[
https://issues.apache.org/jira/browse/HBASE-16651?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15502932#comment-15502932
]
Anoop Sam John commented on HBASE-16651:
----------------------------------------
bq. it was told that going forward the config of having combinedMode= false
will be removed
Where is that comment/discussion?
Ya as per the cur code, we need this return to victimHandler also.
bq. Hope in our code (read path) we don't assume any where that since we get a
block from L1 it can be surely treated as a EXCLUSIVE mode block?
What is the issue even if that is assumed? The returnBlock call will always
happen. Now after this patch, the call come to L1 cache and it passed to L2
victim cache. There we have the SHARED mode check before deciding whether to
decr a counter.
bq.One place to check is that prefetch executore in Storefile open where the
blocks are read and we don't release the blocks. May be we are handling
correctly
The getBlocks calls in prefetch executor is doing the return.
{code}
HFileBlock block = readBlock(offset, onDiskSize, true, false, false, false,
null, null);
// Need not update the current block. Ideally here the readBlock
won't find the
// block in cache. We call this readBlock so that block data is
read from FS and
// cached in BC. So there is no reference count increment that
happens here.
// The return will ideally be a noop because the block is not of
MemoryType SHARED.
returnBlock(block);
{code}
> LRUBlockCache#returnBlock should try return block to Victim Handler L2 cache.
> -----------------------------------------------------------------------------
>
> Key: HBASE-16651
> URL: https://issues.apache.org/jira/browse/HBASE-16651
> Project: HBase
> Issue Type: Sub-task
> Components: regionserver, Scanners
> Affects Versions: 2.0.0
> Reporter: Anoop Sam John
> Assignee: Anoop Sam John
> Fix For: 2.0.0
>
> Attachments: HBASE-16651.patch
>
>
> In case of L1 and L2 cache usage with combinedMode = false, L2 is used as a
> victim handler cache for L1 cache. When a getBlock() request comes, L1 will
> see if block is in it and if not it will try to provide the block from L2
> cache. In such a case, the return block must return the block to L2 cache and
> count down the ref count for the block. But right now we just ignore the
> returnBlock call in LRUCache
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
