On Sun, Jun 17, 2018 at 02:23:53AM +0100, Dibyendu Majumdar wrote:
> Hi,
>
> I am facing a problem with following:
>
> C input:
> long long strlen(const char *s)
> {
> long long n = 0;
> while (s[n] != 0)
> n++;
> return n;
> }
>
> Here is part extract of the linear output from dmr_C - this is the IR
> before any simplifications.
>
> .L2:
> br .L4
>
> .L3:
> load.64 %r9 <- 0[n]
> phisrc.64 %phi1(return) <- %r9
> br .L5
>
> .L5:
> phi.64 %r10 <- %phi1(return)
> ret.64 %r9
>
> Problem is with OP_RET as it should I think return %r10 nor %r9.
>
> I was wondering is this is already fixed somewhere?
Not really a bug so not really fixed but annoying enough that I
removed this optimization in my tree. See:
https://github.com/lucvoo/sparse/commit/1609176c9
-- Luc
--
To unsubscribe from this list: send the line "unsubscribe linux-sparse" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
