- Glasgow-haskell-users

Pattern matching with implicit par. bug
by Jorge Adriano 13 Feb '03

13 Feb '03

There it goes, J.A. type CTPar = ([Double],Int,Int) us :: (?ctPar :: CTPar) => [Double] us = let (us',_,_) = ?ctPar in us' ghc-5.04: panic! (the `impossible' happened, GHC version 5.04): tcSplitTyConApp {?ctPar{-r2kd-} :: ([GHCziFloat.Double{-3u-}], GHCziBase.Int{-3J-}, GHCziBase.Int{-3J-})} Please report it as a compiler bug to glasgow-haskell-bugs(a)haskell.org, or http://sourceforge.net/projects/ghc/.

1 0

GHCi-5.04.2: Windows Open with ... problem.
by Calle Lejdfors 13 Feb '03

13 Feb '03

Hello. I am having a problem using the Open with ... on haskell-files under WinXP. The problem is this: create a file on my desktop and then right-click and try to open it using Open with -> ghci. I get the following error message: can't find module `C:\Documents' (while processing "C:\\Documents") Obviously GHCi incorrectly handles white space in the path. If I run under Cygwin I can use a script like: c:\\ghc\\ghc-5.04.2\\bin\\ghci `cygpath -d $1`; But how do I solve this if I'm not using cygwin? Thanks, Calle

2 4

RE: ghc feature request: core notes
by Simon Marlow 13 Feb '03

13 Feb '03

> Yes in source code (albeit I have not tested this recently). > 'inline' is [was] a magic Id whose unfolding is the relevant > core goop. Like I say, the last time I visted this was years ago! It doesn't appear to exist anymore. But maybe it would be a nice thing to have? Cheers, Simon

1 0

RE: ghc feature request: core notes
by Simon Peyton-Jones 13 Feb '03

13 Feb '03

Hal, OK. If you want to change CoreSyn.Note to have an extra constructor, thus: data Note = SCC CostCentre | Coerce Type Type | InlineMe | InlineCall | NoteString String -- NEW and check that the rest of the compiler does something sensible with NoteString, I do not object. There are only 12 lines of the compiler than mention InlineCall, for example, so I expect NoteString would be similarly low-impact. [Incidentally, InlineCall is an undocumented feature. You can write inline f x to mean the same as (f x), except that this particular call to f will be inlined.] Simon | -----Original Message----- | From: Alastair Reid [mailto:alastair@reid-consulting-uk.ltd.uk] | Sent: 06 February 2003 11:28 | To: Simon Peyton-Jones | Cc: Hal Daume III; GHC Users Mailing List | Subject: Re: ghc feature request: core notes | | | "Hal Daume III" <hdaume(a)ISI.EDU> writes: | > I'm not sure how "generally useful" this would be, but I would | > find it useful to be able to attach notes to core expressions from | > the Haskell code. The idea being something along the lines of a | > code annotation like a pragma. [...] | | Simon Peyton-Jones <simonpj(a)microsoft.com> writes: | > [...] There is an annotation facility in Core, but it's easy for the notes | > to be discarded. So if they are conveying important info, it might | > easily get lost... and if not, what's it doing there in the first | > place. What do you expect to happen to these annotations? [...] | | It seems like _scc_ is (almost) exactly what you need here. When used | for profiling, you mark subexpressions you want profiled and ghc takes | care to preserve and propagate them in a semantically meaningful way. | Any _scc_ annotations that ghc chooses to drop can be reported as | bugs :-) | | I say "almost" because what the semantics used when profiling may not | be the right ones for your purposes. (I'm being vague because I don't | know your purposes.) | | Naturally, you'd want to be able to use these notes independently of | profiling - which can probably be done with a little preprocessor | hackery. Assuming that the 1st argument to 'NOTE' is a literal string | in the following: | | #ifdef TRACK_NOTES | #define NOTE(x,y) (_scc_ "<NOTE " x ">" (y)) | #else | #define NOTE(x,y) {- nothing -} | #endif | | -- | Alastair Reid

3 3

Building snapshot ghc-5.05.20030123 on macOSX
by Arthur Baars 12 Feb '03

12 Feb '03

I want to play a bit with Template Haskell, so I downloaded a cvs snapshot (ghc-5.05.20030123.tar.gz) and tried to compile it on my mac (OSX 10.2). I figured out that I needed a stage2 or stage3 compiler to get TH working, so I tried 'make stage3' . GHC started to compile itself three times. At lot of hours later it finished. Unfortunately, when I tried run the thing on a Prinf example, I got the following message. compiler/stage3-> ./ghc-inplace -c Printf.hs -fglasgow-exts -package haskell-src Template Haskell bracket illegal in a stage-1 compiler [| \s -> s |] The thing in"compiler/stage3" was just a stage-1 compiler, so ghc had achieved nothing useful after re-compiling three times :-( The problem is that GHCi support is not turned on for MacOSX. I guess GHCi is used to do compile-time evaluation of code for TH. In file "mk/config.mk.in" it says: # Include GHCi in the compiler. Default to NO for the time being. ifneq "$(findstring $(HostOS_CPP), mingw32 cygwin32 linux solaris2 freebsd netbsd openbsd)" "" GhcWithInterpreter=YES else GhcWithInterpreter=NO endif In the list of supported systems MacOSX is missing. After adding "darwin" to this list, and waiting for hours, I got a working TH-compiler. Can someone add OSX("darwin") to the "mk/config.mk.in" file? It would also be nice to get a warning if one tries to compile a stage-(N>1) compiler without GHCi support. Cheers, Arthur

3 2

RE: ghc feature request: core notes
by Simon Peyton-Jones 12 Feb '03

12 Feb '03

Yes in source code (albeit I have not tested this recently). 'inline' is [was] a magic Id whose unfolding is the relevant core goop. Like I say, the last time I visted this was years ago! S | -----Original Message----- | From: Simon Marlow [mailto:simonmar@microsoft.com] | Sent: 12 February 2003 16:56 | To: Simon Peyton-Jones; Alastair Reid | Cc: Hal Daume III; GHC Users Mailing List | Subject: RE: ghc feature request: core notes | | | > [Incidentally, InlineCall is an undocumented feature. You can write | > | > inline f x | > | > to mean the same as (f x), except that this particular call | > to f will be | > inlined.] | | Not in source code, surely? It certainly doesn't look like 'inline' is | a keyword. In external Core, you can write '%note InlineCall' to get | this effect though. | | Cheers, | Simon | | _______________________________________________ | Glasgow-haskell-users mailing list | Glasgow-haskell-users(a)haskell.org | http://www.haskell.org/mailman/listinfo/glasgow-haskell-users

1 0

RE: ghc feature request: core notes
by Simon Marlow 12 Feb '03

12 Feb '03

> [Incidentally, InlineCall is an undocumented feature. You can write > > inline f x > > to mean the same as (f x), except that this particular call > to f will be > inlined.] Not in source code, surely? It certainly doesn't look like 'inline' is a keyword. In external Core, you can write '%note InlineCall' to get this effect though. Cheers, Simon

1 0

RE: Using large amounts of system time
by Simon Marlow 12 Feb '03

12 Feb '03

> I appear to be using huge amounts of CPU time, and I am trying to work > out why. The commands I ran and their output are in systime_log at > http://urchin.earth.li/~ian/systime/, and all the source and other > output is there too. Essentially the program uses more system > time than > user time according to time (bash built-in, Linux x86), and > I'd like to > know why (or more precisely I'd like to reduce it, but > knowing why seems > like a good first move :-) ). > > I'm somewhat confused as to why the numbers given by time and > strace -c > don't match, but nevertheless it looks to me like the time must be due > to one or more of: > > rt_sigprocmask(SIG_SETMASK, [], NULL, 8) = 0 > rt_sigprocmask(SIG_BLOCK, [], [], 8) = 0 > --- SIGVTALRM (Virtual timer expired) --- > sigreturn() = ? (mask now []) > > I'm not sure where to go from here. Any suggestions? >From the strace -c ouput, I'd say the system time isn't being eaten up by system calls and signals. The only other system time sink is page faults and swapping activity (I believe), but from the look of it your program is only using a few MB. What does the output from +RTS -Sstderr look like? I'm stumped too, but I'm intrigued to know what's going on. Cheers, Simon

2 1

RE: GHC 5.05 daily snapshots
by Simon Marlow 12 Feb '03

12 Feb '03

> The last source snapshot is dated from the 24th of january. > Is there any reason for that ? Sorry about that, a small hiccup in some recent changes I'd been making to the nightly build system. Should be fixed now, hopefully a source tarball will pop out of tonight's build. Cheers, Simon

1 0

GHC 5.05 daily snapshots
by Alain Cremieux 11 Feb '03

11 Feb '03

Hi, The last source snapshot is dated from the 24th of january. Is there any reason for that ? Thanks, Alain

1 0