- Glasgow-haskell-users

RE: [Haskell-cafe] Packages and modules
by Simon Peyton-Jones 07 Jul '06

07 Jul '06

Simon and I have been thinking about fixing this, and we think we might actually do so for GHC 6.6. Your message provoked us to write up the design. It's here http://hackage.haskell.org/trac/ghc/wiki/GhcPackages Feedback welcome It's worth reading the old threads; for example http://www.haskell.org//pipermail/libraries/2005-August/004281.html But there are many others! Simon | -----Original Message----- | From: haskell-cafe-bounces(a)haskell.org [mailto:haskell-cafe-bounces@haskell.org] On Behalf Of Brian | Hulley | Sent: 25 June 2006 10:16 | To: Haskell-cafe | Subject: [Haskell-cafe] Packages and modules | | Hi - | At the moment there is a problem in that two packages P and Q could contain | the same hierarchical module eg Data.Foo, and the only way for user code to | ensure the "right" Data.Foo is used is to ensure that packages P and Q are | searched in the right order. | However suppose P and Q also contain another module with the same name, eg | Data.Bar. | And suppose a user module wants to use Data.Foo from P but Data.Bar from | Q!!!

12 30

Weird bug with FDs
by Niklas Broberg 06 Jul '06

06 Jul '06

I encounter a strange behavior with functional dependencies. Assume we have a class defined as class Foo x y | x -> y where foo :: x -> y and another class class Bar x y where bar :: x -> y -> Int and I want to write the instance declaration instance (Foo x y, Bar y z) => Bar x z where bar x z = bar (foo x) z Compiling (with 6.4.2, -fallow-undecidable-instances and -fglasgow-exts) I get the following error message: <error> Foo.hs:12:0: Context reduction stack overflow; size = 21 Use -fcontext-stack20 to increase stack size to (e.g.) 20 `$dBar :: Bar y z' arising from use of `bar' at Foo.hs:13:11-13 [... same thing 20 times ...] `$dBar :: Bar y z' arising from use of `bar' at Foo.hs:13:11-13 `bar :: {bar at [y z]}' arising from use of `bar' at Foo.hs:13:11-13 When trying to generalise the type inferred for `bar' Signature type: forall x y z. (Foo x y, Bar y z) => x -> z -> Int Type to generalise: x -> z -> Int In the instance declaration for `Bar x z' </error> The declaration requires undecidable instances, but I doubt that the problem comes from that. What makes it even more weird is that I can get this to compile, and behave as expected, if I do one of a) declare an instance of Bar for any type, or b) add an explicit type signature (foo x :: y) in the definition of Bar. The first seems weird since how could a different, unrelated instance affect the typeability of the second instance? The second, b), is weird since by the FD x -> y we should already know that foo x :: y. GHC 6.4.1 shows the same behavior. Hugs (with -98 +O) accepts the code. I've submitted a trac ticket, but it would be interesting to hear if anyone has an explanation to this? :-) Thanks, /Niklas

1 0

Packages: exposed, hidden, available (was: Packages and Modules)
by Dimitry Golubovsky 05 Jul '06

05 Jul '06

Simon Marlow wrote: In fact, we can imagine three states that a package could be in: - exposed: the package's modules populate the global module namespace, explicit "from" imports may be used to resolve ambiguity - hidden: the package cannot be used at all - available: the package can be used only by an explicit "from" import (this is rather like a qualified import at the package level, and might be useful if you want to use a package with module names that overlap with local modules). --------------------------------------------------------------------------------------------------------- I'd suggest the following sub-definition for "available" (other name could be "transient"): A package which is "available" is not "installed" at (immediately known to) a particular developer's system, as opposed to "exposed" and "hidden" which are "installed". When a package/program is being built that requires a package which is not installed, three options are available: - a package is located (on the Internet, using Cabal-get or Cabalfind, or manual search), downloaded, and installed (i. e. becomes known to the developer's system), and gets into "exposed" or "hidden" category - a package is located, downloaded temporarily (i. e. cached) on the developer's system*, used for a particular package that requires it and discarded afterwards (or lives for some time in the cache).** - package cannot be found (Internet search fails) ============ * just due to the fact that existing OS's generally distinguish between local and network resources. If we imagine some sort of "http file system" or "darcs file system" that mount URLs/repos to local file system entries then explicit downloading is not necessary. The latter may be (or already is) implemented as user-level file systems with recent Linux kernels. ** one might think about "lazy package resolution" i. e. only minimal subset of packages like base, mtl, network, etc that generally come with GHC/other compiler are installed on a developer's system, and the rest of packages are transient/cached on demand. -- Dimitry Golubovsky Anywhere on the Web

2 1

Re: [Haskell-cafe] Deducing Show for GADTs
by Niklas Broberg 28 Jun '06

28 Jun '06

On 6/28/06, David Roundy <droundy(a)darcs.net> wrote: > On Wed, Jun 28, 2006 at 11:52:51AM +0200, Joel Bjrnson wrote: > > Hi. I came a cross the following phenomena which, at least to me, > > occurs kind of awkward. The code below: > > > > data MyData a where > > DC1 :: (Show a ) => a -> MyData a > > GADTs don't yet work right with classes. :( The above, however, > doesn't need to be expressed as a GADT, I believe you can write > something like: > > data MyData a = (forall a. Show a) => DC1 a > > which (this is untested) should do what you want. Only if "what he wants" is something that type checks, but doesn't do the same thing. ;-) In Joel's definition of MyData, values constructed with DC1 applied to a value of type b will have type MyData b. In your definition they will have type MyData a, for any a. In other words, your definition would be identical to the GADT data MyData a where DC1 :: forall a b . (Show b) => b -> MyData a As to Joel's question, this seems really really weird. In particular since adding the completely useless wrapper type solves the problem. In fact, giving DC1 any return type other than MyData a solves the problem. This has to be a bug of some sort. /Niklas

1 0

progress on Darwin x86 port?
by Rodney D Price 27 Jun '06

27 Jun '06

Hi, Some time ago I persuaded my employer to buy me a new Intel-based Mac. While it runs fast, it doesn't compile many open-source programs, including ghc. On perusing the list archives, I found one or two people working on a port. What's the status? Right now I'm writing Haskell code using ghci, which is running under Rosetta. Is it possible to build an unregisterised version of ghc that would at the very least emit executable binaries on Intel Macs? Thanks, -Rod

2 2

Illegal instruction in GHC 6.5 generated code.
by Caio Marcelo 26 Jun '06

26 Jun '06

(If this is the wrong list for this, please tell me. I tried Glasgow-Haskell-Bugs, but message got hold in moderation) Hello, Using GHC 6.5 from darcs repo ("pulled" in 19/Jun), the following snippet of code seems to cause illegal instruction error on the generated code. The error disappears if I use the Dummy instance on top of Refeable. The code is from my bindings to Judy library, revision 10752 in subversion. http://svn.openfoundry.org/pugs/third-party/HsJudy/ http://svn.openfoundry.org/pugs/third-party/judy/ # original C lib code To test it you'll need to compile (./configure; make # but not necessarily install) the original library in third-party/judy/Judy-1.0.3. Back in HsJudy, change the Makefile to static link option. HsJudy$ make testmap HsJudy$ ./testmap Sorry, I tried to come up with an simpler instance of the problem but didn't found one. The program depend on WordPtr which isn't available in GHC 6.4.2, so I couldn't test it in stable version. It was tested on my amd64, linux 2.6.16 and in audreyt's machine (OSX I think). ---------8<------------ --class Dummy a --instance Dummy a --instance Dummy a => Refeable a where instance Refeable a where toRef a = do a' <- newStablePtr a return (ptrToWordPtr (castStablePtrToPtr a')) fromRef v = do a <- deRefStablePtr (castPtrToStablePtr (wordPtrToPtr v)) return a -- Don't need to StablePtr "simple" things like Int instance Refeable Int where toRef i = return $ toEnum i fromRef v = return $ fromEnum v ---------->8------------- Thanks in advance, -- Caio Marcelo

3 4

Compiling with -threaded under Solaris 10
by Florian Lorenzen 26 Jun '06

26 Jun '06

Hi all, I'm trying to run the following program on Solaris 10 with GHC 6.4.1 (6.4.2 does not really work under Solaris 10 right now). module Main where import IO import Control.Concurrent import System.Process main = do (inp, out, err, ph) <- runInteractiveProcess "cat" [] Nothing Nothing forkIO (do hPutStr inp "this\nis\na\ntest\n" hClose inp) forkIO (do r <- hGetContents out putStr r) waitForProcess ph I compile it with $ ghc -o try Try2.hs -threaded -lrt When running, I get $ ./try try: waitForProcess: interrupted (Interrupted system call) and the program aborts. Under Linux, everything is fine. Do I have to set some macros explicitly? And if so, which ones? Any help appreciated. Best regards, Florian

2 2

GHC driver behaviour switch
by kyra 23 Jun '06

23 Jun '06

With GHC 6.4.x I was able to use a ghc driver to compile .cpp sources. Now, with GHC HEAD I cannot. GHC HEAD now ALWAYS enforces "treat as .c" option. Let's compare "DriverPipeline.hs". GHC 6.4.x: let langopt | hcc = [ SysTools.Option "-x", SysTools.Option "c"] | otherwise = [ ] SysTools.runCc dflags ( -- force the C compiler to interpret this file as C when -- compiling .hc files, by adding the -x c option. -- Also useful for plain .c files, just in case GHC saw a -- -x c option. langopt ++ ... GHC HEAD: SysTools.runCc dflags ( -- force the C compiler to interpret this file as C when -- compiling .hc files, by adding the -x c option. -- Also useful for plain .c files, just in case GHC saw a -- -x c option. [ SysTools.Option "-x", SysTools.Option "c"] ++ ... This completely breaks my build infrastructure. Could it be reversed? Cheers, Kyra

3 4

FFI: number of worker threads?
by Li, Peng 22 Jun '06

22 Jun '06

Hello, The paper "Extending the Haskell FFI with Concurrency" mentioned the following in Section 6.3: "GHC's run-time system employs one OS thread for every bound thread; additionally, there is a variable number of so-called "worker" OS threads that are used to execute the unbounded (lightweight) threads." How does the runtime system determine the number of worker threads? Is the number hardcoded in the RTS or dynamically adjustable? Can a programmer specify it as an RTS option or change it using an API? I would like to use a large number (say, 2000) of unbounded threads, each calling a blocking, safe foreign function via FFI import. What is supposed to happen if all the worker threads are used up? I tried this in the recent GHC 6.5 and got some kind of "runaway worker threads?" RTS failure message when more than 32 threads are used. Is it a current limitation of the RTS, or should I file a bug report for it? Thanks, Peng

8 22

partial x86_64 binary snapshots?
by Jens Petersen 22 Jun '06

22 Jun '06

I wanted to download a nightly binary stable snapshot for Linux x86_64 to test if the problem with selinux on Fedora Core has been fixed... but AFAICT recent snapshots (I tried too) are broken: % bunzip2 -t ghc-6.4.2.20060621-x86_64-unknown-linux.tar.bz2 bunzip2: ghc-6.4.2.20060621-x86_64-unknown-linux.tar.bz2: file ends unexpectedly You can use the `bzip2recover' program to attempt to recover data from undamaged sections of corrupted files. Just wondering what is up with this? Jens

1 1