Informatique parallèle

offload error (trying my first program)

I am just trying to get our XeonPhi to work properly. It worked for offload computing a couple of months ago when I tried it, but now I get the following error:

offload error: cannot find offload entry __offload_entry_Conceptual_Transformation_cpp_40f_7e694fe6ef39bee36b1d18aad1ea599dicc43479367WrTpxg
offload error: process on the device 0 unexpectedly exited with code 1

icc version 14.0.2 (gcc version 4.7.2 compatibility)

MPSS 3.5

Code is just a simple example to try out offloading. Any help appreciated.

analyzing .optrpt file


With the intention to ensure/check if a given loop is parallel or not, I generated an .optrpt file.

I used the following flags for the same:   -parallel -opt-report-phase=par -opt-report:5

However I have three doubts persisting that the auto-parallelization flags failed to answer.

1) I had a two level nested loop like this
do k=1,km-1
                   do kk=1,2
<code is here>

Haswell Transactional Memory read/write-set information

Recently, Intel release haswell machines which support hardware transactional memory called transactional synchronization extension(TSX).

As Intel manual said, Speculative memory operations, write-set and read-set, are buffered in L1 cache and L2 cache each. (not exactly)

Then, Can I track transactional memory operations and get information like address, and values of read/write-set?

I have a problem with igzip

I am studying about compression algorithm and software.
I have question about igzip. I download igzip library in intel homepage.
But I don`t know how to make wrapper.
Can you send me 'example of wrapper' or 'example code' or 'manual'?
I read homepage and saw a simple application.
I don`t know how to input target file for compression and to output compression file
and how to decompression?
Do I make code about 'fast_lz and init_stream' function by myself?
Plz help me.
thank you

use gfortran to link mkl libraries

I am writing to ask help for linking mkl with a fortran code on a Surface Pro with win 8.1 OS. I am using gfortran 4.9.2 (TDM-GCC-65). The Fortran code is paralleled using OpenMP.

I am using the following command:

gfortran -fopenmp -o mycode *.o mkl_rt.lib libiomp5md.lib

It links ok. But during execution, the executable crashes and reports Program received signal SIGSEGV: Segmentation fault - invalid memory reference.

I also tried the following

gfortran -fopenmp -o mycode *.o mkl_rt.lib libiomp5md.lib

I got the following linking error:

PCM reporting lower than expected memory read counts

I have a piece of code on which I'm running PCM (Performance Counter Monitor). It is essentially the following:

uint64_t *a,*b;
a = new uint64_t[LEN];
b = new uint64_t[LEN];
for( int i=0;i<LEN;i++ ) a[i] = b[i];

With LEN set to 402,653,184 (384 Mi), PCM is reporting 0.72 GB under READ and 6.30 GB under WRITE. Given that each array is 3 GiB, I would expect that both arrays would be read (since processor uses write-allocate), giving a READ of about 6 GiB. I would expect array "a" to be written back, giving a write count of 3 GiB.

Updated from IPP 7.1.1 to 8.2.1, seeing segmentation faults on AVX (e9)

We have been using the Intel IPP's for many years now (Dialogic was once an Intel Company :)).  A few years back we updated to version 7.1.1 and all was well until we ran into some segmentation faults on certain newer systems.  The crashes were on systems which supported AVX and AVX2 processors. We found that we were able to work around this by limiting the CPU type to AVX.

We recently updated to IPP 8.2.1 hoping that this limitation would no longer be required.  However, we are seeing more frequent segmentation faults on systems which support AVX using the e9 IPP functions.

S’abonner à Informatique parallèle