Intel® C++ Compiler


You're completly right I used outdated documentation which only covered SSE itself but not version 2.

Wow, tons of great instruction *happy*

Thanks a lot, lg Clemens

SIMD and integer computations :(

Hi there,

I am creating an application which needs to process a lot of integer-data very quickly - however I had yesterday a first look at the x86 SIMD extensions (MMX, SSE1/2) and realized that it does not look that good for integer calculations.

As it seems its not possible to simd' shift operations and doing a bit-wise AND is also just possible with MMX - which means I am limited to 64bits a time.
Is there something I've overlooked or some tricks to make SSE more useful for integer calculations?

Thanks in advance, lg Clemens

incremental linking without libirc_s

I'm trying to to an incremental link, passing


the icpc driver ends up passing -lirc_s to the linux linker, despite my adding

-nostdlib -nodefaultlibs -nostartfiles -i_dynamic

to the icpc command line as well.

Is there ANY way to prevent libirc_s.a from being searched?

Here is the command line I feed icpc, and the output showing the -lirc_s.

Does icc support "x86 built-in functions"?


I am new to icc and however was a bit dissapointed by the fact that icc is not able to generate simd-instructions out of my very simply code.

However since I need highest performance I would be interrested in using SSE directly via x86-builtin functions like gcc does provide.

128-bit Encryption using Intel SSE2 Extensions

Anybody need some source code to encrypt basic text messages using Intel SSE2 Extensions. Here we are using 128 bit XOR _mm_xor_si128() intrinsic function to obtain higher performance results. You must make a input.txt file in notepad or other ascii source editor and store in same directory as compiled program. Tell me how it works for you. And for those of you who are beginners and don't understand how the XOR is encrypting the data here is a sample data pattern... I attached the source code since itdidn't look correct in this HTML window.

Choice between 'extern inline' and macros

I want to know how the Intel C compiler handles the use of 'extern inline'. I intend to use either 'extern inline' or macros on a Linux P4 Xeon processor system and am using -O2 level of optimization. How do I confirm whether or not the Intel C compiler has actual expanded an inline function? Given that I want to avoid the overhead of making a function call, would you recommend using 'extern inline' or macros? Please let me know if my question is clear or more information is needed.


Subscribe to Intel® C++ Compiler