Re: Re: [PATCH v2 0/9] Introduce Data Access MONitor (DAMON)

[sjpark]

On Tue, 28 Jan 2020 05:20:35 -0500 Qian Cai <cai@xxxxxx> wrote:

> 
> 
> > On Jan 28, 2020, at 3:58 AM, sjpark@xxxxxxxxxx wrote:
> > 
> > This patchset introduces a new kernel module for practical monitoring of data
> > accesses, namely DAMON.
> > 
> > The patches are organized in the following sequence.  The first four patches
> > implements the core logic of DAMON one by one.  After that, the fifth patch
> > implements DAMON's debugfs interface for users.  To provide a minimal reference
> > to the low level interface and for more convenient use/tests of the DAMON, the
> > sixth patch implements an user space tool.  The seventh patch adds a document
> > for administrators of DAMON, and the eightth patch provides DAMON's kunit
> > tests.  Finally, the ninth patch implements a tracepoint for DAMON.  As the
> > tracepoint prints every monitoring results, it will be easily integrated with
> > other tracers supporting tracepoints including perf.
> 
> I am a bit surprised that this patchset did not include perf maintainers which makes me wonder if there is any attempt to discuss first if we actually need a whole new subsystem for it or a existing tool can be enhanced.

For the comments from perf maintainers, I added Steven Rostedt and Arnaldo
Carvalho de Melo first, but I might missed someone.  If you recommend some more
people, I will add them to recipients.

I made DAMON as a new subsystem because I think no existing subsystem fits well
to be a base of DAMON, due to DAMON's unique goals and mechanisms described
below in the original cover letter.

The existing subsystem that most similar to DAMON might be 'mm/page_idle.c'.
However, there are many conceptual differences with DAMON.  One biggest
difference I think is the target.  'page_idle' deals with physical page frames
while DAMON deals with virtual address of specific processes.

Nevertheless, if you have some different opinion, please let me know.


Thanks,
SeongJae Park