Hi,

I've encountered a strange problem using ippsConv_32f function. It happens once in a while and isn't reproducible all the time.

This function gets 2 pointers to inputs (sizes M and N) and one pointer for output (size N+M-1). 

Even though the inputs are perfectly fine, the function crushes due to a memory problem.

I found that what solves the problem is allocating more space for the first input (N+A instead of N).

I tried putting different values in the extra A places in the first input array and saw that the result isn't effected by these values (as it should be).

So it seems that the function doesn't actually use the extra values but it uses the extra allocated space.

Most of the times the function runs fine, this problem happens only in one case from time to time.

Is it something known? If so, what is the general way to deal with this?

Thank you very much,

Sasha.

Hi ,

I am using IPP 8.0 with Intel C++ Compiler 9.1 .

a. ) Is IPP 8.0 compatible with this compiler version?

b.) Will I get any performance improvement if I use latest Compiler version?

Thanks,

Anchal

Buy International English Language Testing System(IELTS)
Buy Registered IELTS certificates without attending the Exam

(ieltscertificate99@hotmail.com)
We sell registered IELTS & TOEFL, ESOL, and CELTA/DELTA and other English Language 

certificates.

We deal and specialize in the production of registered TOEFL, IELTS, ESOL, CELTA/DELTA & 

other English Language Certificates. Please note that Our IELTS & TOEFL Certificates are 

Original and registered in the data base and Can be verified.After your order is placed it 

takes just few days for us to get your details in the system Once your details are imputed 

in the system it will be in the IELTS or TOEFL web sites/system once for ever and will 

appear real, legit and verifiable for ever.

WE can also help you to get valid Work permits,Driver's license ,second passport and Visas 

to European ,USA,Canada and Australia .

an enquiry send an email : contact email ieltscertificate99@hotmail.com

contact email ieltscertificate99@hotmail.com

ieltscertificate99@hotmail.com

skype ielts.certificate

we provide the beow documents and certificates 
Ielts certificate
Canada Cards
United States Cards
Student Cards
International Cards
Private Cards
Adoption Certificates
Baptism Certificates
Birth Certificates
Death Certificates
Divorce Certificates
Marriage Certificates
Custom Certificates
High School Diplomas
G.E.D. Diplomas
Home School Diplomas
College Degrees
University Degrees
Trade Skill Certificates
Social Security
Validate SSN Number

contact email ieltscertificate99@hotmail.com

Hello,

We have been using IPP 7 and their related version of h264 codec samples (vm, umc,…) but we have recently been facing some issues with new targets (Windows 8.1 tablets).

To benefit from Windows 8.1 support as stated in IPP 9 release note, we have bought ipp 9.

But in this version, it turns out that samples are deprecated...

Is there a h264 codec sample compliant with IPP 9 ?

How can I upgrade our h264 codec implementation based on samples and ipp 7 to ipp 9 ?

Thanks for your help

I'm developing an application using OpenCV 3.0, which uses the Intel IPP-CV libraries to accelerate certain image processing functions. Everything will be statically linked with my executable. What is the licensing situation here? Am I allowed to commercially distribute my application if IPP-CV is present? Do I need a separate licence? I couldn't find this information in the IPP-CV faq and the licence document inside the ippicv folder has a generic licence.

We used IPP 7.1.1 and looking to upgrade to IPP 9.0.1 but found ippiFilterMedian_8u_C1R to be missing.  In fact, the median filter functions were reduced from 16 in v7.1.1 to only 3 in v9.0.1.  Can someone help me understand this change please, and what I can do to fill in this missing library. THANK YOU.  

I have a Community License and it seems only the latest version of IPP is available which is Version 2016 (Update 1) or Version 9.0.1.150. How can I download the older version? In other words, does the Community License allow access to older version?

I have read this page https://software.intel.com/en-us/articles/older-version-product but don't I don't see the older versions of IPP being available or selectable for download.

    The ippsWTFwd creates one level of wavelet.
ippsWTFwd_32f( pSrc, pLow, pHigh, 6, pFwdState );

If i want to create another level of wavelet, shall i set pLow as input?

Such as:

ippsWTFwd_32f( pLow, pLow2, pHigh2, 6, pFwdState2 );

And shall i  create another pFwdState2 as it is said in other topics?

So the forward and inverse wavelet transform is iterative？

The wavelet transform example uses 'db2' filter which can be find in below link:

https://software.intel.com/en-us/forums/intel-integrated-performance-pri...

According to the wavelet transform examples, I get the  'db4' filter values from matlab using wfilters('db4') command.

The length of Lo_D,Hi_D,Lo_R,Hi_R is 8, so I define 

static const int fwdFltLenL = 8;
static const int fwdFltLenH = 8;

....

And copy the values from matlab:

static const float fwdFltL[8] = {-0.0106......}; 

...................

Other code are the same with the example.

....

ippsWTFwd_32f (src, low, high, 6, state);

........

But the low and high values are not equal to the values get by matlab.

So how to create a 'db4' wavelet transform?

I'm trying to update a library built with IPP 7.0 to the latest version, 9.0 Update 1. It used the IPP speech codec libippsc which has been removed from official IPP 9.0 product, so I downloaded and install the IPP 9.0 legacy libraries (available with 9.0 release).

I've updated the corresponding sources, it's compiling fine, but when doing [static] linking, I get some errors, conflict between libippsc90lgc and libippcore :I've reproduced below some interesting parts of the linking command and the resulting error messages :

/usr/bin/cc  -fPIC -m64  -Wall -Wextra -Wbad-function-cast -Wstrict-prototypes -march=core2 -mfpmath=sse -pthread -g -O3 -D_XOPEN_SOURCE=600 -D_DEFAULT_SOURCE -D_BSD_SOURCE -DORTP_INET6 -DUSE_INTEL_IPP90LEGACY -DUSE_INTEL_IPP -DG729  [...] /opt/intel/ipp90legacy_lin/linux/lib/intel64/threaded/libippsc90lgc.a /opt/intel/ipp-9.0/ipp/lib/intel64/threaded/libipps.a /opt/intel/ipp-9.0/ipp/lib/intel64/threaded/libippcore.a -lcrypto -lpthread -lm -lortp -lrt -laio
/opt/intel/ipp-9.0/ipp/lib/intel64/threaded/libippcore.a(owncpufeatures.o): In function `ownGetMaskFeatures':
/localdisk/commander/production/ipp901gold/linux_intel64/sources/core/src/owncpufeatures.c:(.text+0x0): multiple definition of `ownGetMaskFeatures'
/opt/intel/ipp90legacy_lin/linux/lib/intel64/threaded/libippsc90lgc.a(owncpufeatures.o):/localdisk/commander/lgc90/sources/core/src/owncpufeatures.c:(.text+0x0): first defined here
/opt/intel/ipp-9.0/ipp/lib/intel64/threaded/libippcore.a(owncpufeatures.o): In function `ownGetFeature':
/localdisk/commander/production/ipp901gold/linux_intel64/sources/core/src/owncpufeatures.c:(.text+0x3c0): multiple definition of `ownGetFeature'
/opt/intel/ipp90legacy_lin/linux/lib/intel64/threaded/libippsc90lgc.a(owncpufeatures.o):/localdisk/commander/lgc90/sources/core/src/owncpufeatures.c:(.text+0x300): first defined here
/opt/intel/ipp-9.0/ipp/lib/intel64/threaded/libippcore.a(owncpufeatures.o): In function `ownSetFeatureMask':
/localdisk/commander/production/ipp901gold/linux_intel64/sources/core/src/owncpufeatures.c:(.text+0x3e0): multiple definition of `ownSetFeatureMask'
/opt/intel/ipp90legacy_lin/linux/lib/intel64/threaded/libippsc90lgc.a(owncpufeatures.o):/localdisk/commander/lgc90/sources/core/src/owncpufeatures.c:(.text+0x320): first defined here
/opt/intel/ipp-9.0/ipp/lib/intel64/threaded/libippcore.a(owncpufeatures.o): In function `ownMaskIsFeatures':
/localdisk/commander/production/ipp901gold/linux_intel64/sources/core/src/owncpufeatures.c:(.text+0x420): multiple definition of `ownMaskIsFeatures'
/opt/intel/ipp90legacy_lin/linux/lib/intel64/threaded/libippsc90lgc.a(owncpufeatures.o):/localdisk/commander/lgc90/sources/core/src/owncpufeatures.c:(.text+0x340): first defined here
collect2: error: ld returned 1 exit status

How could I fix this linking problem ?

   Thanks.

Sorry to make my problem so noisy. I'm currently in an urgent case.

I had bought a Floating License for 5 user (IPP XE 2013 SP1 update 3). The registration screen notice that my license is expired on 27 Jun 15. But I think the expiration affects only the "Support Status" , doesn't it ? Since I do not have any demain on downloading to latest version or newer than IPP XE 2013 SP1 update 3, is it ok for me to keep building source code with just my currently version ? or I need to renew my license (i don't think so :(  )

Hi

I'm looking for IPP static libraries for Android to use with Intel ATOM platform, both x86 and x86_64 variants.

I have installed both Intel Parallel Stdio XE Cluster Edition on Win8.1 and on Ubuntu 14.04.

I cannot find the libraries for Android at all. Under the Windows installation I can find the Windows x86 and x86_64 libraries, and under Linux I can find only the Linux x86_64 variant.

What am I missing? Thank you for the help.

Regards,

Morag Agmon (Intel)

Hi,

I have recently installed Intel Integrated Performance Primitive. But I am unable to set the environment variables by running the ippvars.bat script file. I followed the instructions at this link: https://software.intel.com/en-us/node/503898. The problem is with MySQL that I have installed on my PC. whenever I try to run this code in the command prompt (as Administrator): 

ippvars.bat intel64

I get the error saying: "MySQL" is not an internal command. I tried to run the script using windows powerShell, but even it showed the same error. I cannot remove MySQL since I use it quite often.

So, I would like to know :

1) If there is any way to make the script file work. OR

2) If there is any manual method to set the environment variables.? ( I observed that at he end of the instructions link, posted above, it is mentioned as to what environment variables are set by the script file, but I don't know what exact path/location to refer to in those variable).

Hello,

Has anyone been through the process of converting C or C++ code which performed JPEG compression and decompression using IJL to using IPP? In particular, we create buffers of data to be compressed and decompressed since we have proprietary file formats. Below is a section of the code which compress a buffer. Note, this is a method in a class and some of the data is set in the class constructor.

As you will see, this code is very clean and simple unlike the IPP examples which seem to be only file based, i.e. reading headers and writing data to files, and also more complex. The documentation is also vast and broken up into small sections. So far it seems impenetrable. 

I will be most grateful for some guidance to usage documentation or an example. 

Thank you.

uint32
JpegCompressor::compress( uint8* buffer, uint32 factor,
                                           uint16 photometric_interpretation )
{
  JPEG_CORE_PROPERTIES jcprops;

  IJLERR stat;
  
  if ((stat = ijlInit(&jcprops)) != IJL_OK)
  {
    const char* err = (const char*)ijlErrorStr(stat);
    
    cerr << "IJLError: "<< err << " UNABLE TO INIT JPEG CORE PROPS\n";
    
    ijlFree(&jcprops);
    return -1;
  }
  
  jcprops.DIBWidth = width;
  jcprops.DIBHeight = height;
  jcprops.DIBChannels = bands;
  jcprops.DIBColor = ( bands == 1 ) ? IJL_G : IJL_RGB;
  jcprops.DIBPadBytes = 0;
  jcprops.DIBBytes = buffer;
  
  int buffer_size = width*height*bands;
  compressed_data = new uint8 [buffer_size];
  jcprops.JPGBytes = compressed_data;
  jcprops.JPGSizeBytes = buffer_size;
  jcprops.JPGWidth = width;
  jcprops.JPGHeight = height;
  jcprops.JPGChannels = bands;

  IJL_COLOR out_photometric = IJL_G;
  if(bands == 3) 
  {
    out_photometric = (photometric_interpretation == PHOTOMETRIC_RGB) ?
      IJL_RGB : IJL_YCBCR;
  }
  jcprops.JPGColor = out_photometric;
  jcprops.JPGSubsampling = (IJL_JPGSUBSAMPLING) IJL_NONE;
  jcprops.jquality = factor;
  
  if((stat = ijlWrite(&jcprops, IJL_JBUFF_WRITEWHOLEIMAGE)) != IJL_OK)
  {
    const char* err = (const char*)ijlErrorStr(stat);
    cerr << "IJLError: "<< err << " UNABLE TO COMPRESS JPEG IMAGE\n";

    ijlFree(&jcprops);
    return -1;
  }
  
  compressed_size = jcprops.JPGSizeBytes;
  
  ijlFree(&jcprops);  
  return compressed_size;  

}

We have lived happily with this old library but now plan to build a 64bit application and so we must upgrade to IPP. 

Hi,

I recently downloaded (purchased) IPP crypto library. Our use case is to do encryption using AES-256 in XTS mode.

Unfortunately I could not find any API for XTS mode. Though API is available for all other modes.

ippsAESEncryptCBC, ippsAESEncryptCTR, ippsAESEncryptOFB etc.

Can someone tell me is there any C API I can use for XTS mode ?

Thanks.

Hi staff,

I am using Parallel Studio 2011 and Visual Studio 2008. 

But I'd like to install more Parallel Studio XE 2016 and Visual Studio 2015.

First, I've installed Visual Studio 2015. Then, I continue to install Parallel Studio XE 2016 and get the message:

 "Intel(R) Parallel Composer 2011 Integration(s) in Microsoft Visual Studio* is already installed.
If you proceed to install, the Intel Parallel Composer 2011 IDE integration(s) will be uninstalled and the Intel(R) Parallel Studio XE 2016 Composer Edition for C++ Windows* IDE integration will be installed. You will no longer be able to select Intel Parallel Composer 2011, 12.0 or older Intel C++ Compilers from the Visual Studio IDE (if it is currently installed)."

When I read the release note of Parallel Studio XE 2016, can mention:

"You do not need to uninstall previous versions or updates before installing a newer version – the new version will coexist with the older versions"

That's why I'd like to ask " Is it possible to make two version Parallel Studio in one PC" or "Can two version of Parallel Studio be coexisted"?

Looking forward your reply,

Thank you

Hello,

We're experiencing segmentation fault randomly in ippsVLCCountBits_16s32s with IPP 6.1.6 as below:

(gdb) bt
#0  0x00007f52beeec38f in y8_ippsVLCCountBits_16s32s () from ./libDSPMP3.so
#1  0x00007f52beeabad1 in mp3enc_quantChooseTableLong () from ./libDSPMP3.so
#2  0x00007f52beeabd0f in mp3enc_quantCalcBitsLong () from ./libDSPMP3.so
#3  0x00007f52beeabfd8 in mp3enc_quantCalcBits () from ./libDSPMP3.so
#4  0x00007f52bee8bc6f in mp3encQuantization () from ./libDSPMP3.so
#5  0x00007f52bee8390b in mp3encGetFrame () from ./libDSPMP3.so
#6  0x00007f52bee81c6a in UMC::MP3Encoder::GetFrame () from ./libDSPMP3.so

Here are the media data parameters passed to :GetFrame ():

(gdb) p InMediaData
$17 = {
  <> = {<No data fields>},
  members of UMC::AudioData:
  m_info = {
    <UMC::StreamInfo> = {<No data fields>},
    members of UMC::sAudioStreamInfo:
    channels = 2,
    sample_frequency = 32000,
    bitrate = 0,
    bitPerSample = 16,
    duration = 0,
    stream_type = UMC::PCM_AUDIO,
    stream_subtype = UMC::UNDEF_AUDIO_SUBTYPE,
    channel_mask = 0,
    streamPID = 0,
    is_protected = false,
    header = 0
  }
}
(gdb) p OutMediaData
$18 = {
  <> = {<No data fields>},
  members of UMC::AudioData:
  m_info = {
    <UMC::StreamInfo> = {<No data fields>},
    members of UMC::sAudioStreamInfo:
    channels = 0,
    sample_frequency = 0,
    bitrate = 0,
    bitPerSample = 0,
    duration = 0,
    stream_type = UMC::UNDEF_AUDIO,
    stream_subtype = UMC::UNDEF_AUDIO_SUBTYPE,
    channel_mask = 0,
    streamPID = 0,
    is_protected = false,
    header = 0
  }
}

What could cause the seg fault?

Thanks,

Shifeng

I tried
Ipp64u before,after;
int count;
for(count=0;count&lt;10;count++)

{

before=IppGetCpuClocks();

after=IppGetCpuClocks();

printf("Total no. of clocks used in between : %f",(double)(after-before));

}

though there is nothing in between before and after still the values I am getting is not stable it returns different values . Why is it so ?
I want to measure value of no. of clocks used for xyz function using

Ipp64u before,after;

before=IppGetCpuClocks();

xyz();

after=IppGetCpuClocks();

printf("Total no. of clocks used in between : %f",(double)(after-before));

So How to get it ?

Hi

I want to resize an image in a particular way that duplicates a pixel vertically and horizontally. For example

source image                       destination image

    1   2                                    1   1   2   2

    3   4                                    1   1   2   2

                                              3   3   4   4

                                              3   3   4   4

Is there an IPP function that can do this?

Thanks

Hi,

Using 7.0.x I'm able to use the px/mx generic versions no problems in Kernel mode for both Linux/Windows.   I'm thinking of moving up to 7.1.x (don't want to break too much stuff) and there is no px/mx available.   I recall reading somewhere you can't use SSE2 or anything that does task switching  in Linux kernel and for Windows x64 kernel the MMX registers.   So the question is, is that true and if so does that mean kernel modes stuck at staying with px/mx versions from 7.0.x ?   Would be using for encryption and/or compression.  No floating point.

Right now it works nice in apps too, users can use ipp generic mode old hardware or use the newer support (all built using static libraries, simply running some macros and a couple passes to the IPPAPI macro to prepend px## or mx## along with macros which choose what to call).

TIA!!