RFR (S): 8034868 Extract G1 From Card Cache into separate class

[Jon Masamitsu]

Thomas,

Change looks correct.

Would you consider  "contains_or_replace()" as a better name for
"contains_and_update()"?

Did you consider dropping the "_from_card" from the
variable names

   53   static int** _from_card_cache;
   54   static uint _from_card_cache_max_regions;
   55   static size_t _from_card_cache_mem_size;


You dropped the "_from_card_cache" from some names.
"init_from_card_cache" became "initialize" for example.

That's all.  Small points so I'm fine with whatever you decide.

Jon


On 3/3/14 3:49 AM, Thomas Schatzl wrote:
> Hi all,
>
>    can I get reviews for the following small change? It extracts the from
> card cache previously embedded into the OtherRegionsTable into a
> separate class, named FromCardCache.
>
> This cleans up the remembered set somewhat.
>
> While looking through the code I found that the remembered set code
> still using int's for thread id's. I did not clean that up here, created
> a new CR instead (https://bugs.openjdk.java.net/browse/JDK-8036116)
> since that change would ripple through to quite a few places.
>
> Webrev:
> http://cr.openjdk.java.net/~tschatzl/8034868/webrev/
>
> CR:
> https://bugs.openjdk.java.net/browse/JDK-8034868
>
> Testing:
> JPRT
>
> Thanks,
>    Thomas
>
>

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://mail.openjdk.org/pipermail/hotspot-gc-dev/attachments/20140303/ccbc2823/attachment.htm>