Discussion:
[PATCH] cpuset confine pdflush to its cpuset
Hirokazu Takahashi
2005-10-24 05:52:58 UTC
Permalink
Hi Paul,

I realized CPUSETS has another problem around pdflush.

Some cpuset may make most of pages in it dirty, while the others don't.
In this case, pdflush may not start since the ratio of the dirty pages
in the box may be less than the watermark, which is defined globally.
This may probably make it hard to allocate pages from the cpuset
or the nodes it depends on. This wouldn't be good for NUMA machine
without cpusets either.

Do you have any plans about it?
This patch keeps pdflush daemons on the same cpuset as their
parent, the kthread daemon.
Some large NUMA configurations put as much as they can of
kernel threads and other classic Unix load in what's called a
bootcpuset, keeping the rest of the system free for dedicated
jobs.
This effort is thwarted by pdflush, which dynamically destroys
and recreates pdflush daemons depending on load.
Thanks,
Hirokazu Takahashi.
Paul Jackson
2005-10-24 06:32:37 UTC
Permalink
Post by Hirokazu Takahashi
I realized CPUSETS has another problem around pdflush.
Excellent observation. I had not realized this.

Thank-you for pointing it out.

I don't have plans. Do you have any suggestions?

( Anyone know what the "pd" stands for in pdflush ?? )
--
I won't rest till it's the best ...
Programmer, Linux Scalability
Paul Jackson <***@sgi.com> 1.925.600.0401
Andrew Morton
2005-10-24 06:40:32 UTC
Permalink
Post by Paul Jackson
Post by Hirokazu Takahashi
I realized CPUSETS has another problem around pdflush.
Excellent observation. I had not realized this.
Thank-you for pointing it out.
I don't have plans. Do you have any suggestions?
Per-zone dirty thresholds (quite messy), per-zone writeback (horrific,
linear searches or data structure proliferation everywhere).

Let's see a (serious) worload/testcase first, hey? vmscan.c writeback off
the LRU is a bit slow, but we should be able to make it suffice.
Post by Paul Jackson
( Anyone know what the "pd" stands for in pdflush ?? )
"page dirty"? It's what bdflush became when writeback went from
being block-based to being page-based.
Paul Jackson
2005-10-24 06:49:18 UTC
Permalink
Post by Andrew Morton
Let's see a (serious) worload/testcase first, hey?
A reasonable request.
Post by Andrew Morton
Post by Paul Jackson
( Anyone know what the "pd" stands for in pdflush ?? )
"page dirty"? It's what bdflush became ...
Ah - thanks.
--
I won't rest till it's the best ...
Programmer, Linux Scalability
Paul Jackson <***@sgi.com> 1.925.600.0401
Hirokazu Takahashi
2005-10-24 07:13:26 UTC
Permalink
Hi Paul,
Post by Paul Jackson
Post by Andrew Morton
Let's see a (serious) worload/testcase first, hey?
A reasonable request.
Can you do this?
I think you may probably use a large NUMA machine.
Post by Paul Jackson
Post by Andrew Morton
Post by Paul Jackson
( Anyone know what the "pd" stands for in pdflush ?? )
"page dirty"? It's what bdflush became ...
Ah - thanks.
Paul Jackson
2005-10-24 07:37:04 UTC
Permalink
Post by Hirokazu Takahashi
Post by Paul Jackson
A reasonable request.
Can you do this?
I think you may probably use a large NUMA machine.
In theory, yes. I certainly have access to large NUMA machines.

However, it is likely not a priority for me. My focus is on work that
will benefit workloads that do not depend on pdflush (except to want to
be sure that pdflush is -not- running in a cpuset containing a dedicated
job.)

That seems to keep me busy enough (and keep my employer paying me),
so I might never get to this problem. I might, but the odds are
not good.

Sorry.
--
I won't rest till it's the best ...
Programmer, Linux Scalability
Paul Jackson <***@sgi.com> 1.925.600.0401
Loading...