UCLA Samueli Computer Science
Yang-gon Kim
CSI-256 : GPU Microarchitecture
SnakeByte: A TLB Design with Adaptive and Recursive Page Merging in GPUs
Jiwon Lee, Won Woo Ro et al. Yonsei University. HPCA 2023
Contents
2
Introduction : TLB
[image] : https://www.csa.iisc.ac.in/~arkapravab/papers/asiaccs-gpu-covert-channel-slides.pdf
3
Introduction : GMMU
SM
…
GMMU
GPU
4
SIMD
SIMD
SIMD
L1$
page walk cache
page table walker
TLB
L1/L2
L2$
ctl
Contribution : Summary
NW : Heterogeneous computing benchmark
BH : Irregular programs on GPU
XSB : Monte Carlo reactor Analysis
5
Contribution : observations
NW : Heterogeneous computing benchmark
BH : Irregular programs on GPU
XSB : Monte Carlo reactor Analysis
6
TLB coverage
page size
Contribution : Promotion
7
4KB
256KB
32KB
x8
x8
8
Contribution : Promotion
9
Contribution : Recursive Promotion
10
Result
11
12
Thank You
Thank You
Thank You
Appendix
Q0. How does SnakeByte help page fault(not-present page)?
A0. SnakeByte does not support for page fault. It only supports TLB misses and PTW only. It doesn’t interfere with OS/driver.
Q1. If GMMU heavily relies on hardware only, how are programs controlled? (OS usually stalled the program and resume after page fault)
A1. when a warp is stalled by long memory access, then the warp scheduler activates the other warps to hide the latency.
13
Appendix
14
Appendix
15
Appendix
16