Messages in this thread | | | From | Dan Streetman <> | Date | Sat, 23 Nov 2013 15:35:35 -0500 | Subject | Re: [PATCH v3] mm/zswap: change zswap to writethrough cache |
| |
On Fri, Nov 22, 2013 at 9:37 PM, Weijie Yang <weijie.yang.kh@gmail.com> wrote: > Hello Dan, > > On Sat, Nov 23, 2013 at 6:10 AM, Dan Streetman <ddstreet@ieee.org> wrote: >> Currently, zswap is writeback cache; stored pages are not sent >> to swap disk, and when zswap wants to evict old pages it must >> first write them back to swap cache/disk manually. This avoids >> swap out disk I/O up front, but only moves that disk I/O to >> the writeback case (for pages that are evicted), and adds the >> overhead of having to uncompress the evicted pages, and adds the >> need for an additional free page (to store the uncompressed page) >> at a time of likely high memory pressure. Additionally, being >> writeback adds complexity to zswap by having to perform the >> writeback on page eviction. >> >> This changes zswap to writethrough cache by enabling >> frontswap_writethrough() before registering, so that any >> successful page store will also be written to swap disk. All the >> writeback code is removed since it is no longer needed, and the >> only operation during a page eviction is now to remove the entry >> from the tree and free it. > > I agree with Seth, It is not good to embedded device. > May be we can find its place in others like server. > I guess it is good to medium workload when swap io is not frequent. > > My suggestion is would you please make it configurable so that user > can choice to use writethrough or writeback mode? >
Having to support both significantly increases complexity and I think would make further improvements more difficult. My opinion is the writeback code should be removed. Is there anyone else who thinks both should be available by a param?
| |