drag chan wrote:
> my question is why every constructor/destructor of class A have a
> couple of same code blocks in the object file?
The ABI requires two separate versions of the ctor and dtor, one for
when the object being initialized is a base class and one where it's
not. Because you used -C
hi list,
I'm using g++ to compile my project under linux. When checking c++
object files, I have a question about it.
there is a example:
--
class A {
public:
A();
A(int _i);
~A();
void hello();
private:
int i;
};
A::A()
{
i = 0;
}
A::A(int _i)
{
Hi,
I am not sure if these is due to some incorrect setting on my work environment.
I have a C++ server which in turn spawns the child process to do the execution
of actual task. Using Gcov, I am getting the code coverage of the main
server(listener) process but the child processes are not gett
Hi Peter,
Peter O'Gorman wrote:
Andreas Tobler wrote:
Jack Howarth wrote:
On i686-apple-darwin9, I am seeing massive regressions in the libjava
testsuite in revision 140713 compared to my previous test on 20080925.
Since the libtool updates went in (which would see the likely culprit),
we w
Andreas Tobler wrote:
> Jack Howarth wrote:
>>On i686-apple-darwin9, I am seeing massive regressions in the libjava
>> testsuite in revision 140713 compared to my previous test on 20080925.
>> Since the libtool updates went in (which would see the likely culprit),
>> we went from zero to 835 un
Jack Howarth wrote:
On i686-apple-darwin9, I am seeing massive regressions in the libjava
testsuite in revision 140713 compared to my previous test on 20080925.
Since the libtool updates went in (which would see the likely culprit),
we went from zero to 835 unexpected failures.
http://gcc.gnu
These new libjava failures all seem to be of the form...
set_ld_library_path_env_vars:
ld_library_path=.:/sw/src/fink.build/gcc44-4.3.999-20080926/darwin_objdir/i686-apple-darwin9/./libjava/.libs
Executing on host:
/sw/src/fink.build/gcc44-4.3.999-20080926/darwin_objdir/i686-apple-darwin9/lib
On i686-apple-darwin9, I am seeing massive regressions in the libjava
testsuite in revision 140713 compared to my previous test on 20080925.
Since the libtool updates went in (which would see the likely culprit),
we went from zero to 835 unexpected failures.
http://gcc.gnu.org/ml/gcc-testresult
I've a question about the way IRA accumulates information from
nested regions into parent regions. ira-int.h says:
/* Accumulated usage references of the allocno. Here and below,
word 'accumulated' means info for given region and all nested
subregions. In this case, 'accumulated' me
