'What the `ipa` pipeline is about in CUDA architecture?
When looking into ncu --query-metrics it turns out that several counters are about this ipa pipeline that isn't even cited in NSight docs, smsp__inst_executed_pipe_ipa for example. While for all of the other pipelines a proper explanation is provided, for ipa I wasn't able to find any reference at all.
Solution 1:[1]
IPA is the Interpolate Attribute pipeline used in pixel/fragment shaders to interpolate a varying attribute over a quad (4 threads). This pipeline is not accessible to a compute shader.
Sources
This article follows the attribution requirements of Stack Overflow and is licensed under CC BY-SA 3.0.
Source: Stack Overflow
| Solution | Source |
|---|---|
| Solution 1 | Greg Smith |