EXCEPTION_ACCESS_VIOLATION and others stories

EXCEPTION_ACCESS_VIOLATION and others stories

[Etienne Baudin]

EXCEPTION_ACCESS_VIOLATION
    Appears when I compile with
<<gnatmake -gnatpn -O2 -mpentium -funroll-loops -fomit-frame-pointer %1
-largs -s e:\gllib\opengl32.lib e:\gllib\glu32.lib e:\gllib\libglut32.a>>

with
<<gnatmake %1
-largs -s e:\gllib\opengl32.lib e:\gllib\glu32.lib e:\gllib\libglut32.a>>
there isn't problem

else : is "tab(1..3):=anothertab(1..3)" faster than
tab(1):=anothertab(1)
tab(2):=anothertab(2)
tab(3):=anothertab(3) or is it the same execution time ?

Then, does anyone know this compilar message and could tell me what is wrong
?
<<access-to-variable designates constant>>

Thanks
Etienne Baudin

EXCEPTION_ACCESS_VIOLATION and others stories

[Etienne Baudin]

EXCEPTION_ACCESS_VIOLATION
    Appears when I compile with
<<gnatmake -gnatpn -O2 -mpentium -funroll-loops -fomit-frame-pointer %1
-largs -s e:\gllib\opengl32.lib e:\gllib\glu32.lib e:\gllib\libglut32.a>>

with
<<gnatmake %1
-largs -s e:\gllib\opengl32.lib e:\gllib\glu32.lib e:\gllib\libglut32.a>>
there isn't problem

else : is "tab(1..3):=anothertab(1..3)" faster than
tab(1):=anothertab(1)
tab(2):=anothertab(2)
tab(3):=anothertab(3) or is it the same execution time ?

Then, does anyone know this compilar message and could tell me what is wrong
?
<<access-to-variable designates constant>>

Thanks
Etienne Baudin

Re: EXCEPTION_ACCESS_VIOLATION and others stories

[Stephen Leake]

"Etienne Baudin" <etienne.baudin@wanadoo.fr> writes:

> EXCEPTION_ACCESS_VIOLATION
>     Appears when I compile with
> <<gnatmake -gnatpn -O2 -mpentium -funroll-loops -fomit-frame-pointer %1
> -largs -s e:\gllib\opengl32.lib e:\gllib\glu32.lib e:\gllib\libglut32.a>>
> 
> with
> <<gnatmake %1
> -largs -s e:\gllib\opengl32.lib e:\gllib\glu32.lib e:\gllib\libglut32.a>>
> there isn't problem

-fomit-frame-pointer is probably dangerous with GNAT; why are you
 specifying it?

> else : is "tab(1..3):=anothertab(1..3)" faster than
> tab(1):=anothertab(1) tab(2):=anothertab(2) tab(3):=anothertab(3) or
> is it the same execution time ?

Time it on your machine and your compiler and your set of options; no
other way to know.

> Then, does anyone know this compilar message and could tell me what
> is wrong ? <<access-to-variable designates constant>>

It means you have an object of an access-to-variable type that you are
trying to make point to a constant (probably an "in" parameter). This
is illegal, because later code could then try to change the constant
by assigning via the access object. If it is an "in" parameter, change
it to either an "in out" parameter or an "access" parameter, or
redefine the subprogram called to take an access-to-constant type.

-- 
-- Stephe