pti-gpu
pti-gpu copied to clipboard
intel
Profiling Tools Interfaces for GPU (PTI for GPU) is a set of Getting Started Documentation and Tools Library to start performance analysis on Intel(R) Processor Graphics easily
Is there a way for a use code to add self logging to the one trace output? For example, I could wrap my MPI calls or certain code sections. This...
Hi, I'm planning to use onetrace as part of automation to collect profiling information about GPU offloaded scientific applications. Since onetrace is a light-weight cmd line tool, it is quite...
Not really a bug in PTI (because it produce correct results) maybe more a bug in the implementation in `icpx`. You seem to be the POC for this kind of...
Hi Anton, @Kerilk and I are also developing a L0 tracer (https://github.com/argonne-lcf/THAPI). Recently we found that we don't handle the use case when a user resets an event with `zeCommandListAppendEventReset`....
Is there an API that can be used to measure realtime vram consumption and processes using Arc dGPU. If there can you please give an example on how to do...
1. A simple application that creates a queue with the following line runs fine both without and with onetrace: ``` cl_command_queue queue = clCreateCommandQueue(context, deviceID, 0, &status); ``` However if...
Hello, Would it be possible to add onetrace formatting options so that all subsections of output are separate files (and remain comma delimited)? This would ease automatic parsing of output....
SYCL graph is an experimental feature released in oneapi 2024.0, and it seems that unitrace or onetrace doesn't support tracing of sycl graph kernels. In the following image, the device...
I am trying to build the unitrace tool on PVC. One PVC succeeded while another failed. I followed https://github.com/intel/pti-gpu/tree/master/tools/unitrace. After the terminal runs `cmake -DCMAKE_BUILD_TYPE=Release -DBUILD_WITH_MPI=0 ..`, it shows `Looking...
