https://gcc.gnu.org/bugzilla/show_bug.cgi?id=125607
--- Comment #5 from ktkachov at gcc dot gnu.org ---
(In reply to Martin Jambor from comment #4)
> Can you please build the benchmark with -fdump-ipa-cp-details and paste here
> the bit when it reasons about cloning the function? That should contain
> everything that is fed into good_cloning_opportunity_p and what it comes up
> with. Thanks.
Evaluating opportunities for _ZN5Token5MatchEPKS_PKci.part.0/15763.
- considering value 0 for param #2 int (caller_count: 54)
good_cloning_opportunity_p (time: 6.79752, size: 381, freq_sum: 174.456,
single_call) -> evaluation: 2645.64, thres
hold: 1650
Creating a specialized node of _ZN5Token5MatchEPKS_PKci.part.0/15763.
replacing param #2 int with const 0
Accounting size:98.00, time:380.69 on predicate exec:(true)
Accounting size:3.00, time:2.00 on new predicate exec:(not
inlined)
Node: _ZN5Token5MatchEPKS_PKci.part.0/15763:
param [2]: VARIABLE
0 [scc: 3721, from: 15781(8.090909) 15764(0.660000) 15791(18.149951)
15791(5.989484)
15765(6.916027) 15765(6.422025) 15784(11.444430) 15773(1.641193)
15795(0.011889)
15795(0.003431) 15795(0.031189) 15770(5.691225) 15770(17.246136)
15976(2.227273)
15976(0.066758) 15976(0.303445) 15976(0.033379) 15976(0.034814)
15976(0.158246)
15976(0.227977) 15976(1.349320) 15976(0.944524) 15976(0.541073)
15976(0.061250)
15769(6.916027) 15769(6.422025) 15783(8.833840) 15783(9.347979)
15786(0.700700)
15786(1.676818) 15934(0.125049) 15934(0.062525) 15934(6.598936)
15934(15.400878)
15934(5.082290) 15934(0.154109) 15934(0.025428) 15934(0.006866)
15935(5.886713)
15935(4.033010) 15777(1.641193) 15782(8.833840) 15966(1.000000)
15966(0.488800)
15966(0.307374)]
[loc_time: 1, loc_size: 110, prop_time: 6.79752, prop_size: 381]
Estimating effects for _ZN5Token5MatchEPKS_PKci.part.0/15763.
Estimating body: _ZN5Token5MatchEPKS_PKci.part.0/15763
size:110 time:520.923427 nonspec time:520.923427 loops with known
iterations:0 known strides:0
- estimates for value 0 for param #2 int: time_benefit: 1, size: 110 (local)
