Looking at `sudo perf top -F 15000 -a -n -z -e 'cpu/event=0xD4,cmask=0x00,edge=0,inv=0,umask=0x0/'` I get counts on anything but divides. I have not seen it actually count on or near a divide, ever. The PPR is wrong.
Measuring on a Ryzen 2700X Stepping 2
Samples: 41K of event 'cpu/event=0xD4,cmask=0x00,edge=0,inv=0,umask=0x0/', 15000 Hz, Event count (approx.
update_blocked_averages /proc/kcore [Percent: local period]
0.02 │ cmp %rdx,%rbx
0.02 │ ↓ je 781
│100: mov 0x150(%rbx),%rax
1.43 │ cmp $0xffffffffb8d5c3c0,%rax
│ ↓ je 156
0.06 │ mov 0x100(%rbx),%rcx
0.44 │ mov 0xa0(%rbx),%rdx
0.01 │ sub %rcx,%rdx
0.01 │ shr $0x6,%rcx
0.04 │ mov %rdx,%rdi
│ sar $0x3f,%rdi
0.09 │ mov %rdi,%rsi
0.02 │ xor %rdx,%rsi
0.13 │ sub %rdi,%rsi
0.37 │ cmp %rcx,%rsi
0.03 │ ↓ jbe 156
│ lock add %rdx,0x140(%rax)
0.88 │ mov 0xa0(%rbx),%rax
0.00 │ mov %rax,0x100(%rbx)
0.01 │14f: mov 0x150(%rbx),%rax
0.21 │156: mov 0xf0(%rax),%rax
7.58 │ mov -0x38(%rbp),%rcx
0.05 │ mov (%rax,%rcx,1),%r12
23.36 │ test %r12,%r12
0.02 │ ↓ je 46c
0.60 │ cmpq $0x0,0x1a0(%r12)
35.91 │ ┌──jne 19a
0.65 │ │ cmpq $0x0,0x1b0(%r12)
0.20 │ │↓ jne 19a
0.29 │ │ mov 0x160(%r12),%rax
4.24 │ │ cmpq $0x0,0x108(%rax)
1.47 │ │↓ je 46c
0.02 │19a:└─→mov 0x158(%r12),%r13
0.24 │ mov 0x18c(%r13),%r8d
0.77 │ mov 0x130(%r13),%r15
0.25 │ test %r8d,%r8d
0.02 │ ↓ jne 87b
│ cmpl $0x1,0x998(%r15)
0.07 │ mov %r15,%rdx
│ ↓ jbe 891
0.02 │1ca: mov 0x9a8(%r15),%r10
│ sub 0x180(%r13),%r10
│1d8: cmpq $0x0,0x180(%r12)
│ ↓ je 1ff
│ mov 0xa20(%rdx),%esi
│ mov %r10,%rdi
│ mov %r12,%rcx
0.01 │ mov %r13,%rdx
0.01 │ mov %r10,-0x48(%rbp)
0.01 │ → callq __update_load_avg_se
│ mov -0x48(%rbp),%r10
0.01 │1ff: mov 0xc4(%r13),%r15d
