On Thu, 5 Oct 2006, Lei Jin wrote:
Hi mec,
Thank you for your answer.
Do you know how the memory are managed on a multicore system? Especially, does
the system aware the underlying memory structure and location?
I feel the page fault handling procedure is a little difficult to follow. And
the materials I can find online are old and do not provide detail. Do you have
any good reference?
Hmm, a starting point could be something like this:
#!/usr/sbin/dtrace -Fn
fbt::pagefault:entry /execname="mytestprogram"/ {self->t=1}
fbt::: /self->t/ {}
fbt::pagefault:return {self->t=0;exit();}
to get yourself a codeflow. Start your test program, kick off the dtrace,
watch what happens and then read the corresponding sourcecode ?
By letting your test program create different kinds of mappings and
triggering different kinds of faults, the above is an easy starting point
to come to terms with pagefault handling, and understand the segment
drivers wrt. to that.
Greetings,
FrankH.
_______________________________________________
opensolaris-code mailing list
[email protected]
http://mail.opensolaris.org/mailman/listinfo/opensolaris-code
