GPU-unlock: FP/BF16 instruction issue rate on Nvidia 3000's cards
A technical guide for a Linux kernel module that unlocks higher FP/BF16 instruction rates on Nvidia 3000-series graphics cards. The tool allows users to achieve professional-grade performance without needing to manually patch drivers after every update.
Why it matters
This provides a significant performance boost for hobbyists and researchers running AI or high-performance computing workloads on consumer hardware.
Unlock FP/BF16 instruction issue rate on Nvidia 3000's cards to their professional equivalents
This Linux kernel module combines the CMP unlocker's V67/PLM technique with the pre-driver timing of the UEFI approach. It runs the unlock, writes the speed selectors (SS0 0x88888888 , which selects the full BF16 rate, and the 4-bit SS1 0x8 IMLA4) and performs FLR before the unmodified NVIDIA driver binds the card. It is neither a UEFI application nor a patched NVIDIA driver.
The benefit of this early kmod driver is that it does not require manual patching each time nvidia releases a new driver version.
The effect of this unlock is that it doubles the instruction issuing speed of FP/BF16 FMA into FP32 operations on GA102, GA103, GA104 and GA106 cards.
3090 Ti dense 8192³ GEMM at 450 W, five-sample medians with FP32 accumulation :
Get smarter about the news
Sign up free for a feed built around what you actually care about, Dive Deeper research on any story, and the full text of every article.
Create free accountAlready have an account? Sign in