- Glasgow-haskell-users

building ghc from source
by johago＠web.de 19 May '03

19 May '03

Hi there! I'm trying to build ghc from scratch under Windows XP... which gives me problems, problems, problems... Steps taken: 1. Installed latest ghc (v5.04.3) 2. Installed happy (v1.13) 3. Unpacked current ghc source distribution (ghc-5.04.3-src.tar.bz2) to c:/build/ghc 4. Downloaded the latest Cygwin distribution and installed it to c:/cygwin 5. Got latest MinGW distribution (MinGW-2.0.0-3.exe) and installed it to c:/MinGW 6. Added /cygdrive/c/MinGW/bin to path in /etc/profile 7. Started Cygwin bash 8. Changed directory to /cygdrive/c/build/ghc 9. Entered "./configure --host=i386-unknown-mingw32 --with-gcc=/cygdrive/c/MinGW/bin/gc c". No errors occurred. --- snip --- $ ./configure --host=i386-unknown-mingw32 --with-gcc=/cygdrive/c/MinGW/bin/gcc creating cache ./config.cache *** The top of your build tree is: /cygdrive/c/build/ghc checking host system type... i386-unknown-mingw32 checking target system type... i386-unknown-mingw32 checking build system type... i386-unknown-mingw32 Canonicalised to: i386-unknown-mingw32 checking for ghc... /cygdrive/c/ghc/ghc-5.04.3/bin/ghc checking version of ghc... 5.04.3 checking for nhc... no checking for nhc98... no checking for hbc... no checking whether #! works in shell scripts... yes checking for perl... /bin/perl checking if your perl works in shell scripts... yes checking for gcc... /cygdrive/c/MinGW/bin/gcc checking whether the C compiler (/cygdrive/c/MinGW/bin/gcc ) works... yes checking whether the C compiler (/cygdrive/c/MinGW/bin/gcc ) is a cross-compile r... no checking whether we are using GNU C... yes checking whether /cygdrive/c/MinGW/bin/gcc accepts -g... yes checking whether you have an ok gcc... yes checking how to run the C preprocessor... /cygdrive/c/MinGW/bin/gcc -E checking whether /cygdrive/c/MinGW/bin/gcc accepts -mno-cygwin... yes checking for ok way to do context diffs... diff -C 1 checking for find... /usr/bin/find checking for bison... no checking for byacc... no checking if it is an OK yacc... checking for what... : yacc checking for flex... flex checking for a BSD compatible install... /usr/bin/install -c checking for ar... /cygdrive/c/MinGW/bin/ar checking whether ln -s works... yes checking for sed... /usr/bin/sed checking for time... /usr/bin/time checking for gtar... no checking for tar... /usr/bin/tar checking for gzip... /usr/bin/gzip checking for openjade... no checking for jade... no checking for DocBook CATALOG... no Warning: You will not be able to build the documentation. checking for happy... /cygdrive/c/happy/happy-1.13/bin/happy checking for version of happy... 1.13 checking for haddock... no checking for ANSI C header files... yes checking for Files.h... no checking for arpa/inet.h... no checking for assert.h... yes checking for console.h... no checking for ctype.h... yes checking for dirent.h... yes checking for errno.h... yes checking for fcntl.h... yes checking for float.h... yes checking for ftw.h... no checking for grp.h... no checking for ieee754.h... no checking for inttypes.h... yes checking for limits.h... yes checking for malloc.h... yes checking for memory.h... yes checking for nlist.h... no checking for pascal.h... no checking for pwd.h... no checking for sgtty.h... no checking for siginfo.h... no checking for signal.h... yes checking for stat.h... no checking for stdint.h... yes checking for stdlib.h... yes checking for stddef.h... yes checking for stdarg.h... yes checking for string.h... yes checking for sys/fault.h... no checking for sys/file.h... yes checking for sys/ioctl.h... no checking for sys/limits.h... no checking for sys/mman.h... no checking for sys/param.h... yes checking for sys/procfs.h... no checking for sys/resource.h... no checking for sys/signal.h... no checking for sys/socket.h... no checking for netdb.h... no checking for netinet/in.h... no checking for netinet/tcp.h... no checking for sys/stat.h... yes checking for sys/syscall.h... no checking for sys/time.h... yes checking for sys/timeb.h... yes checking for sys/timers.h... no checking for sys/times.h... no checking for sys/types.h... yes checking for sys/un.h... no checking for sys/utsname.h... no checking for sys/vadvise.h... no checking for sys/wait.h... no checking for termio.h... no checking for termios.h... no checking for time.h... yes checking for types.h... no checking for unistd.h... yes checking for utime.h... no checking for values.h... yes checking for vfork.h... no checking for bfd.h... yes checking for winsock.h... yes checking for pthread.h... no checking for sys/uio.h... no checking for readline/readline.h... no checking for readline/history.h... no checking for dos.h... yes checking for conio.h... yes checking for io.h... yes checking for std.h... no checking for windows.h... yes checking for GL/gl.h... yes checking whether time.h and sys/time.h may both be included... yes checking for dlfcn.h... no checking for dl.h... no checking for alloc.h... no checking for malloc.h... (cached) yes checking for valloc... no checking for regex.h... no checking whether struct tm is in sys/time.h or time.h... time.h checking for tm_zone in struct tm... no checking for tzname... no checking timezone... yes checking type of timezone... time_t checking altzone... no checking for st_blksize in struct stat... no checking whether /cygdrive/c/MinGW/bin/gcc supports long long types... yes checking for msg_accrights in struct msghdr... no checking for msg_control in struct msghdr... no checking size of char... 1 checking size of double... 8 checking size of float... 4 checking size of int... 4 checking size of long... 4 checking size of long long... 8 checking size of short... 2 checking size of unsigned char... 1 checking size of unsigned int... 4 checking size of unsigned long... 4 checking size of unsigned long long... 8 checking size of unsigned short... 2 checking size of void *... 4 checking alignment of char... 1 checking alignment of double... 8 checking alignment of float... 4 checking alignment of int... 4 checking alignment of long... 4 checking alignment of long long... 8 checking alignment of short... 2 checking alignment of unsigned char... 1 checking alignment of unsigned int... 4 checking alignment of unsigned long... 4 checking alignment of unsigned long long... 8 checking alignment of unsigned short... 2 checking alignment of void *... 4 checking Haskell type for char... Int8 checking Haskell type for signed char... Int8 checking Haskell type for unsigned char... Word8 checking Haskell type for short... Int16 checking Haskell type for unsigned short... Word16 checking Haskell type for int... Int32 checking Haskell type for unsigned int... Word32 checking Haskell type for long... Int32 checking Haskell type for unsigned long... Word32 checking Haskell type for long long... Int64 checking Haskell type for unsigned long long... Word64 checking Haskell type for float... Float checking Haskell type for double... Double checking Haskell type for ptrdiff_t... Int32 checking Haskell type for size_t... Word32 checking Haskell type for wchar_t... Word16 checking Haskell type for sig_atomic_t... Int32 checking Haskell type for clock_t... Int32 checking Haskell type for time_t... Int32 checking Haskell type for dev_t... Word32 checking Haskell type for ino_t... Int16 checking Haskell type for mode_t... Word16 checking Haskell type for off_t... Int32 checking Haskell type for pid_t... Int32 checking Haskell type for gid_t... NotReallyAType checking Haskell type for uid_t... NotReallyAType checking Haskell type for cc_t... NotReallyAType checking Haskell type for speed_t... NotReallyAType checking Haskell type for tcflag_t... NotReallyAType checking Haskell type for nlink_t... NotReallyAType checking Haskell type for ssize_t... NotReallyAType checking whether we can open files in binary mode... no checking return type of signal handlers... void checking for strcasecmp... yes checking for _stricmp... yes checking for stricmp... yes checking for strcmpi... yes checking for strcmp... yes checking for realpath... no checking for _fullpath... yes checking for PBHSetVolSync... no checking for macsystem... no checking for fgetpos... yes checking for fsetpos... yes checking for fseek... yes checking for ftell... yes checking for vsnprintf... yes checking for _vsnprintf... yes checking for snprintf... yes checking for _snprintf... yes checking for popen... yes checking for _popen... yes checking for pclose... yes checking for _pclose... yes checking for access... yes checking for ftime... yes checking for getclock... no checking for getpagesize... yes checking for getrusage... no checking for gettimeofday... no checking for mktime... yes checking for mprotect... yes checking for readlink... no checking for setitimer... no checking for stat... yes checking for lstat... no checking for symlink... no checking for sysconf... no checking for timelocal... no checking for times... no checking for vadvise... no checking for vfork... no checking for localtime_r... no checking for gmtime_r... no checking for readdir_r... no checking for __gmpz_fdiv_qr in -lgmp... no checking for __gmpz_fdiv_qr in -lgmp3... no checking for closedir in -lmingwex... yes checking for GNU regex in libc... no checking for xmalloc in -liberty... yes checking for bfd_init in -lbfd... yes checking for tputs in -lncurses... no checking for tputs in -ltermcap... no checking for tputs in -lcurses... no checking for main in -lm... yes checking for library containing glEnd... no checking for library containing gluNewQuadric... no checking for library containing glXWaitX... no checking for library containing glutMainLoop... no checking OpenGL flags... checking OpenGL libs... -lX11 -lXext -lXmu -lXt -lXi -lm dl: not found checking for dlopen in -ldl... no checking for dlopen... no checking for RTLD_NEXT from dlfcn.h... no checking for RTLD_LOCAL from dlfcn.h... no checking for function_atan... yes checking for in_addr_t in netinet/in.h... no checking for gtk-config... no checking for gtk12-config... no configure: warning: GTK+ not usable; need at least version 1.2 checking for working alloca.h... no checking for alloca... yes checking for working const... yes checking prototypes... yes checking whether byte ordering is bigendian... no checking for nlist in -lelf... no checking leading underscore in symbol names... yes checking for end of text section marker... etext checking for end of text section marker declaration... etext asm("etext") checking for end of data section marker... end checking for end of data section marker declaration... end checking if code section appears before data... yes checking for ld... /cygdrive/c/MinGW/bin/ld checking whether ld understands -x... yes updating cache ./config.cache creating ./config.status creating mk/config.mk creating mk/config.h configuring in ghc running /bin/sh ./configure --host=i386-unknown-mingw32 --with-gcc=/cygdrive/c/ MinGW/bin/gcc --cache-file=.././config.cache --srcdir=. loading cache .././config.cache updating cache .././config.cache creating ./config.status creating ghc.spec creating docs/index.html --- snap --- 10. Entered "make". Some warnings were given frequently: --- snip --- Warning: foreign declaration uses deprecated non-standard syntax --- snap --- and --- snip --- Warning: .type pseudo-op used outside of .def/.endef ignored. --- snap --- and --- snip --- Warning: rest of line ignored; first ignored character is `_' --- snap --- Finally there was an error: --- snip --- [...] /bin/sh ./libtool --mode=link /cygdrive/c/MinGW/bin/gcc -g -O2 -o libgmp.la - rpath /usr/local/lib -version-info 4:1:1 assert.lo compat.lo errno.lo memory.lo mp_set_fns.lo mp_clz_tab.lo mp_minv_tab.lo rand.lo randclr.lo randlc.lo randlc2x .lo randraw.lo randsd.lo randsdui.lo version.lo stack-alloc.lo mp_bpl.lo extract -dbl.lo insert-dbl.lo mpz/abs.lo mpz/add.lo mpz/add_ui.lo mpz/addmul_ui.lo mpz/ and.lo mpz/array_init.lo mpz/bin_ui.lo mpz/bin_uiui.lo mpz/cdiv_q.lo mpz/cdiv_q_ ui.lo mpz/cdiv_qr.lo mpz/cdiv_qr_ui.lo mpz/cdiv_r.lo mpz/cdiv_r_ui.lo mpz/cdiv_u i.lo mpz/clear.lo mpz/clrbit.lo mpz/cmp.lo mpz/cmp_si.lo mpz/cmp_ui.lo mpz/cmpab s.lo mpz/cmpabs_ui.lo mpz/com.lo mpz/divexact.lo mpz/dump.lo mpz/fac_ui.lo mpz/f div_q.lo mpz/fdiv_q_2exp.lo mpz/fdiv_q_ui.lo mpz/fdiv_qr.lo mpz/fdiv_qr_ui.lo mp z/fdiv_r.lo mpz/fdiv_r_2exp.lo mpz/fdiv_r_ui.lo mpz/fdiv_ui.lo mpz/fib_ui.lo mpz /fits_sint_p.lo mpz/fits_slong_p.lo mpz/fits_sshort_p.lo mpz/fits_uint_p.lo mpz/ fits_ulong_p.lo mpz/fits_ushort_p.lo mpz/gcd.lo mpz/gcd_ui.lo mpz/gcdext.lo mpz/ get_d.lo mpz/get_si.lo mpz/get_str.lo mpz/get_ui.lo mpz/getlimbn.lo mpz/hamdist. lo mpz/init.lo mpz/inp_raw.lo mpz/inp_str.lo mpz/invert.lo mpz/ior.lo mpz/iset.l o mpz/iset_d.lo mpz/iset_si.lo mpz/iset_str.lo mpz/iset_ui.lo mpz/jacobi.lo mpz/ kronsz.lo mpz/kronuz.lo mpz/kronzs.lo mpz/kronzu.lo mpz/lcm.lo mpz/legendre.lo m pz/mod.lo mpz/mul.lo mpz/mul_2exp.lo mpz/mul_si.lo mpz/mul_ui.lo mpz/neg.lo mpz/ nextprime.lo mpz/out_raw.lo mpz/out_str.lo mpz/perfpow.lo mpz/perfsqr.lo mpz/pop count.lo mpz/pow_ui.lo mpz/powm.lo mpz/powm_ui.lo mpz/pprime_p.lo mpz/random.lo mpz/random2.lo mpz/realloc.lo mpz/remove.lo mpz/root.lo mpz/rrandomb.lo mpz/scan 0.lo mpz/scan1.lo mpz/set.lo mpz/set_d.lo mpz/set_f.lo mpz/set_q.lo mpz/set_si.l o mpz/set_str.lo mpz/set_ui.lo mpz/setbit.lo mpz/size.lo mpz/sizeinbase.lo mpz/s qrt.lo mpz/sqrtrem.lo mpz/sub.lo mpz/sub_ui.lo mpz/swap.lo mpz/tdiv_ui.lo mpz/td iv_q.lo mpz/tdiv_q_2exp.lo mpz/tdiv_q_ui.lo mpz/tdiv_qr.lo mpz/tdiv_qr_ui.lo mpz /tdiv_r.lo mpz/tdiv_r_2exp.lo mpz/tdiv_r_ui.lo mpz/tstbit.lo mpz/ui_pow_ui.lo mp z/urandomb.lo mpz/urandomm.lo mpz/xor.lo mpn/mp_bases.lo mpn/udiv.lo mpn/umul.lo mpn/inlines.lo mpn/add_n.lo mpn/sub_n.lo mpn/mul_1.lo mpn/addmul_1.lo mpn/submu l_1.lo mpn/lshift.lo mpn/rshift.lo mpn/diveby3.lo mpn/divrem.lo mpn/divrem_1.lo mpn/divrem_2.lo mpn/mod_1.lo mpn/mod_1_rs.lo mpn/pre_mod_1.lo mpn/dump.lo mpn/mu l.lo mpn/mul_fft.lo mpn/mul_n.lo mpn/mul_basecase.lo mpn/sqr_basecase.lo mpn/ran dom.lo mpn/random2.lo mpn/sqrtrem.lo mpn/get_str.lo mpn/set_str.lo mpn/scan0.lo mpn/scan1.lo mpn/popcount.lo mpn/hamdist.lo mpn/cmp.lo mpn/perfsqr.lo mpn/bdivmo d.lo mpn/gcd_1.lo mpn/gcd.lo mpn/gcdext.lo mpn/tdiv_qr.lo mpn/bz_divrem_n.lo mpn /sb_divrem_mn.lo mpn/jacbase.lo mpn/copyi.lo mpn/copyd.lo mkdir .libs libtool: link: warning: undefined symbols not allowed in i386-unknown-mingw32 sh ared libraries rm -fr .libs/libgmp.la .libs/libgmp.* .libs/libgmp.* ar cru .libs/libgmp.a assert.o compat.o errno.o memory.o mp_set_fns.o mp_clz_ta b.o mp_minv_tab.o rand.o randclr.o randlc.o randlc2x.o randraw.o randsd.o randsd ui.o version.o stack-alloc.o mp_bpl.o extract-dbl.o insert-dbl.o mpz/abs.o mpz/a dd.o mpz/add_ui.o mpz/addmul_ui.o mpz/and.o mpz/array_init.o mpz/bin_ui.o mpz/bi n_uiui.o mpz/cdiv_q.o mpz/cdiv_q_ui.o mpz/cdiv_qr.o mpz/cdiv_qr_ui.o mpz/cdiv_r. o mpz/cdiv_r_ui.o mpz/cdiv_ui.o mpz/clear.o mpz/clrbit.o mpz/cmp.o mpz/cmp_si.o mpz/cmp_ui.o mpz/cmpabs.o mpz/cmpabs_ui.o mpz/com.o mpz/divexact.o mpz/dump.o mp z/fac_ui.o mpz/fdiv_q.o mpz/fdiv_q_2exp.o mpz/fdiv_q_ui.o mpz/fdiv_qr.o mpz/fdiv _qr_ui.o mpz/fdiv_r.o mpz/fdiv_r_2exp.o mpz/fdiv_r_ui.o mpz/fdiv_ui.o mpz/fib_ui .o mpz/fits_sint_p.o mpz/fits_slong_p.o mpz/fits_sshort_p.o mpz/fits_uint_p.o mp z/fits_ulong_p.o mpz/fits_ushort_p.o mpz/gcd.o mpz/gcd_ui.o mpz/gcdext.o mpz/get _d.o mpz/get_si.o mpz/get_str.o mpz/get_ui.o mpz/getlimbn.o mpz/hamdist.o mpz/in it.o mpz/inp_raw.o mpz/inp_str.o mpz/invert.o mpz/ior.o mpz/iset.o mpz/iset_d.o mpz/iset_si.o mpz/iset_str.o mpz/iset_ui.o mpz/jacobi.o mpz/kronsz.o mpz/kronuz. o mpz/kronzs.o mpz/kronzu.o mpz/lcm.o mpz/legendre.o mpz/mod.o mpz/mul.o mpz/mul _2exp.o mpz/mul_si.o mpz/mul_ui.o mpz/neg.o mpz/nextprime.o mpz/out_raw.o mpz/ou t_str.o mpz/perfpow.o mpz/perfsqr.o mpz/popcount.o mpz/pow_ui.o mpz/powm.o mpz/p owm_ui.o mpz/pprime_p.o mpz/random.o mpz/random2.o mpz/realloc.o mpz/remove.o mp z/root.o mpz/rrandomb.o mpz/scan0.o mpz/scan1.o mpz/set.o mpz/set_d.o mpz/set_f. o mpz/set_q.o mpz/set_si.o mpz/set_str.o mpz/set_ui.o mpz/setbit.o mpz/size.o mp z/sizeinbase.o mpz/sqrt.o mpz/sqrtrem.o mpz/sub.o mpz/sub_ui.o mpz/swap.o mpz/td iv_ui.o mpz/tdiv_q.o mpz/tdiv_q_2exp.o mpz/tdiv_q_ui.o mpz/tdiv_qr.o mpz/tdiv_qr _ui.o mpz/tdiv_r.o mpz/tdiv_r_2exp.o mpz/tdiv_r_ui.o mpz/tstbit.o mpz/ui_pow_ui. o mpz/urandomb.o mpz/urandomm.o mpz/xor.o mpn/mp_bases.o mpn/udiv.o mpn/umul.o m pn/inlines.o mpn/add_n.o mpn/sub_n.o mpn/mul_1.o mpn/addmul_1.o mpn/submul_1.o m pn/lshift.o mpn/rshift.o mpn/diveby3.o mpn/divrem.o mpn/divrem_1.o mpn/divrem_2. o mpn/mod_1.o mpn/mod_1_rs.o mpn/pre_mod_1.o mpn/dump.o mpn/mul.o mpn/mul_fft.o mpn/mul_n.o mpn/mul_basecase.o mpn/sqr_basecase.o mpn/random.o mpn/random2.o mpn /sqrtrem.o mpn/get_str.o mpn/set_str.o mpn/scan0.o mpn/scan1.o mpn/popcount.o mp n/hamdist.o mpn/cmp.o mpn/perfsqr.o mpn/bdivmod.o mpn/gcd_1.o mpn/gcd.o mpn/gcde xt.o mpn/tdiv_qr.o mpn/bz_divrem_n.o mpn/sb_divrem_mn.o mpn/jacbase.o mpn/copyi. o mpn/copyd.o creating libgmp.la (cd .libs && rm -f libgmp.la && ln -s ../libgmp.la libgmp.la) make[5]: Leaving directory `/cygdrive/c/build/ghc/ghc/rts/gmp' make[4]: Leaving directory `/cygdrive/c/build/ghc/ghc/rts/gmp' ../utils/ghc-pkg/ghc-pkg-inplace --update-package <rts.conf.inplace Reading package info from stdin... done. Expanding embedded variables...done. warning: can't find GHCi lib `HSrts.o' Saving old package config file... done. Writing new package config file... done. ../utils/ghc-pkg/ghc-pkg-inplace -f ../driver/package.conf --update-package <rt s.conf.installed Reading package info from stdin... done. Expanding embedded variables...done. warning: can't find GHCi lib `HSrts.o' Saving old package config file... done. Writing new package config file... done. ../../ghc/compiler/ghc-inplace -optc-mno-cygwin -optc-O -optc-Wall -optc-W - optc -Wstrict-prototypes -optc-Wmissing-prototypes -optc-Wmissing-declarations -o ptc- Winline -optc-Waggregate-return -optc-Wbad-function-cast -optc-Wcast-align - optc -I../includes -optc-I. -optc-Iparallel -optc-DCOMPILING_RTS -optc-fomit-fram e-po inter -O2 -static -package-name rts -O -Rghc-timing -c Adjustor.c -o Adjusto r.o <<ghc: 629196 bytes, 2 GCs, 19832/19832 avg/max bytes residency (1 samples), 5M in use, 0.01 INIT (0.02 elapsed), 0.00 MUT (0.02 elapsed), 0.02 GC (0.03 elapsed ) :ghc>> make[2]: *** [Adjustor.o] Error 1 make[1]: *** [all] Error 1 make[1]: Leaving directory `/cygdrive/c/build/ghc/ghc' make: *** [all] Error 1 --- snap --- Thanks very much for your suggestions in advance Bye, Johannes johago(a)web.de

4 4

FFI under Windows
by Roman Beslik 17 May '03

17 May '03

How can I compile C/Haskell program with my own WinMain function? I try ghc -fglasgow-exts -c Foo.hs gcc -Ie:\ghc\include -c win_ghc.c ghc -o win_ghc.exe -no-hs-main Foo_stub.o win_ghc.o Foo.o and got an error > ghc -o win_ghc.exe -no-hs-main -lgdi32 Foo_stub.o win_ghc.o Foo.o e:/ghc/libHSrts.a(Main.o)(.text+0x82):Main.c: undefined reference to `__stginit_Main' e:/ghc/libHSrts.a(Main.o)(.text+0x93):Main.c: undefined reference to `Main_zdmain_closure' Similar works properly: ghc -fglasgow-exts -c Foo.hs gcc -Ie:\ghc\include -c ffi.c ghc -o ffi.exe -no-hs-main Foo_stub.o ffi.o Foo.o ffi.c contains main(), but win_ghc.c contains WinMain() instead of main(). Best regards, Roman Beslik.

1 0

BU FIRSATTAN YARARLANIN.
by Erotik Market 16 May '03

16 May '03

DIKKAT VIDEO CD : Iddia ediyoruz.. Hic bir yerden temin edemeyeceginiz ses ve g�runtu kalitesi ile yuzlerce porno video CD. arSivimiz yenilenmistir. istemis oldugunuz video CD.ler bire bir yollanir kesinlikle isteginiz harici alakasiz baSka video CD.ler yollanmaz. Anal. Oral. Vajinal. Grup. Zenci. FethiS. Ayak fethiS. Gay. Zenci gay. Trans. Transexual. Animal, Lezbiyen ve T�rk daha bircok ceSit ....Ayrica Unlu taninmis porno yildizlari Rocco & silvia saint http://www.geceshop.com ZENGIN URUN CESITLERIMIZ : Sisme Bebekler ............... (Erkek & Bayan) Kesinlikle size hayir demeyecek. Vibrat�rler ..................... Istediginiz boy ve ebatlarda (Vajinal/Anal/catal.Pilli.Motorlu.TitreSimli) Suni Vajinalar ................. Asla gerceginden ayirt edemeyeceksiniz (Gercek ten hassasiyetinde) Reailistik Penisler........... Gercek ten hassasiyetinde ve dokusunda (Vantuzlu/Deri kemer kilotlu) Vakum Pompalari ......... Ereksiyonu kolaylastirici ve duzenli kullanimlarda peniste irilesme saglar. Geciktiriciler ................. Erken boSalmayi dert etmeyin (Sprey ceSitleri/Kremler) Kremler ....................... Anal ve Vajinal iliSkilerde kullanabileceginiz kayganlaStirici krem ceSitleri Uyandiricilar ............... Cinsel istek uyandirici haplar ve damlalar. Yapmaniz gereken tek Sey http://www.geceshop.com TIKLAMAK .. NOT : BU MAIL REKLAM AMAcLI OLUP HIcBIR SEKILDE TARAFIMIZDA KAYDINIZ BULUNMAMAKTADIR. ILGI ALANINIZIN DISINDA ISE EGER LUTFEN DIKKATE ALMAYINIZ TESEKKURLER..

1 0

-O and -prof
by ketil＠ii.uib.no 16 May '03

16 May '03

Hi, Is there any reason not to use these options in combination? Out of habit, I don't usually combine them, but it'd probably be more interesting to do time profiling with optimized code, since the compiler tend to optimize away some of the more pedestrian stuff on its own. When I recently tried this, I got a surprising stack overflow -- surprising, as I don't get that with either option by itself. -kzm -- If I haven't seen further, it is by standing in the footprints of giants

3 4

RE: -O and -prof
by Simon Marlow 16 May '03

16 May '03

> John Meacham <john(a)repetae.net> writes: > > > I always use them in combination. simply because optimization can > > drastically change the memory useage/profile. profiling the > unoptimized > > version seems rather moot. > > Exactly. If this is indeed the intention, I guess the stack overflow > is a bug? Yes, strictly speaking. It may be a spurious case of the code generated with -O -prof being slightly different from that generated by -O alone - this happens because having cost centres floating around in the intermediate code disables some transformations that would normally be applicable. Nevertheless, if you can provide a test case we'll look into it. > > BTW. does O2 still do not much more than -O? it seems to reduce the > > memory footprint of some of my apps pretty noticbly. > > Not sure. The docs seem to indicate the speed improvement is > negligible. It's normally negligible. -O2 turns on one more optimisation pass, which *might* have a significant impact on your program if it hits the inner loop, but in most cases just costs you extra compilation time for not much benefit. > I thought I saw some hints in the GHC docs, including using > -fvia-C, but I couldn't find them, and I'm not sure if they would be > still current. > > Memory footprint is a problem, I wonder if GHC makes any > effort to pack strict data types? I.e. > > data D1 = A | B > data D2 = A2 | B2 | C2 > > data D3 = D !D1 !D2 -- could fit inside e.g. a Word8? We don't do any useful optimsation of the representation of D3 here, but we could (it's been on my ToDo list since I implemented -funbox-strict-fields some time ago). In a similar vein, the strictness analyser doesn't take advantage of strict enumerated types - it could map them to Int#, for example. Semitagging (an optimisation in the works along with optimistic evaluation) will provide some of the benefit that a more efficient representation would yield here. > Is there an elegant way to achieve this manually (if I know I'll need > large arrays of D3s, for instance -- can I map them to arrays of Word8 > or a similar type?) If you map D1 and D2 to Int by hand, then use data D3 = D !Int !Int you'll get the speed benefit, but not all the space (each field will still take up a 32-bit word). Cheers, Simon

1 0

RE: ghc-5.04.3-1.src.rpm -i
by Simon Marlow 15 May '03

15 May '03

> There's a hidden gotcha in the rpm spec file (the file which gives > instructions for building) - it only does a stage1 build, > whereas ghci > needs a stage2 build. This is OK as long as it's understood by rpm > builders that they need to build binary rpms using a previous install > of 5.04.3. We could probably trick up the spec file to be clever and > trigger a stage2 build if it's necessary. For version 6.0, let's have the spec file do a stage2 build by default. It's much easier these days, you only need one source tree and just saying "make" at the top-level will cause a stage2 build to happen. Cheers, Simon

3 2

RE: Alignment of doubles in ghc
by Simon Peyton-Jones 12 May '03

12 May '03

Try http://research.microsoft.com/~simonpj/papers/eval-apply/ This paper discusses the eval/apply stuff in some detail Simon | -----Original Message----- | From: glasgow-haskell-users-admin(a)haskell.org [mailto:glasgow-haskell-users-admin@haskell.org] | On Behalf Of Jerzy Karczmarczuk | Sent: 12 May 2003 12:59 | To: glasgow-haskell-users(a)haskell.org | Subject: Re: Alignment of doubles in ghc | | Simon Marlow wrote: | | > We've already made the switch to eval/apply for the next release of GHC. | > However, deferring the entirety of stack management to an underlying | > compiler is not that easy: we do accurate garbage collection, so we need | > to retain enough information about the compiler's idea of stack layout | > in order to be able to find all the pointers during GC. We therefore | > can't compile via C using an ordinary C compiler and leave the stack | > management up to the C compiler. | | Is there anything published/unofficially available about it? | | I have pure academic interest in all that. I must say that rarely had I | the occasion to learn something more crunchy about the implementation | of Haskell and FP than from the description of concrete solutions, such | as STG (and the implementation of Clean, and Appel's paper on the imple- | mentation of ML). | | Now, my white/bald hair tell me that the eval/apply model is something | quite oriented towards strict evaluation paradigms, and I would enjoy very | much a thorough discussion of the implementation of laziness. | | | Jerzy Karczmarczuk | | _______________________________________________ | Glasgow-haskell-users mailing list | Glasgow-haskell-users(a)haskell.org | http://www.haskell.org/mailman/listinfo/glasgow-haskell-users

1 0

RE: file too large
by Simon Marlow 12 May '03

12 May '03

> Ferenc Wagner writes: > > > - Macro: _FILE_OFFSET_BITS > > Shouldn't Autoconf determine what needs to be done for > large-file support? If you call the AC_SYS_LARGEFILE macro > in the configure script, your build will be transparently > switched to large-file support by providing the necessary > compiler flags (like -D_FILE_OFFSET_BITS=64). > > The configure.in script coming with GHC 5.04.3 doesn't seem > to make use of this. Maybe adding the macro call is all that > is to fixing this problem? This has come as something of a shock to me: it seems I've been living in a dream world in which off_t was always 64 bits and large files were supported by default :-) The AC_SYS_LARGEFILE macro seems to be a relatively new addition to autoconf (i.e. not in 2.13), but I'll see what I can do about enabling large file support in GHC. Cheers, Simon

1 0

RE: file too large
by Simon Marlow 12 May '03

12 May '03

> There seems to be a 2Gb file limit when using GHC, I don't suppose > there's any way to get around it, apart from manually splitting the > contents over several smaller files? (I'm even unable to use > writeFile on a file that's larger than 2Gb) That's probably a bug, as long as the underlying OS uses 64-bit file offsets. How does the 2Gb limit manifest itself? Cheers, Simon

4 7

RE: Alignment of doubles in ghc
by Simon Marlow 12 May '03

12 May '03

> I've recently browsed some assembly code generated by GHC > (via gcc). It appeared that most values were accessed via > memory, because the stack is managed by GHC explicity. Even > intermediate results seem accessed via memory, probably due > to shortcomings of aliasing analysis in gcc. (?) > > This brings on the topic of eval/apply vs push/enter. In the > eval/apply model, the stack management can (possibly) be left to an > underlying compiler, which would remove most memory accesses, > and leave most of the alignment issue to the back-end. > (http://research.microsoft.com/Users/simonpj/papers/eval-apply/) > > From that article, I can only assume that GHC will switch to > the eval/apply model. Can we expect that soon? The > "alignment" thing would > become easier to deal with... We've already made the switch to eval/apply for the next release of GHC. However, deferring the entirety of stack management to an underlying compiler is not that easy: we do accurate garbage collection, so we need to retain enough information about the compiler's idea of stack layout in order to be able to find all the pointers during GC. We therefore can't compile via C using an ordinary C compiler and leave the stack management up to the C compiler. C-- has a solution to this problem, but there is neither a C-- compiler or a C-- back end for GHC (yet). Cheers, Simon

2 1