No, it does not mean that we can refrain from setting `DYLD_LIBRARY_PATH`. in f7be53ac9dac85b83e7fe5ecede01b98a572ba48, the rpaths are made relative to the final install location. During testing,
#8721: Testsuite not reporting errors for DYN way on OS X -------------------------------+---------------------------------------- Reporter: darchon | Owner: Type: bug | Status: new Priority: normal | Milestone: Component: Test Suite | Version: 7.6.3 Resolution: | Keywords: dynamic linking Operating System: MacOS X | Architecture: Unknown/Multiple Type of failure: Other | Test Case: Blocked By: | Blocking: Related Tickets: | Differential Rev(s): Wiki Page: | -------------------------------+---------------------------------------- Changes (by thomie): * status: closed => new * resolution: fixed => Comment: Replying to [comment:3 darchon]: the libraries are still in their build locations, so we still need `DYLD_LIBRARY_PATH` to find the libraries. We can do better. A normal `validate` //installs// the in-tree GHC into a directory `bindisttest/`, and then starts a testsuite run with `BINDIST=YES`. The testsuite driver then uses this //installed// GHC. Not setting `DYLD_LIBRARY_PATH` when `BINDIST=YES` should give us better coverage for DYN way on OS X, as this ticket requested. I don't have access to a Mac, but reopening for someone who does. I still wonder why on Linux we don't have to set `LD_LIBRARY_PATH`. -- Ticket URL: http://ghc.haskell.org/trac/ghc/ticket/8721#comment:5 GHC http://www.haskell.org/ghc/ The Glasgow Haskell Compiler