August 2006 - Yhc - Haskell.org

still not building on FreeBSD
by David Frech 31 Aug '06

31 Aug '06

latest error: gcc -o inst/bin/yhi src/runtime/BCKernel/process.o src/runtime/BCKernel/external.o src/runtime/BCKernel/hashtable.o src/runtime/BCKernel/heap.o src/runtime/BCKernel/hsffi.o src/runtime/BCKernel/info.o src/runtime/BCKernel/integer.o src/runtime/BCKernel/iofuncs.o src/runtime/BCKernel/jonkers.o src/runtime/BCKernel/main.o src/runtime/BCKernel/make.o src/runtime/BCKernel/mark.o src/runtime/BCKernel/module.o src/runtime/BCKernel/mutator.o src/runtime/BCKernel/primitive.o src/runtime/BCKernel/pretty.o src/runtime/BCKernel/sanity.o src/runtime/BCKernel/stopcopy.o src/runtime/BCKernel/profile.o src/runtime/BCKernel/foreign.o src/runtime/BCKernel/thread.o src/runtime/BCKernel/stable.o src/runtime/BCKernel/builtin/Array.o src/runtime/BCKernel/builtin/FFI.o src/runtime/BCKernel/builtin/IO.o src/runtime/BCKernel/builtin/Prelude.o src/runtime/BCKernel/builtin/Concurrent.o src/runtime/BCKernel/builtin/PackedString.o src/runtime/BCKernel/builtin/RuntimeAPI.o src/runtime/BCKernel/builtin/System.o depends/ctypes/libffi/src/prep_cif.o depends/ctypes/libffi/src/cfield.o depends/ctypes/libffi/src/x86/ffi.o depends/ctypes/libffi/src/x86/sysv.o -L/usr/local/lib -lgmp -ldl -lm -lpthread /usr/bin/ld: cannot find -ldl scons: *** [inst/bin/yhi] Error 1 scons: building terminated because of errors. The issue, I think, is that, unlike on Linux, the dl* functions are in libc, not libdl. - David -- The reasonable man adapts himself to the world; the unreasonable one persists in trying to adapt the world to himself. Therefore all progress depends on the unreasonable man. -- George Bernard Shaw

2 2

Yhc Core injection
by Neil Mitchell 30 Aug '06

30 Aug '06

Hi, I know we have a -core output mode, what I was wondering is how easy it would be to inject some kind of PosLambda like code into the compiler, skipping the type checker and the desugarer? The reason that I ask is that I am currently writing a pile of Haskell manipulations, which make the Haskell code simpler, first order, remove redundant stuff etc. The transformations are for another purpose, but it might be interesting to see if they have an effect on the optimisation potential of Yhc. The reason Yhc is attractive for this is that the code is not type safe after its been reduced and simplified - for Yhc I suspect this won't be a problem? The code is "type safe" still, just not hinley milner type safe (or any kind of non-dependant type safe). Thanks Neil

1 0

Verified compiler
by William DUCK 27 Aug '06

27 Aug '06

Hello Misters, As you are compiler experts, I am asking your help. I am attempting to build up a team. I believe it's time for a full-fledge verified OS. If somebody is interested in functional programming and in formal methods to help to implement a verified compiler , he is welcome :) Here are two examples of a compiler formalisation : http://pauillac.inria.fr/~xleroy/compcert-backend/ http://www.score.is.tsukuba.ac.jp/~okuma/vc/ Another intereesting link : http://pes.cs.tu-berlin.de/cocv2007/ And finally the ultimate goal : http://www.cse.ogi.edu/~hallgren/House/ http://web.cecs.pdx.edu/~rebekah/presentations/ssp2005abs.pdf http://web.cecs.pdx.edu/~rebekah/papers/icfp05_bits.pdf Haskell also depends on a complicated run-time system. Such run-time systems are often unverified. Implementing a minimal, high-assurancerun-time system is a topic for future work. http://www.oreilly.com/catalog/opensources/book/linus.html GCC Unix itself is a great success story in terms of portability. The Unix kernel, like many kernels, counts on the existence of C to give it the majority of the portability it needs. Likewise for Linux. For Unix the wide availability of C compilers on many architectures made it possible to port Unix to those architectures. So Unix underscores how important compilers are. The importance of compilers was one reason I chose to license Linux under the GNU Public License (GPL). The GPL was the license for the GCC compiler. I think that all the other projects from the GNU group are for Linux insignificant in comparison. GCC is the only one that I really care about. A number of them I hate with a passion; the Emacs editor is horrible, for example. While Linux is larger than Emacs, at least Linux has the excuse that it needs to be. But basically compilers are really a fundamental need. Now that the Linux kernel follows a generally portable design, at least for reasonably sane architectures, portability should be possible as long as a reasonably good compiler is available. For the upcoming chips I don't worry much about architectural portability when it comes to the kernel anymore; I worry about the compilers. Intel's 64-bit chip, the Merced, is an obvious example, because Merced is very different for a compiler. So the portability of Linux is very much tied to the fact that GCC is ported to major chip architectures. If you want to contact me, my mail is guillaume_dot_fortaine_at_wanadoo_dot_fr I will set up a mailing-list, a web server, a wiki and an IRC Best Regards, Guillaume FORTAINE

2 1

I can't get foreign export to compile
by Isaac 20 Aug '06

20 Aug '06

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 (the only reason I'm trying _this_ is to test if the Yhc implementation of concurrency described on the wiki works properly; Yhc seems to compile foreign import successfully, though I haven't yet tried running the bytecode it generates for that) - -------- import Foreign.C (CInt) return3 :: IO CInt return3 = return 3 foreign export ccall return3 :: IO CInt main = putStrLn "okay" - -------- I've tried commenting out various combinations of the type-signatures, reordering the definition and the foreign export, making the return be IO or not, () or not... ghc -fffi only complains when I comment out the type-signature on the foreign export line, as expected based on the FFI spec. This is always just yhc -c dying, with various error messages. Isaac -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.2.2 (GNU/Linux) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iD8DBQFE5a46HgcxvIWYTTURAjBPAKCF8fyuZCrwxOdK4hVsdDljMFWwZwCcCMxh ULKutHs0sNcQreGf86fauC8= =pXnT -----END PGP SIGNATURE-----

3 3

"default" keyword bugs
by Isaac 18 Aug '06

18 Aug '06

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 I just tried some uses of the "default" keyword with Yhc that failed, and I didn't find any existing Yhc test-cases for this keyword. These all run with Hugs. Fails with Yhc: - ---------- import Ratio ( Ratio -- :: * -> * ) default (Ratio Integer) main = putStrLn "okay" - ---------- but works with Yhc: - ---------- import Ratio ( Ratio -- :: * -> * ) type RatioInteger = Ratio Integer default (RatioInteger) main = putStrLn "okay" - ---------- Fails (differently) with Yhc: - ---------- default(Rational) main = print (0.1 * 3 == 0.3) - ---------- but works with Yhc: - ---------- default(Rational) main = print ((0.1 :: Rational) * 3 == 0.3) - ---------- Isaac P.S. I know that type Rational = Ratio Integer -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.2.2 (GNU/Linux) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iD8DBQFE5ZhcHgcxvIWYTTURAl9dAKCLUs9q1sxFVdWee2DJK2LDZf7HjgCgs/aK EI0FuUgMqK6OdAlJ+BP6zfE= =37e3 -----END PGP SIGNATURE-----

2 3

Build error, when no C compiler
by Neil Mitchell 17 Aug '06

17 Aug '06

Hi, At home now, of course with no Visual Studio, and as a result: scons: Reading SConscript files ... Checking for architecture... Windows implies x86 Checking for operating system... win32 Checking for Scons binary... "C:\Python24\scons.bat" Checking for Darcs binary... "c:\bin\darcs.exe" Checking for Darcs version... Found version 1.0.7 (release) . Checking for Subversion binary... "C:\Program Files\Subversion\bin\svn.exe" Checking for Subversion version... Found version 1.3.2. Checking for GHC binary... "C:\ghc\ghc-6.4.2\bin\ghc.exe" Checking for GHC version 6.4.1 or later... Found version 6.4.2. Checking whether byte ordering is bigendian... failed Checking size of char... failed Checking size of short... failed Checking size of int... failed Checking size of long... failed Checking size of long long... failed Checking size of float... failed Checking size of double... failed Checking size of void*... failed Checking for pthread_create() in C library pthread... no Generating config.h with the following settings: WORDS_BIGENDIAN None HAVE_LIBPTHREAD 0 SIZEOF_LONG_LONG None SIZEOF_LONG None WIN32 1 SIZEOF_VOIDP None SIZEOF_CHAR None SIZEOF_DOUBLE None SIZEOF_FLOAT None SIZEOF_INT None HAVE_GCC_LABELS 0 SIZEOF_SHORT None TypeError: int argument required: File "SConstruct", line 38: configure(env, Configure, ARGUMENTS) # Configure File "C:\Neil\yhc-devel\build\configure.py", line 97: config_h_build('config.h', 'config.h.new.in', vals) File "C:\Neil\yhc-devel\build\config_h.py", line 21: text = config_h_in.read() % defines I guess we should check for a C compiler first. Just for note, I'll probably try installing gcc rather than Visual Studio (in fact gcc might already be installed here) - will gcc + Windows work at all? Thanks Neil

2 2

darcs patch: Fix some warnings.
by Andrew Wilkinson 15 Aug '06

15 Aug '06

Tue Aug 15 11:49:03 BST 2006 Andrew Wilkinson <aw(a)cs.york.ac.uk> * Fix some warnings.

2 1

BOOM!!!!!
by Thomas Davie 14 Aug '06

14 Aug '06

gcc -o depends/ctypes/libffi/src/x86/ffi.o -c -DX86 -Wall -I. -I/usr/ local/include -Idepends/ctypes/libffi/include -Idepends/ctypes/libffi/ src/x86 -Idepends/ctypes/libffi/include -Idepends/ctypes/libffi/src/ x86 -Idepends/ctypes/libffi depends/ctypes/libffi/src/x86/ffi.c depends/ctypes/libffi/src/x86/ffi.c: In function 'ffi_prep_args': depends/ctypes/libffi/src/x86/ffi.c:108: warning: implicit declaration of function 'bcopy' depends/ctypes/libffi/src/x86/ffi.c:108: warning: incompatible implicit declaration of built-in function 'bcopy' depends/ctypes/libffi/src/x86/ffi.c: In function 'ffi_prep_args_raw': depends/ctypes/libffi/src/x86/ffi.c:395: warning: incompatible implicit declaration of built-in function 'bcopy' gcc -I. -I/usr/local/include -Idepends/ctypes/libffi/include - Idepends/ctypes/libffi/src/x86 -Idepends/ctypes/libffi/include - Idepends/ctypes/libffi/src/x86 -Idepends/ctypes/libffi -c -o depends/ ctypes/libffi/src/x86/sysv.o depends/ctypes/libffi/src/x86/sysv.S depends/ctypes/libffi/src/x86/sysv.S:38:Unknown pseudo-op: .type depends/ctypes/libffi/src/x86/sysv.S:38:Rest of line ignored. 1st junk character valued 102 (f). depends/ctypes/libffi/src/x86/sysv.S:131:Unknown pseudo-op: .size depends/ctypes/libffi/src/x86/sysv.S:131:Rest of line ignored. 1st junk character valued 102 (f). depends/ctypes/libffi/src/x86/sysv.S:136:Unknown pseudo-op: .type depends/ctypes/libffi/src/x86/sysv.S:136:Rest of line ignored. 1st junk character valued 102 (f). depends/ctypes/libffi/src/x86/sysv.S:158:Missing operand value assumed absolute 0. depends/ctypes/libffi/src/x86/sysv.S:158:junk `[.-1b]' after expression depends/ctypes/libffi/src/x86/sysv.S:159:junk `@PLT' after expression depends/ctypes/libffi/src/x86/sysv.S:194:Unknown pseudo-op: .size depends/ctypes/libffi/src/x86/sysv.S:194:Rest of line ignored. 1st junk character valued 102 (f). depends/ctypes/libffi/src/x86/sysv.S:206:Unknown pseudo-op: .type depends/ctypes/libffi/src/x86/sysv.S:206:Rest of line ignored. 1st junk character valued 102 (f). depends/ctypes/libffi/src/x86/sysv.S:259:Unknown pseudo-op: .size depends/ctypes/libffi/src/x86/sysv.S:259:Rest of line ignored. 1st junk character valued 102 (f). depends/ctypes/libffi/src/x86/sysv.S:262:unknown section type: @progbits depends/ctypes/libffi/src/x86/sysv.S:262:Rest of line ignored. 1st junk character valued 46 (.). scons: *** [depends/ctypes/libffi/src/x86/sysv.o] Error 1 scons: building terminated because of errors. this is not a system V machine! this is an x86-darwin machine. Bob

2 1

Buildbot setup
by Thomas Davie 13 Aug '06

13 Aug '06

Hi, Can I get two buldbot username and passwords One should be tatd2/pass and will have ppc-darwin and i386-darwin machines building. The other should be twig/pass and will have centos-i386 building. Bob

1 0

Re: [Yhc] Problem building on Mac OS X
by Brent Fulgham 13 Aug '06

13 Aug '06

On Jul 2, 2006, at 10:20 PM, Goetz Isenmann wrote: >> File "./build/tools/haskell.py", line 163: >> haskellLibrary = SCons.Builder.Builder(action = >> SCons.Defaults.ArAction, >> Chaz:~/Projects/Yhc/yhc-devel brent$ > > I had to change > > { > hunk ./build/tools/haskell.py 163 > - haskellLibrary = SCons.Builder.Builder(action = > SCons.Defaults.ArAction, > + haskellLibrary = SCons.Builder.Builder(action = > SCons.Action.Action("$ARCOM"), > } > > to build with scons 0.96.92 (on freebsd). > -- > Goetz Isenmann Thank you -- this helped quite a bit. I did get a bit further: ghc -i./src/compiler98 -idepends/cpphs -c depends/cpphs/Text/ ParserCombinators/HuttonMeijer.hs -o depends/cpphs/Text/ ParserCombinators/HuttonMeijer.o Getting dependencies for src/compiler98/Bind.hs. An error ocurred running `ghc -M -optdep-f -optdepsrc/ compiler98/.Bind.hs.dep src/compiler98/Bind.hs`: >Could not find module `RenameLib': > use -v to see a list of the files searched for > (imported from src/compiler98/Bind.hs) > Dependencies will not be calculated. ghc -i./src/compiler98 -idepends/cpphs -c src/compiler98/Bind.hs -o src/compiler98/Bind.o src/compiler98/Bind.hs:11:0: Failed to load interface for `State': Could not find module `State': use -v to see a list of the files searched for scons: *** [src/compiler98/Bind.o] Error 1 scons: building terminated because of errors. Chaz:~/Projects/Yhc/yhc-devel brent$ ======= Details: Chaz:~/Projects/Yhc/yhc-devel brent$ ghc -v Glasgow Haskell Compiler, Version 6.4.2, for Haskell 98, compiled by GHC version 6.4.2 Using package config file: /opt/local/lib/ghc-6.4.2/package.conf Using package config file: /Users/brent/.ghc/powerpc-darwin-6.4.2/ package.conf Hsc static flags: -static *** Deleting temp files Deleting: ghc-6.4.2: no input files Usage: For basic information, try the `--help' option. Chaz:~/Projects/Yhc/yhc-devel brent$

2 1