This is a metabug to group together a bunch of related issues do with our current generation of LLVM IR.
The IR we produce is very verbose, this causes an inordinate amount of time to be spent during the optimization passes that we could trivially prevent by simply not outputting these sequences.
Instrumentation and testing issues
Too many copies of functions issues
(see http://people.mozilla.org/~graydon/symbols-by-name.txt)
Too much codegen issues
Incorrect or LLVM-frustrating codegen issues
Overall use of LLVM issues
This is a metabug to group together a bunch of related issues do with our current generation of LLVM IR.
The IR we produce is very verbose, this causes an inordinate amount of time to be spent during the optimization passes that we could trivially prevent by simply not outputting these sequences.
Instrumentation and testing issues
rustc: improve -Z trans-stats to report per-fn LLVM instruction counts and translation timing #7456 : improved trans statsLLVM codegen-equivalence tests #7461 : LLVM codegen equivalence testsToo many copies of functions issues
(see http://people.mozilla.org/~graydon/symbols-by-name.txt)
rewrite type_use #8651 : rewrite type_usenormalize_for_monomorphization could normalize recursively #2529 : normalize for monomorphization recursivelyIn trans::type_use::mark_for_expr, don't count fields after the chosen field #2537 : type_use should only consider fields before the selected fieldToo much codegen issues
Incorrect or LLVM-frustrating codegen issues
Overall use of LLVM issues
Discover better llvm optimizations for Rust #3519 : better LLVM optimizations