December 2001 - Glasgow-haskell-users

GHC-3.02 (maybe OFF-time)
by Rafael Martínez Torres 20 Dec '01

20 Dec '01

Hi all: I'm hacking with ghc-3.02.src.tar.gz on several platforms, namely i386 and sparc ... Both platforms run with gcc-2.7.2.3 ( higher versions may be a risk , 2.95 was bad...) In sparc I have no problems, but in i386 I do obtain the next results (File : fptools/ghc/lib/posix/PosixProcPrim.hc) "../../../ghc/driver/ghc -i../misc -recomp -cpp -fglasgow-exts -fvia-C -Rghc-timing -O '-#include"cbits/libposix.h"' -c PosixProcPrim.hc -o PosixProcPrim.o -osuf o PosixProcPrim.hc:2244: fixed or forbidden register was spilled. This may be due to a compiler bug or to impossible asm statements or clauses." I think cc1 is reducing uncorrectly to asm code. Reading documents on gcc, I got this... ( File: /usr/docs/gcc/PROBLEMS ) May be has nothing to do... "3. When find_reloads is used to count number of spills needed it does not take into account the fact that a reload may turn out to be a dummy." May be this is rather a question for gcc , but... Does any body know where to add any flag to help gcc to reduce OK ? I tried no -O flag , -O2 ,-O1, but no sucess... Regards... Martinez-Torres.

1 0

RE: GHC-3.02 (maybe OFF-time)
by Simon Marlow 20 Dec '01

20 Dec '01

> I'm hacking with ghc-3.02.src.tar.gz on several platforms, namely i386 > and sparc ... > > Both platforms run with gcc-2.7.2.3 ( higher versions may be a risk , > 2.95 was bad...) > > In sparc I have no problems, but in i386 I do obtain the next results > (File : fptools/ghc/lib/posix/PosixProcPrim.hc) > > "../../../ghc/driver/ghc -i../misc -recomp -cpp -fglasgow-exts -fvia-C > -Rghc-timing -O '-#include"cbits/libposix.h"' -c > PosixProcPrim.hc -o > PosixProcPrim.o -osuf o > PosixProcPrim.hc:2244: fixed or forbidden register was spilled. > This may be due to a compiler bug or to impossible asm > statements or clauses." You need to give GHC the -monly-3-regs (or possibly -monly-2-regs) option. Cheers, Simon

1 0

RE: unexpected space leak
by Julian Seward (Intl Vendor) 20 Dec '01

20 Dec '01

It's a hack we put in, because fixing the bytecode generator properly for this relatively obscure case was a lot of effort and other calls on our time (we were stabilising for the 5.02 release at the time) were more pressing. It is (nearly) harmless. What usually causes it is having strict polymorphic fields on constructors: data T a = T !a The warning is intended to mean that the strictness annotation will be ignored in interpreted code. This shouldn't have any effect on the overall outcome of your program, unless it depends on evaluation order changes induced by the !, for some gruesome reason. However, the space behaviour may well be different from what it would be if ! was observed. If you do experience problems as a result of this, I'd be interested to hear of them. J | -----Original Message----- | From: Hal Daume III [mailto:hdaume@ISI.EDU] | Sent: Wednesday, December 19, 2001 11:22 PM | To: GHC Users Mailing List | Subject: unexpected space leak | | | what does it mean when ghci says: | | WARNING: ignoring polymorphic case in interpreted mode. | Possibly due to strict polymorphic/functional constructor args. | Your program may leak space unexpectedly. | | | -- | Hal Daume III | | "Computer science is no more about computers | hdaume(a)isi.edu | than astronomy is about telescopes." -Dijkstra | www.isi.edu/~hdaume | | | _______________________________________________ | Glasgow-haskell-users mailing list | Glasgow-haskell-users(a)haskell.org | http://www.haskell.org/mailman/listinfo/glasgow-haskell-users |

1 0

(no subject)
by mailer.daemon＠server1.home4.com 20 Dec '01

20 Dec '01

1 0

unexpected space leak
by Hal Daume III 19 Dec '01

19 Dec '01

what does it mean when ghci says: WARNING: ignoring polymorphic case in interpreted mode. Possibly due to strict polymorphic/functional constructor args. Your program may leak space unexpectedly. -- Hal Daume III "Computer science is no more about computers | hdaume(a)isi.edu than astronomy is about telescopes." -Dijkstra | www.isi.edu/~hdaume

1 0

RE: ghc-5.02.2
by Simon Peyton-Jones 19 Dec '01

19 Dec '01

"in due course" is a native English way of saying "I'm not quite sure" :-) We are planning a release before Christmas. Whether we will actually get it out in the next 3 days remains to be seen, but if not then it'll be early in the new year. Simon | -----Original Message----- | From: Sven Eric Panitz [mailto:sep@softwareag.com] | Sent: 19 December 2001 10:41 | To: glasgow-haskell-users(a)haskell.org | Subject: ghc-5.02.2 | | | | Some time ago Simon Peyton-Jones posted on this list: | 'We will do 5.02.2 in due course' | | Being not a native English speaker, I am not sure about all | subleties of the English language; I wonder what time frame | 'in due course' indicates? Are there any concrete plans, when | ghc-5.02.2 will be available? | | (otherwise I might seek the thrill of compiling ghc from cvs | sources ...) | | | hope this does not sound too much like nagging | | Sven Eric | | | _______________________________________________ | Glasgow-haskell-users mailing list | Glasgow-haskell-users(a)haskell.org | http://www.haskell.org/mailman/listinfo/glasgow-| haskell-users |

1 0

Re: Optimization and GCC?
by Marcin 'Qrczak' Kowalczyk 19 Dec '01

19 Dec '01

Wed, 19 Dec 2001 09:56:36 +0100, Till Doerges <till(a)doerges.net> pisze: > /u/joe/built/lib/ghc-5.00.2/includes/Regs.h:255: warning: call-clobbered register used for global register variable I don't know about this one. > /tmp/ghc15733.hc:5391: warning: deprecated initialization of zero-length array > /tmp/ghc15733.hc:5391: warning: (near initialization for `Main_main_closure.payload') This is already fixed in CVS. ghc3 is not officially supported yet. Nevertheless I'm using it at home and it can successfully compile almost everything from ghc on ix86. This is my mk/build.mk - I'm compiling three files with gcc-2.95.3: ProjectsToBuild = glafp-utils ghc hslibs happy PrimOps_HC_OPTS = -pgmcgcc-2.95.3 ArrayBase_HC_OPTS = -pgmcgcc-2.95.3 PosixTTY_HC_OPTS = -pgmcgcc-2.95.3 This was for gcc-3.0.1 and ghc a few weeks ago. I'll see if it's better now, with gcc-3.0.2. The problem with PrimOps looked like gcc bug, as if it ignored -fomit-frame-pointer for this file. -- __("< Marcin Kowalczyk * qrczak(a)knm.org.pl http://qrczak.ids.net.pl/ \__/ ^^ QRCZAK

1 0

ghc-5.02.2
by Sven Eric Panitz 19 Dec '01

19 Dec '01

Some time ago Simon Peyton-Jones posted on this list: 'We will do 5.02.2 in due course' Being not a native English speaker, I am not sure about all subleties of the English language; I wonder what time frame 'in due course' indicates? Are there any concrete plans, when ghc-5.02.2 will be available? (otherwise I might seek the thrill of compiling ghc from cvs sources ...) hope this does not sound too much like nagging Sven Eric

1 0

RE: converting to & from unboxed types
by Simon Marlow 19 Dec '01

19 Dec '01

> > I'm sure this is common knowledge but I can't find it > anywhere in the ghc > docs. How do you take, say, an Integer and "unbox" it and > get an Int# or > something? import GlaExts unbox :: Int -> Int# unbox (I# i) = i box :: Int# -> Int box i = I# i Cheers, Simon

1 0

Re: ghc image size
by Albert Lai 19 Dec '01

19 Dec '01

Mike Gunter <m(a)ryangunter.com> writes: > Why is executable size a barrier? 1.64 megabytes (that's the size of > the executable I built with GHC most recently) of disk space costs less > than half a cent. I don't like this argument. Can I go to a computer store, pay a cent, and get a hard disk with space 1.64 megabytes or more? Until then, I can't believe that 1.64 megabytes of disk space costs less than half a cent. When a compiler does not perform as good as other compilers (e.g., in terms of generated code size), it is important to ask: Why does it happen? Is there anything we can do to improve it? Being critical is the first step towards progress. (Of course these questions should be asked in a constructive rather than whining way.) Why would anyone optimize code for time --- a second of electricity and labour cost less than a cent...

3 2