murph
November 24th, 2005, 13:07
I have two questions.
(1) I've scoured the documentation for a while, and it seems to me that there is no way to automatically execute a small set of instructions whenever a MEMORY ACCESS breakpoint is hit, as you can do with the conditional logging breakpoints that always break on a particular address. (This seems evident to me from the fact that setting these instructions-to-be-executed-upon-hit involves setting a NM_PLUGINCMD string that is firmly tied to a particular address.) Am I correct?
(2) Is there a way to tell, when back-tracing a program, that when we break on a certain instruction, that we have in fact jumped to that instruction as opposed to simply stepping to it from the immediately previous instruction? Like a flag in a register somewhere, some "fingerprint" that tells us we have taken a jump to this instruction?
(1) I've scoured the documentation for a while, and it seems to me that there is no way to automatically execute a small set of instructions whenever a MEMORY ACCESS breakpoint is hit, as you can do with the conditional logging breakpoints that always break on a particular address. (This seems evident to me from the fact that setting these instructions-to-be-executed-upon-hit involves setting a NM_PLUGINCMD string that is firmly tied to a particular address.) Am I correct?
(2) Is there a way to tell, when back-tracing a program, that when we break on a certain instruction, that we have in fact jumped to that instruction as opposed to simply stepping to it from the immediately previous instruction? Like a flag in a register somewhere, some "fingerprint" that tells us we have taken a jump to this instruction?