7 papers
Towards an Accurate GPU Data Race Detector
Ajay Nayak, Anubhab Ghosh, Arkaprava Basu
Data races in GPU programs pose a threat to the reliability of GPU-accelerated software stacks. Prior works proposed various dynamic (runtime) and static (compile-time) techniques…
LLM-42: Enabling Determinism in LLM Inference with Verified Speculation
Raja Gond, Aditya K Kamath, Ramachandran Ramjee +1
In LLM inference, the same prompt may yield different outputs across different runs. At the system level, this non-determinism arises from floating-point non-associativity combined…
Chasing Elusive Memory Bugs in GPU Programs
Anubhab Ghosh, Ajay Nayak, Dhananjay Rao Thallikar Shyam +1
Memory safety bugs, such as out-of-bound accesses (OOB) in GPU programs, can compromise the security and reliability of GPU-accelerated software. We report the existence of input-d…
PyGraph: Robust Compiler Support for CUDA Graphs in PyTorch
Abhishek Ghosh, Ajay Nayak, Ashish Panwar +1
Machine learning (ML) workloads launch hundreds to thousands of short-running GPU kernels per iteration. With GPU compute throughput growing rapidly, CPU-side launch latency of ker…
iMIV: in-Memory Integrity Verification for NVM
Rajat Jain, Aravinda Prasad, Sreenivas Subramoney +1
Non-volatile Memory (NVM) could bridge the gap between memory and storage. However, NVMs are susceptible to data remanence attacks. Thus, multiple security metadata must persist al…
Leveraging Architectural Support of Three Page Sizes with Trident
Venkat Sri Sai Ram, Ashish Panwar, Arkaprava Basu
Large pages are commonly deployed to reduce address translation overheads for big-memory workloads. Modern x86-64 processors from Intel and AMD support two large page sizes -- 1GB…