[GHC] #15575: Investigate Haskell rewrite of testsuite driver

GHC ghc-devs at haskell.org
Sun Sep 2 22:22:53 UTC 2018


#15575: Investigate Haskell rewrite of testsuite driver
-------------------------------------+-------------------------------------
        Reporter:  tdammers          |                Owner:  (none)
            Type:  task              |               Status:  new
        Priority:  normal            |            Milestone:  Research
                                     |  needed
       Component:  Test Suite        |              Version:  8.4.3
      Resolution:                    |             Keywords:
Operating System:  Unknown/Multiple  |         Architecture:
                                     |  Unknown/Multiple
 Type of failure:  None/Unknown      |            Test Case:
      Blocked By:                    |             Blocking:
 Related Tickets:  #15363            |  Differential Rev(s):
       Wiki Page:                    |
-------------------------------------+-------------------------------------

Comment (by lantti):

 I haven't yet given a proper look to exactly what we would need for the
 testsuite as whole, but from the actual test cases most (6033/6618) are
 using the test language and canned functionality provided by the testsuite
 driver and should be straight-forward to migrate automatically (for
 example by modifing the existing test language interpreter), some
 (577/6618) use their own makefiles to do various tasks that might need
 manual work or elevated levels of smartness to migrate. Few (8/6618) run
 plain bash commands that probably will be migrated manually just because
 writing the automation would take longer.

 For the driver code, I can't yet at this point even start guessing which
 tools would be the most suitable and which compatibility problems they
 would have between our supported platforms. Counting the number of special
 cases and workarounds in the current driver might not be the most fruitful
 way as our set of tools and libraries would be completely different and
 have different problem spots.

 I'll still need to do some reading...

-- 
Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/15575#comment:2>
GHC <http://www.haskell.org/ghc/>
The Glasgow Haskell Compiler


More information about the ghc-tickets mailing list