Cdiffs Udiffs Sdiffs Frames Old New Patch Raw src/share/vm/opto/addnode.cpp

rev 10168 : 8003585: strength reduce or eliminate range checks for power-of-two sized arrays
Summary: change ((x & m) u<= m) to always true and ((x & (m - 1)) u< m) into (m > 0)
Reviewed-by: kvn, roland
rev 10221 : 8149745: C2 should optimize long accumulations in a counted loop
summary: Look for parallel iv for long adds
Reviewed-by:

10 lines changed: 10 ins; 0 del; 0 mod; 931 unchg

Cdiffs Udiffs Sdiffs Frames Old New Patch Raw src/share/vm/opto/addnode.hpp

rev 10221 : 8149745: C2 should optimize long accumulations in a counted loop
summary: Look for parallel iv for long adds
Reviewed-by:

4 lines changed: 1 ins; 0 del; 3 mod; 249 unchg

Cdiffs Udiffs Sdiffs Frames Old New Patch Raw src/share/vm/opto/loopnode.cpp

rev 10221 : 8149745: C2 should optimize long accumulations in a counted loop
summary: Look for parallel iv for long adds
Reviewed-by:

35 lines changed: 18 ins; 0 del; 17 mod; 3812 unchg

Cdiffs Udiffs Sdiffs Frames Old New Patch Raw src/share/vm/opto/mulnode.cpp

rev 10221 : 8149745: C2 should optimize long accumulations in a counted loop
summary: Look for parallel iv for long adds
Reviewed-by:

10 lines changed: 10 ins; 0 del; 0 mod; 1357 unchg

Cdiffs Udiffs Sdiffs Frames Old New Patch Raw src/share/vm/opto/mulnode.hpp

rev 10221 : 8149745: C2 should optimize long accumulations in a counted loop
summary: Look for parallel iv for long adds
Reviewed-by:

3 lines changed: 1 ins; 0 del; 2 mod; 264 unchg

Cdiffs Udiffs Sdiffs Frames Old New Patch Raw src/share/vm/opto/subnode.cpp

rev 10168 : 8003585: strength reduce or eliminate range checks for power-of-two sized arrays
Summary: change ((x & m) u<= m) to always true and ((x & (m - 1)) u< m) into (m > 0)
Reviewed-by: kvn, roland
rev 10221 : 8149745: C2 should optimize long accumulations in a counted loop
summary: Look for parallel iv for long adds
Reviewed-by:

11 lines changed: 11 ins; 0 del; 0 mod; 1557 unchg

Cdiffs Udiffs Sdiffs Frames Old New Patch Raw src/share/vm/opto/subnode.hpp

rev 10221 : 8149745: C2 should optimize long accumulations in a counted loop
summary: Look for parallel iv for long adds
Reviewed-by:

5 lines changed: 2 ins; 0 del; 3 mod; 502 unchg

This code review page was prepared using ../../c2-inlining/webrev.ksh (vers 23.18-hg-never-dev).