Mike, thank you for the example, but i think this is not the equivalent of the WinApi Messagebox.
The concerns that were brought forward were not about forwarding a string to the screen,
but rather about properly reading the return value by the main program when one of the buttons is clicked.
Michael, thanks for your reply.
It is clear to me that any operation that involves more than one instruction may be interrupted
by thread switching. From Pauls reply i learned that the operation A=B involves also two instructions.
Edwin, thanks. I will try that.
Paul, thanks. Your reply is very clear: the alignment issue is not present in PB.
But on the downside, the operation A = B does not involve one instruction, as i supposed, but two,
and may therefore be interrupted by thread switching.
Therefore it may be concluded that a 32 bit variable is not guaranteed put in place in one single
operation, as i supposed. This implies that synchronisation techniques are required at all times.
Arie Verheul
The concerns that were brought forward were not about forwarding a string to the screen,
but rather about properly reading the return value by the main program when one of the buttons is clicked.
Michael, thanks for your reply.
It is clear to me that any operation that involves more than one instruction may be interrupted
by thread switching. From Pauls reply i learned that the operation A=B involves also two instructions.
Edwin, thanks. I will try that.
Paul, thanks. Your reply is very clear: the alignment issue is not present in PB.
But on the downside, the operation A = B does not involve one instruction, as i supposed, but two,
and may therefore be interrupted by thread switching.
Therefore it may be concluded that a 32 bit variable is not guaranteed put in place in one single
operation, as i supposed. This implies that synchronisation techniques are required at all times.
Arie Verheul
Comment