pellegrini/AC3-TP1
1
1
Fork 0
Code Issues Pull requests Projects Releases 3 Packages Wiki Activity Actions
3 releases 4 tags
  • v0.1.2 6954d3cc94

    v0.1.2: fix text issues noticed by reviewers, improve docs and make scripts more robust Stable

    pellegrini released this 2026-10-09 22:34:04 +00:00 | -21 commits to master since this release

    Signed by pellegrini
    GPG key ID: 9AC2AAA7E7C254FC

    This release is the camera-ready version for SSCAD-WIC: the paper was tightened to the 8-page body limit while answering the reviews, supporting numbers that did not fit were moved into report documentation, and a latexrun hang under TinyTeX was fixed.

    What Changed

    • Camera-ready manuscript (report/main.tex): rewritten abstract, related work (added Dave & Kotak 2024 and Saha 2026; softened the gap claim), methodology, results, and conclusion to address reviews. Corrected analytical model (Amdahl form of Eq. 1 now uses the proper f_{32} = 2f/(1+f); residual for ATAX decomposed into latency rise + sub-linear miss reduction), added predicted-vs-measured line-size column, clarified core-model realism and determinism, made the warm-start explicit (caches are not flushed at ROI entry) and used it to attribute ATAX’s capacity gain to L3 ≥ 32 MiB, replaced the unmeasured “compulsory and capacity misses” claim with the actual L1D miss counts from the associativity sweep, and fixed access-pattern / live-data occupancy descriptions.
    • New report notes (report/README.md): records what the page limit forced us to compress or omit (dropped Table 4, full f_stall sweep, original model error, associativity/capacity per-preset numbers, determinism details, etc.), the corrections to the accepted text (arithmetic fixes, attribution of capacity, warm-start mechanism, residual of Eq. 1, etc.), and deferred reviewer requests.
    • Compact line-size figure: visualize_results.py gains create_cache_line_compact_figure() so L1D miss rate and IPC fit a single short two-panel figure in the report.
    • latexrun freeze under TinyTeX: pin max_print_line=79 in report/.justfile and report/Makefile, and guard the unwrap loop in the vendored latexrun so a non-79 wrap no longer spins forever.
    • Bibliography: added Smith (1987), Hill & Smith (1989), Cantin & Hill (2001), Williams et al. (2009), Hennessy & Patterson (2017), Gholami et al. (2024), Dave & Kotak (2024), Saha (2026); repository citation moved to a footnote and the old repo:2026 entry removed.
    • Tag / clone instructions: default branch tag is now v0.1.2 (README and demo script).

    Setup/Docs Improvements

    • Top-level README points readers at report/README.md for paper-specific constraints, omissions, and how to rebuild only the PDF.
    • Report build docs: one-command zig build report (or just all inside report/) when figures are already present.
    Downloads
  • v0.1.1 d0fe8b4643

    v0.1.1: improve code robustness, performance, and reproducibility Stable

    pellegrini released this 2026-05-24 13:20:51 +00:00 | 15 commits to master since this release

    Signed by pellegrini
    GPG key ID: 9AC2AAA7E7C254FC

    This release focuses on making the artifact easier to build/run reproducibly across Linux distros, speeding up gem5 builds, and making the simulation/plot pipeline more automation-friendly.

    What Changed

    • Python toolchain bootstrapping switches to uv (replaces the prior pyenv-based setup). The build graph now pins Python 3.14.3 via uv and installs requirements into the project venv.
    • Virtualenv location moved to repo root: gem5/venv is now at .venv to avoid a build race where the venv could be initialized before gem5/ exists.
    • Report build no longer requires GNU Make: the report build is driven by Just via a new report/.justfile (still using the vendored latexrun flow).
    • gem5 builds are faster by default:
      • Build only the needed target (no more ALL target builds).
      • Use the gem5.fast preset (gem5/build/X86/gem5.fast) for the simulator build.
    • Plot generation is incremental: visualize_results.py skips regenerating figures when outputs are already up-to-date with their input stats.txt files, and prints a generated/skipped count at the end.
    • Better failure signaling for automation: cache_config.py now exits non-zero when the simulation doesn’t end in the expected “normal exit” condition (and prints both the raw exit reason and translated exit event when available).
    • README installation instructions were tightened up across distros (Debian/Fedora/RHEL/Arch), including missing libc/libstdc++ static/runtime packages for dnf-based systems and better ordering of dependencies.
    • mise support was added:
      • clearer tasks + descriptions, and “run the whole pipeline” ergonomics
      • TeX package installation is conditional/idempotent
      • adds missing perl tool dependency needed by TeX package installs in some environments
    Downloads
  • v0.1.0 a331462c68

    v0.1.0: finish assignment for submission Stable

    pellegrini released this 2026-04-06 02:07:10 +00:00 | 44 commits to master since this release

    Signed by pellegrini
    GPG key ID: 9AC2AAA7E7C254FC
    Downloads