topic Re: About ippsEncodeHuffInit_BZ2_16u8u() in Intel® Integrated Performance Primitives

About ippsEncodeHuffInit_BZ2_16u8u()

Arcady_Chernyak — Mon, 28 Dec 2009 21:04:23 GMT

The function ippsEncodeHuffInit_BZ2_16u8u() takes to much time on my linux embedded.
Can I split it to 2 (or more) other (or same) API calls with reduced parameters values?
Does sombody have "c" like algorithm for this function?
Or/and type of struct IppEncodeHuffState_BZ2{}???

I am trying implement BZIP2 based on IPP samples.
It is Linux embeded and I have restictions for:
1. Time elapsed in single IPP call
2. No semaphores or other blocking calls inside IPP LIB
3. All memory alloc/free should be done outside IPP lib

Thanks
Arcady

Re: About ippsEncodeHuffInit_BZ2_16u8u()

Alexander_Naduev__In — Wed, 30 Dec 2009 07:57:21 GMT

Hello Arcady,
first of all, you can not split this function into 2 or more functions.Next, please, use for an example our ipp_bzip2 sample, which is based on these IPP functions.

Re: About ippsEncodeHuffInit_BZ2_16u8u()

Ying_H_Intel — Thu, 31 Dec 2009 02:04:58 GMT

Hello Arcady,

Just for your information, The IPP bzip2 sample can be downloaded from

This sample implements the bzip2/libbzip2 application and library (http://www.bzip.org/) using the Intel IPP data compression functions library. The Intel IPP version of the bzip2 library implements a lossless, block-sorting data compression algorithm that includes threading optimization improvements.

http://software.intel.com/en-us/articles/intel-integrated-performance-primitives-code-samples/

Regards,
Ying