f



OS PL/1 Compiler Error?

Hi,

Anyone knows what is "IEL0935I U" means(layman terms:)?
I am trying to add a proc to a large program, and I got that compiler error
in OS PL/1 v2r3.

Thanks in advance
0
ospl1
9/30/2006 12:45:22 AM
comp.lang.pl1 1741 articles. 0 followers. Post Follow

17 Replies
820 Views

Similar Articles

[PageSpeed] 55

<ospl1> wrote in message
news:451dbe21$0$16375$c3e8da3@news.astraweb.com...
> Anyone knows what is "IEL0935I U" means(layman terms:)?
> I am trying to add a proc to a large program, and I got that compiler 
> error in OS PL/1 v2r3.

From the manual:
<quote>
IEL0935I U THE SIZE REQUIRED FOR ADDRESSING CONSTANTS IN STATIC STORAGE 
FOR THIS PROGRAM EXCEEDS 4095 BYTES.  COMPILATION TERMINATED.

Explanation: The compiler cannot generate correct addressing code if the
addressing constants at the start of static storage occupy more than
4095 bytes. The addressing constants in question consist of three for
each procedure, ON-unit, and begin-block, and one for each additional
entry point, plus address constants for resident library routines,
branching within the object program, and address constants to address
the remainder of static storage beyond 4K.

This message can be issued when the TEST compile-time option is 
specified, because PLITEST hooks and the PLITEST symbol table 
locator/descriptor initialization code increases the size of the program 
CSECT. This increases the number of addressing constants for branching 
within the program.

Programmer Response: Split the external procedure into two or more 
external procedures, or change some large static variables to CONTROLLED 
or AUTOMATIC.

If you received this message while using the TEST compile-time option,
you can reduce the program CSECT size by specifying a TEST suboption
that causes fewer PLITEST hooks to be generated. You can also split the
external procedure into two or more external procedures to eliminate
this problem.
</quote>

Yes, and what is it exactly that you can't understand about the 
"Programmer response" statement?

Robert
-- 
Robert AH Prins
prino at prino dot plus dot com 


0
prino6362 (7)
9/30/2006 10:32:03 AM
Hi Robert,

Thanks for your reply.
But is there a way to see a list(compiler listing?) of what are the stuff 
that makes the 4k limit. Splitting the program into 2 pieces is not viable 
right now.

Thanks again.


"Robert AH Prins" <prino@onetel.com> wrote in news:4o6rs0Fd78dcU1
@individual.net:

> <ospl1> wrote in message
> news:451dbe21$0$16375$c3e8da3@news.astraweb.com...
>> Anyone knows what is "IEL0935I U" means(layman terms:)?
>> I am trying to add a proc to a large program, and I got that compiler 
>> error in OS PL/1 v2r3.
> 
> From the manual:
> <quote>
> IEL0935I U THE SIZE REQUIRED FOR ADDRESSING CONSTANTS IN STATIC STORAGE 
> FOR THIS PROGRAM EXCEEDS 4095 BYTES.  COMPILATION TERMINATED.
> 
> Explanation: The compiler cannot generate correct addressing code if the
> addressing constants at the start of static storage occupy more than
> 4095 bytes. The addressing constants in question consist of three for
> each procedure, ON-unit, and begin-block, and one for each additional
> entry point, plus address constants for resident library routines,
> branching within the object program, and address constants to address
> the remainder of static storage beyond 4K.
> 
> This message can be issued when the TEST compile-time option is 
> specified, because PLITEST hooks and the PLITEST symbol table 
> locator/descriptor initialization code increases the size of the program 
> CSECT. This increases the number of addressing constants for branching 
> within the program.
> 
> Programmer Response: Split the external procedure into two or more 
> external procedures, or change some large static variables to CONTROLLED 
> or AUTOMATIC.
> 
> If you received this message while using the TEST compile-time option,
> you can reduce the program CSECT size by specifying a TEST suboption
> that causes fewer PLITEST hooks to be generated. You can also split the
> external procedure into two or more external procedures to eliminate
> this problem.
> </quote>
> 
> Yes, and what is it exactly that you can't understand about the 
> "Programmer response" statement?
> 
> Robert

0
ospl1
9/30/2006 7:00:29 PM
ospl1 wrote:
> Hi Robert,
> 
> Thanks for your reply.
> But is there a way to see a list(compiler listing?) of what are the stuff 
> that makes the 4k limit. Splitting the program into 2 pieces is not viable 
> right now.

No, because the whole point of a U message is that the compiler is dying 
half-way through. You /must/ split the program; it's too big for this 
particular compiler to handle, and that's that.

-- 
John W. Kennedy
"The blind rulers of Logres
Nourished the land on a fallacy of rational virtue."
   -- Charles Williams.  "Taliessin through Logres: Prelude"
0
jwkenne (1442)
10/1/2006 4:46:00 AM
Should I mention that another solution is to upgrade to a currently supported 
IBM mainframe PL/I compiler?  (When did that one go out of service?  About 5 
years ago?)

-- 
Bill Klein
 wmklein <at> ix.netcom.com
"John W. Kennedy" <jwkenne@attglobal.net> wrote in message 
news:qKHTg.76$NL1.45@newsfe10.lga...
> ospl1 wrote:
>> Hi Robert,
>>
>> Thanks for your reply.
>> But is there a way to see a list(compiler listing?) of what are the stuff 
>> that makes the 4k limit. Splitting the program into 2 pieces is not viable 
>> right now.
>
> No, because the whole point of a U message is that the compiler is dying 
> half-way through. You /must/ split the program; it's too big for this 
> particular compiler to handle, and that's that.
>
> -- 
> John W. Kennedy
> "The blind rulers of Logres
> Nourished the land on a fallacy of rational virtue."
>   -- Charles Williams.  "Taliessin through Logres: Prelude" 


0
wmklein (2605)
10/1/2006 3:43:32 PM
"John W. Kennedy" <jwkenne@attglobal.net> wrote in message
news:qKHTg.76$NL1.45@newsfe10.lga...
> ospl1 wrote:
> > Hi Robert,
> >
> > Thanks for your reply.
> > But is there a way to see a list(compiler listing?) of what are the stuff
> > that makes the 4k limit. Splitting the program into 2 pieces is not viable
> > right now.
>
> No, because the whole point of a U message is that the compiler is dying
> half-way through. You /must/ split the program; it's too big for this
> particular compiler to handle, and that's that.

Please read the message.
    It says that you can reduce the requirements by
making constants automatic.
    You could also do this by using INITIAL CALL and doing the
initialization there.
    You could ATTACH the new procedure, and/or some
of the existing ones.


0
robin_v (2737)
10/2/2006 1:07:01 PM
<ospl1> wrote in message news:451ebecc$0$16352$c3e8da3@news.astraweb.com...
> Hi Robert,
>
> Thanks for your reply.
> But is there a way to see a list(compiler listing?) of what are the stuff
> that makes the 4k limit.

Please do read the message in the programmer response.
It tells you exactly.

> Splitting the program into 2 pieces is not viable
> right now.
>
> Thanks again.
>
> "Robert AH Prins" <prino@onetel.com> wrote in news:4o6rs0Fd78dcU1
> @individual.net:
>
> > <ospl1> wrote in message
> > news:451dbe21$0$16375$c3e8da3@news.astraweb.com...
> >> Anyone knows what is "IEL0935I U" means(layman terms:)?
> >> I am trying to add a proc to a large program, and I got that compiler
> >> error in OS PL/1 v2r3.
> >
> > From the manual:
> > <quote>
> > IEL0935I U THE SIZE REQUIRED FOR ADDRESSING CONSTANTS IN STATIC STORAGE
> > FOR THIS PROGRAM EXCEEDS 4095 BYTES.  COMPILATION TERMINATED.
> >
> > Explanation: The compiler cannot generate correct addressing code if the
> > addressing constants at the start of static storage occupy more than
> > 4095 bytes. The addressing constants in question consist of three for
> > each procedure, ON-unit, and begin-block, and one for each additional
> > entry point, plus address constants for resident library routines,
> > branching within the object program, and address constants to address
> > the remainder of static storage beyond 4K.
> >
> > This message can be issued when the TEST compile-time option is
> > specified, because PLITEST hooks and the PLITEST symbol table
> > locator/descriptor initialization code increases the size of the program
> > CSECT. This increases the number of addressing constants for branching
> > within the program.
> >
> > Programmer Response: Split the external procedure into two or more
> > external procedures, or change some large static variables to CONTROLLED
> > or AUTOMATIC.
> >
> > If you received this message while using the TEST compile-time option,
> > you can reduce the program CSECT size by specifying a TEST suboption
> > that causes fewer PLITEST hooks to be generated. You can also split the
> > external procedure into two or more external procedures to eliminate
> > this problem.
> > </quote>
> >
> > Yes, and what is it exactly that you can't understand about the
> > "Programmer response" statement?
> >
> > Robert


0
robin_v (2737)
10/2/2006 1:07:01 PM
robin wrote:
> "John W. Kennedy" <jwkenne@attglobal.net> wrote in message
> news:qKHTg.76$NL1.45@newsfe10.lga...
> > ospl1 wrote:
> > > Hi Robert,
> > >
> > > Thanks for your reply.
> > > But is there a way to see a list(compiler listing?) of what are the stuff
> > > that makes the 4k limit. Splitting the program into 2 pieces is not viable
> > > right now.
> >
> > No, because the whole point of a U message is that the compiler is dying
> > half-way through. You /must/ split the program; it's too big for this
> > particular compiler to handle, and that's that.
>
> Please read the message.
>     It says that you can reduce the requirements by
> making constants automatic.
>     You could also do this by using INITIAL CALL and doing the
> initialization there.

Actually, if you had properly read the original posting and quote from
the manual, you should have realised that adding yet another proc is an
absolute no-no in this case!

>     You could ATTACH the new procedure, and/or some
> of the existing ones.

Robert

0
prino (21)
10/3/2006 6:59:39 PM
"prino" <prino@bigfoot.com> wrote in message
news:1159901979.348669.186870@k70g2000cwa.googlegroups.com...
> robin wrote:
> > "John W. Kennedy" <jwkenne@attglobal.net> wrote in message
> > news:qKHTg.76$NL1.45@newsfe10.lga...
> > > ospl1 wrote:
> > > > Hi Robert,
> > > >
> > > > Thanks for your reply.
> > > > But is there a way to see a list(compiler listing?) of what are the
stuff
> > > > that makes the 4k limit. Splitting the program into 2 pieces is not
viable
> > > > right now.
> > >
> > > No, because the whole point of a U message is that the compiler is dying
> > > half-way through. You /must/ split the program; it's too big for this
> > > particular compiler to handle, and that's that.
> >
> > Please read the message.
> >     It says that you can reduce the requirements by
> > making constants automatic.
> >     You could also do this by using INITIAL CALL and doing the
> > initialization there.
>
> Actually, if you had properly read the original posting and quote from
> the manual, you should have realised that adding yet another proc is an
> absolute no-no in this case!

Actually, if you had properly read the second (the follow-on) posting,
you would have read:

"Programmer Response: Split the external procedure into
" two or more external procedures, or change some large
" static variables to CONTROLLED or AUTOMATIC."

and had you done so, you would have realized that
changing some large static variables to CTL of AUTO
are appropriate remedies.
    I suggested converting static initializations to
automatic in a separate procedure.

I also suggested making some procedures ATTACH-able.

> >     You could ATTACH the new procedure, and/or some
> > of the existing ones.
>
> Robert


0
robin_v (2737)
10/5/2006 8:01:38 AM
On Thu, 05 Oct 2006 01:01:38 -0700, robin <robin_v@bigpond.com> wrote:

> "prino" <prino@bigfoot.com> wrote in message
> news:1159901979.348669.186870@k70g2000cwa.googlegroups.com...
>> robin wrote:
>> > "John W. Kennedy" <jwkenne@attglobal.net> wrote in message
>> > news:qKHTg.76$NL1.45@newsfe10.lga...
>> > > ospl1 wrote:
>> > > > Hi Robert,
>> > > >
>> > > > Thanks for your reply.
>> > > > But is there a way to see a list(compiler listing?) of what are  
>> the
> stuff
>> > > > that makes the 4k limit. Splitting the program into 2 pieces is  
>> not
> viable
>> > > > right now.
>> > >
>> > > No, because the whole point of a U message is that the compiler is  
>> dying
>> > > half-way through. You /must/ split the program; it's too big for  
>> this
>> > > particular compiler to handle, and that's that.
>> >
>> > Please read the message.
>> >     It says that you can reduce the requirements by
>> > making constants automatic.
>> >     You could also do this by using INITIAL CALL and doing the
>> > initialization there.
>>
>> Actually, if you had properly read the original posting and quote from
>> the manual, you should have realised that adding yet another proc is an
>> absolute no-no in this case!
>
> Actually, if you had properly read the second (the follow-on) posting,
> you would have read:
>
> "Programmer Response: Split the external procedure into
> " two or more external procedures, or change some large
> " static variables to CONTROLLED or AUTOMATIC."
>
> and had you done so, you would have realized that
> changing some large static variables to CTL of AUTO
> are appropriate remedies.
>     I suggested converting static initializations to
> automatic in a separate procedure.

It may not be significant in this case, but statics are only initialized
once, when the program is compiled.  Automatics are initialized everytime
the program is run.

>
> I also suggested making some procedures ATTACH-able.
>
>> >     You could ATTACH the new procedure, and/or some
>> > of the existing ones.
>>
>> Robert
>
>



-- 
Using Opera's revolutionary e-mail client: http://www.opera.com/mail/
0
tom294 (608)
10/5/2006 12:21:54 PM
Tom Linden wrote:
> 
> It may not be significant in this case, but statics are only initialized
> once, when the program is compiled.  Automatics are initialized everytime
> the program is run.
> 
This statement is inaccurate.  Static variables are initialized when the program 
is loaded - most by simply loading constant data generated by the compiler but 
some by the relocating loader.  For example,

dcl x bin float static init(100.25), p ptr static init((addr(s))), addr builtin;

The floating point representation of 100.25 can be generated by the compiler but 
the address of x is not known until load time or at least link time depending on 
the addressing architecture of the machine.

Moreover, automatics are initialized each time the block in which they occur is 
entered.
0
jjw (608)
10/5/2006 8:33:18 PM
James J. Weinkam <jjw@cs.sfu.ca> wrote:
> Tom Linden wrote:
 
>> It may not be significant in this case, but statics are only initialized
>> once, when the program is compiled.  Automatics are initialized everytime
>> the program is run.
 
> This statement is inaccurate.  Static variables are initialized when 
> the program is loaded - most by simply loading constant data generated 
> by the compiler but some by the relocating loader.  For example,

For a suitably generalized meaning of program, I suppose it
is fine.  If, for example, you consider a procedure (in the
absence of BEGIN) a program, or even a BEGIN block as a program.
 
> dcl x bin float static init(100.25), p ptr static init((addr(s))), 
> addr builtin;
 
> The floating point representation of 100.25 can be generated by the 
> compiler but the address of x is not known until load time or at least 
> link time depending on the addressing architecture of the machine.

On most systems, the address as known is inserted, and relocated
at link and load time.  If, for example, you made it:

dcl 1 x,
    2 x bin float static init(100.25), 
    2 p ptr static init((addr(s)));

The value compiled after 100.25 may not be the final address,
but it will likely be related to the actual address.

-- glen
0
gah1 (524)
10/5/2006 9:52:11 PM
"Tom Linden" <tom@kednos-remove.com> wrote in message
news:op.tgx7aspwtte90l@hyrrokkin...
> On Thu, 05 Oct 2006 01:01:38 -0700, robin <robin_v@bigpond.com> wrote:
>
> > "prino" <prino@bigfoot.com> wrote in message
> > news:1159901979.348669.186870@k70g2000cwa.googlegroups.com...
> >> robin wrote:
> >> > "John W. Kennedy" <jwkenne@attglobal.net> wrote in message
> >> > news:qKHTg.76$NL1.45@newsfe10.lga...
> >> > Please read the message.
> >> >     It says that you can reduce the requirements by
> >> > making constants automatic.
> >> >     You could also do this by using INITIAL CALL and doing the
> >> > initialization there.
> >>
> >> Actually, if you had properly read the original posting and quote from
> >> the manual, you should have realised that adding yet another proc is an
> >> absolute no-no in this case!
> >
> > Actually, if you had properly read the second (the follow-on) posting,
> > you would have read:
> >
> > "Programmer Response: Split the external procedure into
> > " two or more external procedures, or change some large
> > " static variables to CONTROLLED or AUTOMATIC."
> >
> > and had you done so, you would have realized that
> > changing some large static variables to CTL of AUTO
> > are appropriate remedies.
> >     I suggested converting static initializations to
> > automatic in a separate procedure.
>
> It may not be significant in this case, but statics are only initialized
> once, when the program is compiled.  Automatics are initialized everytime
> the program is run.

That's true, but there are ways of ensuring that the
data is initialized once only.
Anyway, it's up to OP to choose the method that best suits.


0
robin_v (2737)
10/8/2006 2:24:18 PM
William M. Klein wrote:
> Should I mention that another solution is to upgrade to a currently supported 
> IBM mainframe PL/I compiler?  (When did that one go out of service?  About 5 
> years ago?)
> 
Well, close; only 3 years ago...
Here's the 18+ year 'history' of OS PL/I Version 2:
> =====================================================================
>            OS PL/I Version 2
> Program                                Marketing  Service      Replaced
> Number     VRM   Announced  Available  Withdrawn  Discontinued   By
> --------- ------ ---------- ---------- ---------- ----------   --------
>  5668-909 2.03.0 1990/06/26 1990/07/27 2000/09/06 2003/12/31   5655-H31
>  5668-909 2.02.1 1989/06/20 1989/06/30 2000/09/06 1991/12/31   5655-H31
>  5668-909 2.01.1 1989/06/20 1989/06/30 2000/09/06 1990/12/31   5655-H31
>  5668-909 2.01.0 1987/05/19 1988/03/04 2000/09/06 1990/12/31   5655-H31
> =====================================================================
And, the listed replacement, Enterprise PL/I, is now a year into the 
life of the current release, Version 3 Release 5.

Carl
0
10/9/2006 3:39:18 AM
We are running into the same problem with a large pl/I program.

With a compile option of LIST,MAP for the last good compile of your
program you should see a static storage map that lists the PROGRAM
ACONS. See the Disappearing Static Storage post.

STATIC INTERNAL STORAGE MAP
000000  E0000074              PROGRAM ADCON
000004  00000008              PROGRAM ADCON
000008  00000096              PROGRAM ADCON
00000C  000000B8              PROGRAM ADCON <-Entry
000010  0000010E              PROGRAM ADCON <-Exit
000014  00000128              PROGRAM ADCON <-Entry
000018  0000017E              PROGRAM ADCON <-Exit
00001C  00000198              PROGRAM ADCON <-Entry
000020  000001EE              PROGRAM ADCON <-Exit
000024  00000208              PROGRAM ADCON <-Entry
000028  0000025E              PROGRAM ADCON <-Exit
00002C  0000025E              PROGRAM ADCON <-Repeating Why?
000030  0000025E              PROGRAM ADCON <-Repeating Why?
000034  0000025E              PROGRAM ADCON <-Repeating Why?
000038  0000025E              PROGRAM ADCON <-Repeating Why?
00003C  0000025E              PROGRAM ADCON <-Repeating Why?
000040  0000025E              PROGRAM ADCON <-Repeating Why?

This is static map from four procedures. The last 6 program adcons are
repeating adcons for reasons unknown. There are two adcons for each
procedure entry and exit.  The last adcon exit repeats for the number
of procedures + 2. This is the big mystery.

Do your PROGRAM ADCONs repeat on your last good compile of the program?
If so how many repeating PROGRAM ADCONs do you have?

I have seen this with the pl/I compilers on two different mainframes
z/VM and z/OS.

Could this be the cause of you exceeding the 4095 address constant
limit?

0
Fred.Horn (6)
10/17/2006 2:31:54 AM
Fred.Horn@kable.com wrote:
> We are running into the same problem with a large pl/I program.
 
> With a compile option of LIST,MAP for the last good compile of your
> program you should see a static storage map that lists the PROGRAM
> ACONS. See the Disappearing Static Storage post.
 
> STATIC INTERNAL STORAGE MAP
> 000000  E0000074              PROGRAM ADCON
> 000004  00000008              PROGRAM ADCON
> 000008  00000096              PROGRAM ADCON
> 00000C  000000B8              PROGRAM ADCON <-Entry
> 000010  0000010E              PROGRAM ADCON <-Exit

How many ENTRY statements do you have?  

PL/I ENTRY/EXIT code can be pretty complicated.  Among
others, PL/I will convert the return values used in
any RETURN statement to any of the types required
by the main or any ENTRY points.  The compiler doesn't
try to figure out the path between each ENTRY and RETURN,
so generates all the conversion code for any combination.

I do like that better than Fortran, which EQUIVALENCEs
all the function return variables.

-- glen
0
gah1 (524)
10/17/2006 6:01:51 PM
Glen,

More info on this issue. Below is the program with no real executable
code, just empty procedures 001 to 254. You can use any number of procs
254 or less. The compile options must be LIST(1),MAP to get the static
storage map.

    CALL PROC001;
    PROC001: PROC;
END PROC001;
.................
    PROC254: PROC;
END PROC254;


STATIC INTERNAL STORAGE MAP


000000  E0000C44              PROGRAM ADCON
............................................
0007F4  00006F68              PROGRAM ADCON
0007F8  00006FBE              PROGRAM ADCON <- Procedure PROC254 exit
0007FC  00006FBE              PROGRAM ADCON <- first repeating adcon
............................................
000C10  00006FBE              PROGRAM ADCON <- last repeating adcon


The program above with 254 empty procedures totaled 773 PROGRAM ADCONS.

There were a total of 262 repeating adcons with the same exit address
of the last procedure '00006FBE'. 33.8% of static storage used due to
repeating adcons. Seems like a waste of static storage to me.

We have a call in to IBM to get an answer on this issue.

Is anyone else seeing the same thing with their pl/I compilers?

0
Fred.Horn (6)
10/17/2006 10:50:56 PM
Fred.Horn@kable.com wrote:
> More info on this issue. Below is the program with no real executable
> code, just empty procedures 001 to 254. You can use any number of procs
> 254 or less. The compile options must be LIST(1),MAP to get the static
> storage map.
(snip) 
> The program above with 254 empty procedures totaled 773 PROGRAM ADCONS.
(snip) 
> There were a total of 262 repeating adcons with the same exit address
> of the last procedure '00006FBE'. 

I might say that it was saving memory by not repeating the
prologue/epilogue code for each, but reusing one copy.

Actually, I don't know the details yet, but that is a guess
why they are all the same.  I know Fortran compilers that
generate prologue/epilogue code for each entry point,
even if the code is the same.

-- glen
0
gah1 (524)
10/17/2006 11:24:57 PM
Reply:

Similar Artilces:

Error compiling 1.9.1 on Mac OS X
When making ruby-1.9.1-p0 on Mac OS X (10.5.6), I get the following error: cc -dynamic -bundle -undefined suppress -flat_namespace -o .ext/i386- darwin9/enc/trans/transdb.bundle enc/trans/transdb.o -L. -L. -L. -L/ opt/local/lib -L/usr/local/lib -lruby1.9 -lpthread -ldl -lobjc ld: in /usr/local/lib/libstdc++.6.dylib, file is not of required architecture collect2: ld returned 1 exit status make[1]: *** [.ext/i386-darwin9/enc/trans/transdb.bundle] Error 1 make: *** [trans] Error 2 Ideas? ...

PL/1 Version 1.1 and z/OS support
Hi guys, Can anyone tell me please if z/OS version 1.7 still supports PL/1 ver 1.1? Thanks, Ron. ronsho.shoshani@gmail.com wrote: > Hi guys, > > Can anyone tell me please if z/OS version 1.7 still supports PL/1 ver > 1.1? > I would expect it would probably *run*, baring LE incompatibalities (score one for static libraries). Hasn't EOS been announced for 1.1? Most of the incompatibilities with Enterprise PL/I have now been corrected. ronsho.shoshani@gmail.com wrote: > Hi guys, > > Can anyone tell me please if z/OS version 1.7 still supports PL/1 ver > 1.1? > > > Thanks, > Ron. > I have used it in vendor software running in a number of shops - some quite large - and no reports of problems so far. I would expect programs compiled in it to run correctly for quite a few years to come. So, in summary: No worries! PL/I for MVS and VM still works great. Ah yes. "The last of the mainframe compilers...". The compilers written in PL/I (VisualAge and later) cannot produce a load module which can (correctly) be marked as RENT. Not a big issue for the programmer I suppose, but I do like that subpool 252 storage protection. (Actually, in my current job, I'm using OS PL/I V2R3 (with PL/I transient library ahead of the LE runtime library) on OS/390 2.8. PL/I for MVS and VM would be luxury. <sigh> DB2 V5. IMS V6. Gotta love it. <g>) Cheers, Greg ...

Attach and PL/1 using Enterprise PL/1 Compiler V 3.3
Hallo, does anybody have a simple PL/1 sample Program which does subtasking using ATTACH-statement or using POSIX-calls? If you have, please send it to me. Thanks Gottfried ...

Error Compile 1.8.0 and 1.8.1-pre2
I'm trying to compile they latest. But i'm getting a SEGMENTATION FAULT error. It happens on miniruby. I found a post that said to add --enable-shared. So I did and this got me to the next file libruby.so with an error saying 'Unkown nod type 0'. I tried it on 1.8.1 and got the same segementation fault on miniruby. --enabled-shared didn't help. I'll keep searching and hacking but I thought I'd put this out these to see if anyone has any ideas. thanks, -t0 p.s. I'm doing this on GoboLinux. So maybe the alternate file hiearchry is the problem, but I doubt I w...

Re: Error Compile 1.8.0 and 1.8.1-pre2
> Could you show stack trace? Sorry, I don't know much about compiling programs. How do I do what you are asking? -t0 Hi, At Wed, 12 Nov 2003 09:23:55 +0900, T. Onoma wrote: > > Could you show stack trace? > > Sorry, I don't know much about compiling programs. How do I do what you are asking? Run miniruby from gdb or run "gdb miniruby core" after segfault, and type "where" (without quotes, with return). -- Nobu Nakada Hi, At Wed, 12 Nov 2003 10:06:43 +0900, nobu.nokada@softhome.net wrote: > Run miniruby from gdb or run "gdb minirub...

ruby version 1.9.2 dev with memcached-0.1.18 native compilation error/syntax errors #2
Hi there Hopefully these errors somehow relate to REE, if not, highly appreciate any feedback and a direction! Installed latest (from git master) enterprise server and attempting to build memcached-0.18.0 with ruby compiler version 1.9.2 dev , encountered 2 exceptions (one during native compilation of memcached gem and second during ruby compilation of memcached rails.rb (syntax error on the usage of alias) Symptom 1: errors during gem execution of memcache gem Symptom 2: errors during ruby compilation of memcached .rb file Using environment: A) Ruby Server deployment using ruby enterp...

compile error re: Open Office 1.1.2 -> com_sun_star_lib_connections_pipe_PipeConnection.c: 558 syntax error before "void"
Anyone solved this ? Thank you, Courtney ...

Re: Error Compile 1.8.0 and 1.8.1-pre2 #3
took me a bit to get back to... > Check `ulimit -c'. returns: 0 > > the error specifically says: 'mkmf.rb:354: [BUG] Segmentation > > Fault' if that helps. > > Hmmm, only "false" line? > What's your configure options? ./configure --prefix=/Programs/Ruby/1.8.1 \ --exec-prefix=/Programs/Ruby/1.8.1 \ --enable-shared \ --with-sitedir=/Programs/Ruby/1.8.1/lib/site_ruby \ --localstatedir=/Programs/Ruby/Variable \ --sysconfdir=/Programs/Ruby/Settings like i said, Gobo has non-standard ...

Compiler 4.1.1 Error: perl not recognized
Hi everyone, I just installed the Matlab Compiler 4.1.1 (Service Pack 1+) (which is not included in the CD distribution of SP1, must download from Mathworks homepage). Now, everytime I try to compile (mcc -m test.m), I get this error: 'perl' is not recognized as an internal or external command, operable program or batch file. And no .exe-file is created... With Compiler 4.1 I do not have this problem, I do everything else the same... Best Regards, Stuart I'm having the same error using the latest Compiler in order to create a DLL with the comtool. Nick, Stuart Bonasona wrot...

krb5-1.9.1 compilation errors on NetBSD
Hi all, I get a fatal error when I try to build krb5-1.9.1 on NetBSD 5.1 (amd64). /home/iulius/autobuild/bin/gcc-core-4.6.1/bin/gcc -fPIC -DSHARED -DLIBDIR=\"/home/iulius/autobuild/bin/krb5-1.9.1/lib\" -I../../../include -I../../../include -I./../os -I../../.. -DKRB5_DEPRECATED=1 -g -O2 -Wall -Wcast-align -Wshadow -Wmissing-prototypes -pedantic -Wno-format-zero-length -Woverflow -Wstrict-overflow -Wmissing-format-attribute -Wmissing-prototypes -Wreturn-type -Wmissing-braces -Wparentheses -Wswitch -Wunused-function -Wunused-label -Wunused-variable -Wunused-value -Wunknown-pragmas -Wsign-compare -Werror=declaration-after-statement -Werror=variadic-macros -Werror-implicit-function-declaration -pthread -c init_ctx.c -o init_ctx.so.o && mv -f init_ctx.so.o init_ctx.so In file included from ../../../include/k5-int.h:132:0, from init_ctx.c:56: init_ctx.c: In function 'init_common': init_ctx.c:134:5: error: implicit declaration of function 'krb5int_initialize_library' [-Werror=implicit-function-declaration] cc1: some warnings being treated as errors I know I can remove the "-Werror-implicit-function-declaration" flag, but it is not the default configure options. Also, I noticed other few (non-blocking) warnings: .../../include/k5-platform.h: In function 'store_64_be': .../../include/k5-platform.h:594:1: warning: use of C99 long long integer constant [-Wlong-long] .../../include/k5-platform.h: In function ...

compiler 4.1.1, execution-time error
Hi! I have just upgraded both my malab (to r14sp1) and my compiler (4.1.1). Updating matlab didn't give as much problems as I feared (just my 'pause(0)' which needed to be replaced by 'drawnow'). Great! Now with the stand-alone version I just created: - my GUI window does not appear at the same position on the screen as in normal mode (ie non-stand-alone). Someone who experimented the same problem and found a solution? Almost my whole window is off-screen! - loading a neural network that was saved in matlab does not work, it is not recognised (Warning at run-time: Class...

Netbeans compiler always error with error code 1.
Hi All, After re-install netbeans IDE with wireless toolkit, my IDE show a problem when I compiler java code every time, ".........\nbproject\build-impl.xml:228: Execution failed with error code 1." Could you please help? I re-install so many times, it still happen, if I run it in WTK directly, it will success, why ... Best regards, Boki. boki wrote: > Hi All, > After re-install netbeans IDE with wireless toolkit, my IDE > show a problem when I compiler java code every time, > > > ".........\nbproject\build-impl.xml:228: Execution failed ...

Cross-compiling error when compiling 2.6.1...
I come across the following error: checking for chflags... configure: error: cannot run test program while cross compiling See `config.log' for more details. make-3.81[1]: *** [/nobackup/garrcoop/python_upgrade/contrib/python/obj-mips32/Makefile] Error 1 make-3.81[1]: Leaving directory `/nobackup/garrcoop/python_upgrade' make-3.81: *** [all] Error 2 The blurb from configure that does this test says: # On Tru64, chflags seems to be present, but calling it will # exit Python Basically this testcase (while valid for Tru64) doesn't work with many cross-compilation environments. Cou...

Compiling 8.13.4 with GCC 3.3--error, stdarg.h v. varargs.h on Mac OS X 10.4.1?
I'm trying to compile Sendmail 8.13.4 on Mac OS X 10.4.1, using GCC 3.3. I'm getting the message that GCC no longer supports varargs.h and code should be updated to use stdarg.h. I know this is a long-standing problem, and I used to be able to solve it (with Sendmail 8.12.x on Mac OS X 10.2.x and 10.3.x) by reverting to GCC 3.1. Unfortunately, GCC 3.1 is not installed on Mac OS X 10.4. The weird thing is that the last machine I build this way was Sendmail 8.13.3 on Mac OS X 10.3.7 and it compiled just fine with GCC 3.3. Possibly this had something to do with the fact that GCC 3...

My executable compiled with Real Time Engine 5.1.1 in App Builder gives error when trying to run in App Builder with RTE 6.1
Hello.&nbsp; I built an executable using App Builder in LV 5.1.1 and now, years later, I am trying to run it in LV 6.1 but I get an error stating a requirement for&nbsp;Run Time Engine version 5.1.1.&nbsp; To be able to run this, must I install&nbsp;the older version of Run Time Engine?&nbsp; If so, where can I find the older version?&nbsp; Do I have to install the older version of LabVIEW to get the RTE?&nbsp; Will having both versions on my pc be a problem? &nbsp; I assume that once I am able to run the exe, that I will be able to recompile with the newer vers...

Who can help us getting SciLab started? Console Error Messages when attempting to start scilab-5.1.1 on Mac OS X 10.6.1
These are the Console Error Messages when attempting to start scilab-5.1.1 on Mac OS X 10.6.1 by double clicking the icon. Who can help us getting SciLab started? ZK 11/8/09 1:59:06 AM [0x0-0xd60d6].org.scilab.modules.jvm.Scilab[2494] dyld: Symbol not found: __ZNSo9_M_insertIlEERSoT_ 11/8/09 1:59:06 AM [0x0-0xd60d6].org.scilab.modules.jvm.Scilab[2494] Referenced from: /System/Library/Frameworks/CoreServices.framework/Versions/A/Frameworks/LaunchServices.framework/Versions/A/LaunchServices 11/8/09 1:59:06 AM [0x0-0xd60d6].org.scilab.modules.jvm.Scilab[2494] Expected in: /Applications/scilab-5.1.1-alpha-1.app/Contents/MacOS/share/scilab/../../lib/thirdparty//libstdc++.6.dylib 11/8/09 1:59:06 AM [0x0-0xd60d6].org.scilab.modules.jvm.Scilab[2494] in /System/Library/Frameworks/CoreServices.framework/Versions/A/Frameworks/LaunchServices.framework/Versions/A/LaunchServices 11/8/09 1:59:09 AM [0x0-0xd60d6].org.scilab.modules.jvm.Scilab[2494] /Applications/scilab-5.1.1-alpha-1.app/Contents/MacOS/bin/scilab: line 313: 2505 Trace/BPT trap "$SCILABBIN" "$@" 11/8/09 1:59:09 AM com.apple.launchd.peruser.501[112] ([0x0-0xd60d6].org.scilab.modules.jvm.Scilab[2494]) Exited with exit code: 133 11/8/09 1:59:09 AM ReportCrash[2507] Saved crash report for scilab-bin[2505] version ??? (???) to /Users/heinz/Library/Logs/DiagnosticReports/scilab-bin_2009-11-08-015909_iMacDuo.crash ...

Compile error on AIX5.1 with ACE5.2.1 use REPORT_FORM
ACE VERSION: 5.2.1 HOST MACHINE and OPERATING SYSTEM: AIX myhost 1 5 0005C39A4C00 TARGET MACHINE and OPERATING SYSTEM, if different from HOST: Same as before COMPILER NAME AND VERSION (AND PATCHLEVEL): xlC information: xlC.adt.include 6.0.0.1 COMMITTED C Set ++ Application xlC.aix50.rte 6.0.0.9 COMMITTED C Set ++ Runtime for AIX 5.0 xlC.cpp 5.0.2.0 COMMITTED C for AIX Preprocessor xlC.msg.en_US.cpp 5.0.2.0 COMMITTED C for AIX Preprocessor xlC.msg.en_US.rte 6.0.0.0 COMMITTED C Set ++ Runtime xlC.rte 6.0.0.0 COMMITTED C Set ++ Runtime DOES THE PROBLEM AFFECT: COMPILATION? If so, what do your $ACE_ROOT/ace/config.h and $ACE_ROOT/include/makeinclude/platform_macros.GNU contain? LINKING? config.h -> config-aix5.1.h more config.h // config-aix5.1.h,v 1.1 2001/09/05 15:51:55 shuston Exp // // Config file for AIX 5.1 #include "ace/config-aix-4.x.h" platform_macros.GNU -> platform_aix_ibm.GNU On Unix systems, did you run make realclean first? I don't know, this is compiled by other guys. SYNOPSIS: [Brief description of the problem] When use fellow command to compile my program,I will got link error while I can success do this on another machine with same library. My code is: ACE_Reactor reactor; reactor.run_reactor_event_loop(); Compil...

handling PL/SQL compile errors, 'show errors' as procedure
Q: Which table and timestamp column is used for 'show errors' definition of 'the previously defined object'? Longer description: Concider you have the schema definition for a larger schema with tables and PL/SQL objects in a text file. Everything works for the current version but you want some error control for new pices of code. So, you want SQLPLUS to abort when the first defective object definition is found and report an exist status unequal to zero in case of problems. I tried to catch compilation errors with WHENEVER SQLERROR EXIT FAILURE and noticed sqlplus exits for errors in SQL code, but errors in PL/SQL code will pass unnoticed. Sqlplus will report the errors for the last compiled object from USER_ERRORS with 'show errors', but it will not abort the script. WHENEVER SQLERROR does not catch this type of errors. Google found an article from T.Kyte with a PLS/SQL procedure which addresses this problem. The procedure raises an exception if it finds an error message in USER_ERRORS. But the original procedure required me to enter and update the object name as a parameter of the procedure for each use of the procedure. Would't it be nice to let the method it self find the 'most recently compiled object' and raise only if that object had an error? I tried to modify the procedure, but failed to find the proper timestamp column. Do you know how which view and column finds the 'most recently compiled object'? Thank...

Error compiling gcc/g++ 4.1.1 on AIX 5.2
Hi, I'm trying to compile gcc/g++ 4.1.1 on AIX 5.2. I called configure with parameters --disable-aix64 --disable-nls --enable-languages=c++. At first everything seams to work fine, but then I get lots of errors while compiling <builddir>/gcc/insn-conditions.c, like <srcdir>/gcc/config/rs6000.md:4695: error 'TARGET_POWER2' undeclared here (not in a function) <srcdir>/gcc/config/rs6000.md:4695: error 'TARGET_HARD_FLOAT' undeclared here (not in a function) followed by <srcdir>/gcc/config/rs6000.md:4696: warning: missing initializer <srcdir>/gcc/config/rs6000.md:4696: warning: (near initialization for 'insn_conditions[1].value') Any ideas what's going wrong here? Michael On 24 Jun., 22:01, "Michael H." <aixrepair...@yahoo.de> wrote: > Hi, > > I'm trying to compile gcc/g++ 4.1.1 on AIX 5.2. I called configure > with parameters --disable-aix64 --disable-nls --enable-languages=c++. > At first everything seams to work fine, but then I get lots of errors > while compiling <builddir>/gcc/insn-conditions.c, like > Hi, I found the answer in mailing list gnu-bugs. The reason is a bug in gawk (see GCC Bugzilla Bug 19601). After installing a new version of gawk (3.1.5) the error does no longer occur. Michael ...

wxMac-2.5.1 does not compile with gcc3.1 MacOSX (error on checklst.cpp)
Hi, I've just downloaded wx-cvs-20030925.tar.gz (wxMac-2.5.1) and it does not compile on MacOSX. My configuration: G4, MacOS X (10.2.4), gcc 3.1. In the main directory I did: >mkdir release >cd release >../configure --disable-shared --enable-monolithic (this ran well) >make and the error is: ../bk-deps g++ -c -o monolib_checklst.o -DNO_GCC_PRAGMA -Ilib/wx/include/mac- 2.5 -I../include -D_FILE_OFFSET_BITS=64 -D_LARGE_FILES -fpascal-strings - I../src/mac/morefilex -I/Developer/Headers/FlatCarbon -D__WXMAC__ - I../include -I../src/expat/lib -DwxUSE_BASE...

[ANN] Graphviz for Mac OS X 1.12 (v12) [ANN] Graphviz for Mac OS X 1.12 (v8) [ANN] Graphviz for Mac OS X 1.12 (v8)
Dear All: Them pesky bugs. A few more squashed courtesy of the sleepy pixel. http://www.pixelglow.com/graphviz/ What's new in v11 ------------ Fixed some comprehensive help [NRi]. Fixed scale option placeholder [NRi]. Improved application and document icons. Example files now double-click to open in application. What's new in v12 ------------ Added layout option tooltips [NRi]. Fixed layout popup button changing wrong graph [MKe]. Clicking on warning icon now opens Activity window [NRi, AM]. Revert menu item now disabled. Cheers, Glen Low --- pixelglow software | simply brilliant stuff www.pixelglow.com ...

[ANN] Graphviz for Mac OS X 1.13 (v13) [ANN] Graphviz for Mac OS X 1.12 (v8) [ANN] Graphviz for Mac OS X 1.12 (v8)
Dear All, I've just released the newest version of Mac Graphviz, featuring shapefile support and enhanced zoom. http://www.pixelglow.com/graphviz/ Shapefiles supported include PDF, EPS, PS, JPEG, PNG and all Quicktime formats. Either specify an absolute or relative (to working directory) file path, or a URL using the shapefile attribute. Developers can now also use the graphviz.framework directly with #include headers in C e.g. using Xcode "Add Frameworks..."; documentation is available from the main Graphviz site -- http://www.research.att.com/sw/tools/graphviz/libguide.pdf Changes ------- Added drawer and zoom menu commands [DWa, NRi]. Added intelligent window zooming [NRi]. Added shapefile support. Added UTF-8 support [RSc]. Added cvtgxl, gvpack and gvpr tools [BSw]. Fixed page setup then close unexpectedly invoking save dialog [PRo]. HTML-like labels now work on 10.2 (use embedded expat instead of libxml2). Improved internal frameworks (added headers, consolidated dylibs). Tracked main build of 23 June. Graphviz is still free, but I'm now accepting donations for it. Enjoy! Cheers, Glen Low --- pixelglow software | simply brilliant stuff www.pixelglow.com ...

[ANN] Graphviz for Mac OS X 1.12 (v10) [ANN] Graphviz for Mac OS X 1.12 (v8) [ANN] Graphviz for Mac OS X 1.12 (v8)
Hi all, Yet another Graphviz version. The old application icon had been voted off the island, and brand new application and document icons flown in for the task. Comprehensive help features in this version too. http://www.pixelglow.com/graphviz/ What's new: ------------ Added new application and document icons. Added comprehensive help. Fixed changes not affecting graph size displaying incorrectly [BTr]. Fixed export then close crashing the export of an open window. Improved shadowed frame for graph [NRi]. Improved status display [NRi]. Changed sources to pure BSD license. Cheers, Glen Low --- pixelglow software | simply brilliant stuff www.pixelglow.com ...

Re: error compiling bc 1.06 with readline 6.1 and flex 2.54
On Sat, 18 Dec 2010 13:17:51 -0600, Allan wrote: [snip] > I got it to build by manually editing bc\scan.l; just comment out the > (re)definition of the string pointer `readline'. [snip] That is wrong. The bc-1.06 port ist 10 years old so that certain readline declarations do no longer match the declarations used in the bc headers. The readline function declaration in readline header is an external declaration and you should make match the readline declaration in the .l file with the one in the readline header . Below a small patch that will make compile the bc port with djdev204 ...

Web resources about - OS PL/1 Compiler Error? - comp.lang.pl1

Compiler - Wikipedia, the free encyclopedia
... , or external linking . The most common reason for wanting to transform source code is to create an executable program. The name "compiler" ...

Compiler - Wikipedia, the free encyclopedia
"Compile" and "compiling" redirect here. For the software company, see Compile (publisher) . For other uses, see Compilation . This article has ...

Facebook Open-Sources HipHop PHP Compiler Software
Earlier this morning, Facebook officially made their new PHP “compiler,” called HipHop, available as open source software. In the blog post by ...

Mobile C ( Offline Compiler ) on the App Store
Read reviews, compare customer ratings, see screenshots, and learn more about Mobile C ( Offline Compiler ). Download Mobile C ( Offline Compiler ...

Art in the Age of Matter Compilers
jurvetson posted a photo: Sheba may be the harbinger of art in the digital age — a mathematical sculptor with digital matter. She manipulates ...

Interpreters and Compilers (Bits and Bytes, Episode 6) - YouTube
This animation explains the difference between interpreters and compilers. It is from Episode 6 of the classic 1983 television series, Bits and ...

Typesafe cofounder forking Scala compiler
The main contributor to the Scala compiler, Paul Phillips, has announced on GitHub that he is forking the compiler to “fix some of the innumerable ...

Does Apple's new developer agreement ban Adobe's Flash-to-iPhone compiler?
Given that any kind of formal truce between Apple and Adobe was essentially blown out of the water by Steve Job's very public slating of Flash ...

Xcode 5.0 released with iOS 7 SDK, 64-bit app compiler
... has just pushed out Xcode 5.0 to the Mac App Store. The free app includes the SDKs for iOS 7 and Mountain Lion as well as the 64-bit app compiler: ...

NVIDIA and Continuum Analytics Announce NumbaPro, A Python CUDA Compiler
... are announcing that they are bringing Python support to CUDA. Specifically, Continuum Analytics’ will be introducing a new Python CUDA compiler, ...

Resources last updated: 3/19/2016 6:38:04 PM