- Gpc - gnu.de

GPC with gcc-4.[23]
by Waldek Hebisch 18 Apr '13

18 Apr '13

I did some work on updating gpc to newer gcc versions. You can find it at GitHub: https://github.com/hebisch/gpc This is experimental work. 4.2 and 4.3 ports (on "master" Git branch) are basically in "works for my purpose" state: on my machine compiler builds and passes test suite with small (but nonzero) number of failures. On 4.5.1 there is initial port to gcc-4.[45] -- this one is broken from user point of view and may be of some use only for people who want to seriously work on the compiler. Note that this is _not_ a release. There are relatively easy to fix but non working things (in particular some features of Makefiles are broken). Testing was limited to running testsute just for one configuration. I estimate that fixing easily fixable stuff and polishing things would take me about a week -- unfortunately I am busy with other things. So I decided to make the code available "as is". If there will be bug reports, I will look at them and there is reasonable chance that I will fix them. But otherwise I do not plan to work on gpc in near future. If you want to try the code note that there are changes to build process, look at toplevel README file. -- Waldek Hebisch hebisch(a)math.uni.wroc.pl

4 15

Re: GPC compiler crash
by Garth Fletcher 19 Feb '13

19 Feb '13

Just to clarify the differences a bit. Space for a global variable is allocated when the program starts, essentially from the "heap" - which is the collection of available memory. Space for variables allocated by New() or NewPtr() are also allocated from the heap. These allocations can be released by using Dispose. For example: type myDataType = array... {defines the size of the structure} in the main program var myData : myDataType; {allocates a global structure in the heap} however, within a procedure or function procedure myProc; var localdata : myDataType; {allocates a LOCAL structure in the Stack} The reasoning is simple - the structure is local to that procedure or function, so the intent is to allocate it only as long as the procedure is executing. The Stack was a reasonable way to handle such local structures so that exiting the procedure, unstacks anything that was placed on it when the procedure is called and thus releases that allocated memory. This is OK only as long as the structures are small... The right way to handle local large structures is to explicitly allocate them upon entry to the procedure and then DEallocate them before exiting the procedure. For example: type myDataTypePtr : ^myDataType; procedure myProc; var localdataP : myDataTypePtr; {allocates a POINTER (4 or 8 bytes) on the stack} begin localdataP := NewPtr(sizeOf(myDataType)); {this allocates the structure on the heap, with localdataP} { pointing to that allocated space} {use the structure via localdataP^ } dispose(localdataP); {release the allocated memory} end; There is another gotcha - the difference between passing arguments "by reference" versus "by value". The former uses the VAR prefix to indicate that the procedure will have direct access to the actual data structure. The latter indicates that the called procedure will be given a copy of the original structure, i.e., a LOCAL copy. procedure myCalledProc (VAR byRef : myDataType; byVal : myDataType); "VAR byRef" specifies a POINTER to the original data to be passed on the Stack. The compiler takes care of converting any access to byRef within the procedure into a byRef^ and also of creating the pointer. "byVal" specifies a LOCAL COPY of the data structure to be passed on the Stack, which causes overflow if the copy is too big In other words, calling myCalledProc (myData, myData) creates a pointer to myData and passes that on the stack as byRef it also creates a complete copy of myData and passes that on the Stack as byVal. For big structures the difference is huge... I hope the above clarifies more than it confuses... -- Garth Fletcher

4 5

Re: GPC compiler crash
by Prof A Olowofoyeku (The African Chief) 19 Feb '13

19 Feb '13

Thomas Schneider wrote: >When I set it to 2003001 it compiles but gives a segmention fault even >though it's global. Maybe there's a limitation to allocating that >way? > >Then I switched dataprism to be a pointer using 'dataprism: >^trisquare;' as you suggested and changed the calls to be 'dataprism^' >(a rather cute trick!) and ... it compiles but still gives me a >segmentation fault! Actually, I was able to reproduce this crash. It happened at the point of "dispose (dataprism)". I changed the memory allocation/deallocation calls, and all works well (with "trimax=8006000"): procedure themain(var da: text); var dataprism: ^trisquare; (* the data structure for diana *) begin getmem (dataprism, sizeof (trisquare)); writeln(output,' trimdiana -gpc bug',version:4:2); writeprism(da, dataprism^); freemem (dataprism, sizeof (trisquare)); end; begin writeln ('sizeof trisquare=', sizeof (trisquare) div 1024 div 1024, 'mb'); themain(da); end. Best regards, The Chief -------- Prof. Abimbola A. Olowofoyeku (The African Chief) web: http://www.greatchief.plus.com/

3 5

GPC compiler crash
by Thomas Schneider 19 Feb '13

19 Feb '13

Folks: I discovered a program that crashes the GPC compiler, attached. Regards, Tom Thomas D. Schneider, Ph.D. Senior Investigator National Institutes of Health National Cancer Institute Frederick National Laboratory for Cancer Research Gene Regulation and Chromosome Biology Laboratory Molecular Information Theory Group Frederick, Maryland 21702-1201 schneidt(a)mail.nih.gov http://alum.mit.edu/www/toms

4 15

Re: Understanding {$I-} {$I+} ioResult
by Prof. A Olowofoyeku (The African Chief) 26 Nov '12

26 Nov '12

On 26 Nov 2012 at 14:59, Baudouin Le Charlier wrote: > > Le 26 nov. 2012 à 12:39, Prof. A Olowofoyeku (The African Chief) a > écrit : > > > On 26 Nov 2012 at 9:10, Baudouin Le Charlier wrote: > > > > [...] > >>> This is a historical factor, and there was no need to duplicate > >>> features that Borland already documented well, and which were well > >>> known to Borland Pascal/Delphi programmers. > >> > >> I am not sure they are so well known since you did not explain to > >> me why (i.e. by which miracle) a program that stops abruptly (when > >> compiled with $I-), stops normally if you add a call to ioResult > >> *after* the io statement that makes the original program fail. I > >> doubt very much that this could be explained in the original > >> Borland documentation and also that this behavior should be well > >> known of Borland Pascal programmers. > > > > As I said, the Borland Pascal/Delphi help files are freely available > > on the net. If you want to use Borland Pascal/Delphi compatibility > > features in GPC, you would do well to study the Borland > > documentation on those features. > > > > This is from a 20 year old Borland Turbo Pascal help file: > > > > [--- excerpt begins ---] > > $I: Input/Output-Checking Switch > > Enables or disables the automatic code generation that checks the > > result of a call to an I/O procedure > > > > Syntax: $I+ or $I- > > Default: $I+ > > Type: Local > > Command-Line: /$I+ or /$I- > > Menu Command: Options|Compiler|I/O Checking > > > > Remarks > > If an I/O procedure returns a non-zero I/O result when the $I switch > > is on, the program terminates, displaying a run-time error message. > > > > > > When the $I switch is off, you must use the IOResult function to > > check for I/O errors. [--- excerpt ends ---] > > > > Please let us stop this meaningless debate now. > > To me, it is not meaningless. I sent you an example where gnu pascal > does not conform to the excerpt you just sent me : the program is > compiled with the switch off but nevertheless it stops with an error > message. And where in that code did you use the IOResult function to check for errors? Best regards, The Chief ------- Prof. Abimbola Olowofoyeku (The African Chief) Web: http://www.greatchief.plus.com

2 1

Bus error when writing in a typed file
by Baudouin Le Charlier 26 Nov '12

26 Nov '12

Hi ! I am using a Mac OS X 10.6.8. I have downloaded GNU Pascal for Mac Os fromhttp://www.microbizz.nl/gpc.html. I have been using GNU Pascal for writing very simple programs for an introductory programming course and it worked fine until I wrote the following program for which I got a "bus error" when writing a first record in the file (at line 128). Notice that the program executes without errors when compiled with free pascal compiler. I should be happy if anyone can help me to solve this "problem". Baudouin Le Charlier

6 19

Re: running gpc on Mac OS X 10.7.4
by Thomas Schneider 06 Sep '12

06 Sep '12

Kevan: > I'm running GPC just fine on 10.7.4. I may have installed on an earlier > version of 10.7 and then upgraded. I installed the Mac OS X 10.6 Intel > version I found here: > > http://www.microbizz.nl/gpc.html > > What did you install? I hadn't installed, but I went to that page and installed from â¢ GNU Pascal 3.4.6u4 installer for Mac OS X 10.6 Intel native and PowerPC cross-compiler15.7 MB) It SAYS it's supposed to be ONLY for 10.6, but I went ahead anyway on 10.7. In /Developer/Pascal it installed cleanly and replaced directory gpc346u2 with gpc346u4. I can now compile: egg 18% cat h.p program helloworld (output); begin writeln(output,'hello world'); end. egg 19% gpc h.p ld: warning: -macosx_version_min not specified, assuming 10.7 ld: warning: PIE disabled. Absolute addressing (perhaps -mdynamic-no-pic) not allowed in code signed PIE, but used in __start from /Developer/SDKs/MacOSX10.5.sdk//usr/lib/crt1.o. To fix this warning, don't compile with -mdynamic-no-pic or link with -Wl,-no_pie egg 20% a.out hello world Thanks, Tom Thomas D. Schneider, Ph.D. Senior Investigator National Institutes of Health National Cancer Institute Frederick National Laboratory for Cancer Research Gene Regulation and Chromosome Biology Laboratory Molecular Information Theory Group Frederick, Maryland 21702-1201 http://alum.mit.edu/www/toms

2 1

collect2 no PATH override
by Jay Michael 03 Sep '12

03 Sep '12

I think "collect2" should honor all of the "where to find compiler components" switches fed to GPC. I think it would be helpful if GPC defaulted to acting as if it were run with "-B <directory containing gpc>" (unless the user provides a "-B" switch). The "collect2" in the binary distribution of gpc-2007-09-04 for Windows doesn't seem to make any effort to locate the "ld" it runs -- if there is no "ld" in the PATH, it prints ld-orig: No such file or directory I have been running GPC without putting GPC\bin in my PATH. It has been working. Then I discovered that when I installed Free Pascal, it took the liberty of updating my system-wide PATH. When I removed it from my PATH, running GPC\bin\gpc resulted in gpc: installation problem, cannot exec `as': No such file or directory Adding "-B GPC\bin" solves that, but has no effect on "collect2" finding "ld".

3 2

RangeCheckError from RuntimeError
by Jay Michael 03 Sep '12

03 Sep '12

My program was terminating with no runtime error message. When run in gdb, gdb told me it terminated with exit code 52 (I think octal). I had a Range Check Error. In gdb, I see that _p_RangeCheckError calls _p_RuntimeError. Then, _p_RuntimeError has code that might call _p_RangeCheckError. I didn't trace further -- gdb suddenly becomes unable to single-step, says "Unable to set breakpoint number -37". (The specific number changes if I keep trying to single-step. It's always negative.) I don't know how much better it would have been, but shouldn't _p_RuntimeError have been compiled with range checking suppressed? (I've found source for RangeCheckError and RuntimeError. I can't find source for CString2String -- is it an intrinsic function? It seems to have been inline coded in RuntimeError.)

3 2

MoveLeft / MoveRight special cases
by Jay Michael 03 Sep '12

03 Sep '12

The procedures MoveLeft and MoveRight start with a test for whether special case code will be useful/possible. procedure MoveLeft (const Source; var Dest; Count: SizeType); if (Count >= SizeOf (LongestCard)) and not (pDst - pSrc in [0 .. 2 * SizeOf (TWord) - 1]) then procedure MoveRight (const Source; var Dest; Count: SizeType); if (Count >= SizeOf (LongestCard)) and not (pSrc - pDst in [0 .. 2 * SizeOf (TWord) - 1]) then Procedure Move calls MoveRight when the address of the source is less than the source of the destination, so the difference will always be negative (except when source and destination are the same address), so the difference will never be in the range [0..2*SizeOf(TWord)-1]. Shouldn't those differences be reversed? Shouldn't the threshold be either SizeOf(LongestCard) in both clauses, or 2*SizeOf(TWord) in both clauses?

1 0