#15426: `elemIndex` and `findIndex` still can't fuse -------------------------------------+------------------------------------- Reporter: kabuhr | Owner: kabuhr Type: bug | Status: new Priority: normal | Milestone: 8.8.1 Component: libraries/base | Version: 8.4.3 Resolution: | Keywords: Operating System: Unknown/Multiple | Architecture: | Unknown/Multiple Type of failure: None/Unknown | Test Case: | testsuite/tests/perf/should_run/T15426.hs Blocked By: | Blocking: Related Tickets: | Differential Rev(s): Wiki Page: | https://phabricator.haskell.org/D5063 -------------------------------------+------------------------------------- Comment (by kabuhr): No, these differences were for the "TotalMem" column from the "nofib- analyse" tool. I gather this corresponds to the memory "in use" in "-RTS +t" output, which is peak OS allocation and presumably sensitive to GC timing. After doing some more testing, it seems both the `cacheprof` and `maillist` differences are spurious (e.g., `maillist` either "uses" `8M` or `16M` during each run, and I just happened to have a string of good luck on the `INLINABLE` test). There were no differences in allocations for these or any of the other benchmarks. -- Ticket URL: http://ghc.haskell.org/trac/ghc/ticket/15426#comment:8 GHC http://www.haskell.org/ghc/ The Glasgow Haskell Compiler