Re: Windows faliures

2010-01-29 Thread Andrew Suffield
On Fri, Jan 29, 2010 at 09:30:38AM +, Simon Marlow wrote: > The openTempFile001 failure is probably harmless, it just indicates > that we tried to write a character that isn't known in your current > locale (maybe you can set your locale to a Unicode one, I can't > remember how to do that on Wi

Re: rpath for shared libraries

2010-01-12 Thread Andrew Suffield
On Tue, Jan 12, 2010 at 08:22:48AM +, Duncan Coutts wrote: > man ld.so says: > > The shared libraries needed by the program are searched for in the > following order: > > * (ELF only) Using the directories specified in the DT_RPATH > dynamic section attribute of the binary if

Re: rpath for shared libraries

2010-01-11 Thread Andrew Suffield
On Tue, Jan 12, 2010 at 12:16:31AM +, Duncan Coutts wrote: > On Mon, 2010-01-11 at 20:19 +0000, Andrew Suffield wrote: > > > To reiterate, this is the exemplar use case for relocatable objects: > > > > Install ghc and some libraries to /usr/local/ > > Build

Re: rpath for shared libraries

2010-01-11 Thread Andrew Suffield
On Mon, Jan 11, 2010 at 04:32:02PM +, Duncan Coutts wrote: > There are advantages to not doing so. In particular multiple instances > of the same version of a package, or multiple instances of the same ghc > version can co-exist happily if they're isolated. Why is this useful? I can't imagine

Re: Windows DLLs should be working.

2010-01-10 Thread Andrew Suffield
On Mon, Jan 11, 2010 at 05:31:46AM +0100, Lars Viklund wrote: > > If you want to get this working with COM, then building an assembly > > out of the DLLs is probably mandatory. So that's looking like the way > > to go. > > If memory serves me right, you can register a COM DLL anywhere on the > sys

Re: rpath for shared libraries

2010-01-10 Thread Andrew Suffield
On Sun, Jan 10, 2010 at 07:11:43PM +, Duncan Coutts wrote: > Right. I think it's much more sensible to use isolated locations for > installed library files. It makes it much easier to have multiple > instances of things. This does not appear to be an issue because the .so files are uniquely na

Re: Windows DLLs should be working.

2010-01-10 Thread Andrew Suffield
On Sun, Jan 10, 2010 at 12:55:35PM +, Neil Mitchell wrote: > This isn't a patch that changes the default - I suspect Windows will > build static exes forever more. However, there are some circumstances > where you need Haskell to build with DLL's (writing Haskell COM > components for example),

Re: Windows DLLs should be working.

2010-01-10 Thread Andrew Suffield
On Sun, Jan 10, 2010 at 11:15:08PM +1100, Majestic Perch wrote: > The problem we have is that when running the testsuite there are > many directories containing many individual .hs files that need to > be built and linked against the GHC libraries.. Adding a copy to > each directory isn't really an

Re: Windows DLLs should be working.

2010-01-10 Thread Andrew Suffield
On Sun, Jan 10, 2010 at 05:24:17PM +1100, Ben Lippmeier wrote: > I haven't pushed a patch to do this yet because the DLLs for the > runtime and libraries currently require manual installation. After > building GHC I've been doing a find . -name "*.dll" and copying > all the results to c:\Windows

Type checking errors & foreign exports (was: #3806)

2010-01-05 Thread Andrew Suffield
> In short, a `foreign export` is not a type signature at all. It's a > ''use'' of `foo`. Yes, but the diagnostic doesn't point you to the right location. After sleeping on it, this is an instance of a more general issue with type checking errors: they're following the model of an explicitly ty

Re: Validate failure

2010-01-05 Thread Andrew Suffield
On Tue, Jan 05, 2010 at 09:54:41AM +, Simon Marlow wrote: > I've also seen failures that look suspiciously like attempts to run > more than one ghc-pkg at the same time, which as far as I know the > build system is not supposed to do. It's difficult to get make to understand this sort of thing

Re: rpath for shared libraries

2009-12-31 Thread Andrew Suffield
On Fri, Jan 01, 2010 at 01:38:58AM +0100, Lars Viklund wrote: > On Wed, Dec 30, 2009 at 06:29:23PM +0000, Andrew Suffield wrote: > > I've been thinking about this. The only reason why rpath is needed is > > because the .so files are being stuck in lib/ghc-$version/ instead o

Re: Validate failure

2009-12-31 Thread Andrew Suffield
On Thu, Dec 31, 2009 at 02:34:33PM +, Ian Lynagh wrote: > * Why does HSdph-prim-par-0.4.0-ghc6.13.20091222 have filename > libHSdph-prim-seq-0.4.0-ghc6.13.20091222.so > (note par vs seq) Probably because -par-*.so is linked against -seq-*.so; the error message is from dlopen(), which

rpath for shared libraries

2009-12-30 Thread Andrew Suffield
On Wed, Dec 30, 2009 at 02:20:41PM +, Duncan Coutts wrote: > On Wed, 2009-12-30 at 12:09 +0100, Maxime Henrion wrote: > > > - Is there a plan to deal with the ldconfig cache on UNIX systems? As > > things are now, I had to manually add all the package directories > > under /usr/local/lib/ghc-6

Re: Patch for shared libraries support on FreeBSD

2009-12-30 Thread Andrew Suffield
On Wed, Dec 30, 2009 at 12:09:08PM +0100, Maxime Henrion wrote: > - Is there a plan to deal with the ldconfig cache on UNIX systems? As > things are now, I had to manually add all the package directories > under /usr/local/lib/ghc-6.12.1/ to be able to run the generated > executable. I guess Cabal