- Glasgow-haskell-users

Re: Threaded runtime error
by Simon Marlow 22 Mar '06

22 Mar '06

Volker Stolz wrote: > * Volker Stolz <stolz(a)i2.informatik.rwth-aachen.de>: > >>timeout@menelaos [20:17:08]> ./timeout 10 /usr/bin/true >>FFatal error '_pq_insert_tail: Already in priority queue' at line 200 in file /usr/src/lib/libpthread/thread/thr_priority_queue.c (errno = 0) > > > A workaround seems to use another threading library here. Both > libthr and libc_r work at least for this small example. > > The error occurs on createOSThread(). I'm not sure where exactly > (ie., for the child or the parent, but I guess the child). > I'll keep on looking. If you can get me a backtrace, that would be most helpful. Cheers, Simon

1 0

Re: 6.4.2 release plans
by Simon Marlow 22 Mar '06

22 Mar '06

Duncan Coutts wrote: > Chris Parrott found that ghc-6.4.1 doesn't work with gcc 4.1 on amd64, > although it does work on x86. > > The error manifests itself as link errors due to undefined symbols: > > /var/tmp/portage/ghc-6.4.1-r2/work/ghc-6.4.1/ghc/rts/libHSrts.a(Linker.o):(.data+0x48): undefined reference to `StgReturn' > /var/tmp/portage/ghc-6.4.1-r2/work/ghc-6.4.1/ghc/rts/libHSrts.a(RtsStartup.o): In function `hs_add_root': > RtsStartup.c:(.text+0x1a8): undefined reference to `StgRun' Coincidentally, I just upgraded our x86_64 box to FC5, and the nightly build fell over with exactly these errors. It turns out that the HEAD already had fixes in some arch's for this, but it was fixed by making StgRunIsImplementedInAssembler() non-static. The attribute fix is better, so I've applied it consistently and merged to 6.4.2. Cheers, Simon

1 0

Show-ing GHC internal data structures
by David William Simmons-Duffin 21 Mar '06

21 Mar '06

I'm working with a friend on a project that involves hacking at the GHC internals. We'd like to be able to print some of the large data structures involved -- specifically a TcGblEnv. TcGblEnv is a record with a number of fields of different types, and it would be silly to go through the source adding "deriving Show" to the end of each type definition. It seems like Show instances for internal data structures would be useful for the GHC developers (for debugging purposes -- perhaps they were left out of the GHC distribution to keep the compiled code small?), and we were wondering if anyone had the relevant instance declarations lying around. Thanks, David

2 1

RE: RULES pragma with class constraint
by Simon Peyton-Jones 20 Mar '06

20 Mar '06

Definitely not at present, and I see no easy way to implement it. RULES are implemented by simple matching in Core. A call to nub will have an Eq dictionary, but Core knows nothing of instance declarations, and has no clue how to make an Ord dictionary. But an Ord dictionary is what you want on the RHS of the rule. The type checker, on the other hand, knows about dictionary construction, but nothing about RULES. Simon | -----Original Message----- | From: glasgow-haskell-users-bounces(a)haskell.org [mailto:glasgow-haskell-users- | bounces(a)haskell.org] On Behalf Of John Meacham | Sent: 20 March 2006 11:49 | To: glasgow-haskell-users(a)haskell.org | Subject: RULES pragma with class constraint | | Is it possible to create a RULES that fires only if a type has a given | class constraint? something like: | | > snub :: Ord a => [a] -> [a] | > snub xs = f Set.empty xs where | > f _ [] = [] | > f (x:xs) set | > | x `Set.member` set = f xs | > | otherwise = x:f xs (Set.insert x set) | > | > | > {-# RULES "nub/snub" Ord a => forall a xs . nub (xs::[a]) = snub xs #-} | | Such a rule would probably only be able to fire on monotypes with known | instances since otherwise the dictionary won't necessarily be available. | but it seems like it might be a useful feature even with that | restriction. | | John | | -- | John Meacham - ⑆repetae.net⑆john⑈ | _______________________________________________ | Glasgow-haskell-users mailing list | Glasgow-haskell-users(a)haskell.org | http://www.haskell.org/mailman/listinfo/glasgow-haskell-users

2 1

RULES pragma with class constraint
by John Meacham 20 Mar '06

20 Mar '06

Is it possible to create a RULES that fires only if a type has a given class constraint? something like: > snub :: Ord a => [a] -> [a] > snub xs = f Set.empty xs where > f _ [] = [] > f (x:xs) set > | x `Set.member` set = f xs > | otherwise = x:f xs (Set.insert x set) > > > {-# RULES "nub/snub" Ord a => forall a xs . nub (xs::[a]) = snub xs #-} Such a rule would probably only be able to fire on monotypes with known instances since otherwise the dictionary won't necessarily be available. but it seems like it might be a useful feature even with that restriction. John -- John Meacham - ⑆repetae.net⑆john⑈

2 1

ghc for SuSE. Reply
by Serge D. Mechveliani 20 Mar '06

20 Mar '06

To my request on GHC-6.4.1 for SuSE 10.0 Christian Maeder <maeder(a)tzi.de> writes on 17 Mar 2006 >> First, we install it to the system area from binary found on the GHC www >> page as >> Binary x86, Generic Linux with glib 2.3. > This binary is not for a 64 bit arch, I think. >> /usr/lib64/gcc/x86_64-suse-linux/4.0.2/../../../../x86_64-suse-linux/bin/ld: cannot find -lreadline > you may try: > http://www.haskell.org/ghc/dist/6.4.1/ghc-6.4.1-x86_64-unknown-linux.tar.bz2 Thank you. I have installed it to my user directory. Now, again, it reports --------- ghc $doconCpOpt --make Main Chasing modules from: Main Skipping Main ( Main.hs, Main.o ) Linking ... /usr/lib64/gcc/x86_64-suse-linux/4.0.2/../../../../x86_64-suse-linux/bin/ld: cannot find -lreadline collect2: ld returned 1 exit status ---------- > which ghc /home/mechvel/ghc/6.4.1/binary/inst/bin/ghc -- this is of ghc-6.4.1-x86_64-unknown-linux.tar.bz2 . There also is visible /home/mechvel/ghc/6.4.1/binary/inst/lib/ghc-6.4.1/libHSreadline.a But it, probably, searches for shared library in other place. Maybe, we need to remove this old ghc- x86 Generic Linux with glib 2.3, to think of path, etc. We shall try. ----------------- Serge Mechveliani mechvel(a)botik.ru

2 2

RE: 6.4.2 release plans
by Simon Marlow 20 Mar '06

20 Mar '06

On 18 March 2006 18:32, Ian Lynagh wrote: > On Wed, Mar 15, 2006 at 04:24:14PM +0000, Simon Marlow wrote: >> >> If you have anything else for 6.4.2, please let me know. > > Just a small thing, but it looks like, while the HEAD has been changed > to send --help output to stdout, ghc-6.4.2.20060316 still has the > following in ghc/compiler/main/DriverFlags.hs > > dump ('$':'$':s) = hPutStr stderr progName >> dump s > dump (c:s) = hPutChar stderr c >> dump s > > Is this deliberate, or could it be changed to > > dump ('$':'$':s) = putStr progName >> dump s > dump (c:s) = putChar c >> dump s > > please? Done Simon

1 0

ForeignPtr's - why can't they be passed directly to foreign functions?
by Brian Hulley 19 Mar '06

19 Mar '06

Hi - I've got the beginnings of an API for a GUI system as follows: data DWindow a data DEdit a type Window = DWindow type Edit a = DWindow (DEdit a) foreign import ccall duma_init :: IO () foreign import ccall duma_run :: IO () foreign import ccall duma_release :: FunPtr (Ptr (Window a) -> IO ()) foreign import ccall duma_createEdit :: IO (Ptr (Edit a)) foreign import ccall duma_addTop :: Ptr (Window a) -> IO () createEdit :: IO (ForeignPtr (Edit a)) createEdit = do edit <- duma_createEdit newForeignPtr duma_release edit addTop :: ForeignPtr (Window a) -> IO () addTop w = withForeignPtr w duma_addTop This works, but it seems a bit of a pain to have to manually convert between ForeignPtr's and Ptr's all the time. In particular, for the definition of addTop, I tried: foreign import ccall "duma_addTop" addTop :: ForeignPtr (Window a) -> IO () but got an error because ForeignPtr's are not allowed as part of the type of a foreign function. Since the definition of ForeignPtr is just void *, I wonder why this restriction exists - ie is a ForeignPtr not just the same address as the corresponding Ptr? My other question is what happens if I want to have a function that takes more than one ForeignPtr as argument ie foreign import ccall duma_test :: Ptr (Window a) -> Ptr (Window a) -> IO () test :: ForeignPtr (Window a) -> ForeignPtr (Window a) -> IO () test p q = withForeignPtr p (\p' -> withForeignPtr q $ duma_test p') Is this the only way to achieve this? It seems a bit long-winded and possibly a bit inefficient... One other question: if I use forkIO within Haskell, am I right in thinking that the lightweight concurrent threads are safe to use with my single threaded C code ie that there is no danger of a thread being preemptively halted while it is inside a foreign function? Thanks, Brian.

6 9

6.4.1 for SuSE-10.0 lib64
by Serge D. Mechveliani 17 Mar '06

17 Mar '06

Dear GHC team, We are trying to install ghc-6.4.1 on SuSE Linux 10.0, lib64. First, we install it to the system area from binary found on the GHC www page as Binary x86, Generic Linux with glib 2.3. Now, ~> rpm -qa | grep compat says compat-curl2-32bit-7.11.0-7.2 compat-32bit-2004.11.13-4 compat-libstdc++-5.0.7-6 compat-readline4-4.3-4 compat-2004.11.13-4 compat-curl2-7.11.0-7.2 Then, ghc makes the project (under Cabal), creates .o, .a libraries, after runhaskell Setup.hs install --user Then, ghc $dmCpOpt --make Main 'makes' the user module and needs to link all libraries. It reports --------- Chasing modules from: Main Skipping Main ( Main.hs, Main.o ) Linking ... /usr/lib64/gcc/x86_64-suse-linux/4.0.2/../../../../x86_64-suse-linux/bin/ld: cannot find -lreadline collect2: ld returned 1 exit status source/demotest> --------- We simplified the project to a single module module DPrelude (sublists) where sublists :: [a] -> [[a]] sublists [] = [[]] sublists (x:xs) = (map (x:) ls) ++ ls where ls = sublists xs The libraries are made of this module. The "user" program is module Main where import DPrelude main = let ll = sublists [1, 2 :: Integer] in putStr "abc" >> writeFile "fl" "abc" >> getChar >> putStr (shows ll "\n") >> appendFile "fl" "1" And it faills with this"readline". ------------------------------------------------------------------------- Then, we try the module which does not use newly created library, but uses input-output: main = putStr "abc" >> writeFile "fl" "abc" >> getChar >> putStr "abc" >> appendFile "fl" "1" And it makes and runs all right. ------------------------------------------------------------------------ Then, I tried to "make" GHC from source by this binary GHC. "configure" does not tell anything about Alex and Happy -- although I do not see them (forgotten of them). It finishes with exit 0, as if nothing has happened. Then, > make reports something which ends (after a couple of minutes) with utils/PrimPacked.lhs:263:0: Warning: foreign declaration uses deprecated non-standard syntax In file included from /tmp/ghc31185.hc:6: /usr/local/lib/ghc-6.4.1/include/HsReadline.h:5:31: error: readline/readline.h:\ No such file or directory /usr/local/lib/ghc-6.4.1/include/HsReadline.h:6:30: error: readline/history.h: \ No such file or directory <<ghc: 101068136 bytes, 20 GCs, 3654882/7854280 avg/max bytes residency (4 samp\ les), 20M in use, 0.00 INIT (0.00 elapsed), 0.27 MUT (0.81 elapsed), 0.11 GC (0\ .13 elapsed) :ghc>> make[2]: *** [stage1/utils/PrimPacked.o] Error 1 make[1]: *** [all] Error 1 make[1]: Leaving directory `/home/mechvel/ghc/6.4.1/ghc-6.4.1/ghc' make: *** [build] Error 1 Please, can you advise, how to make ghc-6.4.1 work under the above system? Does it matter the difference between lib64 and lib32 ? (we could remain with lib32). ----------------- Serge Mechveliani mechvel(a)botik.ru

2 1

newbie: building ghc
by Mark Greenbank 16 Mar '06

16 Mar '06

Hi, Sorry for the newbie question ... I'm trying to build the compiler and I'm getting the following error: [root@quark ghc-6.4.1]# ./configure --prefix=/usr/pkg checking build system type... i386-pc-solaris2.10 checking host system type... i386-pc-solaris2.10 checking target system type... i386-pc-solaris2.10 Canonicalised to: i386-unknown-solaris2 checking for path to top of build tree... /usr/pkg/src/ghc-6.4.1 checking for ghc... no checking for nhc... no checking for nhc98... no checking for hbc... no configure: error: GHC is required unless bootstrapping from .hc files. Does this somehow mean that I need the compiler in order to build the compiler? Thanks, Mark

2 3