June 2003 - Glasgow-haskell-users

RE: 5.04.3: update_fwd: unknown/strange object
by Simon Marlow 18 Jun '03

18 Jun '03

> It occurs that we need first to install ghc-6.0 > on > i386-unknown, RedHat Linux platform. > > Our administrator says here is a difficulty: > > a) it is hard for us to install from sources, > b) binary RPM would not do, due to library versions etc., > c) we need > source RPM distributive, > > similar as we used when installing ghc-5.04.3. > > But ghc-6.0 does not show any source RPM distributive. > > Please, what can we do about this? Use the nosrc RPM, or install locally in your home directory from the binary distribution? (you don't say what version of RedHat you have). Cheers, Simon

2 1

RE: isEmptyChan blocks?
by Simon Marlow 18 Jun '03

18 Jun '03

> I noticed that isEmptyChan blocks if the channel is already > waited on by readChan, is this the intended behaviour? > At least I was a bit surprised by a blocking predicate.. I agree this doesn't seem right. Looking at the implementation though, I can't see an easy way to fix it - any suggestions? Cheers, Simon

1 0

RE: "How do I find out a function's strictness?"
by Simon Marlow 18 Jun '03

18 Jun '03

> In the ghc manual I read : > > "How do I find out a function's strictness?" > Don't guess-look it up. > > Look for your function in the interface file, then for the > third field in the > pragma; it should say __S <string>. The <string> gives the > strictness of the > function's arguments. L is lazy (bad), S and E are strict > (good), P is > "primitive" (good), U(...) is strict and "unpackable" (very > good), and A is > absent (very good). > > -- > > When I look in the *.hi binary files generated by ghc with > ghc --show-iface, I > don't see anything like "__S" anywhere. > > Where should I look ? Make sure the module is compiled with -O, otherwise the strictness analyser isn't run and you don't get any strictness info. Cheers, Simon

3 2

@-bindings broken in 6.0?
by John Meacham 18 Jun '03

18 Jun '03

change in list comprehension parsing? after I upgraded to ghc 6.0 there were some odd syntax errors on list comprehensions which i did not get before, making sure i put spaces between all symbols seems to fix them.. was something changed about how they are parsed which breaks code that used to work? just curious. John -- --------------------------------------------------------------------------- John Meacham - California Institute of Technology, Alum. - john(a)foo.net ---------------------------------------------------------------------------

2 1

isEmptyChan blocks?
by Peter Strand 17 Jun '03

17 Jun '03

Hi, I noticed that isEmptyChan blocks if the channel is already waited on by readChan, is this the intended behaviour? At least I was a bit surprised by a blocking predicate.. Example program: import Control.Concurrent main = do ch <- newChan let loop = do threadDelay 1000000 -- make sure readChan executes isEmptyChan ch print "not reached" loop forkIO loop readChan ch This happens both with 5.04.2 and 6.0, on linux. When compiled, "Fail: thread blocked indefinitely" is printed, ghci just blocks. The corresponding program with MVars doesn't block. /Peter

1 0

RE: 5.04.3: update_fwd: unknown/strange object
by Simon Marlow 17 Jun '03

17 Jun '03

> I build by new DoCon program under ghc-5.04.3, > > installed from RPM to i386-unknown, RedHat Linux. > > Under -Onot, the command make space=-M22m docon > builds the thing successfully. > And for less memory, say, ... 16m ..., > it ends with > > ... > pol/Pgcd_.hs:184: > Warning: No explicit method nor default method for `toSquareFree' > In the instance declaration for `GCDRing (Pol a)' > > =================== INTERFACE HAS CHANGED ==================== > No old interface available > > ghc-5.04.3: fatal error: update_fwd: unknown/strange object 38344 > make: *** [obj] Error 254 > > > I expected of GHC to report of exhausted memory. > Probably, such behavior is not fatal for may aims, > but if you consider this as an interesting bug, I could send the > archive. I'll be interested to know whether you can repeat this with 6.0. If so, yes I'd like to see the sources. Cheers, Simon

2 1

5.04.3: update_fwd: unknown/strange object
by Serge D. Mechveliani 17 Jun '03

17 Jun '03

Dear GHC team, I build by new DoCon program under ghc-5.04.3, installed from RPM to i386-unknown, RedHat Linux. Under -Onot, the command make space=-M22m docon builds the thing successfully. And for less memory, say, ... 16m ..., it ends with ... pol/Pgcd_.hs:184: Warning: No explicit method nor default method for `toSquareFree' In the instance declaration for `GCDRing (Pol a)' =================== INTERFACE HAS CHANGED ==================== No old interface available ghc-5.04.3: fatal error: update_fwd: unknown/strange object 38344 make: *** [obj] Error 254 I expected of GHC to report of exhausted memory. Probably, such behavior is not fatal for may aims, but if you consider this as an interesting bug, I could send the archive. ----------------- Serge Mechveliani mechvel(a)botik.ru

1 0

RE: problems with working with Handles
by Simon Peyton-Jones 17 Jun '03

17 Jun '03

| Prelude Strategies IO> hClose h | Prelude Strategies IO> length cs | 737535 | Prelude Strategies IO> | | Note that you have to use -package Strategies option. I couldn't find | this module in the hierarchical libraries. The same works in GHC 5.04.3 | | To GHC developers: | What is the status of this module? | Can I assume that future versions of GHC will have it? It's called Control.Parallel.Strategies. Current plan is to leave it there, and ship it with future versions of GHC. Simon

1 0

"How do I find out a function's strictness?"
by Francis Girard 17 Jun '03

17 Jun '03

Hi, In the ghc manual I read : "How do I find out a function's strictness?" Don't guess-look it up. Look for your function in the interface file, then for the third field in the pragma; it should say __S <string>. The <string> gives the strictness of the function's arguments. L is lazy (bad), S and E are strict (good), P is "primitive" (good), U(...) is strict and "unpackable" (very good), and A is absent (very good). -- When I look in the *.hi binary files generated by ghc with ghc --show-iface, I don't see anything like "__S" anywhere. Where should I look ? Thank you Francis Girard francis.girard(a)free.fr LE CONQUET, France

1 0

Re: [Gtk2hs-users] ghci-5.04.3 and ghc-6.0
by Duncan Coutts 16 Jun '03

16 Jun '03

On 16 Jun 2003 16:59:07 +0900 Jens Petersen <petersen(a)redhat.com> wrote: > Did anyone ever manage to get gtk2hs working with ghci-5.04.3? > When I try to load it I get: > > % ghci-5.04.3 -package gtk2 > : > Loading package data ... linking ... done. > Loading package gtk2 ... > > GHCi runtime linker: fatal error: I found a duplicate definition for > symbol > __stginit_Arrow > whilst processing object file > /usr/lib/ghc-5.04.3/gtk2hs/gtk2hs.o My brief investigation into this revealed that __stginit_Arrow is defined in HSlang.o (ghci's object file for package lang) and this is the only standard package in which it is defined. __stginit_Arrow is indeed also defined in gtk2hs.o It is defined there because it is defined in gtk/misc/Arrow.o (in the buid tree). Itis defined there because Arrow is the name of the module and that symbol refers to the module initialisation function (or something like that, perhaps it's a module dependancy marker symbol). What seems to be happening is that the Arrow module in gtk2hs is clashing with the Arrow module in the lang package. As far as I can tell this all looks perfectly normal, so why don't we this kind of problem all the time? What's special about Arrow? How does it get resolved in the libHS*.a case? How does the system linker resolve the ambiguity when ghci's linker cannot? Duncan

1 0