On Mon, Aug 14, 2017 at 3:28 AM, Dibyendu Majumdar <mobile@xxxxxxxxxxxxxxx> wrote: > lua_settop: > .L0: > <entry-point> > setge.32 %r2 <- %arg2, $0 > cbr %r2, .L6, .L2 > > .L6: > load.64 %r5 <- 0[VOID] > load.64 %r7 <- 8[VOID] <=========== Thanks for the report. So you mean here the arg1 has been optimized to void. I am able to duplicate this problem with rc5. Haven't have time to dig deeper. But it seems before the optimization, the first round of linearization has the output that arg1 was still there. Regard to this release. I don't think we have time to make all the IR correct for RC5. I suggest we make the 0.5.1 release this week. Then immediate start to work on 0.5.2, make the necessary fix there, then we can have RC1 for 0.5.2. Chris -- 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
