RFR(XXS): 8165018: Missing memory barrier for PPC64 in Unsafe_GetObjectVolatile
martin.doerr at sap.com
Mon Sep 5 08:35:50 UTC 2016
Hi David and Kim,
thanks for reviewing.
Seems like we all share the opinion that the missing G1 barrier is a bug which should get fixed.
Would you prefer to handle it in a separate bug+webrev? Otherwise I can create a webrev for both issues.
From: Kim Barrett [mailto:kim.barrett at oracle.com]
Sent: Samstag, 3. September 2016 02:04
To: David Holmes <david.holmes at oracle.com>
Cc: Doerr, Martin <martin.doerr at sap.com>; hotspot-runtime-dev at openjdk.java.net; hotspot-gc-dev at openjdk.java.net
Subject: Re: RFR(XXS): 8165018: Missing memory barrier for PPC64 in Unsafe_GetObjectVolatile
> On Sep 2, 2016, at 1:03 AM, David Holmes <david.holmes at oracle.com> wrote:
> Hi Martin,
> Adding GC folk
> On 30/08/2016 8:51 PM, Doerr, Martin wrote:
>> we found that a memory barrier for PPC64 is missing in the current Unsafe implementation. get_volatile already contains the memory barrier for "support_IRIW_for_not_multiple_copy_atomic_cpu". The same is needed in Unsafe_GetObjectVolatile.
>> Here's my webrev:
> That looks fine to me.
Looks good to me too.
>> And while looking at it I wonder why Unsafe_GetObjectVolatile does not contain a G1 barrier like Unsafe_GetObject. Is it not possible to use the Volatile version to access the referent field of a Reference?
> That looks like a bug to me. :)
Looks like a bug to me too.
>> Please review. As it is shared code, I will need a sponsor, please.
> I can sponsor.
>> Best regards,
More information about the hotspot-gc-dev