, Junchao via users 
  wrote:
 
 Hello,
   I met a bug with OpenMPI 2.1.1 distributed in the latest Ubuntu 18.04.2 
 LTS. It happens with self to self send/recv using MPI_ANY_SOURCE for 
 message matching.  See the attached test code.  You can reproduce it even 
 with one process.
   It is a severe bug. Since this Ubuntu is widely used and has long term 
 support, could it be somehow fixed?
   Thanks a lot.
 
 --Junchao Zhang
 ___
 users mailing list
 users@lists.open-mpi.org
 https://lists.open-mpi.org/mailman/listinfo/users
>>> 
>>> -- 
>>> Jeff Squyres
>>> jsquy...@cisco.com
>>> 
>> 
> ___
> users mailing list
> users@lists.open-mpi.org
> https://lists.open-mpi.org/mailman/listinfo/users


-- 
Jeff Squyres
jsquy...@cisco.com

___
users mailing list
users@lists.open-mpi.org
https://lists.open-mpi.org/mailman/listinfo/users


Re: [OMPI users] OpenMPI 2.1.1 bug on Ubuntu 18.04.2 LTS

2019-08-01 Thread Gilles Gouaillardet via users

Juanchao,


Is the issue related to https://github.com/open-mpi/ompi/pull/4501 ?


Jeff,


you might have to configure with --enable-heterogeneous to evidence the 
issue




Cheers,


Gilles

On 8/2/2019 4:06 AM, Jeff Squyres (jsquyres) via users wrote:

I am able to replicate the issue on a stock Ubuntu 18.04 install with their 
Open MPI package.

But if I compile my own Open MPI 2.1.1, it works fine.
Also, if I compile my own Open MPI 2.1.6, it works fine.

I filed a bug at Ubuntu about this:

 https://bugs.launchpad.net/ubuntu/+source/xubuntu-meta/+bug/1838684




On Aug 1, 2019, at 2:33 PM, Zhang, Junchao  wrote:

$ aptitude versions libopenmpi-dev
Package libopenmpi-dev:
i   2.1.1-8  bionic  500
Package libopenmpi-dev:i386:
p   2.1.1-8 bionic  500

$ sudo apt-get install libopenmpi-dev=2.1.6
Reading package lists... Done
Building dependency tree
Reading state information... Done
E: Version '2.1.6' for 'libopenmpi-dev' was not found

--Junchao Zhang


On Thu, Aug 1, 2019 at 1:15 PM Jeff Squyres (jsquyres)  
wrote:
Does the bug exist in Open MPI v2.1.6?


On Jul 31, 2019, at 2:19 PM, Zhang, Junchao via users 
 wrote:

Hello,
   I met a bug with OpenMPI 2.1.1 distributed in the latest Ubuntu 18.04.2 LTS. 
It happens with self to self send/recv using MPI_ANY_SOURCE for message 
matching.  See the attached test code.  You can reproduce it even with one 
process.
   It is a severe bug. Since this Ubuntu is widely used and has long term 
support, could it be somehow fixed?
   Thanks a lot.

--Junchao Zhang
___
users mailing list
users@lists.open-mpi.org
https://lists.open-mpi.org/mailman/listinfo/users


--
Jeff Squyres
jsquy...@cisco.com




___
users mailing list
users@lists.open-mpi.org
https://lists.open-mpi.org/mailman/listinfo/users


Re: [OMPI users] OpenMPI 2.1.1 bug on Ubuntu 18.04.2 LTS

2019-08-01 Thread Jeff Squyres (jsquyres) via users
I am able to replicate the issue on a stock Ubuntu 18.04 install with their 
Open MPI package.

But if I compile my own Open MPI 2.1.1, it works fine.
Also, if I compile my own Open MPI 2.1.6, it works fine.

I filed a bug at Ubuntu about this:

https://bugs.launchpad.net/ubuntu/+source/xubuntu-meta/+bug/1838684



> On Aug 1, 2019, at 2:33 PM, Zhang, Junchao  wrote:
> 
> $ aptitude versions libopenmpi-dev
> Package libopenmpi-dev:
> i   2.1.1-8  bionic  500
> Package libopenmpi-dev:i386:
> p   2.1.1-8 bionic  500
> 
> $ sudo apt-get install libopenmpi-dev=2.1.6
> Reading package lists... Done
> Building dependency tree
> Reading state information... Done
> E: Version '2.1.6' for 'libopenmpi-dev' was not found
> 
> --Junchao Zhang
> 
> 
> On Thu, Aug 1, 2019 at 1:15 PM Jeff Squyres (jsquyres)  
> wrote:
> Does the bug exist in Open MPI v2.1.6?
> 
> > On Jul 31, 2019, at 2:19 PM, Zhang, Junchao via users 
> >  wrote:
> > 
> > Hello,
> >   I met a bug with OpenMPI 2.1.1 distributed in the latest Ubuntu 18.04.2 
> > LTS. It happens with self to self send/recv using MPI_ANY_SOURCE for 
> > message matching.  See the attached test code.  You can reproduce it even 
> > with one process.
> >   It is a severe bug. Since this Ubuntu is widely used and has long term 
> > support, could it be somehow fixed?
> >   Thanks a lot.
> > 
> > --Junchao Zhang
> > ___
> > users mailing list
> > users@lists.open-mpi.org
> > https://lists.open-mpi.org/mailman/listinfo/users
> 
> 
> -- 
> Jeff Squyres
> jsquy...@cisco.com
> 


-- 
Jeff Squyres
jsquy...@cisco.com

___
users mailing list
users@lists.open-mpi.org
https://lists.open-mpi.org/mailman/listinfo/users


Re: [OMPI users] OpenMPI 2.1.1 bug on Ubuntu 18.04.2 LTS

2019-08-01 Thread Zhang, Junchao via users
$ aptitude versions libopenmpi-dev
Package libopenmpi-dev:
i   2.1.1-8  bionic  500
Package libopenmpi-dev:i386:
p   2.1.1-8 bionic  500

$ sudo apt-get install libopenmpi-dev=2.1.6
Reading package lists... Done
Building dependency tree
Reading state information... Done
E: Version '2.1.6' for 'libopenmpi-dev' was not found

--Junchao Zhang


On Thu, Aug 1, 2019 at 1:15 PM Jeff Squyres (jsquyres) 
mailto:jsquy...@cisco.com>> wrote:
Does the bug exist in Open MPI v2.1.6?

> On Jul 31, 2019, at 2:19 PM, Zhang, Junchao via users 
> mailto:users@lists.open-mpi.org>> wrote:
>
> Hello,
>   I met a bug with OpenMPI 2.1.1 distributed in the latest Ubuntu 18.04.2 
> LTS. It happens with self to self send/recv using MPI_ANY_SOURCE for message 
> matching.  See the attached test code.  You can reproduce it even with one 
> process.
>   It is a severe bug. Since this Ubuntu is widely used and has long term 
> support, could it be somehow fixed?
>   Thanks a lot.
>
> --Junchao Zhang
> ___
> users mailing list
> users@lists.open-mpi.org
> https://lists.open-mpi.org/mailman/listinfo/users


--
Jeff Squyres
jsquy...@cisco.com

___
users mailing list
users@lists.open-mpi.org
https://lists.open-mpi.org/mailman/listinfo/users

Re: [OMPI users] OpenMPI 2.1.1 bug on Ubuntu 18.04.2 LTS

2019-08-01 Thread Jeff Squyres (jsquyres) via users
Does the bug exist in Open MPI v2.1.6?

> On Jul 31, 2019, at 2:19 PM, Zhang, Junchao via users 
>  wrote:
> 
> Hello,
>   I met a bug with OpenMPI 2.1.1 distributed in the latest Ubuntu 18.04.2 
> LTS. It happens with self to self send/recv using MPI_ANY_SOURCE for message 
> matching.  See the attached test code.  You can reproduce it even with one 
> process.
>   It is a severe bug. Since this Ubuntu is widely used and has long term 
> support, could it be somehow fixed?
>   Thanks a lot.
> 
> --Junchao Zhang
> ___
> users mailing list
> users@lists.open-mpi.org
> https://lists.open-mpi.org/mailman/listinfo/users


-- 
Jeff Squyres
jsquy...@cisco.com

___
users mailing list
users@lists.open-mpi.org
https://lists.open-mpi.org/mailman/listinfo/users


[OMPI users] OpenMPI 2.1.1 bug on Ubuntu 18.04.2 LTS

2019-07-31 Thread Zhang, Junchao via users
Hello,
  I met a bug with OpenMPI 2.1.1 distributed in the latest Ubuntu 18.04.2 LTS. 
It happens with self to self send/recv using MPI_ANY_SOURCE for message 
matching.  See the attached test code.  You can reproduce it even with one 
process.
  It is a severe bug. Since this Ubuntu is widely used and has long term 
support, could it be somehow fixed?
  Thanks a lot.

--Junchao Zhang


any.c
Description: any.c
___
users mailing list
users@lists.open-mpi.org
https://lists.open-mpi.org/mailman/listinfo/users