17 Oct
2008
17 Oct
'08
6:15 p.m.
On Fri, Oct 17, 2008 at 02:33:00PM -0400, Paul Coccoli wrote:
There's no chance that a compiler optimization is
hiding the bug,
because we haven't turned on optimization.
But there is nothing that stops the compiler from writing
read_ptr any number of times it wants unless you make that
variable volatile.
HOWEVER, if you go back
to the original version and REMOVE the volatile qualifiers, then
re-compile with -O2, NOW a compiler optimization is hiding the bug
(and the test passes)! The rb->read_ptr is presumably cached in a
register and hence only modified once (giving the same affect as the
patch).
Indeed. The original version really invited the problem,
by *forcing* the compiler to write read_ptr twice.
Ciao,
--
FA
Laboratorio di Acustica ed Elettroacustica
Parma, Italia
Lascia la spina, cogli la rosa.