Load value prediction via path-based address prediction avoiding mispredictions due to conflicting stores

Current flagship processors excel at extracting instruction-level-parallelism (ILP) by forming large instruction windows. Even then, extracting ILP is inherently limited by true data dependencies. Value prediction was proposed to address this limitation. Many challenges face value prediction, in thi...

Celý popis

Uloženo v:

Podrobná bibliografie
Vydáno v:	MICRO-50 : the 50th annual IEEE/ACM International Symposium on Microarchitecture : proceedings : October 14-18, 2017, Cambridge, MA s. 423 - 435
Hlavní autoři:	Sheikh, Rami, Cain, Harold W., Damodaran, Raguram
Médium:	Konferenční příspěvek
Jazyk:	angličtina
Vydáno:	New York, NY, USA ACM 14.10.2017
Edice:	ACM Conferences
Témata:	Address Prediction Computer systems organization > Architectures > Serial architectures > Pipeline computing Computer systems organization > Architectures > Serial architectures > Reduced instruction set computing Computer systems organization > Architectures > Serial architectures > Superscalar architectures History Machinery Microarchitecture Path-based Predictor Pipelines Prefetching Registers Value Prediction value prediction address prediction path-based predictor microarchitecture
ISBN:	1450349528, 9781450349529
ISSN:	2379-3155
On-line přístup:	Získat plný text
Tagy:	Přidat tag Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!

Buďte první, kdo okomentuje tento záznam!