Re: [fpc-devel] New -Xg option in the last 9778 revision
2) Configuration files (note that these may be fairly complex with IFDEFs and include files. So in Lazarus such an option could be just selecting a configuration file. Seems rather easy. -Michael ___ fpc-devel maillist - fpc-devel@lists.freepascal.org http://lists.freepascal.org/mailman/listinfo/fpc-devel
Re: [fpc-devel] New -Xg option in the last 9778 revision
2) asymmetrical (-Xs -g turns off stripping, but -g -Xs does not turn off debug information ... Do / should the order the options are given in matter ? IMHO this can be the source of major confusion. The Lazarus GUI uses check boxes to select compiler options. This no order is selectable here. And this seems like a good idea to me. -Michael ___ fpc-devel maillist - fpc-devel@lists.freepascal.org http://lists.freepascal.org/mailman/listinfo/fpc-devel
Re: [fpc-devel] New -Xg option in the last 9778 revision
2) asymmetrical (-Xs -g turns off stripping, but -g -Xs does not turn off debug information ... Do / should the order the options are given in matter ? IMHO this can be the source of major confusion. The Lazarus GUI uses check boxes to select compiler options. This no order is selectable here. And this seems like a good idea to me. The order of -Xs and -g does not matter. The stripping is disabled after all parameters are parsed. Peter ___ fpc-devel maillist - fpc-devel@lists.freepascal.org http://lists.freepascal.org/mailman/listinfo/fpc-devel
Re: [fpc-devel] PocketCMD for Windows CE devices
Problem is that it works not anymore on Windows Mobile 5 or 6 Why not simply use Lazarus. I just did a (very simple) test with crosscompiling to CE using Lazarus. It was _very_ easy. -Michael ___ fpc-devel maillist - fpc-devel@lists.freepascal.org http://lists.freepascal.org/mailman/listinfo/fpc-devel
[fpc-devel] Interfaces vs. smartlink issue
Hello, I noticed that, within a unit, all interface wrappers (groups of 'WRPR_*' symbols) are placed into one assembler unit together with unit initialization/finalization code (sorry, do not know the exact term for assember unit). As a consequence, init/final code pulls in *all* interface wrappers, which in turn pull in every class that implement an interface. That effectively disrupts smartlinking of the unit. Does this happen on purpose? Perhaps each interface wrapper should be placed into its own assembler unit? Sergei ___ fpc-devel maillist - fpc-devel@lists.freepascal.org http://lists.freepascal.org/mailman/listinfo/fpc-devel
Re: [fpc-devel] Interfaces vs. smartlink issue
Hello, I noticed that, within a unit, all interface wrappers (groups of 'WRPR_*' symbols) are placed into one assembler unit together with unit initialization/finalization code (sorry, do not know the exact term for assember unit). As a consequence, init/final code pulls in *all* interface wrappers, which in turn pull in every class that implement an interface. That effectively disrupts smartlinking of the unit. Does this happen on purpose? Perhaps each interface wrapper should be placed into its own assembler unit? It is legacy. You can improve it to put the wrappers for one interface definition in its own section. Peter ___ fpc-devel maillist - fpc-devel@lists.freepascal.org http://lists.freepascal.org/mailman/listinfo/fpc-devel
[fpc-devel] Strange .EXE stripped issue
Hi to all, I have a little problem stripping my .EXE compiled file. The final size with the DEBUG symbols (from -gl flag) is: 65132 bytes (A) with strip --strip-debug become: 64583 bytes (B) with strip --strip-unneeded become: 39424 bytes Difference from (A) - (B) = 25159 bytes What data are stored into this 25159 bytes? How I can remove them? -- Best regards... Fabio Dell'Aria. ___ fpc-devel maillist - fpc-devel@lists.freepascal.org http://lists.freepascal.org/mailman/listinfo/fpc-devel
Re: [fpc-devel] Strange .EXE stripped issue
Hi, 2008/1/21, Jonas Maebe [EMAIL PROTECTED]: On 21 Jan 2008, at 13:18, Fabio Dell'Aria wrote: I have a little problem stripping my .EXE compiled file. The final size with the DEBUG symbols (from -gl flag) is: 65132 bytes (A) with strip --strip-debug become: 64583 bytes (B) with strip --strip-unneeded become: 39424 bytes Difference from (A) - (B) = 25159 bytes What data are stored into this 25159 bytes? Symbols. How I can remove them? With --strip-unneeded No my questions is: How I can remove ONLY this symbols (not DEBUG sections)? :) -- Best regards... Fabio Dell'Aria. ___ fpc-devel maillist - fpc-devel@lists.freepascal.org http://lists.freepascal.org/mailman/listinfo/fpc-devel
Re: [fpc-devel] Strange .EXE stripped issue
Hi, 2008/1/21, Jonas Maebe [EMAIL PROTECTED]: On 21 Jan 2008, at 13:18, Fabio Dell'Aria wrote: I have a little problem stripping my .EXE compiled file. The final size with the DEBUG symbols (from -gl flag) is: 65132 bytes (A) with strip --strip-debug become: 64583 bytes (B) with strip --strip-unneeded become: 39424 bytes Difference from (A) - (B) = 25159 bytes What data are stored into this 25159 bytes? Symbols. How I can remove them? With --strip-unneeded No my questions is: How I can remove ONLY this symbols (not DEBUG sections)? :) strip is a GNU binutils tool. Please ask on the binutils mailing lists. Peter ___ fpc-devel maillist - fpc-devel@lists.freepascal.org http://lists.freepascal.org/mailman/listinfo/fpc-devel
Re: [fpc-devel] Strange .EXE stripped issue
Fabio Dell'Aria wrote: The final size with the DEBUG symbols (from -gl flag) is: 65132 bytes (A) with strip --strip-debug become: 64583 bytes (B) with strip --strip-unneeded become: 39424 bytes Difference from (A) - (B) = 25159 bytes What data are stored into this 25159 bytes? Use objdump -h file to see what sections an executable contains. Micha ___ fpc-devel maillist - fpc-devel@lists.freepascal.org http://lists.freepascal.org/mailman/listinfo/fpc-devel
[fpc-devel] gtkliststore.inc
Hello, FPC developers' list. Can someone tell me what for gtk2 package have this strange defines inside sources: {$IFNDEF KYLIX} function gtk_list_store_new(n_columns:gint; args:array of const):PGtkListStore; cdecl; overload; external gtklib; function gtk_list_store_new(n_columns:gint):PGtkListStore; cdecl; overload; varargs; external gtklib; {$ELSE} function gtk_list_store_new(n_columns:gint):PGtkListStore; varargs; cdecl; external gtklib; {$ENDIF} I am about KYLIX. -- Best regards, Paul Ishenin. ___ fpc-devel maillist - fpc-devel@lists.freepascal.org http://lists.freepascal.org/mailman/listinfo/fpc-devel