Herding Cats: Modelling, Simulation, Testing, and Data Mining for Weak Memory.
- Submitting institution
-
Queen Mary University of London
- Unit of assessment
- 11 - Computer Science and Informatics
- Output identifier
- 425
- Type
- D - Journal article
- DOI
-
10.1145/2627752
- Title of journal
- ACM Transactions on Programming Languages and Systems
- Article number
- 7
- First page
- 1
- Volume
- 36
- Issue
- 2
- ISSN
- 0164-0925
- Open access status
- Out of scope for open access requirements
- Month of publication
- July
- Year of publication
- 2014
- URL
-
-
- Supplementary information
-
-
- Request cross-referral to
- -
- Output has been delayed by COVID-19
- No
- COVID-19 affected output statement
- -
- Forensic science
- No
- Criminology
- No
- Interdisciplinary
- No
- Number of additional authors
-
2
- Research group(s)
-
-
- Citation count
- 88
- Proposed double-weighted
- No
- Reserve for an output with double weighting
- No
- Additional information
- "We introduce a novel framework to describe weak memory models of
multiprocessors. We prove that a previous model for Power is flawed. Extensive
testing on ARM revealed a bug acknowledged by ARM. We provide a new simulator
(herd7), now used by the Linux kernel, and exercised a new static analysis tool
on 1500 software packages. The paper was also invited for presentation in PLDI,
the premier conference for programming languages and systems. It is cited in C++
draft amendment C/C++ WG21/P0098R1, Linux Weekly News, and led to work with
NVidia on their memory model plus Alglave being hired by ARM."
- Author contribution statement
- -
- Non-English
- No
- English abstract
- -