Re: [deal.II] Re: Regarding tbb error

2023-04-28 Thread Marc Fehling 
Interestingly, the cmake configuration also complains about the following 
unused variable. This variable is now named DEAL_II_FORCE_BUNDLED_TBB. Can 
someone of the candi developers have a look?

CMake Warning:
  Manually-specified variables were not used by the project:
DEAL_II_FORCE_BUNDLED_THREADS

On Friday, April 28, 2023 at 2:05:36 PM UTC-6 Marc Fehling wrote:

> Hello Jau-Uei Chen,
>
> the configuration did not pick up the TBB version you provided, and 
> instead uses the version that comes bundled with deal.II (i.e. tbb 2018U2). 
> However, the linker still tries to link against tbb 2021.6.0. Something got 
> mixed up in the configuration phase.
>
> Marc
>
>
> -- Include 
> /org/groups/ceo/chenju/dealii/install/tmp/unpack/deal.II-v9.4.0/cmake/configure/configure_10_tbb.cmake
> -- TBB_LIBRARY not found! The call was:
> -- FIND_LIBRARY(TBB_LIBRARY NAMES tbb HINTS PATH_SUFFIXES lib lib64 
> lib)
> -- TBB_DEBUG_LIBRARY not found! The call was:
> -- FIND_LIBRARY(TBB_DEBUG_LIBRARY NAMES tbb_debug HINTS PATH_SUFFIXES 
> lib lib64 lib)
> -- TBB_INCLUDE_DIR not found! The call was:
> -- FIND_PATH(TBB_INCLUDE_DIR tbb/tbb_stddef.h HINTS PATH_SUFFIXES 
> include include/tbb tbb)
> -- Found TBB_INCLUDE_DIR
> --   TBB_VERSION: 2021.6
> --   TBB_LIBRARIES: *** Required variable "TBB_LIBRARY" set to NOTFOUND ***
> -- Could NOT find TBB
> -- DEAL_II_WITH_TBB has unmet external dependencies.
> -- Performing Test DEAL_II_HAVE_FLAG_Wno_parentheses
> -- Performing Test DEAL_II_HAVE_FLAG_Wno_parentheses - Success
> -- DEAL_II_WITH_TBB successfully set up with bundled packages.
>
>
> [ 50%] Linking CXX executable step-6
> /usr/bin/ld.bfd: CMakeFiles/step-6.dir/step-6.cc.o: in function 
> `tbb::detail::d1::wait_context::add_reference(long)':
> /opt/apps/sysnet/intel/oneapi/2022.2/tbb/2021.6.0/include/oneapi/tbb/detail/_task.h:114:
>  
> undefined reference to `tbb::detail::r1::notify_waiters(unsigned long)'
> On Friday, April 28, 2023 at 1:28:04 PM UTC-6 jau...@gmail.com wrote:
>
>> Dear all,
>>
>> I just encounter a similar issue and am wondering if there is a solution 
>> to fix it.
>>
>> Please see the attachment for the error message I got. This message is 
>> from dealii 9.4.0 (I also tried dealii 9.4.2 and got a similar message.). 
>>
>> During the installation of dealii 9.4.0, no issue is showing up though 
>> (The installation is carried out by using candi. I also attach the log 
>> here.). The tbb I used is "tbb/2021.6.0" (Intel(R) oneAPI Threading 
>> Building Blocks for intel64).
>>
>> Any comment or suggestion is greatly appreciated.
>>
>> Best Regards,
>> Jau-Uei Chen
>>
>> Deepika Kushwah 在 2022年9月19日 星期一晚上11:08:55 [UTC-5] 的信中寫道：
>>
>>> Thanks for the reply.
>>>
>>>
>>> I am using the latest version of dealii 9.4.0.
>>> With this version I am getting the error as mentioned in the previous 
>>> email.
>>>
>>>
>>>
>>>
>>> Thanks & Regards
>>> Deepika
>>>
>>> On Tue, Sep 20, 2022 at 2:48 AM Marc Fehling  wrote:
>>>
 Hello Deepika,

 which version of deal.II are you using?

 From your screenshots, it looks like you are using Intel oneAPI TBB. We 
 started to support oneAPI versions with deal.II 9.4.0. Could you try our 
 latest release?

 Marc

 On Sunday, September 18, 2022 at 6:18:51 AM UTC-6 
 deepika...@iitgoa.ac.in wrote:

> Hello Everyone,
>
> I am a beginner in deal.II.
> While running any tutorial on centos7 I am getting the following 
> errors (please see attachments).
>
> How to proceed?
>
>
> Thanks & Regards
> Deepika
>
 -- 
 The deal.II project is located at http://www.dealii.org/
 For mailing list/forum options, see 
 https://groups.google.com/d/forum/dealii?hl=en
 --- 
 You received this message because you are subscribed to the Google 
 Groups "deal.II User Group" group.
 To unsubscribe from this group and stop receiving emails from it, send 
 an email to dealii+un...@googlegroups.com.
 To view this discussion on the web visit 
 https://groups.google.com/d/msgid/dealii/e0f889c4-38b1-403f-bfb8-58ddb01a70d0n%40googlegroups.com
  
 
 .

>>>
>>> 
>>> **
>>> This e-mail is for the sole use of the intended recipient(s) and may
>>> contain confidential and privileged information. If you are not the
>>> intended recipient, please contact the sender by reply e-mail and destroy
>>> all copies and the original message. Any unauthorized review, use,
>>> disclosure, dissemination, forwarding, printing or copying of this email
>>> is strictly prohibited and appropriate legal action will be taken. 
>>> 
>>>  
>>>
>>

-- 
The deal.II project is

Re: [deal.II] Re: Regarding tbb error

2023-04-28 Thread Marc Fehling 
Hello Jau-Uei Chen,

the configuration did not pick up the TBB version you provided, and instead 
uses the version that comes bundled with deal.II (i.e. tbb 2018U2). 
However, the linker still tries to link against tbb 2021.6.0. Something got 
mixed up in the configuration phase.

Marc


-- Include 
/org/groups/ceo/chenju/dealii/install/tmp/unpack/deal.II-v9.4.0/cmake/configure/configure_10_tbb.cmake
-- TBB_LIBRARY not found! The call was:
-- FIND_LIBRARY(TBB_LIBRARY NAMES tbb HINTS PATH_SUFFIXES lib lib64 lib)
-- TBB_DEBUG_LIBRARY not found! The call was:
-- FIND_LIBRARY(TBB_DEBUG_LIBRARY NAMES tbb_debug HINTS PATH_SUFFIXES 
lib lib64 lib)
-- TBB_INCLUDE_DIR not found! The call was:
-- FIND_PATH(TBB_INCLUDE_DIR tbb/tbb_stddef.h HINTS PATH_SUFFIXES 
include include/tbb tbb)
-- Found TBB_INCLUDE_DIR
--   TBB_VERSION: 2021.6
--   TBB_LIBRARIES: *** Required variable "TBB_LIBRARY" set to NOTFOUND ***
-- Could NOT find TBB
-- DEAL_II_WITH_TBB has unmet external dependencies.
-- Performing Test DEAL_II_HAVE_FLAG_Wno_parentheses
-- Performing Test DEAL_II_HAVE_FLAG_Wno_parentheses - Success
-- DEAL_II_WITH_TBB successfully set up with bundled packages.


[ 50%] Linking CXX executable step-6
/usr/bin/ld.bfd: CMakeFiles/step-6.dir/step-6.cc.o: in function 
`tbb::detail::d1::wait_context::add_reference(long)':
/opt/apps/sysnet/intel/oneapi/2022.2/tbb/2021.6.0/include/oneapi/tbb/detail/_task.h:114:
 
undefined reference to `tbb::detail::r1::notify_waiters(unsigned long)'
On Friday, April 28, 2023 at 1:28:04 PM UTC-6 jau...@gmail.com wrote:

> Dear all,
>
> I just encounter a similar issue and am wondering if there is a solution 
> to fix it.
>
> Please see the attachment for the error message I got. This message is 
> from dealii 9.4.0 (I also tried dealii 9.4.2 and got a similar message.). 
>
> During the installation of dealii 9.4.0, no issue is showing up though 
> (The installation is carried out by using candi. I also attach the log 
> here.). The tbb I used is "tbb/2021.6.0" (Intel(R) oneAPI Threading 
> Building Blocks for intel64).
>
> Any comment or suggestion is greatly appreciated.
>
> Best Regards,
> Jau-Uei Chen
>
> Deepika Kushwah 在 2022年9月19日 星期一晚上11:08:55 [UTC-5] 的信中寫道：
>
>> Thanks for the reply.
>>
>>
>> I am using the latest version of dealii 9.4.0.
>> With this version I am getting the error as mentioned in the previous 
>> email.
>>
>>
>>
>>
>> Thanks & Regards
>> Deepika
>>
>> On Tue, Sep 20, 2022 at 2:48 AM Marc Fehling  wrote:
>>
>>> Hello Deepika,
>>>
>>> which version of deal.II are you using?
>>>
>>> From your screenshots, it looks like you are using Intel oneAPI TBB. We 
>>> started to support oneAPI versions with deal.II 9.4.0. Could you try our 
>>> latest release?
>>>
>>> Marc
>>>
>>> On Sunday, September 18, 2022 at 6:18:51 AM UTC-6 
>>> deepika...@iitgoa.ac.in wrote:
>>>
 Hello Everyone,

 I am a beginner in deal.II.
 While running any tutorial on centos7 I am getting the following errors 
 (please see attachments).

 How to proceed?


 Thanks & Regards
 Deepika

>>> -- 
>>> The deal.II project is located at http://www.dealii.org/
>>> For mailing list/forum options, see 
>>> https://groups.google.com/d/forum/dealii?hl=en
>>> --- 
>>> You received this message because you are subscribed to the Google 
>>> Groups "deal.II User Group" group.
>>> To unsubscribe from this group and stop receiving emails from it, send 
>>> an email to dealii+un...@googlegroups.com.
>>> To view this discussion on the web visit 
>>> https://groups.google.com/d/msgid/dealii/e0f889c4-38b1-403f-bfb8-58ddb01a70d0n%40googlegroups.com
>>>  
>>> 
>>> .
>>>
>>
>> 
>> **
>> This e-mail is for the sole use of the intended recipient(s) and may
>> contain confidential and privileged information. If you are not the
>> intended recipient, please contact the sender by reply e-mail and destroy
>> all copies and the original message. Any unauthorized review, use,
>> disclosure, dissemination, forwarding, printing or copying of this email
>> is strictly prohibited and appropriate legal action will be taken. 
>> 
>>  
>>
>

-- 
The deal.II project is located at http://www.dealii.org/
For mailing list/forum options, see 
https://groups.google.com/d/forum/dealii?hl=en
--- 
You received this message because you are subscribed to the Google Groups 
"deal.II User Group" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to dealii+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/dealii/89bbe432-93de-4d9b-9210-c57364025bf3n%40googlegroups.com.

Re: [deal.II] Internal deal.II error when instantiating TrilinosWrappers::block_operator

2023-04-28 Thread Luca Heltai 
Dear Lucas, 

take a look at the tests directory of deal.II:

https://github.com/dealii/dealii/blob/master/tests/lac/block_linear_operator_05.cc

there are many small examples and snippets that show how to use this.

In particular, I think the problem is in the fact that you are calling directly 
dealii::TrilinosWrappers::linear_operator and 
dealii::TrilinosWrappers::block_operator.

These are used internally by the functions in linear_operator_tools, and, btw, 
you are not required to include

#include 

as these are included by linear_operator_tools.h, and the typical usage is 
through the functions dealii::linear_operator, dealii::block_operator, 
dealii::block_diagonal_operator, etc. (notice these are in dealii namespace, 
not in dealii::TrilinosWrappers namespace).

e.g., this should compile:

using vec = dealii::LinearAlgebraTrilinos::MPI::Vector;
using block_vec = dealii::LinearAlgebraTrilinos::MPI::BlockVector;

const auto B = linear_operator(system_matrix.block(0, 0)); // Notice this 
is not dealii::TrilinosWrappers::linear_operator
const auto P_inv = block_diagonal_operator<1, block_vec 
>(std::array({{B}}));

Creating (as for your example) a block operator with a single block.

L.

Ps: I don’t see an obvious reason why your example should not compile too. 
Could you try to modify

https://github.com/dealii/dealii/blob/master/tests/lac/block_linear_operator_05.cc

to reflect your example, and open an issue with a MWE that fails to compile? 
Reading your code, I would have expected it to compile, and I think you have 
spotted a problem in deal.II due to an unusual way in which you are using the 
code. 




> On 28 Apr 2023, at 24:10, Lucas Myers  wrote:
> 
> Hi folks,
> 
> I'm trying to use the block_operator technique on Trilinos vectors and 
> matrices, but I can't quite get it to compile. For simplicity, the relevant 
> lines of code are:
> ```
> #include 
> #include 
> 
> using vec = dealii::LinearAlgebraTrilinos::MPI::Vector;
> using block_vec = dealii::LinearAlgebraTrilinos::MPI::BlockVector;
> const auto B = dealii::TrilinosWrappers::linear_operator vec>(system_matrix.block(0, 0));
> const auto P_inv = dealii::TrilinosWrappers::block_operator<1, 1, 
> block_vec>({B});
> ```
> where system_matrix is of type 
> `dealii::LinearAlgebraTrilinos::MPI::BlockSparseMatrix`.
> 
> The first error that comes up looks like:
> 
> `deal.II/lac/block_linear_operator.h:679:37: error: 
> ‘populate_linear_operator_functions’ was not declared in this scope; did you 
> mean 
> ‘dealii::internal::BlockLinearOperatorImplementation::populate_linear_operator_functions’?`
> 
> Is this a bug in deal.II? That would be my first inclination, given that I 
> cannot see how it would be template-related.
> 
> There are a few other errors that come up, so I'm attaching the error message 
> in a text file. 
> 
> Any tips on debugging this? I'm finding the compiler output basically 
> inscrutable. Any help is appreciated, so sorry for all the spam!
> 
> - Lucas
> 
> -- 
> The deal.II project is located at http://www.dealii.org/
> For mailing list/forum options, see 
> https://groups.google.com/d/forum/dealii?hl=en
> --- 
> You received this message because you are subscribed to the Google Groups 
> "deal.II User Group" group.
> To unsubscribe from this group and stop receiving emails from it, send an 
> email to dealii+unsubscr...@googlegroups.com.
> To view this discussion on the web visit 
> https://groups.google.com/d/msgid/dealii/8a6be404-f323-4edc-9720-fe603c92dfe6n%40googlegroups.com.
> 

-- 
The deal.II project is located at http://www.dealii.org/
For mailing list/forum options, see 
https://groups.google.com/d/forum/dealii?hl=en
--- 
You received this message because you are subscribed to the Google Groups 
"deal.II User Group" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to dealii+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/dealii/D553461D-423F-4B77-A842-96FABE9DF0EA%40gmail.com.

Re: [deal.II] Simple Query about an example in the code gallary

2023-04-28 Thread Abbas Ballout 
Sorry I deleted this before it had any comments on it. 
I thought it would disappear. Won't happen again. 
 
Abbas 

On Wednesday, April 26, 2023 at 7:18:59 PM UTC+2 Wolfgang Bangerth wrote:

> On 4/26/23 03:34, Abbas wrote:
> > Why is it that in step 40 the locally relevant solution is initialised 
> with 
> > both locally_owned_dofs and locally_relevant_dofs, while in the 
> > Distributed_LDG_Method example in the code gallery the locally relevant 
> > solution is just initialised with just the locally_relevant_dofs only?
>
> The code gallery program uses this function [2/5]:
>
> https://dealii.org/developer/doxygen/deal.II/classTrilinosWrappers_1_1MPI_1_1Vector.html#af74a39fa88174f63767d510416c09e6f
> The tutorial, on the other hand, uses [3/5] immediately below.
>
> Both are reasonable. I would gladly accept a patch that switches the use 
> from 
> [2/5] to [3/5] because that version is clearly more obvious.
>
> Best
> W.
>
> -- 
> 
> Wolfgang Bangerth email: bang...@colostate.edu
> www: http://www.math.colostate.edu/~bangerth/
>
>
>

-- 
The deal.II project is located at http://www.dealii.org/
For mailing list/forum options, see 
https://groups.google.com/d/forum/dealii?hl=en
--- 
You received this message because you are subscribed to the Google Groups 
"deal.II User Group" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to dealii+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/dealii/a40acae0-391b-47a3-bceb-1e4c5bc40dbdn%40googlegroups.com.