This discussion has been locked.

You can no longer post new replies to this discussion. If you have a question you can start a new discussion

clEnqueueMap takes long time giving no performance benefit of using CL_MEM_ALLOC_HOST_PTR

Hello everyone,

Based on OpenCL guide for Mali 600 gpu, CL_MEM_ALLOC_HOST_PTR should be used to remove any data copy and to improve performance.

Today, I was testing on memory copy time using CL_MEM_ALLOC_HOST_PTR on Arndale board having a Mali 604 gpu. I tested with CL_MEM_ALLOC_HOST_PTR and with clEnqueueWriteBuffer. I found that overall I do not get much performance improvement if I use CL_MEM_ALLOC_HOST_PTR. Because the clEnqueueMap function takes almost same time as clEnqueueWriteBuffer.

This test has been done on vector addition.

How I tested:

Instead of having a pointer created with malloc and transfer data to device, I created a buffer at first using CL_MEM_ALLOC_HOST_PTR . Then I mapped this buffer using OpenCL API. This returns a pointer and I filled the memory pointed by this pointer with data. The mapping mechanism in this case takes time. The mapping time is almost equal to clENqueuewritebuffer. So, from this example, I did not get any significant improvement using CL_MEM_ALLOC_HOST_PTR.

My question is, why mapping time is so big when I use CL_MEM_ALLOC_HOST_PTR?

Here is the performance measurements:

Element size: 10000000, Kernel : vector addition, All times are in microseconds

Normal read write buffer	time
Kernel QUEUE -> SUBMIT	76
Kernel SUBMIT -> START	2985
Kernel START -> END	34549
Kernel QUEUE -> END	37611
clReleaseMemObject	61169
buffer creation time	20
enqueue write buffer time	108019

CL_MEM_ALLOC_HOST_PTR-with direct data copying inside allocated buffer	time
Kernel QUEUE -> SUBMIT	80
kernel SUBMIT -> START	2904
kernel START -> END	34175
kernel QUEUE -> END	37161
clReleaseMemObject	51675
Filling the pointer returned by clEnqueueMap with data	208009
mapping time	81346
unmapping time	269

CL_MEM_ALLOC_HOST_PTR-with data copying from a malloc pointer to host alloc pointer using memcpy	Time
Kernel QUEUE -> SUBMIT	88
Kernel SUBMIT -> START	3142
Kernel START -> END	33950
Kernel QUEUE -> END	37181
clReleaseMemObject	56681
mapping time	64134
unmapping time	190
memcpy time (copy data from already created malloc pointer to host allocated pinned pointer)	56987

I have also attached the three version of vector addition codes with this post for your kind review.

6205.zip

Top replies

Anthony Barbier over 11 years ago in reply to Md Mainul Hassan +1 verified

Hi, 1. It is still the wrong way of doing things: in a real life appication you would allocate your buffers at initialisation time and then re-use them, you wouldn't free and allocate new ones at every...