[prev in list] [next in list] [prev in thread] [next in thread] 

List:       openjdk-hotspot-gc-dev
Subject:    ZapUnusedHeapArea
From:       bengt.rutisson () oracle ! com (Bengt Rutisson)
Date:       2012-03-28 11:23:31
Message-ID: 4F72F4B3.2080106 () oracle ! com
[Download RAW message or body]


Hi Tom,

I am all for making this a product flag. I just had a quick look at the 
webrev. I think there are two #ifndef ASSERT statements in 
MutableNUMASpace::ensure_parsability() that also should be removed.

Otherwise it looks good.

Bengt

On 2012-03-27 20:16, Tom Rodriguez wrote:
> Is there any interest in adding product support for ZapUnusedHeapArea?  I \
> implemented it while debugging what turned out to be a card marking issue in the \
> compiler but I figured I'd ask before blowing it away.  The webrev is at \
> http://cr.openjdk.java.net/~never/zapunused.  It's a moderately useful flag when \
> debugging stale oop issues with the compiler since it causes you to die pretty much \
> at the first use of the stale pointer. 
> tom


[prev in list] [next in list] [prev in thread] [next in thread]