Meta tags:

Headings (most frequently used words):

edit, branch, prediction, predictor, level, two, navigation, neural, tools, contents, implementation, history, see, also, references, external, links, menu, static, dynamic, random, next, line, one, local, global, alloyed, agree, hybrid, loop, indirect, of, function, returns, overriding, personal, namespaces, views, search, contribute, print, export, languages, saturating, counter, adaptive,

Text of the page (most frequently used words):
the (291), branch (176), prediction (103), and (100), #predictor (95), taken (45), #history (44), that (44), two (43), instruction (41), with (40), not (38), conditional (36), for (34), jump (33), bit (32), this (30), predictors (29), edit (27), level (25), are (23), from (22), counter (21), table (20), will (20), instructions (18), processor (17), unit (17), processors (17), branches (17), more (16), architecture (16), can (16), one (16), was (15), pdf (15), buffer (15), memory (15), target (15), machine (15), intel (15), has (15), local (15), dynamic (14), neural (14), used (14), each (14), pattern (14), retrieved (13), performance (13), indirect (13), adaptive (13), next (13), than (13), have (13), saturating (13), register (12), state (12), amd (12), line (12), per (11), alpha (11), loop (11), global (11), use (10), all (10), different (10), processing (10), address (10), core (10), ibm (10), called (10), pentium (10), first (10), static (10), time (10), may (9), using (9), cache (9), second (9), pipeline (9), based (9), other (9), same (9), entry (9), then (9), wikipedia (8), execution (8), computer (8), cycle (8), international (8), bimodal (8), where (8), any (8), perceptron (8), fetch (8), which (8), last (7), information (7), design (7), data (7), program (7), stack (7), single (7), cycles (7), mips (7), microarchitecture (7), fast (7), also (7), combined (7), fetched (7), predict (7), once (7), whether (7), every (7), before (7), stage (7), about (6), simple (6), random (6), digital (6), power (6), order (6), example (6), jimenez (6), ev8 (6), 2017 (6), proceedings (6), symposium (6), high (6), journal (6), modern (6), misprediction (6), made (6), between (6), bits (6), uses (6), four (6), scheme (6), most (6), its (6), after (6), many (6), hybrid (6), entries (6), some (6), times (6), number (6), shared (6), jumps (6), strongly (6), guess (6), agree (5), articles (5), chip (5), hardware (5), cpus (5), clock (5), quantum (5), cpu (5), pipelined (5), risc (5), architectures (5), turing (5), 2018 (5), 2003 (5), seznec (5), stretch (5), lucian (5), 2000 (5), original (5), method (5), part (5), mechanism (5), counters (5), very (5), large (5), because (5), delay (5), there (5), later (5), long (5), both (5), advantage (5), much (5), return (5), separate (5), likely (5), technique (5), correct (5), figure (5), when (5), discarded (5), page (4), links (4), unsourced (4), statements (4), speculative (4), microprocessor (4), purpose (4), management (4), decoder (4), control (4), multi (4), size (4), network (4), system (4), superscalar (4), stored (4), automaton (4), does (4), they (4), multiple (4), 2016 (4), ryzen (4), new (4), vintan (4), archived (4), alloyed (4), see (4), but (4), choice (4), base (4), only (4), recurrence (4), microprogrammed (4), were (4), proposed (4), implementation (4), microprocessors (4), good (4), function (4), way (4), such (4), well (4), executed (4), evaluated (4), non (3), cite (3), file (3), changes (3), what (3), help (3), current (3), contents (3), main (3), navigation (3), 2009 (3), org (3), array (3), general (3), signal (3), circuit (3), logic (3), sequential (3), rate (3), others (3), systems (3), package (3), pipelining (3), dec (3), sparc (3), vax (3), set (3), access (3), cellular (3), pointer (3), models (3), via (3), accuracy (3), december (3), ieee (3), path (3), 2019 (3), yeh (3), technical (3), analysis (3), mispredictions (3), wrong (3), university (3), mcfarling (3), combining (3), web (3), advanced (3), 2006 (3), caches (3), zero (3), majority (3), result (3), parity (3), rather (3), since (3), 21264 (3), commercial (3), execute (3), b4900 (3), back (3), into (3), implements (3), type (3), opcode (3), particular (3), did (3), predictions (3), until (3), latency (3), developed (3), citation (3), needed (3), histories (3), while (3), over (3), gshare (3), overriding (3), possibly (3), predicted (3), behavior (3), aliasing (3), spec (3), benchmarks (3), been (3), case (3), means (3), possible (3), outcome (3), usually (3), would (3), backward (3), often (3), always (3), attempts (3), statement (2), mobile (2), view (2), contact (2), privacy (2), policy (2), additional (2), terms (2), you (2), 2022 (2), english (2), version (2), wikidata (2), special (2), upload (2), related (2), tools (2), learn (2), article (2), search (2), talk (2), categories (2), 2013 (2), september (2), 2015 (2), short (2), description (2), https (2), index (2), branch_predictor (2), security (2), module (2), ppw (2), watt (2), voltage (2), scaling (2), integrated (2), circuitry (2), barrel (2), binary (2), datapath (2), microcode (2), registers (2), load (2), store (2), floating (2), point (2), arithmetic (2), heterogeneous (2), count (2), word (2), gpu (2), graphics (2), coprocessor (2), application (2), vector (2), types (2), associative (2), taxonomy (2), thread (2), parallelism (2), algorithm (2), visc (2), isa (2), arm (2), x86 (2), motorola (2), series (2), comparison (2), computing (2), fabric (2), secondary (2), hierarchy (2), nondeterministic (2), finite (2), hierarchical (2), jeff (2), software (2), fog (2), agner (2), describes (2), ev6 (2), complex (2), 2002 (2), paper (2), tradeoffs (2), 1996 (2), indexing (2), previous (2), external (2), spectre (2), virtually (2), micro (2), 9000 (2), supercomputer (2), 7030 (2), ars (2), technica (2), might (2), takes (2), black (2), championship (2), daniel (2), annual (2), acm (2), 2001 (2), mexico (2), perceptrons (2), romanian (2), towards (2), conference (2), networks (2), principles (2), 2004 (2), 1997 (2), june (2), 24th (2), skadron (2), techniques (2), doi (2), patt (2), 1991 (2), scott (2), 1993 (2), research (2), lab (2), jim (2), g4e (2), isbn (2), smith (2), s2cid (2), penalty (2), skewed (2), cheng (2), references (2), public (2), predication (2), researchers (2), vulnerability (2), project (2), late (2), had (2), overridden (2), vote (2), gskew (2), 21464 (2), ecc (2), tables (2), sufficient (2), like (2), these (2), rely (2), r8000 (2), trivial (2), slots (2), resolution (2), semantically (2), equivalent (2), represent (2), updated (2), popular (2), generally (2), require (2), 3090 (2), several (2), designs (2), unconditional (2), implemented (2), subsequent (2), implement (2), values (2), corresponding (2), condition (2), early (2), them (2), recovery (2), include (2), disadvantage (2), even (2), period (2), reduce (2), according (2), resource (2), growth (2), override (2), having (2), call (2), returns (2), without (2), among (2), preload (2), best (2), detected (2), meta (2), makes (2), patterns (2), final (2), either (2), remembers (2), past (2), globally (2), better (2), must (2), longer (2), make (2), instead (2), keeps (2), making (2), correlation (2), saturate (2), mmx (2), repetitive (2), initial (2), sequence (2), assume (2), third (2), indicating (2), comes (2), never (2), occurrences (2), contains (2), change (2), toward (2), twice (2), weakly (2), records (2), itself (2), aligned (2), assuming (2), uniform (2), distribution (2), wasted (2), direction (2), simplest (2), get (2), fall (2), hints (2), code (2), immediately (2), executing (2), passed (2), speculatively (2), known (2), sep (2), cookie, statistics, developers, disclaimers, text, available, under, apply, site, registered, trademark, profit, organization, wikimedia, foundation, inc, creative, commons, attribution, sharealike, license, edited, august, utc, українська, türkçe, svenska, suomi, српски, srpski, русский, português, norsk, bokmål, 日本語, lombard, italiano, 한국어, français, فارسی, español, eesti, deutsch, català, العربية, languages, printable, download, print, export, item, permanent, link, pages, here, recent, community, portal, contribute, donate, events, read, views, namespaces, log, create, account, contributions, logged, personal, menu, hidden, april, cs1, errors, missing, periodical, php, title, oldid, 1105472873, carrier, pin, grid, tick, tock, model, semiconductor, device, fabrication, electronics, chronology, gating, frequency, acpi, apm, pmu, switch, analog, boolean, mixed, shifter, sum, addressed, multiplier, adder, demultiplexer, multiplexer, horizontal, rom, write, hardwired, status, gate, glue, combinational, imc, controller, mmu, tlb, translation, lookaside, fpu, agu, generation, alu, functional, units, fifo, bus, coherence, replacement, policies, scratchpad, components, manycore, variable, slicing, 512, 256, 128, baseband, secure, cryptoprocessor, tpu, tensor, dsp, ppu, physics, vpu, vision, image, accelerator, accelerators, noc, psoc, programmable, mpsoc, multiprocessor, soc, soft, asip, ultra, low, notebook, microcontroller, embedded, pop, sip, mcm, cpld, fpoa, fpga, asic, pal, tile, stream, gpgpu, central, orders, magnitude, metrics, sups, synaptic, updates, tps, transactions, flops, operations, ips, ipc, cpi, transistor, spmd, mimd, misd, swar, simt, simd, sisd, flynn, cooperative, preemptive, hyperthreading, simultaneous, temporal, multithreading, distributed, process, task, scalar, serial, dependence, wide, issue, renaming, reservation, station, tomasulo, scoreboarding, out, false, sharing, structural, dependency, hazards, classic, operand, forwarding, stall, epiphany, tilera, 3x0, 390, 370, 360, lmc, microblaze, openrisc, itanium, unicore, m32r, etrax, cris, superh, clipper, powerpc, stanford, pdp, 68000, sets, addressing, modes, zisc, nisc, oisc, misc, epic, vliw, trips, edge, specific, cisc, orthogonal, neuromorphic, cognitive, multiprocessing, storage, virtual, huma, numa, endianness, transport, triggered, dataflow, modified, harvard, von, neumann, machines, belt, zeno, hypercomputation, probabilistic, post, universal, alternating, queue, deterministic, abstract, technologies, yee, alexander, overflow, andrews, 2007, reorganization, prevent, mispredicts, considerations, reconsidering, excellent, job, discussing, how, arrived, their, various, constraints, simulation, studies, demonstrates, impaired, block, ahead, gibbs, samuel, guardian, meltdown, worst, ever, bugs, affect, computers, felix, krishnan, sazeides, aggressive, uniprocessor, zen, now, actually, challenge, www, com, horizon, james, dave, pcgamesn, reviews, news, pricing, availability, brekelbaum, edward, rupley, wilkerson, chris, bryan, 34th, istanbul, turkey, scheduling, windows, 36th, san, diego, usa, walton, jarred, 2012, anandtech, trinity, review, a10, 4600m, hope, lin, hpca, monterrey, 197, 296, bucharest, academy, 287, 301, 1453, 8245, issn, science, technology, powerful, 1999, ijcnn, 2010, tse, sharangpani, apparatus, published, 014628, february, 2014, zenterprise, bc12, guide, march, sa22, 7832, operation, kanter, aaron, 2008, stokes, jon, look, centrino, driesen, karel, hölzle, urs, limits, holdings, cortex, a15, mpcore, reference, manual, section, sprangle, reducing, negative, interference, denver, martonosi, clark, october, robust, solution, philadelphia, parallel, compilation, egan, colin, steven, gordon, quick, anguera, 557, 570, 1016, s1383, 7621, 00095, training, albuquerque, puerto, rico, carnegie, mellon, improves, western, wrl, report, engr, oregonstate, edu, plusquellic, cmsc, 611, chapter, architectural, shen, john, lipasti, mikko, 2005, boston, 057064, 455, mcgraw, hill, higher, education, fundamentals, eyerman, eeckhout, 72217, 4244, 0186, 1109, ispass, 1620789, characterizing, requires, michaud, pierre, andré, uhlig, richard, 3712157, mutlu, onur, 447, lecture, parihar, raj, optimizations, chih, schemes, performances, malishevsky, alexey, beck, douglas, schmid, andreas, landry, eric, stibp, ibrs, restricted, speculation, ibpb, barrier, ibc, prefetching, key, cryptography, rsa, attacks, catastrophic, google, affecting, involves, extracting, private, leftover, cancelled, minimum, voting, another, otherwise, effectively, suffering, error, invalidated, refetched, became, important, introduction, 21064, earlier, just, loss, loses, r4000, r3000, r2000, announced, 1989, performs, cobol, released, around, 1982, during, opcodes, operator, indicated, determines, needs, rewrites, hinted, proper, obtains, hit, granted, patent, 435, 756, burroughs, 1960s, 1980s, beyond, took, however, addition, examples, incorporated, introduced, tom, mcwilliams, curt, widdoes, 1977, lawrence, livermore, national, independently, 1979, cdc, designed, 1950s, pre, executes, depended, production, untaken, changed, indicator, today, codes, designers, considered, hint, decided, against, provided, lookahead, reputation, less, stellar, blamed, required, 1985, exynos, samsung, infinity, art, competition, already, idea, simulators, taking, speed, tricks, computation, relatively, compared, microarchitectures, chooses, weights, concept, monchiero, tarjan, desmet, akkary, aasaraai, michael, etc, ability, exploit, requiring, linear, classical, exponential, reports, improvement, style, year, further, presented, feasible, piledriver, blaga, sibiu, layer, lvq, learning, buffers, handle, provide, inaccurate, cores, cost, lost, sometimes, dealt, slower, complicated, bigger, trade, off, normally, reads, mirror, typically, simply, choose, specialized, newer, kind, contributes, support, given, constructed, adding, immediate, displacement, value, zec12, controls, bottom, repeats, placed, top, goes, easily, detects, track, multiplication, map, situation, turn, suffer, those, beneficial, arrange, least, functions, analogous, caching, his, odd, outputs, xored, give, contentions, opposite, happen, share, combines, tests, indicate, nano, concatenating, gselect, atom, silvermont, concatenates, xors, sizes, rarely, grows, hence, big, exponentially, keep, record, diluted, irrelevant, uncorrelated, iii, replaced, quickly, arbitrary, invented, publication, become, variants, michigan, yale, rule, sub, sequences, 001001001, zeroes, consecutive, ones, consider, choosing, shift, three, decision, depend, upon, scenarios, works, efficiently, saturation, regularly, recurring, illustrated, represents, shown, referred, dimensional, indexed, decoded, maps, unique, though, imperfect, deviate, done, closing, mispredicted, diagram, states, essentially, although, accurate, flip, flop, destination, lines, add, nearly, complete, group, again, placements, slot, points, groups, simplicity, targets, respectively, handles, pseudorandom, pure, guarantee, cannot, improved, worsened, reordering, take, reorder, timing, compilers, gathered, run, decisions, compile, mpc7450, allow, inserted, tell, should, accepts, feature, abandoned, form, presumes, forward, lower, own, loops, pointing, evaluate, decode, fetches, define, utilize, implementations, found, predicts, solely, computed, decoding, encountered, encounters, seen, recognize, equal, stages, tend, quite, pipelines, increases, need, wait, enter, avoid, waste, trying, guessed, partially, starts, incurring, branching, place, continue, certain, calculated, fig, colored, boxes, independent, tries, definitively, improve, flow, play, critical, role, achieving, else, structure, confused, free, encyclopedia, wayback, http, archive, 20221016120420, wiki, timestamps, capture, fail, success, 2023, 2021, nov, oct, aug, 2026, 561, captures,

Text of the page (random words):
diction happen to share the same entry in the pattern history table 17 hybrid predictor edit a hybrid predictor also called combined predictor implements more than one prediction mechanism the final prediction is based either on a meta predictor that remembers which of the predictors has made the best predictions in the past or a majority vote function based on an odd number of different predictors scott mcfarling proposed combined branch prediction in his 1993 paper 12 on the spec 89 benchmarks such a predictor is about as good as the local predictor citation needed predictors like gshare use multiple table entries to track the behavior of any particular branch this multiplication of entries makes it much more likely that two branches will map to the same table entry a situation called aliasing which in turn makes it much more likely that prediction accuracy will suffer for those branches once you have multiple predictors it is beneficial to arrange that each predictor will have different aliasing patterns so that it is more likely that at least one predictor will have no aliasing combined predictors with different indexing functions for the different predictors are called gskew predictors and are analogous to skewed associative caches used for data and instruction caching loop predictor edit a conditional jump that controls a loop is best predicted with a special loop predictor a conditional jump in the bottom of a loop that repeats n times will be taken n 1 times and then not taken once if the conditional jump is placed at the top of the loop it will be not taken n 1 times and then taken once a conditional jump that goes many times one way and then the other way once is detected as having loop behavior such a conditional jump can be predicted easily with a simple counter a loop predictor is part of a hybrid predictor where a meta predictor detects whether the conditional jump has loop behavior indirect branch predictor edit an indirect jump instruction can choose among more than two branches some processors have specialized indirect branch predictors 18 19 newer processors from intel 20 and amd 21 can predict indirect branches by using a two level adaptive predictor this kind of instruction contributes more than one bit to the history buffer the zec12 and later z architecture processors from ibm support a branch prediction preload instruction that can preload the branch predictor entry for a given instruction with a branch target address constructed by adding the contents of a general purpose register to an immediate displacement value 22 23 processors without this mechanism will simply predict an indirect jump to go to the same target as it did last time 8 prediction of function returns edit a function will normally return to where it is called from the return instruction is an indirect jump that reads its target address from the call stack many microprocessors have a separate prediction mechanism for return instructions this mechanism is based on a so called return stack buffer which is a local mirror of the call stack the size of the return stack buffer is typically 4 16 entries 8 overriding branch prediction edit the trade off between fast branch prediction and good branch prediction is sometimes dealt with by having two branch predictors the first branch predictor is fast and simple the second branch predictor which is slower more complicated and with bigger tables will override a possibly wrong prediction made by the first predictor the alpha 21264 and alpha ev8 microprocessors used a fast single cycle next line predictor to handle the branch target recurrence and provide a simple and fast branch prediction because the next line predictor is so inaccurate and the branch resolution recurrence takes so long both cores have two cycle secondary branch predictors that can override the prediction of the next line predictor at the cost of a single lost fetch cycle the intel core i7 has two branch target buffers and possibly two or more branch predictors 24 neural branch prediction edit machine learning for branch prediction using lvq and multi layer perceptrons called neural branch prediction was proposed by lucian vintan lucian blaga university of sibiu 25 one year later he developed the perceptron branch predictor 26 the neural branch predictor research was developed much further by daniel jimenez 27 in 2001 27 the first perceptron predictor was presented that was feasible to implement in hardware the first commercial implementation of a perceptron branch predictor was in amd s piledriver microarchitecture 28 the main advantage of the neural predictor is its ability to exploit long histories while requiring only linear resource growth classical predictors require exponential resource growth jimenez reports a global improvement of 5 7 over a mcfarling style hybrid predictor 29 he also used a gshare perceptron overriding hybrid predictors 29 the main disadvantage of the perceptron predictor is its high latency even after taking advantage of high speed arithmetic tricks the computation latency is relatively high compared to the clock period of many modern microarchitectures in order to reduce the prediction latency jimenez proposed in 2003 the fast path neural predictor where the perceptron predictor chooses its weights according to the current branch s path rather than according to the branch s pc many other researchers developed this concept a seznec m monchiero d tarjan k skadron v desmet akkary et al k aasaraai michael black etc citation needed most of the state of the art branch predictors are using a perceptron predictor see intel s championship branch prediction competition 30 intel already implements this idea in one of the ia 64 s simulators 2003 31 the amd ryzen 32 33 34 multi core processor s infinity fabric and the samsung exynos processor include a perceptron based neural branch predictor history edit the ibm 7030 stretch designed in the late 1950s pre executes all unconditional branches and any conditional branches that depended on the index registers for other conditional branches the first two production models implemented predict untaken subsequent models were changed to implement predictions based on the current values of the indicator bits corresponding to today s condition codes 35 the stretch designers had considered static hint bits in the branch instructions early in the project but decided against them misprediction recovery was provided by the lookahead unit on stretch and part of stretch s reputation for less than stellar performance was blamed on the time required for misprediction recovery subsequent ibm large computer designs did not use branch prediction with speculative execution until the ibm 3090 in 1985 two bit predictors were introduced by tom mcwilliams and curt widdoes in 1977 for the lawrence livermore national lab s 1 supercomputer and independently by jim smith in 1979 at cdc 36 microprogrammed processors popular from the 1960s to the 1980s and beyond took multiple cycles per instruction and generally did not require branch prediction however in addition to the ibm 3090 there are several other examples of microprogrammed designs that incorporated branch prediction the burroughs b4900 a microprogrammed cobol machine released around 1982 was pipelined and used branch prediction the b4900 branch prediction history state is stored back into the in memory instructions during program execution the b4900 implements 4 state branch prediction by using 4 semantically equivalent branch opcodes to represent each branch operator type the opcode used indicated the history of that particular branch instruction if the hardware determines that the branch prediction state of a particular branch needs to be updated it rewrites the opcode with the semantically equivalent opcode that hinted the proper history this scheme obtains a 93 hit rate us patent 4 435 756 and others were granted on this scheme the vax 9000 announced in 1989 is both microprogrammed and pipelined and performs branch prediction 37 the first commercial risc processors the mips r2000 and r3000 and the earlier sparc processors do only trivial not taken branch prediction because they use branch delay slots fetched just one instruction per cycle and execute in order there is no performance loss the later r4000 uses the same trivial not taken branch prediction and loses two cycles to each taken branch because the branch resolution recurrence is four cycles long branch prediction became more important with the introduction of pipelined superscalar processors like the intel pentium dec alpha 21064 the mips r8000 and the ibm power series these processors all rely on one bit or simple bimodal predictors the dec alpha 21264 ev6 uses a next line predictor overridden by a combined local predictor and global predictor where the combining choice is made by a bimodal predictor 38 the amd k8 has a combined bimodal and global predictor where the combining choice is another bimodal predictor this processor caches the base and choice bimodal predictor counters in bits of the l2 cache otherwise used for ecc as a result it has effectively very large base and choice predictor tables and parity rather than ecc on instructions in the l2 cache the parity design is sufficient since any instruction suffering a parity error can be invalidated and refetched from memory the alpha 21464 38 ev8 cancelled late in design had a minimum branch misprediction penalty of 14 cycles it was to use a complex but fast next line predictor overridden by a combined bimodal and majority voting predictor the majority vote was between the bimodal and two gskew predictors in 2018 a catastrophic security vulnerability called spectre was made public by google s project zero and other researchers affecting virtually all modern cpus the vulnerability involves extracting private data from the leftover data caches of branch mispredictions 39 see also edit branch target predictor branch predication branch prediction analysis attacks on rsa public key cryptography instruction unit cache prefetching indirect branch control ibc indirect branch prediction barrier ibpb indirect branch restricted speculation ibrs single thread indirect branch predictor stibp references edit a b malishevsky alexey beck douglas schmid andreas landry eric dynamic branch prediction archived from the original on 2019 07 17 retrieved 2017 03 22 a b cheng chih cheng the schemes and performances of dynamic branch predictors pdf parihar raj branch prediction techniques and optimizations pdf archived from the original pdf on 2017 05 16 retrieved 2017 04 02 mutlu onur 2013 02 11 18 447 computer architecture lecture 11 branch prediction pdf michaud pierre seznec andré uhlig richard september 1996 skewed branch predictors s2cid 3712157 cite journal cite journal requires journal help eyerman s smith j e eeckhout l 2006 characterizing the branch misprediction penalty 2006 ieee international symposium on performance analysis of systems and software ieee 48 58 doi 10 1109 ispass 2006 1620789 isbn 1 4244 0186 0 s2cid 72217 shen john p lipasti mikko 2005 modern processor design fundamentals of superscalar processors boston mcgraw hill higher education pp 455 isbn 0 07 057064 7 a b c d e f fog agner 2016 12 01 the microarchitecture of intel amd and via cpus pdf p 36 retrieved 2017 03 22 the pentium 4 and the g4e an architectural comparison ars technica plusquellic jim cmsc 611 advanced computer architecture chapter 4 part v dynamic branch prediction web engr oregonstate edu archived from the original on 2019 07 17 retrieved 2017 11 01 a b c mcfarling scott june 1993 combining branch predictors pdf digital western research lab wrl technical report tn 36 new algorithm improves branch prediction 3 27 95 pdf carnegie mellon university retrieved 2016 02 02 yeh t y patt y n 1991 two level adaptive training branch prediction proceedings of the 24th annual international symposium on microarchitecture albuquerque new mexico puerto rico acm pp 51 61 egan colin steven gordon quick p anguera r vintan lucian december 2003 two level branch prediction using neural networks journal of systems architecture 49 12 15 557 570 doi 10 1016 s1383 7621 03 00095 x skadron k martonosi m clark d w october 2000 a taxonomy of branch mispredictions and alloyed prediction as a robust solution to wrong history mispredictions proceedings of the 2000 international conference on parallel architectures and compilation techniques philadelphia sprangle e et al june 1997 the agree predictor a mechanism for reducing negative branch history interference proceedings of the 24th international symposium on computer architecture denver cortex a15 mpcore technical reference manual section 6 5 3 indirect predictor arm holdings driesen karel hölzle urs 1997 06 25 limits of indirect branch prediction pdf stokes jon 2004 02 25 a look at centrino s core the pentium m pp 2 3 kanter aaron 2008 10 28 performance analysis for core 2 and k8 part 1 p 5 z architecture principles of operation pdf ibm march 2015 pp 7 40 7 43 sa22 7832 10 ibm zenterprise bc12 technical guide pdf ibm february 2014 p 78 wo 2000 014628 yeh tse yu sharangpani h p a method and apparatus for branch prediction using a second level branch prediction table published 2000 03 16 vintan lucian n 1999 towards a high performance neural branch predictor pdf proceedings international journal conference on neural networks ijcnn archived from the original pdf on 2019 07 13 retrieved 2010 12 02 vintan lucian n 2000 towards a powerful dynamic branch predictor pdf romanian journal of information science and technology bucharest romanian academy 3 3 287 301 issn 1453 8245 a b jimenez d a lin c 2001 dynamic branch prediction with perceptrons proceedings of the 7 th international symposium on high performance computer architecture hpca 7 monterrey nl mexico pp 197 296 walton jarred 2012 05 15 the amd trinity review a10 4600m a new hope anandtech a b jimenez daniel a december 2003 fast path based neural branch prediction pdf the 36th annual ieee acm international symposium on microarchitecture micro 36 san diego usa retrieved 2018 04 08 championship branch prediction brekelbaum edward rupley jeff wilkerson chris black bryan december 2002 hierarchical scheduling windows proceedings of the 34th international symposium on microarchitecture istanbul turkey james dave 2017 12 06 amd ryzen reviews news performance pricing and availability pcgamesn amd takes computing to a new horizon with ryzen processors www amd com retrieved 2016 12 14 amd s zen cpu is now called ryzen and it might actually challenge intel ars technica uk retrieved 2016 12 14 ibm stretch 7030 aggressive uniprocessor parallelism s 1 supercomputer micro architecture of the vax 9000 a b seznec felix krishnan sazeides design tradeoffs for the alpha ev8 conditional branch predictor gibbs samuel 2018 01 04 meltdown and spectre worst ever cpu bugs affect virtually a...