Summary
Jason Turner explores how the 'inline' keyword affects compiler optimization decisions, using Clang's optimization viewer to demonstrate its impact. He shows that while 'inline' doesn't directly force inlining, it can adjust the compiler's internal thresholds for deciding when to inline functions. Jason emphasizes that 'inline' should primarily be used to solve linking issues (multiple definitions across translation units), not as a performance optimization tool, as compilers already make sophisticated cost/benefit calculations for inlining.