On Friday 13 November 2009 15:17, you wrote:> > Are these known to be broken right now? I get failure when using either. > > > > $ llc -march=arm -print-machineinstrs hw.bc > > Seems due to David's patches.Ok, it's faulting in SlotTracker with what looks like a bad Function. One of the Argument values is corrupted. I'm not abdicating responsibility, but at the moment I don't see the connection to my patches. Anton, are you seeing something specific as to the cause? I will keep investigating. -Dave
On Friday 13 November 2009 15:49, David Greene wrote:> On Friday 13 November 2009 15:17, you wrote: > > > Are these known to be broken right now? I get failure when using > > > either. > > > > > > $ llc -march=arm -print-machineinstrs hw.bc > > > > Seems due to David's patches. > > Ok, it's faulting in SlotTracker with what looks like a bad Function. One > of the Argument values is corrupted. > > I'm not abdicating responsibility, but at the moment I don't see the > connection to my patches. Anton, are you seeing something specific > as to the cause? > > I will keep investigating.Somehow a MachineIntr memoperand is pointing to an Instruction. That's not supposed to happen, is it? I saw the symptom change after updating to ToT so I suspect this is a rather subtle memory corruption problem. -Dave
On Nov 13, 2009, at 1:49 PM, David Greene wrote:> On Friday 13 November 2009 15:17, you wrote: >>> Are these known to be broken right now? I get failure when using either. >>> >>> $ llc -march=arm -print-machineinstrs hw.bc >> >> Seems due to David's patches. > > Ok, it's faulting in SlotTracker with what looks like a bad Function. One of > the Argument values is corrupted. > > I'm not abdicating responsibility, but at the moment I don't see the > connection to my patches. Anton, are you seeing something specific > as to the cause? > > I will keep investigating.The cause was isa<Instruction> was returning true for the new FixedStackPseudoSourceValue values. I fixed it on trunk, and added a comment to hopefully protect against others making this mistake in the future. Dan
On Friday 13 November 2009 15:49, David Greene wrote:> On Friday 13 November 2009 15:17, you wrote: > > > Are these known to be broken right now? I get failure when using > > > either. > > > > > > $ llc -march=arm -print-machineinstrs hw.bc > > > > Seems due to David's patches. > > Ok, it's faulting in SlotTracker with what looks like a bad Function. One > of the Argument values is corrupted. > > I'm not abdicating responsibility, but at the moment I don't see the > connection to my patches. Anton, are you seeing something specific > as to the cause? > > I will keep investigating.Dan just fixed it. I was just about to the same point he got to. :) -Dave
On Friday 13 November 2009 16:28, Dan Gohman wrote:> On Nov 13, 2009, at 1:49 PM, David Greene wrote: > > On Friday 13 November 2009 15:17, you wrote: > >>> Are these known to be broken right now? I get failure when using > >>> either. > >>> > >>> $ llc -march=arm -print-machineinstrs hw.bc > >> > >> Seems due to David's patches. > > > > Ok, it's faulting in SlotTracker with what looks like a bad Function. > > One of the Argument values is corrupted. > > > > I'm not abdicating responsibility, but at the moment I don't see the > > connection to my patches. Anton, are you seeing something specific > > as to the cause? > > > > I will keep investigating. > > The cause was isa<Instruction> was returning true for the new > FixedStackPseudoSourceValue values. I fixed it on trunk, and > added a comment to hopefully protect against others making > this mistake in the future.Is there any value is making isa<>, cast<> and friends a full dynamic_cast<> in paranoid build mode? It could help identify problems like this. -Dave