>The 0x405be1 PC is probably the fallthrough target for a loop,
>jumped to only when the loop is complete
Sounds like a LA-PC (lookahead pc)


>Is instruction 0x405bdd a branch of some sort?
1254045000: system.switch_cpus.fetch: [tid:0]: Instruction PC 0x405bdd
(0) created [sn:1].
1254045000: system.switch_cpus.fetch: [tid:0]: Instruction is:
SUB_M_I : limm   t2d, 0x1

>If so, you would get
>different PC values for inst->pcState() and inst->branchTarget()[I think
>that's the right syntax, anyway].
While watching at this line:
1254046000: system.switch_cpus.decode: [tid:0]: Processing instruction
[sn:1] with PC (0x405bdd=>0x405be1).(0=>1)

there is no inst->branchTarget() to watch. Instead,
pc:
   _pc=0x405bdd
   _npc=0x405be1

predPC:
   _pc=0x405bdd
   _npc=0x405be1

-- 
--
// Naderan *Mahmood;
_______________________________________________
gem5-users mailing list
[email protected]
http://m5sim.org/cgi-bin/mailman/listinfo/gem5-users

Reply via email to