Instruction address translation and instruction prefetch engine

IN597600BActive Publication Date: 2026-07-31ADVANCED MICRO DEVICES INC
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
IN202217070863
Authority / Receiving Office
IN · IN
Patent Type
Patents
Current Assignee / Owner
Priority Date
2020-06-26
Filing Date
2022-12-08
Publication Date
2026-07-31
Estimated Expiration
2041-06-25

AI Technical Summary

Technical Problem

Branch instructions in microprocessors cause delays due to the time difference between fetching and evaluating branch outcomes, which existing branch prediction methods fail to adequately mitigate, leading to inefficiencies in instruction fetch throughput.

Method used

The implementation of a method that determines instruction addresses for a primary branch prediction path, requests level 0 TLB caching for these addresses, and further requests level 1 TLB caching for alternate and lookahead control flow path addresses to reduce latency and misprediction penalties by prefetching address translations.

Benefits of technology

This approach reduces the latency associated with branch mispredictions and improves instruction fetch throughput by caching translations in higher levels of the TLB hierarchy, thereby minimizing the impact of misses in the level 0 TLB.

✦ Generated by Eureka AI based on patent content.
Patent Text Reader

Abstract

Techniques for performing instruction fetch operations are provided. The techniques include determining instruction addresses for a primary branch prediction path; requesting that a level 0 translation lookaside buffer ("TLB") caches address translations for the primary branch prediction path; determining either or both of alternate control flow path instruction addresses and lookahead control flow path instruction addresses; and requesting that either the level 0 TLB or an alternative level TLB caches address translations for either or both of the alternate control flow path instruction addresses and the lookahead control flow path instruction addresses.
Need to check novelty before this filing date? Find Prior Art

Description

BACKGROUND

[0002] In a microprocessor, instructions are fetched for executionsequentially until a branch occurs. A branch causes a change in the address fromwhich instructions are fetched and may be associated with delays in instructionfetch throughput. For example, branches may need to be evaluated to determinewhether to take the branch as well as what the branch destination is. However,branches cannot be evaluated until the branch has entered the instructionexecution pipeline. Branch delays are associated with the difference between thetime that the branch is fetched and the time that the branch is evaluated todetermine the outcome of that branch and thus what instructions need to befetched next. Branch prediction helps to mitigate this delay by predicting theexistence and outcome of a branch instruction based upon instruction address.Improvements to the operation of branch predictors are therefore desirable.BRIEF DESCRIPTION OF THE DRAWINGS

[0003] A more detailed understanding may be had from the followingdescription, given by way of example in conjunction with the accompanyingdrawings wherein:

[0004] Figure 1 is a block diagram of an example device in which one or moredisclosed embodiments may be implemented;

[0005] Figure 2 is a block diagram of an instruction execution pipeline,located within the processor of Figure 1;

[0006] Figure 3A illustrates an example instruction fetch subsystem;

[0007] Figure 3B illustrates another example instruction fetch system; and

[0008] Figure 4 is a flow diagram of a method for performing instructionfetch operations, according to an example.DETAILED DESCRIPTION

[0009] Techniques for performing instruction fetch operations are provided.The techniques include determining instruction addresses for a primary branchprediction path; requesting that a level 0 translation lookaside buffer ("TLB")caches address translations for the primary branch prediction path; determiningeither or both of alternate control flow path instruction addresses and lookaheadcontrol flow path instruction addresses; and requesting that a level 1 TLB cachesaddress translations for either or both of the alternate control flow pathinstruction addresses and the lookahead control flow path instruction addresses.

[0010] Figure 11is a block diagram of an example device 100 in which aspectsof the present disclosure are implemented. The device 100 includes, for example,a computer, a gaming device, a handheld device, a set-top box, a television, amobile phone, or a tablet computer. The device 100 includes a processor 102, amemory 104, a storage device 106, one or more input devices 108, and one or moreoutput devices 110. The device 100 may also optionally include an input driver 112and an output driver 114. It is understood that the device 100 may includeadditional components not shown in Figure 1.

[0011] The processor 102 includes a central processing unit (CPU), agraphics processing unit (GPU), a CPU and GPU located on the same die, or oneor more processor cores, wherein each processor core is a CPU or a GPU. Thememory 104 may be located on the same die as the processor 102, or may be locatedseparately from the processor 102. The memory 104 includes a volatile or non-volatile memory, for example, random access memory (RAM), dynamic RAM, or acache.

[0012] The storage device 106 includes a fixed or removable storage, forexample, a hard disk drive, a solid state drive, an optical disk, or a flash drive. Theinput devices 108 include a keyboard, a keypad, a touch screen, a touch pad, adetector, a microphone, an accelerometer, a gyroscope, a biometric scanner, or anetwork connection (e.g, a wireless local area network card for transmissionand / or reception of wireless IEEE 802 signals). The output devices 110 include adisplay, a speaker, a printer, a haptic feedback device, one or more lights, anantenna, or a network connection (e.g., a wireless local area network card fortransmission and / or reception of wireless IEEE 802 signals).

[0013] The input driver 112 communicates with the processor 102 and theinput devices 108, and permits the processor 102 to receive input from the inputdevices 108. The output driver 114 communicates with the processor 102 and theoutput devices 110, and permits the processor 102 to send output to the outputdevices 110. It is noted that the input driver 112 and the output driver 114 areoptional components, and that the device 100 will operate in the same manner ifthe input driver 112 and the output driver 114 are not present.

[0014] Figure 2 is a block diagram of an instruction execution pipeline 200,located within the processor 102 of Figure 1. Although one specific configurationfor an instruction execution pipeline 200 is illustrated, it should be understoodthat any instruction execution pipeline 200 that uses a branch target buffer toprefetch instructions into an instruction cache falls within the scope of the presentdisclosure. The instruction execution pipeline 200 retrieves instructions frommemory and executes the instructions, outputting data to memory and modifyingthe state of elements associated with the instruction execution pipeline 200, suchas registers within register file 218.

[0015] The instruction execution pipeline 200 includes an instruction fetchunit 204 that fetches instructions from system memory (such as memory 104)using an instruction cache 202, a decoder 208 that decodes fetched instructions,functional units 216 that perform calculations to process the instructions, a loadstore unit 214 that loads data from or store data to system memory via a datacache 220, and a register file 218, which includes registers that store working datafor the instructions. A reorder buffer 210 tracks instructions that are currentlyin-flight and ensures in-order retirement of instructions despite allowing out-of-order execution while in-flight. The term "in-flight instructions" refers toinstructions that have been received by the reorder buffer 210 but have not yethad results committed to the architectural state of the processor (e.g., resultswritten to a register file, or the like). Reservation stations 212 maintain in-flightinstructions and track instruction operands. When all operands are ready forexecution of a particular instruction, reservation stations 212 send the instructionto a functional unit 216 or a load / store unit 214 for execution. Completedinstructions are marked for retirement in the reorder buffer 210 and are retiredwhen at the head of the reorder buffer queue 210. Retirement refers to the act ofcommitting results of an instruction to the architectural state of the processor.For example, writing an addition result to a register, by an add instruction,writing a loaded value to a register by a load instruction, or causing instructionflow to jump to a new location, by a branch instruction, are all examples ofretirement of the instruction.

[0016] Various elements of the instruction execution pipeline 200communicate via a common data bus 222. For example, the functional units 216and load / store unit 214 write results to the common data bus 222 which may beread by reservation stations 212 for execution of dependent instructions and bythe reorder buffer 210 as the final processing result of an in-flight instruction thathas finished execution. The load / store unit 214 also reads data from the commondata bus 222. For example, the load / store unit 214 reads results from completedinstructions from the common data bus 222 and writes the results to memory viathe data cache 220 for store instructions.

[0017] Typically, the instruction fetch unit 204 fetches instructionssequentially in memory. Sequential control flow may be interrupted by branchinstructions, which causes the instruction pipeline 200 to fetch instructions froma non-sequential address. Branch instructions may be conditional, causing abranch only if a particular condition is satisfied, or non-conditional, and mayspecify a target directly or indirectly. Direct targets are specified by constants inthe instruction byte itself and indirect targets are specified by values in a registeror in memory. Direct and indirect branches may be conditional or non-conditional.

[0018] As described above, the instruction fetch unit 204 fetchesinstructions from an instruction cache 202 for execution by the rest of the pipeline200. In general, instruction fetch involves several phases: branch prediction,address translation, and micro-operation fetch.

[0019] Branch prediction involves predicting the control flow path througha program stored in memory. More specifically, as is generally known, a programis laid out in a certain manner in memory. Execution without control flow changeinvolves sequentially fetching instructions and executing those instructions.Control flow instructions such as branches and jumps have the capability to alterthe address from which instructions are fetched. Such instructions "jump" to atarget address, either unconditionally or based on the result of a conditionalevaluation. The branch predictor 302 utilizes heuristics to predict the outcomesof these branch and jump instructions, in order to predict a most likely paththrough the program.

[0020] In some implementations, the output of branch prediction isinstruction addresses in a virtual address space. Address translation involvestranslating the addresses of these instructions from the virtual address space to aphysical address space, so that the instructions can be loaded from memory. Ingeneral, an address translation system includes a page table walker that acceptsvirtual addresses, looks those virtual addresses up in a page table, and providesphysical addresses in response. Caches referred to as translation lookaside buffers(TLB's) cache address translations for faster access. The TLBs are, in someimplementations, arranged in a hierarchy (including, for example, level 0 (LO)TLBs, level 1 (IL1) TLBs, level 2 (L.2) TLBs, and so on).

[0021] Micro-operation fetch includes fetching the instructions frommemory based on the physical addresses. In some implementations, this fetchingoccurs through a cache hierarchy. Micro-operation fetch also includes translatingthe fetched instructions into micro-operations via instruction-to-operationtranslation logic. In some implementations, instruction-to-operation translationsare cached (for example, in a micro-operation cache) for quick retrieval at asubsequent time.

[0022] As described, a translation lookaside buffer fetches instructionaddress translations for instructions in the predicted path. Techniques areprovided herein for improving the latency with which address translations arefetched.

[0023] Figure 3A illustrates an example instruction fetch subsystem 300.The instruction fetch subsystem includes a branch predictor 302, an addresstranslation subsystem 303, and a micro-operation subsystem 305.

[0024] The address translation subsystem 3083 includes a level 0 TLB 304, alevel 1 TLB 306, and higher address translation levels 308, which, in variousimplementations, include higher TLB levels and / or a page table walker thatperforms address translation based on the contents of one or more page tables.

[0025] The micro-operation subsystem 305 includes a lowest levelinstruction cache 310 and higher cache levels 312. In some examples, the lowestlevel instruction cache 310 includes a level 0 instruction cache and an operationcache 310. In some examples, the lowest level instruction cache 310 also includesan instruction-to-micro-operation decoder (not shown) that decodes instructionsinto micro-operations for execution by the instruction pipeline 200.

[0026] The branch predictor performs branch prediction operations,identifying predicted path addresses which are the instruction addresses fromwhich the instruction fetch subsystem 300 is to fetch instructions. As shown, thebranch predictor 302 sends these predicted addresses to the lowest level TLLB 304for address translation. The lowest level TLB 304 functions as a cache, storing asubset of all translations in the page tables for quick translation. However, it ispossible for the branch predictor 302 to request a translation that is not stored inthe lowest level TLB 304. In that instance, the lowest level TLB 304 requests suchtranslation from the level 1 TLB 306. In the situation that the level 1 TLB 306has such translation, the level 1 TLB 306 returns that translation to the lowestlevel TLB 304. In the instance that the level 1 TLB 306 does not have thetranslation, then the level 1 TLB 306 requests the translation from the higheraddress translation levels 308, and so on. The LO TLB 304 provides instructionaddress translations to the micro-operation subsystem 305 to obtain theinstructions at that address and convert those instructions into micro-operations.

[0027] In the instruction fetch subsystem 300, the branch predictor 302requests that the level 1 TLB 306 prefetch certain address translations, in orderto reduce the performance decrease associated with a miss in the level 0 TLB 304.Specifically, the branch predictor 302 prefetches one or both of addresstranslations from alternate branch paths and address translations that are in"lookahead" branch paths, which are further advanced in program control flowthan the "primary" branch path instruction addresses sent to the level 0 TLB 304.The alternate branch paths may be the code which is not the same as the predictedpath. If the branch was predicted taken, then the alternate path is the not-takenpath. If the branch was predicted not-taken, then the alternate path is the takenpath. For indirect branches, the alternate path may be a lower confidence targetof the indirect branch predictor.

[0028] In some implementations, the branch predictor 302 selects particularalternate paths for which to fetch address translations into the level 1 TLB 306.In some implementations, the branch predictor 302 maintains or calculatesinformation indicating the confidence that any particular branch decision (forconditional branches) and / or branch target is correct. In some implementations,the alternate branch paths for which address translations are prefetched into thelevel 1 TLB 306 include the paths for which the confidence level is relatively low(the maintained or calculated confidence is below a threshold). In other words,the branch predictor 302 fetches translations into the level 1 TLB 306 for alternatepaths past branch points where the confidence for the path that is taken at thatbranch point is lower than the threshold. In such instances, it is relatively likelythat a branch misprediction will occur, and thus it is relatively likely that thetranslations in the level 1 TLB 306 will be used. In some implementations, thebranch predictor 302 periodically (e.g., once every certain number of cycles)includes, in the alternate paths, not-taken paths for high-confidence branchpredictions, since such high confidence branch predictions will sometimes bewrong.

[0029] In some implementations, the branch predictor 302 detects a loopexecution and attempts to maintain the address translation for the loop exit targetin the L.1 TLB 306. Note that this loop exit may be a taken branch to another pageor it may be a not-taken branch near the end of a page. In some implementations,the branch predictor 302 detects a loop execution by detecting that control flowrepeatedly returns to the same code. The loop exit is the code to be executed whenthe loop is finished. In various implementations, the branch predictor 302periodically requests that the L1 TLB 306 fetch and / or store the addresstranslation for code that will be executed after the loop exit.

[0030] In some implementations, the branch predictor 302 limits thetranslations fetched from alternate paths based on a branch depth. The branchdepth indicates the number of branches beyond a fork point which the branchpredictor 302 obtains translations. In various examples, the branch depth is atunable parameter or a fixed parameter. In various implementations, the tunableparameter is set based on a runtime heuristic or based on a parameter set by anentity such as the operating system or an application.

[0031] In some implementations, the branch predictor 302 limits thetranslations fetched from alternate paths based on an alternate path translationsnumber limit. The alternate path translation number limit indicates the limit fortotal number of translations in all alternate paths that can be prefetched into theL1 TLB 306 in a given sliding time window.

[0032] As described above, the branch predictor 302 alternatively oradditionally prefetches address translations from the "primary" (taken) path thatare further in program order than the predicted path addresses requested to bestored into the level 0 TLB 304. These are sometimes referred to herein as"lookahead" address translations or the lookahead path.

[0033] By fetching, into the level 1 TLB 306, addresses that are further intime and / or that are in alternate paths than those fetched into the level 0 TLB304, the penalty of branch mispredictions, or the overall latency of a miss in thelevel 0 TLB 304, is reduced, since resulting misses in the level 0 TLB 304 wouldbe able to fetch data from the level 1 TLB 306, rather than from higher levels ofthe address translation subsystem 303.

[0034] The micro-operation subsystem 305 fetches micro-operations forexecution by the instruction pipeline 200. The micro-operation subsystem 305includes a lowest level instruction cache 310. In some implementations, the lowestlevel instruction cache 305 includes a lowest level instruction cache memory thatcaches instructions for decoding and an operation cache that caches previouslydecoded instructions. In the situation that the lowest level instruction cache 310receives an instruction for which decoded micro-operations already exist in theoperation cache, the lowest level instruction cache 310 transmits thoseinstructions to the rest of the instruction pipeline 200 for execution. In thesituation that the lowest level instruction cache 310 receives an instruction forwhich decoded micro-operations do not already exist in the operation cache, thelowest level instruction cache 310 fetches the instructions from the cachehierarchy, starting with the lowest level instruction cache memory of the lowestlevel instruction cache 310 and proceeding up the hierarchy to the higher cachelevels 312, and, once received, decodes those instructions into micro-operations fortransmission to the instruction pipeline 200 for execution.

[0035] Figure 3B illustrates another example instruction fetch system 350.The instruction fetch system 350 includes a branch predictor 352, an addresstranslation subsystem 353, and a micro-operation subsystem 355. The addresstranslation subsystem 353 includes a level 0 TLB 354, a level 1 TLB 356, andhigher address translation levels 358. The micro-operation subsystem 355includes a lowest level instruction cache 360, a level 1 cache 362, and higher cachelevels 364.

[0036] The branch predictor 352 performs similar operations as with thebranch predictor 302 of Figure 3A, with regards to the address translationsubsystem 353. Specifically, the branch predictor 352 generates addresses for a"primary" path, sending those addresses to the level 0 TLB 354 for addresstranslation. The branch predictor 352 also generates addresses for alternatepaths and / or lookahead paths for transmission to the level 1 TLB 356. The level0 TLB 354 and level 1 TLB 356, as well as the higher address translation levels358 function in a similar manner s with the level 0 TLB 304, level 1 TLB 306, andhigher address translation levels 308 of Figure 3A.

[0037] In addition to providing the address translation subsystem 353 withindications of the instruction of the alternate path addresses and optionallookahead addresses, the branch predictor 352 also provides data indicating theseinstructions to the level 1 cache 362 for pre-fetching. Thus, in addition toprefetching certain address translations into the level 1 TLB 356, in theinstruction fetch system 350, the level 1 cache 362 pre-fetches the instructions ofthe alternate path and / or lookahead path.

[0038] The instruction fetch system 350 is similar to the instruction fetchsystem 300 of Figure 3A, except that the branch predictor provides information tothe micro-operation subsystem 305 for fetching instructions from the alternatepaths and / or the lookahead paths. Thus in the system 350 of Figure 3B, the micro-operation subsystem 305 fetches instructions from the alternate paths and / orlookahead paths in any of the instances that the address translation subsystem353 prefetches corresponding address translations (where "corresponding addresstranslations" to instructions are address translations for the addresses at whichthe instructions are found). Put differently, the present disclosure contemplatesimplementations of an instruction fetch system 350 that prefetches instructionsaccording to any combination of situations in which the address translationsubsystem 35 prefetches corresponding address translations discussed herein.

[0039] Figure 4 is a flow diagram of a method 400 for performing instructionfetch operations, according to an example. Although the method 400 is describedwith respect to the system of Figures 1-3B, those of skill in the art will recognizethat any system, configured to perform the steps of the method 400 in anytechnically feasible order, falls within the scope of the present disclosure.

[0040] The method 400 begins at step 402, where a branch predictor 302determines instruction addresses for a primary branch prediction path. As statedelsewhere herein, branch prediction is a known technique whereby a branchpredictor uses branch prediction data such as which instruction addresses areknown or predicted to be branch instructions, information derived from at leastportions of branch addresses, targets (instruction addresses jumped to) branchinstructions previously branched to, whether conditional branches werepreviously taken or not taken, as well as previously-seen control flow patterns, tomake a prediction as to the destination of a particular branch. In someimplementations, the branch predictor 302 stores branch prediction data on acache line basis, where the data includes information indicating the number ofbranches in that cache line and information allowing the branch predictor 302 todetermine predicted targets for those branches.

[0041] The branch predictor 302 determines a primary control flow path,which includes instruction addresses for the control flow path that the branchpredictor 302 predicts the instruction execution pipeline 200 will actually follow.This primary control flow path is distinguished from one or more alternate paths,which include at least a portion of instructions that the branch predictor 302predicts the instruction execution pipeline 200 will not follow. The primary controlflow path is distinguished from the lookahead control flow path in that thelookahead control flow path includes instructions further ahead in time in thepredicted control flow path than the primary control flow path.

[0042] At step 404, the branch predictor 302 requests that a level 0translation lookaside buffer caches the address translations for the primarybranch prediction path. In some implementations, the level 0 translationlookaside buffer is the lowest level TLB available for use by the branch predictor302.

[0043] At step 406, the branch predictor 302 determines either or both ofalternate control flow path instruction addresses and lookahead control flow pathinstruction addresses. For lookahead control flow path instruction addresses, insome implementations, the branch predictor 302 performs the same operations fordetermining the instruction addresses of the primary branch prediction path, butobtains instructions that are further in the program control flow than the primarybranch prediction path (also referred to as "further in time" in this document). Foralternate control flow path instructions, the branch predictor 302 identifies controlflow for conditional branch decisions other than the ones that are predicted (e.g,identifies a control flow path past a branch for the "not-taken" direction if thebranch is predicted to be taken) and / or identifies branch targets for indirectbranches that are not predicted to be taken. An indirect branch specifies a branchtarget based on a variable, not on a constant, and therefore the branch predictor302 sometimes stores multiple candidate targets for an indirect branch. In someinstances, the branch predictor predicts a certain target is the taken target for anindirect branch and, for the alternate control flow paths, selects one or more not-taken targets for such branches as indicating the control flow paths for which toobtain instruction addresses.

[0044] At step 408, the branch predictor 302 requests that a level 1 TLE 306cache the translations for the instruction addresses determined at step 406. Withthese addresses cached, the latency from the level 1 TLB 306 to higher portions ofthe address translation subsystem 303 is hidden.

[0045] In some implementations, the granularity for which the branchpredictor 302 requests new address translations is at the memory page level andnot the individual instruction level, since typically, a virtual address and physicaladdress share the same in-page offset but differ by page number. Thus, in someimplementations, the branch predictor 302 sends, to the address translationsubsystem 303, new page addresses for translation, and does not send the samepage address for translation for multiple instructions encountered in a given timeperiod (for example, a given cycle).

[0046] In the implementation of Figure 3B, the L.L1 TLB 356 (for example, atthe direction of the branch predictor 302 or other logic) requests that the L1 cache362 fetch instructions at the addresses corresponding to the alternate path and / orlookahead path. This occurs in addition to the action of requesting the L.1 TLB 356cache address translations for the alternate path and / or lookahead path.

[0047] Regarding the flow diagram 400, it should be understood that actionsdescribed as being performed by the branch predictor 302, 1.0 TL 304, 1.1 TLB 306,and higher address translation levels 308, are performed by the branch predictor352, LO TLB 354, L1 TLB 356, and higher address translation levels 358 in theimplementation of Figure 3B.

[00438] In some examples, the steps of Figure 4 are performed in a singleclock cycle or in certain period of time, and repeated every cycle or every certainperiod of time.

[0049] It should be understood that many variations are possible based onthe disclosure herein. Although features and elements are described above inparticular combinations, each feature or element may be used alone without theother features and elements or in various combinations with or without otherfeatures and elements. In some examples, although alternate-path or fetch-aheadaddress translations or instructions are described as being placed into a level 1TLB or level 1 instruction cache, it should be understood that these items could beplaced into either or both of a different TLB level (e.g., level 0 TLB or level 2 TLB,or level 0 instruction cache or level 2 instruction cache) or a different instructioncache level.

[0050] The various functional units illustrated in the figures and / ordescribed herein (including, where appropriate, the processor 102, the input driver112, the input devices 108, the output driver 114, the output devices 110, theinstruction cache 202, the instruction fetch unit 204, the decoder 208, the reorderbuffer 210, the reservation stations 212, the data cache 220, the load / store unit214, the functional units 216, the register file 218, the common data bus 222, thebranch predictors, and the caches) are implemented as hardware circuitry, or,where appropriate, as software executing on a processor or a combination ofsoftware executing on a processor and hardware. The methods provided may beimplemented in a general purpose computer, a Processor, or a Processor core.Suitable processors include, by way of example, a general purpose processor, aspecial purpose processor, a conventional processor, a digital signal processor(DSP), a plurality of microprocessors, one or more microprocessors in associationwith a DSP core, a controller, a microcontroller, Application Specific IntegratedCircuits (ASICs), Field Programmable Gate Arrays (FPGAs) circuits, any othertype of integrated circuit (IC), and / or a state machine. Such processors may bemanufactured by configuring a manufacturing process using the results ofprocessed hardware description language (HDL) instructions and otherintermediary data including netlists (such instructions capable of being stored ona computer readable media). The results of such processing may be maskworksthat are then used in a semiconductor manufacturing process to manufacture aprocessor which implements aspects of the embodiments.

[0051] The methods or flow charts provided herein may be implemented ina computer program, software, or firmware incorporated in a non-transitorycomputer-readable storage medium for execution by a general purpose computeror a processor. Examples of non-transitory computer-readable storage mediumsinclude a read only memory (ROM), a random access memory (RAM), a register,cache memory, semiconductor memory devices, magnetic media such as internalhard disks and removable disks, magneto-optical media, and optical media suchas CD-ROM disks, and digital versatile disks (DVDs).

Claims

1. A method for performing instruction fetch operations, the method comprising: determining instruction addresses for a primary branch prediction path; requesting that a level 0 translation lookaside buffer ("TLB") caches address translations for the primary branch prediction path; determining either or both of alternate control flow path instruction addresses and lookahead control flow path instruction addresses; and requesting that either the level 0 TLB or an alternative TLB level caches address translations for either or both of the alternate control flow path instruction addresses and the lookahead control flow path instruction addresses.

2. The method of claim 1, wherein the primary branch prediction path includes control flow paths past branch targets predicted as being taken and not past branch targets predicted as being not-taken.

3. The method of claim 1, wherein the alternate control flow path instruction addresses comprise instruction addresses for one or more alternate control flow paths defined by control flow past one or more branch targets predicted as not-taken.

4. The method of claim 1, wherein the lookahead control flow path instruction addresses comprise instruction addresses for control flow paths past branch targets predicted as taken, wherein at least one of the instruction addresses is at a point in program control flow that is past the primary branch prediction path.

5. The method of claim 1, further comprising one or more of: requesting an instruction cache store instructions corresponding to the alternate control flow path instruction addresses; and requesting the instruction cache store instructions corresponding to the lookahead control flow path instruction addresses.

6. The method of claim 1, wherein the alternate control flow path instruction addresses include one or more of a loop exit, and a not-taken branch target for a branch whose taken-target has a low confidence metric.

7. The method of claim 1, further comprising; periodically including, in the alternate control flow path instruction addresses, a not-taken branch target for a branch whose taken-target has a high confidence metric.

8. The method of claim 1, wherein the instruction addresses for the primary path, the instruction addresses for the alternate control flow paths, and the instruction addresses for the lookahead control flow paths comprise page addresses.

9. The method of claim 1, wherein the alternate control flow path instruction addresses include not-taken addresses for indirect branches.

10. An instruction fetch system for performing instruction fetch operations, the instruction fetch system comprising: a level O translation lookaside buffer ("T'LB"); and a branch predictor configured to: determine instruction addresses for a primary branch prediction path; request that the level 0 translation lookaside buffer ("TLB") caches address translations for the primary branch prediction path; determine either or both of alternate control flow path instruction addresses and lookahead control flow path instruction addresses; and request that either the level 0 TLB or an alternative TLB level caches address translations for either or both of the alternate control flow path instruction addresses and the lookahead control flow path instruction addresses.

11. The instruction fetch system of claim 10, wherein the primary branch prediction path includes control flow paths past branch targets predicted as being taken and not past branch targets predicted as being not-taken.

12. The instruction fetch system of claim 10, wherein the alternate control flow path instruction addresses comprise instruction addresses for one or more alternate control flow paths defined by control flow past one or more branch targets predicted as not-taken.

13. The instruction fetch system of claim 10, wherein the lookahead control flow path instruction addresses comprise instruction addresses for control flow paths past branch targets predicted as taken, wherein at least one of the instruction addresses is at a point in program control flow that is past the primary branch prediction path.

14. The instruction fetch system of claim 10, wherein the branch predictor is further configured to perform one or more of: requesting an instruction cache store instructions corresponding to the alternate control flow path instruction addresses; and requesting the instruction cache store instructions corresponding to the lookahead control flow path instruction addresses.

15. The instruction fetch system of claim 10, wherein the alternate control flow path instruction addresses include one or more of a loop exit, and a not-taken branch target for a branch whose taken-target has a low confidence metric.

16. The instruction fetch system of claim 10, wherein the branch predictor is further configured to: periodically include, in the alternate control flow path instruction addresses, a not-taken branch target for a branch whose taken-target has a high confidence metric.

17. The instruction fetch system of claim 10, wherein the instruction addresses for the primary path, the instruction addresses for the alternate control flow paths, and the instruction addresses for the lookahead control flow paths comprise page addresses.

18. The instruction fetch system of claim 10, wherein the alternate control flow path instruction addresses include not-taken addresses for indirect branches.

19. A system for performing instruction fetch operations, the processor comprising: an instruction execution pipeline; and a memory storing instructions for execution by the instruction execution pipeline, wherein the instruction execution pipeline is configured to: determine instruction addresses for a primary branch prediction path; request that a level 0 translation lookaside buffer ("TLB") caches address translations for the primary branch prediction path; determine either or both of alternate control flow path instruction addresses and lookahead control flow path instruction addresses; and request that either the level 0 TLB or an alternative TLB level caches address translations for either or both of the alternate control flow path instruction addresses and the lookahead control flow path instruction addresses.

20. The system of claim 19, wherein the primary branch prediction path includes control flow paths past branch targets predicted as being taken and not past branch targets predicted as being not-taken.