On Fri, Jul 1, 2011 at 4:28 PM, Jakub Jermar <ja...@jermar.eu> wrote: > On 1.7.2011 16:24, Laurent Desnogues wrote: >> On Fri, Jul 1, 2011 at 4:11 PM, Jakub Jermar <ja...@jermar.eu> wrote: >> [...] >>> Actually, the testcase can be further reduced into: >>> >>> .global _start >>> >>> .text >>> >>> .space 0x20 >>> >>> _start: >>> set 110393, %i1 >>> set 0x40, %i2 >>> >>> cmp %i1, %i2 >>> udivx %g0, 1, %g0 >>> movgu %xcc, %i2, %i1 >>> cmp %i1, 512 >>> bgu %xcc, 0f >>> nop >>> >>> succ: >>> ta 0 >>> >>> fail: >>> 0: >>> ta 1 >>> >>> The presence of the `udivx` instruction seems to be essential. Even >>> though it has no effect on the computation, removing it will make the >>> testcase non-reproducible. >> >> Could you try to replace udivx with sdivx? It looks wrong too. > > Yeah, `sdivx` behaves the same wrt. the testcase.
Don't need to. Will send a qemu patch tonight. Great job! Once more, thanks a lot! -- Regards, Artyom Tarasenko solaris/sparc under qemu blog: http://tyom.blogspot.com/
