Displaying 1 result from an estimated 1 matches for "2000011".
Did you mean:
3000011
2015 Apr 24
5
[LLVMdev] Loss of precision with very large branch weights
...is' for function 'main':
block-frequency-info: main
- entry: float = 1.0, int = 8
- for.cond: float = 500001.5, int = 4000011
- for.body: float = 500000.5, int = 4000003
- for.inc: float = 500000.5, int = 4000003
- for.end: float = 1.0, int = 8
- for.cond1: float = 250001.5, int = 2000011
- for.body3: float = 250000.5, int = 2000003
- for.inc4: float = 250000.5, int = 2000003
- for.end6: float = 1.0, int = 8
But if I manually modify the frequencies of both to get close to MAX_INT32,
the ratios between the frequencies do not reflect reality. For example, if
I change branch_weigh...