Messages in this thread | | | Date | Wed, 8 Jan 1997 19:19:53 +0100 (MET) | From | Ingo Molnar <> | Subject | Re: Since no one else has stepped forward: 'ZeroD' patch |
| |
On Wed, 8 Jan 1997, Nigel Metheringham wrote:
> } Ingo Molnar <mingo@pc5829.hil.siemens.at> writes: > } > } > Here is a patch that implements 'ZeroD', a simple kernel thread that uses > } > up CPU idle time to keep a pool of pre-zeroed pages around. Almost > } > anything that doesnt burn 100% CPU time all the time should benefit from > } > this. > } > } Hmm, did you check that this hasn't a bad effect on the cache memory when > } a cpu intensive process has to sleep for short times? > } A /proc/sys interface to tune it would be nice too. > > This would also presumably affect the heat production of the CPU - > especially on hot Cyrixs! > > Does it spend all the idle time doing this, or just a part??
no, just part of it. There is a limited queue of prezeroed pages currently. So the current typical way of operation is that some system activity eats 1-2-3 zero pages, zerod wakes up whenever there is some idle time left and zeroes some 1-2-3 new pages.
i dont even know if it makes sense at all :) This is an experimental patch for an experimental kernel :))
for the cache subsystem thing: if you think about it, it generates slightly >faster< general system operation, mainly because all the zeroing done by the daemon (or lets it call a pinguin :) has to happen anyways. And we should rather do it when the system is idle.
-- mingo
| |