liveBook · Manning

PTX

As explained in GPU Programming with Triton

PTX is NVIDIA's low-level virtual instruction-set representation for GPU programs.

PTX is NVIDIA’s textual virtual instruction set for GPUs. In Triton’s compilation pipeline, LLVM IR is translated into PTX, which NVIDIA’s assembler turns into a CUBIN binary for execution. PTX offers low-level control but is difficult to work with; writing it directly is rarely necessary. section 1.2.1 section 2.3.1

Figure 2.10 The Triton compiler pipeline: from user code through AST, TTIR, and TTGIR to GPU execution via LLVM-IR, PTX, and CUBIN.
CH02 F10 Fartale

Trace PTX’s place in the Triton compiler pipeline, from user code through intermediate representations to GPU execution. section 2.3.1

How GPU Programming with Triton covers it

How deep it goes
Defines PTX as textual GPU instructions and traces its compiler role to CUBIN, without teaching PTX programming.
Examples
Uses Triton’s compiler pipeline from LLVM IR through PTX to CUBIN as its concrete artifact.
What you can do after
Understand how PTX fits between LLVM IR and GPU binary generation in Triton’s compilation pipeline.
How it connects
Links PTX to LLVM IR, CUBIN, CUDA, SASS, and Triton’s compiler pipeline.

Also explained in

Compare how 2 books explain PTX →

Related concepts

This explainer was generated by Manning's knowledge graph from the text of GPU Programming with Triton. The section links point to the parts of the book it draws on.