[Gambas-user] Bug with JIT and external functions

Jussi Lahtinen jussi.lahtinen at ...626...
Tue Jun 19 20:38:02 CEST 2012 

More bugs...

Make shared library (just dummy example [or just use binary that is
included with project source]);

#include <math.h>

int yEta(int x1, int y1, int x2, int y2)
{return (int) hypot(x1-x2,y1-y2);}


Then run the attached project.
With JIT, number of arguments is counted wrong.


Jussi








On 19 June 2012 21:02, Jussi Lahtinen <jussi.lahtinen at ...626...> wrote:

> Yes, it works now!
>
> Jussi
>
>
>
>
> On 18 June 2012 22:34, Emil Lenngren <emil.lenngren at ...626...> wrote:
>
>> Does it work in revision #4847?
>>
>> /Emil
>>
>> 2012/6/18 Jussi Lahtinen <jussi.lahtinen at ...626...>
>>
>> > Starting program: /usr/local/bin/gbx3
>> > [Thread debugging using libthread_db enabled]
>> > Using host libthread_db library
>> "/lib/x86_64-linux-gnu/libthread_db.so.1".
>> > gbx3: ExecutionEngine.cpp:165: void
>> > llvm::ExecutionEngine::addGlobalMapping(const llvm::GlobalValue*,
>> void*):
>> > Assertion `(CurVal == 0 || Addr == 0) && "GlobalMapping already
>> > established!"' failed.
>> >
>> > Program received signal SIGABRT, Aborted.
>> > 0x00007ffff7130445 in __GI_raise (sig=<optimized out>)
>> >    at ../nptl/sysdeps/unix/sysv/linux/raise.c:64
>> > 64    ../nptl/sysdeps/unix/sysv/linux/raise.c: No such file or
>> directory.
>> > (gdb) bt
>> > #0  0x00007ffff7130445 in __GI_raise (sig=<optimized out>)
>> >    at ../nptl/sysdeps/unix/sysv/linux/raise.c:64
>> > #1  0x00007ffff7133bab in __GI_abort () at abort.c:91
>> > #2  0x00007ffff712910e in __assert_fail_base (fmt=<optimized out>,
>> >    assertion=0x7fffed43c3b8 "(CurVal == 0 || Addr == 0) &&
>> \"GlobalMapping
>> > already established!\"", file=0x7fffed43c55b "ExecutionEngine.cpp",
>> >    line=<optimized out>, function=<optimized out>) at assert.c:94
>> > #3  0x00007ffff71291b2 in __GI___assert_fail (
>> >    assertion=0x7fffed43c3b8 "(CurVal == 0 || Addr == 0) &&
>> \"GlobalMapping
>> > already established!\"", file=0x7fffed43c55b "ExecutionEngine.cpp",
>> > line=165,
>> >    function=0x7fffed43d7a0 "void
>> > llvm::ExecutionEngine::addGlobalMapping(const llvm::GlobalValue*,
>> void*)")
>> > at assert.c:103
>> > #4  0x00007fffeca84aae in
>> > llvm::ExecutionEngine::addGlobalMapping(llvm::GlobalValue const*,
>> void*) ()
>> > from /usr/lib/libLLVM-3.2svn.so
>> > #5  0x00007fffedc85ec7 in register_global_symbol (address=0x40ebf6,
>> >    value=0x8b2500, name=...) at jit_codegen.cpp:242
>> > #6  register_global_symbol (name=..., value=0x8b2500, address=0x40ebf6)
>> >    at jit_codegen.cpp:240
>> > #7  0x00007fffedc8600e in get_global_function_real (
>> >    name=0x7fffedcb9e57 "CLASS_free", func=0x40ebf6, ret=<optimized out>,
>> >    args=<optimized out>, vararg=<optimized out>) at jit_codegen.cpp:276
>> > #8  0x00007fffedc8a9dc in operator() (this=<synthetic pointer>)
>> >    at jit_codegen.cpp:991
>> > ---Type <return> to continue, or q <return> to quit---
>> > #9  gen_if<unref_object_no_nullcheck(llvm::Value*)::<lambda()> >
>> (func=...,
>> >    cond=0x8999d0, if_name=<optimized out>, cont_name=<optimized out>)
>> >    at jit_codegen.cpp:562
>> > #10 unref_object_no_nullcheck (ptr=0x898ca8) at jit_codegen.cpp:993
>> > #11 0x00007fffedc9a44c in codegen_get_value (this=0x8abba0)
>> >    at jit_codegen.cpp:1566
>> > #12 PushPureObjectVariableExpression::codegen_get_value (this=0x8abba0)
>> >    at jit_codegen.cpp:1545
>> > #13 0x00007fffedca3f0d in SubrExpression::codegen_get_value
>> (this=0x86a450)
>> >    at jit_codegen.cpp:5898
>> > #14 0x00007fffedc8c21d in DropExpression::codegen (this=0x86a4a0)
>> >    at jit_codegen.cpp:1257
>> > #15 0x00007fffedca6fa7 in codegen_statements () at jit_codegen.cpp:855
>> > #16 JIT_codegen () at jit_codegen.cpp:6490
>> > #17 0x00007fffedcadc64 in JIT_compile_and_execute () at
>> jit_compile.cpp:119
>> > #18 0x000000000040a2fb in EXEC_jit_function_loop () at gbx_exec.c:871
>> > #19 0x000000000040ab22 in EXEC_function_real () at gbx_exec.c:862
>> > #20 0x000000000041e86d in raise_event (observer=<optimized out>,
>> >    object=<optimized out>, func_id=<optimized out>, nparam=<optimized
>> out>)
>> >    at gbx_api.c:711
>> > #21 0x000000000041efa6 in GB_Raise (object=0x8b95c8, event_id=16,
>> nparam=0)
>> >    at gbx_api.c:842
>> > #22 0x00007ffff5fce3d3 in gb_raise_button_Click (sender=<optimized out>)
>> >
>> > Jussi
>> >
>> >
>> >
>> > On 18 June 2012 22:08, Emil Lenngren <emil.lenngren at ...626...> wrote:
>> >
>> > > Hi. Can you please provide more information? A gdb backtrace or
>> > something,
>> > > because I get no errors at all. Everything seems to work for me ...
>> > >
>> > > /Emil
>> > >
>> > > 2012/6/18 Jussi Lahtinen <jussi.lahtinen at ...626...>
>> > >
>> > > > OK, I found it. See attachment.
>> > > >
>> > > > Jussi
>> > > >
>> > > >
>> > > >
>> > > >
>> > > > On 17 June 2012 23:36, Jussi Lahtinen <jussi.lahtinen at ...626...>
>> > wrote:
>> > > >
>> > > > > OK, that is fixed, and I cannot see any problems with compare
>> method
>> > > > > anymore.
>> > > > >
>> > > > > But with my big project, I got this with signal 6;
>> > > > > gbx3: ExecutionEngine.cpp:165: void
>> > > > > llvm::ExecutionEngine::addGlobalMapping(const llvm::GlobalValue*,
>> > > void*):
>> > > > > Assertion `(CurVal == 0 || Addr == 0) && "GlobalMapping already
>> > > > > established!"' failed.
>> > > > >
>> > > > > I'm out of time, but as soon as I have time I try to isolate code
>> > > causing
>> > > > > the problem.
>> > > > >
>> > > > > Jussi
>> > > > >
>> > > > >
>> > > > >
>> > > > > On 17 June 2012 20:24, Emil Lenngren <emil.lenngren at ...626...>
>> > wrote:
>> > > > >
>> > > > >> It should be fixed in revision #4843!
>> > > > >>
>> > > > >> /Emil
>> > > > >>
>> > > > >> 2012/6/17 Jussi Lahtinen <jussi.lahtinen at ...626...>
>> > > > >>
>> > > > >> > Yes, it really works now. And I found  a lot of errors by me.
>> > > > >> >
>> > > > >> > But for some reason JIT raises compare method when it
>> shouldn't,
>> > > > >> > and when I tried to isolate the problem I got crash (signal
>> 11).
>> > > > >> > This seems not to be related to compare method, so just ignore
>> > name
>> > > of
>> > > > >> the
>> > > > >> > attached project.
>> > > > >> > I will look it more closely when this is fixed.
>> > > > >> >
>> > > > >> > Jussi
>> > > > >> >
>> > > > >> >
>> > > > >> >
>> > > > >> >
>> > > > >> > On 17 June 2012 19:01, Emil Lenngren <emil.lenngren at ...626...>
>> > > wrote:
>> > > > >> >
>> > > > >> > > Yes, it works now :)
>> > > > >> > >
>> > > > >> > > 2012/6/17 Benoît Minisini <gambas at ...1...>
>> > > > >> > >
>> > > > >> > > > Le 17/06/2012 02:54, Jussi Lahtinen a écrit :
>> > > > >> > > > > Yeah, terminal gives other position...
>> > > > >> > > > > With my actual project, I got error message; wanted
>> string,
>> > > got
>> > > > >> > > > > blaablaaClass instead.
>> > > > >> > > > > And that error message comes from function (from other
>> > module
>> > > > >> without
>> > > > >> > > > Fast)
>> > > > >> > > > > which takes two objects are parameters.
>> > > > >> > > > >
>> > > > >> > > > > So either terminal isn't showing the real stack, or
>> there is
>> > > > >> > something
>> > > > >> > > > else
>> > > > >> > > > > wrong...
>> > > > >> > > > >
>> > > > >> > > > > But my time is up, I'll see it tomorrow (with latest
>> > > revision).
>> > > > >> > > > >
>> > > > >> > > > > Jussi
>> > > > >> > > > >
>> > > > >> > > > >
>> > > > >> > > >
>> > > > >> > > > The incorrect line number problem, as reported by the
>> debugger
>> > > > (and
>> > > > >> not
>> > > > >> > > > when you run the program normally!), should be fixed in
>> > revision
>> > > > >> #4841.
>> > > > >> > > >
>> > > > >> > > > Regards,
>> > > > >> > > >
>> > > > >> > > > --
>> > > > >> > > > Benoît Minisini
>> > > > >> > > >
>> > > > >> > > >
>> > > > >> > > >
>> > > > >> > >
>> > > > >> >
>> > > > >>
>> > > >
>> > >
>> >
>> ------------------------------------------------------------------------------
>> > > > >> > > > Live Security Virtual Conference
>> > > > >> > > > Exclusive live event will cover all the ways today's
>> security
>> > > and
>> > > > >> > > > threat landscape has changed and how IT managers can
>> respond.
>> > > > >> > Discussions
>> > > > >> > > > will include endpoint security, mobile security and the
>> latest
>> > > in
>> > > > >> > malware
>> > > > >> > > > threats.
>> > > > http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/
>> > > > >> > > > _______________________________________________
>> > > > >> > > > Gambas-user mailing list
>> > > > >> > > > Gambas-user at lists.sourceforge.net
>> > > > >> > > > https://lists.sourceforge.net/lists/listinfo/gambas-user
>> > > > >> > > >
>> > > > >> > >
>> > > > >> > >
>> > > > >> >
>> > > > >>
>> > > >
>> > >
>> >
>> ------------------------------------------------------------------------------
>> > > > >> > > Live Security Virtual Conference
>> > > > >> > > Exclusive live event will cover all the ways today's security
>> > and
>> > > > >> > > threat landscape has changed and how IT managers can respond.
>> > > > >> Discussions
>> > > > >> > > will include endpoint security, mobile security and the
>> latest
>> > in
>> > > > >> malware
>> > > > >> > > threats.
>> > > http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/
>> > > > >> > > _______________________________________________
>> > > > >> > > Gambas-user mailing list
>> > > > >> > > Gambas-user at lists.sourceforge.net
>> > > > >> > > https://lists.sourceforge.net/lists/listinfo/gambas-user
>> > > > >> > >
>> > > > >> >
>> > > > >> >
>> > > > >> >
>> > > > >>
>> > > >
>> > >
>> >
>> ------------------------------------------------------------------------------
>> > > > >> > Live Security Virtual Conference
>> > > > >> > Exclusive live event will cover all the ways today's security
>> and
>> > > > >> > threat landscape has changed and how IT managers can respond.
>> > > > >> Discussions
>> > > > >> > will include endpoint security, mobile security and the latest
>> in
>> > > > >> malware
>> > > > >> > threats.
>> > http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/
>> > > > >> > _______________________________________________
>> > > > >> > Gambas-user mailing list
>> > > > >> > Gambas-user at lists.sourceforge.net
>> > > > >> > https://lists.sourceforge.net/lists/listinfo/gambas-user
>> > > > >> >
>> > > > >> >
>> > > > >>
>> > > > >>
>> > > >
>> > >
>> >
>> ------------------------------------------------------------------------------
>> > > > >> Live Security Virtual Conference
>> > > > >> Exclusive live event will cover all the ways today's security and
>> > > > >> threat landscape has changed and how IT managers can respond.
>> > > > Discussions
>> > > > >> will include endpoint security, mobile security and the latest in
>> > > > malware
>> > > > >> threats.
>> http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/
>> > > > >> _______________________________________________
>> > > > >> Gambas-user mailing list
>> > > > >> Gambas-user at lists.sourceforge.net
>> > > > >> https://lists.sourceforge.net/lists/listinfo/gambas-user
>> > > > >>
>> > > > >
>> > > > >
>> > > >
>> > > >
>> > > >
>> > >
>> >
>> ------------------------------------------------------------------------------
>> > > > Live Security Virtual Conference
>> > > > Exclusive live event will cover all the ways today's security and
>> > > > threat landscape has changed and how IT managers can respond.
>> > Discussions
>> > > > will include endpoint security, mobile security and the latest in
>> > malware
>> > > > threats. http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/
>> > > > _______________________________________________
>> > > > Gambas-user mailing list
>> > > > Gambas-user at lists.sourceforge.net
>> > > > https://lists.sourceforge.net/lists/listinfo/gambas-user
>> > > >
>> > > >
>> > >
>> > >
>> >
>> ------------------------------------------------------------------------------
>> > > Live Security Virtual Conference
>> > > Exclusive live event will cover all the ways today's security and
>> > > threat landscape has changed and how IT managers can respond.
>> Discussions
>> > > will include endpoint security, mobile security and the latest in
>> malware
>> > > threats. http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/
>> > > _______________________________________________
>> > > Gambas-user mailing list
>> > > Gambas-user at lists.sourceforge.net
>> > > https://lists.sourceforge.net/lists/listinfo/gambas-user
>> > >
>> >
>> >
>> ------------------------------------------------------------------------------
>> > Live Security Virtual Conference
>> > Exclusive live event will cover all the ways today's security and
>> > threat landscape has changed and how IT managers can respond.
>> Discussions
>> > will include endpoint security, mobile security and the latest in
>> malware
>> > threats. http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/
>> > _______________________________________________
>> > Gambas-user mailing list
>> > Gambas-user at lists.sourceforge.net
>> > https://lists.sourceforge.net/lists/listinfo/gambas-user
>> >
>>
>> ------------------------------------------------------------------------------
>> Live Security Virtual Conference
>> Exclusive live event will cover all the ways today's security and
>> threat landscape has changed and how IT managers can respond. Discussions
>> will include endpoint security, mobile security and the latest in malware
>> threats. http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/
>> _______________________________________________
>> Gambas-user mailing list
>> Gambas-user at lists.sourceforge.net
>> https://lists.sourceforge.net/lists/listinfo/gambas-user
>>
>
>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: JITBug-0.0.4.tar.gz
Type: application/x-gzip
Size: 8081 bytes
Desc: not available
URL: <http://lists.gambas-basic.org/pipermail/user/attachments/20120619/a86d8256/attachment.bin>

More information about the User mailing list