Re: [PATCH 07/10] Increase compressed page size threshold

[Pekka Enberg]

Hi Nitin,

On Mon, Aug 9, 2010 at 8:26 PM, Nitin Gupta <ngupta@xxxxxxxxxx> wrote:
> Compression takes much more time than decompression. So, its quite
> wasteful in terms of both CPU cycles and memory usage to have a very
> low compressed page size threshold and thereby storing such not-so-well
> compressible pages as-is (uncompressed). So, increasing it from
> PAGE_SIZE/2 to PAGE_SIZE/8*7. A low threshold was useful when we had
> "backing swap" support where we could forward such pages to the backing
> device (applicable only when zram was used as swap disk).
>
> It is not yet configurable through sysfs but may be exported in future,
> along with threshold for average compression ratio.
>
> Signed-off-by: Nitin Gupta <ngupta@xxxxxxxxxx>

The description makes sense but lacks any real data. What kind of
workloads did you test this with? Where does it help most? How much?

                        Pekka
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/