Re: [RFC PATCH v5 16/16] dcache: Add CONFIG_DCACHE_SMO

[Roman Gushchin <guro@fb.com>]

On Wed, May 29, 2019 at 01:54:06PM +1000, Tobin C. Harding wrote:
> On Tue, May 21, 2019 at 02:05:38AM +0000, Roman Gushchin wrote:
> > On Tue, May 21, 2019 at 11:31:18AM +1000, Tobin C. Harding wrote:
> > > On Tue, May 21, 2019 at 12:57:47AM +0000, Roman Gushchin wrote:
> > > > On Mon, May 20, 2019 at 03:40:17PM +1000, Tobin C. Harding wrote:
> > > > > In an attempt to make the SMO patchset as non-invasive as possible add a
> > > > > config option CONFIG_DCACHE_SMO (under "Memory Management options") for
> > > > > enabling SMO for the DCACHE.  Whithout this option dcache constructor is
> > > > > used but no other code is built in, with this option enabled slab
> > > > > mobility is enabled and the isolate/migrate functions are built in.
> > > > > 
> > > > > Add CONFIG_DCACHE_SMO to guard the partial shrinking of the dcache via
> > > > > Slab Movable Objects infrastructure.
> > > > 
> > > > Hm, isn't it better to make it a static branch? Or basically anything
> > > > that allows switching on the fly?
> > > 
> > > If that is wanted, turning SMO on and off per cache, we can probably do
> > > this in the SMO code in SLUB.
> > 
> > Not necessarily per cache, but without recompiling the kernel.
> > > 
> > > > It seems that the cost of just building it in shouldn't be that high.
> > > > And the question if the defragmentation worth the trouble is so much
> > > > easier to answer if it's possible to turn it on and off without rebooting.
> > > 
> > > If the question is 'is defragmentation worth the trouble for the
> > > dcache', I'm not sure having SMO turned off helps answer that question.
> > > If one doesn't shrink the dentry cache there should be very little
> > > overhead in having SMO enabled.  So if one wants to explore this
> > > question then they can turn on the config option.  Please correct me if
> > > I'm wrong.
> > 
> > The problem with a config option is that it's hard to switch over.
> > 
> > So just to test your changes in production a new kernel should be built,
> > tested and rolled out to a representative set of machines (which can be
> > measured in thousands of machines). Then if results are questionable,
> > it should be rolled back.
> > 
> > What you're actually guarding is the kmem_cache_setup_mobility() call,
> > which can be perfectly avoided using a boot option, for example. Turning
> > it on and off completely dynamic isn't that hard too.
> 
> Hi Roman,
> 
> I've added a boot parameter to SLUB so that admins can enable/disable
> SMO at boot time system wide.  Then for each object that implements SMO
> (currently XArray and dcache) I've also added a boot parameter to
> enable/disable SMO for that cache specifically (these depend on SMO
> being enabled system wide).
> 
> All three boot parameters default to 'off', I've added a config option
> to default each to 'on'.
> 
> I've got a little more testing to do on another part of the set then the
> PATCH version is coming at you :)
> 
> This is more a courtesy email than a request for comment, but please
> feel free to shout if you don't like the method outlined above.
> 
> Fully dynamic config is not currently possible because currently the SMO
> implementation does not support disabling mobility for a cache once it
> is turned on, a bit of extra logic would need to be added and some state
> stored - I'm not sure it warrants it ATM but that can be easily added
> later if wanted.  Maybe Christoph will give his opinion on this.

Perfect!

Thanks.