This is the mail archive of the
mailing list for the systemtap project.
Re: djprobes status
On Fri, 2006-09-22 at 17:34 -0400, Satoshi Oshima wrote:
> Plan E: freeze_processes()/thow_processes()
> I realize that Plan C&D is highly archtecture dependent
> and Intel CPU's erratta may bring another problem.
> And someone pointed me out that current djprobes
> implementation is very hard to port to some other
> So We would like to try Plan E first. The code will be
> simpler and mainly architecture independent.
agreed. Plan E combined with "batching" would be my preference too,
because it allows the fastest implementation for probes. (probes would
not have to be aware of preemption or code modification at all, because
whatever modification is done to the code it will be in a completely
"quiet" state of the system.) This would speed up normal kprobes too.