RE: Can't compile HaskellDirect with GHC
George Russell [EMAIL PROTECTED] writes: .. ../../ghc/driver/ghc -fglasgow-exts -Rghc-timing -H16m -W -recomp-O -O2-for-C -H30m -c Utils.lhs -o Utils.o -osuf o ghc: ignoring heap-size-setting option (-H16m)...not the largest seen Utils.lhs:22: parse error on input `(#' Compilation had errors "s|(#|( #|g" - the parser thinks it's the start of an unboxed tuple, which is not the case here. --sigbjorn
Can't compile HaskellDirect with GHC
The latest GHC chokes on the latest HaskellDirect file src/Utils.lhs.
Here is a copy of error message and attached is the src/Utils.lhs file.
It will be seen from the file that ghc is objecting to an explicit
universal quantification. What should I do?
===fptools== Recursively making `all' in src lib examples doc ...
PWD = /export/ger/fptools/hdirect
==fptools== gmake all -r;
in /export/ger/fptools/hdirect/src
../../ghc/driver/ghc -syslib exts -Rghc-timing -H16m -W -recomp-O -O2-for-C -H30m
-c Utils.lhs -o Utils.o -osuf o
ghc: ignoring heap-size-setting option (-H16m)...not the largest seen
Utils.lhs:345: parse error on input `.'
Compilation had errors
%
% @(#) $Docid: May. 24th 1999 17:42 Sigbjorn Finne $
% @(#) $Contactid: [EMAIL PROTECTED] $
%
\begin{code}
module Utils
( showOct
, showHex
, mapFromMb
, mapMb
, mapMbM
, concMaybe
, split
, prefix
, traceIf
, elemBy
, mapUnzip
, diff
, deEscapeString
, (#)
--,UNUSED: catMapMaybes
, dropSuffix
-- re-exported
, trace
, tryOpen
, basename
, prefixDir
, hdirect_root
, bailIf
, decons
, safe_init
, mapAccumLM
, readTaggedFields
, read_int
, read_string
, read_bool
, read_qid
, FieldInfo(..)
) where
--import NumExts
import Char (chr, ord, readLitChar, isSpace, isAlphaNum)
import IOExts
import IO
{- BEGIN_GHC_ONLY
import Directory
END_GHC_ONLY -}
import Monad ( when )
import List ( mapAccumL, nub, isPrefixOf )
infixl 1 #
\end{code}
A convenience operator for invoking methods on objects:
\begin{code}
obj # meth = meth obj
\end{code}
Until NumExts is commonly available, we define the following show functions here:
\begin{code}
showIntAtBase :: Integral a = a - (a - Char) - a - ShowS
showIntAtBase base toChr n r
| n 0 = '-':showIntAtBase 10 toChr (negate n) r
| otherwise =
case quotRem n base of { (n', d) -
case toChr dof { ch -
let
r' = ch : r
in
if n' == 0 then r' else showIntAtBase base toChr n' r'
}}
showHex :: Integral a = a - ShowS
showHex n r =
showString "0x" $
showIntAtBase 16 (toChrHex) n r
where
toChrHex d
| d 10= chr (ord_0 + fromIntegral d)
| otherwise = chr (ord 'a' + fromIntegral (d - 10))
showOct :: Integral a = a - ShowS
showOct n r =
showString "0o" $
showIntAtBase 8 (toChrOct) n r
where toChrOct d = chr (ord_0 + fromIntegral d)
ord_0 :: Num a = a
ord_0 = fromInt (ord '0')
\end{code}
Mapping from a Maybe:
\begin{code}
mapFromMb :: b - (a - b) - Maybe a - b
mapFromMb d f mb = case mb of Nothing - d ; Just v - f v
\end{code}
\begin{code}
split :: Eq a = a - [a] - [[a]]
split a [] = []
split a as =
case break (==a) as of
(xs,[]) - [xs]
(xs,_:ys) - xs:split a ys
\end{code}
\begin{code}
prefix :: Eq a = [a] - [a] - Maybe [a] -- what's left
prefix [] ls = Just ls
prefix ls [] = Nothing
prefix (x:xs) (y:ys)
| x == y= prefix xs ys
| otherwise = Nothing
\end{code}
\begin{code}
traceIf :: Bool - String - a - a
traceIf True str v = trace str v
traceIf _ _ v = v
elemBy :: (a - Bool) - [a] - Bool
elemBy isEqual [] = False
elemBy isEqual (y:ys) = isEqual y || elemBy isEqual ys
mapUnzip :: (a - (b,c)) - [a] - ([b],[c])
mapUnzip f [] = ([],[])
mapUnzip f (x:xs) =
let
(a, b) = f x
(as,bs) = mapUnzip f xs
in
(a:as,b:bs)
\end{code}
Returns list of deltas, i.e,
@
diff [x0,x1..xp,xn] = [x0, x1-x0, .., xp - xn]
@
\begin{code}
diff :: Num a = [a] - [a]
diff ls = snd (mapAccumL ( \ acc v - (v, v - acc)) 0 ls)
\end{code}
\begin{code}
catMapMaybes :: (a - b) - [Maybe a] - [b]
catMapMaybes f ls = [f x | Just x - ls]
\end{code}
Dropping the extension off of a filename:
\begin{code}
dropSuffix :: String - String
dropSuffix str =
case dropWhile (\ch - ch /= '.' ch /= '/' ch /= '\\' )
(reverse str) of
('.':rs) - reverse rs
_- str
-- give up if we reach a separator (/ or \) or end of list.
dropPrefix :: Eq a = [a] - [a] - [a]
dropPrefix [] ys = ys
dropPrefix xs [] = []
dropPrefix (x:xs) (y:ys)
| x == y = dropPrefix xs ys
| otherwise= y:ys
\end{code}
Slightly generalised version of code found in Green Card's front end:
\begin{code}
tryOpen :: Bool
- [FilePath]
- [String]
- FilePath
- IO (Maybe FilePath)
tryOpen verbose path exts name =
doUntil (mbOpenFile verbose) (allFileNames path name exts)
RE: compiling CVS of June 99
-- - ghc -cpp -fglasgow-exts -Rghc-timing -I. -IcodeGen -InativeGen -Iparser -iutils:basicTypes:types:hsSyn:prelude:rename:typecheck:deSugar :coreSyn:specialise:simplCore:stranal:stgSyn:simplStg:codeGen :absCSyn:main:profiling:parser:usageSP:cprAnalysis:nativeGen -recomp -optCrts-G3 -optCrts-F1.5 -optCrts-M112m -Onot -H45m -dcore-lint -fno-warn-incomplete-patterns -dcore-lint -c parser/Parser.hs -o parser/Parser.o -osuf o ghc: 2896826300 bytes, 2495 GCs, 20862924/58736116 avg/max bytes residency (25 samples), 96M in use, 0.02 INIT (0.00 elapsed), 93.62 MUT (1541.24 elapsed), 93.58 GC (3792.84 elapsed) :ghc ghc: module version unchanged at 1 make[2]: *** [parser/Parser.o] Error 1 make[1]: *** [all] Error 1 make: *** [all] Error 1 - What might this mean? It means your machine is thrashing, for a start. Buy some more memory ;-) Seriously, it looks like something crashed. The fact that you got the stats from the compiler means it probably wasn't the compiler, so it could have been perl. Try compiling again with EXTRA_HC_OPTS=-v, to see exactly where the compilation got to. I'm a bit concerned that the mem in use figure is 96M, and yet the maximum residency is reported as 58M - I'd expect the in use figure to be at least twice the maximum residency. Very strange. Cheers, Simon
RE: Can't compile HaskellDirect with GHC
Untested, but changing the occurrence of '-syslib exts' to '-fglasgow-exts' in the first setting of SRC_HC_OPTS in src/Makefile may just take you past this one.. --sigbjorn George Russell [EMAIL PROTECTED] writes: The latest GHC chokes on the latest HaskellDirect file src/Utils.lhs. Here is a copy of error message and attached is the src/Utils.lhs file. It will be seen from the file that ghc is objecting to an explicit universal quantification. What should I do? -- -- ===fptools== Recursively making `all' in src lib examples doc ... PWD = /export/ger/fptools/hdirect -- -- -- -- ==fptools== gmake all -r; in /export/ger/fptools/hdirect/src -- -- ../../ghc/driver/ghc -syslib exts -Rghc-timing -H16m -W -recomp-O -O2-for-C -H30m -c Utils.lhs -o Utils.o -osuf o ghc: ignoring heap-size-setting option (-H16m)...not the largest seen Utils.lhs:345: parse error on input `.' Compilation had errors
Re: Can't compile HaskellDirect with GHC
George Russell wrote: "Sigbjorn Finne (Intl Vendor)" wrote: Untested, but changing the occurrence of '-syslib exts' to '-fglasgow-exts' in the first setting of SRC_HC_OPTS in src/Makefile may just take you past this one.. In that case I get another error message: ===fptools== Recursively making `all' in src lib examples doc ... PWD = /export/ger/fptools/hdirect ==fptools== gmake all -r; in /export/ger/fptools/hdirect/src ../../ghc/driver/ghc -fglasgow-exts -Rghc-timing -H16m -W -recomp-O -O2-for-C -H30m -c Utils.lhs -o Utils.o -osuf o ghc: ignoring heap-size-setting option (-H16m)...not the largest seen Utils.lhs:22: parse error on input `(#' Compilation had errors . . . but if (as well as Sigborn's change to the Makefile) I replace "(#)" by "( # )" on line 22 of Utils.lhs, it goes through!
Re: Can't compile HaskellDirect with GHC
The struggle continues . . . after the hacks suggested by Sigborn and myself, Utils went through, as did a lot of other HaskellDirect files, but now I get the following messages. So we have a bug in GHC itself, with a non-exhaustive match in ghc/compiler/typecheck/TcMatches.lhs. The version of GHC is up to date (as of last night) from the CVS sources. ===fptools== Recursively making `all' in src lib examples doc ... PWD = /export/ger/fptools/hdirect ==fptools== gmake all -r; in /export/ger/fptools/hdirect/src ../../ghc/driver/ghc -fglasgow-exts -Rghc-timing -H16m -W -recomp-O -O2-for-C -H30m -c CodeGen.lhs -o CodeGen.o -osuf o ghc: ignoring heap-size-setting option (-H16m)...not the largest seen TcMatches.lhs:274: Non-exhaustive patterns in function tcStmts gmake[1]: *** [CodeGen.o] Error 1 gmake: *** [all] Error 1
Re: Can't compile HaskellDirect with GHC
"Sigbjorn Finne (Intl Vendor)" wrote: Untested, but changing the occurrence of '-syslib exts' to '-fglasgow-exts' in the first setting of SRC_HC_OPTS in src/Makefile may just take you past this one.. In that case I get another error message: ===fptools== Recursively making `all' in src lib examples doc ... PWD = /export/ger/fptools/hdirect ==fptools== gmake all -r; in /export/ger/fptools/hdirect/src ../../ghc/driver/ghc -fglasgow-exts -Rghc-timing -H16m -W -recomp-O -O2-for-C -H30m -c Utils.lhs -o Utils.o -osuf o ghc: ignoring heap-size-setting option (-H16m)...not the largest seen Utils.lhs:22: parse error on input `(#' Compilation had errors
Re: Integer - Int conversion
Sigbjorn: : The question is what should 0x8000 :: Integer : become when its coerced into an int. : : GHC takes the lowest 32 bits. : Advantage: Equality above maxBound still "works" aka above. : Disadvantage: 0x + 1 == 0 ! : : Perhaps constant Int's larger than maxBound (or smaller : than minBound) should be a compile time error? : : : Int arithmetic is modulo maxBound (at least that's what : ghc and Hugs implements), so making the Integer-Int conversion : fall into line with that, has some merit. On my machine the following program: main = do putStr "According to ghc-4.02 (1-maxBound)^2 = " putStr $ show mult1 putStr "\nAccording to ghc-4.02 (maxBound-1)^2 = " putStr $ show mult2 where mult1, mult2 :: Int mult1 = (1-maxBound)*(1-maxBound) mult2 = (maxBound-1)*(maxBound-1) outputs: According to ghc-4.02 (1-maxBound)^2 = 4 According to ghc-4.02 (maxBound-1)^2 = 4 Mod maxBound the result of both multiplication should have been 1. Maybe it is better to rely on Integers. Regards, Marc van Dongen
RE: Integer - Int conversion
On my machine the following program: main = do putStr "According to ghc-4.02 (1-maxBound)^2 = " putStr $ show mult1 putStr "\nAccording to ghc-4.02 (maxBound-1)^2 = " putStr $ show mult2 where mult1, mult2 :: Int mult1 = (1-maxBound)*(1-maxBound) mult2 = (maxBound-1)*(maxBound-1) outputs: According to ghc-4.02 (1-maxBound)^2 = 4 According to ghc-4.02 (maxBound-1)^2 = 4 Mod maxBound the result of both multiplication should have been 1. Int arithmetic on GHC and Hugs is modulo 2^32, i.e. maxBound + 1. Cheers, Simon
GHC panic when building prelude
When building the GHC prelude (latest sources again), the following happened: rm -f Weak.o ; if [ ! -d Weak ]; then mkdir Weak; else find Weak -name '*.o' -print | xargs rm -f __rm_food ; fi ; ../../../ghc/driver/ghc -recomp -cpp -fglasgow-exts -fvia-C -Rghc-timing -O -split-objs -odir Weak -H20m -static-O -O2-for-C -H30m -Onot -c Weak.lhs -o Weak.o -osuf o ghc: ignoring heap-size-setting option (-H20m)...not the largest seen panic! (the `impossible' happened): cgEvalAlts: dodgy case of unboxed tuple type Please report it as a compiler bug to [EMAIL PROTECTED]
Re: Can't compile HaskellDirect with GHC
George Russell wrote: [...] So we have a bug in GHC itself, with a non-exhaustive match in ghc/compiler/typecheck/TcMatches.lhs. The version of GHC is up to date (as of last night) from the CVS sources. [...] The same happens during the compilation of Green Card: [...] ghc -fvia-C -fglasgow-exts -recomp -Rghc-timing -O -H10m -c Lex.lhs -o Lex.o -osuf o TcMatches.lhs:274: Non-exhaustive patterns in function tcStmts Cheers, Sven -- Sven PanneTel.: +49/89/2178-2235 LMU, Institut fuer Informatik FAX : +49/89/2178-2211 LFE Programmier- und Modellierungssprachen Oettingenstr. 67 mailto:[EMAIL PROTECTED]D-80538 Muenchen http://www.informatik.uni-muenchen.de/~Sven.Panne
Re: Integer - Int conversion
Lennart Augustsson wrote: Simon Marlow wrote: Int arithmetic on GHC and Hugs is modulo 2^32, i.e. maxBound + 1. Actually, if we are in nit-picking mode, arithmetic is modulo (maxBound+1)*2. PS. I.e., if we interpret "modulo" in the right way since the result is interpreted as a signed number. Int arithmetic in all Haskell implementations I know of operate in the same way as C usually does. It has the advantage that you have the ring properties of + and *, which arithmetic with exceptions would not give you. -- Lennart
