Diagnosing differences between bochs backend and Hyper-V backend
warwick-davis opened this issue · comments
warwick-davis commented
Is there any defined protocol or recommended method to figure out why a testcase may be crashing on Hyper-V but not on Bochs (nor on real CPU)?
I wanted to try and get a trace to see where they diverge but it is only supported on Bochs unfortunately. Any help would be appreciated.
Axel Souchet commented
One of the way I've done that before is by using the code coverage trace
(you get an address per basic block) and you can start comparing this with
a full RIP bochs trace. It's tedious but it should give you an idea where
execution diverges and why.
Cheers
…On Thu, Apr 27, 2023 at 12:05 AM warwick-davis ***@***.***> wrote:
Is there any defined protocol or recommended method to figure out why a
testcase may be crashing on Hyper-V but not on Bochs (nor on real CPU)?
I wanted to try and get a trace to see where they diverge but it is only
supported on Bochs unfortunately. Any help would be appreciated.
—
Reply to this email directly, view it on GitHub
<#160>, or unsubscribe
<https://github.com/notifications/unsubscribe-auth/AALIORMLGJWXEG42PNCDWZLXDILE7ANCNFSM6AAAAAAXNNM4E4>
.
You are receiving this because you are subscribed to this thread.Message
ID: ***@***.***>
warwick-davis commented
Unfortunately I had to make a new dump regardless and I ran into bigger issues, but useful for the future, thanks!