> On Mon, Jul 21, 2014 at 01:15:03AM -0300, Flávio Etrusco wrote:
> Is there any thought to someone extending Exuberant CTags
> to handle innovations in Pascal since Knuth? This CTags
> program has a Pascal mode that handles old-style Turbo
> Pascal source with procedural code reasonably well, bu
Hi all
Does anybody know if Lazarus IDE has something similar to View units (Ctrl+F12)
tool, but searching in all search paths defined in project settings?
It would allow to open modules much quicker than using standard Open file
dialog.
--
Regards,
Denis Golovan
__
> This patch is totally, totally wrong. variants and varutils are no longer
> part of the RTL, and this patch tries to reinstate them.
>
> It should simply work as is. Yes, there are some old dependencies left in
> the old makefile.fpc, but that doesn't matter.
>
> Something else in your setup is
>
> fpc trunk 26427 compiles here under Linux 64bit with fpc 2.6.2.
>
> Mattias
Hi Mattias
Even though it currently works, I believe at least Makefile.fpc in rtl/linux
directory is not updated.
It still assumes varutils.pp, variants.pp in old places.
You can try to look at my roughly made patc
Hi
Looks like since 26398 rev. fpc cannot be built.
It fails with following when running "make all".
classes.pp(56) Warning: Implicit uses of Variants unit
Fatal: Can't find unit variants used by Classes
Fatal: Compilation aborted
Any hints to resolve?
P.S. I am under Linux, svn rev. 26427
-
> You can use full/relative path for resources: {$R /path/to/resource/file} or
> {$R ../relative/resource/file}
Yes. I can.
The problem is that my .res files are not always in the same directory.
My build scripts use separate dirs for every arch-os-debug combination.
Now the question to the compi
Hi
Fortunately I don't use Windows when doing development for Windows :)
Cross-compiling under Linux is way better.
BR,
Denis
___
fpc-pascal maillist - fpc-pascal@lists.freepascal.org
http://lists.freepascal.org/cgi-bin/mailman/listinfo/fpc-pascal
Hi
Could anybody give a hint on which fpc option to use to force compiler search
.res files in some directory?
I've added -vt switch and it shows that the only file name being searched is in
the same directory of referring unit.
It looks like search paths for res files are not implemented at al
Ok, I see.
Thanks for the explanation.
Regards, Denis Golovan
___
fpc-pascal maillist - fpc-pascal@lists.freepascal.org
http://lists.freepascal.org/mailman/listinfo/fpc-pascal
Hi all
I experience inconsistency using fpc wrapper.
See below.
When I invoke fpc using command line "fpc -Px86_64 ..." it correctly executes
cross-compiler (ppcrossx64) under Linux 32 bit.
But if I try to push -Px86_64 into a separate file (opts.cfg) and run with "fpc
@opts.cfg" it just execu
Hi all
I am trying to get stack traces for Win64 for exception, but BackTraceStrFunc
still returns only hex representation.
My stack dumper is called via ExceptProc global variable.
Basically the procedure dumping exception stack follows below.
procedure TLogger.DumpExceptionBackTrace;
var
Fra
Thanks a ton.
Building FPC with -dTEST_WIN64_SEH makes the trick!
02.03.2013, 14:47, "Reinier Olislagers" :
> On 2-3-2013 11:42, denisgolovan wrote:
> Yep, it seems like issue is exception handling (similar to issue
> http://bugs.freepascal.org/view.php?id=21581
> etc)
&g
Hi all
Currently I am trying to make work attached code under Win64.
It uses LuaJIT 2 as drop-in replacement to Lua.
The code works fine under Linux32, Linux64, Win32, but not under Win64.
Under Win64 it crashes with unhandled exception like following:
An unhandled exception occurred at $000
Hi all
I experience some strange issue, which I happen to find accidentally.
When I try to build fpc using "make -B all" (executing in fpc sources root), it
crashes with following error.
$ make -B all
/bin/mkdir -p x86_64/units/x86_64-linux
/usr/local/lib64/fpc/source/compiler/ppc3 -Ur -Xs -O2 -
Hi Flávio
> You can use sets or TBits. Unless you mean packed arrays of byte or
> word, etc. In this case you just declare "myvar: packed array of
> byte".
>
> -Flávio
I meant bitpacked arrays of boolean. That's when boolean occupies only one bit
in memory.
Yes, I can use TBits, but unfortunate
Hi guys
It looks like currently FPC does not support dynamic bitpacked arrays.
I wonder how much work it will take to make support for it?
For some things, it's a really useful functionality.
--
Regards,
Denis Golovan
___
fpc-pascal maillist - fpc-
> Indeed, adding inline assembler to a routine disables many
> optimizations.
>
> Jonas
Hm.
Yeap. Looks like I completely missed the point :)
Thanks for the explanation.
--
Regards,
Denis Golovan
___
fpc-pascal maillist - fpc-pascal@lists.freepasc
Hi all
Recently I discovered that fpc refuses to use registers in my loops even with
optimization turned on (-O1,-O3 used).
I tested using both Linux x86 and Windows x64 version.
The results are basically the same.
The compiler just uses stack variables.
I've also included naive asm version of
> PasDoc can produce 'dot' files of unit dependencies. (No idea if it will
> handle circular refs though). GraphViz can then be used to produce
> images based on the 'dot' files. I gave up with it because it did not
> parse the implementaion sections, but you won't be needing that for unit
> depend
Hi all
Now and again, I am trying to debug complex circular references.
By complex circular references I mean those which are produced by cyclic graphs
of units, not simple one unit uses another and that uses the first one.
Unfortunately, currently fpc just shows that some circular references ex
> > Same for Lazarus, although the numbering is different.
> > There is always a svn branch "fixes", while the svn "trunk" is the
> > development version. The fixes receives only bug fixes.
> > The release was 0.9.30. The development version was 0.9.31. Then came
> > several minor releases 0.9.30.
Hi all
I am considering to subscribe to commercial support in FPC/Lazarus, but I don't
have a clear picture what actually support is :)
I am looking at http://www.lazarussupport.com/lazarus/Support and it is
somewhat too abstract.
I thought maybe it is easier to see what I am interested in and
101 - 122 of 122 matches
Mail list logo