- Glasgow-haskell-users

IO concurrency
by Max A . K . 10 Oct '01

10 Oct '01

Hello. Playing with GHC, I met some oddity. Consider the example: > import IO > import Concurrent > import Posix > import PosixIO > > main = do > (fdIn, fdIn_send) <- createPipe > hIn_send <- fdToHandle fdIn_send > hIn <- fdToHandle fdIn > -- The "waiter" is a simple tk/tcl script which > -- prints "click" anytime I click a button. > runProcess "./waiter" [] Nothing Nothing > Nothing (Just hIn_send) Nothing > -- hSetBuffering hIn LineBuffering {-2-} > -- hSetBuffering stdin LineBuffering > forkIO(interact stdin) > -- forkIO(interact hIn) {-1-} > main_cycle > where > interact hIn = do > line <- hGetLine hIn > numIn <- handleToInt {-made from handleToFd & fdToInt-} > -- threadWaitRead numIn {-3-} > putStrLn line > interact hIn > main_cycle = do > putStrLn "tick" > threadDelay 1000000 > main_cycle The example works OK. It prints "tick" once a second and dups my input as I type it. But, when I uncomment the line {-1-} to handle input from child as well, everything is hanged on this read operation. Uncommenting the {-3-} line helps, but, first, I'm not sure it is a right way and, second, the matter is still unclear - what is the reason of such difference? Uncommenting the {-2-} lines doesn't help. I thing the reason is in some inner properties of the handlers, but have no idea in which one. Max.

1 1

ghc 5.02 and Solaris ld
by Akop Pogosian 10 Oct '01

10 Oct '01

$ grep 'ld.*-x' `find . -name Makefile` ./ghc-5.02/ghc/lib/std/Makefile: ld -r -x -o $@ $(GHCI_LIBOBJS) ./ghc-5.02/ghc/lib/std/Makefile: ld -r -x -o HSstd1.o $(filter Prel%, $(GHCI_LIBOBJS)) ./ghc-5.02/ghc/lib/std/Makefile: ld -r -x -o HSstd2.o $(filter-out Prel%, $(GHCI_LIBOBJS)) ./ghc-5.02/hslibs/win32/Makefile: ld -r -x -o HSwin321.o $(filter Win32M% Win32N% Win32R% Win32W%, \ ./ghc-5.02/hslibs/win32/Makefile: ld -r -x -o HSwin322.o $(filter-out Win32M% Win32N% Win32R% Win32W%, \ It looks like ghc makefiles assume that I am using GNU ld. The -x option does not exist in Solaris version of ld. Is this just a bug or ghc was not meant to be linked with other versions of ld? Just wondering, I could always use GNU ld instead (its just not the first thing on my PATH) According to the SunOS ld manual '-z redlocsym' seems to do what -x does. -z redlocsym Eliminates all local symbols except for the SECT symbols from the symbol table SHT_SYMTAB. All relocations that refer to local symbols will be updated to refer to the corresponding SECT sym- bol. So, I'll try to edit the makefiles accordingly and see what happens. --akop

1 0

Re: HOpengl on Ghc 5.02
by cwitty＠newtonlabs.com 09 Oct '01

09 Oct '01

Nicolas <Nicolas.Oury(a)ens-lyon.fr> writes: > Hi there, > Sorry for this stupid question: > Is there a distrib of a HOpenGl package working with ghc 5.02. I tried > the CVS but don't manage to make it work (ghc 5.03 panic). > Can someone help me? I got HOpenGL to work without trouble. On September 29, I checked out the ghc-5-02 branch from CVS, configured it with --enable-hopengl, and built and installed it. (I actually built a Debian package, using the Debian packager's build scripts.) I doubt if it matters, but I'm using HOpenGL with gtk+hs and gtkglarea instead of Glut. Carl Witty

1 0

RE: ghc-pkg
by Simon Marlow 09 Oct '01

09 Oct '01

> Additionally, ghc --make lacks (AFAIK) several useful > features found in > other make tools (although not all in the same tool...): > > 1. The ability to distingush directories containing source > code to be > compiled from directories containing previously compiled code > (hmake has the flags -P and -i, hbcmake has -i and -I). (I guess > you could use -package-conf as a cumbersome substitute, though.) > > 2. The ability to specify compiler flags for individual modules > without putting them in the source code. (Some flags > are 'static' > and can not be put in the source code.) > 3. The ability to compile several modules in parallel, on a > multi-processor machine, or a network of workstations. > 4. The ability to automatically invoke program generators > (e.g. happy)... > 5. A graphical user interface. All valid arguments, of course. But in --make's favour, it *is* much faster than individual compiles. In a little test I did today, ghc --make beats hmake/nhc98 on a reasonably sized program (nofib/real/anna). I timed both with -H32m and no optimisation, ghc --make compiled it in 48 seconds compared to 51 seconds for hmake/nhc98. On small individual modules, nhc98 wins hands down though. Ok, so speed isn't everything, and I hear the arguments you enumerated above. Most of these can be achieved through extra compiler support, and the last one (a make GUI) should actually be easier with GHC once we get around to specifying the compiler's programmatic API more precisely. Cheers, Simon

2 1

HOpengl on Ghc 5.02
by Nicolas 08 Oct '01

08 Oct '01

Hi there, Sorry for this stupid question: Is there a distrib of a HOpenGl package working with ghc 5.02. I tried the CVS but don't manage to make it work (ghc 5.03 panic). Can someone help me? Cheers, Nicolas Oury

1 0

RE: what is :a
by Simon Marlow 08 Oct '01

08 Oct '01

> Who would tell, please, what is the real difference between > :load and :add in ghci-5.02 ? There is a set of current target modules, call that set S. Then: :load M1..Mn ===> S := { M1..Mn } :add M1..Mn ===> S := S u { M1..Mn } In addition, :load unloads all currently loaded modules. Both commands initiate a reload starting from the target modules. Cheers, Simon

1 0

hdirect and ghc 5.02
by Sven Eric Panitz 08 Oct '01

08 Oct '01

Just a short question: is there a ghc 5.02 compatible distribution of hdirect available? Sven Eric -- ______________________________________________________________ Sven Eric Panitz Uhlandstr. 12 D-64297 Darmstadt Software AG sep(a)softwareag.com (+49)6151-92-1426 --- when I cannot sing my heart, I can only speak my mind ---

3 2

UniCode
by Krasimir Angelov 08 Oct '01

08 Oct '01

Why Char is 32 bit. UniCode characters is 16 bit. __________________________________________________ Do You Yahoo!? NEW from Yahoo! GeoCities - quick and easy web site hosting, just $8.95/month. http://geocities.yahoo.com/ps/info1

5 5

5.02 .deb?
by Ashley Yakeley 07 Oct '01

07 Oct '01

Is there a Debian package for GHC 5.02? -- Ashley Yakeley, Seattle WA

1 0

Re: UniCode
by Marcin 'Qrczak' Kowalczyk 06 Oct '01

06 Oct '01

Fri, 5 Oct 2001 23:23:50 +1000, Andrew J Bromage <andrew(a)bromage.org> pisze: > There is a set of one million (more correctly, 1M) Unicode characters > which are only accessible using surrogate pairs (i.e. two UTF-16 > codes). There are currently none of these codes assigned, This information is out of date. AFAIR about 40000 of them is assigned. Most for Chinese (current, not historic). > So rare, in fact, that the cost of strings taking up twice the > space that the currently do simply isn't worth the cost. In Haskell strings already have high overhead. In GHC a Char# value (inside Char object) always takes the same size as the pointer (32 or 64 bits), no matter how much of it is used. > It just goes to show that strings are not merely arrays of characters > like some languages would have you believe. In Haskell String = [Char]. It's true that Char values don't necessarily correspond to glyphs, but Strings are composed of Chars. -- __("< Marcin Kowalczyk * qrczak(a)knm.org.pl http://qrczak.ids.net.pl/ \__/ ^^ SYGNATURA ZASTÊPCZA QRCZAK

2 1