Re: C++ Threads, what's the status quo?
Greg Herlihy wrote:
Jeff Koftinoff wrote:
With that code, the compiler is STILL permitted to re-order main() to
be:
int main()
{
count++;
StMutex lockMe( &gMutex );
}
There are no guarantees, since the compiler knows that StMutex does not
modify count. Even if you try to 'hide' the definition of StMutex, even
if it involves a kernel trap/syscall/etc - there is nothing in the c++
standard that forbids this optimization - and yes, this kind of
optimization is important to be able to have for non-mutex classes. So
how does the compiler know that this class is very special and is a
'multi-threaded sequence point'?
No, the compiler may possibly eliminate count++ (in some cases), but it
is not allowed to execute it before StMutex's constructor executes.
After all, there is a sequence point between the two statements:
"At certain specified points in the execution sequence called sequence
points, all side effects of previous evaluations shall be complete and
no side effects of subsequent evaluations shall have taken
place."[?1.9/7]
The above rule applies only to "actual code generated." The compiler is
not obligiated to generate code for every statement - statements that
have no effect on the observable state of the program can be
eliminated. But otherwise, unless the statement is eliminated, then it
must execute as ?1.9/7 requires - that is, in a specific order
relative to the program's other statements. Otherwise, if the compiler
could reorder sequence poitnts at will, it would not be possible to
write single-threaded C++ programs and be sure that they would run
correctly.
Don't forget the "as if" rule: if a conforming program can't tell the
difference, the compiler is free to do anything, so long as the program
acts as if all the rules had been followed. Since, by assumption, the
lock is not affected by and does not affect the value of count, the
compiler is free to reorder the calls. A conforming program will have
exactly the same behavior both ways. It's only the non-comforming ones,
with a sneak path to modify count from another thread, that don't
necessarily act the same.
Incidentally, you should mention that your paraphrase "acual code
generated" is from a footnote, not from the normative text of the standard.
--
-- Pete
Roundhouse Consulting, Ltd. (www.versatilecoding.com)
Author of "The Standard C++ Library Extensions: a Tutorial and
Reference." (www.petebecker.com/tr1book)
[ See http://www.gotw.ca/resources/clcm.htm for info about ]
[ comp.lang.c++.moderated. First time posters: Do this! ]