STEP_SKIPS_DELAY question, sort of

[Orjan Friberg]

Andrew Cagney wrote:
Ok,

When given:

  N+0: branch foo
  N+2: nop
PC=N (or N+1)
SPC=N+2
would this result in a double single-step leaving PC==foo?  If it does 
then you've at least got STEP_SKIPS_DELAY like behavior.
Actually, both when PC=N (at the branch instruction) and when PC=N+1 (in 
the delay slot) a *single* single-step will leave us at PC==foo - the 
single-step exception fires after the instruction pointed out by SPC has 
been executed.

I'll go and have a closer look at STEP_SKIPS_DELAY.

--
Orjan Friberg
Axis Communications