RFR(two-liner): 8212707: GlobalCounter padding is too optimistic (original) (raw)
Robbin Ehn robbin.ehn at oracle.com
Mon Oct 22 11:14:46 UTC 2018
- Previous message: RFR(two-liner): 8212707: GlobalCounter padding is too optimistic
- Next message: RFR(two-liner): 8212707: GlobalCounter padding is too optimistic
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
Thanks Claes!
/Robbin
On 10/22/18 1:18 PM, Claes Redestad wrote:
Looks good to me, and as you imply in the comments the DEFAULTCACHELINESIZE should be 64 on x8664
/Claes On 2018-10-22 13:06, Robbin Ehn wrote: Hi all, please review.
We should not divide DEFAULTCACHELINESIZE by 2. More info in issue: https://bugs.openjdk.java.net/browse/JDK-8212707 Code: below Sanity with T1. Thanks, Robbin diff -r bca2b63dd839 src/hotspot/share/utilities/globalCounter.hpp --- a/src/hotspot/share/utilities/globalCounter.hpp Mon Oct 22 14:08:07 2018 +0800 +++ b/src/hotspot/share/utilities/globalCounter.hpp Mon Oct 22 11:48:17 2018 +0200 @@ -48,1 +48,1 @@ - DEFINEPADMINUSSIZE(0, DEFAULTCACHELINESIZE/2, 0); + DEFINEPADMINUSSIZE(0, DEFAULTCACHELINESIZE, 0); @@ -50,1 +50,1 @@ - DEFINEPADMINUSSIZE(1, DEFAULTCACHELINESIZE/2, sizeof(volatile uintx)); + DEFINEPADMINUSSIZE(1, DEFAULTCACHELINESIZE, sizeof(volatile uintx));
- Previous message: RFR(two-liner): 8212707: GlobalCounter padding is too optimistic
- Next message: RFR(two-liner): 8212707: GlobalCounter padding is too optimistic
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]