- Glasgow-haskell-users

RE: unsafePerformIO
by Simon Peyton-Jones 24 Sep '02

24 Sep '02

The actions performed by unsafePerformIO are simply done at some entirely unpredictable time, perhaps interleaved with actions on the main execution path. The formal semantics in the notes doesn't have a good way to express that because the purely-functional part is given a denotational semantics... yet unsafePerformIO can occur in the middle of that. So it's tiresome to characterise in theory, and hard to predict in practice. You should only use it when the state it is fiddling with is well partitioned. Simon | -----Original Message----- | From: David Sabel [mailto:dsabel@stud.uni-frankfurt.de] | Sent: 20 September 2002 13:48 | To: Simon Peyton-Jones; glasgow-haskell-users(a)haskell.org | Subject: unsafePerformIO | | In read your paper ""Tackling the Awkward Squad: monadic input / output, | concurrency, exceptions, and foreign-language calls in Haskell", and have | a question about unsafePerformIO. | | In your operational semantic of the IO-Monad you tell nothing about, how | 'unsafe' IO actions are performed, is there another paper / documentation | about | this available, or can you - or someone else - give me a review about that? | | David | JWGU Frankfurt |

3 2

RE: unsafePerformIO
by Simon Peyton-Jones 24 Sep '02

24 Sep '02

Koen Based on this info, how would you like to write the notes you would like to see on unsafePerformIO? The starting point is at http://haskell.cs.yale.edu/ghc/docs/latest/html/base/GHC.IOBase.html#uns afePerformIO If you write the notes, we'll check their veracity and add them. We'll do the formatting too! Simon | -----Original Message----- | From: Simon Marlow [mailto:simonmar@microsoft.com] | Sent: 24 September 2002 13:59 | To: Koen Claessen; glasgow-haskell-users(a)haskell.org | Subject: RE: unsafePerformIO | | > But it is a fact that many of us have at least some idea of | > what happens "under the hood" when we use unsafePerformIO. | > This is also described in your paper "Stretching the storage | > manager: weak pointers and stable names in Haskell". | > | > However, I for example have no idea what happens when | > unsafely executing something that throws exceptions, | > performs a forkIO, something that uses MVar's, etc. | | Exceptions: an exception raised by the computation inside | unsafePerformIO will be propagated to the enclosing expression, as per | the normal semantics for imprecise exceptions. The exception raised by | an application of unsafePerformIO may of course also be imprecise; | consider 'unsafePerformIO $ (1/0 + error "foo") `seq` return ()'. | | The behaviour of forkIO and other side-effecting operations inside | unsafePerforIO can I think be explained by the following statement: if | an expression "unsafePerformIO e" has been reduced to head normal form, | then all the I/O computations in e have been performed. The exact | timing of the evaluation is underspecified, just as any other expression | in Haskell. However, if you're using unsafePerformIO with real side | effects, then I suspect that what you're doing is highly dodgy and you | should find another way to do it :-) | | Does that help? | | As for sharing, we currently don't provide any guarnatees, although we | should. It is currently the case that if you write | | a = unsafePerformIO (putStr "hello") | b = unsafePerformIO (putStr "hello") | | and both a and b are evaluated, then you may get either one or two | "hello"s on stdout. You can currently make things more deterministic by | (a) adding NOINLINE pragmas for a and b, and (b) using the flag -fno-cse | to disable common sub-expression elimination. Then you'll get exactly | two instances of "hello" on stdout, although we won't guarantee that | behaviour for ever. At some point we'll fix it so that unsafePerformIO | applications are never duplicated or coalesced. | | If you're interested in when unsafePerformIO is "safe" to use, see: | | http://www.haskell.org/pipermail/glasgow-haskell-users/2002-July/003683. | html | | (and see my followups for an alternative view). | | Cheers, | Simon | _______________________________________________ | Glasgow-haskell-users mailing list | Glasgow-haskell-users(a)haskell.org | http://www.haskell.org/mailman/listinfo/glasgow-haskell-users

1 0

ghc-inplace exiting with "error 1" foolishness
by Dominic Cooney 24 Sep '02

24 Sep '02

Hi, I complained late last week that GHC failed to build because ghc-inplace couldn't do anything more complicated than print its version number. I tried again on a clean machine, this time remembering to replace Cygwin's sh.exe with bash.exe, and mounting C:\mingw as /mingw and C:\mingw\include as /mingw/mingw32/include. GHC apparently builds now (although the build hasn't finished... I've only 256 MB RAM and serious thrashing goes on when the build is doing this xargs ar thing!). Thanks for your help and patience! Dominic Cooney

2 1

RE: Newbie building GHC
by Simon Peyton-Jones 24 Sep '02

24 Sep '02

| In retrospect I think that unless "configure" is doing something really | tricky, if you want the Mingw32 "ar" and "ld" to be picked up instead of the | Cygwin equivalents then you must have the Mingw32 bin directory in your path | ahead of the Cygwin bin directory. The right thing to do is what the building guide says. The GHC build system is careful to use the newly-built GHC for compiling even .c files, and for doing linking. The newly-built GHC will use the gcc that you point to with the --with-gcc configure command. It will not use a path in any shape or form (unless you omit --with-gcc in which case it'll pick whatever random gcc is lying around; probably not the right one). The only way I know to make all this work is to use a straightforward Cygwin environment for building (with absolutely no mingw stuff in your path) and use --with-gcc in your configure line. Other things may work, but you're on your own. Simon | | You should also remove the Mingw bin "make.exe" as it is seriously flawed | and will not suffice to do much at all, let alone build GHC. | | Cheers | | Mike Thomas. | | | ----- Original Message ----- | From: "Simon Peyton-Jones" <simonpj(a)microsoft.com> | To: "Mike Thomas" <miketh(a)brisbane.paradigmgeo.com>; | <glasgow-haskell-users(a)haskell.org> | Cc: <mthomas(a)gil.com.au> | Sent: Tuesday, September 24, 2002 5:30 PM | Subject: RE: Newbie building GHC | | | | - Run the following following commands: | | | | export PATH=/cygdrive/c/lang/mingw32/bin::${PATH} | | | | make clean; ./configure --build=mingw >& configure.log | | Uh oh. You aren't following the instructions in Section 12.4 of the GHC | building guide! | http://haskell.cs.yale.edu/ghc/docs/latest/html/building/winbuild.html | | (a) I believe that it's important to run autoconf | (b) I've never seen this --build option for configure. What I use (as | the building guide says) is: | | ./configure --host=i386-unknown-mingw32 --with-gcc=/mingw/bin/gcc | | You need to run configure in fptools/ghc first actually. (No need to | autoconf there.) | | | What you did may work, but I am not confident in it. For example where | are you saying where to find the (mingw) gcc that GHC should use? | | Simon |

1 0

RE: Newbie building GHC
by Simon Peyton-Jones 24 Sep '02

24 Sep '02

| - Run the following following commands: | | export PATH=/cygdrive/c/lang/mingw32/bin::${PATH} | | make clean; ./configure --build=mingw >& configure.log Uh oh. You aren't following the instructions in Section 12.4 of the GHC building guide! http://haskell.cs.yale.edu/ghc/docs/latest/html/building/winbuild.html (a) I believe that it's important to run autoconf (b) I've never seen this --build option for configure. What I use (as the building guide says) is: ./configure --host=i386-unknown-mingw32 --with-gcc=/mingw/bin/gcc You need to run configure in fptools/ghc first actually. (No need to autoconf there.) What you did may work, but I am not confident in it. For example where are you saying where to find the (mingw) gcc that GHC should use? Simon

2 2

RE: Dynamic/runtime loading with GHC
by Simon Peyton-Jones 24 Sep '02

24 Sep '02

Great! Simon M is going to take a look when he has a moment. Meanwhile I'd like to encourage those of you who are keen on this stuff [Duncan, Andre?] to write up (a) how to use it (in SGML, to add to the user manual); see fptools/ghc/docs/using (b) how it is implemented (in HTML, to add to the GHC Commentary), including not only the new stuff, but what you know about the existing stuff see fptools/ghc/docs/comm (c) patches for what needs to change in the GHC and/or libraries code base to support it That way it'll get into our main code base and get maintained. Simon M has way too much to do! Simon | -----Original Message----- | From: Andre Pang [mailto:ozone@algorithm.com.au] | Sent: 21 September 2002 15:19 | To: glasgow-haskell-users(a)haskell.org | Cc: Martin Norbäck; Ganesh Sittampalam; Shae Matijs Erisson; Manuel Chakravarty; | Duncan Coutts | Subject: Dynamic/runtime loading with GHC | | Hi all, | | I've whipped up an example tarball of how to do "runtime loading" | (similar to dynamic loading, a.k.a. dlopen() or LoadLibrary()) in | GHC. You can find it here: | | http://www.algorithm.com.au/wiki/hacking/haskell.ghc_runtime_loading | | It's a little example stdin text filter program, that loads in .o | and .so "plugins" at runtime. New in version 1.1 is the ability | to load .so files as plugins :). I think that with a bit more | polish, we may even be able to rename this to a dynamic loader | one day. | | Have fun! | | | -- | #ozone/algorithm <ozone(a)algorithm.com.au> - trust.in.love.to.save | _______________________________________________ | Glasgow-haskell-users mailing list | Glasgow-haskell-users(a)haskell.org | http://www.haskell.org/mailman/listinfo/glasgow-haskell-users

1 0

RE: Newbie building GHC
by Simon Marlow 24 Sep '02

24 Sep '02

> I ran strace with -ff option. Attached is the trace file. I > can see many "No > such file or directory" in the trace. Can you see the problem? Here's the problem line: [pid 1514] execve("n/sh", ["/bin/sh", "-c", "gcc Adjustor.c -o /tmp/ghc1513.s"...], [/* 36 vars */]) = -1 ENOENT (No such file or directory) so for some unknown reason, execve() is trying to run "n/sh" rather than "/bin/sh". execve() is called by the Haskell System.system function, via execl(). I have absolutely no idea what is causing this :) Simon PJ said it happened to him at some point, but neither of us can remember the cause. Just in case it gives us any clues, can you tell us: - what OS flavour/version is on your system - what version of GHC is installed, and where you got it from, - what GHC sources you are trying to build (source bundle or CVS?) also, could you try building & running a small program that calls System.system on your machine with the installed GHC, and see if it works as expected? Cheers, Simon

1 0

RE: ghci takes all available CPU time
by Simon Marlow 24 Sep '02

24 Sep '02

> On our system (Solaris) the interpreter takes all available > CPU time just > waiting for input. The problem seems to be restricted to our site and > to the latest release of ghc (5.04). We really need to fix > this problem > since many of us are using SunRays with a shared server. Any > idea of what may cause this behaviour? I've seen this before, and I think it was caused by an old version of the readline library. The current release (5.04.1) certainly doesn't eat all the CPU on our Sparc here. Perhaps you could try 5.04.1? Cheers, Simon

1 0

ghci takes all available CPU time
by Niklas Sörensson 24 Sep '02

24 Sep '02

Hi, On our system (Solaris) the interpreter takes all available CPU time just waiting for input. The problem seems to be restricted to our site and to the latest release of ghc (5.04). We really need to fix this problem since many of us are using SunRays with a shared server. Any idea of what may cause this behaviour? /Niklas

1 0

RE: ghc-inplace exiting with "error 1" foolishness
by Simon Peyton-Jones 24 Sep '02

24 Sep '02

| I tried again on a clean machine, this time remembering to replace | Cygwin's sh.exe with bash.exe, and mounting C:\mingw as /mingw and | C:\mingw\include as /mingw/mingw32/include. You should not need the latter two step, if you use the --with-gcc option to configure, as the building guide advises. Simon

1 0