[mpich-discuss] Cannot catch std::bac_alloc?

[Joachim Protze]

Jeff,

as I understand the original question, this is the intention of the 
statement: reproducibly trigger an allocation exception, which should be 
caught by the try-catch block.


Hui,

As you will probably experience, your intended behavior will not be 
reached with replacing the vector resize by allocation with new:

std::vector<double*> a(100);
...
a[i] = new double[1000000000];

The reason is, that new allocates virtual memory. You will see a 
segfault, when you try to initialize the whole array and request 
physical memory for the array.
The vector resize operation allocates the memory and immediately 
initializes the memory. Therefore not the allocation itself fails, but 
finally the access to the memory during initialization.
I have no explanation why you can see the bad_alloc when executing it 
only on one rank.

Best
Joachim


On 4/3/19 6:51 PM, Jeff Hammond via discuss wrote:
> This error has nothing to do with MPI.
> 
> These two lines are the problem:
> 
> std::vector<std::vector<double> > a(100);
> a[i].resize(1000000000);
> 
> You are trying to allocate 1e9 std::vector<double>, which I assume is 
> many gigabytes of memory.
> 
> Jeff
> 
> 
> On Wed, Apr 3, 2019 at 8:58 AM Zhou, Hui via discuss <discuss at mpich.org 
> <mailto:discuss at mpich.org>> wrote:
>  >
>  > I just tested on my Mac, the system keep popping out message asking 
> me to force kill some application to free up memory, but the code never 
> crashed, just stalled. So I guess the out-of-memory behavior is 
> operating system dependent. My guess is when you have multiple process 
> competing for memory, it may cause some race issue in the os that the OS 
>   returns a valid address but discover out-of memory later because the 
> memory is stolen by another process.
>  >
>  > —
>  > Hui Zhou
>  > T: 630-252-3430
>  >
>  >
>  >
>  >
>  >
>  >
>  >
>  >
>  > On Apr 3, 2019, at 8:42 AM, Zhen Wang via discuss <discuss at mpich.org 
> <mailto:discuss at mpich.org>> wrote:
>  >
>  > Hi,
>  >
>  > I have difficulty catching std::bac_alloc in an MPI environment. The 
> code is attached. I'm uisng gcc 6.3 on SUSE Linux Enterprise Server 11 
> (x86_64). mpich is built from source. The commands are as follows:
>  >
>  > Build
>  > g++ -I<mpich-3.3-opt/include> -L<mpich-3.3-opt/lib> -lmpi memory.cpp
>  >
>  > Run
>  > <mpich-3.3-opt/bin/mpiexec> -n 2 a.out
>  >
>  > Output
>  > 0
>  > 0
>  > 1
>  > 1
>  >
>  > 
> ===================================================================================
>  > =   BAD TERMINATION OF ONE OF YOUR APPLICATION PROCESSES
>  > =   PID 16067 RUNNING AT <machine name>
>  > =   EXIT CODE: 9
>  > =   CLEANING UP REMAINING PROCESSES
>  > =   YOU CAN IGNORE THE BELOW CLEANUP MESSAGES
>  > 
> ===================================================================================
>  > YOUR APPLICATION TERMINATED WITH THE EXIT STRING: Killed (signal 9)
>  > This typically refers to a problem with your application.
>  > Please see the FAQ page for debugging suggestions
>  >
>  > If I uncomment the line //if (rank == 0), i.e., only rank 0 allocates 
> memory, I'm able to catch bad_alloc as I expected. It seems that I am 
> misunderstanding something. Could you please help? Thanks a lot.
>  >
>  >
>  > Best regards,
>  > Zhen
>  > <memory.cpp>_______________________________________________
>  > discuss mailing list discuss at mpich.org <mailto:discuss at mpich.org>
>  > To manage subscription options or unsubscribe:
>  > https://lists.mpich.org/mailman/listinfo/discuss
>  >
>  >
>  > _______________________________________________
>  > discuss mailing list discuss at mpich.org <mailto:discuss at mpich.org>
>  > To manage subscription options or unsubscribe:
>  > https://lists.mpich.org/mailman/listinfo/discuss
> 
> 
> 
> --
> Jeff Hammond
> jeff.science at gmail.com <mailto:jeff.science at gmail.com>
> http://jeffhammond.github.io/
> 
> _______________________________________________
> discuss mailing list     discuss at mpich.org
> To manage subscription options or unsubscribe:
> https://lists.mpich.org/mailman/listinfo/discuss
> 


-- 
Dipl.-Inf. Joachim Protze

IT Center
Group: High Performance Computing
Division: Computational Science and Engineering
RWTH Aachen University
Seffenter Weg 23
D 52074  Aachen (Germany)
Tel: +49 241 80- 24765
Fax: +49 241 80-624765
protze at itc.rwth-aachen.de
www.itc.rwth-aachen.de

-------------- next part --------------
A non-text attachment was scrubbed...
Name: smime.p7s
Type: application/pkcs7-signature
Size: 4915 bytes
Desc: S/MIME Cryptographic Signature
URL: <http://lists.mpich.org/pipermail/discuss/attachments/20190403/54d380c5/attachment.p7s>