Volatile Ranges

Sorry for being late here.

I wanted to propose some further discussion on the volatile ranges concept.

Basically trying to sort out a coherent story around:

* My attempts at volatile ranges for shared tmpfs files (similar functionality as Android's ashmem provides)

* Minchan's volatile ranges for anonymous memory

* How to track page volatility & purged state (via VMAs vs file address_space)

* Purged data semantics (ie: Mozilla's request for SIGBUS on purged data access vs zero fill)

* Aging anonymous pages in swapless systems

thanks
-john

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in the body to [email protected]. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"[email protected]"> [email protected] </a>

Hi John,

FYI,
I have a plan to redesign anon volatile range by some reasons. In old version, I have been working with Jason who is jemalloc author and he has a very interested in anon volatile.

I saw 2x faster with certain webserver workload with jemalloc which is tweaked by Jason and me. Even, I have a plan to enhance it much higher. If this topic is selected and I will be there,
I will share it with pleasure.

Thanks.

Kind regards,
Minchan Kim

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to [email protected].  For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"[email protected]"> [email protected] </a>