https://bugs.kde.org/show_bug.cgi?id=381553
--- Comment #20 from Julian Seward <jsew...@acm.org> ---
For the actual cost of doing register allocation, for /usr/bin/date on
Memcheck,
I have the following profile results from Callgrind. These numbers are for
doRegisterAllocation_{v2,v3} and all of their callees.
v2 193.4 million insns
v3 156.7 million insns
The v3 allocator beats the v2 allocator on all Callgrind-measured numbers
(memory accesses, misses, etc) except that it is about 10% worse in
mispredicted conditional branches, for reasons I don't know yet.
--
You are receiving this mail because:
You are watching all bug changes.
