Thread (20 messages) 20 messages, 3 authors, 2017-09-16

Re: [PATCH 2/5] powerpc/kprobes: Do not suppress instruction emulation if a single run failed

From: Naveen N. Rao <hidden>
Date: 2017-09-14 10:03:20

On 2017/09/14 02:45AM, Masami Hiramatsu wrote:
On Thu, 14 Sep 2017 12:08:07 +0530
"Naveen N. Rao" [off-list ref] wrote:
quoted
On 2017/09/13 04:53PM, Masami Hiramatsu wrote:
quoted
On Thu, 14 Sep 2017 02:50:33 +0530
"Naveen N. Rao" [off-list ref] wrote:
quoted
Currently, we disable instruction emulation if emulate_step() fails for
any reason. However, such failures could be transient and specific to a
particular run. Instead, only disable instruction emulation if we have
never been able to emulate this. If we had emulated this instruction
successfully at least once, then we single step only this probe hit and
continue to try emulating the instruction in subsequent probe hits.
Hmm, would this mean that the instruction is emulatable or not depends
on context? What kind of situation is considerable?
Yes, as an example, a load/store instruction can cause exceptions 
depending on the address. In some of those cases, we will have to single 
step the instruction, but we will be able to emulate in most scenarios.
OK, I got it.
Could you add this example as comment in the code so that readers can
easily understand?
Sure.

Thanks,
Naveen
Keyboard shortcuts
hback out one level
jnext message in thread
kprevious message in thread
ldrill in
Escclose help / fold thread tree
?toggle this help