packages/unix.git
7 years agoBump version to 2.5.1.1 ghc-7.4 ghc-7.4.2-release unix-2.5.1.1-release
Paolo Capriotti [Sun, 1 Apr 2012 19:00:21 +0000 (20:00 +0100)] 
Bump version to 2.5.1.1

7 years agofix documentation for getAnyProcessStatus/getGroupProcessStatus (#5946)
Simon Marlow [Fri, 23 Mar 2012 15:10:17 +0000 (15:10 +0000)] 
fix documentation for getAnyProcessStatus/getGroupProcessStatus (#5946)

MERGED from commit e518038a0eca7dddbbd3d4326355db9c1d6f0068

7 years agoDon't expose System.Posix.DynamicLinker.Common ghc-7.4.1-release unix-2.5.1.0-release
Ian Lynagh [Thu, 15 Dec 2011 14:47:03 +0000 (14:47 +0000)] 
Don't expose System.Posix.DynamicLinker.Common

The other .Common modules aren't expose, and we think exposing this
one was a mistake.

7 years agoUpdate dependencies
Ian Lynagh [Mon, 12 Dec 2011 21:39:01 +0000 (21:39 +0000)] 
Update dependencies

7 years agoforkProcess now works with +RTS -N; update docs and tests
Simon Marlow [Tue, 6 Dec 2011 15:20:28 +0000 (15:20 +0000)] 
forkProcess now works with +RTS -N; update docs and tests

7 years agoRoll back the sigset capi changes
Ian Lynagh [Sat, 3 Dec 2011 16:43:18 +0000 (16:43 +0000)] 
Roll back the sigset capi changes

They broken the build on OSX. See #2979.

7 years agoUse capi to define the fcntl FFI imports
Ian Lynagh [Mon, 28 Nov 2011 18:59:18 +0000 (18:59 +0000)] 
Use capi to define the fcntl FFI imports

7 years agoUse capi some more (part of #5480)
Ian Lynagh [Mon, 28 Nov 2011 18:20:04 +0000 (18:20 +0000)] 
Use capi some more (part of #5480)

7 years agoConvert come FFI bindings to use the capi calling convention
Ian Lynagh [Sat, 26 Nov 2011 16:16:51 +0000 (16:16 +0000)] 
Convert come FFI bindings to use the capi calling convention

7 years agoFix Solaris breakage (I hope)
Simon Marlow [Mon, 28 Nov 2011 09:35:36 +0000 (09:35 +0000)] 
Fix Solaris breakage (I hope)

7 years agoFixes for NetBSD
Ian Lynagh [Fri, 25 Nov 2011 01:17:02 +0000 (01:17 +0000)] 
Fixes for NetBSD

Based on a patch from Arnaud Degroote <degroote@NetBSD.org> in
trac #5480.

7 years agoMove openPseudoTerminal into System.Posix.Terminal{.ByteString}
Simon Marlow [Wed, 23 Nov 2011 09:35:29 +0000 (09:35 +0000)] 
Move openPseudoTerminal into System.Posix.Terminal{.ByteString}

It may depend on getSlaveTerminalName if !defined(HAVE_OPENPTY)

7 years agoBump version 2.5.0.0 -> 2.5.1.0 (added RawFilePath API)
Simon Marlow [Thu, 17 Nov 2011 13:25:23 +0000 (13:25 +0000)] 
Bump version 2.5.0.0 -> 2.5.1.0 (added RawFilePath API)

7 years agoProvide a raw ByteString version of FilePath and environment APIs
Simon Marlow [Fri, 11 Nov 2011 16:18:48 +0000 (16:18 +0000)] 
Provide a raw ByteString version of FilePath and environment APIs

The new module System.Posix.ByteString provides exactly the same API
as System.Posix, except that:

  - There is a new type: RawFilePath = ByteString

  - All functions mentioning FilePath in the System.Posix API
    use RawFilePath in the System.Posix.ByteString API

  - RawFilePaths are not subject to Unicode locale encoding and
    decoding, unlike FilePaths.  They are the exact bytes passed to and
    returned from the underlying POSIX API.

  - Similarly for functions that deal in environment
    strings (System.Posix.Env): these use untranslated ByteStrings
    in System.Posix.Environment

  - There is a new function

     System.Posix.ByteString.getArgs :: [ByteString]

    returning the raw untranslated arguments as passed to exec() when
    the program was started.

7 years agoexecuteFile001: omit profiling ways
Simon Marlow [Mon, 21 Nov 2011 14:37:15 +0000 (14:37 +0000)] 
executeFile001: omit profiling ways

The RTS doesn't get a chance to produce the profile output after
executeFile, so tests that check for correct profiling output will fail.

7 years agoAdd a test for #5319
Ian Lynagh [Wed, 16 Nov 2011 18:21:32 +0000 (18:21 +0000)] 
Add a test for #5319

7 years agoFollow change to FFI decls: Import constructors of newtypes
Ian Lynagh [Thu, 20 Oct 2011 23:52:38 +0000 (00:52 +0100)] 
Follow change to FFI decls: Import constructors of newtypes

7 years agoFollow the removal of the Eq superclass of Num
Ian Lynagh [Thu, 15 Sep 2011 21:03:51 +0000 (22:03 +0100)] 
Follow the removal of the Eq superclass of Num

7 years agoRemove old darcs boring file
David Terei [Wed, 10 Aug 2011 19:35:13 +0000 (12:35 -0700)] 
Remove old darcs boring file

7 years agoFix conditional pragma to work with 6.12
David Terei [Tue, 9 Aug 2011 17:50:03 +0000 (10:50 -0700)] 
Fix conditional pragma to work with 6.12

7 years agoUse Safe Haskell when GHC >= 7.2
David Terei [Thu, 4 Aug 2011 00:06:28 +0000 (17:06 -0700)] 
Use Safe Haskell when GHC >= 7.2

7 years agoRemove some antiquated C constructs
Ian Lynagh [Mon, 1 Aug 2011 00:08:55 +0000 (01:08 +0100)] 
Remove some antiquated C constructs

Fixes validate on amd64/Linux with:

SRC_CC_OPTS += -Wmissing-parameter-type
SRC_CC_OPTS += -Wold-style-declaration
SRC_CC_OPTS += -Wold-style-definition

7 years agoRelax base dep ghc-7.2 ghc-7.2.1-release ghc-7.2.2-release
Ian Lynagh [Thu, 7 Jul 2011 22:07:19 +0000 (23:07 +0100)] 
Relax base dep

7 years agoBump version number
Ian Lynagh [Wed, 6 Jul 2011 17:13:30 +0000 (18:13 +0100)] 
Bump version number

7 years agoprefer bsd/libutil.h (if available) to libutil.h
Ross Paterson [Wed, 29 Jun 2011 17:58:04 +0000 (18:58 +0100)] 
prefer bsd/libutil.h (if available) to libutil.h

libutil.h is deprecated in recent releases of libbsd

8 years agoFix typos in future process group API
Favonia [Tue, 24 May 2011 22:47:35 +0000 (18:47 -0400)] 
Fix typos in future process group API

8 years agoelaborate the deprecation messages, and add commented-out new
Simon Marlow [Mon, 23 May 2011 10:30:40 +0000 (11:30 +0100)] 
elaborate the deprecation messages, and add commented-out new
versions, for setProcessGroupID and createProcessGroup

8 years agoadd processGroup001/processGroup002
Simon Marlow [Mon, 23 May 2011 10:13:13 +0000 (11:13 +0100)] 
add processGroup001/processGroup002

8 years agoTest cases for the new process group API.
Favonia [Thu, 12 May 2011 02:35:53 +0000 (22:35 -0400)] 
Test cases for the new process group API.

8 years agoChange the POSIX process group API. (trac #5167)
Favonia [Wed, 4 May 2011 10:52:44 +0000 (06:52 -0400)] 
Change the POSIX process group API. (trac #5167)

Make it possible to query the process group of an existing
process (through 'getProcessGroupIDOf') and try to make
function names more consistent. Here is the full list of
API changes in System.Posix.Process in this patch:

getProcessGroupID     => same
getProcessGroupIDOf   => new
createProcessGroup    => deprecated
createProcessGroupFor => new
joinProcessGroup      => same
setProcessGroupID     => deprecated
setProcessGroupIDOf   => new

8 years agoImproved Unicode support in the light of PEP383
Max Bolingbroke [Sat, 14 May 2011 21:45:27 +0000 (22:45 +0100)] 
Improved Unicode support in the light of PEP383

8 years agoallow some syscalls in System.Posix.Directory to return EINTR (#5184)
Simon Marlow [Tue, 10 May 2011 10:16:55 +0000 (11:16 +0100)] 
allow some syscalls in System.Posix.Directory to return EINTR (#5184)

8 years agoUpdate source repo location to be the git repo in .cabal file
Ian Lynagh [Fri, 8 Apr 2011 10:23:09 +0000 (11:23 +0100)] 
Update source repo location to be the git repo in .cabal file

8 years agoUse _NSGetEnviron on OS X: fixes #2458
Max Bolingbroke [Wed, 6 Apr 2011 06:57:21 +0000 (07:57 +0100)] 
Use _NSGetEnviron on OS X: fixes #2458

8 years agoAdd initial .gitignore
Max Bolingbroke [Tue, 5 Apr 2011 07:54:28 +0000 (08:54 +0100)] 
Add initial .gitignore

8 years agoInclude HsUnix.h earlier in System/Posix/Files.hsc ghc-darcs-git-switchover
Ian Lynagh [Thu, 24 Mar 2011 00:10:29 +0000 (00:10 +0000)] 
Include HsUnix.h earlier in System/Posix/Files.hsc

8 years agoAdd NondecreasingIndentation to the list of extensions
Ian Lynagh [Mon, 17 Jan 2011 19:24:30 +0000 (19:24 +0000)] 
Add NondecreasingIndentation to the list of extensions

8 years agoProper autoconf test for sem_close's library; fixes trac #4523
Ian Lynagh [Sat, 1 Jan 2011 20:11:50 +0000 (20:11 +0000)] 
Proper autoconf test for sem_close's library; fixes trac #4523

8 years agorun forkprocess01 with +RTS -ls too, to test #4512
Simon Marlow [Fri, 10 Dec 2010 09:46:45 +0000 (09:46 +0000)] 
run forkprocess01 with +RTS -ls too, to test #4512

8 years agoClarify behavior of "awaitSignal Nothing", export SignalSet that includes all signals...
Dmitry Astapov [Wed, 8 Dec 2010 18:38:49 +0000 (18:38 +0000)] 
Clarify behavior of "awaitSignal Nothing", export SignalSet that includes all signals reserved by RTS (#4504)

8 years agoRemove unnecessary fromIntegral calls
simonpj@microsoft.com [Tue, 16 Nov 2010 17:25:36 +0000 (17:25 +0000)] 
Remove unnecessary fromIntegral calls

8 years agoBump version to 2.4.1.0
Ian Lynagh [Fri, 17 Sep 2010 15:01:14 +0000 (15:01 +0000)] 
Bump version to 2.4.1.0

8 years agoavoid Foreign.unsafePerformIO
Ross Paterson [Thu, 9 Sep 2010 12:54:59 +0000 (12:54 +0000)] 
avoid Foreign.unsafePerformIO

9 years agoAdd bindings for setting effective user-id and group-id
Eric Mertens [Mon, 17 May 2010 18:19:45 +0000 (18:19 +0000)] 
Add bindings for setting effective user-id and group-id

Having the ability to set the effective user-id and group-id
enables the application to switch between the real user-id
and set-user-id multiple times.

9 years agoAllow base-4.3
Simon Marlow [Wed, 2 Jun 2010 08:25:09 +0000 (08:25 +0000)] 
Allow base-4.3

9 years agoadd a comment about 3816 failing due to Samba
Simon Marlow [Tue, 18 May 2010 08:44:22 +0000 (08:44 +0000)] 
add a comment about 3816 failing due to Samba

9 years agoProvide a dummy input file for queryfdoption01
Matthias Kilian [Sun, 18 Apr 2010 17:26:19 +0000 (17:26 +0000)] 
Provide a dummy input file for queryfdoption01

9 years agomention that forkProcess is not supported with +RTS -Nn where n > 1
Simon Marlow [Fri, 30 Apr 2010 08:28:41 +0000 (08:28 +0000)] 
mention that forkProcess is not supported with +RTS -Nn where n > 1

9 years agomake getAllGroupEntries work when called again; fixes #3816
Simon Marlow [Mon, 29 Mar 2010 11:27:13 +0000 (11:27 +0000)] 
make getAllGroupEntries work when called again; fixes #3816

9 years agoadd test for #3816
Simon Marlow [Mon, 29 Mar 2010 11:26:46 +0000 (11:26 +0000)] 
add test for #3816

9 years agofix warnings
Simon Marlow [Mon, 22 Mar 2010 13:16:15 +0000 (13:16 +0000)] 
fix warnings

9 years agohandleToFd: close both sides of a DuplexHandle (#3914)
Simon Marlow [Fri, 19 Mar 2010 21:08:02 +0000 (21:08 +0000)] 
handleToFd: close both sides of a DuplexHandle (#3914)

9 years agocheck for EINTR in openFd
Simon Marlow [Wed, 27 Jan 2010 11:46:00 +0000 (11:46 +0000)] 
check for EINTR in openFd

9 years agoaccept --with-cc to set the path to gcc (#2966)
Simon Marlow [Wed, 27 Jan 2010 11:43:29 +0000 (11:43 +0000)] 
accept --with-cc to set the path to gcc (#2966)

9 years agoSystem.Posix.Temp: rm whitespace
gwern0@gmail.com [Tue, 19 Jan 2010 00:34:37 +0000 (00:34 +0000)] 
System.Posix.Temp: rm whitespace

9 years agoSystem.Posix.Temp: pad input filenames with 6 Xs
gwern0@gmail.com [Tue, 19 Jan 2010 00:30:22 +0000 (00:30 +0000)] 
System.Posix.Temp: pad input filenames with 6 Xs
If the argument doesn't terminate in capital Xs, the C mkstemp will simply bomb out
with zero warning. This was not documented.
By arbitrarily sticking a bunch of Xes at the end of all arguments, we guarantee that
this exception will not be thrown, the type signature will not change, and no existing
code can break (since if it was manually avoiding the exception by adding "XXX" itself,
the temp files will now be simply 3 random characters longer, nothing worse).

9 years agoSystem.Posix.Temp: improve haddocks for mkstemp to explain what that returned String...
gwern0@gmail.com [Tue, 19 Jan 2010 00:27:55 +0000 (00:27 +0000)] 
System.Posix.Temp: improve haddocks for mkstemp to explain what that returned String actually is

9 years agofix base dependency: should be >= 4.2 (#3780), and bump verison to 2.4.0.1
Simon Marlow [Wed, 13 Jan 2010 11:38:03 +0000 (11:38 +0000)] 
fix base dependency: should be >= 4.2 (#3780), and bump verison to 2.4.0.1

9 years agono good reason these tests should be expect_fail, as far as I can tell
Simon Marlow [Thu, 31 Dec 2009 14:57:39 +0000 (14:57 +0000)] 
no good reason these tests should be expect_fail, as far as I can tell

9 years agoSupport for DragonFly
Simon Marlow [Mon, 21 Dec 2009 11:07:09 +0000 (11:07 +0000)] 
Support for DragonFly
Not sure where this patch came from, I found it in my validate tree.

9 years agoDon't use absolute paths to headers (#3728)
Simon Marlow [Mon, 21 Dec 2009 11:06:34 +0000 (11:06 +0000)] 
Don't use absolute paths to headers (#3728)

9 years agoAdd Haiku (#3727)
Simon Marlow [Mon, 21 Dec 2009 11:05:54 +0000 (11:05 +0000)] 
Add Haiku (#3727)

9 years ago#include <fcntl.h>, not <sys/fcntl.h> (#3723)
Simon Marlow [Wed, 16 Dec 2009 10:41:54 +0000 (10:41 +0000)] 
#include <fcntl.h>, not <sys/fcntl.h> (#3723)

9 years agoforking works in GHCi too
Simon Marlow [Wed, 11 Nov 2009 14:20:55 +0000 (14:20 +0000)] 
forking works in GHCi too
Or at least, it works well enough to run this test.  The main GHCi
thread is gone after forking, but the current evaluation continues to
run.

9 years agoadd a test for #1185
Simon Marlow [Wed, 21 Oct 2009 09:07:41 +0000 (09:07 +0000)] 
add a test for #1185

9 years agorun fdReadBuf only the threaded ways (fixed unregisterised failure)
Simon Marlow [Thu, 8 Oct 2009 11:01:26 +0000 (11:01 +0000)] 
run fdReadBuf only the threaded ways (fixed unregisterised failure)

9 years agoAdd a test from trac #2969
Ian Lynagh [Wed, 23 Sep 2009 19:48:37 +0000 (19:48 +0000)] 
Add a test from trac #2969

9 years agoBump version to 2.4.0.0
Ian Lynagh [Sun, 20 Sep 2009 14:20:01 +0000 (14:20 +0000)] 
Bump version to 2.4.0.0

9 years agoAdd comments about why user001 might fail on Linux.
Simon Marlow [Thu, 17 Sep 2009 13:40:46 +0000 (13:40 +0000)] 
Add comments about why user001 might fail on Linux.
See  http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=466647

9 years agoAllow building with base 4.2
Ian Lynagh [Fri, 11 Sep 2009 16:07:36 +0000 (16:07 +0000)] 
Allow building with base 4.2

9 years agoremove dead code
Simon Marlow [Tue, 8 Sep 2009 14:46:27 +0000 (14:46 +0000)] 
remove dead code

9 years agoUse Foreign.Concurrent for Haskell finalizers (#3473)
Simon Marlow [Tue, 8 Sep 2009 14:25:36 +0000 (14:25 +0000)] 
Use Foreign.Concurrent for Haskell finalizers (#3473)

9 years agoUpdate to follow RTS tidyp changes
Simon Marlow [Sat, 1 Aug 2009 19:34:59 +0000 (19:34 +0000)] 
Update to follow RTS tidyp changes

9 years agoNetBSD does not have support for symbol versioning, so updated systen
Simon Marlow [Thu, 23 Jul 2009 07:57:25 +0000 (07:57 +0000)] 
NetBSD does not have support for symbol versioning, so updated systen
functions need to be given a new name, and the header files contain
some __asm hackery in order to let the program call the correct function.

This mean that you need to use the header files in order to call the
correct system functions, which prevents things like "foreign import ccall" from working.

Ghc solves this with wrapper functions for some of the renamed functions,
but it has not been updated for newer versions of NetBSD that has recently
versioned some more functions.

The attached patches introduces wrapper functions for all currently
NetBSD-versioned functions used in libraries/unix.  Solves ~20 testsuite
failures.

Contributed by: Krister Walfridsson <krister.walfridsson@gmail.com>

9 years agoFix warning
Ian Lynagh [Sat, 11 Jul 2009 11:27:01 +0000 (11:27 +0000)] 
Fix warning

9 years agoFix some "warn-unused-do-bind" warnings where we want to ignore the value
Ian Lynagh [Fri, 10 Jul 2009 20:08:06 +0000 (20:08 +0000)] 
Fix some "warn-unused-do-bind" warnings where we want to ignore the value

9 years agoRemove unused imports
Ian Lynagh [Tue, 7 Jul 2009 11:59:00 +0000 (11:59 +0000)] 
Remove unused imports

9 years agoMove directory stuff from base to here
Simon Marlow [Thu, 25 Jun 2009 09:32:58 +0000 (09:32 +0000)] 
Move directory stuff from base to here
leaving out Windows-specific hacks

10 years agorename cache variables to keep recent autoconfs happy
Ross Paterson [Tue, 16 Jun 2009 14:06:11 +0000 (14:06 +0000)] 
rename cache variables to keep recent autoconfs happy

10 years agoDecouple from base (almost) 2009-06-25
Simon Marlow [Thu, 18 Jun 2009 13:20:52 +0000 (13:20 +0000)] 
Decouple from base (almost)
by copying foreign imports here from System.Posix.Internals

10 years agoUpdate to work with the new GHC IO library internals
Simon Marlow [Thu, 11 Jun 2009 14:18:33 +0000 (14:18 +0000)] 
Update to work with the new GHC IO library internals

10 years agoadd test for fdReadBuf/fdWriteBuf
Simon Marlow [Fri, 29 May 2009 12:56:09 +0000 (12:56 +0000)] 
add test for fdReadBuf/fdWriteBuf

10 years agoadd fdReadBuf, fdWriteBuf
Simon Marlow [Fri, 29 May 2009 12:19:41 +0000 (12:19 +0000)] 
add fdReadBuf, fdWriteBuf

-- | Read data from an 'Fd' into memory.  This is exactly equivalent
-- to the POSIX @read@ function.
fdReadBuf :: Fd
          -> Ptr Word8    -- ^ Memory in which to put the data
          -> ByteCount    -- ^ Maximum number of bytes to read
          -> IO Bytecount -- ^ Number of bytes read (zero for EOF)

-- | Write data from memory to an 'Fd'.  This is exactly equivalent
-- to the POSIX @write@ function.
fdWriteBuf :: Fd
           -> Ptr Word8    -- ^ Memory containing the data to write
           -> ByteCount    -- ^ Maximum number of bytes to write
           -> IO ByteCount -- ^ Number of bytes written

10 years agoGive bounds for the base dependency; fixes #3142
Ian Lynagh [Sun, 24 May 2009 15:55:04 +0000 (15:55 +0000)] 
Give bounds for the base dependency; fixes #3142

10 years agoMake get{Group,User}EntryBy{ID,Name} more portable.
Matthias Kilian [Sun, 29 Mar 2009 14:42:52 +0000 (14:42 +0000)] 
Make get{Group,User}EntryBy{ID,Name} more portable.
Retry with a larger buffer whenever getgrgid_r(3), getgrnam_r(3),
getpwuid_r(3) or getpwnam_r(3) return ERANGE. Suggested in the
examples sections of IEEE Std 1003.1-2008.

While here, change the default for grBufSize back to 1024.

10 years agofix this test: we were overflowing the IO manager's pipe with too many signals
Simon Marlow [Tue, 10 Mar 2009 09:09:16 +0000 (09:09 +0000)] 
fix this test: we were overflowing the IO manager's pipe with too many signals

10 years agoAdd config.guess config.sub install-sh as extra-source-files
Ian Lynagh [Sat, 7 Mar 2009 16:19:11 +0000 (16:19 +0000)] 
Add config.guess config.sub install-sh as extra-source-files

10 years agoTweak an internal detail
Ian Lynagh [Wed, 4 Mar 2009 18:28:36 +0000 (18:28 +0000)] 
Tweak an internal detail
We now use an EmptyDataDecl rather than recursive newtype as an
argument to Ptr. As well as being prettier, this also avoids an infinite
loop bug in haddock (trac #3066).

10 years agoRemove an incorrect comment
Ian Lynagh [Wed, 4 Mar 2009 16:25:31 +0000 (16:25 +0000)] 
Remove an incorrect comment

10 years agoRemove some debugging CPP
Ian Lynagh [Thu, 26 Feb 2009 00:16:36 +0000 (00:16 +0000)] 
Remove some debugging CPP

10 years agoRewrite of signal-handling.
Simon Marlow [Thu, 19 Feb 2009 10:05:32 +0000 (10:05 +0000)] 
Rewrite of signal-handling.

The API is the same (for now).  The new implementation has the
capability to define signal handlers that have access to the siginfo
of the signal (#592), but this functionality is not exposed in this
patch.

#2451 is the ticket for the new API.

The main purpose of bringing this in now is to fix race conditions in
the old signal handling code (#2858).  Later we can enable the new
API in the HEAD.

Implementation differences:

 - More of the signal-handling is moved into Haskell.  We store the
   table of signal handlers in an MVar, rather than having a table of
   StablePtrs in the RTS.

 - In the threaded RTS, the siginfo of the signal is passed down the
   pipe to the IO manager thread, which manages the business of
   starting up new signal handler threads.  In the non-threaded RTS,
   the siginfo of caught signals is stored in the RTS, and the
   scheduler starts new signal handler threads.

10 years agoDon't put inline'd functions in HsUnix.h; fixes trac #2969
Ian Lynagh [Wed, 11 Feb 2009 18:29:06 +0000 (18:29 +0000)] 
Don't put inline'd functions in HsUnix.h; fixes trac #2969
If they are included into a C file which also has certain symbols
defined, then the behaviour of the HsUnix.h functions can change
(e.g. lstat can become the 32bit, rather than 64bit, version).

10 years agofix warnings
Simon Marlow [Tue, 3 Feb 2009 10:02:54 +0000 (10:02 +0000)] 
fix warnings

10 years agoAdd check for -lrt to get the shm* functions. Subst. in buildinfo
Don Stewart [Fri, 30 Jan 2009 11:35:02 +0000 (11:35 +0000)] 
Add check for -lrt to get the shm* functions. Subst. in buildinfo

10 years agoSharedMem.hsc wasn't including HsUnixConfig.h, so no #defines were propagating
Don Stewart [Fri, 30 Jan 2009 11:34:51 +0000 (11:34 +0000)] 
SharedMem.hsc wasn't including HsUnixConfig.h, so no #defines were propagating

10 years agoRequire Cabal version >= 1.6
Ian Lynagh [Thu, 22 Jan 2009 01:13:31 +0000 (01:13 +0000)] 
Require Cabal version >= 1.6

10 years agoAdd "bug-reports" and "source-repository" info to the Cabal file
Ian Lynagh [Wed, 21 Jan 2009 18:28:42 +0000 (18:28 +0000)] 
Add "bug-reports" and "source-repository" info to the Cabal file
Also switched to the modern Cabal file format

10 years agogeneralise type of executeFile (#2948)
Simon Marlow [Wed, 14 Jan 2009 12:47:26 +0000 (12:47 +0000)] 
generalise type of executeFile (#2948)

10 years agoAvoid using IOError internals
Ian Lynagh [Sun, 4 Jan 2009 17:32:21 +0000 (17:32 +0000)] 
Avoid using IOError internals

10 years agofix pthread linkage problem for openbsd
Matthias Kilian [Sat, 29 Nov 2008 00:06:38 +0000 (00:06 +0000)] 
fix pthread linkage problem for openbsd
This should make my openbsd build slave happy when SplitObjs=NO.
May be useful for other BSDs and even Linux, regardless wether you
need -pthread or -lpthread. Time will tell...