phd-thesis/plan/40_a72_frontend.md

82 lines
2.6 KiB
Markdown

# Beyond ports: manually modelling the A72 frontend
## Necessity to go beyond ports
* Palmed: concerned mostly with ports
* Noticed the importance of the frontend while investigating its performances
* heatmap representation: uops gone wild
* example of a frontend-bound microkernel
* Palmed's vision of a frontend
* UiCA: OK, but it's more complicated
## Cortex A72
* General intro
* ARMv8-A
* Out of order
* Designed as general-purpose, high-performance core for low-power applications
"Next generation of high-efficiency compute"
* Raspberry Pi 4 (BCM2711)
* Backend
* 2x Int
* IntM
* Load
* Store
* FP0
* FP1
* Frontend: 3 insn/cycle
* very limiting: eg. (?)
* Pure Palmed results
## Manual frontend
### Base methodology
* Basis: throughput model
* eg. Palmed, uops, official reference
* Simple instructions: 1μop, single port.
* Categorize by Palmed quads: a~b iff ∀i, Cyc(ai) = Cyc(bi)
* (not necessary, but reduces number of experiments required)
* Find the impact of insn i on frontend.
* The frontend must be bottleneck; build a benchmark B = i + (simples) so
that the simples do not cause a bottleneck backend-wise
* Add until Cyc(B) > Cyc(i)
* Should need at most 3 x Cyc(i) - 1 simples
* Measure with Pipedream
* General case: F(i) = 3xCyc(B) - |simples|
### Bubbles
* Instructions:
* `ADD_RD_SP_W_SP_RN_SP_W_SP_AIMM`, eg. `add w0, w1, #0x10`: 1 μop, Int01
* `MUL_RD_W_RN_W_RM_W`, eg. `mul w0,w1,w2`, 1μop, IntM
* `FMIN_FD_D_FN_D_FM_D`, eg `fmin d0, d1, d2`, 1μop, FP01
* `ADDV_FD_H_VN_V_8H`, eg. `addv h0, v0.8h`, 2μop, FP01
[doc](https://developer.arm.com/documentation/ddi0602/2022-12/Base-Instructions/ADD--immediate---Add--immediate--?lang=en)
* Example:
* `add`: 1/2 cycle (backend bound)
* `addv`: 1 cycle (backend bound)
* `add; fmin`: 2/3 cycle (frontend bound)
* `add; addv`: 1 cycle (frontend bound)
* [find nice example]
### Adopted model
* Question: given a kernel K, how many (frac.) cycles does it take to be
decoded in steady state?
* Frontend state ∈ [|0,2|]: how many μops already decoded this cycle
* Assumption: if st > 0 and i s.t. F(i) > 1 cannot be decoded fully without
crossing a cycle boundary, we leave a bubble and start decoding it next cycle
* next: st \mapsto st after executing K
* longest "cycle" of next is at most 3
* thus next^3(0) brings us into steady state
* Execute K enough (t) times to reach the same state as the first
* Result is C(K^t) / t
### Evaluation on Palmed
* Add this model to Palmed: for each kernel, simply take
max(frontend(K), Palmed(k))
* Results