- Glasgow-haskell-users

RE: source RPM ghc-6.0
by Simon Marlow 19 Jun '03

19 Jun '03

> Is this a correct way to detect versions needed for the GHC > installation? > > more /etc/redhat-release > --> > Red Hat Linux release 7.3 (Valhalla) > > ls /lib/libc* > --> > /lib/libc-2.2.5.so > ... The binary distribution (not the RPM) is basically just a tarball. It does no checking that you have the right versions of things, so it is a "use at your own risk". The download instructions do say that it is suitable for RedHat 7.2, though, so you might be ok with 7.3. If you want to build from source, you can use the "nosrc" RPM. This is just like the source RPM, except that the source isn't included. To build it, you need to download the source too (your sysadmin should know how to do this). Cheers, Simon

2 1

how and where to {-# specialize #-} ?
by Johannes Waldmann 19 Jun '03

19 Jun '03

Dear all, I am looking for some advice on where to place specialize-pragmas for ghc (to optimize speed - and heap usage) Typically I have some module A (containing polymorphic functions A.f, ..) and a module B uses this. The interesting case is when I want to specialize A.f (which is NOT exported) to some types that are only defined in B (but normally A should not know about B). Is there some kind of `deep specialisation' where one would say, `specialize ALL the functions that are reachable from here'? - or how can one get a similar effect? ( Of course I would be willing to pay some compilation time for this. Even a LOT of time. ) Thanks, -- -- Johannes Waldmann ---- http://www.informatik.uni-leipzig.de/~joe/ -- -- joe(a)informatik.uni-leipzig.de -- phone/fax (+49) 341 9732 204/209 --

1 0

RE: state of ghc6 on sparc
by Simon Peyton-Jones 19 Jun '03

19 Jun '03

| > often...). The problem around this is in my opinion to make a | > Haskell98 | > compliant trimmed down readline library which doesn't depend on the c | > library. It shouldn't be too difficult. I don't expect that people are | > using the full power of readline anyway. | | Making a working readline isn't that simple - you need to know something | about the terminal properties for one, which means being able to use | termcap or terminfo. But if someone produced an adequate one that did the job, and that we could distribute with GHC, we'd surely use it, right? To avoid these perpetual readline problems. Simon

1 0

RE: current CVS version doesn't compile
by Simon Marlow 18 Jun '03

18 Jun '03

> Simon Marlow writes: > > > What version of GHC are you compiling with? > > The Glorious Glasgow Haskell Compilation System, version 6.1 > > I built this version from CVS as well, maybe two or three days before > the 6.0 version was released. We only support compiling the CVS sources with: a. A released compiler (i.e. one ending in an even number), or b. itself. So, building with a random 6.1 isn't supported, hence the build error. You should start from 6.0. Cheers, Simon

2 1

RE: state of ghc6 on sparc
by Simon Marlow 18 Jun '03

18 Jun '03

> On Wed, 18 Jun 2003, Isaac Jones wrote: > > Can anyone enlighten me as to the state of the ghc6 sparc > > distribution? The file on Hal's web page mentioned some > time back is > > not the same file as on the GHC web page (I note that the filenames > > are different also). > > > > Are any of them known to work if installed in a non-standard place? > > > Here at Chalmers we use the first shipped version (I believe > this is the > one on the GHC web page or an even earlier one...). I patched > it myself, > it was very easy. I expect that the version on Hal's web page > should work since he said he fixed the bug. Hal's version still has problems, I believe. He said that he will try to fix the problems soon. > The only thing that I don't like is that the dist isn't compiled with > readline. It makes working in ghci a nightmare (I need to use > backspace > often...). The problem around this is in my opinion to make a > Haskell98 > compliant trimmed down readline library which doesn't depend on the c > library. It shouldn't be too difficult. I don't expect that people are > using the full power of readline anyway. Making a working readline isn't that simple - you need to know something about the terminal properties for one, which means being able to use termcap or terminfo. Cheers, Simon

1 0

RE: current CVS version doesn't compile
by Simon Marlow 18 Jun '03

18 Jun '03

> I'm trying to compile the latest version of GHC from CVS on Linux, but > the build reproducibly fails at this point: > > main/SysTools.lhs:703: > Couldn't match `IO ExitCode' against `t -> t1' > Expected type: IO ExitCode > Inferred type: t -> t1 > Probable cause: `rawSystem' is applied to too many > arguments in the call > (rawSystem pgm real_args) > In a 'do' expression: exit_code <- rawSystem pgm real_args > make[2]: *** [stage1/main/SysTools.o] Error 1 > make[1]: *** [all] Error 1 > make[1]: Leaving directory `/usr/local/src/ghc-current/ghc' > > Does anyone have an idea what's going wrong here? What version of GHC are you compiling with? Cheers, Simon

2 1

current CVS version doesn't compile
by Peter Simons 18 Jun '03

18 Jun '03

I'm trying to compile the latest version of GHC from CVS on Linux, but the build reproducibly fails at this point: main/SysTools.lhs:703: Couldn't match `IO ExitCode' against `t -> t1' Expected type: IO ExitCode Inferred type: t -> t1 Probable cause: `rawSystem' is applied to too many arguments in the call (rawSystem pgm real_args) In a 'do' expression: exit_code <- rawSystem pgm real_args make[2]: *** [stage1/main/SysTools.o] Error 1 make[1]: *** [all] Error 1 make[1]: Leaving directory `/usr/local/src/ghc-current/ghc' Does anyone have an idea what's going wrong here? Peter

1 0

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