RFR (XXS): 8134856: Incorrect use of PLAB::min_size() in MaxPLABSizeBounds

[Thomas Schatzl]

Hi,

On Tue, 2015-09-01 at 12:31 -0400, Tom Benson wrote:
> Hi Thomas,
> Code change looks good to me.  The hardcoded knowledge of min PLAB size 
> to min TLAB size ratio in the test seems unfortunate, because if it 
> changes, the test will still work, but not be checking anything useful. 
>    (Tho honestly, I'm not really sure why a test is needed for this 
> particular change...)

  by ratio, you mean the calculation of HeapWordSize in bytes? I think
that ratio will stay in the future.

However I think it is pretty bad that the *PLABSize are in HeapWords but
the *TLABSize options are in bytes...

Thanks,
  Thomas

> Tom
> 
> 
> On 9/1/2015 11:59 AM, Thomas Schatzl wrote:
> > Hi all,
> >
> >    can I have reviews for this small change that fixes some error message
> > when you specify a too large Young- or OldPLABSize?
> >
> > Instead of printing the maximum PLAB size, the error message contains
> > the minimum PLAB size.
> >
> > The change is just one line, but the test is a bit longer.
> >
> > CR:
> > https://bugs.openjdk.java.net/browse/JDK-8134856
> > Webrev:
> > http://cr.openjdk.java.net/~tschatzl/8134856/webrev/
> > Testing:
> > jprt
> >
> > Thanks,
> >    Thomas
> >
> >
>