__sec_reduce_all_nonzero crashing compiler with backend signals

[MHlav]

Why does the following code crash the compiler with backend signals?  Am I doing something wrong?

#include <mathimf.h>
#include <float.h>

double avec[] = {0, 1, 2, 3, 4};

__sec_reduce_all_nonzero(_finite(avec[0:sizeof(avec)/sizeof(avec[0])]));

Thanks.

[SergeyKostrov]

I'd like to follow up: - What version of Intel C++ compiler do you use? - What is __sec_reduce_all_nonzero function? I do not see it declared in mathimf.h header file.

[MHlav]

It documented here:  http://software.intel.com/sites/products/documentation/doclib/iss/2013/compiler/cpp-lin/GUID-6FB572F4-A455-41D1-A799-9E6C20AF2C25.htm

I'm using Composer XE 2013 Update 5 (Intel(R) C++ Compiler XE on IA-32, version 13.1.3 Package ID: w_ccompxe_2013.5.198)

[Amanda_S_Intel]

Thanks for the nice test case - I will pass this on to the developers. The issue ID is DPD200246880.

[SergeyKostrov]

>>It documented here: http://software.intel.com/sites/products/documentation/doclib/iss/2013/compiler/ >>cpp-lin/GUID-6FB572F4-A455-41D1-A799-9E6C20AF2C25.htm >> >>I'm using Composer XE 2013 Update 5 (Intel(R) C++ Compiler XE on IA-32, version 13.1.3 Package ID: >>w_ccompxe_2013.5.198) Thanks. I see that the function you're talking about is a built-in function and is Not declared in any header files.

[Amanda_S_Intel]

This is targeted to be fixed in an upcoming release.

[Amanda_S_Intel]

This problem is fixed in Intel® C++ Composer XE 2013 SP1 Update 1. Thanks for your report.