[aarch64-port-dev ] RFR: JDK-8169177 Aarch64: SIGSEGV when "-XX:+ZeroTLAB" is specified along with GC options

White, Derek Derek.White at cavium.com
Wed Nov 16 17:08:08 UTC 2016


Hi Andrew,

I withdraw my suggestion " c1_MacroAssembler_aarch64.cpp:   - Think about calling zero_words instead."

But I still suggest calling zero_words in MacroAssembler::tlab_refill().

The whole point of the ZeroTLAB flag is if bulk clearing the TLAB is faster than piecemeal clearing as needed. Doing a simpler zero_memory just has the potential to pollute the cache unnecessarily.  The extra size of calling zero_words here only comes into play if ZeroTLAB flag is on, and the macro should only be expanded once in any case.

Thanks!

 - Derek

-----Original Message-----
From: aarch64-port-dev [mailto:aarch64-port-dev-bounces at openjdk.java.net] On Behalf Of Andrew Haley
Sent: Friday, November 11, 2016 12:28 PM
To: aarch64-port-dev at openjdk.java.net
Subject: Re: [aarch64-port-dev ] RFR: JDK-8169177 Aarch64: SIGSEGV when "-XX:+ZeroTLAB" is specified along with GC options

On 11/11/16 17:21, Kavitha Natarajan wrote:
> Thanks a lot for your review. MacroAssembler::zero_words() seems more 
> optimized than MacroAssembler::zero_memory() as you suggested.

It is, but it's a larger.  The smaller one is preferred for use in C1- generated code, just to try to keep the code size reasonable.

Andrew.


More information about the aarch64-port-dev mailing list