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
Creating this pull request - so we can discuss CallBack API here. Anyone, the feedback is very welcome! After we more or less agree on the API - I will...
When building with CMake build directory passed to command line, ``` cmake -Bbuild -H. -DCMAKE_BUILD_TYPE=Release ``` The get_itt.py fails to build, ``` Traceback (most recent call last): File "/tmp/pti-gpu/tools/unitrace/scripts/get_itt.py", line...
I am getting following error Build started at 12:04... 1>------ Build started: Project: ze_gen_headers, Configuration: Release x64 ------ 2>------ Build started: Project: level_zero_lib, Configuration: Release x64 ------ 3>------ Build started:...
## Reproduce - OS: ubuntu 24.04 - kernel: 6.13.4 - compute runtime: https://github.com/intel/compute-runtime/releases/tag/25.05.32567.17 - dGPU: B580 ```bash dpkg -l | grep intel ii intel-gsc 0.9.5-112~u24.04 amd64 Intel(R) Graphics System Controller...
I have tried to use unitrace to analyze my pytorch program with xpu device. Here is my simple program: ```python import torch from torch.profiler import profile, ProfilerActivity a = torch.randn(3).xpu()...
PTI plans to implement Callback functionality. Some work related to API IDs is already in progress. Please, provide your comments. # PTI Callback API ## Overview PTI Callback API allows...
In the README the metrics for a csv are queried here: https://github.com/intel/pti-gpu/blob/master/tools/unitrace/README.md#list-contents-of-the-metric-data-file Two of the metrics listed for that file are `L3_BYTE_READ[bytes]` and `L3_BYTE_WRITE[bytes]`. These are metrics I would like...
Recently L0 added Mutable Command List Extension that will be a part of a SYCL Graph for AI workloads. L0 adds these functions to DDI table - can you add...
Unitrace hangs in following scenarios for --chrome-kernel-logging or --chrome-device-logging flags: 1. If cmdListDesc.flags |= ZE_COMMAND_LIST_FLAG_IN_ORDER is set 2. If zeCommandListAppendWriteGlobalTimestamp is used, Ways to reproduce: Run the test on compute...
### Description This is an initial implementation of PTI Synchronous Callback API. Current implementation serves to implement on top of it PTIMetricsScope API (will be added soon) - to collect...