[15] RFR 8245240: Shenandoah: support nesting evacuation OOM scope

Zhengyu Gu zgu at redhat.com
Wed May 20 15:08:15 UTC 2020


Okay, then added assertion to make sure:
http://cr.openjdk.java.net/~zgu/JDK-8245240/webrev.01/index.html


Test:
    hotspot_gc_shenandoah (Linux/64 and 32)
    Built on Windows


Thanks,

-Zhengyu

On 5/19/20 5:23 PM, Roman Kennke wrote:
> On Tue, 2020-05-19 at 23:14 +0200, Aleksey Shipilev wrote:
>> On 5/19/20 9:38 PM, Roman Kennke wrote:
>>> I don't have a strong opinion regarding packed fields or separate
>>> fields. Bitfield access seems slightly awkward (I had to look it up
>>> to
>>> see what it actually is), otoh, space is limited in TL, so... ?
>>
>> I would prefer separate fields to make the code clearer. Are we
>> really down to squeezing in a new
>> byte in TL?
> 
> True. As long as it fits and doesn't exceed the current limit, I guess
> we're ok. I expect that it also means simpler generated machine code.
> 
> Roman
> 



More information about the hotspot-gc-dev mailing list