On Tue, 28 Jun 2011, junghun lee wrote:
Dear huangyongbing
That may true or false.
If you have 32bit machine, you can't allocate physical memory 4096mb (
http://comments.gmane.org/gmane.comp.emulators.m5.users/4865)
But if you have 64bit machine, you and me have same problems.
During I have debuging, I have find* the astonishing fact* that this problem
happen when I allocate 4,8,16,20GB.
That is, this means* four **multiple.*
*
*
*I have thought that allocation boundary is not allocated properly.
Maybe...*
*
*
Nilay Vaish and all
Please, could you fixed this problems.
And
Dear Nilay
According to your request, I have tried to print the values of arguments in
memcpy().
However,
I couldn't print exact value, but I surely find obj.m_data value is wrong at
some point
It means that is garbage value.
The reason of my thought is that this problem still happen when just print
values (m_data, obj.m_data, RubySystem::getBlockSizeBytes()) before
memcpy().
And this problem happen at point that I write code to print value of
obj.m_data,
the others are ok
Please help me.
Once the segmentation fault happens under GDB, can't you switch to stack
frame 1 and print the values for arguments passed to memcpy()?
--
Nilay
_______________________________________________
gem5-users mailing list
[email protected]
http://m5sim.org/cgi-bin/mailman/listinfo/gem5-users
