As you probably know by now, one can't look at a bug of this sort without a compilable test case. Andrew correctly pointed out that this optimization is affected by (for instance) inlining.
Hi Daniel, The function referenced is in a separate compilation unit and even if it were in the same unit, it is referenced multiple times and is sufficiently complex that inlining is highly improbable. Anyway, there is no code path that does not set the value and there is no code path that references the initial value, even if it were inlined. The warning was issued because "xxx" was used after (one of) the calls to the setting functions. If that is a bug with the flow analysis and warning generation, then I will construct a simple example that shows it and provide detailed compiler information. I confess to assuming that the analysis was expected to be beyond GCC's abilities. The examples regarding flow analysis abilities do not cover this scenario: int x; if (do_init) init_x(&x); else set_x(&x); use_x(x); Thank you again. Regards, Bruce
