Custom Query (6519 matches)
Results (19 - 21 of 6519)
|#317||fixed||Simplifier doesn't preserve bottoms sometimes||simonmar|
Here's a test due to Iavor Diatchki [email@example.com]: >> test2 :: Bool >> test2 = stops (bot ()) >> where >> stops :: Int# -> Bool >> stops _ = True ghc 6.4 gives the value True for test2.
|#318||fixed||GHCi support on x86_64||nobody||nobody|
I just installed the 64bit version of ghc on an athlon 64. ghc does work and produces correct 64bit code, but ghci fails to start. Here is the complete output: [phil@athlon64 ~]$ ghci ___ ___ _ / _ \ /\ /\/ __(_) / /_\// /_/ / / | | GHC Interactive, version 6.2.2, for Haskell 98. / /_\\/ __ / /___| | http://www.haskell.org/ghc/ \____/\/ /_/\____/|_| Type :? for help. Loading package base ... /usr/lib64/ghc-6.2.2/HSbase.o: unknown architecture ghc-6.2.2: panic! (the `impossible' happened, GHC version 6.2.2): loadObj: failed Please report it as a compiler bug to firstname.lastname@example.org, or http://sourceforge.net/projects/ghc/. I'm using Fedora Core 3 and installed ghc using yum. If you have any questions, contact me: philipp.classen[AT]gmx.net Thanks, Philipp
|#323||fixed||Exponential behaviour with type synonyms||simonpj||simonpj|
You're quite right. GHC has a simple but non- performant representation of type synonyms in types, so as to be able to generate good error messages, In particular, the type S t where S is a type synonym defined by 'type S a = s', is represented as SynNote (S t) (s [t/a]) That is, (S t) is represented by *both* its un-expanded and expanded form. The SynNote is ignored by unification, but the un- expanded form is useful for error messages. Unfortunately, t is duplicated, as you can see, and that leads to the behaviour you describe. I don't see myself fixing this soon, at least partly because I can't see an obvious way to fix it that doesn't lose error message behaviour. I'm going to open a SourceForge bug for it. If anyone has good ideas, let me know. Simon | -----Original Message----- | From: email@example.com [mailto:glasgow-haskell-bugs- | firstname.lastname@example.org] On Behalf Of Iavor Diatchki | Sent: 17 February 2005 01:27 | To: email@example.com | Subject: 'type' declarations | | hello, | ghc seems to be having trouble with 'type' declarations. | while compiling (i guess kind checking is the correct word here) | the following program for a very long time, ghc (6.2) runs out of 300Mb of heap. | | module Test where | | type S = Maybe | type S2 n = S (S n) | type S4 n = S2 (S2 n) | type S8 n = S4 (S4 n) | type S16 n = S8 (S8 n) | type S32 n = S16 (S16 n) | | type N64 n = S32 (S32 n) | | type N64' = | S ( S ( S ( S ( S ( S ( S ( S ( | S ( S ( S ( S ( S ( S ( S ( S ( | S ( S ( S ( S ( S ( S ( S ( S ( | S ( S ( S ( S ( S ( S ( S ( S ( | S ( S ( S ( S ( S ( S ( S ( S ( | S ( S ( S ( S ( S ( S ( S ( S ( | S ( S ( S ( S ( S ( S ( S ( S ( | S ( S ( S ( S ( S ( S ( S ( S ( | Int | )))))))) | )))))))) | )))))))) | )))))))) | )))))))) | )))))))) | )))))))) | )))))))) | | if i remove the N64 definition things work. i guess something | exponential is happening | (substitution?). | | -iavor
Note: See TracQuery for help on using queries.