Displaying 2 results from an estimated 2 matches for "inlineddi".
2015 Jul 30
4
[LLVMdev] RFC: Callee speedup estimation in inline cost analysis
...ruction and a long latency instruction. Instead of
using InstructionCount(BB)), we could use Sum_I(Weight(I)) where the
summation is over all instructions I in B and Weight(I) represents the time
it takes to execute instruction I.
The dynamic instruction count of F into a callsite C after inlining
InlinedDI(F, C) can be similary computed taking into account the
instructions that are simplified due to inlining. The estimated speedup is
Speedup(F, C) = (DI(F) - InlinedDI(F, C)) / DI(F)
Speedup above a pre-determined threshold implies there is an expected
benefit in inlining the callee F and hence a bo...
2015 Jul 31
0
[LLVMdev] RFC: Callee speedup estimation in inline cost analysis
...struction. Instead of
> using InstructionCount(BB)), we could use Sum_I(Weight(I)) where the
> summation is over all instructions I in B and Weight(I) represents the time
> it takes to execute instruction I.
>
> The dynamic instruction count of F into a callsite C after inlining
> InlinedDI(F, C) can be similary computed taking into account the
> instructions that are simplified due to inlining. The estimated speedup is
> Speedup(F, C) = (DI(F) - InlinedDI(F, C)) / DI(F)
>
> Speedup above a pre-determined threshold implies there is an expected
> benefit in inlining the...