Re: [GHC] #4213: LLVM: Add support for TNTC to LLVM compiler suite

#4213: LLVM: Add support for TNTC to LLVM compiler suite -------------------------------------+------------------------------------ Reporter: dterei | Owner: dterei Type: feature request | Status: new Priority: low | Milestone: 7.6.2 Component: Compiler (LLVM) | Version: 6.13 Resolution: | Keywords: Operating System: Unknown/Multiple | Architecture: Unknown/Multiple Type of failure: None/Unknown | Difficulty: Unknown Test Case: | Blocked By: Blocking: | Related Tickets: -------------------------------------+------------------------------------ Comment (by altaic): Is there any reason to not propose to change LLVM's prefix semantics such that the function's symbol points to the function body rather than the prefix data? To keep the optimization passes happy, it seems like one could have symbols for both the prefix data and the function body. The prefix data seems to be stuffed in a return instruction, so I'd think with both symbols, dead code elimination and other optimization passes wouldn't mess with the prefix data. I see two advantages to this approach: • no extra unconditional branch • prefix would no longer require the front end to construct target specific instructions -- Ticket URL: http://ghc.haskell.org/trac/ghc/ticket/4213#comment:21 GHC http://www.haskell.org/ghc/ The Glasgow Haskell Compiler
participants (1)
-
GHC