RFR (L): 7173584: Implement arraycopy as a macro node (original) (raw)

Vladimir Kozlov vladimir.kozlov at oracle.com
Fri Sep 28 15:05:11 PDT 2012

Thank you for the case. I did go through the code and this change looks fine.

Thanks, Vladimir

Roland Westrelin wrote:

Vladimir,

Thanks for taking a look at this.

I am not comfortable with this predicate change. The check come from generatelimitguard() :

if (a.length u< offset+copylength) uncommontrap; else arraycopy; which is reversed by igvn to if (offset+copylength u< a.length) arraycopy; else uncommontrap; So it should not be IfFalse projection. Can you explain the case which you hit? This is the java code that triggers this: for (int i=0; i<a1.length; i++) { a2 = new …[a3.length+a1.length-i-1]; System.arraycopy(a3, 0, a2, 0, a3.length); System.arraycopy(a1, i+1, a2, a3.length, a1.length-i-1); } For the first arraycopy, one of the check is: if (a2.length < a3.length) uncommontrap; else arraycopy; if (a3.length + a1.length-i-1 < a3.length) uncommontrap; else arraycopy; a1.length is 5 if (a3.length + 4 - i < a3.length) uncommontrap; else arraycopy; a3.length is a loop invariant. Also instead of duplicating gensubtypecheck() in macroArrayCopy.cpp can you modify the original one to work for you (by passing additional flag)? Ok I'll do that. Roland.

More information about the hotspot-compiler-dev mailing list