CPU instructions not accurately profiled
Hello everyone, I've been using codeXL for a while, and it works pretty good, except for source code viewing,when most of the time I see that just one single instruction takes most of the execution...
View ArticleRe: Missing libraries from omega catalyst?
Yes. That is because there is no /usr/lib/fglrx/libGL.so, so if you try to load this first, you will get the mesa libGL. After system loads it, it caches all relevant info and sees that there is a...
View ArticleRe: Do you plan to keep AMD Core Math Library (ACML) for Linux built with...
Hi,I try to use ACML 6.1.0 on Intel CPU E5-2670. It's 2 CPU systems with 16 cores per CPU.flags : fpu vme de pse tsc msr pae mce cx8 apic mtrr pge mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2...
View Articlepower profiling in latest CodeXL 1.6.7247.0
Hi there. Recently I installed the latest CodeXL (v1.6.7247.0) and found that it can do power profiling, which is a good function for AMD APUs because APUs highlight energy efficiency. However, I have...
View ArticleRe: Most cache friendly tiling of OpenCL image ?
Hi,There can be many possible physical memory layouts for OpenCL images. To accelerate the accessing of images, run-time can rearrange the data layout to take advantage of special hardware (say texture...
View ArticleOcl debuggers
In the User's Guide there is a whole section on CodeXL, with very interesting points. There is another debugger, opencl-emu, for which I know very little. Is it any good? How does it compare to CodeXL?...
View ArticleRe: glCopyImageSubData for small compressed mips
Hi Baldur, From your description, I think you are right, glCopyImageSubData shouldn't return GL_INVALID_VALUE. But the sample code you provide is too less. Would you please give me more code (such as...
View ArticleRe: clEnqueueWriteBufferRect does not work when region width is not equal to...
Thanks. I'll try to forward your OpenGL issue to concerned team.
View ArticleRe: clEnqueueWriteBufferRect does not work when region width is not equal to...
Thanks for the confirmation.I'm not sure what is the actual reason behind this issue. I've filed an internal bug report against it. Hope this will be fixed in future release but cannot comment about...
View ArticleRe: fm2+ dual socket apu-support
http://www.asus.com/us/Notebooks_Ultrabooks/G73Jh/Aside from the fact that this is more an idea rather than a question, I have thought of the same thing (and believe I have posted it earlier, but Jive...
View ArticleRe: power profiling in latest CodeXL 1.6.7247.0
Hi acekiller,Your AMD APU contains 4 CPU Cores: Core 0, Core 1, Core 2 and Core 3. The cores are organized in pairs. Each pair is called a Compute Unit. Thus:Compute Unit 0 = Core 0 + Core 1Compute...
View ArticleRe: Can't Debug Kernel in Teapot Example
I would really appreciate it if someone could help with this problem. I have upgraded to OpenXL 1.6, but I am still having the previously mentioned problem. I have tried different GPUs with the same...
View ArticleRe: Problem when install sdk 3.0
Are you root (sudo if using Ubuntu)? Do you have space in your filesystem? On Fri, Dec 19, 2014 at 3:41 PM, davies1828 <developer.forums@amd.com>
View ArticleRe: Problem when install sdk 3.0
Yeah I login as a root and I also have about 30G in my disk
View ArticleRe: NEW - GPU PerfStudio 3.1 has been released
Hi, A new version of GPU PerfStudio (3.1.323) has been posted to the usual place (http://developer.amd.com/tools-and-sdks/graphics-development/gpu-perfstudio/), which addresses the Shader Analyzer...
View ArticleRe: Re: Reuse of Kernel in aparapi / memoryleak ?
Hi, I have found a workaround:aparapi seams to have troubles with 2 dim arrays. So if I convert the kernel to a 1 dim array it works with explizit. Both methods are affected the get(int[][] array) and...
View ArticleRe: Most cache friendly tiling of OpenCL image ?
dipak, Didn't know the hw could possibly remap the image to enable better cache coherency, good info. Sorry for hijacking your thread boxerab.
View ArticleRe: Few problems with new release 1.5.6571.0
Hi,I have now reproduced the VS hang you reported. It will take a few days to investigate this properly. I'll keep you posted.
View ArticleRe: NEW - GPU PerfStudio 3.1 has been released
Thank you, now I can compile one shader for multiple ISAs in a single click without issues
View ArticleRe: Execution issues of CodeXL 1.6 on 64bit Linux
With respect to your original question, CodeXL 1.6 seeks the path dri/libGL.so.1 in x64 systems. I think a better workaround in terms of architecture, is to create /usr/lib/fglrx/dri and link there the...
View Article