RFR: 6869327: Add new C2 flag to keep safepoints in counted loops.

[Aleksey Shipilev]

Hi,

On 11/27/2015 04:01 PM, Andreas Eriksson wrote:
> Please review this change that adds a flag to keep a safepoint in
> counted loops.
> 
> Currently C2 removes safepoints in counted loops.
> This can force other safepointing threads to wait for the counted loop
> thread for long periods of time.
> This change adds a flag, UseCountedLoopSafepoints, which keeps a
> safepoint in the loop. Its default value is false.
> 
> Bug: 6869327: Add new C2 flag to keep safepoints in counted loops.
> https://bugs.openjdk.java.net/browse/JDK-6869327
> 
> Webrev: http://cr.openjdk.java.net/~aeriksso/6869327/webrev.00/

Do I get it right that this logic works good with loop unrolling? Can we
check?

E.g. I would expect that a counted loop like this:

 int s = 0;
 for (int c = 0; c < N; c++) {
    s += c;
    <safepoint>
 }

...would get unrolled and stripped to:

 int s = 0;
 for (int c = 0; c < N; c += 4) {
   s += c + 0;
   s += c + 1;
   s += c + 2;
   s += c + 3;
   <safepoint> // only one safepoint here
 }

This would basically perform a limited variant of what Vladimir mentions
as "strip mining" (aka transform a countable loop into two nested loops
and place safepoint into external loop), where the size of "internal"
loop is limited by unrolling.

Thanks,
-Aleksey

-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 819 bytes
Desc: OpenPGP digital signature
URL: <http://mail.openjdk.java.net/pipermail/hotspot-compiler-dev/attachments/20151127/61852d10/signature.asc>