We read every piece of feedback, and take your input very seriously.
To see all available qualifiers, see our documentation.
Updated General Guide of AMD Triton Performance Optimization (markdown)
Update section on occupancy
Clean up software pipelining section. Add appendix.
Update software pipelining guidelines
Clarify that tt.trans for the given example does not lower to multiple LDS reads and writes.
Add note to try avoiding doing transposes in the kernel
Created General Guide of AMD Triton Performance Optimization (markdown)
Initial Home page