Gpc

gpc@gnu.de

5400 discussions

PPC
by f.c. kuechmann 24 May '00

24 May '00

Has anyone compiled gpc to run on PowerPC platforms under LinuxPPC? I'd hate to go to the trouble if somebody already did it. Cheers fk

1 0

Re: revised faq
by Frank Heckenbach 23 May '00

23 May '00

Russ Whitaker wrote: > Attached is a .gz file. It should look ok. Yep. > > (Note: The syntax given here is valid for GPC releases of May 2000 > > and newer.) > > > ok entered change > visited agnes earlier today - No May releases yet. I didn't claim there were any. ;-) (Which means that the syntax doesn't yet work with the existing releases, but since I expect Peter to make the next release soon, I think we can put this forward-looking information in the FAQ already.) Since there were no more comments (besides the ones by The Chief and mine which you've integrated), I'm putting this version of the FAQ into the GPC distribution now. In the future, you will be able to put updates into the distribution yourself via CVS. (Please contact Peter after he's announced the CVS server.) Frank -- Frank Heckenbach, frank(a)g-n-u.de, http://fjf.gnu.de/ GPC To-Do list, latest features, fixed bugs: http://agnes.dida.physik.uni-essen.de/~gnu-pascal/todo.html

1 0

Re: revised faq
by Russ Whitaker 23 May '00

23 May '00

Hi Andreas Jaeger writes> > >>>>> Russ Whitaker writes: > > Mandrake is NOT compatible with RedHat despite claims - you need less > > manual tweaks then in case with SuSE or Debian but ONLY RedHat is > Why do you need manual tweaks at all with SuSE? SuSE 6.4 ships with: > rpm -q -f /usr/bin/gpc > gpc-2.95.2-19 > (Version from October 1999) > Any given binary should install easily on the target it was intended for. Any other target you may have problems. For instance, SuSE 5.1, 5.2, 6.1, 6.2 did not have GPC, so you might run into problems trying to install the SuSE 6.4 package into a 5.1 box. Russ <russwhit(a)mind.net>

1 0

Re: revised faq
by Frank Heckenbach 23 May '00

23 May '00

Khimenko Victor wrote: > > {$ifdef DJGPP} > > var > > MinStackSize : Cardinal = $200000; asmname '_stklen'; > > {$endif} > > Hmm. If here we said about DJGPP's stack handling perhaps setrlimit(2) and > ulimit from bash(1) should be mentioned as well ? In SVr4 and BSD 4.3-based > systems (including Linux) you can raise stack size limit via setrlimit(2) in > run-time or from bash's command line BEFORE starting program with ulimit -s ... > Of course in Linux you'll need not `array[1..4000000] of byte' but > `array[1..10000000] of byte' to get exception (with default 8MiB limit) but > anyway ... That's one difference. The main difference, of course, is that a Linux user can set the stack size on the command line (and if s/he or the admin limits it, they might have their reasons, so a program probably should not usually override it) while under DJGPP the stack size has to be written into the executable (and the whole stack has to be allocated in physical memory at program startup due to a weakness of the DPMI specification), so one cannot just set it to some large value without problems. Finally, the method of setting the stack size is quite different (DJGPP: declare a constant(*) with a certain asmname; Unix: call a function at any time), so these are almost completely different matters, IMHO. Now, I wouldn't object to putting the information for Unix into the FAQ, but since the DJGPP information is in the DJGPP specific section, one would have to find a suitable place for it... (*) I wrote `var', but it should rather be `const'. Russ, can you please change it in the FAQ? Frank -- Frank Heckenbach, frank(a)g-n-u.de, http://fjf.gnu.de/ GPC To-Do list, latest features, fixed bugs: http://agnes.dida.physik.uni-essen.de/~gnu-pascal/todo.html

1 0

Re: posible bug question...
by Frank Heckenbach 23 May '00

23 May '00

Maurice Lombardi wrote: > Frank Heckenbach a Ã©crit : > > This would mean that the loop variable could practically not be used > > outside of the for loop. If this is really true, I'd really favour a > > syntax that makes this clear (like `for i : Integer = 1 to 10 do ...'), > > so i has only the corresponding scope. > > In fact this very syntax was used in one of the intermediate Wirth > languages between algol and pascal (we used at Grenoble an algolW > compiler implementing it at that time). It was done so exactly to forbid > any use of the for control variable outside of the loop, because in many > CPU for control variables are maintained in registers (CX in 16 bit 386 > CPU for example) and need not at any time to be affected into memory (at > least the for control variable of the innermost loop). > So a good idea was lost in going into EP. > But who may update the standard ? The commitees... ;-) But we could perhaps implement it in GPC (like we've already implemented a lot of other non-standard extensions)... Frank -- Frank Heckenbach, frank(a)g-n-u.de, http://fjf.gnu.de/ GPC To-Do list, latest features, fixed bugs: http://agnes.dida.physik.uni-essen.de/~gnu-pascal/todo.html

1 0

Re: revised faq
by Russ Whitaker 22 May '00

22 May '00

Hi Frank wrote: > Russ Whitaker wrote: > > > COPYING.LIB GNU Library Public License > > The precise name is "GNU Library General Public License". > ok > > http://www.cis.ohio-state.edu/hypertext/faq/usenet/x-faq/part6/faq-doc-2.htm > > l > > Bad line break. > Have 2 computers on my desk - the old one runs Windows 95 and it is used strictly for email. When I insert a .txt file it breaks lines where it wants to. Attached is a .gz file. It should look ok. > > Another way is to add the following code to your program to define a > minimum stack size (here: 2 MB). This value will be honored even if > a user sets a lower value by using stubedit, so this method might be > a little safer. > > (Note: The syntax given here is valid for GPC releases of May 2000 > and newer.) > ok entered change visited agnes earlier today - No May releases yet. Did find version 0.4 of the FAQ Russ <russwhit(a)mind.net>

1 0

Re: revised faq
by Frank Heckenbach 22 May '00

22 May '00

Russ Whitaker wrote: > hi > here's the monthly :-) revised faq > > > COPYING.LIB GNU Library Public License The precise name is "GNU Library General Public License". > http://www.cis.ohio-state.edu/hypertext/faq/usenet/x-faq/part6/faq-doc-2.htm > l Bad line break. > I'm accessing an `array[1..4000000] of byte' and I got an exception. > ==================================================================== > > Per default, the maximum stack size of a djgpp application is 256K. > If you need more, you have to adjust it with the stubedit program, i.e.: > > stubedit your_app.exe minstack=5000K Another way is to add the following code to your program to define a minimum stack size (here: 2 MB). This value will be honored even if a user sets a lower value by using stubedit, so this method might be a little safer. (Note: The syntax given here is valid for GPC releases of May 2000 and newer.) {$ifdef DJGPP} var MinStackSize : Cardinal = $200000; asmname '_stklen'; {$endif} Frank -- Frank Heckenbach, frank(a)g-n-u.de, http://fjf.gnu.de/ GPC To-Do list, latest features, fixed bugs: http://agnes.dida.physik.uni-essen.de/~gnu-pascal/todo.html

2 1

Re: posible bug question...
by J. David Bryan 22 May '00

22 May '00

Hello Marco, I have copied my reply to the list. On 21 May 2000, at 15:56, Marco van de Voort wrote: > Or do they mean: > > procedure x; > > var i : integer; > > procedure nestedx; > > begin > i:=5; > end; > > begin > for i:=1 to 5 do nestedx; > end; I believe you are correct, and my original reading that: i := 0; for i := 1 to 10 do ; was incorrect is itself incorrect. The key part of the statement in the standard is that the "...procedure-and-function-definition part of the block that closest-contains a for-statement..." cannot threaten the control variable. In your example, the "procedure x" block (the second begin-end block) is the "block that closest-contains a for-statement." Therefore, your first block, that of "procedure nestedx", forms the "procedure-and- function-definition-part of the block...." So your example breaks the requirement listed in the standard, meaning your interpretation is correct. In my example, the threat ("i := 0") is not in either the for-statement nor the procedure-and-function-definition part, so it is legal. -- Dave

1 0

Re: posible bug question...
by Russ Whitaker 21 May '00

21 May '00

Hi 1) p[h] defaults to zero 2) the value of h can exceed the array size 3) In BP 7, If you modify the index inside a for-loop the results are undefined After you exit the for-loop the index is undefined unless you exited with a goto in C, You can mdify the index inside the for-loop The value of the index is visable after you exit the loop (ref K&R 2nd Ed, p61) So what does the Pascal specs say? (I wish I had a copy of each) Russ <russwhit(a)mind.net> ---------- > From: Frank Heckenbach <frank(a)g-n-u.de> > To: gpc(a)gnu.de > Subject: Re: posible bug question... > Date: Friday, May 19, 2000 10:17 AM > > Richard Sharman wrote: > > > Frank Heckenbach writes: > > | Norberto Alfredo Bensa wrote: > > | The problem is the following: You use a for loop counter (h) after > > | the loop, and in Pascal this is undefined (this might be one of the > > | most despised features of Pascal, but it has its reasons). > > | Therefore, it's quite possible that one compiler does what you want > > | and another compiler (or another version of the same compiler) > > | doesn't. > > | > > > > Isn't it also wrong to attempt to modify a FOR loop index variable > > inside the body for the FOR statement? > > Yes, it is. I'd missed that one. (Using a while loop will avoid this > problem as well, of course.) > > > for h := 1 to j-1 do > > if 2*p[h]>i then h := j { <------- } > > else if i mod p[h]= 0 then break; > > > > I thought this was not allowed in Pascal, but I see gpc accepts it. > > I agree, GPC should check this more strictly (at least give a > warning). > > Frank > > -- > Frank Heckenbach, frank(a)g-n-u.de, http://fjf.gnu.de/ > GPC To-Do list, latest features, fixed bugs: > http://agnes.dida.physik.uni-essen.de/~gnu-pascal/todo.html >

2 1

Re: posible bug question...
by Frank Heckenbach 21 May '00

21 May '00

Russ Whitaker wrote: > 1) p[h] defaults to zero I don't think that's the problem since h < j and p [j] is initialized before each increment of j. > 2) the value of h can exceed the array size Only if more than 50 primes <= 100 are found which does not usually happen. ;-) > 3) In BP 7, > If you modify the index inside a for-loop the results > are undefined > After you exit the for-loop the index is undefined unless > you exited with a goto > > in C, > You can mdify the index inside the for-loop > The value of the index is visable after you exit the loop > (ref K&R 2nd Ed, p61) Yes, because there does not even have to be a loop counter in a C for loop (or there can be several counters, or anything). C for loops are really only syntactic sugar for while loops... > So what does the Pascal specs say? Like BP (in this case). > (I wish I had a copy of each) You can find them at ftp://ftp.digital.com/pub/DEC/Pascal/ (but you need a lot of patience to make sense of their formal language)... Frank -- Frank Heckenbach, frank(a)g-n-u.de, http://fjf.gnu.de/ GPC To-Do list, latest features, fixed bugs: http://agnes.dida.physik.uni-essen.de/~gnu-pascal/todo.html

1 0

Jump to page:

2025

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

2006

2005

2004

2003

2002

2001

2000

1999

1998

1997

1996

1995

Gpc