Ticket #3550: ghc-macosx-dylibs-v4.darcs-send

Tue Sep 29 09:34:00 NZDT 2009  Stephen Blackheath <oversensitive.pastors.stephen@blacksapphire.com>
  * Add -install-name option to GHC so the install name can be set for dynamic libs on Mac OS/X.

Tue Sep 29 09:38:00 NZDT 2009  Stephen Blackheath <oversensitive.pastors.stephen@blacksapphire.com>
  * Force -fPIC when linking against dynamic libraries on Mac OS/X.
  Otherwise you get
  /tmp/ghc7602_0/ghc7602_0.s:207:0:
     non-relocatable subtraction expression, "___stginit_Lib_dyn" minus "L1x2;4"
  /tmp/ghc7602_0/ghc7602_0.s:207:0:
     symbol: "___stginit_Lib_dyn" can't be undefined in a subtraction expression

Tue Sep 29 09:41:58 NZDT 2009  Stephen Blackheath <oversensitive.pastors.stephen@blacksapphire.com>
  * Link all dynamic libraries with the correct install_name on Mac OS/X.

New patches:

[Add -install-name option to GHC so the install name can be set for dynamic libs on Mac OS/X.
Stephen Blackheath <oversensitive.pastors.stephen@blacksapphire.com>**20090928203400
 Ignore-this: 4cb302364ead896ab95cf3bb6fa1c258
] {
hunk ./compiler/main/DriverPipeline.hs 1624
     --  later, so that it will not complain about the use of the option
     --  -undefined dynamic_lookup above.
     -- -install_name
-    --   Causes the dynamic linker to ignore the DYLD_LIBRARY_PATH when loading
-    --   this lib and instead look for it at its absolute path.
-    --   When installing the .dylibs (see target.mk), we'll change that path to
-    --   point to the place they are installed. Therefore, we won't have to set
-    --  up DYLD_LIBRARY_PATH specifically for ghc.
+    --   Mac OS/X stores the path where a dynamic library is (to be) installed
+    --   in the library itself.  It's called the "install name" of the library.
+    --   Then any library or executable that links against it before it's
+    --   installed will search for it in its ultimate install location.  By
+    --   default we set the install name to the absolute path at build time, but
+    --   it can be overridden by the -install-name option passed to ghc.  Cabal
+    --   does this.
     -----------------------------------------------------------------------------
 
     let output_fn = case o_file of { Just s -> s; Nothing -> "a.out"; }
hunk ./compiler/main/DriverPipeline.hs 1635
 
-    pwd <- getCurrentDirectory
+    instName <- case installName dflags of
+        Just n -> return n
+        Nothing -> do
+            pwd <- getCurrentDirectory
+            return $ pwd `combine` output_fn
     SysTools.runLink dflags
         ([ SysTools.Option verb
          , SysTools.Option "-dynamiclib"
hunk ./compiler/main/DriverPipeline.hs 1649
         ++ map SysTools.Option (
            md_c_flags
         ++ o_files
-        ++ [ "-undefined", "dynamic_lookup", "-single_module", "-Wl,-macosx_version_min","-Wl,10.5", "-install_name " ++ (pwd </> output_fn) ]
+        ++ [ "-undefined", "dynamic_lookup", "-single_module", "-Wl,-macosx_version_min","-Wl,10.5",
+              "-Wl,-read_only_relocs,suppress", "-install_name", instName ]
         ++ extra_ld_inputs
         ++ lib_path_opts
         ++ extra_ld_opts
hunk ./compiler/main/DynFlags.hs 379
 
   -- paths etc.
   objectDir             :: Maybe String,
+  installName           :: Maybe String,
   hiDir                 :: Maybe String,
   stubDir               :: Maybe String,
 
hunk ./compiler/main/DynFlags.hs 621
         thisPackage             = mainPackageId,
 
         objectDir               = Nothing,
+        installName             = Nothing,
         hiDir                   = Nothing,
         stubDir                 = Nothing,
 
hunk ./compiler/main/DynFlags.hs 764
   | verbosity dflags >= 3  = "-v"
   | otherwise =  ""
 
-setObjectDir, setHiDir, setStubDir, setOutputDir,
+setObjectDir, setHiDir, setStubDir, setOutputDir, setInstallName,
          setObjectSuf, setHiSuf, setHcSuf, parseDynLibLoaderMode,
          setPgmP, setPgmL, setPgmF, setPgmc, setPgmm, setPgms, setPgma, setPgml, setPgmdll, setPgmwindres,
          addOptL, addOptP, addOptF, addOptc, addOptm, addOpta, addOptl, addOptwindres,
hunk ./compiler/main/DynFlags.hs 779
   -- -stubdir D adds an implicit -I D, so that gcc can find the _stub.h file
   -- \#included from the .hc file when compiling with -fvia-C.
 setOutputDir  f = setObjectDir f . setHiDir f . setStubDir f
+setInstallName  f d = d{ installName = Just f}
 
 setObjectSuf  f d = d{ objectSuf  = f}
 setHiSuf      f d = d{ hiSuf      = f}
hunk ./compiler/main/DynFlags.hs 1324
   , Flag "tmpdir"         (HasArg (upd . setTmpDir)) Supported
   , Flag "stubdir"        (HasArg (upd . setStubDir)) Supported
   , Flag "outputdir"      (HasArg (upd . setOutputDir)) Supported
+  , Flag "install-name"   (HasArg (upd . setInstallName)) Supported
   , Flag "ddump-file-prefix" (HasArg (upd . setDumpPrefixForce . Just))
          Supported
 
}
[Force -fPIC when linking against dynamic libraries on Mac OS/X.
Stephen Blackheath <oversensitive.pastors.stephen@blacksapphire.com>**20090928203800
 Ignore-this: 465433af2349779b510f500dc79768f3
 Otherwise you get
 /tmp/ghc7602_0/ghc7602_0.s:207:0:
    non-relocatable subtraction expression, "___stginit_Lib_dyn" minus "L1x2;4"
 /tmp/ghc7602_0/ghc7602_0.s:207:0:
    symbol: "___stginit_Lib_dyn" can't be undefined in a subtraction expression
] hunk ./compiler/main/StaticFlags.hs 268
 opt_PIC :: Bool
 #if darwin_TARGET_OS && x86_64_TARGET_ARCH
 opt_PIC                         = True
+#elif darwin_TARGET_OS
+opt_PIC                         = lookUp (fsLit "-fPIC") || not opt_Static
 #else
 opt_PIC                         = lookUp (fsLit "-fPIC")
 #endif
[Link all dynamic libraries with the correct install_name on Mac OS/X.
Stephen Blackheath <oversensitive.pastors.stephen@blacksapphire.com>**20090928204158
 Ignore-this: aef6dfa45d84bd7f4e4358df19490907
] 
<
[We now tell the linker macosx_version_min is 10.5; trac #3521
Ian Lynagh <igloo@earth.li>**20090924162450
 We used to say 10.3, but this gives
     -rpath can only be used when targeting Mac OS X 10.5 or later 
 when building shared libraries.
 Patch from mwotton.
] 
[Add -install-name option to GHC so the install name can be set for dynamic libs on Mac OS/X.
Stephen Blackheath <oversensitive.pastors.stephen@blacksapphire.com>**20090928203400
 Ignore-this: 4cb302364ead896ab95cf3bb6fa1c258
] 
> {
hunk ./ghc.mk 717
        $(INSTALL_DIR) $(DESTDIR)$(bindir)
        for i in $(INSTALL_BINS); do \
                $(INSTALL_PROGRAM) $(INSTALL_BIN_OPTS) $$i $(DESTDIR)$(bindir) ;  \
-                if test "$(darwin_TARGET_OS)" = "1"; then \
-                   sh mk/fix_install_names.sh $(ghclibdir) $(DESTDIR)$(bindir)/$$i ; \
-                fi ; \
        done
 
 install_libs: $(INSTALL_LIBS)
hunk ./ghc.mk 731
                  *.so) \
                    $(INSTALL_SHLIB) $(INSTALL_OPTS) $$i $(DESTDIR)$(ghclibdir) ;; \
                  *.dylib) \
-                   $(INSTALL_SHLIB) $(INSTALL_OPTS) $$i $(DESTDIR)$(ghclibdir); \
-                   install_name_tool -id $(DESTDIR)$(ghclibdir)/`basename $$i` $(DESTDIR)$(ghclibdir)/`basename $$i` ;; \
+                   $(INSTALL_SHLIB) $(INSTALL_OPTS) $$i $(DESTDIR)$(ghclibdir) ;; \
                  *) \
                    $(INSTALL_DATA) $(INSTALL_OPTS) $$i $(DESTDIR)$(ghclibdir); \
                esac; \
hunk ./ghc.mk 854
        $(wildcard libraries/*/dist-install/doc/) \
     $(filter-out extra-gcc-opts,$(INSTALL_LIBS)) \
     $(filter-out %/project.mk mk/config.mk %/mk/install.mk,$(MAKEFILE_LIST)) \
-       mk/fix_install_names.sh \
        mk/project.mk \
        mk/install.mk.in \
        bindist.mk \
hunk ./libffi/ghc.mk 85
 libffi_DYNAMIC_LIBS = $(libffi_HS_DYN_LIB)
 else
 libffi_DYNAMIC_PROG =
+ifeq "$(darwin_TARGET_OS)" "1"
+libffi_DYNAMIC_LIBS = libffi/libffi$(soext) libffi/libffi.5$(soext) libffi/libffi.5.0.9$(soext)
+else
 libffi_DYNAMIC_LIBS = libffi/libffi.so libffi/libffi.so.5 libffi/libffi.so.5.0.9
 endif
hunk ./libffi/ghc.mk 90
+endif
 
 ifeq "$(BuildSharedLibs)" "YES"
 libffi_EnableShared=yes
hunk ./libffi/ghc.mk 196
 # Rename libffi.so to libHSffi...so
 $(libffi_HS_DYN_LIB): $(libffi_DYNAMIC_LIBS)
        "$(CP)" $(word 1,$(libffi_DYNAMIC_LIBS)) $(libffi_HS_DYN_LIB)
+ifeq "$(darwin_TARGET_OS)" "1"
+       # Ensure library's install name is correct before anyone links with it.
+       install_name_tool -id $(ghclibdir)/$(libffi_HS_DYN_LIB_NAME) $(libffi_HS_DYN_LIB)
+endif
 
 $(eval $(call all-target,libffi,$(libffi_HS_DYN_LIB)))
 endif
hunk ./mk/fix_install_names.sh 1
-# Darwin / Mac OS X only
-#   sh fix_install_names.sh directory binary
-#
-# Changes 'binary' to assume that all libHS*_dyn.dylib libraries
-# are to be found in 'directory'.
-
-prefix=$1
-file=$2
-
-type=`file "$file"`
-
-if `test "${type/Mach-O}" == "$type"`
-then
-    exit
-fi
-
-if `test x${prefix%/} != x"" `
-then
-    prefix=${prefix%/}/
-fi
-
-for i in `otool -L $file \
-         | grep 'libHS.*_dyn.dylib' \
-         | sed 's/.\(.*libHS.*_dyn.dylib\).*/\1/'`
-do
-    install_name_tool -change $i "$prefix`basename $i`" $file
-done
-
-if `test "${file%.dylib}" != "${file}"`
-then
-    install_name_tool -id "$prefix`basename $file`" $file
-fi
rmfile ./mk/fix_install_names.sh
hunk ./rts/ghc.mk 101
 $(call c-suffix-rules,rts,dist,$1,YES)
 $(call cmm-suffix-rules,rts,dist,$1)
 
-rts_$1_LIB = rts/dist/build/libHSrts$$($1_libsuf)
+rts_$1_LIB_NAME = libHSrts$$($1_libsuf)
+rts_$1_LIB = rts/dist/build/$$(rts_$1_LIB_NAME)
 
 rts_$1_C_OBJS   = $$(patsubst rts/%.c,rts/dist/build/%.$$($1_osuf),$$(rts_C_SRCS)) $$(patsubst %.c,%.$$($1_osuf),$$(rts_$1_EXTRA_C_SRCS))
 rts_$1_S_OBJS   = $$(patsubst rts/%.S,rts/dist/build/%.$$($1_osuf),$$(rts_S_SRCS))
hunk ./rts/ghc.mk 117
        "$$(RM)" $$(RM_OPTS) $$@
        "$$(rts_dist_HC)" -shared -dynamic -dynload deploy \
          -no-auto-link-packages `cat rts/libs.depend` $$(rts_$1_OBJS) -o $$@
+ifeq "$(darwin_TARGET_OS)" "1"
+       # Ensure library's install name is correct before anyone links with it.
+       install_name_tool -id $(ghclibdir)/$$(rts_$1_LIB_NAME) $$@
+endif
 else
 $$(rts_$1_LIB) : $$(rts_$1_OBJS)
        "$$(RM)" $$(RM_OPTS) $$@
hunk ./rules/build-package-way.mk 50
        "$$($1_$2_HC)" $$($1_$2_$3_ALL_OBJS) \
          `$$($1_$2_$3_MKSTUBOBJS)` \
          -shared -dynamic -dynload deploy \
+         -install-name $(ghclibdir)/`basename "$$@" | sed 's/^libHS//;s/[-]ghc.*//'`/`basename "$$@"` \
          -no-auto-link-packages $$(addprefix -package,$$($1_$2_DEPS)) \
          -o $$@
 else
}

Context:

[emitRetUT: cope with arguments overlapping with results (#3546)
Simon Marlow <marlowsd@gmail.com>**20090928124455
 Ignore-this: 899c4be65a7c31d05bf5e79ea423147e
 
 In decodeFloat_Int# we have the C-- code:
 
     mp_tmp1  = Sp - WDS(1);
     mp_tmp_w = Sp - WDS(2);
     
     /* arguments: F1 = Float# */
     arg = F1;
     
     /* Perform the operation */
     foreign "C" __decodeFloat_Int(mp_tmp1 "ptr", mp_tmp_w "ptr", arg) [];
     
     /* returns: (Int# (mantissa), Int# (exponent)) */
     RET_NN(W_[mp_tmp1], W_[mp_tmp_w]);
 
 Which all looks quite reasonable.  The problem is that RET_NN() might
 assign the results to the stack (with an unregisterised back end), and
 in this case the arguments to RET_NN() refer to the same stack slots
 that will be assigned to.
 
 The code generator should do the right thing here, but it wasn't - it
 was assuming that it could assign the results sequentially.  A 1-line
 fix to use emitSimultaneously rather than emitStmts (plus comments).
] 
[fix cut-and-paste bugs in .cmm -> .hc/.s rules
Simon Marlow <marlowsd@gmail.com>**20090928115139
 Ignore-this: fe9a834cb94b53c7d4d6dc238c0d348d
] 
[Add a way to generate tracing events programmatically
Simon Marlow <marlowsd@gmail.com>**20090925150243
 Ignore-this: 7f17f3474b06a1fb4a527c62e003e7aa
 
 added:
 
  primop  TraceEventOp "traceEvent#" GenPrimOp
    Addr# -> State# s -> State# s
    { Emits an event via the RTS tracing framework.  The contents
      of the event is the zero-terminated byte string passed as the first
      argument.  The event will be emitted either to the .eventlog file,
      or to stderr, depending on the runtime RTS flags. }
 
 and added the required RTS functionality to support it.  Also a bit of
 refactoring in the RTS tracing code.
] 
[Under utils/ext-core, update README to mention that ext-core is now on Hackage
Tim Chevalier <chevalier@alum.wellesley.edu>**20090927230845
 Ignore-this: 8d8d2cad45c07931f0ed1ccfed002d66
] 
[Update the OS X package build scripts
Ian Lynagh <igloo@earth.li>**20090925132151
 The docs are no longer built and installed separately.
] 
[Use $topdir in the RTS's package.conf file when doing a relocatable build
Ian Lynagh <igloo@earth.li>**20090927145309] 
[Fix invoking windres in a directory containing spaces
Ian Lynagh <igloo@earth.li>**20090927144119] 
[Add support for relocatable builds in the new build system
Ian Lynagh <igloo@earth.li>**20090927010605] 
[Fix building of the RTS with the NCG under Windows
Ben.Lippmeier@anu.edu.au**20090926075507] 
[Make some sed more portable: Use \{0,1\} instead of \?
Ian Lynagh <igloo@earth.li>**20090924171025] 
[teach boot-pkgs about -snapshot tarballs
Ian Lynagh <igloo@earth.li>**20090921203904] 
[We now tell the linker macosx_version_min is 10.5; trac #3521
Ian Lynagh <igloo@earth.li>**20090924162450
 We used to say 10.3, but this gives
     -rpath can only be used when targeting Mac OS X 10.5 or later 
 when building shared libraries.
 Patch from mwotton.
] 
[Errors talking about a left section should refer to the "first" argument
Ian Lynagh <igloo@earth.li>**20090924124128
 rather than the "second" argument. Fixes trac #3505.
] 
[Make count_lines work with the new directory layout
Ian Lynagh <igloo@earth.li>**20090923182251] 
[Add count_lines back into the repo
Ian Lynagh <igloo@earth.li>**20090923174945] 
[Remove a haskeline release note that doesn't apply to ghci
Ian Lynagh <igloo@earth.li>**20090923145130
 Spotted by Judah.
] 
[fix warnings
Simon Marlow <marlowsd@gmail.com>**20090923125059
 Ignore-this: 6cc0a727c7a55a5bf79b0190c47666de
] 
[Add erf, erfc, erfff, erfcf (#3536)
Simon Marlow <marlowsd@gmail.com>**20090923105240
 Ignore-this: 4c90186bf779bec6b9c72e2989a1876a
] 
[Various updates/additions
Simon Marlow <marlowsd@gmail.com>**20090923094647
 Ignore-this: 7551a50a3f7723f21bc8c17e4c827764
] 
[Document -package-id, and use the term "package ID" consistently
Simon Marlow <marlowsd@gmail.com>**20090923094639
 Ignore-this: 234a5a2840b7a9608c9a5d4bc475080f
] 
[Make it so that -Ds implies -ls
Simon Marlow <marlowsd@gmail.com>**20090923105119
 Ignore-this: 52662a8e69ff4b8e8d9a07e4ca3e181
] 
[Don't build haddock if HADDOC_DOCS = NO, and disable HADDOC_DOCS if GhcWithInterpreter = NO
Matthias Kilian <kili@outback.escape.de>**20090920181319
 Ignore-this: 1268ae31d74e746b09287814ee435f65
 Haddock uses TcRnDriver.tcRnGetInfo, which is only available if
 GHCI is built. Set HADDOC_DOCS to NO if GhcWithInterpreter is NO,
 and disable the haddock build if HADDOC_DOCS = NO.
] 
[Remove syb; it is no longer needed
Ian Lynagh <igloo@earth.li>**20090922222308] 
[Remove base3-compat
Ian Lynagh <igloo@earth.li>**20090922200657] 
[clean some Haddock-generated files
Simon Marlow <marlowsd@gmail.com>**20090921114210
 Ignore-this: a2569ec63c9aa6b975345872ba4f2f2a
] 
[Set version number to 6.13
Ian Lynagh <igloo@earth.li>**20090922152400] 
[boot-pkgs now automatically works out which libraries are from tarballs
Ian Lynagh <igloo@earth.li>**20090921165559] 
[remove "touch %_hsc.c" - what was it for?
Simon Marlow <marlowsd@gmail.com>**20090921155720
 Ignore-this: 48d5d7ad35e1f2bdcebf4e197d46c609
] 
[Validate now cleans regardless of whether mk/config.mk exists
Ian Lynagh <igloo@earth.li>**20090921182636
 Cleaning should always succeed now
] 
[Update 6.12.1 release notes
Ian Lynagh <igloo@earth.li>**20090921182431
 Add changes that we inherit from haskeline
] 
[Add sdist to ghc-cabal's bag of tricks
Ian Lynagh <igloo@earth.li>**20090921162003] 
[Add the --checked-out flag; expand the docs/comments at the top
Simon Marlow <marlowsd@gmail.com>**20090921145201
 Ignore-this: f714387e94beb5e03ce2ec5432bca38e
] 
[move way-related stuff into a separate file
Simon Marlow <marlowsd@gmail.com>**20090921130024
 Ignore-this: f75bfb3b60ee92d6add3e35b000da60d
 This fixes at least one problem: libHSffi-ghc<version>.so wasn't being
 cleaned, because $(dyn_libsuf) wasn't defined while cleaning.
] 
[Fix bindist creation
Ian Lynagh <igloo@earth.li>**20090920183631] 
[Update dependencies
Ian Lynagh <igloo@earth.li>**20090920152402] 
[Add release notes for 6.12.1
Ian Lynagh <igloo@earth.li>**20090920141820] 
[remove encoding of output using Haskeline; the IO library does it now (#3398)
Simon Marlow <marlowsd@gmail.com>**20090918144041
 Ignore-this: 155e3d4253462af10874882ca2d0b2e
] 
[ignore unused-import warnings from Haskeline
Simon Marlow <marlowsd@gmail.com>**20090918142718
 Ignore-this: 619c5c1d572842b0b19268f87dffe75a
] 
[Fix the nursery size calculation when -N is used with -H
Simon Marlow <marlowsd@gmail.com>**20090918134438
 Ignore-this: 182b62d1c9fc1568fe453e375cb34bc4
] 
[implement case-on-Word in the byte code generator/interpreter (#2881)
Simon Marlow <marlowsd@gmail.com>**20090918133204
 Ignore-this: f8c29e1cdb4fb70e9d9ba6230a5531dc
] 
[Fix #3439: -debug implies -ticky, and -ticky code links with any RTS
Simon Marlow <marlowsd@gmail.com>**20090918114631
 Ignore-this: 2bc27f861f33a5db471539b0d13618b4
] 
[Fix #3525 - we were still checking for ParsePkgConf.hs, which is gone
Simon Marlow <marlowsd@gmail.com>**20090918104327
 Ignore-this: d461a9cf7213c292bf2021edfadf08fc
] 
[improve the cyclic module error message as per #1856
Simon Marlow <marlowsd@gmail.com>**20090916205036
 Ignore-this: 2e679e7f6f44e613717742cc2edee10b
 
  - remove the main: stuff
  - show the filenames
  - don't clutter the output with imports that aren't involved in
    the cycle
] 
[Use CONF_CC_OPTS
Manuel M T Chakravarty <chak@cse.unsw.edu.au>**20090917132706
 Ignore-this: 3dd1a67c04f8bb8e10c161c7c5549753
 - Needed to get arch-dependent options, eg, on Snow Leopard
] 
[fix install on Windows
Simon Marlow <marlowsd@gmail.com>**20090917121921
 Ignore-this: e94a2d765b618e3adc9e2af7bfaf8559
] 
[fix install on Windows
Simon Marlow <marlowsd@gmail.com>**20090917103350
 Ignore-this: 513bb0336183cd23535e5d1c6d7e2d58
] 
[Allow -package-id to override the default package shadowing semantics
Simon Marlow <marlowsd@gmail.com>**20090917121510
 Ignore-this: 30d211e9989085f1de40552569b544b9
 So that Cabal, if it wants, can use a more general algorithm to find a
 consistent set of packages to use.
] 
[fix xref
Simon Marlow <marlowsd@gmail.com>**20090917100544
 Ignore-this: 1bdfc1c3cd79de5be808e9ec2c8d3542
] 
[add -eventlog
Simon Marlow <marlowsd@gmail.com>**20090917100403
 Ignore-this: bb3f722cb2becfca08a971c927e8ea28
] 
[document -eventlog and the associated runtime options
Simon Marlow <marlowsd@gmail.com>**20090916152815
 Ignore-this: 3efd91b35ac73487d94b887fdd85a04d
] 
[typo in error message
Simon Marlow <marlowsd@gmail.com>**20090916142324
 Ignore-this: 37bda5b3aac4c383e21f1c26ab054d0b
] 
[only create the .eventlog file if any tracing options are enabled
Simon Marlow <marlowsd@gmail.com>**20090916142315
 Ignore-this: 144e9261e008e9e8b641c596d9892fd2
] 
[Fix build on Mac OS 10.6 (Snow Leopard)
Manuel M T Chakravarty <chak@cse.unsw.edu.au>**20090917044121
 Ignore-this: 16fbdb0fc58acf3d1497da4a8b3b516e
 - We have -m32 as machine-dependent option for gcc for a 32 bit build
 - Like on OpenBSD, SL requires -fno-stack-protector to avoid triggering the
   stack smashing checks inserted by gcc by default on this platform.
] 
[Follow the builtin:rts vs. builtin_rts renaming
Matthias Kilian <kili@outback.escape.de>**20090915204254
 Ignore-this: d205dfce1a912796c355ba1c1e660d7a
 
 This unbreaks make install again.
 
] 
[Put context information for warnings in errMsgExtraInfo.
Thomas Schilling <nominolo@googlemail.com>**20090915150353
 
 For type checker warnings, the context information ("In the expression
 ...") was simply appended to the main message while for proper errors
 they live in errMsgExtraInfo.  This allows GHC API clients to drop
 that information if not needed.
] 
[Update and improve packages documentation
Simon Marlow <marlowsd@gmail.com>**20090915161024
 Ignore-this: a0253dbee88fe3fe1a0a87e6e1f67bf
] 
[FIX #3434 (improve vi tags: add non-exported symbols, kinds, regex tags)
Peter Hercek <phercek@gmail.com>**20090817184100
 Ignore-this: 929dd59429be967bb3c9ec4409f6f850
] 
[Event tracing: put the capability in the block marker, omit it from the events
Simon Marlow <marlowsd@gmail.com>**20090915111227
 Ignore-this: 6c1e336e077f50aca1c3e4c094e033a9
 This makes events smaller and tracing quicker, and speeds up reading
 and sorting the trace file.
 
 HEADS UP: this changes the format of event log files.  Corresponding
 changes to the ghc-events package are required (and will be pushed
 soon).  Normally we would make backwards-compatible changes, but this
 changes the format of every event (to remove the capability) so I'm
 breaking the rules this time.  This will be the only time we can do
 this, since the format becomes public in 6.12.1.
] 
[Add event block markers
Simon Marlow <marlowsd@gmail.com>**20090913212546
 Ignore-this: f9de0afe4730560fa79a5a8450e08dcf
 These indicate the size and time span of a sequence of events in the
 event log, to make it easier to sort and navigate a large event log.
] 
[Optimise the code generated at trace points
Simon Marlow <marlowsd@gmail.com>**20090912213504
 Ignore-this: eaad8b3a8f204f0cbfed31d7e25672d1
] 
[Improve the default parallel GC settings, and sanitise the flags (#3340)
Simon Marlow <marlowsd@gmail.com>**20090915084000
 Ignore-this: 1aec6ccccb26ed80286de468e72657a1
 
 Flags (from +RTS -?):
 
   -qg[<n>]  Use parallel GC only for generations >= <n>
             (default: 0, -qg alone turns off parallel GC)
   -qb[<n>]  Use load-balancing in the parallel GC only for generations >= <n>
             (default: 1, -qb alone turns off load-balancing)
 
 these are good defaults for most parallel programs.  Single-threaded
 programs that want to make use of parallel GC will probably want +RTS
 -qg1 (this is documented).
 
 I've also updated the docs.
] 
[Update the "Hints for using SMP parallelism" section
Simon Marlow <marlowsd@gmail.com>**20090915083431
 Ignore-this: e0147b4b025b356e9395b47f085002b7
 In particular, remove the claim that the GC is single-threaded!
] 
[Fix collateral damage after distdir-opts.mk was refactored away
Simon Marlow <marlowsd@gmail.com>**20090914144427
 Ignore-this: bc47d134ce4cdd47e664bf4cbb28d7bd
] 
[alow macros to redefine builtin GHCi commands (implements #3084)
Peter Hercek <phercek@gmail.com>**20090512172459
 Ignore-this: 62f3aa596e922bd1cb8c4875ca674570
] 
[Fix build on Windows
Simon Marlow <marlowsd@gmail.com>**20090914110550
 Ignore-this: b2aaa2183edfd175d23ea3b0dd8dd2b0
] 
[Add "ghc-pkg init" command for creating a new package DB
Simon Marlow <marlowsd@gmail.com>**20090914105823
 Ignore-this: 1f788fb090f28c0ac005695fb79b3c7f
] 
[Fix a space leak in the native code gen (again)
Simon Marlow <marlowsd@gmail.com>**20090911152812
 Ignore-this: 8c678698faf4f3cfa4be5c923fb4f6a2
] 
[Use _ rather than : for the builtin package filenames
Ian Lynagh <igloo@earth.li>**20090913204016
 Using "builtin:foo" creates a file called "builtin" on Windows
] 
[Don't register the non-munged stage1 ghc package
Ian Lynagh <igloo@earth.li>**20090913161945
 It confuses the build system.
] 
[Fixes for compiling the HEAD with iteslf
Ian Lynagh <igloo@earth.li>**20090913142122] 
[Don't try and put HaddockLex and HaddockParse into src dists
Ian Lynagh <igloo@earth.li>**20090912175050
 They no longer exist.
] 
[ParsePkgConf no longer exists, so don't try to put it in source dists
Ian Lynagh <igloo@earth.li>**20090912173253] 
[Tweak an error message
Ian Lynagh <igloo@earth.li>**20090912173223] 
[Refactor sdist so that we can see which command is being run
Ian Lynagh <igloo@earth.li>**20090912171627] 
[Fix the location of the package database in the installed ghc-pkg
Ian Lynagh <igloo@earth.li>**20090912144046
 It was using the old filename, rather than the new directory name.
] 
[More bindist fixes
Ian Lynagh <igloo@earth.li>**20090912143248] 
[Add includes/ghcautoconf.h and includes/ghcplatform.h to bindists
Ian Lynagh <igloo@earth.li>**20090912132842] 
[Update the URL in the usage info to point to the current docs location
Ian Lynagh <igloo@earth.li>**20090911220735] 
[move hschooks.c to ghc/, and enable it
Simon Marlow <marlowsd@gmail.com>**20090911140702
 Ignore-this: 97a3cc9b1fa2ce756058c4c708da4dd2
] 
[Fix #3408: lengthen the idle GC time to 5s for GHC/GHCi.
Simon Marlow <marlowsd@gmail.com>**20090911124547
 Ignore-this: 238db99e1fcca0d15a7e5c2a948ac05c
] 
[Comments and Cmm notes
simonpj@microsoft.com**20090911133513
 Ignore-this: 9fd8719a14824dbe2fcb07b655e972cd
] 
[Remove the old package.conf parser, use read instead (fixed #3410)
Simon Marlow <marlowsd@gmail.com>**20090911114928
 Ignore-this: 4f677ceb1f05c55515da6dd5045fa6c5
 Performance isn't an issue for reading the old-style package.conf
 files, so we might as well revert to using read and fix a bug at the
 same time.
] 
[Fix some makefile logic, and remove the now unused *_USE_BOOT_LIBS vars
Ian Lynagh <igloo@earth.li>**20090911123135] 
[Change a sed separator character to make it easier to read
Ian Lynagh <igloo@earth.li>**20090911121827] 
[Use -fno-warn-name-shadowing when compiling Binary.hs
Ian Lynagh <igloo@earth.li>**20090911114756] 
[Improve optimisation of cost centres
simonpj@microsoft.com**20090911115630
 Ignore-this: 9964cfe9d39b12eb10724cd1f3fee6b5
 
 This patch fixes test failures for the profiling way for drv001.
 The problem was that the arity of a function was decreasing during
 "optimisation" because of interaction with SCC annotations.
 In particular
       f = /\a. scc "f" (h x)    -- where h had arity 2
 and h gets inlined, led to
       f = /\a. scc "f" let v = scc "f" x in \y. <blah>
 
 Two main changes:
 
 1.  exprIsTrivial now says True for (scc "f" x)
     See Note [SCCs are trivial] in CoreUtils
 
 2.  The simplifier eliminates nested pushing of the same cost centre:
        scc "f" (...(scc "f" e)...) 
        ==>  scc "f" (...e...)
 
 
 
] 
[Fix build with GHC <= 6.10.1
Simon Marlow <marlowsd@gmail.com>**20090911103142
 Ignore-this: 7a378cd66edf9ada5ea813051a1a7c30
] 
[loosen base dependency to allow base-3
Simon Marlow <marlowsd@gmail.com>**20090911093139
 Ignore-this: 7b07363b4e9003acbfc310a66d0746e8
] 
[remove superfluous trailing backslash
Simon Marlow <marlowsd@gmail.com>**20090911093124
 Ignore-this: e13b6a5ca640c216db477485b92356d7
] 
[Work around bug in old bytestring versions by reading the package DB strictly
Simon Marlow <marlowsd@gmail.com>**20090911093103
 Ignore-this: 96778e3c876a1ab012888368bbcd9777
] 
[Add cmm-notes, describing Simon and John's work on Cmm pipeline
simonpj@microsoft.com**20090911105316
 Ignore-this: f042ad4d869a667d84de9d432e8d96ff
] 
[Wibble to error message
simonpj@microsoft.com**20090911090819
 Ignore-this: 930f47d57a9d33643dd87f19821df55f
] 
[add comment: __thread is not supported by gcc on OS X yet
Simon Marlow <marlowsd@gmail.com>**20090910152146
 Ignore-this: 44abcd5ece4baeb91c9461f3cb35df52
] 
[Fix #3320: we forgot to save/restore the GC register variable
Simon Marlow <marlowsd@gmail.com>**20090910151623
 Ignore-this: 95fe8e80d1977feca7ba61222a992db7
] 
[put back F_, just in case the via-C codegen uses it
Simon Marlow <marlowsd@gmail.com>**20090910124037
 Ignore-this: 4bbfcf2765422a9c965052120b7643fb
] 
[remove debugging/testing code that got here by mistake
Simon Marlow <marlowsd@gmail.com>**20090910122607
 Ignore-this: 76a733fc1d310fc5bf38820b8d925bf1
] 
[fix installation of header files (#3451)
Simon Marlow <marlowsd@gmail.com>**20090910121831
 Ignore-this: d49dcadc1cbd62d7c593b06888aa3d5b
] 
[Three improvements to Template Haskell (fixes #3467)
simonpj@microsoft.com**20090910125848
 Ignore-this: 40e480334a8cd06d128a845cefd4632a
 
 This patch implements three significant improvements to Template Haskell.
   
 Declaration-level splices with no "$" 
 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
 This change simply allows you to omit the "$(...)" wrapper for 
 declaration-level TH splices.  An expression all by itself is
 not legal, so we now treat it as a TH splice.  Thus you can now
 say
        data T = T1 | T2
        deriveMyStuff ''T
 
 where deriveMyStuff :: Name -> Q [Dec]
 This makes a much nicer interface for clients of libraries that use
 TH: no scary $(deriveMyStuff ''T).
 
 Nested top-level splices
 ~~~~~~~~~~~~~~~~~~~~~~~~
 Previously TH would reject this, saying that splices cannot be nested:
        f x = $(g $(h 'x))
 But there is no reason for this not to work.  First $(h 'x) is run,
 yielding code <blah> that is spliced instead of the $(h 'x). Then (g
 <blah>) is typechecked and run, yielding code that replaces the
 $(g ...) splice.  
 
 So this simply lifts the restriction.
 
 Fix Trac #3467: non-top-level type splices
 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
 
 It appears that when I added the ability to splice types in TH
 programs, I failed to pay attention to non-top-level splices -- that
 is, splices inside quotatation brackets.  
 
 This patch fixes the problem.  I had to modify HsType, so there's a
 knock-on change to Haddock.
 
 Its seems that a lot of lines of code has changed, but almost all the
 new lines are comments!
 
 General tidying up
 ~~~~~~~~~~~~~~~~~~
 As a result of thinking all this out I re-jigged the data type ThStage,
 which had far too many values before.  And I wrote a nice state transition
 diagram to make it all precise; 
    see Note [Template Haskell state diagram] in TcSplice
 
 Lots more refactoring in TcSplice, resulting in significantly less code.
 (A few more lines, but actually less code -- the rest is comments.)
 
 I think the result is significantly cleaner.
 
] 
[Refactor the parsing of data type declarations
simonpj@microsoft.com**20090908192327
 Ignore-this: ee7969cd804c1901e4cf4f36d832fc2f
 
 This is a minor change to the parser that tidies it up a bit,
 and allows us to parse
 
        data T :: *
         data S :: * -> *
 
 just like  
 
        data T 
        data S a
 
] 
[Remove unused data constructor ITpragma in type Lexer.Token
simonpj@microsoft.com**20090905171511
 Ignore-this: a0f80c136477921ca8df3d2853e65e6b
] 
[Comments in Cmm
simonpj@microsoft.com**20090910122928
 Ignore-this: 8a3bdd5a63ebd217ab5c9e02d482974c
] 
[fix warnings
Simon Marlow <marlowsd@gmail.com>**20090910111156
 Ignore-this: b3a5e52a9c35b7ecc796b1efaa855856
] 
[binary has unused-import warnings
Simon Marlow <marlowsd@gmail.com>**20090910111115
 Ignore-this: 31ec6d9a0f2f6d5c463a81d0234f8798
] 
[Change the representation of the package database
Simon Marlow <marlowsd@gmail.com>**20090910102703
 Ignore-this: 7c9b38ded7f753d5bb95743695700dbc
 
  - the package DB is a directory containing one file per package
    instance (#723)
 
  - there is a binary cache of the database (#593, #2089)
 
  - the binary package is now a boot package
 
  - there is a new package, bin-package-db, containing the Binary
    instance of InstalledPackageInfo for the binary cache.
 
 Also included in this patch
 
  - Use colour in 'ghc-pkg list' to indicate broken or hidden packages
   
    Broken packages are red, hidden packages are 
   
    Colour support comes from the terminfo package, and is only used when
     - not --simple-output
     - stdout is a TTY
     - the terminal type has colour capability
 
  - Fix the bug that 'ghc-pkg list --user' shows everything as broken
] 
[Improve the user guide section on shared libs
Duncan Coutts <duncan@well-typed.com>**20090908183241
 Ignore-this: ee48f925b9e5441be0fb004b774613cd
 Make it clear that Haskell code to be used by other Haskell code
 must be built as a package.
] 
[FIX #711 implement osFreeAllMBlocks for unix
Austin Seipp <mad.one@gmail.com>**20090910084630
 Ignore-this: 5c773ab5f7f25350dda33d34b77f89f
] 
[NO_CLEAN_GMP fixes
Simon Marlow <marlowsd@gmail.com>**20090909093443
 Ignore-this: 7d490a47c837f1a002937a2b229d0049
] 
[Omit visibility pragmas on Windows (fixes warnings/validate failures)
Simon Marlow <marlowsd@gmail.com>**20090909093217
 Ignore-this: 14cd79e7cded8c0a353544d272f3b974
] 
[Comments only, on Cmm data types
simonpj@microsoft.com**20090910091703
 Ignore-this: ff152a48718289c8125af052b30f8813
] 
[Fix up logic to decide whether shared libs are being built or not
Simon Marlow <marlowsd@gmail.com>**20090909115757
 Ignore-this: 7709d980f416318eae71e8bae255fa2d
 
 Now, adding dyn to $(GhcLibWays) is how shared libs are enabled.
 Everything else keys off that, rather than testing
 $(BuildSharedLibs).
] 
[remove a ToDo that is Done
Simon Marlow <marlowsd@gmail.com>**20090908154411
 Ignore-this: 44b2ada6bc9fe2ed33813bd18159d316
] 
[comment fix
Simon Marlow <marlowsd@gmail.com>**20090908150351
 Ignore-this: 65b1708f41cfdc2f633e66940dc85be5
] 
[refactor *_OPTS settings in the build system; no functional changes
Simon Marlow <marlowsd@gmail.com>**20090909094408
 Ignore-this: 34f51913eef30e4f0afc2c199ad1a42f
] 
[add $(CONF_*_OPTS) for options that come from ./configure (fixes #3426)
Simon Marlow <marlowsd@gmail.com>**20090908151908
 Ignore-this: e70ca8071b79d25fc0d2fa45aafb319e
] 
[Add -package-id, improve package shadowing behaviour and error messages
Simon Marlow <marlowsd@gmail.com>**20090908140906
 Ignore-this: 7180370e7ad19a6d632e8e3c9732267c
 A few changes bundled together here:
 
  - Add -package-id flag, like -package but takes an
    InstalledPackageId as an argument (mainly for Cabal)
    (docs to follow)
 
  - Fix the shadowing behaviour: if we end up with
    two packages with the same name/version that came from
    different package databases, then one will shadow the
    other.  This may mean we have to disable other packages
    that depended on the now-shadowed package.  Lots of
    refactoring to ensure that we can get reasonable diagnostics when
    this happens
 
 <command line>: cannot satisfy -package shadowdep: 
     shadowdep-1-XXX is unusable due to missing or recursive dependencies:
       shadow-1-XXX
     (use -v for more information)
] 
[remove --enable-hc-boot-unregisterised (implied by --enable-hc-boot now)
Simon Marlow <marlowsd@gmail.com>**20090908141919
 Ignore-this: ee36b12d19c28557ebb2334a87197da2
 and fix up related cruft
] 
[fix a warning
Simon Marlow <marlowsd@gmail.com>**20090908131551
 Ignore-this: f61240f6e4e4cae7c1272773ef6c6e01
] 
[fix warnings
Simon Marlow <marlowsd@gmail.com>**20090908130854
 Ignore-this: 46db903fb4bc7b5e6be2fe591548f98e
] 
[fix warnings
Simon Marlow <marlowsd@gmail.com>**20090908114200
 Ignore-this: c39be5ba8646bfb3241f02e15ad97683
] 
[remove Haddock-lexing/parsing/renaming from GHC
Isaac Dupree <id@isaac.cedarswampstudios.org>**20090826145924
 Ignore-this: 6cecf43747cd1f50e3d8e21485af781
] 
[Fix Trac #3403: interaction of CPR and pattern-match failure
simonpj@microsoft.com**20090908131400
 Ignore-this: 38bbb145cdc2c814a45122144ef84a4d
 
 A fine bug report (#3403) demonstrated that we were losing the tail
 call property when a complicated pattern match was involved.  After
 a bit of investigation I discovered that the culprit was the failure
 join-point introduced by the pattern matcher.  It was a zero-argument
 thunk, which is not very CPR-friendly, and that interacted badly with
 CPR worker/wrapper.
 
 It's easy to fix, the same way that we fix other join points, by supplying
 a dummy argument (that is not really passed at runtime.
 
] 
[Fix Trac #3468: improve checking for hs-boot interfaces
simonpj@microsoft.com**20090908130350
 Ignore-this: df9e84e963b1c69328e020772a481f5e
 
 When checking the interface exported by a hs-boot file against the
 Real Thing, I'd failed to check the kind of a type constructor.  If you
 get it wrong, the inconsistency leads to all manner of mischief, as 
 'wkahl' reports in #3468.
 
 This patch should do the job.
 
] 
[Allow TH/annotations to be used with -dynamic
Simon Marlow <marlowsd@gmail.com>**20090908101644
 Ignore-this: ba9e132a946e1f7cd56775d4237405f0
] 
[Fix warnings on 64-bit platforms; fixes validate on x86-64
Simon Marlow <marlowsd@gmail.com>**20090908092639
 Ignore-this: 6bf1d21759aa5ac43807f010c9a9908f
] 
[fix -u flags (after changes for #3310)
Simon Marlow <marlowsd@gmail.com>**20090903072318
 Ignore-this: 518e20973adf1136caf866770f5980ed
] 
[Unify event logging and debug tracing.
Simon Marlow <marlowsd@gmail.com>**20090829094727
 Ignore-this: c137a22fb32a7a5b50fe6906b464344e
 
   - tracing facilities are now enabled with -DTRACING, and -DDEBUG
     additionally enables debug-tracing.  -DEVENTLOG has been
     removed.
 
   - -debug now implies -eventlog
 
   - events can be printed to stderr instead of being sent to the
     binary .eventlog file by adding +RTS -v (which is implied by the
     +RTS -Dx options).
 
   - -Dx debug messages can be sent to the binary .eventlog file
     by adding +RTS -l.  This should help debugging by reducing
     the impact of debug tracing on execution time.
 
   - Various debug messages that duplicated the information in events
     have been removed.
] 
[waitForReturnCapability: fix logic bug
Simon Marlow <marlowsd@gmail.com>**20090831064908
 Ignore-this: 9a9efaee2ab33fc26a04d075f93424b1
 The check for whether a Capability was free was inverted, which harmed
 performance for callbacks.
] 
[Handle renames from #3310
Simon Marlow <marlowsd@gmail.com>**20090830153123
 Ignore-this: 7973f406c726dab27e28b2145dac81dc
 
 Also add a panic for resurrecting a thread blocked on an exception,
 since it should never happen.
] 
[add RTS_PRIVATE attribute
Simon Marlow <marlowsd@gmail.com>**20090829171752
 Ignore-this: 5566c6ecdd4f40c10c4d996856c39113
] 
[Fix incorrectly hidden RTS symbols
Simon Marlow <marlowsd@gmail.com>**20090829132814
 Ignore-this: 6ec9d80873930cfaf05e4ffccce498f6
] 
[Maintain Task/Capability invariants in performPendingThrowTos
Simon Marlow <marlowsd@gmail.com>**20090829092607
 Ignore-this: f33aa00f326155172c30906c7db9ec5b
 Fixes an ASSERTION failure with concprog001, -threaded -debug, +RTS -N2
] 
[Declare RTS-private prototypes with __attribute__((visibility("hidden")))
Simon Marlow <marlowsd@gmail.com>**20090805102159
 Ignore-this: 7e298c7ecbc1216502463fa05a984d3b
 This has no effect with static libraries, but when the RTS is in a
 shared library it does two things:
 
  - it prevents the function from being exposed by the shared library
 
  - internal calls to the function can use the faster non-PLT calls,
    because the function cannot be overriden at link time.
] 
[Configurable iconv header and library locations
Matthias Kilian <kili@outback.escape.de>**20090826155913
 Ignore-this: f3ef2490c61fea6f1cfff648422b919b
 Should help to fix the build on OpenBSD (together with a corresponding
 patch to libraries/base).
] 
[Use -W, not -Werror, for gcc older than 3.4
Matthias Kilian <kili@outback.escape.de>**20090826210044
 Ignore-this: d2fdfef7c2915f58ebe716986847a22d
] 
[Don't mess up absolute INCLUDE_DIRS
Matthias Kilian <kili@outback.escape.de>**20090826153340
 Ignore-this: 3b8c2cceee0bcf9397fb803991ce27ab
 When there are some absolute include dirs, don't create arguments
 like -Ilibraries/base//usr/local/include.
 
 I need this fix for injecting the iconv include dir (which is
 /usr/local/include on OpenBSD) into the build of libraries/base.
] 
[Fix #3461: protect the use of keepCAFs with #ifdef DYNAMIC
Simon Marlow <marlowsd@gmail.com>**20090828125802
 Ignore-this: 96e2dd803e548ec080d701791eab06c
] 
[fix 'darcs-all rec' (amongst other things)
Simon Marlow <marlowsd@gmail.com>**20090827141554
 Ignore-this: fb91b85e37c9be9d153f381c27d9d37c
] 
[REDO: Add -r option to darcs-all, and remove push-all (#3375)
Simon Marlow <marlowsd@gmail.com>**20090827135717
 Ignore-this: 270dbe385afb807c32eaa4b02a3f7d7
 
 rolling back:
 
 Mon Aug  3 11:44:13 BST 2009  Simon Marlow <marlowsd@gmail.com>
   UNDO: Add -r option to darcs-all, and remove push-all (#3375)
   Contributed by: seliopou@gmail.com
       
   This patch modifies darcs-all to have feature parity with push-all by
   recognizing two new options.
   
       * -i, equivalent to --ignore-failure in push-all
       * -r <repo>, specifies the remote repository darcs commands will use
   
   Some example commands:
   
   Get the libraries from a repository of your choosing. This is useful
   when working with a git mirror:
   
       $ ./darcs-all -r http://darcs.haskell.org get
   
   Pull changes. Used to be:
   
       $ ./push-all --pull http://darcs.haskell.org
   
   Is now:
   
       $ ./darcs-all -r http://darcs.haskell.org pull
   
   Or to use the default remote of the ghc repository:
   
       $ ./darcs-all pull
 
     M ./darcs-all -79 +33
     A ./push-all
] 
[Add a link to hp2any from the profiling section.
Simon Marlow <marlowsd@gmail.com>**20090827114050
 Ignore-this: 998f316811cacaa3a9ee6a25aae99e80
] 
[Fix "make install"
Ian Lynagh <igloo@earth.li>**20090826155638
 We need to change a dependency on
     pkg-inplace
 to
     pkg-$abihash
 when installing
] 
["ghc-cabal install" now needs to know where GHC is, to get the ABI hash
Ian Lynagh <igloo@earth.li>**20090826133359] 
[Fix bindist creation
Ian Lynagh <igloo@earth.li>**20090826122953
 We were running into problems like:
     for f in  LICENSE configure config.sub config.guess [...]
     make[2]: execvp: /bin/sh: Argument list too long
 This patch moves the loop into make, rather than the shell.
] 
[UNDO: fix the inplace runghc (it broke the installed runghc)
Simon Marlow <marlowsd@gmail.com>**20090826104937
 Ignore-this: 4a9639a93dbe37d81baafcec0029cdcb
] 
[Fix part of #3398: pretty-printing always goes via the I/O library encoding
Simon Marlow <marlowsd@gmail.com>**20090826104052
 Ignore-this: 7ce68105539ff38f4853de149a43a1a3
 That is, unless we're printing in LeftMode, where we bypass encoding
 for speed.  This is safe, because LeftMode is used for outputting C or
 asm, where everyting is Z-encoded and hence ASCII.
 
 Error messages and other compiler output containing Unicode will now
 appear correctly according to the locale settings.
] 
[Tidy up file headers and copyrights; point to the wiki for docs
Simon Marlow <marlowsd@gmail.com>**20090825095047
 Ignore-this: d34994b9910fd24ff7498102d2a12648
 
 I've updated the wiki page about the RTS headers
   http://hackage.haskell.org/trac/ghc/wiki/Commentary/SourceTree/Includes
 to reflect the new layout and explain some of the rationale.  All the
 header files now point to this page.
] 
[Follow changes in Cabal: package -> sourcePackageId
Simon Marlow <marlowsd@gmail.com>**20090824160020
 Ignore-this: c1f9d41c2ed75cf47842fa04c4b2fac5
] 
[fix an unused-import warning
Simon Marlow <marlowsd@gmail.com>**20090824153657
 Ignore-this: 35e87617fc289237cfa12133ee9d5dca
] 
[Add unique package identifiers (InstalledPackageId) in the package DB
Simon Marlow <marlowsd@gmail.com>**20090820110920
 Ignore-this: 5e09becc51e6888ec5186811e7e33d46
 See commentary at
 http://hackage.haskell.org/trac/ghc/wiki/Commentary/Packages
] 
[Move the standalone-deriving flag test from parser to renamer
simonpj@microsoft.com**20090825073324
 Ignore-this: 10e3760c5443fc88b0f498ad0a2595ae
 
 This is just a tiny refactoring.  In general, we're trying to
 get rid of parser errors in favour of later, more civlised, errors.
 
] 
[Error message wibble
simonpj@microsoft.com**20090825073259
 Ignore-this: 85257158393634c41057a0849d95490f
] 
[Fix Trac #3406 (albeit not very satisfactorily): scoped type variables
simonpj@microsoft.com**20090825073059
 Ignore-this: 1509e0565ccd4b6708ac4cf4b2eb451e
 
 The issue here is this:
 
   type ItemColID a b = Int  -- Discards a,b
 
   get :: ItemColID a b -> a -> ItemColID a b
   get (x :: ItemColID a b) = x :: ItemColID a b
 
 The pattern signature for 'x' doesn't actually rigidly bind a,b.
 This crashed GHC 6.10 with a 'readFilledBox' panic.  Now we fail
 with an erroe message
 
 With the new outside-in algorithm we'll be able to accept this program.
 
] 
[Make FastString thread-safe.
Thomas Schilling <nominolo@googlemail.com>**20090824182252
   
 This is needed both for per-session parallelism and for allowing
 multiple concurrent sessions in the same process.  With the help of
 atomicModifyIORef and unsafePerformIO it is also quite fast--an MVar
 would most likely be slower.  On a full compilation of Cabal's head
 branch it was about 1-2 percent slower, but then overall compilation
 times varied by about 4 percent, so I think it's worth it.
] 
[Improve docs on -XFlexibleContexts
simonpj@microsoft.com**20090824113722
 Ignore-this: 16c4b05f52bca250ae6eff59feb08cc4
] 
[fix do-notation warnings
Simon Marlow <marlowsd@gmail.com>**20090824080400
 Ignore-this: 74e0b085478ec858ce56cc925749e0ea
] 
[Use explicit Word32/Int32 in place of Int in the on-disk .hi file
Simon Marlow <marlowsd@gmail.com>**20090821155028
 Ignore-this: e48ea5045f9460c347d457ada0e2b0da
 For: FastStrings, Names, and Bin values.  This makes .hi files smaller
 on 64-bit platforms, while also making the format a bit more robust.
] 
[Fix the interface-file incompatibility crash (#3435)
Simon Marlow <marlowsd@gmail.com>**20090821154737
 Ignore-this: 5b7b9aca621fd378a7f911c50429c100
 
 We now have a dummy 32/64-bit field near the beginning of a .hi file
 for backward-compatibility reasons; see comments (Note [dummy iface
 field]) in BinIface.hs.
] 
[trim Data.Sequence import, in preparation for expanding its API
Ross Paterson <ross@soi.city.ac.uk>**20090824080533
 Ignore-this: 97f8c3e309c14a132f20c565d33d46a9
] 
[Remove bitrotted IA64 code in Linker.c
Ian Lynagh <igloo@earth.li>**20090823151600
 It breaks the unregisterised build on IA64.
] 
[Make consIORef atomic.
Thomas Schilling <nominolo@googlemail.com>**20090822230937
 Ignore-this: c4570e17a21be724b45c574bd25d0330
] 
[Make updates to external package state idempotent.
Thomas Schilling <nominolo@googlemail.com>**20090821235355
 
 Without this, concurrent updates to the EPS could introduce
 overlapping instances (even though they came from the same module).
] 
[Fix Trac #3423: missed instantiation for newtype-derived instances
simonpj@microsoft.com**20090821210700
 Ignore-this: 73fa66b45a9affe5b4e0276551e5861f
 
 Somehow I'd forgotten to instantiate the coercion that is stored in a
 'NewtypeDerived' constructor in an InstInfo.  The necessary code is
 in TcInstDcls.tc_inst_decl2.
 
 The result was ghc: panic! (the 'impossible' happened)
    (GHC version 6.10.3 for x86_64-unknown-linux):
    No match in record selector Var.tcTyVarDetails
 because we were looking at an (uninstantiated) TyVar instead of
 an (instantiated) TcTyVar.
 
] 
[Put "dl" back in rts/package.conf if HAVE_DL is defined
Ian Lynagh <igloo@earth.li>**20090821145423
 Fixes linking with -dynamic
] 
[Link CMM objects into dynamic libraries
Ian Lynagh <igloo@earth.li>**20090821132809
 This fixes linking hello world with -dynamic. I've also added some more
 variables, so there is less duplication between the different ways of
 linking.
] 
[-fPIC -fvia-C issues a warning and ignores -fvia-C
Simon Marlow <marlowsd@gmail.com>**20090821144544
 Ignore-this: 2ba1902cf7eec7f149f38cadeac8c245
 Also, -fPIC causes an error if the target is registerised and has no
 native code generator.
] 
[Use allocateLocal() rather than allocate() in the interpreter
Simon Marlow <marlowsd@gmail.com>**20090820152325
 Ignore-this: 42cc9250f46d95a945394b82acc76be4
 This gives about a 15% performance boost in GHCi for me.  nice!
] 
[Another tiny tidy-up to RnPat
simonpj@microsoft.com**20090821100826
 Ignore-this: 4290aab233ee1b0fcd3768009950a213
] 
[Fix Trac #3437: strictness of specialised functions
simonpj@microsoft.com**20090821095251
 Ignore-this: 24345ff251a517e540a9666a78ed8176
 
 'lilac' helpful pin-pointed a space leak that was due to a specialised
 function being insufficiently strict.  Here's the new comment in SpecConstr:
 
 Note [Transfer strictness]
 ~~~~~~~~~~~~~~~~~~~~~~~~~~
 We must transfer strictness information from the original function to
 the specialised one.  Suppose, for example
 
   f has strictness     SS
         and a RULE     f (a:as) b = f_spec a as b
 
 Now we want f_spec to have strictess  LLS, otherwise we'll use call-by-need
 when calling f_spec instead of call-by-value.  And that can result in 
 unbounded worsening in space (cf the classic foldl vs foldl')
 
 See Trac #3437 for a good example.
 
 The function calcSpecStrictness performs the calculation.
 
 
] 
[Wibbles to field-label puns
simonpj@microsoft.com**20090821090637
 Ignore-this: ed6dc679b906b2a928e1b9a6db590108
] 
[Fix library installation; fixes #3374
Ian Lynagh <igloo@earth.li>**20090820173707
 When configuring packages, enable library profiling and shared libraries
 based on the ways in GhcLibWays.
] 
[Escape some $s in makefiles for consistency
Ian Lynagh <igloo@earth.li>**20090820165033] 
[Wibble to RnPat refactoring
simonpj@microsoft.com**20090820161520
 Ignore-this: 77c328334e088113afb25275f0aca364
] 
[Rollback: use cas() to claim the closure in copyPart(), to match copy_tag()
Simon Marlow <marlowsd@gmail.com>**20090820155833
 Ignore-this: e62da4d667c15c0dafb2660f4b915b36
 
 rolling back:
 
   * use cas() to claim the closure in copyPart(), to match copy_tag()
   * rename whitehole_spin to evac_collision, and update it properly
 
 This introduced a new failure in parallel GC.  I'll rollback for now
 until I've fixed it.
] 
[Relax the assumption that all objects fit in a single block (#3424)
Simon Marlow <marlowsd@gmail.com>**20090820144308
 Ignore-this: 44ec5ed31491241b6f03182c80c0681e
 
 It is possible for the program to allocate single object larger than a
 block, without going through the normal large-object mechanisms that
 we have for arrays and threads and so on.  
 
 The GC was assuming that no object was larger than a block, but #3424
 contains a program that breaks the assumption.  This patch removes the
 assumption.  The objects in question will still be copied, that is
 they don't get the normal large-object treatment, but this case is
 unlikely to occur often in practice.
 
 In the future we may improve things by generating code to allocate
 them as large objects in the first place.
] 
[remove a bogus assertion
Simon Marlow <marlowsd@gmail.com>**20090820143305
 Ignore-this: 1b58b8a73d58257630832d37475795bb
] 
[Restore the GHC version check
Simon Marlow <marlowsd@gmail.com>**20090820134234
 Ignore-this: a1a0640e985a9eaf78804465276aadcd
 I'm not sure what happened here, but the ordering of tests was messed
 up, with the result that the GHC version check was being omitted.
] 
[fix the inplace runghc
Simon Marlow <marlowsd@gmail.com>**20090820122538
 Ignore-this: 54be86fcd9ff96b5a37e77f64bf3c1d9
 
 It was invoking $(TOP)/inplace/bin/ghc rather than
 $(TOP)/inplace/bin/ghc-stage2
] 
[Add a case for IND (and a comment).  Fixes #3424, perhaps only partially.
Simon Marlow <marlowsd@gmail.com>**20090820131537
 Ignore-this: 75287786343359fdceb0691f88cb4f5f
] 
[Improvements to record puns, wildcards
simonpj@microsoft.com**20090820123443
 Ignore-this: 2ac26fcc045853dab346280bcc440431
 
 * Make C { A.a } work with punning, expanding to C { A.a = a }
 
 * Make it so that, with -fwarn-unused-matches, 
         f (C {..}) = x
   does not complain about the bindings introduced by the "..".
 
 * Make -XRecordWildCards implies -XDisambiguateRecordFields.
 
 * Overall refactoring of RnPat, which had become very crufty. 
   In particular, there is now a monad, CpsRn, private to RnPat,
   which deals with the cps-style plumbing.  This is why so many
   lines of RnPat have changed.
 
 * Refactor the treatment of renaming of record fields into two passes
        - rnHsRecFields1, used both for patterns and expressions,
             which expands puns, wild-cards
        - a local renamer in RnPat for fields in patterns
        - a local renamer in RnExpr for fields in construction and update
   This make it all MUCH easier to understand
  
 * Improve documentation of record puns, wildcards, and disambiguation
 
] 
[Make -dppr-debug print locations in HsSyn
simonpj@microsoft.com**20090820123245
 Ignore-this: 673f8100c2744e58635e75e6cf585ce5
 
 Show SrcSpans for Located things might be overkill, but it's sometimes
 useful.
 
 I also added
   ppWhen, ppUnless :: Bool -> SDoc -> SDoc
 to Outputable
 
] 
[Make -dynamic a proper way, so we read the .dyn_hi files
Simon Marlow <marlowsd@gmail.com>**20090820121208
 Ignore-this: 252d2091ff90041e055efd0d62ebe185
 Also, I cleaned up some of the way-related infrastructure, removing
 two global variables.  
 
 There's more that could be done here, but it's a start.  The way flags
 probably don't need to be static any more.
] 
[fix warning
Simon Marlow <marlowsd@gmail.com>**20090820110532
 Ignore-this: ce16d3661b6bc2f0d382f9025c007019
] 
[generalise the type of "on"
Simon Marlow <marlowsd@gmail.com>**20090820094516
 Ignore-this: 5b789c36663accb7284c26b1bd7e019d
] 
[need to check $(HADDOCK_DOCS) around contents/index generation
Simon Marlow <marlowsd@gmail.com>**20090820092220
 Ignore-this: ef5352164fd6aac41ba3a1542182d213
] 
[WAY_dyn_LIB_TARGET is not used anywhere; kill it
Simon Marlow <marlowsd@gmail.com>**20090819155952
 Ignore-this: 36b784db6d85acc2374e781fcd77d10d
] 
[Rolling back: Add a kludge to fix building shared libs
Simon Marlow <marlowsd@gmail.com>**20090819145507
 Ignore-this: fd2cc70119796be7a0ba62131f7e9ddb
] 
[rename whitehole_spin to evac_collision, and update it properly
Simon Marlow <marlowsd@gmail.com>**20090819141518
 Ignore-this: c6bad4a99758b7a60749cc769305aa90
] 
[use cas() to claim the closure in copyPart(), to match copy_tag()
Simon Marlow <marlowsd@gmail.com>**20090819141106
 Ignore-this: 6baad6aa65e640d9dbfd3d18b0f77644
 
 copyPart() was still using the old WHITEHOLE mechanism for locking the
 closure.  I don't think this fixes any actual bugs, but it removes a
 gratuitous difference between two functions that should look similar.
] 
[Always yieldCapabilty() when a bound thread blocks
Simon Marlow <marlowsd@gmail.com>**20090819130856
 Ignore-this: 25303050a3095760972fc550445865e6
 Fixes crash in concprog002(threaded2_qw), and possibly other problems
] 
[FIX #2767 & original problem of #3208
Manuel M T Chakravarty <chak@cse.unsw.edu.au>**20090820072608
 Ignore-this: cb8326cf885c2da0b497ef10f8fed942
] 
[Improve the "Stack space overflow" error; fixes trac #3296
Ian Lynagh <igloo@earth.li>**20090819202112] 
[Detect C finalizer callbacks in rts_lock() instead of schedule()
Simon Marlow <marlowsd@gmail.com>**20090819123208
 Ignore-this: 522b05d928b711ee3473539c9db421b4
 Otherwise, finalizer callbacks cause a deadlock in the threaded RTS
 (including GHCi)
] 
[Bump Happy requirement from 1.15 to 1.16
Simon Marlow <marlowsd@gmail.com>**20090819093929
 Ignore-this: d550a42d284b8585faf546231046134b
 Now that the haskell98 dependency of GHC has been removed, the
 Happy-generated parsers must import Data.Array rather than Array, and
 that change was made in Happy 1.16.
] 
[Restore the entry field in StgInfoTable when !defined(TABLES_NEXT_TO_CODE)
Simon Marlow <marlowsd@gmail.com>**20090819113550
 Ignore-this: 491038d9d526888f31e12e7349024e37
 Somehow this got lost, probably in the recent RTS tidy-up.  
 Fixes segfaults in unregisterised compilation.
] 
[Fix warnings
Ian Lynagh <igloo@earth.li>**20090819125440] 
[Put a newtype wrapper around ModuleEnv
Ian Lynagh <igloo@earth.li>**20090818192903] 
[Make the thr_dyn RTS ways optional on GhcUnregisterised
Simon Marlow <marlowsd@gmail.com>**20090819095538
 Ignore-this: 7bbeddde0fcd7f2030087f3292416492
 
 Fixes this failure in the unreigsterised build:
   /usr/bin/ld: cannot find -lHSrts_thr
] 
[fix an include
Simon Marlow <marlowsd@gmail.com>**20090819095237
 Ignore-this: d2988008e40589f615e56f92df01c19b
] 
[remove some redundant declarations
Simon Marlow <marlowsd@gmail.com>**20090819094338
 Ignore-this: ce62ec28b40d00f6e8b072807b2f7591
] 
[Remove the lock around NameCache for readBinIface.
Thomas Schilling <nominolo@googlemail.com>**20090818213243
 
 Turns out using atomic update instead of a full-blown lock was easier
 than I thought.  It should also be safe in the case where we
 concurrently read the same interface file.  Whichever thread loses the
 race will simply find that all of the names are already defined and
 will have no effect on the name cache.
] 
[Fix #3429: a tricky race condition
Simon Marlow <marlowsd@gmail.com>**20090818112942
 Ignore-this: c6f47c6bd3b12ecbd7aa3cfcd25bf9a7
 
 There were two bugs, and had it not been for the first one we would
 not have noticed the second one, so this is quite fortunate.
 
 The first bug is in stg_unblockAsyncExceptionszh_ret, when we found a
 pending exception to raise, but don't end up raising it, there was a
 missing adjustment to the stack pointer.  
 
 The second bug was that this case was actually happening at all: it
 ought to be incredibly rare, because the pending exception thread
 would have to be killed between us finding it and attempting to raise
 the exception.  This made me suspicious.  It turned out that there was
 a race condition on the tso->flags field; multiple threads were
 updating this bitmask field non-atomically (one of the bits is the
 dirty-bit for the generational GC).  The fix is to move the dirty bit
 into its own field of the TSO, making the TSO one word larger (sadly).
] 
[comment fixes (install.mk.in, not dirs.mk.in)
Simon Marlow <marlowsd@gmail.com>**20090817094724
 Ignore-this: 1ca40b832e7cec7a560762a40daa8d4f
] 
[Fix configure summary: --enable-shared has gone
Simon Marlow <marlowsd@gmail.com>**20090817091210
 Ignore-this: 97dce2f98f13c00fd4475f78f3c04e84
] 
[comment fix
Simon Marlow <marlowsd@gmail.com>**20090817091105
 Ignore-this: 880c1948e6f1cfc2841c20a9f93232c1
] 
[remove unused file
Simon Marlow <marlowsd@gmail.com>**20090806112859
 Ignore-this: b0aea4ddcd651e04806a4d2a0f250c3d
] 
[Fix unregisterised build
Simon Marlow <marlowsd@gmail.com>**20090806095417
 Ignore-this: 93ad8435aa581fd767fb639b10c1ed4a
] 
[FIX #3405
Manuel M T Chakravarty <chak@cse.unsw.edu.au>**20090818121425
 Ignore-this: 6ca07059605e9cdf5ab5aa5bb8cc4455
] 
[Make the dynamic linker thread-safe.
Thomas Schilling <nominolo@googlemail.com>**20090817142352
   
 The current implementation is rather pessimistic.  The persistent
 linker state is now an MVar and all exported Linker functions are
 wrapped in modifyMVar calls.  This is serves as a big lock around all
 linker functions.
 
 There might be a chance for more concurrency in a few places. E.g.,
 extending the closure environment and loading packages might be
 independent in some cases.  But for now it's better to be on the safe
 side.
] 
[Make access to NameCache atomic.  Sometimes needs a lock.
Thomas Schilling <nominolo@googlemail.com>**20090817004819
 
 'readBinIface' updates the name cache in a way that is hard to use
 with atomicModifyIORef, so this patch introduces a lock for this case.
 All other updates use atomicModifyIORef.
 
 Having a single lock is quite pessimistic, so it remains to be seen
 whether this will become a problem.  In principle we only need to make
 sure that we do not load the same file concurrently (or that it's
 idempotent).  In practice we also need to ensure that concurrent reads
 do not cancel each other out (since the new NameCache may be based on
 an outdated version).
] 
[Make updates to the Finder caches atomic.  Well, almost.
Thomas Schilling <nominolo@googlemail.com>**20090816231920
 
 Flushing and uncaching a single module is not completely atomic since
 both caches a cleared separately.  However, flushing is only done when
 changing the working directory which shouldn't be done concurrently to
 other threads.  Uncaching is only done in 'summariseModule' which
 requires some more work to become thread-safe anyway.
] 
[Make updates to the external package state atomic.
Thomas Schilling <nominolo@googlemail.com>**20090816231316
 Ignore-this: a700ff0c6725264941d6a5458ce5dade
] 
[Minor documentation fixes.
Thomas Schilling <nominolo@googlemail.com>**20090722231342] 
[bindist fix
Ian Lynagh <igloo@earth.li>**20090816200326] 
[Build fixes
Ian Lynagh <igloo@earth.li>**20090816190043] 
[Remove the old platform code from the bindist configure.ac
Ian Lynagh <igloo@earth.li>**20090814225431] 
[Make our install variables etc compliant with GNU standards; fixes #1924
Ian Lynagh <igloo@earth.li>**20090814224549] 
[Improve fix to Trac #3007
simonpj@microsoft.com**20090814110214
 Ignore-this: ba7d9b91b80f66aa5ece93dcd4ca4bab
 
 This patch tides up Ian's fix a little. In particular, if if you
 {-# SOURCE #-} import a module from a different package, you now 
 get a much more civlised error message.
 
] 
[Only look up whether a module's SOURCE-imported if it's in the current package
Ian Lynagh <igloo@earth.li>**20090813162435
 Suppose we import anotherPackage:M, which exports things from
 anotherPackage:Internal. Then GHC will want to read
 anotherPackage:Internal.hi.
 
 However, if we have also SOURCE-imported thisPackage:Internal then
 we don't want GHC to try to read anotherPackage:Internal.hi-boot
 instead.
 
 The mapping that tells us whether a module is SOURCE-imported uses just
 the module name for the key, so we have to check the package ID before
 looking it up.
 
 Fixes #3007.
] 
[Fix Trac #3409: type synonyms that discard their arguments
simonpj@microsoft.com**20090813161154
 Ignore-this: d2ab9292c281323280fc74e04a9d29dd
 
 Type synonyms that don't mention one of their type parameters on the 
 RHS are a pain in the neck.  This patch fixes a long-standing bug (that
 simply has not appeared before) in that exprType could return a type
 mentioning an existentially-quantified variable (in one of those ignored
 argument positions).
 
 See CoreUtils Note [Existential variables and silly type synonyms]
 
 The fix is not entirely beautiful, but it works, and is very localised.
 
] 
[Add support for multi-line deprecated pragmas; trac #3303
Ian Lynagh <igloo@earth.li>**20090812185912] 
[Fix a sanity check; fixes #3089
Ian Lynagh <igloo@earth.li>**20090812133817] 
[Skip "Cabal check" for certain packages that we know will fail
Ian Lynagh <igloo@earth.li>**20090811220146] 
[Fix "Cabal check" warnings
Ian Lynagh <igloo@earth.li>**20090811215839] 
[Check Cabal packages when validating
Ian Lynagh <igloo@earth.li>**20090811212559
 This checks that hackage would accept the packages.
 Currently warnings are printed, but don't result in failure.
] 
[Include a pointer to GHC.Exts in the generated GHC.Prim docs
Ian Lynagh <igloo@earth.li>**20090811145300] 
[Refactor, and improve error messages (cf Trac #3395)
simonpj@microsoft.com**20090811143655
 Ignore-this: e7205f5bcf8408ff791ba19156e48461
 
 The Convert stuff should not panic if the programmer hands over an
 invalid TH term; instead it should give a graceful error message.
 Largely this had been done, but not for do-blocks, so this patch
 fixes that problem.
 
 Moreover, I did some refactoring and tidying up, which is why
 so many lines of code have changed
 
] 
[Fix Trac #3421: a typo in TysPrim
simonpj@microsoft.com**20090810164320
 Ignore-this: 67af95909862d672c6eb738c52458873
 
 This is just a blatant typo, where Any1 :: *->* was getting mixed
 up with Any :: *.
 
] 
[Improve the recent changes to overlap-checking for view patters
simonpj@microsoft.com**20090810141158
 Ignore-this: 8b97dfef44aa9951c31587a884ec92f
 
 The previous patch simply gave up for view patterns; this version
 instead treats them like n+k patterns and gives signficantly better
 results.
 
 Less code, too.
 
] 
[Turn group into a special_id when TransformListComp is on
Max Bolingbroke <batterseapower@hotmail.com>**20090717224903
 Ignore-this: eabef2f2044ee6cc6bc876529f4044ca
] 
[Make Constants shareable with the base package
Ian Lynagh <igloo@earth.li>**20090809180701] 
[Minor refactoring
Ian Lynagh <igloo@earth.li>**20090809154258] 
[FIX #2395 (ViewPatterns trigger bad Check errors)
Alexander Dunlap <alexander.dunlap@gmail.com>**20090807184841
 Ignore-this: af83dc1d93b735bd89d5f9dcb2f5ca6b
] 
[Pass -m32 to gcc on i386 and ppc OS X
Ian Lynagh <igloo@earth.li>**20090808222537
 This makes GHC work even if you are actually running it in 64bit mode,
 e.g. on OS X 10.6 Snow.
] 
[Fix bindist creation
Ian Lynagh <igloo@earth.li>**20090808184407
 This is a bit kludgy. We've hit the maximum number of arguments you can
 have in the for loop for adding files to the bindist list, so this just
 splits the list in 2.
] 
[Put the library haddock docs in the html subdirectory of docdir
Ian Lynagh <igloo@earth.li>**20090808175756] 
[Install the main doc index page, and also put it in bindists
Ian Lynagh <igloo@earth.li>**20090808112034] 
[FIX BUILD: Remove harmful dnls and identifier typo in aclocal.m4
Alexander Dunlap <alexander.dunlap@gmail.com>**20090801012335
 Ignore-this: 825b21e13d4b3f04fab91bf07f4e693b
] 
[Fix the build on OS X
Ian Lynagh <igloo@earth.li>**20090807152151] 
[Add -r option to darcs-all, and remove push-all (#3375)
Simon Marlow <marlowsd@gmail.com>*-20090803104413
 Ignore-this: 6871c15294113898902be90210d9a7a9
 Contributed by: seliopou@gmail.com
     
 This patch modifies darcs-all to have feature parity with push-all by
 recognizing two new options.
 
     * -i, equivalent to --ignore-failure in push-all
     * -r <repo>, specifies the remote repository darcs commands will use
 
 Some example commands:
 
 Get the libraries from a repository of your choosing. This is useful
 when working with a git mirror:
 
     $ ./darcs-all -r http://darcs.haskell.org get
 
 Pull changes. Used to be:
 
     $ ./push-all --pull http://darcs.haskell.org
 
 Is now:
 
     $ ./darcs-all -r http://darcs.haskell.org pull
 
 Or to use the default remote of the ghc repository:
 
     $ ./darcs-all pull
] 
[Add a kludge to fix building shared libs
Ian Lynagh <igloo@earth.li>**20090807130731
 The .dyn_hi files currently depend on the .dyn_hi files of modules that
 they import. But they actually want the .hi files of modules from
 another package.This we make the .dyn_hi files depend on the .hi files
 so we are sure that they exist and are up-to-date.
] 
[Leave dyn in GhcLibWays when validating
Ian Lynagh <igloo@earth.li>**20090807005445] 
[Build shared library support by default on platforms that support it
Ian Lynagh <igloo@earth.li>**20090806164320] 
[Update the libffi filenames to follow the new version number
Ian Lynagh <igloo@earth.li>**20090806141359
 Stops make going into an infinite rebuilding loop
] 
[Fix ticky build
Simon Marlow <marlowsd@gmail.com>**20090806093021
 Ignore-this: dbabf96061763e3a6cb7259e3ea71bed
] 
[add #include <sys/types.h> (hopefully fixes OS X build)
Simon Marlow <marlowsd@gmail.com>**20090806081747
 Ignore-this: 82d02b32ec5f7ad1966729402a6610
] 
[profiling build fix
Simon Marlow <marlowsd@gmail.com>**20090805102217
 Ignore-this: 610a09b10b9f326e392f96d92f0fcb20
] 
[profiling build fixes
Simon Marlow <marlowsd@gmail.com>**20090805093539
 Ignore-this: 90015383cdb908227508363b28de76e4
] 
[#include fix
Simon Marlow <marlowsd@gmail.com>**20090805092537
 Ignore-this: f48f53a20e0ad367d78ecdd675b99b7c
] 
[use C99-style array initialisers
Simon Marlow <marlowsd@gmail.com>**20090803214545
 Ignore-this: 87a2adbaf9495d8ee9b9191dd863dfa7
] 
[use C99-style array initialisers
Simon Marlow <marlowsd@gmail.com>**20090803210948
 Ignore-this: fadcc60df270797beec7dcc4dca5da04
] 
[move termios prototypes into a public header
Simon Marlow <marlowsd@gmail.com>**20090803210304
 Ignore-this: ef87e8e25aac38ba4b92df4f0078d9cc
] 
[move StgEntCounter type into its own header
Simon Marlow <marlowsd@gmail.com>**20090803210229
 Ignore-this: 6f9896a3e575e347ca9c89108d12469e
] 
[Common up two closure type -> string tables
Simon Marlow <marlowsd@gmail.com>**20090803203252
 Ignore-this: e4a07b24c945a5181339ce769a178c33
 Also, use C99-style array initialisers
] 
[Rename primops from foozh_fast to stg_foozh
Simon Marlow <marlowsd@gmail.com>**20090803202940
 Ignore-this: eb6022a346e27ad8e19b1be307d4618e
 For consistency with other RTS exported symbols
] 
[remove the GUM closure types
Simon Marlow <marlowsd@gmail.com>**20090803202151
 Ignore-this: 871ab05276a311af7330c38a49e415f0
] 
[use C99-style array initialisers
Simon Marlow <marlowsd@gmail.com>**20090803201148
 Ignore-this: 3490e28f536bfb424510628b0d75d4a7
] 
[Remove final bits of the old .NET support
Simon Marlow <marlowsd@gmail.com>**20090803195935
 Ignore-this: a02948975e17219512910d530f2faab2
] 
[fast make omits all GHC stages
Simon Marlow <marlowsd@gmail.com>**20090802190752
 Ignore-this: 8462306d70cbb95df532e50caec3d20f
] 
[fast make omits dependencies in ghc/ too
Simon Marlow <marlowsd@gmail.com>**20090802190733
 Ignore-this: 5f622f40ce8ce604573ee9867047ce0a
] 
[remove unnecessary -#include options
Simon Marlow <marlowsd@gmail.com>**20090802190651
 Ignore-this: 4e6765f51914cbc258f9b5e75c86fd83
] 
[mention that INCLUDE pragmas are accepted, but ignored
Simon Marlow <marlowsd@gmail.com>**20090802190614
 Ignore-this: 533952df3b9c4350b1d2a19bda40f324
] 
[Deprecate the -#include flag
Simon Marlow <marlowsd@gmail.com>**20090802190535
 Ignore-this: 851c0c0d791af34eac7578b59affcf79
] 
[remove docuumentation for -#include option
Simon Marlow <marlowsd@gmail.com>**20090801222118
 Ignore-this: 7b8a03145a2b767e6f71cf768a2e97
] 
[fix off-by-one in memory allocation
Simon Marlow <marlowsd@gmail.com>**20090804092115
 Ignore-this: a92a5e27d1acd27706b96fff701af0a2
] 
[Windows build fix
Simon Marlow <marlowsd@gmail.com>**20090803150533
 Ignore-this: 425a5572d84152f5bf98890030ed2b34
] 
[Windows build fixes
Simon Marlow <marlowsd@gmail.com>**20090803142346
 Ignore-this: e1ede00e0ecb223ce222767c29f92063
] 
[move gc_alloc_block to make it visible on 32-bit
Simon Marlow <marlowsd@gmail.com>**20090803132847
 Ignore-this: 1859f0a8e527a37012587ae385f2c4f4
] 
[Windows build fixes
Simon Marlow <marlowsd@gmail.com>**20090803131933
 Ignore-this: 9a381956cbe00a3be1487d68ee70f67f
] 
[x86_64 warning fix
Simon Marlow <marlowsd@gmail.com>**20090803120415
 Ignore-this: 2f5484f0285849f090cedc2e24cae161
] 
[x86_64 warning fixes
Simon Marlow <marlowsd@gmail.com>**20090803115955
 Ignore-this: 7a0a2a5e98441b1bc1d40848e9265891
] 
[x86_64 build fix: declare gc_alloc_block_sync
Simon Marlow <marlowsd@gmail.com>**20090803113603
 Ignore-this: c719572e261edd98f9b7fadf43cf7616
] 
[RTS tidyup sweep, first phase
Simon Marlow <marlowsd@gmail.com>**20090802213204
 Ignore-this: 1f7aabac4d5f9d32f6d91f1ccb1a80a4
 
 The first phase of this tidyup is focussed on the header files, and in
 particular making sure we are exposinng publicly exactly what we need
 to, and no more.
 
  - Rts.h now includes everything that the RTS exposes publicly,
    rather than a random subset of it.
 
  - Most of the public header files have moved into subdirectories, and
    many of them have been renamed.  But clients should not need to
    include any of the other headers directly, just #include the main
    public headers: Rts.h, HsFFI.h, RtsAPI.h.
 
  - All the headers needed for via-C compilation have moved into the
    stg subdirectory, which is self-contained.  Most of the headers for
    the rest of the RTS APIs have moved into the rts subdirectory.
 
  - I left MachDeps.h where it is, because it is so widely used in
    Haskell code.
  
  - I left a deprecated stub for RtsFlags.h in place.  The flag
    structures are now exposed by Rts.h.
 
  - Various internal APIs are no longer exposed by public header files.
 
  - Various bits of dead code and declarations have been removed
 
  - More gcc warnings are turned on, and the RTS code is more
    warning-clean.
 
  - More source files #include "PosixSource.h", and hence only use
    standard POSIX (1003.1c-1995) interfaces.
 
 There is a lot more tidying up still to do, this is just the first
 pass.  I also intend to standardise the names for external RTS APIs
 (e.g use the rts_ prefix consistently), and declare the internal APIs
 as hidden for shared libraries.
] 
[replace sparc-specific Int64 code with calls to platform-independent macros
Simon Marlow <marlowsd@gmail.com>**20090727215103
 Ignore-this: 4f65892caeb42f8f58724883cc6097eb
] 
[remove old incarnation of .NET support
Simon Marlow <marlowsd@gmail.com>**20090628211147
 Ignore-this: b8e0e316e371b91e676cacf08ed8e1c3
] 
[Windows build fix
Simon Marlow <marlowsd@gmail.com>**20090803144931
 Ignore-this: c066b5b75e58f3c4389fc199f700dce5
] 
[rts_stop_on_exception is a C int, not a W_
Simon Marlow <marlowsd@gmail.com>**20090803124916
 Ignore-this: d0432287956e7291a503a554424d6897
 amazing this hasn't caused any problems before now
] 
[Fix #3412: the worker of an Id might not be a local Id
Simon Marlow <marlowsd@gmail.com>**20090803112803
 Ignore-this: c433025fd02c997a05555f5c7e6a6a7b
] 
[Add -r option to darcs-all, and remove push-all (#3375)
Simon Marlow <marlowsd@gmail.com>**20090803104413
 Ignore-this: 6871c15294113898902be90210d9a7a9
 Contributed by: seliopou@gmail.com
     
 This patch modifies darcs-all to have feature parity with push-all by
 recognizing two new options.
 
     * -i, equivalent to --ignore-failure in push-all
     * -r <repo>, specifies the remote repository darcs commands will use
 
 Some example commands:
 
 Get the libraries from a repository of your choosing. This is useful
 when working with a git mirror:
 
     $ ./darcs-all -r http://darcs.haskell.org get
 
 Pull changes. Used to be:
 
     $ ./push-all --pull http://darcs.haskell.org
 
 Is now:
 
     $ ./darcs-all -r http://darcs.haskell.org pull
 
 Or to use the default remote of the ghc repository:
 
     $ ./darcs-all pull
] 
[remove dead code
Simon Marlow <marlowsd@gmail.com>**20090620214113
 Ignore-this: 6da2bda8efd2dc63c8931243ab7dec0a
] 
[fix warnings
Simon Marlow <marlowsd@gmail.com>**20090729134348
 Ignore-this: 98128ac0b78d861439ebf807af1ca417
] 
[Implement "ghc --abi-hash M1 M2 ..."
Simon Marlow <marlowsd@gmail.com>**20090729132125
 Ignore-this: d0a71b2fadefdfdefa6d287d28bad40f
 This prints a combined hash of the ABIs exposed by the modules listed
 on the command line.  It will be used by Cabal for generating a
 package Id based on the actual ABI of a package.
] 
[Clean GMP properly; fixes #3411
Ian Lynagh <igloo@earth.li>**20090802195759] 
[Fix permissions when installing
Ian Lynagh <igloo@earth.li>**20090802161237] 
[Fix creation of library doc index, and put the library docs in bindists
Ian Lynagh <igloo@earth.li>**20090802114202] 
[Remove a TODO item that's been done
Ian Lynagh <igloo@earth.li>**20090802103944] 
[Add a publish-docs rule
Ian Lynagh <igloo@earth.li>**20090801224155] 
[Add docs to bindists
Ian Lynagh <igloo@earth.li>**20090801222551] 
[Fix "make show" in a bindist
Ian Lynagh <igloo@earth.li>**20090801212604] 
[Make the new build system install the documentation
Ian Lynagh <igloo@earth.li>**20090801204624] 
[whitespace tweaks in rules/docbook.mk
Ian Lynagh <igloo@earth.li>**20090801195221] 
[Fix configure when alex/happy are installed to a directory containing spaces
Ian Lynagh <igloo@earth.li>**20090801182817] 
[Allow more than 64k instructions in a BCO; fixes #789
Ian Lynagh <igloo@earth.li>**20090801153203] 
[If ghci runs out of labels, panic
Ian Lynagh <igloo@earth.li>**20090801132829] 
[Fix the 64k insns overflow check in ghci, and add more checks
Ian Lynagh <igloo@earth.li>**20090801130014] 
[Fix cleaning the integer package
Ian Lynagh <igloo@earth.li>**20090730143129] 
[Fix warnings when building with the HEAD
Ian Lynagh <igloo@earth.li>**20090730143031] 
[Remove some redundant fromIntegral's
Ian Lynagh <igloo@earth.li>**20090730105532] 
[Fix space problems in ghci
Ian Lynagh <igloo@earth.li>**20090730105351
 We were making arrays with range (0, n-1) which is bad if n == 0 now
 that we are using Word types.
] 
[Make the types we use when creating GHCi bytecode better match reality
Ian Lynagh <igloo@earth.li>**20090729130911
 We were keeping things as Int, and then converting them to Word16 at
 the last minute, when really they ought to have been Word16 all along.
] 
[Add an (Outputable Word16) instance
Ian Lynagh <igloo@earth.li>**20090728134436] 
[Fix whitespace in ByteCodeAsm.lhs
Ian Lynagh <igloo@earth.li>**20090728123444] 
[Add CHECK(p), like ASSERT(p) but works even when !defined(DEBUG)
Simon Marlow <marlowsd@gmail.com>**20090729075433
 Ignore-this: a723f456e4b1eaeaa617a675da276aa2
 For inexpensive assertions
] 
[enable the x86-specific versions of atomic_inc()/atomic_dec()
Simon Marlow <marlowsd@gmail.com>**20090729075307
 Ignore-this: a271b7ade5502ec4d8444aac19f4b4c3
] 
[fix warning
Simon Marlow <marlowsd@gmail.com>**20090728103855
 Ignore-this: ee626aa5de0a2aed8f44ae1131cc341d
] 
[fix warning
Simon Marlow <marlowsd@gmail.com>**20090728102707
 Ignore-this: ad4f07a163921d17d242a9452b4bb578
] 
[fix a warning
Simon Marlow <marlowsd@gmail.com>**20090728101731
 Ignore-this: 73ead0a0004723757c0a51b56681c991
] 
[Be a bit more sensible about choosing external OccNames
Simon Marlow <marlowsd@gmail.com>**20090728100434
 Ignore-this: 4adcd661e76440deb2b4ab498ebd2d1e
 
 Instead of chr_$wchr, we now just get $wchr.  In general, when an
 OccName is system-generated, we leave it out of the final external
 name, preferring to use the name of the exported parent instead (which
 is necessarily a user-written name).
 
 Names should be no less deterministic, but should be shorter and more
 readable.
] 
[Remove old 'foreign import dotnet' code
Simon Marlow <marlowsd@gmail.com>**20090727144524
 Ignore-this: 821ebea2c3897415195318f107421472
 It still lives in darcs, if anyone wants to revive it sometime.
] 
[remove a couple of ToDos
Simon Marlow <marlowsd@gmail.com>**20090727094719
 Ignore-this: ef00fc481821dff4381ba9efcd792708
] 
[buildinfo files need a $$(wildcard)
Simon Marlow <marlowsd@gmail.com>**20090727091012
 Ignore-this: b4c3201dfa81fef32ee254dd9c955b2d
] 
[Slight tweak to avoid overflowing the command-line size in bindist
Simon Marlow <marlowsd@gmail.com>**20090727090946
 Ignore-this: 1e2ff207d03fed08576ac59f0b46c08c
 Not a real fix: if this bites us again we'll have to rethink
] 
[avoid (benign) error about overriding rules for binary-dist
Simon Marlow <marlowsd@gmail.com>**20090727090903
 Ignore-this: ffb7f7bf1290f2faf96ac177f76a1422
] 
[Give a better error message for hidden packages when building Cabal package
Ian Lynagh <igloo@earth.li>**20090726194915
 Fixes #3168
] 
[Add a -fbuilding-cabal-package flag
Ian Lynagh <igloo@earth.li>**20090726181934
 This means GHC knows whether it's building a Cabal package, or just
 Haskell sources. For example, it may wish to give different error
 messages when building a Cabal package.
] 
[Add an extension to disable n+k patterns
Ian Lynagh <igloo@earth.li>**20090725134703] 
[Fix a warning on Windows
Ian Lynagh <igloo@earth.li>**20090724221244] 
[Remove GHC's haskell98 dependency
Ian Lynagh <igloo@earth.li>**20090724210825] 
[add number of bytes to +RTS -DS leak reports
Simon Marlow <marlowsd@gmail.com>**20090724150010
 Ignore-this: 3a66585c8fd2b58ce96abab1e154fb6e
] 
[free the gc_thread structures during shutdown
Simon Marlow <marlowsd@gmail.com>**20090724145956
 Ignore-this: 25efeb189cbfb549af4550d266604f0e
] 
[Add atomic_inc()/atomic_dec(), and use them to replace gc_running_mutex
Simon Marlow <marlowsd@gmail.com>**20090724142620
 Ignore-this: d775eeaf85fd0f9064d87a0909134bc0
 This also fixes a memory leak on Windows with -threaded, because we
 were calling initMutex(&gc_running_mutex) for each GC, which allocates
 memory.
] 
[Rewrite the foreign import string parser using ReadP
Simon Marlow <marlowsd@gmail.com>**20090723152138
 Ignore-this: 1c7db770a29d48710b05e2a3d216b2a8
 And kill the duplicate one in HsSyn.Convert
] 
[point to the wiki
Simon Marlow <marlowsd@gmail.com>**20090723132345
 Ignore-this: c11300bac62ce2f56d7fc271aa26dbcd
] 
[Remove note about avoiding use of #def in libraries
Simon Marlow <marlowsd@gmail.com>**20090723111026
 Ignore-this: 8a027ed37b2d10094f7a31548aee2535
 It should be safe to use now that we aren't relying on C prototypes
 for foreign functions in via-C code.
] 
[refactorings
Simon Marlow <marlowsd@gmail.com>**20090723091230
 Ignore-this: 836feb0e819127603dd56623af6e48dc
] 
[Fix Trac #3391: make generic to/from bindings only for newly-declared types
simonpj@microsoft.com**20090723155803
 Ignore-this: bb56c2ec054397d421dce13d5eb6c73f
 
 Before this patch we were bogusly making to/from bindings for all data types
 in the TcGblEnv.  But that is wrong when we have multiple "chunks" of 
 bindings in Template Haskell.  We should start from the declarations 
 themselves.  Easy.
 
] 
[Print explicit braces and semicolons in do-notation
simonpj@microsoft.com**20090723152411
 Ignore-this: a97ddf19774d27d15a01d63787708b20
 
 By printing explicit braces we make it more likely that pretty-printed
 code will be acceptable if fed back into GHC.
 
 See http://www.haskell.org/pipermail/glasgow-haskell-users/2009-July/017554.html
 
] 
[Documentation for stand-alone deriving (Trac #3012)
simonpj@microsoft.com**20090723132558
 Ignore-this: 54445c5984594eb7f82151b2ac118695
] 
[Windows only: set the encoding on stdin to utf8
Simon Marlow <marlowsd@gmail.com>**20090723121913
 Ignore-this: d65115d9711b5fb68e77786565ef6de
 Otherwise it defaults to latin1.
] 
[Fix Trac #3012: allow more free-wheeling in standalone deriving
simonpj@microsoft.com**20090723130145
 Ignore-this: 357580b9388ccbe1da3c1da3ba90e456
 
 In standalone deriving, we now do *not* check side conditions.
 We simply generate the code and typecheck it.  If there's a type
 error, it's the programmer's problem. 
 
 This means that you can do 'deriving instance Show (T a)', where
 T is a GADT, for example, provided of course that the boilerplate
 code does in fact typecheck.
 
 I put some work into getting a decent error message.  In particular
 if there's a type error in a method, GHC will show the entire code
 for that method (since, after all, the user did not write it).
 Most of the changes are to achieve that goal.
 
 Still to come: changes in the documentation.
 
] 
[Use the ErrMsg record type
simonpj@microsoft.com**20090723130108
 Ignore-this: 7fb6dd78d3185da0c33901b8aac8d108
] 
[Stop generating redundant parens in 'deriving' code
simonpj@microsoft.com**20090723125903
 Ignore-this: 6fc82df9648a82bcf7bf6fdfa9b4dad3
 
 This makes the code printed by -ddump-deriv look prettier
 
] 
[Wibble to printing tuple sections
simonpj@microsoft.com**20090723125756
 Ignore-this: af2a1b9784f6447fea0e11d454cf082f
] 
[Fix Trac #3193: improve line number reporting for equality constraints
simonpj@microsoft.com**20090723065504
 Ignore-this: b45a68071bcaca48cad7855dccb9c9eb
 
 When reporting an error from a failed equality constraint, we were
 setting the *context* but not the *line number* in TcTyFuns.eqInstMisMatch
 As a result, the line number didn't match the context at all.  It's
 trivial to fix.
 
 I'm 99% certain this fixes #3193, but it's too complicated to
 reproduce, so I have not actually tested it.
 
] 
[Add fmapM_maybe :: Monad m => (a -> m b) -> Maybe a -> m (Maybe b)
simonpj@microsoft.com**20090723064932
 Ignore-this: db5f6319b52a5e6b5f85d76985f2a7c9
 
 This function isn't used at the moment, but Max added it, and it
 looks useful.
 
] 
[Add tuple sections as a new feature
simonpj@microsoft.com**20090723063859
 Ignore-this: d42a26fc1efff112b852b5c1135c1746
 
 This patch adds tuple sections, so that
 
        (x,,z)  means   \y -> (x,y,z)
 
 Thanks for Max Bolinbroke for doing the hard work.
 
 In the end, instead of using two constructors in HsSyn, I used
 just one (still called ExplicitTuple) whose arguments can be
        Present (LHsExpr id)
 or     Missing PostTcType
 
 While I was at it, I did a bit of refactoring too.
 
] 
[NetBSD defines _REENTRANT in its header files, so compiling ghc gives
Simon Marlow <marlowsd@gmail.com>**20090723075030
 Ignore-this: 4722c4ff0541c6080de8f433e498684
 redefinition warnings for all files that are including includes/Rts.h.
 
 Contributed by: Krister Walfridsson <krister.walfridsson@gmail.com>
] 
[includes/TSO.h: kill trailing whitespace
Samuel Bronson <naesten@gmail.com>**20090722170354
 Ignore-this: 36d4afd1a21188d604ed6b432942dcdc
] 
[Say what StgTSOBlockInfo is for, where to read about it.
Samuel Bronson <naesten@gmail.com>**20090722163011
 Ignore-this: 6c09e11f23594251cdc2db1bc642edc9
] 
[Make the Integer library used directly configurable in GHC and base
Ian Lynagh <igloo@earth.li>**20090722151048
 Rather than indirecting through an integer package
] 
[Fix cleaning with the new integer changes
Ian Lynagh <igloo@earth.li>**20090722142545] 
[Add integer-simple as a build option
Ian Lynagh <igloo@earth.li>**20090722013137] 
[Tweak whitespace
Ian Lynagh <igloo@earth.li>**20090719221303] 
[thenIO, bindIO, returnIO moved to GHC.Base
Simon Marlow <marlowsd@gmail.com>**20090722102219
 Ignore-this: 5d6c5988e0abab2c5169540aa8ddedb9
] 
[remove unused $(HscIfaceFileVersion)
Simon Marlow <marlowsd@gmail.com>**20090720145053
 Ignore-this: d725cbbde3c68673e2342b370460d87f
] 
[Choose external names more predictably
Simon Marlow <marlowsd@gmail.com>**20090720144751
 Ignore-this: 17513bc93af646108f21bbed1c8f4a3a
 
 Now, for a compiler-generated binding "x", if "x" is referred to by
 the exported "f", then it will be named "f_x" rather than something
 like "x23".  This means that hopefully
 
   - compilation will more often product the same results given the
     same input (the choice of names is not dependent on the
     non-deterministic order of bindings within the compiler).
 
   - less recompilation will be necessary after making changes
 
   - navigating Core might be a bit easier.
 
 unfortunately, compilation with -O still does not consistently produce
 the same ABI.  The simplifier sometimes does different things,
 apparently.
 
 Names will be longer, but I can't see a way around that.
] 
[Use stable ordering in the dependencies
Simon Marlow <marlowsd@gmail.com>**20090717123449
 Ignore-this: e20bac233cf6f834e69c027ff60b5b50
 Fixes another cause of wobbly interface files and unnecessary recompilation.
] 
[fall back on libffi for 'foreign import "wrapper"' if necessary
Simon Marlow <marlowsd@gmail.com>**20090716134549
 Ignore-this: e1073e1ad77e720326865a6d3c4f3790
] 
[Take account of GADTs when reporting patterm-match overlap
simonpj@microsoft.com**20090722050933
 Ignore-this: 7dcbdcb91021e83e6e6208a2e68c50c9
 
 When matching against a GADT, some of the constructors may be impossible.
 For example
        data T a where
           T1 :: T Int
           T2 :: T Bool
           T3 :: T a
 
         f :: T Int -> Int
         f T1 = 3
         f T3 = 4
 
 Here, does not have any missing cases, despite omittting T2, because
 T2 :: T Bool.
 
 This patch teaches the overlap checker about GADTs, which happily
 turned out to be rather easy even though the overlap checker needs
 a serious rewrite.
 
] 
[Fix Trac #3382: desugaring of NPats
simonpj@microsoft.com**20090720061226
 Ignore-this: 4dccdaf2b7d6428141dcf174cb455a20
 
 Max spotted that the short-cut rules for desugaring NPats (where
 we compare against a literal) were wrong now that we have overloaded
 strings.
 
] 
[Add a -fwarn-dodgy-exports flag; fixes #1911
Ian Lynagh <igloo@earth.li>**20090719200124
 This is used to control warnings that were previously unconditional.
] 
[Build terminfo if we /aren't/ on Windows, not if we /are/
Ian Lynagh <igloo@earth.li>**20090719111709] 
[Change how PACKAGES is constructed, so that everything gets cleaned properly
Ian Lynagh <igloo@earth.li>**20090718210058
 If Windows wasn't defined properly then the Win32 package wasn't being
 cleaned, as it wasn't added to PACKAGES. Now we always add everything to
 PACKAGES when CLEANING=YES.
] 
[temporarily turn off unused import warnings for the time library
Ian Lynagh <igloo@earth.li>**20090718183445] 
[Follow the split directory rename in the GHC build system rules
Ian Lynagh <igloo@earth.li>**20090718155618] 
[Add osuf to the name we use for the split dir
Ian Lynagh <igloo@earth.li>**20090718145522
 This avoids a collision between the directories we use when compiling
 multiple ways, which in turn leads to a race condition in parallel
 builds.
] 
[Temporarily turn off unused-do-bind warnings for the time package
Ian Lynagh <igloo@earth.li>**20090718134536] 
[Make ghc-cabal handle "Custom" Setup.hs files that have a configure script
Ian Lynagh <igloo@earth.li>**20090718131555] 
[Add the time library, and support for libraries in tarballs
Ian Lynagh <igloo@earth.li>**20090718121649] 
[Always serialise Int as 64bit values; fixes trac #3041
Ian Lynagh <igloo@earth.li>**20090717224203
 This means that, provided the values are small enough, files
 serialized are portable between architectures. In particular,
 .haddock files are portable.
] 
[Remove some code that has always been commented out
Ian Lynagh <igloo@earth.li>**20090717224100] 
[Fix Trac #3346: tcSimplify for LHS of RULES with type equalities
simonpj@microsoft.com**20090717155722
 Ignore-this: dfdd0f9a62d78d63276a4d558831099c
] 
[Allow mixed case in the LINE pragma; patch from squadette; fixes #1817
Ian Lynagh <igloo@earth.li>**20090717133522] 
[Comment only
simonpj@microsoft.com**20090717120154
 Ignore-this: f96b11e602fe4b311c1e466af9aa1908
] 
[Add missing case for eq_note.
t-peterj@microsoft.com**20090624134407] 
[Rename parameters to make debugging code compile.
t-peterj@microsoft.com**20090626105440] 
[Comment fix: use the same variable names in the conclusion as in the premise.
t-peterj@microsoft.com**20090618092235] 
[Typo fixes, from Alexey Mahotkin
Ian Lynagh <igloo@earth.li>**20090717010817] 
[Use names like '$fOrdInt' for dfuns (and TF instances), rather than '$f21'
Simon Marlow <marlowsd@gmail.com>**20090716125643
 Ignore-this: d0b4632cf8ed9e05b67a19aa19ab3e19
 
 2 reasons for this:
   - compilation is more predictable.  Adding or removing an instance
     is less likely to force unnecessary recompilation due to
     renumbering other dfun names.
   - it makes it easier to read Core / C-- / asm
 
 The names aren't completely deterministic.  To do that, we'd have to
 include package and module names, which would make the symbol names
 long and reduce readability.  So the compromise is that if there's a
 clash, we disambiguate by adding an integer suffix.  This is fairly
 unlikely in practice unless you're using overlapping instances.
 
 Type family instances are handled in the same way, with the same
 disambiguation strategy.
] 
[Use a stable ordering for the export list in the interface
Simon Marlow <marlowsd@gmail.com>**20090716122601
 Ignore-this: 847dd7adc8b52e56f28d2478c78c925
 The export list was ordered according to the whim of FastStrings,
 which meant that interface fingerprints could change for no good
 reason, causing spurious recompilation.
] 
[Don't put all of $CFLAGS into $SRC_CC_OPTS
Ian Lynagh <igloo@earth.li>**20090716131309
 Instead, we just put the flags we need in there (e.g. -m64 on OS X 64).
 This fixes a problem found by Simon M, where we were compiling
 everything with -g, leading to a bloated RTS.
] 
[Move showOpt into DynFlags
Ian Lynagh <igloo@earth.li>**20090716005314] 
[Make the --info values printable with "ghc --print-foo"; trac #3122
Ian Lynagh <igloo@earth.li>**20090716001718
 Also, libdir is now part of the --info output, so this subsumes the old
 --print-libdir flag.
 The mode parsing was getting rather adhoc, so I've tidied it up a bit
 in the process.
] 
[whitespace only
Simon Marlow <marlowsd@gmail.com>**20090716104217
 Ignore-this: 38cff291d9ef15c30e3ed685ffc3c9f9
] 
[refactor: use packageConfigId in place of mkPackageId . package
Simon Marlow <marlowsd@gmail.com>**20090716104145
 Ignore-this: f3d73e7bd1b307a67d26585c49f3d89f
] 
[Fix a flag name in the docs
Ian Lynagh <igloo@earth.li>**20090714165943] 
[Add the -fno-shared-implib flag
Ian Lynagh <igloo@earth.li>**20090714165631
 Patch from
     Max Bolingbroke <batterseapower@hotmail.com>
 Rerecorded to avoid conflicts.
] 
[Derived Foldable instances should use Data.Foldable.foldr
m.niloc@gmail.com**20090711130647
 Ignore-this: e3eb841e9535a842a98bb1ae0532c6e8
] 
[remove Solaris-specific hacks, now unnecessary
Simon Marlow <marlowsd@gmail.com>**20090713083524
 Ignore-this: 500077008e463532e0677ee82f5284bb
] 
[Simplify timestamp restoration
Matthias Kilian <kili@outback.escape.de>**20090711100244
 Ignore-this: 7eaede224befa6b5368c91b92366211
] 
[FIX #3272
Manuel M T Chakravarty <chak@cse.unsw.edu.au>**20090714054559
 Ignore-this: 225fe4d82d4eed02e9b1377687661bac
] 
[Fix warnings
Roman Leshchinskiy <rl@cse.unsw.edu.au>**20090713092032
 Ignore-this: 3631b87164fc54d82e3a02875dc08f7d
] 
[Separate length from data in DPH arrays
Roman Leshchinskiy <rl@cse.unsw.edu.au>**20090713044212
 Ignore-this: aa2cc3b5ae43bd2c493ce4b330c883cd
] 
[Stop using -fno-warn-unused-do-bind when compiling the libraries
Ian Lynagh <igloo@earth.li>**20090709160422
 They're now fixed to not generate those warnings
] 
[Remove maybePrefixMatch, using stripPrefix instead
Ian Lynagh <igloo@earth.li>**20090709160412
 We already require GHC 6.8 to build, and that included stripPrefix
 in Data.List.
] 
[TFs: FIX #2203 (second half)
Manuel M T Chakravarty <chak@cse.unsw.edu.au>**20090710064834
 Ignore-this: 46a46feaa73f74feb08524b9e7547414
] 
[TFs: Fix should_compile/Simple8
Manuel M T Chakravarty <chak@cse.unsw.edu.au>**20090710042728
 Ignore-this: 471ab67e3df1c5245921be5286a45f93
] 
[workaround new Cygwin bash CRLF behaviour
Simon Marlow <marlowsd@gmail.com>**20090709132850
 Ignore-this: 5cfa2cc9d776ebe315c0f6ad7ab56d98
] 
[Use /usr/bin/test if it exists, and fix test syntax.
Simon Marlow <marlowsd@gmail.com>**20090709124616
 Ignore-this: 83a75ba7c3ce2a1d02bddb7bfe414bfe
 Should fix Solaris build failures
] 
[Allow mixed case pragmas; #1817. Patch from squadette
Ian Lynagh <igloo@earth.li>**20090709153737
 This patch allow you to use "Language CPP", or even "LaNgUaGe CPP",
 if you wish, as the manual claims you can.
] 
[don't create inplace/bin/ghc-<version>
Simon Marlow <marlowsd@gmail.com>**20090706092031
 Ignore-this: 2584d7bf56e77b27ca5b7b557c152c5e
] 
[Fix ignored-monadic-result warnings
Ian Lynagh <igloo@earth.li>**20090707181857] 
[Fix an unused import warning
Ian Lynagh <igloo@earth.li>**20090707144706] 
[Fix unused import warnings
Ian Lynagh <igloo@earth.li>**20090707143216] 
[Fix unused import warnings
Ian Lynagh <igloo@earth.li>**20090707133537] 
[When exporting F(..), all the children of F are also exported
Ian Lynagh <igloo@earth.li>**20090707133427
 This fixes the unused imports warning when
     Foo (F(x,y,z))
 is imported and
     Foo (F(..))
 is exported.
] 
[Remove unused imports
Ian Lynagh <igloo@earth.li>**20090707121548] 
[Major patch to fix reporting of unused imports
simonpj@microsoft.com**20090706112503
 Ignore-this: 3b5ecdd880474493d73bdbdc0fa0b782
 
 This patch, joint work between and Ian and Simon, fixes Trac #1074
 by reporting unused import declarations much more accuratly than 
 before.  The specification is described at 
 
 http://hackage.haskell.org/trac/ghc/wiki/Commentary/Compiler/UnusedImports
 
 The implementation is both easier to understand than before, and shorter
 too.  
 
 Also fixed are #1148, #2267
 
 Also fixed is -ddump-minimal imports, which now works properly, fixing 
 Trac #1792.
 
 
] 
[Trim unused imports detected by new unused-import code
simonpj@microsoft.com**20090706112201
 Ignore-this: c6ca46d3a750c1cd1d58ea2c0de9f14f
] 
[Avoid unnecessary recompilation after ./configure (helps #3228)
Simon Marlow <marlowsd@gmail.com>**20090707085040
 Ignore-this: f8b3e7a2a96bc23cd29505ab9c8dbd7d
 We cache the old versions of files generated by configure, so that if
 configure touches the file without changing it, we can detect that and
 restore the timestamp.
] 
[check for tabs in compiler/ghc.cabal.in (#3344)
Simon Marlow <marlowsd@gmail.com>**20090707081845
 Ignore-this: 6073db47eafd52e13e76c58ef738afcf
] 
[remove tabs
Simon Marlow <marlowsd@gmail.com>**20090707081823
 Ignore-this: 3d65831fc019f76cefac03291904842a
] 
[fix cleaning of libraries (now 'make clean' in libraries/* works again)
Simon Marlow <marlowsd@gmail.com>**20090703114638
 Ignore-this: b3af731d50ff5bfbd453f94aa40cb92c
] 
[FIX #2677
Manuel M T Chakravarty <chak@cse.unsw.edu.au>**20090707055442
 Ignore-this: e224dd09d0d1c9ec4f3b46c7accb8d57
] 
[Update driver/Makefile for the new build system
Ian Lynagh <igloo@earth.li>**20090705204041] 
[Fix generational GC bug (#3348)
Simon Marlow <marlowsd@gmail.com>**20090706112227
 Ignore-this: 5938338efa0ad1550968c664a5a76f31
] 
[Windows fixes to build system: use the 'find' and 'sort' found by configure
simonpj@microsoft.com**20090706103413
 Ignore-this: a96197917f388a637118bafefb427495
 
 The build system should use 'find' and 'sort' that are discovered by
 configure, not the ones in your path.  On Windows the ones in your path
 might well be the non-Unixy Windows versions.
 
 This patch fixes the ones I tripped over. There may be more.
 
] 
[Follow Cabal changes
Ian Lynagh <igloo@earth.li>**20090705180414] 
[Update TODO list
Ian Lynagh <igloo@earth.li>**20090705165009] 
[Make -fext-core a dynamic flag (it was a static flag)
Ian Lynagh <igloo@earth.li>**20090705132420] 
[Update a few points about shared libs in other sections
Duncan Coutts <duncan@well-typed.com>**20090704212212
 And add links to the new shared libs section.
] 
[Document -dynload flag. Also add it and -shared to the flags reference.
Duncan Coutts <duncan@well-typed.com>**20090704212119] 
[Add new section on using shared libs
Duncan Coutts <duncan@well-typed.com>**20090704212003] 
[Document foreign import prim in the user guide
Duncan Coutts <duncan@well-typed.com>**20090704180547
 Basically just stat that it exists and refer to the ghc dev wiki
 for the details, because we don't really want people using it.
] 
[For now, use -fno-warn-unused-do-bind when building the libraries
Ian Lynagh <igloo@earth.li>**20090704210654] 
[Make changes to -fwarn-unused-do-bind and -fwarn-wrong-do-bind suggested by SPJ
Max Bolingbroke <batterseapower@hotmail.com>**20090702150943
 Ignore-this: 595368298d2e11623c0bd280ff89d8de
] 
[Support for -fwarn-unused-do-bind and -fwarn-wrong-do-bind, as per #3263
Max Bolingbroke <batterseapower@hotmail.com>**20090701200344
 Ignore-this: 511117ffc10d4b656e530b751559b8b8
] 
[Improved infrastructure for fast-rebuilding of parts of the tree
Simon Marlow <marlowsd@gmail.com>**20090703074527
 Ignore-this: ab348d0988d8bbc28c2b4babbd6bbfb8
 
 e.g.
 
   cd compiler
   make FAST=YES stage1/build/HscTypes.o
 
 builds just the specified .o file, without rebuilding dependencies,
 and omitting some of the makefile phases.  FAST=YES works anywhere, to
 omit depenencies and phases.  'make fast' is shorthand for 'make
 all FAST=YES'.
] 
[Fix Trac #3342: missed zonking in TcHsSyn
simonpj@microsoft.com**20090702124331
 Ignore-this: 9b97b2142dfc665b503f59df7c55dd17
 
 The type in a ViewPat wasn't being zonked.  Easily fixed.
 
] 
[Type synonym families may be nullary
Manuel M T Chakravarty <chak@cse.unsw.edu.au>**20090702084826
 Ignore-this: bcfe6ed62c901206daf5a5088890bbea
] 
[New syntax for GADT-style record declarations, and associated refactoring
simonpj@microsoft.com**20090702094657
 Ignore-this: bd9817230d3773b3b01fae3d7f04c57d
 
 The main purpose of this patch is to fix Trac #3306, by fleshing out the
 syntax for GADT-style record declraations so that you have a context in 
 the type.  The new form is
    data T a where
      MkT :: forall a. Eq a => { x,y :: !a } -> T a
 See discussion on the Trac ticket.
 
 The old form is still allowed, but give a deprecation warning.
 
 When we remove the old form we'll also get rid of the one reduce/reduce
 error in the grammar. Hurrah!
 
 While I was at it, I failed as usual to resist the temptation to do lots of
 refactoring.  The parsing of data/type declarations is now much simpler and
 more uniform.  Less code, less chance of errors, and more functionality.
 Took longer than I planned, though.
 
 ConDecl has record syntax, but it was not being used consistently, so I
 pushed that through the compiler.
 
] 
[White space only
simonpj@microsoft.com**20090702094627
 Ignore-this: 19f654cbf371c8dcc6517fd4934855b4
] 
[Comments only
simonpj@microsoft.com**20090702094531
 Ignore-this: 384fc2729c7c50a1680775a1f9ff89e4
] 
[Look through Notes when matching
simonpj@microsoft.com**20090702094444
 Ignore-this: 7daea81e905ec6061d3e0fd588d7e61b
] 
[FIX #3197
Manuel M T Chakravarty <chak@cse.unsw.edu.au>**20090702070905
 Ignore-this: ebf829f0ae025e82bccdfa4345828ffe
] 
[Fix #2197 (properly this time)
Simon Marlow <marlowsd@gmail.com>**20090701122354
 Ignore-this: 39b6e4b0bcdd8c2f4660f976b7db768d
 
 $ ./inplace/bin/ghc-stage2 --interactive
 GHCi, version 6.11.20090701: http://www.haskell.org/ghc/  :? for help
 ghc-stage2: GHCi cannot be used when compiled with -prof
 [1]    32473 exit 1     ./inplace/bin/ghc-stage2 --interactive
] 
[make GhcProfiled work, and add a "prof" flavour to build.mk
Simon Marlow <marlowsd@gmail.com>**20090701114211
 Ignore-this: 386d347e4ad8b6c2bd40a2ba7da31ba6
 
 Building a profiled GHC is as simple as adding
 
 GhcLibWays += p
 GhcProfiled = YES
 
 to your build.mk and saying 'make'.  Then you have a profiled
 inplace/bin/ghc-stage2.
] 
[remove unnecessary $(RM)s
Simon Marlow <marlowsd@gmail.com>**20090701110609
 Ignore-this: f326ec8931d0d484a66b67ce1270cc6e
] 
['make html' in a library builds the Haddock docs
Simon Marlow <marlowsd@gmail.com>**20090630111137
 Ignore-this: 781bf10e2d4bca23b7f70c6f0465d120
] 
[fix GC bug introduced with the C finalizer support
Simon Marlow <marlowsd@gmail.com>**20090630080834
 Ignore-this: 3567e3adb5ae4a5dcbce81733487f348
] 
[Add a configure test for whether or not __mingw_vfprintf exists
Ian Lynagh <igloo@earth.li>**20090627150501] 
[Fix #3319, and do various tidyups at the same time
Simon Marlow <marlowsd@gmail.com>**20090626095421
 Ignore-this: ea54175f6bd49e101d7b33392764f643
  - converting a THSyn FFI declaration to HsDecl was broken; fixed
  - pretty-printing of FFI declarations was variously bogus; fixed
  - there was an unused "library" field in CImport; removed
] 
[rename cache variable to keep recent autoconfs happy
Ross Paterson <ross@soi.city.ac.uk>**20090626131410
 Ignore-this: 187091bbe78f2b14402162acfb98180f
] 
[TAG 2009-06-25
Ian Lynagh <igloo@earth.li>**20090625155528] 
Patch bundle hash:
bc96e1d367a9b638acc2fd6ebd38991bec3dd684