Tootfinder

Spec Sheets Are Not Kernels: An ISA- and Source-Level Audit of INT8 Availability on NVIDIA Blackwell Ultra
Teng-Ruei Chen
https://arxiv.org/abs/2608.11693 https://arxiv.org/pdf/2608.11693 https://arxiv.org/html/2608.11693
arXiv:2608.11693v1 Announce Type: new
Abstract: NVIDIA's published specifications give the Blackwell Ultra GPU (B300) a dense-compute ratio of roughly 30:1 between FP8 and INT8 tensor-core throughput; its predecessors, H200 and B200, both provide 1:1. We audit what this deprioritization means in practice by tracing INT8 W8A8 support through four layers of the stack: the published specifications, the PTX ISA, NVIDIA's CUTLASS kernel library, and the two major open-source LLM serving engines (vLLM and SGLang). We find a consistent, layered withdrawal: (i) the PTX ISA never exposes the fifth-generation tensor-core integer path (tcgen05.mma with .kind::i8) on sm_103a, even though the same PTX revision extends the FP4 kinds to that target, leaving legacy warp-level IMMA as the only architecturally legal integer tensor-core path on B300; (ii) CUTLASS's kernel generator explicitly skips INT8 UMMA generation for any build targeting 103a, while generating FP8 unconditionally; (iii) vLLM ships no INT8 GEMM for Blackwell and fails with a hard runtime error at the first forward pass, after the model has loaded; and (iv) SGLang's ahead-of-time INT8 GEMM stops at Sm90, while its FP8 tuning configurations already cover B200. We document an escape hatch (rerouting vLLM's INT8 path to a JIT-compiled Triton backend via an environment variable), a false-negative trap in the obvious profiler methodology for detecting "native INT8" on sm_103, and the practical failure semantics that make naive testing expensive. Together, these findings show that a quantization format's availability is a property of the whole stack rather than of the model or the spec sheet. Four distinct layers, three of them NVIDIA's own, withdrew INT8 support in mutually consistent ways, and a format that is nominally present on the datasheet is, by default, undeployable on this hardware.
toXiv_bot_toot

Good morning! Since it's become a drag to support in mainline, I've extracted svcs's Pyramid support into an own package svcs-pyramid.
That means, the next svcs version will ship WITHOUT a `svcs.pyramid` module.
You can start using svcs-pyramid TODAY to avoid breakage the moment 26.2.0 drops – there are no code changes, just adjust the import statement:

GitHub - hynek/svcs-pyramid: Pyramid integration for the svcs Python service locator
Pyramid integration for the svcs Python service locator - hynek/svcs-pyramid

Spec-Driven Hardware Evolution via Executable Contract Refinement and Proof-Guided RTL Update
Shibo Zhao, Yang Zhang, Mengxia Tao, Baoqi Zhang, Kezhi Li, Qiang Xu, Binwu Zhu, Hao Yan, Min Li
https://arxiv.org/abs/2608.12684 https://arxiv.org/pdf/2608.12684 https://arxiv.org/html/2608.12684
arXiv:2608.12684v1 Announce Type: new
Abstract: Hardware development is inherently evolutionary: major revisions typically begin by changing intended behavior and then updating a previously validated implementation, rather than regenerating RTL from scratch. Yet most recent LLM-based hardware research still frames the task primarily as prompt-to-RTL generation, offering limited support for semantic version evolution of trusted legacy designs. We present spec-driven hardware evolution, a contract-centered formulation for RTL version iteration. Instead of treating a new feature request as a direct prompt for RTL generation, we refine it into a reviewed executable contract for the next version. This contract specifies what must hold at the externally visible transactional level through a behavior-level reference together with explicit observation and checking semantics, while leaving how the change is realized in RTL to the evolution process. Based on this formulation, we organize hardware evolution into four stages: Specify, Plan, Implement, and Validate. After contract approval, the remaining stages proceed automatically: Plan derives cross-version semantic deltas and localizes affected RTL regions, aided by mutation-based semantic probing; Implement and Validate then perform legacy-aware RTL update under proof-guided checking and iterative repair. We evaluate the framework on a controlled version-evolution case study of a representative TPU datapath block under data-format changes. The results support the feasibility of contract-driven hardware evolution and demonstrate that the proposed backend workflow can effectively drive validated legacy RTL toward next-version functional convergence under a reviewed executable contract. An anonymous artifact for reproducibility is available at https://anonymous.4open.science/r/SDHE-3A6C.
toXiv_bot_toot

from my link log —
Lisp: the many faces of an undying programming language.
http://jakob.space/blog/thoughts-on-lisps.html
saved 2020-07-21 https://

Air Force Secretary Troy Meink made the first public declaration Monday that the United States has placed “space control weapons” in orbit, an announcement that will surely reverberate in the power centers of Beijing and Moscow: #SpaceWar