[Harbour] HB_GTGUI_HACK

2008-09-23 Thread Szakáts Viktor
Hi all, I intend to remove HB_GTGUI_HACK undocumented build-time option from Harbour, as it just adds up the build-time and GT selection confusion. Users needing GUI apps, should very simply use these in their code with any default Windows builds: -- PROCEDURE HB_GTSYS REQUEST HB_GT_GUI_DEFAU

[Harbour] 2008-09-24 02:26 UTC+0200 Viktor Szakats (harbour.01 syenar hu)

2008-09-23 Thread Szakáts Viktor
2008-09-24 02:26 UTC+0200 Viktor Szakats (harbour.01 syenar hu) * make_gcc.sh * make_gnu_os2.cmd * make_gnu.bat * make_gnu.sh * make_tgz.sh % Some additional GT defaulting logic removed. -- Brgds, Viktor ___ Harbour mailing list Harbour@har

[Harbour] HB_NO_DEBUG - getting rid of it

2008-09-23 Thread Szakáts Viktor
Hi Przemek, I still think we should have no HB_NO_DEBUG switch to avoid debug overhead. The debug overhead comes from these few sources: - hb_vmDo() - hb_vmSend() - HB_P_LINE opcode - HB_P_MODULENAME opcode From these, HB_P_LINE can easily be suppressed by /l switch for maximum speed and small

[Harbour] Compile error...

2008-09-23 Thread elart
I have a prg was compiling without any issue, now i get: /usr/lib/gcc/i486-linux-gnu/4.3.2/../../../../lib/libgttrm.a(gttrm.o): In function `hb_gt_trm_termOutTrans': gttrm.c:(.text+0x5b1): undefined reference to `hb_vmCDP' /usr/lib/gcc/i486-linux-gnu/4.3.2/../../../../lib/libgttrm.a(gttrm.o):

[Harbour] Re: mingw results and Windows compiler roundup

2008-09-23 Thread Szakáts Viktor
The other conclusion is that MSVS2008 seems to generate far the best code, 1.1.0 MT mode even beats 1.0.1 ST mode with all compilers. Update/Typo: Except itself. Brgds and good night, Viktor ___ Harbour mailing list Harbour@harbour-project.org http:/

[Harbour] Re: mingw results and Windows compiler roundup

2008-09-23 Thread Szakáts Viktor
[ Would be interesting to see how removing -gc3 would change above results, but I'll leave that for another time. ] Update: Thinking about it, -gc3 for Harbour itself only (that was the case) shouldn't affect test results too much, as speedtst doesn't seem to call .prg RTL code, so the results

[Harbour] mingw results and Windows compiler roundup

2008-09-23 Thread Szakáts Viktor
Hi all, I've built mingw with -DHB_USE_TLS (the rest is default), then I got 'undefined reference to '__emutls_get_address' errors on linking. Such symbol doesn't BTW exist in MinGW 4.3.2 supplied libs. Then I went with the fully default build, and it worked okay. Final mingw speedtst results a

Re: [Harbour] memtst and HB_FM_DL_ALLOC

2008-09-23 Thread [EMAIL PROTECTED]
>Here bcc go always fast, but dla have "WAITS" in various steps and the >results are variable, but always bad than bcc. Very strange. My first test was done under Windows Vista. The same test with a notebook and windows XP and DLA works better, with no "waits". Regards, Toninho. _

Re: [Harbour] Linux default GT

2008-09-23 Thread Szakáts Viktor
No problem at all Przemek. Brgds, Viktor On 2008.09.24., at 1:20, Przemyslaw Czerpak wrote: On Wed, 24 Sep 2008, Szak�ts Viktor wrote: I have it ready in the cue already :) just let me commit it, and please review / complete it. Ups, sorry Viktor, I commited before I received this message :

[Harbour] 2008-09-24 01:41 UTC+0200 Viktor Szakats (harbour.01 syenar hu)

2008-09-23 Thread Szakáts Viktor
2008-09-24 01:41 UTC+0200 Viktor Szakats (harbour.01 syenar hu) * contrib/hbwhat/tests/whatdemo.prg ! SVN ID. + contrib/hbwhat/tests/bld_b32.bat + contrib/hbwhat/tests/bld_vc.bat + Added missing non-GNU make files. -- Brgds, Viktor ___ Har

[Harbour] 2008-09-24 01:37 UTC+0200 Viktor Szakats (harbour.01 syenar hu)

2008-09-23 Thread Szakáts Viktor
2008-09-24 01:37 UTC+0200 Viktor Szakats (harbour.01 syenar hu) * source/rtl/gtsys.c * Added HB_WINCE with GTWVT to the set of defaults. * config/darwin/global.cf * config/hpux/global.cf * config/linux/global.cf * config/sunos/global.cf * config/bsd/global.cf % Removed unnecess

Re: [Harbour] Linux default GT

2008-09-23 Thread Przemyslaw Czerpak
On Wed, 24 Sep 2008, Szak�ts Viktor wrote: > I have it ready in the cue already :) just let me commit it, > and please review / complete it. Ups, sorry Viktor, I commited before I received this message :-( best regards, Przemek ___ Harbour mailing list

[Harbour] 2008-09-23 16:19 UTC-0800 Pritpal Bedi ([EMAIL PROTECTED])

2008-09-23 Thread Pritpal Bedi
2008-09-23 16:19 UTC-0800 Pritpal Bedi ([EMAIL PROTECTED]) * harbour/contrib/hbwhat + /tests + /tests/whatdemo.prg A demo program, very basic, will be expanded in future, together with gtwvt. Regards Pritpal Bedi -- View this message in context: http://www.nabble.com/2008-

[Harbour] Re: 2008-09-19 22:27 UTC+0200 Przemyslaw Czerpak (druzus/at/priv.onet.pl)

2008-09-23 Thread David Arturo Macias Corona
Przemek, Maurilio: [Maurilio] >> I did the change to the makefile, I get an immediate SIGSEGV simply >>starting >> hbrun. >> (E:\repository\harbour-svn\tests)\harbour\bin\hbrun mt\mttest01.prg >> Killed by SIGSEGV [Przemek] >Then there is probably sth wrong with your build. >I would like to wa

[Harbour] 2008-09-24 00:44 UTC+0200 Przemyslaw Czerpak (druzus/at/priv.onet.pl)

2008-09-23 Thread Przemyslaw Czerpak
2008-09-24 00:44 UTC+0200 Przemyslaw Czerpak (druzus/at/priv.onet.pl) * harbour/source/rtl/hbgtcore.c * harbour/source/rtl/gtsys.c * use GTTRM as default GT in *nixes best regards Przemek ___ Harbour mailing list Harbour@harbour-project.org http:

Re: [Harbour] Linux default GT

2008-09-23 Thread Szakáts Viktor
I have it ready in the cue already :) just let me commit it, and please review / complete it. Brgds, Viktor On 2008.09.24., at 0:38, Przemyslaw Czerpak wrote: On Wed, 24 Sep 2008, Szak�ts Viktor wrote: Hi Viktor, In hbgtcore.c currently for HB_OS_LINUX the default GT is GTCRS. Shouldn't we

Re: [Harbour] Linux default GT

2008-09-23 Thread Przemyslaw Czerpak
On Wed, 24 Sep 2008, Szak�ts Viktor wrote: Hi Viktor, > In hbgtcore.c currently for HB_OS_LINUX the > default GT is GTCRS. Shouldn't we change this > to GTTRM? Yes it should and probably also for all other *nixes. I'll commit modification in a while. best regards, Przemek __

[Harbour] Linux default GT

2008-09-23 Thread Szakáts Viktor
Hi Przemek, In hbgtcore.c currently for HB_OS_LINUX the default GT is GTCRS. Shouldn't we change this to GTTRM? Brgds, Viktor ___ Harbour mailing list Harbour@harbour-project.org http://lists.harbour-project.org/mailman/listinfo/harbour

[Harbour] 2008-09-24 00:18 UTC+0200 Viktor Szakats (harbour.01 syenar hu)

2008-09-23 Thread Szakáts Viktor
2008-09-24 00:18 UTC+0200 Viktor Szakats (harbour.01 syenar hu) * tests/memtst.prg + Added #include "simpleio.ch" to allow output redirection. * bin/bld.bat * bin/bld_os2.cmd * Removed any HB_GT_LIB "selection" logic. Selection is to be done from source code, from now on all s

Re: [Harbour] 2008-09-23 15:38 UTC+0200 Przemyslaw Czerpak (druzus/at/priv.onet.pl)

2008-09-23 Thread Lorenzo Fiorini
On Tue, Sep 23, 2008 at 3:39 PM, Przemyslaw Czerpak <[EMAIL PROTECTED]> wrote: > Please make speed tests for MT mode with different C compilers > and different platforms. Under Fedora 9, gcc version 4.3.0 20080428 (Red Hat 4.3.0-8) (GCC) MT total real time:

Re: [Harbour] Re: 2008-09-19 22:27 UTC+0200 Przemyslaw Czerpak (druzus/at/priv.onet.pl)

2008-09-23 Thread Przemyslaw Czerpak
On Tue, 23 Sep 2008, David Arturo Macias Corona wrote: Hi David, > >2008-09-23 17:09 UTC+0200 Przemyslaw Czerpak (druzus/at/priv.onet.pl) > > * harbour/config/os2/gcc.cf > >* enabled -DTCPV40HDRS in default builds, can be disabled by > > setting HB_OS2_NEW_TCP=yes environment variable >

RE: [Harbour] Re: MSVS 2008: error with new alloc

2008-09-23 Thread Massimo Belgrano
>>> use //build switch and if necessary redirect stderr to >>> stdout by //stderr:1 >> I know these, but couldn't we make it easier? >> I may look into this later. memtest output can >> only be screen grabbed. >So probably we should follow Massimo suggestion and use >SET ALTERNATE for logging out

Re: [Harbour] Proposed changes to EXTRAP.C

2008-09-23 Thread Szakáts Viktor
Hi Randy, What you say is probably possible to control by the return value of the exception handler. I'd feel a Set() to be too early to control such an isolated, moreover platform-dependent detail in Harbour, but if we can somehow broaden the context this may also be an option. For the time be

[Harbour] Re: 2008-09-19 22:27 UTC+0200 Przemyslaw Czerpak (druzus/at/priv.onet.pl)

2008-09-23 Thread David Arturo Macias Corona
Przemek, Maurilio: [Maurilio] >I see that the OS/2 build has -DTCPV40HDRS disabled by default, now, >IMHO this >is an error since the biggest part of OS/2 systems still out there won't >be >able to run harbour .EXEs if this define is not enabled. >On the contrary, on newer systems this define

Re: [Harbour] Re: MSVS 2008: error with new alloc

2008-09-23 Thread Przemyslaw Czerpak
On Tue, 23 Sep 2008, Szak�ts Viktor wrote: Hi Viktor, >> use //build switch and if necessary redirect stderr to >> stdout by //stderr:1 > I know these, but couldn't we make it easier? > I may look into this later. memtest output can > only be screen grabbed. So probably we should follow Massimo

Re: [Harbour] hbrun and mt

2008-09-23 Thread Przemyslaw Czerpak
On Tue, 23 Sep 2008, Maurilio Longo wrote: Hi Maurilio, > I did the change to the makefile, I get an immediate SIGSEGV simply starting > hbrun. > (E:\repository\harbour-svn\tests)\harbour\bin\hbrun mt\mttest01.prg > Killed by SIGSEGV > pid=0x7e87 ppid=0x0253 tid=0x0001 slot=0x00b2 pri=0x0200 mc=0

Re: [Harbour] Proposed changes to EXTRAP.C

2008-09-23 Thread Randy Portnoff
...or is it Harbour that is telling Windows that a GPF occurred? If so, we should make this optional and just allow the EXE to exit. Hi Viktor, What would also be great would be if you could optionally (eg. _SET_FORCE_FATER_TERMINATION) force the app to close instead of being held open by the

Re: [Harbour] .cdx corruption detected

2008-09-23 Thread Szakáts Viktor
Hi Przemek, The apps are 100% exactly the same. I just suspect the index was created by the Clipper version of my app, but nevertheless 1.0.1 opens it, while 1.1.0 doesn't. And this is the answer. Header validation is too strong and recognize this index as corrupted because it sets that for ex

Re: [Harbour] Proposed changes to EXTRAP.C

2008-09-23 Thread Randy Portnoff
Hi Viktor, What would also be great would be if you could optionally (eg. _SET_FORCE_FATER_TERMINATION) force the app to close instead of being held open by the O/S GPF alert message. For example, a server application that runs into a GPF, is held open/running since Windows attempts to displa

[Harbour] 2008-09-23 21:42 UTC+0200 Przemyslaw Czerpak (druzus/at/priv.onet.pl)

2008-09-23 Thread Przemyslaw Czerpak
2008-09-23 21:42 UTC+0200 Przemyslaw Czerpak (druzus/at/priv.onet.pl) * harbour/bin/hb-func.sh * use HB_IMPORT instead of hardcoded attributes in hb* scripts * harbour/source/common/hbver.c * generate different info for POCC and XCC compilers * harbour/source/rdd/dbfcdx/dbfcdx1.c

Re: [Harbour] .cdx corruption detected

2008-09-23 Thread Przemyslaw Czerpak
On Tue, 23 Sep 2008, Szak�ts Viktor wrote: Hi Viktor, > The apps are 100% exactly the same. I just suspect > the index was created by the Clipper version of my > app, but nevertheless 1.0.1 opens it, while 1.1.0 > doesn't. And this is the answer. Header validation is too strong and recognize thi

[Harbour] 2008-09-23 20:41 UTC+0200 Viktor Szakats (harbour.01 syenar hu)

2008-09-23 Thread Szakáts Viktor
2008-09-23 20:41 UTC+0200 Viktor Szakats (harbour.01 syenar hu) * source/vm/extrap.c ! Fixed to close .log file before showing MessageBox(). [TOMERGE 1.0] + Don't attempt to write into .log file if the _SET_HBOUTLOG filename is empty. ; TOFIX: Remove all remaining Messa

Re: [Harbour] Proposed changes to EXTRAP.C

2008-09-23 Thread Szakáts Viktor
Hi Randy, Hi all, I suggest the following changes to EXTAP.C: 1. The log should be closed BEFORE any alert messages are displayed. True and thanks, I'll commit this ASAP. 2. There needs to be a way to suppress any alert messages (ie. create log and exit) for applications that do not req

Re: [Harbour] .cdx corruption detected

2008-09-23 Thread Szakáts Viktor
Hi Przemek, It's possible that two applications where using different national collating. Please be sure that it's not in your code. If your code is correct then it's possible that I make some mistake when I was updating CDP internals moving hb_code_page to HB_STACK. In such case any example whi

Re: [Harbour] 2008-09-23 18:56 UTC+0200 Viktor Szakats (harbour.01 syenar hu)

2008-09-23 Thread [EMAIL PROTECTED]
>2008-09-23 18:56 UTC+0200 Viktor Szakats (harbour.01 syenar hu) > * contrib/rddads/adsmgmnt.c >! Blind fix to ADSMGGETWORKERTHREADACTIVITY() where the > first element in list wasn't returned. > Please test, I couldn't. Thanks Viktor. Tested and working. Regards, Toninho. ___

Re: Re: [Harbour] 2008-09-23 08:45 UTC-0300 Antonio Carlos Pantaglione

2008-09-23 Thread [EMAIL PROTECTED]
>Hi Toninho, > >On 2008.09.23., at 13:54, [EMAIL PROTECTED] wrote: > >> 2008-09-23 08:45 UTC-0300 Antonio Carlos Pantaglione >> <[EMAIL PROTECTED]> >> * contrib/rddads/adsmgmnt.c >> ! ADSMGGETUSERNAMES() that was skipping first user > >Many thanks. Probably the same fix should be done for >AD

[Harbour] Proposed changes to EXTRAP.C

2008-09-23 Thread Randy Portnoff
Hi all, I suggest the following changes to EXTAP.C: 1. The log should be closed BEFORE any alert messages are displayed. 2. There needs to be a way to suppress any alert messages (ie. create log and exit) for applications that do not require user activity (eg. server programs). Regards, Ran

Re: [Harbour] hbrun and mt

2008-09-23 Thread Maurilio Longo
Przemyslaw, I did the change to the makefile, I get an immediate SIGSEGV simply starting hbrun. (E:\repository\harbour-svn\tests)\harbour\bin\hbrun mt\mttest01.prg Killed by SIGSEGV pid=0x7e87 ppid=0x0253 tid=0x0001 slot=0x00b2 pri=0x0200 mc=0x0001 E:\HARBOUR\BIN\HBRUN.EXE HBRUN 0:00026f26 cs:ei

Re: [Harbour] Re: MSVS 2008: error with new alloc

2008-09-23 Thread Szakáts Viktor
Hi Przemek, I've done the tests, MT speed got ~5% better. The MT overhead for MSVS 2008 on my system is 41%. How much do you think this could be further enhanced in case we'd pass the HVM context to all Harbour API functions? ST ST MT

[Harbour] PellesC 5.0 64-bit issues confirmed

2008-09-23 Thread Szakáts Viktor
See: http://forum.pellesc.de/index.php?topic=2590.0 Brgds, Viktor ___ Harbour mailing list Harbour@harbour-project.org http://lists.harbour-project.org/mailman/listinfo/harbour

Re: [Harbour] 2008-09-23 06:51 UTC-0800 Pritpal Bedi ([EMAIL PROTECTED])

2008-09-23 Thread Szakáts Viktor
Thanks a lot Pritpal. Brgds, Viktor On 2008.09.23., at 15:53, Pritpal Bedi wrote: 2008-09-23 06:51 UTC-0800 Pritpal Bedi ([EMAIL PROTECTED]) * harbour/contrib/hbwin/win_ole.c % hb_vmPushDynDym() => hb_vmPushDynSym() Regards Pritpal Bedi -- View this message in context: http://www.nabble

Re: [Harbour] 2008-09-22 19:19 UTC+0200 Viktor Szakats (harbour.01 syenar hu)

2008-09-23 Thread Szakáts Viktor
Hi Miguel, I think is better this: #ifdef HB_C52_STRICT if( uiLen > CDX_MAXKEY ) uiLen = CDX_MAXKEY; #endif ... #ifndef HB_C52_STRICT else if ( bType == 'C' && uiLen > CDX_MAXKEY ) { if( hb_cdxErrorRT( pArea, EG_DATAWIDTH, EDBF_INVALIDKEY, NULL, 0, EF_CANDEFAULT

Re: [Harbour] Re: MSVS 2008: error with new alloc

2008-09-23 Thread Szakáts Viktor
Hi Przemek, Okay, I'll redo the tests using DLALLOC only to keep it simple. BTW, can we change memtst to output to stdout, not require key presses, and also for both tests to include build-time switches in their outputs. These would ease testing. use //build switch and if necessary redirect s

[Harbour] 2008-09-23 18:56 UTC+0200 Viktor Szakats (harbour.01 syenar hu)

2008-09-23 Thread Szakáts Viktor
2008-09-23 18:56 UTC+0200 Viktor Szakats (harbour.01 syenar hu) * contrib/rddads/adsmgmnt.c ! Blind fix to ADSMGGETWORKERTHREADACTIVITY() where the first element in list wasn't returned. Please test, I couldn't. * source/vm/fm.c ! Stupid typo. (I guess I'm completely confu

Re: [Harbour] hbrun and mt

2008-09-23 Thread Przemyslaw Czerpak
On Tue, 23 Sep 2008, Maurilio Longo wrote: > Hi Przemyslaw, > is hbrun mt ready? I was trying to use it to run mt\mttestxx but I get errors, > sometimes, like this: > (E:\repository\harbour-svn\tests\mt)\harbour\bin\hbrun mttest01.prg > Harbour 1.1.0dev (Rev. 9479) > Error BASE/3012 Argument error

Re: [Harbour] 2008-09-23 15:38 UTC+0200 Przemyslaw Czerpak (druzus/at/priv.onet.pl)

2008-09-23 Thread Saulius Zrelskis
Some speed test results: ST | MT --- -TLS +TLS | -TLS +TLS =|==

Re: [Harbour] 2008-09-23 17:09 UTC+0200 Przemyslaw Czerpak (druzus/at/priv.onet.pl)

2008-09-23 Thread Maurilio Longo
Very neat, thanks a lot. Maurilio. Przemyslaw Czerpak wrote: > 2008-09-23 17:09 UTC+0200 Przemyslaw Czerpak (druzus/at/priv.onet.pl) > * harbour/config/os2/gcc.cf > * enabled -DTCPV40HDRS in default builds, can be disabled by > setting HB_OS2_NEW_TCP=yes environment variable > > bes

Re: [Harbour] os/2 build

2008-09-23 Thread Maurilio Longo
Przemyslaw Czerpak wrote: > Yes of course. These are real OS2 users developers decisions. > I have no idea what this macro exactly does and where is necessary. This macro enables the use of older version of TCP/IP headers, like those that were available on OS/2 before eComStation and OS/2 Warp Serv

[Harbour] PillHarbour: website

2008-09-23 Thread Massimo Belgrano
I have started the website of harbour documentation http://sites.google.com/site/pillharbour/ I think that everyone on harbour community have in heads same strong capability of this tools. 90% of this knowledge in head and 10% in files With this site I want collect documentation from any so

Re: [Harbour] 2008-09-22 19:19 UTC+0200 Viktor Szakats (harbour.01 syenar hu)

2008-09-23 Thread Miguel Angel Marchuet
I think is better this: #ifdef HB_C52_STRICT if( uiLen > CDX_MAXKEY ) uiLen = CDX_MAXKEY; #endif ... #ifndef HB_C52_STRICT else if ( bType == 'C' && uiLen > CDX_MAXKEY ) { if( hb_cdxErrorRT( pArea, EG_DATAWIDTH, EDBF_INVALIDKEY, NULL, 0, EF_CANDEFAULT ) == E_DE

[Harbour] 2008-09-23 17:09 UTC+0200 Przemyslaw Czerpak (druzus/at/priv.onet.pl)

2008-09-23 Thread Przemyslaw Czerpak
2008-09-23 17:09 UTC+0200 Przemyslaw Czerpak (druzus/at/priv.onet.pl) * harbour/config/os2/gcc.cf * enabled -DTCPV40HDRS in default builds, can be disabled by setting HB_OS2_NEW_TCP=yes environment variable best regards Przemek ___ Harbour ma

Re: [Harbour] os/2 build

2008-09-23 Thread Przemyslaw Czerpak
On Tue, 23 Sep 2008, Maurilio Longo wrote: Hi Maurilio, > I see that the OS/2 build has -DTCPV40HDRS disabled by default, now, IMHO this > is an error since the biggest part of OS/2 systems still out there won't be > able to run harbour .EXEs if this define is not enabled. > On the contrary, on n

[Harbour] hbrun and mt

2008-09-23 Thread Maurilio Longo
Hi Przemyslaw, is hbrun mt ready? I was trying to use it to run mt\mttestxx but I get errors, sometimes, like this: (E:\repository\harbour-svn\tests\mt)\harbour\bin\hbrun mttest01.prg Harbour 1.1.0dev (Rev. 9479) Error BASE/3012 Argument error: HB_THREADJOIN Called from HB_THREADJOIN(0) Called

[Harbour] os/2 build

2008-09-23 Thread Maurilio Longo
Prezemyslaw, I see that the OS/2 build has -DTCPV40HDRS disabled by default, now, IMHO this is an error since the biggest part of OS/2 systems still out there won't be able to run harbour .EXEs if this define is not enabled. On the contrary, on newer systems this defines will just cause very mino

Re: [Harbour] SIGFPE during build of hbtest

2008-09-23 Thread Przemyslaw Czerpak
On Tue, 23 Sep 2008, Maurilio Longo wrote: Hi Maurilio, > I removed those lines and a full rebuild was completed without errors. Many thanks for confirmations. best regards, Przemek ___ Harbour mailing list Harbour@harbour-project.org http://lists.har

[Harbour] 2008-09-23 07:05 UTC-0800 Pritpal Bedi ([EMAIL PROTECTED])

2008-09-23 Thread Pritpal Bedi
2008-09-23 07:05 UTC-0800 Pritpal Bedi ([EMAIL PROTECTED]) * harbour/contrib/hbwhat/wincorec.c % hb_retptr() => HB_RETWI() Regards Pritpal Bedi -- View this message in context: http://www.nabble.com/2008-09-23-07%3A05-UTC-0800-Pritpal-Bedi-%28pritpal%40vouchcac.com%29-tp19628541p19628541.

[Harbour] 2008-09-23 06:51 UTC-0800 Pritpal Bedi ([EMAIL PROTECTED])

2008-09-23 Thread Pritpal Bedi
2008-09-23 06:51 UTC-0800 Pritpal Bedi ([EMAIL PROTECTED]) * harbour/contrib/hbwin/win_ole.c % hb_vmPushDynDym() => hb_vmPushDynSym() Regards Pritpal Bedi -- View this message in context: http://www.nabble.com/2008-09-23-06%3A51-UTC-0800-Pritpal-Bedi-%28pritpal%40vouchcac.com%29-tp1962820

Re: [Harbour] SIGFPE during build of hbtest

2008-09-23 Thread Maurilio Longo
Przemyslaw, I removed those lines and a full rebuild was completed without errors. Thanks a lot. Maurilio. Przemyslaw Czerpak wrote: > On Tue, 23 Sep 2008, Maurilio Longo wrote: > > Hi Maurilio, > > >> I'm not able to get a stack dump :( (it's a long story, gdb is not up to date >> with the

RE: [Harbour] PillHarbour: Harbour MT/ST mode Please revise and replycorrected version

2008-09-23 Thread Massimo Belgrano
I have shared a document with title "Harbour MT Mode and ST Mode": is achivied on online on Google Document. clic for open: http://docs.google.com/Doc?id=dhmtv9fs_164d7q4qf4&invite= Require me authorization for modification or use as tem

[Harbour] 2008-09-23 15:38 UTC+0200 Przemyslaw Czerpak (druzus/at/priv.onet.pl)

2008-09-23 Thread Przemyslaw Czerpak
2008-09-23 15:38 UTC+0200 Przemyslaw Czerpak (druzus/at/priv.onet.pl) * harbour/source/vm/codebloc.c * updated old and not longer valid comment * harbour/source/vm/memvars.c * minor code formatting * harbour/source/common/expropt2.c ! fixed bug in power math expression optimizat

Re: [Harbour] 2008-09-15 13:38 UTC+0200 Przemyslaw Czerpak (druzus/at/priv.onet.pl)

2008-09-23 Thread Przemyslaw Czerpak
On Thu, 18 Sep 2008, Mindaugas Kavaliauskas wrote: Hi Mindaugas, > it's possible the goal could be reached, but I'm unable to write inline > function for tls accessing. > Perhaps we should use HB_THREAD_STUB or pass pStack parameter to functions. I've just implement it for estack.c and hvm.c. I

Re: [Harbour] Re: MSVS 2008: error with new alloc

2008-09-23 Thread Przemyslaw Czerpak
On Tue, 23 Sep 2008, Szak�ts Viktor wrote: Hi Viktor, Many hanks for your tests. Now the MT overhead is much smaller. I'll commit last modification with TLS buffering and this will be the final version. Probably we can improve the speed yet a little bit but it will not be anything hardly noticeab

[Harbour] PillHarbour: Harbour MT/ST mode Please revise and reply corrected version

2008-09-23 Thread Massimo Belgrano
Harbour MT Mode and ST Mode How build st mode remove if present all instances of HB_BUILD_ST from your settings. Please also remove all HB_MT_VM macros you may have in C_USR How Build in mt mode Creating both libraries is default as long as you do not change it by some personal settings f.e

Re: [Harbour] .cdx corruption detected

2008-09-23 Thread Przemyslaw Czerpak
On Tue, 23 Sep 2008, Szak�ts Viktor wrote: Hi Viktor, > I've now built my app with 1.1.0dev (ST), and while everything > seem to work fine and much snappier than in 1.0.1, I'm getting > "1012 Corruption detected" errors on so far perfectly working > .cdx indexes. After a reindex the problem goes

Re: [Harbour] SIGFPE during build of hbtest

2008-09-23 Thread Przemyslaw Czerpak
On Tue, 23 Sep 2008, Maurilio Longo wrote: Hi Maurilio, > I'm not able to get a stack dump :( (it's a long story, gdb is not up to date > with the rest of the GCC tool chain I'm using) > But I've found that the problem comes from rt_hvma.prg lines 191 and 192 > (commented as operators precedence

Re: [Harbour] Re: 2008-09-19 22:27 UTC+0200 Przemyslaw Czerpak (druzus/at/priv.onet.pl)

2008-09-23 Thread Przemyslaw Czerpak
On Tue, 23 Sep 2008, David Arturo Macias Corona wrote: Hi David, > Results below for OS/2:mttest08.prg:mt Mant thanks for your all tests. > Below are results of current Harbour under eComStation 1.2MR > gcc 3.3.5 > ST >total application time: 47.13 >total re

Re: [Harbour] 2008-09-23 08:45 UTC-0300 Antonio Carlos Pantaglione

2008-09-23 Thread Szakáts Viktor
Hi Toninho, On 2008.09.23., at 13:54, [EMAIL PROTECTED] wrote: 2008-09-23 08:45 UTC-0300 Antonio Carlos Pantaglione <[EMAIL PROTECTED]> * contrib/rddads/adsmgmnt.c ! ADSMGGETUSERNAMES() that was skipping first user Many thanks. Probably the same fix should be done for ADSGETWORKERTHREAD

[Harbour] 2008-09-23 14:50 UTC+0200 Viktor Szakats (harbour.01 syenar hu)

2008-09-23 Thread Szakáts Viktor
2008-09-23 14:50 UTC+0200 Viktor Szakats (harbour.01 syenar hu) * contrib/rddads/adsmgmnt.c * Minor. -- Brgds, Viktor ___ Harbour mailing list Harbour@harbour-project.org http://lists.harbour-project.org/mailman/listinfo/harbour

[Harbour] 2008-09-23 14:38 UTC+0200 Viktor Szakats (harbour.01 syenar hu)

2008-09-23 Thread Szakáts Viktor
2008-09-23 14:38 UTC+0200 Viktor Szakats (harbour.01 syenar hu) * contrib/hbole/ole2.c * contrib/hbgf/hbgfwin/winapi.c * contrib/hbgf/hbgfos2/os2pm.c * contrib/hbwin/win_ole.c * contrib/hbwhat/wincorec.c * source/rtl/do.c * source/rdd/workarea.c % hb_vmPushSymbol( hb_dynsymSymbol(

[Harbour] memtst and HB_FM_DL_ALLOC

2008-09-23 Thread [EMAIL PROTECTED]
Hi, in memtst, with bcc 6.10 and windows: "testing single large memory blocks allocation and freeing..." same results using bcc and dla "testing many large memory blocks allocation and freeing..." bcc 7.39, dla 4.94 "testing large memory block reallocation with intermediate allocations..." "

[Harbour] 2008-09-23 08:45 UTC-0300 Antonio Carlos Pantaglione

2008-09-23 Thread [EMAIL PROTECTED]
2008-09-23 08:45 UTC-0300 Antonio Carlos Pantaglione <[EMAIL PROTECTED]> * contrib/rddads/adsmgmnt.c + ADSMGSETHANDLE( nHandle ) to set an existing connection handle ! ADSMGGETUSERNAMES() that was skipping first user Regards, Toninho. __

[Harbour] .cdx corruption detected

2008-09-23 Thread Szakáts Viktor
Hi all, I've now built my app with 1.1.0dev (ST), and while everything seem to work fine and much snappier than in 1.0.1, I'm getting "1012 Corruption detected" errors on so far perfectly working .cdx indexes. After a reindex the problem goes away, but this is not very good even in this case. Th

Re: [Harbour] SIGFPE during build of hbtest

2008-09-23 Thread Maurilio Longo
Przemyslaw, I'm not able to get a stack dump :( (it's a long story, gdb is not up to date with the rest of the GCC tool chain I'm using) But I've found that the problem comes from rt_hvma.prg lines 191 and 192 (commented as operators precedence). If I comment them the build goes on, If I uncomme

[Harbour] 2008-09-23 11:27 UTC+0200 Viktor Szakats (harbour.01 syenar hu)

2008-09-23 Thread Szakáts Viktor
2008-09-23 11:27 UTC+0200 Viktor Szakats (harbour.01 syenar hu) * source/vm/fm.c ! Enabled USE_DL_PREFIX for _MSC_VER compilers. This may be refined, but without this setting MSVS 2008 would give an error on exit and GPF. -- Brgds, Viktor

[Harbour] 2008-09-23 11:24 UTC+0200 Viktor Szakats (harbour.01 syenar hu)

2008-09-23 Thread Szakáts Viktor
2008-09-23 11:24 UTC+0200 Viktor Szakats (harbour.01 syenar hu) * include/hbthread.h % Enabled TLS for _MSC_VER compilers. ; Tested with MSVS 2008 32-bit. Someone pls test with older/other MSVC compilers (especially MSVC6 and WinCE mode) to refine the default setting if neede

[Harbour] Re: 2008-09-19 22:27 UTC+0200 Przemyslaw Czerpak (druzus/at/priv.onet.pl)

2008-09-23 Thread David Arturo Macias Corona
Przemek: For mttest05: [...] result: 0 errors Thank you very much for your tests. Looks that it works correctly anyhow I do not like the way it is done. I hope that OS2 users will clean this code in the future. Maurilio, are you reading ? :-) ( yes, I know you can clean it ) 2008-09-2

Re: [Harbour] Re: SIGFPE during build of hbtest

2008-09-23 Thread Maurilio Longo
David Arturo Macias Corona wrote: >> It's possible that this problem is local to GCC builds anyhow >> it should be located and fixed. > Maybe it's because I'm using a SMP kernel? (on a PIV HT cpu, though, not a real dual or more core). I'll try a debug build asap. Maurilio. -- __ | |