Intel® Integrated Performance Primitives

ippsMalloc_...(int len)

what is the parameter 'len' referring to in ippsMalloc_XYZ(int len) function?

does it refer to "number of bytes", or "number" of elements"?

using examples on the reference:  https://software.intel.com/en-us/ipp-9.0-ipps-manual-pdf

page 35:

     Ipp8u* pBuf = ippsMalloc_8u(8*sizeof(Ipp8u));

seems referring to "number of bytes"

but the example on page 168:  

Transpose error

I use ipp library in my codes(FFTwd, MulC, ie.) and they work right. but I couldnt use ipp transpose functions. I added ipp library and path. I wrote this code from the ipp documents:
IppStatus transpose_m_32f(void) {
/* Source matrix with width=4 and height=3 */
Ipp32f pSrc[3*4] = { 1, 2, 3, 4,
5, 6, 7, 8,
9, 0, 1, 2 };

Alpha composite onto destination without alpha

I have two cases where I need to do fast composition of images and I can't find a suitable method in IPP.

In both cases the destination image doesn't have an alpha channel, but I have alpha information for the image I'm trying to composite on top. I don't want to create an alpha channel in the destination image just for this operation because the ROI might be very small.

So my use cases are:

problems with ippiMalloc_8u_C4()

I am having all sorts of pain chasing down a bug that crashes my app when exiting, that is, when I call all of the destructors and exit. Using gdb, I have noticed something that seems strange to me.   When I first call ippiMalloc_8u_C4() I get a very large/high pointer, but subsequent calls return much lower memory.  For example, my log will look like this:

Dec 08 16:04:29.038  ***: ippiMalloc() returning ptr: 0x7f6a09347040 
Dec 08 16:04:29.053  ***: ippiMalloc() returning ptr: 0x2e8f9c0 
Dec 08 16:04:29.057  ***: ippiMalloc() returning ptr: 0x269ed80 

How can I try the MIMO MMSE function?

It is said it is integreated in the Intel® System Studio suite. I have downloaded the 2016 Intel® System Studio suite. But I still can't find the related function?

As it is said that MMSE is included in the ippe.h. But I can't find ippe.h in the IPP library. I want to verfy its speed. How can I get it and verified?

Thanks,

zuri

the result image data is always flip vertical

Hi I am using Intel IPL library and ImageMagicK Library.
When I firstly get image date from ImageMagicK I put the data to IPL.
Then IPL can get image data.

When I display what I got a image data form IPL. I found the result image data is always flip verticalized. you can see as following the problem.

original image

and then I've got a image as following.

Switching ippSetNumThreads in threads

I have an application that has several threads. We are using IPP 7.0.

I want certain functions to run IPP in multi threaded model and some in single threaded model

So I switch between ippSetNumThreads(1) to ippSetNumThreads(N) where N is the number of cores in the machine.

Please answer these questions

1) Thread1 sets ippSetNumThreads(1)  and makes a call. Is that guaranteed to run with only one thread ? What if there was a context switch and the other thread also made a call to ippSetNumThreads(N) ?

ippSetNumThreads fails to create threads

We are using IPP 7.0.

I have a small console application in which I am calling the FFT function from here

https://software.intel.com/en-us/articles/how-to-use-intel-ipp-s-1d-four...

When i call ippSetNumThreads(10); and get the number of threads using ppGetNumThreads(&origTh); its returning 10.

However when my FFT is running if i monitor number of threads in task manager i only see one thread.

Dont understand why. Please help.

Shashi

S’abonner à Intel® Integrated Performance Primitives