I'm all for this, if Roman and/or Don care to help -- thank you!  A little 
readme to explain how to add a new test would be good.

Ian and/or Simon may want to comment on dependencies, or technology (eg Haskell 
vs Python).

Simon

| -----Original Message-----
| From: Roman Leshchinskiy [mailto:[EMAIL PROTECTED]
| Sent: 08 February 2008 00:53
| To: Simon Peyton-Jones
| Cc: [email protected]
| Subject: Simplifier tests (was Re: patch applied (testsuite): Add a new 
category of "eyeball" tests)
|
| Simon Peyton Jones wrote:
| > Thu Feb  7 08:22:44 PST 2008  [EMAIL PROTECTED]
| >   * Add a new category of "eyeball" tests
| >
| >   These tests are hard to do automatically, but they record examples that
| >   provoked changes to the optimiser.  Each one has notes that says what you
| >   should expect to see!
|
| I think we should try to do them automatically. Usually it's just a
| matter of dumping the simplifier output and grepping for things. For the
| tests you added:
|
| >     A ./tests/ghc-regress/eyeball/dead1.hs
|
| Make sure "bar" doesn't occur in the output of phase 0.
|
| >     A ./tests/ghc-regress/eyeball/inline1.hs
|
| Make sure "myIndex" doesn't occur in the output of phase 2.
|
| >     A ./tests/ghc-regress/eyeball/inline2.hs
|
| Count the number of simplifier iterations in -dshow-passes.
|
|  >     A ./tests/ghc-regress/eyeball/inline3.hs
|
| Count the number of calls to (+#).
|
| And so on. I'll try to implement a simple framework. I'm just not sure
| if it should be in Haskell, Python or shell?
|
| Roman

_______________________________________________
Cvs-ghc mailing list
[email protected]
http://www.haskell.org/mailman/listinfo/cvs-ghc

Reply via email to