summaryrefslogtreecommitdiff
path: root/rules
Commit message (Collapse)AuthorAgeFilesLines
* Fix a commentIan Lynagh2013-06-021-1/+1
| | | | Make expands things even in comments
* Add a commentIan Lynagh2013-06-011-0/+4
|
* Don't link base/rts/... into C wrapper programsIan Lynagh2013-05-161-2/+2
|
* Make dynamic GHC no Windows installable tooIan Lynagh2013-05-161-11/+32
| | | | | We need different paths in the wrapper, as teh installed tree is a different shape to the build tree.
* Don't try to build bindist wrappers on WindowsIan Lynagh2013-05-151-0/+2
|
* Link to the right RTS whenever we build a .dll on WindowsIan Lynagh2013-05-151-12/+0
| | | | | When GHCi makes temporary DLLs, those also need to be linked against the right RTS, or we won't be able to load them.
* Define the right RTS config in the Windows dyn wrapper programsIan Lynagh2013-05-141-0/+17
| | | | | | This is particularly important as without it validate fails, as it tries to pass RTS options to haddock, and with the default RTS config those options aren't permitted.
* Fix the GHC package DLL-splittingIan Lynagh2013-05-142-1/+9
| | | | | | | There's now an internal -dll-split flag, which we use to tell GHC how the GHC package is split into 2 separate DLLs. This is used by Packages.isDllName to determine whether a call is within the same DLL, or whether it is a call to another DLL.
* Simplify ghc-cabalIan Lynagh2013-05-142-2/+2
| | | | | | It now consistently takes directory and distDirectory as its first 2 arguments. Also, it only supports configuring 1 package at a time now (we weren't using the ability to configure more than one at once).
* Merge win:/cygdrive/c/ghc/git/dtIan Lynagh2013-05-122-1/+32
|\
| * More work towards dynamic programs on WindowsIan Lynagh2013-05-121-1/+20
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Dynamic GHC is now working in-place, but pathologically slow due to the DLL split. (GHC assumes that all intra-package calls are in the same DLL, but that isn't true when we split the GHC package into 2 DLLs. That means that GHC's startup time is around 22 seconds, as it is doing run-time linking). Also, ghci isn't actually working yet: $ inplace/bin/ghc-stage2 --interactive GHCi, version 7.7.20130512: http://www.haskell.org/ghc/ :? for help Loading package ghc-prim ... <command line>: can't load .so/.DLL for: HSghc-prim-0.3.1.0.dll (addDLL: could not load DLL) ghc-stage2.exe: HSghc-prim-0.3.1.0: The specified module could not be found.
| * We actually need to use -threaded/-debug when linking /all/ DLLsIan Lynagh2013-05-121-0/+12
| | | | | | | | Not just base, integer-gmp and ghc-prim.
* | Generate dependencies for .cmm files properlyIan Lynagh2013-05-092-6/+13
|/
* Use -rpath flags on all Elf OSes, not just LinuxIan Lynagh2013-04-271-1/+1
| | | | Patch from pgj; part of #7819.
* When linking with $ORIGIN rpaths, use the "-z origin" linker flag tooIan Lynagh2013-04-271-1/+1
| | | | | On FreeBSD, one needs use "-z origin" in order to enable resolution of $ORIGIN in RPATH. Part of #7819.
* pass the correct -Dxxx_HOST_ARCH and -Dxxx_HOST_OS to hsc2hs; Fixes #7761.Ian Lynagh2013-04-211-2/+0
| | | | Patch from Stephen Blackheath.
* Fix "make 2" in ghc/Ian Lynagh2013-04-211-3/+12
| | | | The ghc-stage1_INPLACE variable wasn't being defined
* Use different exeext variables for each stage; fixes #7709Ian Lynagh2013-04-201-1/+1
| | | | | Currently they are all set to the same value, but when cross-compiling they could be set to different values.
* Follow build system changes in integer-gmpIan Lynagh2013-04-203-2/+10
| | | | Also a couple of other fixes and sanity checks along the way.
* Some build system refactoringIan Lynagh2013-04-208-27/+27
|
* Add a check that the Windows DLL split is OK; fixes #7780Ian Lynagh2013-04-071-0/+11
|
* Build system fix: really print the number of symbols in DLLsIan Lynagh2013-04-061-1/+1
|
* When making a DLL, print the number of symbols that are in itIan Lynagh2013-04-061-0/+1
|
* Tweak the _HC_OPTS variablesIan Lynagh2013-03-232-6/+9
| | | | | We no longer pass -hisuf/-osuf flags to "ghc -M". Doing so didn't really make sense with the way the -dep-suffix flags now work.
* Remove some commented out build system codeIan Lynagh2013-03-191-11/+0
|
* Fix building when $(CC_STAGE0) contains spacesIan Lynagh2013-03-191-2/+2
| | | | Spotted by Bill Tutt
* Optimistically assume that LD_LIBRARY_PATH works for all platformsIan Lynagh2013-03-191-3/+3
| | | | | | other than Windows and OS X. It's known to work on Linux and FreeBSD.
* Fix ohi-sanity-check when we have ways other than v and dyn enabledIan Lynagh2013-03-171-1/+5
| | | | | It thought that /all/ ways should create the dyn o/hi files, but only the v way does.
* Partial support for dynamic ghc on WindowsIan Lynagh2013-03-161-1/+9
|
* By default, use the dynamic way for programs in the GHC treeIan Lynagh2013-03-154-6/+6
| | | | | In particular, this means that GHCi will use DLLs, rather than loading object files itself.
* Finish adding support for 2 DLLs in the ghc package; fixes #5987Ian Lynagh2013-03-131-3/+17
|
* Merge branch 'master' of win:c:/ghc/git/cygwin/.Ian Lynagh2013-03-121-8/+33
|\
| * Fix building the GHC package DLL on WindowsIan Lynagh2013-03-111-8/+33
| | | | | | | | | | | | We now put a handful of modules in a separate DLL. For now the list is hand-written, but we could automate it in the future.
* | Update CabalIan Lynagh2013-03-111-7/+6
| |
* | Fix construction of ALL_ALEX_OPTS and ALL_AS_OPTSMax Bolingbroke2013-03-102-2/+2
|/ | | | Signed-off-by: David Waern <david.waern@gmail.com>
* Don't pass -dynamic-too to GHC when compiling C filesIan Lynagh2013-03-091-1/+1
|
* Add some extra sanity checking to the (l)hs->o rulesIan Lynagh2013-03-091-0/+17
|
* Automatically add the $(exeext) to program namesIan Lynagh2013-03-033-11/+25
| | | | | | | We now define _PROGNAME, and _PROG is automatically defined with $(exeext). This will shortly automatically use the right exeext depending on what stage it is being compiled with (exeext may be different for different stages when cross-compiling).
* Automatically define _INPLACE variables for perl programs tooIan Lynagh2013-03-031-2/+7
| | | | and use them for split
* Remove the _INSTALL_IN support in the build systemIan Lynagh2013-03-031-10/+0
| | | | | It doesn't seem to do anything that _INSTALL and _INSTALL_INPLACE can't do.
* Use unlit_INPLACE rather than UNLIT in the build systemIan Lynagh2013-03-031-2/+2
|
* Use more of the automatically generated INPLACE variablesIan Lynagh2013-03-034-9/+9
|
* Define $(prog)_INPLACE variables automatically, and start using themIan Lynagh2013-03-033-8/+14
|
* Add PACKAGES_STAGE0 to build-dirs too; fixes #7700Ian Lynagh2013-03-031-1/+4
| | | | | | | | This is sometimes needed when cross-compiling, as some packages may be built in stage 0 but not stage 1. In order to make everything work out, this also removes the requirement that the build-dirs are in dependency order
* Give hsc2hs different options in different stages; fixes #7705Ian Lynagh2013-03-021-2/+3
| | | | Based on patch from Stephen Blackheath.
* Rename $(Windows) to $(Windows_Host)Ian Lynagh2013-03-012-3/+3
|
* Change how the build system handles packagesIan Lynagh2013-03-013-49/+44
| | | | | This makes the build system a little simpler, and in particular will make it easier to handle the changes needed for cross-compilation.
* Tweak the sed command so that it works on OSXIan Lynagh2013-02-231-1/+11
| | | | | Using { } with sed on OS X requires using newlines rather than semicolons. I've rewritten the sed so that no { } are necessary.
* Add another chapter to the hi-rule novelIan Lynagh2013-02-221-1/+19
|
* Make the build system use -dynamic-too if both way v and dyn are being builtIan Lynagh2013-02-225-8/+38
|
View Lorry Controller Status