US20080034190A1 - Method and apparatus for suspending execution of a thread until a specified memory access occurs - Google Patents
Method and apparatus for suspending execution of a thread until a specified memory access occurs Download PDFInfo
- Publication number
- US20080034190A1 US20080034190A1 US11/891,076 US89107607A US2008034190A1 US 20080034190 A1 US20080034190 A1 US 20080034190A1 US 89107607 A US89107607 A US 89107607A US 2008034190 A1 US2008034190 A1 US 2008034190A1
- Authority
- US
- United States
- Prior art keywords
- thread
- monitor
- processor
- instruction
- address
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Abandoned
Links
- 238000000034 method Methods 0.000 title claims abstract description 28
- 230000004044 response Effects 0.000 claims abstract description 18
- 238000005192 partition Methods 0.000 claims description 17
- 239000000872 buffer Substances 0.000 claims description 11
- 238000012544 monitoring process Methods 0.000 claims description 10
- 238000013519 translation Methods 0.000 claims description 10
- 230000001419 dependent effect Effects 0.000 claims description 3
- 238000013461 design Methods 0.000 description 12
- 238000012360 testing method Methods 0.000 description 8
- 238000010586 diagram Methods 0.000 description 7
- 239000003795 chemical substances by application Substances 0.000 description 6
- 238000012545 processing Methods 0.000 description 6
- 238000000638 solvent extraction Methods 0.000 description 5
- 239000000725 suspension Substances 0.000 description 5
- 230000008859 change Effects 0.000 description 4
- 238000001514 detection method Methods 0.000 description 4
- 238000004519 manufacturing process Methods 0.000 description 4
- 238000004088 simulation Methods 0.000 description 4
- 230000004913 activation Effects 0.000 description 3
- 230000007246 mechanism Effects 0.000 description 3
- 230000003287 optical effect Effects 0.000 description 3
- 230000005540 biological transmission Effects 0.000 description 2
- 230000001934 delay Effects 0.000 description 2
- 238000011010 flushing procedure Methods 0.000 description 2
- 230000006870 function Effects 0.000 description 2
- 238000001693 membrane extraction with a sorbent interface Methods 0.000 description 2
- 239000000203 mixture Substances 0.000 description 2
- 230000009471 action Effects 0.000 description 1
- 230000003213 activating effect Effects 0.000 description 1
- 230000003466 anti-cipated effect Effects 0.000 description 1
- 230000008901 benefit Effects 0.000 description 1
- 230000001427 coherent effect Effects 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 238000012938 design process Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 230000010354 integration Effects 0.000 description 1
- 230000005055 memory storage Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 239000004065 semiconductor Substances 0.000 description 1
- 230000008054 signal transmission Effects 0.000 description 1
- 238000010200 validation analysis Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/46—Multiprogramming arrangements
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/30—Arrangements for executing machine instructions, e.g. instruction decode
- G06F9/30003—Arrangements for executing specific machine instructions
- G06F9/30076—Arrangements for executing specific machine instructions to perform miscellaneous control operations, e.g. NOP
- G06F9/3009—Thread control instructions
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/30—Arrangements for executing machine instructions, e.g. instruction decode
- G06F9/30003—Arrangements for executing specific machine instructions
- G06F9/30076—Arrangements for executing specific machine instructions to perform miscellaneous control operations, e.g. NOP
- G06F9/30079—Pipeline control instructions, e.g. multicycle NOP
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/30—Arrangements for executing machine instructions, e.g. instruction decode
- G06F9/38—Concurrent instruction execution, e.g. pipeline, look ahead
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/30—Arrangements for executing machine instructions, e.g. instruction decode
- G06F9/38—Concurrent instruction execution, e.g. pipeline, look ahead
- G06F9/3836—Instruction issuing, e.g. dynamic instruction scheduling or out of order instruction execution
- G06F9/3851—Instruction issuing, e.g. dynamic instruction scheduling or out of order instruction execution from multiple instruction streams, e.g. multistreaming
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/46—Multiprogramming arrangements
- G06F9/48—Program initiating; Program switching, e.g. by interrupt
- G06F9/4806—Task transfer initiation or dispatching
- G06F9/4812—Task transfer initiation or dispatching by interrupt, e.g. masked
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/46—Multiprogramming arrangements
- G06F9/48—Program initiating; Program switching, e.g. by interrupt
- G06F9/4806—Task transfer initiation or dispatching
- G06F9/4843—Task transfer initiation or dispatching by program, e.g. task dispatcher, supervisor, operating system
Definitions
- the present disclosure pertains to the field of processors. More particularly, the present disclosure pertains to multi-threaded processors and techniques for temporarily suspending the processing of one thread in a multi-threaded processor.
- a multi-threaded processor is capable of processing multiple different instruction sequences concurrently.
- a primary motivating factor driving execution of multiple instruction streams within a single processor is the resulting improvement in processor utilization.
- Highly parallel architectures have developed over the years, but it is often difficult to extract sufficient parallelism from a single stream of instructions to utilize the multiple execution units.
- Simultaneous multi-threading processors allow multiple instruction streams to execute concurrently in the different execution resources in an attempt to better utilize those resources. Multi-threading can be particularly advantageous for programs that encounter high latency delays or which often wait for events to occur. When one thread is waiting for a high latency task to complete or for a particular event, a different thread may be processed.
- processors detect particular long latency events such as L2 cache misses and switch threads in response to these detected long latency events. While detection of such long latency events may be effective in some circumstances, such event detection is unlikely to detect all points at which it may be efficient to switch threads. In particular, event based thread switching may fail to detect points in a program where delays are intended by the programmer.
- FIG. 1 illustrates one embodiment of a multi-threaded processor having a monitor to monitor memory accesses.
- FIG. 2 is a flow diagram illustrating operation of the multi-threaded processor of FIG. 1 according to one embodiment.
- FIG. 3 illustrates further details of one embodiment of a multi-threading processor.
- FIG. 4 illustrates resource partitioning, sharing, and duplication according to one embodiment.
- FIG. 5 is a flow diagram illustrating suspending and resuming execution of a thread according to one embodiment.
- FIG. 6 a is a flow diagram illustrating activation and operation of monitoring logic according to one embodiment.
- FIG. 6 b is a flow diagram illustrating enhancement of the observability of writes according to one embodiment.
- FIG. 7 is a flow diagram illustrating monitor operations according to one embodiment.
- FIG. 8 illustrates a system according to one embodiment.
- FIGS. 9 a - 9 c illustrate various embodiments of software sequences utilizing disclosed processor instructions and techniques.
- FIG. 10 illustrates an alternative embodiment which allows a monitored address to remain cached.
- FIG. 11 illustrates various design representations or formats for simulation, emulation, and fabrication of a design using the disclosed techniques.
- the disclosed techniques may allow a programmer to implement a waiting mechanism in one thread while letting other threads harness processing resources.
- a monitor may be set up such that a thread may be suspended until a particular memory access such as a write to a specified memory location occurs.
- a thread may be resumed upon a specified event without executing a processor-resource-wasting routine like a spin-wait loop.
- partitions previously dedicated to the suspended thread may be relinquished while the thread is suspended.
- FIG. 1 illustrates one embodiment of a multi-threaded processor 100 having a memory access monitor 110 to monitor memory accesses.
- a “processor” may be formed as a single integrated circuit in some embodiments. In other embodiments, multiple integrated circuits may together form a processor, and in yet other embodiments, hardware and software routines (e.g., binary translation routines) may together form the processor.
- a bus/memory controller 120 provides instructions for execution to a front end 130 . The front end 130 directs the retrieval of instructions from various threads according to instruction pointers 170 . Instruction pointer logic is replicated to support multiple threads.
- the front end 130 feeds instructions into thread partitionable resources 140 for further processing.
- the thread partitionable resources 140 include logically separated partitions dedicated to particular threads when multiple threads are active within the processor 100 . In one embodiment, each separate partition only contains instructions from the thread to which that portion is dedicated.
- the thread partitionable resources 140 may include, for example, instruction queues. When in a single thread mode, the partitions of the thread partitionable resources 140 may be combined to form a single large partition dedicated to the one thread.
- the processor 100 also includes replicated state 180 .
- the replicated state 180 includes state variables sufficient to maintain context for a logical processor. With replicated state 180 , multiple threads can execute without competition for state variable storage. Additionally, register allocation logic may be replicated for each thread. The replicated state-related logic operates with the appropriate resource partitions to prepare incoming instructions for execution.
- the thread partitionable resources 140 pass instructions along to shared resources 150 .
- the shared resources 150 operate on instructions without regard to their origin.
- scheduler and execution units may be thread-unaware shared resources.
- the partitionable resources 140 may feed instructions from multiple threads to the shared resources 150 by alternating between the threads in a fair manner that provides continued progress on each active thread.
- the shared resources may execute the provided instructions on the appropriate state without concern for the thread mix.
- the shared resources 150 may be followed by another set of thread partitionable resources 160 .
- the thread partitionable resources 160 may include retirement resources such as a re-order buffer and the like. Accordingly, the thread partitionable resources 160 may ensure that execution of instructions from each thread concludes properly and that the appropriate state for that thread is appropriately updated.
- the processor 100 of FIG. 1 includes the memory access monitor 110 .
- the memory access monitor 110 is programmable with information about a memory access cycle for which the monitor 110 can be enabled to watch.
- the monitor 110 includes a monitor cycle information register 112 , which is compared against bus cycle information received from the bus/memory controller 120 by comparison logic 114 . If a match occurs, a resume thread signal is generated to re-start a suspended thread.
- Memory access information may be obtained from internal and/or external buses of the processor.
- the monitor cycle information register 112 may contain details specifying the type of cycle and/or the address which should trigger the resumption of a thread.
- the monitor cycle information register 112 stores a physical address, and the monitor watches for any bus cycle that indicates an actual or potential write to that physical address.
- Such a cycle may be in the form of an explicit write cycle and/or may be a read for ownership or an invalidating cycle by another agent attempting to take exclusive ownership of a cacheable line so that it can write to that line without an external bus transaction.
- the monitor may be programmed to trigger on various transactions in different embodiments.
- the instruction set of the processor 100 includes a MONITOR opcode (instruction) which sets up the monitor transaction information.
- the MONITOR opcode is received as a part of the sequence of instructions of a first thread (T 1 ).
- the processor 100 in response to the MONITOR opcode, the processor 100 enables the monitor 110 to monitor memory accesses for the specified memory access.
- the triggering memory access may be specified by an implicit or explicit operand. Therefore, executing the MONITOR opcode may specify the monitor address as the monitor address can be stored in advance in a register or other location as an implicit operand.
- the monitor tests whether the specified cycle is detected. If not, the monitor continues monitoring memory accesses. If the triggering cycle is detected, then a monitor event pending indicator is set as indicated in block 220 .
- the execution of the MONITOR opcode triggers the activation of the monitor 110 .
- the monitor 110 may begin to operate in parallel with other operations in the processor.
- the MONITOR instruction itself only sets up the monitor 110 with the proper memory cycle information and activates the monitor 110 , without unmasking monitor events.
- monitor events may accrue, but may not be recognized unless they are explicitly unmasked.
- triggering of a memory wait is indicated as a separate event.
- a memory wait (WAIT) opcode may be used to trigger the recognition of monitor events and the suspension of TI. Using two separate instructions to set up and trigger the thread suspension may provide a programmer added flexibility and allow more efficient programming. An alternative embodiment, however, triggers the memory wait from the first opcode which also set up the monitor 110 . In either case, one or more instructions arm the monitor and enable recognition of monitor events.
- T 1 may be suspended as shown in block 235 .
- the processor With T 1 suspended, the processor enters an implementation dependent state which allows other threads to more fully utilize the processor resources.
- the processor may relinquish some or all of the partitions of partitionable resources 140 and 160 that were dedicated to TI.
- different permutations of the MONITOR opcode or settings associated therewith may indicate which resources to relinquish, if any. For example, when a programmer anticipates a shorter wait, the thread may be suspended, but maintain its resource partitions. Throughput is still enhanced because the shared resources may be used exclusively by other threads during the thread suspension period. When a longer wait is anticipated, relinquishing all partitions associated with the suspended thread allows other threads to have additional resources, potentially increasing the throughput of the other threads. The additional throughput, however, comes at the cost of the overhead associated with removing and adding partitions when threads are respectively suspended and resumed.
- T 1 remains in a suspended state until a monitor event is pending.
- the monitor 110 operates independently to detect and signal monitor events (blocks 215 - 220 ). If the processor detects that a monitor event is pending in block 240 , then T 1 is resumed, as indicated in block 250 . No active processing of instructions in T 1 needs to occur for the monitor event to wake up TI. Rather T 1 remains suspended and the enabled monitor 110 signals an event to the processor. The processor handles the event, recognizes that the event indicates T 1 should be resumed, and performs the appropriate actions to resume T 1 .
- FIGS. 1 and 2 provide techniques to allow a thread suspended by a program to be resumed upon the occurrence of a specified memory access.
- other events also cause T 1 to be resumed.
- an interrupt may cause T 1 to resume.
- Such an implementation advantageously allows the monitor to be less than perfect in that it may miss (not detect) certain memory accesses or other conditions that should cause the thread to resume.
- TI may be awakened unnecessarily at times.
- Such an implementation reduces the likelihood that T 1 will become permanently frozen due to a missed event, simplifying hardware design and validation.
- the unnecessary awakenings of T 1 may be only a minor inconvenience as a loop may be constructed to have T 1 double-check whether the condition it was awaiting truly did occur, and if not to suspend itself once again.
- the thread partitionable resources, the replicated resources, and the shared resources may be arranged differently. In some embodiments, there may not be partitionable resources on both ends of the shared resources. In some embodiments, the partitionable resources may not be strictly partitioned, but rather may allow some instructions to cross partitions or may allow partitions to vary in size depending on the thread being executed in that partition or the total number of threads being executed. Additionally, different mixes of resources may be designated as shared, duplicated, and partitioned resources.
- FIG. 3 illustrates further details of one embodiment of a multi-threading processor.
- the embodiment of FIG. 3 includes coherency related logic 350 , one implementation of a monitor 310 , and one specific implementation of thread suspend and resume logic 377 , among other things.
- a bus interface 300 includes a bus controller 340 , event detect logic 345 , a monitor 310 , and the coherency related logic 350 .
- the bus interface 300 provides instructions to a front end 365 , which performs micro-operand (uOP) generation, generating uOPs from macroinstructions.
- Execution resources 370 receive uOPs from the front end 365 , and back end logic 380 retires the various uOPs after they are executed. In one embodiment, out-of-order execution is supported by the front end, back end, and execution resources.
- a MONITOR opcode may enter the processor through the bus interface 300 and be prepared for execution by the front end 365 .
- a special MONITOR uOP is generated for execution by the execution resources 370 .
- the MONITOR uOP may be treated similarly to a store operation by the execution units, with the monitor address being translated by address translation logic 375 into a physical address, which is provided to the monitor 310 .
- the monitor 310 communicates with thread suspend and resume logic 377 to cause resumption of threads.
- the thread suspend and resume logic may perform partition and anneal resources as the number of active threads changes.
- FIG. 4 illustrates the partitioning, duplication, and sharing of resources according to one embodiment.
- Partitioned resources may be partitioned and annealed (fused back together for re-use by other threads) according to the ebb and flow of active threads in the machine.
- duplicated resources include instruction pointer logic in the instruction fetch portion of the pipeline, register renaming logic in the rename portion of the pipeline, state variables (not shown, but referenced in various stages in the pipeline), and an interrupt controller (not shown, generally asynchronous to pipeline).
- Shared resources in the embodiment of FIG. 4 include schedulers in the schedule stage of the pipeline, a pool of registers in the register read and write portions of the pipeline, execution resources in the execute portion of the pipeline.
- a trace cache and an L1 data cache may be shared resources populated according to memory accesses without regard to thread context.
- consideration of thread context may be used in caching decisions.
- Partitioned resources in the embodiment of FIG. 4 include two queues in queuing stages of the pipeline, a re-order buffer in a retirement stage of the pipeline, and a store buffer. Thread selection multiplexing logic alternates between the various duplicated and partitioned resources to provide reasonable access to both threads.
- FIG. 4 For exemplary purposes, it is assumed that the partitioning, sharing, and duplication shown in FIG. 4 is utilized in conjunction with the embodiment of FIG. 3 in further describing operation of an embodiment of the processor of FIG. 3 . In particular, further details of operation of the embodiment of FIG. 3 will now be discussed with respect to the flow diagram of FIG. 5 .
- the processor is assumed to be executing in a multi-threading mode, with at least two threads active.
- the front end 365 receives a MONITOR opcode during execution of a first thread (TI).
- a special monitor uOP is generated by the front end 365 in one embodiment.
- the MONITOR uOP is passed to the execution resources 370 .
- the monitor uOP has an associated address which indicates the address to be monitored (the monitor address).
- the associated address may be in the form of an explicit operand or an implicit operand (i.e., the associated address is to be taken from a predetermined register or other storage location).
- the associated address “indicates” the monitor address in that it conveys enough information to determine the monitor address (possibly in conjunction with other registers or information).
- the associated address may be a linear address which has a corresponding physical address that is the appropriate monitor address.
- the monitor address could be given in virtual address format, or could be indicated as a relative address, or specified in other known or convenient address-specifying manners. If virtual address operands are used, it may be desirable to allow general protection faults to be recognized as break events,
- the monitor address may indicate any convenient unit of memory for monitoring.
- the monitor address may indicate a cache line.
- the monitor address may indicate a portion of a cache line, a specific/selected size portion or unit of memory which may bear different relationships to the cache line sizes of different processors, or a single address.
- the monitor address thus may indicate a unit that includes data specified by the operand (and more data) or may indicate specifically an address for a desired unit of data.
- the monitor address is provided to the address translation logic 375 and passed along to the monitor 310 , where it is stored in a monitor address register 335 .
- the execution resources 370 then enable and activate the monitor 310 as indicated in block 510 and further detailed in FIG. 6 .
- block 510 is shown as occurring in parallel with block 505 because the monitor 310 continues to operate in parallel with other operations until a break event occurs once it is activated by the MONITOR opcode in this embodiment.
- a memory wait (MWAIT) opcode is received in thread 1 , and passed to execution. Execution of the MWAIT opcode unmasks monitor events in the embodiment of FIG. 5 .
- a test is performed, as indicated in block 515 , to determine whether a monitor event is pending. If no monitor event is pending, then a test is performed in block 520 to ensure that the monitor is active. For example, if an MWAIT is executed without previously executing a MONITOR, the monitor 310 would not be active. If either the monitor is inactive or a monitor event is pending, then thread 1 execution is continued in block 580 .
- the thread suspend/resume logic 377 includes pipeline flush logic 382 , which drains the processor pipeline in order to clear all instructions as indicated in block 530 .
- partition/anneal logic 385 causes any partitioned resources associated exclusively with thread 1 to be relinquished for use by other threads as indicated in block 535 . These relinquished resources are annealed to form a set of larger resources for the remaining active threads to utilize. For example, referring to the two thread example of FIG. 4 , all instructions related to thread 1 are drained from both queues. Each pair of queues is then combined to provide a larger queue to the second thread.
- various events are tested to determine whether thread 1 should be resumed. Notably, these tests are not performed by instructions being executed as a part of thread 1 . Rather, these operations are performed by the processor in parallel to its processing of other threads.
- the monitor itself checks whether a monitor write event has occurred and so indicates by setting an event pending indicator.
- the event pending indicator is provided via a WRITE DETECTED signal to the suspend/resume logic 377 (e.g., microcode).
- Microcode may recognize the monitor event at an appropriate instruction boundary in one embodiment (block 540 ) since this event was unmasked by the MWAIT opcode in block 505 .
- Event detect logic 345 may detect other events, such as interrupts, that are designated as break events (block 545 ). Additionally, an optional timer may be used periodically exit the memory wait state to ensure that the processor does not become frozen due to some particular sequence of events (block 550 ). If none of these events signal an exit to the memory wait state, then thread 1 remains suspended.
- the thread/suspend resume logic 377 is again activated upon detection of the appropriate event. Again, the pipeline is flushed, as indicated in block 560 , to drain instructions from the pipeline so that resources can be once again partitioned to accommodate the soon-to-be-awakened thread 1 . In block 570 , the appropriate resources are re-partitioned, and thread 1 is resumed in block 580 .
- FIG. 6 a illustrates further details of the activation and operation of the monitor 310 .
- the front end fetching for thread 1 is stopped to prevent further thread 1 operations from entering the machine.
- the associated address operand is converted from being a linear address to a physical address by the address translation logic 375 .
- the observability of writes to the monitored address are increased. In general, the objective of this operation is to force caching agents to make write operations which would affect the information stored at the monitor address visible to the monitor 310 itself. More details of one specific implementation are discussed with respect to FIG. 6 b .
- the physical address for monitoring is stored, although notably this address may be stored earlier or later in this sequence.
- the monitor is enabled.
- the monitor monitors bus cycles for writes to the physical address which is the monitor address stored in the monitor address register 335 . Further details of the monitoring operation are discussed below with respect to FIG. 7 .
- a store fence operation is executed as indicated in block 625 .
- the store fence helps ensure that all stores in the machine are processed at the time the MONITOR opcode completes execution. With all stores from before the MONITOR being drained from the machine, the likelihood that a memory wait state will be entered erroneously is reduced.
- the store fence operation is a precaution, and can be a time consuming operation.
- This store fence is optional because the MONITOR/MWAIT mechanism of this embodiment has been designed as a multiple exit mechanism. In other words, various events such as certain interrupts, system or on board timers, etc., may also cause exit from the memory wait state. Thus, it is not guaranteed in this embodiment that the only reason the thread will be awakened is because the data value being monitored has changed. Accordingly (see also FIG. 9 a - c below), in this implementation, software—should double-check whether the particular value stored in memory has changed. In one embodiment, some events including assertion of INTR, NMI and SMI interrupts; machine check interrupts; and faults are break events, and others including powerdown events are not. In one embodiment, assertion of the A20M pin is also a break event.
- the monitor continues to test whether bus cycles occurring indicate or appear to indicate a write to the monitor address. If such a bus cycle is detected, the monitor event pending indicator is set, as indicated in block 635 . After execution of the MWAIT opcode (block 505 , FIG. 5 ), this event pending indicator is serviced as an event and causes thread resumption in blocks 560 - 580 of FIG. 5 . Additionally, events that change address translation may cause thread 1 to resume. For example, events that cause a translation look-aside buffer to be flushed may trigger resumption of thread 1 since the translation made to generate the monitor address, from a linear to a physical address may no longer be valid. For example, in an x86 Intel Architecture compatible processor, writes to control registers CR 0 , CR 3 and CR 4 , as well as certain machine specific registers may cause exit of the memory wait state.
- FIG. 6 b illustrates further details of the enhancement of observability of write to the monitor address (block 610 in FIG. 6 a ).
- the processor flushes the cache line associated with the monitor address from all internal caches of the processor as indicated in block 650 . As a result of this flushing, any subsequent write to the monitor address reaches the bus interface 300 , allowing detection by the monitor 310 which is included in the bus interface 300 .
- the MONITOR uOP is modeled after and has the same fault model as a cache line flush CLFLUSH instruction which is an existing instruction in an x86 instruction set. The monitor uOP proceeds through linear to physical translation of the address, and flushing of internal caches much as CLFLUSH does; however. the bus interface recognizes the difference between MONITOR and CLFLUSH and treats the MONITOR uOP appropriately.
- the coherency related logic 350 in the bus interface 300 activates read line generation logic 355 to generate a read line transaction on the processor bus.
- the read line transaction to the monitor address ensures that no other caches in processors on the bus store data at the monitor address in either a shared or exclusive state (according to the well known MESI protocol). In other protocols, other states may be used; however, the transaction is designed to reduce the likelihood that another agent can write to the monitor address without the transaction being observable by the monitor 310 . In other words, writes or write-indicating transactions are subsequently broadcast so they can be detected by the monitor.
- the monitor 310 begins to monitor transactions on the bus.
- the coherency related logic continues to preserve the observability of the monitor address by attempting to prevent bus agents from taking ownership of the cache line associated with the monitored address. According to one bus protocol, this may be accomplished by hit generation logic 360 asserting a HIT# signal during a snoop phase of any read of the monitor address as indicated in block 660 .
- the assertion of HIT# prevents other caches from moving beyond the Shared state in the MESI protocol to the Exclusive and then potentially the Modified state.
- no agents in the chosen coherency domain (the memory portion which is kept coherent) can have data in the modified or exclusive state (or their equivalents).
- the processor effectively appears to have the cache line of the monitor address cached even though it has been flushed from internal caches in this embodiment.
- FIG. 7 illustrates further details of operation of the monitor 310 .
- the monitor 310 receives request and address information from a bus controller 340 for a bus transaction.
- the monitor 310 examines the bus cycle type and the address(es) affected.
- cycle compare logic 320 determines whether the bus cycle is a specified cycle.
- an address comparison circuit 330 compares the bus transaction address to the monitor address stored in the monitor address register 335 , and write detect logic 325 decodes the cycle type information from the bus controller 340 to detect whether a write has occurred.
- a monitor event pending indicator is set as indicated in block 720 .
- a signal (WRITE DETECTED) is provided to the thread suspend/resume logic 377 to signal the event (and will be serviced assuming it has been enabled by executing WAIT).
- the monitor 310 is halted as indicated in block 730 . Halting the monitor saves power, but is not critical as long as false monitor events are masked or otherwise not generated.
- the monitor event indicator may also be reset at this point. Typically, servicing the monitor event also masks the recognition of further monitor events until MWAIT is again executed.
- the coherency related logic 350 is activated. As indicated in block 740 , a signal (such as HIT#) is asserted to prevent another agent from gaining ownership which would allow future writes without coherency broadcasts.
- the monitor 310 remains active and returns to block 700 after and is unaffected by a read of the monitor address. Additionally, if a transaction is neither a read nor a write to the monitor address, the monitor remains active and returns to block 700 .
- the MONITOR instruction is limited such that only certain types of accesses may be monitored. These accesses may be ones chosen as indicative of efficient programming techniques, or may be chosen for other reasons.
- the memory access must be a cacheable store in write-back memory that is naturally aligned.
- a naturally aligned element is an N bit element that starts at an address divisible by N.
- a single cache line needs to be accessed (rather than two cache lines as would be needed in the case where data is split across two cache lines) in order to write to the monitored address.
- using naturally aligned memory addresses may simplify bus watching.
- FIG. 8 illustrates one embodiment of a system that utilizes disclosed multithreaded memory wait techniques.
- a set of N multithreading processors, processors 805 - 1 through 805 -N are coupled to a bus 802 .
- a single processor or a mix of multi-threaded processors and single-threaded processors may be used.
- other known or otherwise available system arrangements may be used.
- the processors may be connected in a point-to-point fashion, and parts such as the memory interface may be integrated into each processor.
- a memory interface 815 coupled to the bus is coupled to a memory 830 and a media interface 820 .
- the memory 830 contains a multiprocessing ready operating system 835 , and instructions for a first thread 840 and instructions for a second thread 845 .
- the instructions 830 include an idle loop according to disclosed techniques, various versions of which are shown in FIGS. 9 a - 9 c.
- the appropriate software to perform these various functions may be provided in any of a variety of machine readable mediums.
- the media interface 820 provides an interface to such software.
- the media interface 820 may be an interface to a storage medium (e.g., a disk drive, an optical drive, a tape drive, a volatile memory, a nonvolatile memory, or the like) or to a transmission medium (e.g., a network interface or other digital or analog communications interface).
- the media interface 820 may read software routines from a medium (e.g., storage medium 792 or transmission medium 795 ).
- Machine readable mediums are any mediums that can store, at least temporarily, information for reading by a machine interface. This may include signal transmissions (via wire, optics, or air as the medium) and/or physical storage media 792 such as various types of disk and memory storage devices.
- FIG. 9 a illustrates an idle loop according to one embodiment.
- the MONITOR command is executed with address 1 as its operand, the monitor address.
- the MWAIT command is executed in block 910 within the same thread.
- the MWAIT instruction causes the thread to be suspended, assuming other conditions are properly met.
- the routine moves on to block 920 to determine if the value stored at the monitor address changed. If the value at the monitor address did change, then execution of the thread continues, as indicated in block 922 . If the value did not change, then a false wake event occurred. The wake event is false in the sense that the MWAIT was exited without a memory write to the monitor address occurring. If the value did not change, then the loop returns to block 905 where the monitor is once again set up.
- This loop software implementation allows the monitor to be designed to allow false wake events.
- FIG. 9 b illustrates an alternative idle loop.
- the embodiment of FIG. 9 b adds one additional check to further reduce the chance that the MWAIT instruction will fail to catch a write to the monitored memory address.
- the flow begins in FIG. 9 b with the MONITOR instruction being executed with address 1 as its operand, as indicated in block 925 .
- the software routine reads the memory value at the monitor address.
- the software double checks to ensure that the memory value has not changed from the value indicating that the thread should be idled. If the value has changed, then thread execution is continued, as indicated in block 952 . If the value has not changed, then the MWAIT instruction is executed, as indicated in block 940 .
- the thread is suspended until a break event occurs in block 945 . Again, however, since false break events are allowed, whether the value has changed is again checked in block 950 . If the value has not changed, then the loop returns to once again enable the monitor to track address 1 , by returning to block 925 . If the value has changed, then execution of the thread continue in block 952 . In some embodiments, the MONITOR instruction may not need to be executed again after a false wake event before the MWAIT instruction is executed to suspend the thread again.
- FIG. 9 c illustrates another example of a software sequence utilizing MONITOR and MWAIT instructions.
- the loop does not idle unless two separate tasks within the thread have no work to do.
- a constant value CVI is stored in work location WL 1 when there is work to be done by a first routine.
- a second constant value CV 2 is stored in WL 2 when there is work to be done by a second routine.
- WL 1 and WL 2 are chosen to be memory locations in the same cache line.
- a single work location may also be used to store status indicators for multiple tasks. For example, one or more bits in a single byte or other unit may each represent a different task.
- the monitor is set up to monitor WL 1 .
- WL 1 stores the constant value indicating that there is work to be done. If so, the work related to WL 1 is performed, as indicated in block 965 . If not, in block 970 , it is tested whether WL 2 stores CV 2 indicated that there is work to be done related to WL 2 . If so, the work related to WL 2 is performed, as indicated in block 975 . If not, the loop may proceed to determine if it is appropriate to call a power management handler in block 980 .
- the logical processor may be placed in a reduced power consumption state (e.g., one of a set of “C” states defined under the Advanced Configuration and Power Interface (ACPI) Specification, Version 1.0b (or later), published Feb. 8, 1999, available at www.acpi.info as of the filing of the present application).
- ACPI Advanced Configuration and Power Interface
- the power management handler is called in block 985 .
- the thread does that work, and then loops back to make the same determinations again after setting the monitor in block 955 .
- the loop back from blocks 965 , 975 , and 985 could be to block 960 as long as the monitor remains active.
- the W A I T instruction is executed as indicated in block 990 .
- the thread suspended state caused by MWAIT is eventually exited when a break event occurs as indicated in block 995 .
- the loop returns to block 955 to set the monitor and thereafter determine whether either WL 1 or WL 2 indicate that there is work to be done. If no work is to be done (e.g., in the case of a false wake up event), the loop will return to MWAIT in block 990 and again suspend the thread until a break event occurs.
- FIG. 10 illustrates one alternative embodiment of a processor that allows the monitor value to remain cached in the L1 cache.
- the processor in FIG. 10 includes execution units 1005 , an L1 cache 1010 , and write combining buffers between the L1 cache and an inclusive L2 cache 1030 .
- the write combining buffers 1020 include a snoop port 1044 which ensures coherency of the internal caches with other memory via operations received by a bus interface 1040 from a bus 1045 . Since coherency-affecting transactions reach the write combining buffers 1020 via the snoop port 1044 , a monitor may be situated at the L1 cache level and still receive sufficient information to determine when a memory write event is occurring on the bus 1045 . Thus, the line of memory corresponding to the monitor address may be kept in the L1 cache. The monitor is able to detect both writes to the L1 cache from the execution units and writes from the bus 1045 via the snoop port 1044 .
- Another alternative embodiment supports a two operand monitor instruction.
- One operand indicates the memory address as previously discussed.
- the second operand is a mask which indicates which of a variety of events that would otherwise not break from the memory wait state should cause a break from this particular memory wait. For example, one mask bit may indicate that masked interrupts should be allowed to break the memory wait despite the fact that the interrupts are masked (e.g., allowing a wake up event even when the EFLAGS bit IF is set to mask interrupts). Presumably, then one of the instructions executed after the memory wait state is broken unmasks that interrupt so it is serviced. Other events that would otherwise not break the memory wait state can be enabled to break the memory wait, or conversely events that normally break the memory wait state can be disabled. As discussed with the first operand, the second operand may be explicit or implicit.
- FIG. 11 illustrates various design representations or formats for simulation, emulation, and fabrication of a design using the disclosed techniques.
- Data representing a design may represent the design in a number of manners.
- the hardware may be represented using a hardware description language or another functional description language which essentially provides a computerized model of how the designed hardware is expected to perform.
- the hardware model 1110 may be stored in a storage medium 1100 such as a computer memory so that the model may be simulated using simulation software 1120 that applies a particular test suite 1130 to the hardware model 1110 to determine if it indeed functions as intended.
- the simulation software is not recorded, captured, or contained in the medium.
- a circuit level model with logic and/or transistor gates may be produced at some stages of the design process.
- This model may be similarly simulated, sometimes by dedicated hardware simulators that form the model using programmable logic. This type of simulation, taken a degree further, may be an emulation technique.
- re-configurable hardware is another embodiment that may involve a machine readable medium storing a model employing the disclosed techniques.
- the data representing the hardware model may be the data specifying the presence or absence of various features on different mask layers for masks used to produce the integrated circuit.
- this data representing the integrated circuit embodies the techniques disclosed in that the circuitry or logic in the data can be simulated or fabricated to perform these techniques.
- the data may be stored in any form of a computer readable medium.
- An optical or electrical wave 1160 modulated or otherwise generated to transmit such information, a memory 1150 , or a magnetic or optical storage 1140 such as a disc may be the medium.
- the set of bits describing the design or the particular part of the design are an article that may be sold in and of itself or used by others for further design or fabrication.
Abstract
Techniques for suspending execution of a thread until a specified memory access occurs. In one embodiment, a processor includes multiple execution units capable of executing multiple threads. A first thread includes an instruction that specifies a monitor address. Suspend logic suspends execution of the first thread, and a monitor causes resumption of the first thread in response to an access to the specified monitor address.
Description
- This Application claims the benefit of priority on U.S. patent application Ser. No. 10/039,579, now U.S. Pat. No. ______.
- 1. Field
- The present disclosure pertains to the field of processors. More particularly, the present disclosure pertains to multi-threaded processors and techniques for temporarily suspending the processing of one thread in a multi-threaded processor.
- 2. Description of Related Art
- A multi-threaded processor is capable of processing multiple different instruction sequences concurrently. A primary motivating factor driving execution of multiple instruction streams within a single processor is the resulting improvement in processor utilization. Highly parallel architectures have developed over the years, but it is often difficult to extract sufficient parallelism from a single stream of instructions to utilize the multiple execution units. Simultaneous multi-threading processors allow multiple instruction streams to execute concurrently in the different execution resources in an attempt to better utilize those resources. Multi-threading can be particularly advantageous for programs that encounter high latency delays or which often wait for events to occur. When one thread is waiting for a high latency task to complete or for a particular event, a different thread may be processed.
- Many different techniques have been proposed to control when a processor switches between threads. For example, some processors detect particular long latency events such as L2 cache misses and switch threads in response to these detected long latency events. While detection of such long latency events may be effective in some circumstances, such event detection is unlikely to detect all points at which it may be efficient to switch threads. In particular, event based thread switching may fail to detect points in a program where delays are intended by the programmer.
- In fact, often, the programmer is in the best position to determine when it would be efficient to switch threads to avoid wasteful spin-wait loops or other resource-consuming delay techniques. Thus, allowing programs to control thread switching may enable programs to operate more efficiently. Explicit program instructions that affect thread selection may be advantageous to this end. For example, a “Pause” instruction is described in U.S. patent application Ser. No. 09/489,130, filed Jan. 21, 2000. The Pause instruction allows a thread of execution to be temporarily suspended either until a count is reached or until an instruction has passed through the processor pipeline. Different techniques may be useful in allowing programmers to more efficiently harness the resources of a multi-threaded processor.
- The present invention is illustrated by way of example and not limitation in the Figures of the accompanying drawings.
-
FIG. 1 illustrates one embodiment of a multi-threaded processor having a monitor to monitor memory accesses. -
FIG. 2 is a flow diagram illustrating operation of the multi-threaded processor ofFIG. 1 according to one embodiment. -
FIG. 3 illustrates further details of one embodiment of a multi-threading processor. -
FIG. 4 illustrates resource partitioning, sharing, and duplication according to one embodiment. -
FIG. 5 is a flow diagram illustrating suspending and resuming execution of a thread according to one embodiment. -
FIG. 6 a is a flow diagram illustrating activation and operation of monitoring logic according to one embodiment. -
FIG. 6 b is a flow diagram illustrating enhancement of the observability of writes according to one embodiment. -
FIG. 7 is a flow diagram illustrating monitor operations according to one embodiment. -
FIG. 8 illustrates a system according to one embodiment. -
FIGS. 9 a-9 c illustrate various embodiments of software sequences utilizing disclosed processor instructions and techniques. -
FIG. 10 illustrates an alternative embodiment which allows a monitored address to remain cached. -
FIG. 11 illustrates various design representations or formats for simulation, emulation, and fabrication of a design using the disclosed techniques. - The following description describes techniques for suspending execution of a thread until a specified memory access occurs. In the following description, numerous specific details such as logic implementations, opcodes, means to specify operands, resource partitioning/sharing/duplication implementations, types and interrelationships of system components, and logic partitioning/integration choices are set forth in order to provide a more thorough understanding of the present invention. It will be appreciated, however, by one skilled in the art that the invention may be practiced without such specific details. In other instances, control structures, gate level circuits and full software instruction sequences have not been shown in detail in order not to obscure the invention. Those of ordinary skill in the art, with the included descriptions, will be able to implement appropriate functionality without undue experimentation.
- The disclosed techniques may allow a programmer to implement a waiting mechanism in one thread while letting other threads harness processing resources. A monitor may be set up such that a thread may be suspended until a particular memory access such as a write to a specified memory location occurs. Thus, a thread may be resumed upon a specified event without executing a processor-resource-wasting routine like a spin-wait loop. In some embodiments, partitions previously dedicated to the suspended thread may be relinquished while the thread is suspended. These and/or other disclosed techniques may advantageously improve overall processor throughput.
-
FIG. 1 illustrates one embodiment of amulti-threaded processor 100 having amemory access monitor 110 to monitor memory accesses. A “processor” may be formed as a single integrated circuit in some embodiments. In other embodiments, multiple integrated circuits may together form a processor, and in yet other embodiments, hardware and software routines (e.g., binary translation routines) may together form the processor. In the embodiment ofFIG. 1 , a bus/memory controller 120 provides instructions for execution to afront end 130. Thefront end 130 directs the retrieval of instructions from various threads according toinstruction pointers 170. Instruction pointer logic is replicated to support multiple threads. - The front end 130 feeds instructions into thread
partitionable resources 140 for further processing. The threadpartitionable resources 140 include logically separated partitions dedicated to particular threads when multiple threads are active within theprocessor 100. In one embodiment, each separate partition only contains instructions from the thread to which that portion is dedicated. The threadpartitionable resources 140 may include, for example, instruction queues. When in a single thread mode, the partitions of the threadpartitionable resources 140 may be combined to form a single large partition dedicated to the one thread. - The
processor 100 also includes replicatedstate 180. The replicatedstate 180 includes state variables sufficient to maintain context for a logical processor. With replicatedstate 180, multiple threads can execute without competition for state variable storage. Additionally, register allocation logic may be replicated for each thread. The replicated state-related logic operates with the appropriate resource partitions to prepare incoming instructions for execution. - The thread
partitionable resources 140 pass instructions along to sharedresources 150. The sharedresources 150 operate on instructions without regard to their origin. For example, scheduler and execution units may be thread-unaware shared resources. Thepartitionable resources 140 may feed instructions from multiple threads to the sharedresources 150 by alternating between the threads in a fair manner that provides continued progress on each active thread. Thus, the shared resources may execute the provided instructions on the appropriate state without concern for the thread mix. - The shared
resources 150 may be followed by another set of threadpartitionable resources 160. The threadpartitionable resources 160 may include retirement resources such as a re-order buffer and the like. Accordingly, the threadpartitionable resources 160 may ensure that execution of instructions from each thread concludes properly and that the appropriate state for that thread is appropriately updated. - As previously mentioned, it may be desirable to provide programmers with a technique to implement the functionality of a spin-wait loop without requiring constant polling of a memory location or even execution of instructions. Thus, the
processor 100 ofFIG. 1 includes thememory access monitor 110. The memory access monitor 110 is programmable with information about a memory access cycle for which themonitor 110 can be enabled to watch. Accordingly, themonitor 110 includes a monitorcycle information register 112, which is compared against bus cycle information received from the bus/memory controller 120 bycomparison logic 114. If a match occurs, a resume thread signal is generated to re-start a suspended thread. Memory access information may be obtained from internal and/or external buses of the processor. - The monitor cycle information register 112 may contain details specifying the type of cycle and/or the address which should trigger the resumption of a thread. In one embodiment, the monitor cycle information register 112 stores a physical address, and the monitor watches for any bus cycle that indicates an actual or potential write to that physical address. Such a cycle may be in the form of an explicit write cycle and/or may be a read for ownership or an invalidating cycle by another agent attempting to take exclusive ownership of a cacheable line so that it can write to that line without an external bus transaction. In any case, the monitor may be programmed to trigger on various transactions in different embodiments.
- The operations of the embodiment of
FIG. 1 may be further explained with reference to the flow diagram ofFIG. 2 . In one embodiment, the instruction set of theprocessor 100 includes a MONITOR opcode (instruction) which sets up the monitor transaction information. Inblock 200, the MONITOR opcode is received as a part of the sequence of instructions of a first thread (T1). As indicated inblock 210, in response to the MONITOR opcode, theprocessor 100 enables themonitor 110 to monitor memory accesses for the specified memory access. The triggering memory access may be specified by an implicit or explicit operand. Therefore, executing the MONITOR opcode may specify the monitor address as the monitor address can be stored in advance in a register or other location as an implicit operand. As indicated inblock 215, the monitor tests whether the specified cycle is detected. If not, the monitor continues monitoring memory accesses. If the triggering cycle is detected, then a monitor event pending indicator is set as indicated inblock 220. - The execution of the MONITOR opcode triggers the activation of the
monitor 110. Themonitor 110 may begin to operate in parallel with other operations in the processor. In one embodiment, the MONITOR instruction itself only sets up themonitor 110 with the proper memory cycle information and activates themonitor 110, without unmasking monitor events. In other words, in this embodiment, after the execution of the MONITOR opcode, monitor events may accrue, but may not be recognized unless they are explicitly unmasked. - Thus, in
block 225, triggering of a memory wait is indicated as a separate event. In some embodiments, a memory wait (WAIT) opcode may be used to trigger the recognition of monitor events and the suspension of TI. Using two separate instructions to set up and trigger the thread suspension may provide a programmer added flexibility and allow more efficient programming. An alternative embodiment, however, triggers the memory wait from the first opcode which also set up themonitor 110. In either case, one or more instructions arm the monitor and enable recognition of monitor events. - In embodiments where separate opcodes are used to arm the
monitor 110 and to trigger the recognition of monitor events, it may be advantageous to perform a test to ensure that the monitor has been activated before suspending the thread as shown inblock 230. Additionally, by testing if a monitor event is already pending (not shown), suspension of T1 may be avoided, and operation may continue inblock 250. Assuming themonitor 110 has been enabled and no monitor events are already pending, T1 may be suspended as shown inblock 235. - With T1 suspended, the processor enters an implementation dependent state which allows other threads to more fully utilize the processor resources. In some embodiments, the processor may relinquish some or all of the partitions of
partitionable resources - T1 remains in a suspended state until a monitor event is pending. As previously discussed, the
monitor 110 operates independently to detect and signal monitor events (blocks 215-220). If the processor detects that a monitor event is pending inblock 240, then T1 is resumed, as indicated inblock 250. No active processing of instructions in T1 needs to occur for the monitor event to wake up TI. Rather T1 remains suspended and theenabled monitor 110 signals an event to the processor. The processor handles the event, recognizes that the event indicates T1 should be resumed, and performs the appropriate actions to resume T1. - Thus, the embodiments of
FIGS. 1 and 2 provide techniques to allow a thread suspended by a program to be resumed upon the occurrence of a specified memory access. In one embodiment, other events also cause T1 to be resumed. For example, an interrupt may cause T1 to resume. Such an implementation advantageously allows the monitor to be less than perfect in that it may miss (not detect) certain memory accesses or other conditions that should cause the thread to resume. As a result, TI may be awakened unnecessarily at times. However, such an implementation reduces the likelihood that T1 will become permanently frozen due to a missed event, simplifying hardware design and validation. The unnecessary awakenings of T1 may be only a minor inconvenience as a loop may be constructed to have T1 double-check whether the condition it was awaiting truly did occur, and if not to suspend itself once again. - In some embodiments, the thread partitionable resources, the replicated resources, and the shared resources may be arranged differently. In some embodiments, there may not be partitionable resources on both ends of the shared resources. In some embodiments, the partitionable resources may not be strictly partitioned, but rather may allow some instructions to cross partitions or may allow partitions to vary in size depending on the thread being executed in that partition or the total number of threads being executed. Additionally, different mixes of resources may be designated as shared, duplicated, and partitioned resources.
-
FIG. 3 illustrates further details of one embodiment of a multi-threading processor. The embodiment ofFIG. 3 includes coherency relatedlogic 350, one implementation of amonitor 310, and one specific implementation of thread suspend and resumelogic 377, among other things. In the embodiment ofFIG. 3 , abus interface 300 includes abus controller 340, event detectlogic 345, amonitor 310, and the coherency relatedlogic 350. - The
bus interface 300 provides instructions to afront end 365, which performs micro-operand (uOP) generation, generating uOPs from macroinstructions.Execution resources 370 receive uOPs from thefront end 365, andback end logic 380 retires the various uOPs after they are executed. In one embodiment, out-of-order execution is supported by the front end, back end, and execution resources. - Various details of operations are further discussed with respect to
FIGS. 5-9 . Briefly, however, a MONITOR opcode may enter the processor through thebus interface 300 and be prepared for execution by thefront end 365. In one embodiment, a special MONITOR uOP is generated for execution by theexecution resources 370. The MONITOR uOP may be treated similarly to a store operation by the execution units, with the monitor address being translated byaddress translation logic 375 into a physical address, which is provided to themonitor 310. Themonitor 310 communicates with thread suspend and resumelogic 377 to cause resumption of threads. The thread suspend and resume logic may perform partition and anneal resources as the number of active threads changes. - For example,
FIG. 4 illustrates the partitioning, duplication, and sharing of resources according to one embodiment. Partitioned resources may be partitioned and annealed (fused back together for re-use by other threads) according to the ebb and flow of active threads in the machine. In the embodiment ofFIG. 4 , duplicated resources include instruction pointer logic in the instruction fetch portion of the pipeline, register renaming logic in the rename portion of the pipeline, state variables (not shown, but referenced in various stages in the pipeline), and an interrupt controller (not shown, generally asynchronous to pipeline). Shared resources in the embodiment ofFIG. 4 include schedulers in the schedule stage of the pipeline, a pool of registers in the register read and write portions of the pipeline, execution resources in the execute portion of the pipeline. Additionally, a trace cache and an L1 data cache may be shared resources populated according to memory accesses without regard to thread context. In other embodiments, consideration of thread context may be used in caching decisions. Partitioned resources in the embodiment ofFIG. 4 include two queues in queuing stages of the pipeline, a re-order buffer in a retirement stage of the pipeline, and a store buffer. Thread selection multiplexing logic alternates between the various duplicated and partitioned resources to provide reasonable access to both threads. - For exemplary purposes, it is assumed that the partitioning, sharing, and duplication shown in
FIG. 4 is utilized in conjunction with the embodiment ofFIG. 3 in further describing operation of an embodiment of the processor ofFIG. 3 . In particular, further details of operation of the embodiment ofFIG. 3 will now be discussed with respect to the flow diagram ofFIG. 5 . The processor is assumed to be executing in a multi-threading mode, with at least two threads active. - In
block 500, thefront end 365 receives a MONITOR opcode during execution of a first thread (TI). A special monitor uOP is generated by thefront end 365 in one embodiment. The MONITOR uOP is passed to theexecution resources 370. The monitor uOP has an associated address which indicates the address to be monitored (the monitor address). The associated address may be in the form of an explicit operand or an implicit operand (i.e., the associated address is to be taken from a predetermined register or other storage location). The associated address “indicates” the monitor address in that it conveys enough information to determine the monitor address (possibly in conjunction with other registers or information). For example, the associated address may be a linear address which has a corresponding physical address that is the appropriate monitor address. Alternatively, the monitor address could be given in virtual address format, or could be indicated as a relative address, or specified in other known or convenient address-specifying manners. If virtual address operands are used, it may be desirable to allow general protection faults to be recognized as break events, - The monitor address may indicate any convenient unit of memory for monitoring. For example, in one embodiment, the monitor address may indicate a cache line. However, in alternative embodiments, the monitor address may indicate a portion of a cache line, a specific/selected size portion or unit of memory which may bear different relationships to the cache line sizes of different processors, or a single address. The monitor address thus may indicate a unit that includes data specified by the operand (and more data) or may indicate specifically an address for a desired unit of data.
- In the embodiment of
FIG. 3 , the monitor address is provided to theaddress translation logic 375 and passed along to themonitor 310, where it is stored in amonitor address register 335. In response to the MONITOR opcode, theexecution resources 370 then enable and activate themonitor 310 as indicated inblock 510 and further detailed inFIG. 6 . As will be further discussed below with respect toFIG. 6 , it may be advantageous to fence any store operations that occur after the MONITOR opcode to ensure that stores are processed and therefore detected before any thread suspension occurs. Thus, some operations may need to occur as a result of activating themonitor 310 before any subsequent instructions can be undertaken in this embodiment. However, block 510 is shown as occurring in parallel withblock 505 because themonitor 310 continues to operate in parallel with other operations until a break event occurs once it is activated by the MONITOR opcode in this embodiment. - In
block 505, a memory wait (MWAIT) opcode is received inthread 1, and passed to execution. Execution of the MWAIT opcode unmasks monitor events in the embodiment ofFIG. 5 . In response to the MWAIT opcode, a test is performed, as indicated inblock 515, to determine whether a monitor event is pending. If no monitor event is pending, then a test is performed inblock 520 to ensure that the monitor is active. For example, if an MWAIT is executed without previously executing a MONITOR, themonitor 310 would not be active. If either the monitor is inactive or a monitor event is pending, thenthread 1 execution is continued inblock 580. - If the
monitor 310 is active and no monitor event is pending, thenthread 1 execution is suspended as indicated inblock 525. The thread suspend/resumelogic 377 includes pipelineflush logic 382, which drains the processor pipeline in order to clear all instructions as indicated inblock 530. Once the pipeline has been drained, partition/anneal logic 385 causes any partitioned resources associated exclusively withthread 1 to be relinquished for use by other threads as indicated inblock 535. These relinquished resources are annealed to form a set of larger resources for the remaining active threads to utilize. For example, referring to the two thread example ofFIG. 4 , all instructions related tothread 1 are drained from both queues. Each pair of queues is then combined to provide a larger queue to the second thread. Similarly, more registers from the register pool are made available to the second thread, more entries from the store buffer are freed for the second thread, and more entries in the re-order buffer are made available to the second thread. In essence, these structures are returned to single dedicated structures of twice the size. Of course, different proportions may result from implementations using different numbers of threads. - In blocks 540, 545, and 550, various events are tested to determine whether
thread 1 should be resumed. Notably, these tests are not performed by instructions being executed as a part ofthread 1. Rather, these operations are performed by the processor in parallel to its processing of other threads. As will be discussed in further detail with respect toFIG. 6 , the monitor itself checks whether a monitor write event has occurred and so indicates by setting an event pending indicator. The event pending indicator is provided via a WRITE DETECTED signal to the suspend/resume logic 377 (e.g., microcode). Microcode may recognize the monitor event at an appropriate instruction boundary in one embodiment (block 540) since this event was unmasked by the MWAIT opcode inblock 505. Event detectlogic 345 may detect other events, such as interrupts, that are designated as break events (block 545). Additionally, an optional timer may be used periodically exit the memory wait state to ensure that the processor does not become frozen due to some particular sequence of events (block 550). If none of these events signal an exit to the memory wait state, thenthread 1 remains suspended. - If
thread 1 is resumed, the thread/suspendresume logic 377 is again activated upon detection of the appropriate event. Again, the pipeline is flushed, as indicated inblock 560, to drain instructions from the pipeline so that resources can be once again partitioned to accommodate the soon-to-be-awakened thread 1. Inblock 570, the appropriate resources are re-partitioned, andthread 1 is resumed inblock 580. -
FIG. 6 a illustrates further details of the activation and operation of themonitor 310. Inblock 600, the front end fetching forthread 1 is stopped to preventfurther thread 1 operations from entering the machine. Inblock 605, the associated address operand is converted from being a linear address to a physical address by theaddress translation logic 375. Inblock 610, the observability of writes to the monitored address are increased. In general, the objective of this operation is to force caching agents to make write operations which would affect the information stored at the monitor address visible to themonitor 310 itself. More details of one specific implementation are discussed with respect toFIG. 6 b. Inblock 615, the physical address for monitoring is stored, although notably this address may be stored earlier or later in this sequence. - Next, as indicated in
block 620, the monitor is enabled. The monitor monitors bus cycles for writes to the physical address which is the monitor address stored in themonitor address register 335. Further details of the monitoring operation are discussed below with respect toFIG. 7 . After the monitor is enabled, a store fence operation is executed as indicated inblock 625. The store fence helps ensure that all stores in the machine are processed at the time the MONITOR opcode completes execution. With all stores from before the MONITOR being drained from the machine, the likelihood that a memory wait state will be entered erroneously is reduced. The store fence operation, however, is a precaution, and can be a time consuming operation. - This store fence is optional because the MONITOR/MWAIT mechanism of this embodiment has been designed as a multiple exit mechanism. In other words, various events such as certain interrupts, system or on board timers, etc., may also cause exit from the memory wait state. Thus, it is not guaranteed in this embodiment that the only reason the thread will be awakened is because the data value being monitored has changed. Accordingly (see also
FIG. 9 a-c below), in this implementation, software—should double-check whether the particular value stored in memory has changed. In one embodiment, some events including assertion of INTR, NMI and SMI interrupts; machine check interrupts; and faults are break events, and others including powerdown events are not. In one embodiment, assertion of the A20M pin is also a break event. - As indicated in
block 630, the monitor continues to test whether bus cycles occurring indicate or appear to indicate a write to the monitor address. If such a bus cycle is detected, the monitor event pending indicator is set, as indicated inblock 635. After execution of the MWAIT opcode (block 505,FIG. 5 ), this event pending indicator is serviced as an event and causes thread resumption in blocks 560-580 ofFIG. 5 . Additionally, events that change address translation may causethread 1 to resume. For example, events that cause a translation look-aside buffer to be flushed may trigger resumption ofthread 1 since the translation made to generate the monitor address, from a linear to a physical address may no longer be valid. For example, in an x86 Intel Architecture compatible processor, writes to control registers CR0, CR3 and CR4, as well as certain machine specific registers may cause exit of the memory wait state. - As noted above,
FIG. 6 b illustrates further details of the enhancement of observability of write to the monitor address (block 610 inFIG. 6 a). In one embodiment, the processor flushes the cache line associated with the monitor address from all internal caches of the processor as indicated inblock 650. As a result of this flushing, any subsequent write to the monitor address reaches thebus interface 300, allowing detection by themonitor 310 which is included in thebus interface 300. In one embodiment, the MONITOR uOP is modeled after and has the same fault model as a cache line flush CLFLUSH instruction which is an existing instruction in an x86 instruction set. The monitor uOP proceeds through linear to physical translation of the address, and flushing of internal caches much as CLFLUSH does; however. the bus interface recognizes the difference between MONITOR and CLFLUSH and treats the MONITOR uOP appropriately. - Next, as indicated in
block 655, the coherency relatedlogic 350 in thebus interface 300 activates readline generation logic 355 to generate a read line transaction on the processor bus. The read line transaction to the monitor address ensures that no other caches in processors on the bus store data at the monitor address in either a shared or exclusive state (according to the well known MESI protocol). In other protocols, other states may be used; however, the transaction is designed to reduce the likelihood that another agent can write to the monitor address without the transaction being observable by themonitor 310. In other words, writes or write-indicating transactions are subsequently broadcast so they can be detected by the monitor. Once the read line operation is done, themonitor 310 begins to monitor transactions on the bus. - As additional transactions occur on the bus, the coherency related logic continues to preserve the observability of the monitor address by attempting to prevent bus agents from taking ownership of the cache line associated with the monitored address. According to one bus protocol, this may be accomplished by
hit generation logic 360 asserting a HIT# signal during a snoop phase of any read of the monitor address as indicated inblock 660. The assertion of HIT# prevents other caches from moving beyond the Shared state in the MESI protocol to the Exclusive and then potentially the Modified state. As a result, as indicated inblock 665, no agents in the chosen coherency domain (the memory portion which is kept coherent) can have data in the modified or exclusive state (or their equivalents). The processor effectively appears to have the cache line of the monitor address cached even though it has been flushed from internal caches in this embodiment. - Referring now to
FIG. 7 , further details of the operations associated withblock 620 inFIG. 6 a are detailed. In particular,FIG. 7 illustrates further details of operation of themonitor 310. Inblock 700, themonitor 310 receives request and address information from abus controller 340 for a bus transaction. As indicated inblock 710, themonitor 310 examines the bus cycle type and the address(es) affected. In particular, cycle comparelogic 320 determines whether the bus cycle is a specified cycle. In one embodiment, anaddress comparison circuit 330 compares the bus transaction address to the monitor address stored in themonitor address register 335, and write detectlogic 325 decodes the cycle type information from thebus controller 340 to detect whether a write has occurred. If a write to the monitor address occurs, a monitor event pending indicator is set as indicated inblock 720. A signal (WRITE DETECTED) is provided to the thread suspend/resumelogic 377 to signal the event (and will be serviced assuming it has been enabled by executing WAIT). Finally, themonitor 310 is halted as indicated inblock 730. Halting the monitor saves power, but is not critical as long as false monitor events are masked or otherwise not generated. The monitor event indicator may also be reset at this point. Typically, servicing the monitor event also masks the recognition of further monitor events until MWAIT is again executed. - In the case of a read to the monitor address, the coherency related
logic 350 is activated. As indicated inblock 740, a signal (such as HIT#) is asserted to prevent another agent from gaining ownership which would allow future writes without coherency broadcasts. Themonitor 310 remains active and returns to block 700 after and is unaffected by a read of the monitor address. Additionally, if a transaction is neither a read nor a write to the monitor address, the monitor remains active and returns to block 700. - In some embodiments, the MONITOR instruction is limited such that only certain types of accesses may be monitored. These accesses may be ones chosen as indicative of efficient programming techniques, or may be chosen for other reasons. For example, in one embodiment, the memory access must be a cacheable store in write-back memory that is naturally aligned. A naturally aligned element is an N bit element that starts at an address divisible by N. As a result of using naturally aligned elements, a single cache line needs to be accessed (rather than two cache lines as would be needed in the case where data is split across two cache lines) in order to write to the monitored address. As a result, using naturally aligned memory addresses may simplify bus watching.
-
FIG. 8 illustrates one embodiment of a system that utilizes disclosed multithreaded memory wait techniques. In the embodiment ofFIG. 8 , a set of N multithreading processors, processors 805-1 through 805-N are coupled to abus 802. In other embodiments, a single processor or a mix of multi-threaded processors and single-threaded processors may be used. In addition, other known or otherwise available system arrangements may be used. For example, the processors may be connected in a point-to-point fashion, and parts such as the memory interface may be integrated into each processor. - In the embodiment of
FIG. 8 , amemory interface 815 coupled to the bus is coupled to amemory 830 and amedia interface 820. Thememory 830 contains a multiprocessingready operating system 835, and instructions for afirst thread 840 and instructions for asecond thread 845. Theinstructions 830 include an idle loop according to disclosed techniques, various versions of which are shown inFIGS. 9 a-9 c. - The appropriate software to perform these various functions may be provided in any of a variety of machine readable mediums. The
media interface 820 provides an interface to such software. Themedia interface 820 may be an interface to a storage medium (e.g., a disk drive, an optical drive, a tape drive, a volatile memory, a nonvolatile memory, or the like) or to a transmission medium (e.g., a network interface or other digital or analog communications interface). Themedia interface 820 may read software routines from a medium (e.g.,storage medium 792 or transmission medium 795). Machine readable mediums are any mediums that can store, at least temporarily, information for reading by a machine interface. This may include signal transmissions (via wire, optics, or air as the medium) and/orphysical storage media 792 such as various types of disk and memory storage devices. -
FIG. 9 a illustrates an idle loop according to one embodiment. Inblock 905, the MONITOR command is executed withaddress 1 as its operand, the monitor address. The MWAIT command is executed inblock 910 within the same thread. As previously discussed, the MWAIT instruction causes the thread to be suspended, assuming other conditions are properly met. When a break event occurs inblock 915, the routine moves on to block 920 to determine if the value stored at the monitor address changed. If the value at the monitor address did change, then execution of the thread continues, as indicated inblock 922. If the value did not change, then a false wake event occurred. The wake event is false in the sense that the MWAIT was exited without a memory write to the monitor address occurring. If the value did not change, then the loop returns to block 905 where the monitor is once again set up. This loop software implementation allows the monitor to be designed to allow false wake events. -
FIG. 9 b illustrates an alternative idle loop. The embodiment ofFIG. 9 b adds one additional check to further reduce the chance that the MWAIT instruction will fail to catch a write to the monitored memory address. Again, the flow begins inFIG. 9 b with the MONITOR instruction being executed withaddress 1 as its operand, as indicated inblock 925. Additionally, inblock 930, the software routine reads the memory value at the monitor address. Inblock 935, the software double checks to ensure that the memory value has not changed from the value indicating that the thread should be idled. If the value has changed, then thread execution is continued, as indicated inblock 952. If the value has not changed, then the MWAIT instruction is executed, as indicated inblock 940. As previously discussed, the thread is suspended until a break event occurs inblock 945. Again, however, since false break events are allowed, whether the value has changed is again checked inblock 950. If the value has not changed, then the loop returns to once again enable the monitor to trackaddress 1, by returning to block 925. If the value has changed, then execution of the thread continue inblock 952. In some embodiments, the MONITOR instruction may not need to be executed again after a false wake event before the MWAIT instruction is executed to suspend the thread again. -
FIG. 9 c illustrates another example of a software sequence utilizing MONITOR and MWAIT instructions. In the example ofFIG. 9 c, the loop does not idle unless two separate tasks within the thread have no work to do. A constant value CVI is stored in work location WL1 when there is work to be done by a first routine. Similarly, a second constant value CV2 is stored in WL2 when there is work to be done by a second routine. In order to use a single monitor address, WL1 and WL2 are chosen to be memory locations in the same cache line. Alternatively, a single work location may also be used to store status indicators for multiple tasks. For example, one or more bits in a single byte or other unit may each represent a different task. - As indicated in
block 955, the monitor is set up to monitor WL1. Inblock 960, it is tested whether WL1 stores the constant value indicating that there is work to be done. If so, the work related to WL1 is performed, as indicated inblock 965. If not, inblock 970, it is tested whether WL2 stores CV2 indicated that there is work to be done related to WL2. If so, the work related to WL2 is performed, as indicated inblock 975. If not, the loop may proceed to determine if it is appropriate to call a power management handler inblock 980. For example, if a selected amount of time has elapsed, then the logical processor may be placed in a reduced power consumption state (e.g., one of a set of “C” states defined under the Advanced Configuration and Power Interface (ACPI) Specification, Version 1.0b (or later), published Feb. 8, 1999, available at www.acpi.info as of the filing of the present application). If so, then the power management handler is called inblock 985. In any of thecases block 955. In an alternative embodiment, the loop back fromblocks - If no work to be done is encountered through
blocks block 990. The thread suspended state caused by MWAIT is eventually exited when a break event occurs as indicated inblock 995. At this point, the loop returns to block 955 to set the monitor and thereafter determine whether either WL1 or WL2 indicate that there is work to be done. If no work is to be done (e.g., in the case of a false wake up event), the loop will return to MWAIT inblock 990 and again suspend the thread until a break event occurs. -
FIG. 10 illustrates one alternative embodiment of a processor that allows the monitor value to remain cached in the L1 cache. The processor inFIG. 10 includesexecution units 1005, anL1 cache 1010, and write combining buffers between the L1 cache and aninclusive L2 cache 1030. Thewrite combining buffers 1020 include a snoopport 1044 which ensures coherency of the internal caches with other memory via operations received by a bus interface 1040 from abus 1045. Since coherency-affecting transactions reach thewrite combining buffers 1020 via the snoopport 1044, a monitor may be situated at the L1 cache level and still receive sufficient information to determine when a memory write event is occurring on thebus 1045. Thus, the line of memory corresponding to the monitor address may be kept in the L1 cache. The monitor is able to detect both writes to the L1 cache from the execution units and writes from thebus 1045 via the snoopport 1044. - Another alternative embodiment supports a two operand monitor instruction. One operand indicates the memory address as previously discussed. The second operand is a mask which indicates which of a variety of events that would otherwise not break from the memory wait state should cause a break from this particular memory wait. For example, one mask bit may indicate that masked interrupts should be allowed to break the memory wait despite the fact that the interrupts are masked (e.g., allowing a wake up event even when the EFLAGS bit IF is set to mask interrupts). Presumably, then one of the instructions executed after the memory wait state is broken unmasks that interrupt so it is serviced. Other events that would otherwise not break the memory wait state can be enabled to break the memory wait, or conversely events that normally break the memory wait state can be disabled. As discussed with the first operand, the second operand may be explicit or implicit.
-
FIG. 11 illustrates various design representations or formats for simulation, emulation, and fabrication of a design using the disclosed techniques. Data representing a design may represent the design in a number of manners. First, as is useful in simulations, the hardware may be represented using a hardware description language or another functional description language which essentially provides a computerized model of how the designed hardware is expected to perform. The hardware model 1110 may be stored in astorage medium 1100 such as a computer memory so that the model may be simulated usingsimulation software 1120 that applies a particular test suite 1130 to the hardware model 1110 to determine if it indeed functions as intended. In some embodiments, the simulation software is not recorded, captured, or contained in the medium. - Additionally, a circuit level model with logic and/or transistor gates may be produced at some stages of the design process. This model may be similarly simulated, sometimes by dedicated hardware simulators that form the model using programmable logic. This type of simulation, taken a degree further, may be an emulation technique. In any case, re-configurable hardware is another embodiment that may involve a machine readable medium storing a model employing the disclosed techniques.
- Furthermore, most designs, at some stage, reach a level of data representing the physical placement of various devices in the hardware model. In the case where conventional semiconductor fabrication techniques are used, the data representing the hardware model may be the data specifying the presence or absence of various features on different mask layers for masks used to produce the integrated circuit. Again, this data representing the integrated circuit embodies the techniques disclosed in that the circuitry or logic in the data can be simulated or fabricated to perform these techniques.
- In any representation of the design, the data may be stored in any form of a computer readable medium. An optical or
electrical wave 1160 modulated or otherwise generated to transmit such information, amemory 1150, or a magnetic oroptical storage 1140 such as a disc may be the medium. The set of bits describing the design or the particular part of the design are an article that may be sold in and of itself or used by others for further design or fabrication. - Thus, techniques for suspending execution of a thread until a specified memory access occurs are disclosed. While certain exemplary embodiments have been described and shown in the accompanying drawings, it is to be understood that such embodiments are merely illustrative of and not restrictive on the broad invention, and that this invention not be limited to the specific constructions and arrangements shown and described, since various other modifications may occur to those ordinarily skilled in the art upon studying this disclosure.
Claims (23)
1-69. (canceled)
70. A processor comprising:
a plurality of execution units to execute a plurality of threads;
suspend logic to suspend execution of a first thread of the plurality of threads upon execution of a first instruction within the first thread while at least one of the plurality of threads remains active; and
a memory access monitor to signal resumption of the first thread in response to a memory access to a specified memory location.
71. The processor of claim 70 wherein the memory access monitor is to cause resumption of the first thread in response to events that cause a translation look-aside buffer to be flushed.
72. The processor of claim 70 wherein the memory access monitor is to cause resumption of the first thread in response to a write to a predetermined control register.
73. The processor of claim 72 further comprising resume logic receive signals to cause resumption in the execution of the first thread.
74. The processor of claim 70 wherein the suspend logic further sets a monitor address as the specified memory location in response to execution of a second instruction in the first thread so that resumption of the first thread occurs only in response to the memory access to the monitor address.
75. The processor of claim 70 wherein the suspend logic is only to suspend the first thread in response to a write access to the specified memory location.
76. The processor of claim 70 wherein the suspend logic is only to suspend the first thread if the memory access is a write access to a selected memory.
77. The processor of claim 76 wherein the selected memory type is a write-back type of memory.
78. The processor of claim 70 further comprising an instruction buffer adapted as either a single partition dedicated to one thread or multiple partitions to be used by the plurality of threads.
79. The processor of claim 74 wherein the processor is capable of out-of-order execution and wherein the second instruction is followed by a store fence that ensures all store operations are processed at the time the first instruction completes execution.
80. The processor of claim 74 wherein the second instruction is an instruction having only implicit operands.
81. The processor of claim 70 further comprising:
coherency logic to perform a read line transaction in conjunction with suspending the first thread.
82. The processor of claim 81 wherein the coherency logic is to perform a cache line flush to flush internal caches in conjunction with suspending the first thread.
83. A processor comprising:
a front end to receive a first instruction and a second instruction, the first instruction having an implicit operand from a predetermined register indicating a monitor address;
execution resources to execute the first instruction and the second instruction and to enter a first implementation dependent state in response to the second instruction if the first instruction has been executed and no break events have occurred after execution of the first instruction; and
a monitor to cause an exit from the first implementation dependent state in response to a memory access to the monitor address.
84. The processor of claim 83 wherein the implicit operand is to indicate a linear address, and wherein the processor further comprises address translation logic to translate the linear address to obtain the monitor address which is a physical address.
85. The processor of claim 83 further comprising:
coherency logic to ensure that no cache in another processor coupled to the processor stores information at the monitor address in a modified or exclusive state.
86. The processor of claim 83 wherein the coherency logic is to assert a hit signal in response to another processor snooping the monitor address.
87. The processor of claim 85 wherein the coherency logic is to assert a hit signal in response to another processor snooping the monitor address.
88. A method comprising:
executing a plurality of threads;
suspending execution of a first thread of the plurality of threads upon execution of a first instruction within the first thread while at least one of the plurality of threads remains active;
passively monitoring for an event; and
resuming execution of the first thread in response to detecting the event.
89. The method of claim 88 , wherein the passively monitoring of the event includes monitoring for a memory access to a specified memory location.
90. The method of claim 88 wherein the passively monitoring of the event includes monitoring for an event without undergoing a spin-wait loop.
91. The method of claim 89 wherein prior to suspending the first thread, the method further comprises setting a monitor address as the specified memory location in response to execution of a second instruction in the first thread so that resumption of the first thread occurs in response to the memory access to the monitor address.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
US11/891,076 US20080034190A1 (en) | 2001-12-31 | 2007-08-08 | Method and apparatus for suspending execution of a thread until a specified memory access occurs |
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
US10/039,579 US7363474B2 (en) | 2001-12-31 | 2001-12-31 | Method and apparatus for suspending execution of a thread until a specified memory access occurs |
US11/891,076 US20080034190A1 (en) | 2001-12-31 | 2007-08-08 | Method and apparatus for suspending execution of a thread until a specified memory access occurs |
Related Parent Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
US10/039,579 Continuation US7363474B2 (en) | 2001-12-31 | 2001-12-31 | Method and apparatus for suspending execution of a thread until a specified memory access occurs |
Publications (1)
Publication Number | Publication Date |
---|---|
US20080034190A1 true US20080034190A1 (en) | 2008-02-07 |
Family
ID=21906217
Family Applications (2)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
US10/039,579 Active 2025-01-08 US7363474B2 (en) | 2001-12-31 | 2001-12-31 | Method and apparatus for suspending execution of a thread until a specified memory access occurs |
US11/891,076 Abandoned US20080034190A1 (en) | 2001-12-31 | 2007-08-08 | Method and apparatus for suspending execution of a thread until a specified memory access occurs |
Family Applications Before (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
US10/039,579 Active 2025-01-08 US7363474B2 (en) | 2001-12-31 | 2001-12-31 | Method and apparatus for suspending execution of a thread until a specified memory access occurs |
Country Status (11)
Country | Link |
---|---|
US (2) | US7363474B2 (en) |
JP (2) | JP4601958B2 (en) |
KR (1) | KR100814993B1 (en) |
CN (1) | CN100383740C (en) |
AU (1) | AU2002364557A1 (en) |
DE (2) | DE10297596B4 (en) |
GB (1) | GB2400947B (en) |
HK (1) | HK1066883A1 (en) |
RU (1) | RU2308754C2 (en) |
TW (2) | TWI322959B (en) |
WO (1) | WO2003058447A2 (en) |
Cited By (37)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20090199197A1 (en) * | 2008-02-01 | 2009-08-06 | International Business Machines Corporation | Wake-and-Go Mechanism with Dynamic Allocation in Hardware Private Array |
US20090199184A1 (en) * | 2008-02-01 | 2009-08-06 | Arimilli Ravi K | Wake-and-Go Mechanism With Software Save of Thread State |
US20090199028A1 (en) * | 2008-02-01 | 2009-08-06 | Arimilli Ravi K | Wake-and-Go Mechanism with Data Exclusivity |
US20090199189A1 (en) * | 2008-02-01 | 2009-08-06 | Arimilli Ravi K | Parallel Lock Spinning Using Wake-and-Go Mechanism |
US20090199029A1 (en) * | 2008-02-01 | 2009-08-06 | Arimilli Ravi K | Wake-and-Go Mechanism with Data Monitoring |
US20100268915A1 (en) * | 2009-04-16 | 2010-10-21 | International Business Machines Corporation | Remote Update Programming Idiom Accelerator with Allocated Processor Resources |
US20100269115A1 (en) * | 2009-04-16 | 2010-10-21 | International Business Machines Corporation | Managing Threads in a Wake-and-Go Engine |
US20100268790A1 (en) * | 2009-04-16 | 2010-10-21 | International Business Machines Corporation | Complex Remote Update Programming Idiom Accelerator |
US20100287341A1 (en) * | 2008-02-01 | 2010-11-11 | Arimilli Ravi K | Wake-and-Go Mechanism with System Address Bus Transaction Master |
US20100293340A1 (en) * | 2008-02-01 | 2010-11-18 | Arimilli Ravi K | Wake-and-Go Mechanism with System Bus Response |
US20110173630A1 (en) * | 2008-02-01 | 2011-07-14 | Arimilli Ravi K | Central Repository for Wake-and-Go Mechanism |
US20110173419A1 (en) * | 2008-02-01 | 2011-07-14 | Arimilli Ravi K | Look-Ahead Wake-and-Go Engine With Speculative Execution |
US20110173420A1 (en) * | 2010-01-08 | 2011-07-14 | International Business Machines Corporation | Processor resume unit |
US20110173625A1 (en) * | 2008-02-01 | 2011-07-14 | Arimilli Ravi K | Wake-and-Go Mechanism with Prioritization of Threads |
US20110173422A1 (en) * | 2010-01-08 | 2011-07-14 | International Business Machines Corporation | Pause processor hardware thread until pin |
US20110173631A1 (en) * | 2008-02-01 | 2011-07-14 | Arimilli Ravi K | Wake-and-Go Mechanism for a Data Processing System |
US20110173417A1 (en) * | 2008-02-01 | 2011-07-14 | Arimilli Ravi K | Programming Idiom Accelerators |
US8082315B2 (en) | 2009-04-16 | 2011-12-20 | International Business Machines Corporation | Programming idiom accelerator for remote update |
US20120144232A1 (en) * | 2010-12-03 | 2012-06-07 | International Business Machines Corporation | Generation of Standby Images of Applications |
US8250396B2 (en) | 2008-02-01 | 2012-08-21 | International Business Machines Corporation | Hardware wake-and-go mechanism for a data processing system |
US8341635B2 (en) | 2008-02-01 | 2012-12-25 | International Business Machines Corporation | Hardware wake-and-go mechanism with look-ahead polling |
US8452947B2 (en) | 2008-02-01 | 2013-05-28 | International Business Machines Corporation | Hardware wake-and-go mechanism and content addressable memory with instruction pre-fetch look-ahead to detect programming idioms |
US8713362B2 (en) | 2010-12-01 | 2014-04-29 | International Business Machines Corporation | Obviation of recovery of data store consistency for application I/O errors |
US8725992B2 (en) | 2008-02-01 | 2014-05-13 | International Business Machines Corporation | Programming language exposing idiom calls to a programming idiom accelerator |
US8732683B2 (en) | 2008-02-01 | 2014-05-20 | International Business Machines Corporation | Compiler providing idiom to idiom accelerator |
WO2014084919A1 (en) * | 2012-11-28 | 2014-06-05 | Intel Corporation | Instruction and logic to provide pushing buffer copy and store functionality |
US8898671B2 (en) | 2010-07-07 | 2014-11-25 | Panasonic Corporation | Processor that executes a plurality of threads by promoting efficiency of transfer of data that is shared with the plurality of threads |
US8990597B2 (en) | 2009-12-18 | 2015-03-24 | Intel Corporation | Instruction for enabling a processor wait state |
US9256477B2 (en) * | 2014-05-29 | 2016-02-09 | Netapp, Inc. | Lockless waterfall thread communication |
US9304702B2 (en) | 2014-05-29 | 2016-04-05 | Netapp, Inc. | System and method for parallelized performance data collection in a computing system |
US9477521B2 (en) | 2014-05-29 | 2016-10-25 | Netapp, Inc. | Method and system for scheduling repetitive tasks in O(1) |
WO2016200567A1 (en) * | 2015-06-09 | 2016-12-15 | Optimum Semiconductor Technologies, Inc. | Processor with advanced operating system support |
US9575802B2 (en) | 2014-10-28 | 2017-02-21 | International Business Machines Corporation | Controlling execution of threads in a multi-threaded processor |
EP3074860A4 (en) * | 2013-11-25 | 2017-10-11 | Marvell World Trade Ltd. | Systems and methods for loop suspension in java programming |
US10275254B2 (en) | 2017-03-08 | 2019-04-30 | International Business Machines Corporation | Spin loop delay instruction |
US11061680B2 (en) | 2014-10-28 | 2021-07-13 | International Business Machines Corporation | Instructions controlling access to shared registers of a multi-threaded processor |
US11231881B2 (en) | 2020-04-02 | 2022-01-25 | Dell Products L.P. | Raid data storage device multi-step command coordination system |
Families Citing this family (100)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US7127561B2 (en) * | 2001-12-31 | 2006-10-24 | Intel Corporation | Coherency techniques for suspending execution of a thread until a specified memory access occurs |
US7653912B2 (en) * | 2003-05-30 | 2010-01-26 | Steven Frank | Virtual processor methods and apparatus with unified event notification and consumer-producer memory operations |
US7213093B2 (en) * | 2003-06-27 | 2007-05-01 | Intel Corporation | Queued locks using monitor-memory wait |
US8984199B2 (en) * | 2003-07-31 | 2015-03-17 | Intel Corporation | Inter-processor interrupts |
US20050108711A1 (en) * | 2003-11-13 | 2005-05-19 | Infineon Technologies North America Corporation | Machine instruction for enhanced control of multiple virtual processor systems |
US7310722B2 (en) * | 2003-12-18 | 2007-12-18 | Nvidia Corporation | Across-thread out of order instruction dispatch in a multithreaded graphics processor |
US7490218B2 (en) * | 2004-01-22 | 2009-02-10 | University Of Washington | Building a wavecache |
US7657882B2 (en) * | 2004-01-22 | 2010-02-02 | University Of Washington | Wavescalar architecture having a wave order memory |
US7823162B1 (en) | 2004-01-30 | 2010-10-26 | Xilinx, Inc. | Thread circuits and a broadcast channel in programmable logic |
US7770179B1 (en) * | 2004-01-30 | 2010-08-03 | Xilinx, Inc. | Method and apparatus for multithreading on a programmable logic device |
US7552042B1 (en) | 2004-01-30 | 2009-06-23 | Xilinx, Inc. | Method for message processing on a programmable logic device |
US8984517B2 (en) * | 2004-02-04 | 2015-03-17 | Intel Corporation | Sharing idled processor execution resources |
US7581214B2 (en) * | 2004-04-15 | 2009-08-25 | Intel Corporation | Live set transmission in pipelining applications |
JP4376692B2 (en) * | 2004-04-30 | 2009-12-02 | 富士通株式会社 | Information processing device, processor, processor control method, information processing device control method, cache memory |
US8607241B2 (en) * | 2004-06-30 | 2013-12-10 | Intel Corporation | Compare and exchange operation using sleep-wakeup mechanism |
US8074030B1 (en) | 2004-07-20 | 2011-12-06 | Oracle America, Inc. | Using transactional memory with early release to implement non-blocking dynamic-sized data structure |
US7703098B1 (en) | 2004-07-20 | 2010-04-20 | Sun Microsystems, Inc. | Technique to allow a first transaction to wait on condition that affects its working set |
US7206903B1 (en) * | 2004-07-20 | 2007-04-17 | Sun Microsystems, Inc. | Method and apparatus for releasing memory locations during transactional execution |
KR101051703B1 (en) | 2004-08-09 | 2011-07-25 | 삼성전자주식회사 | Integrated circuit card and integrated circuit card system with suspend / reset function |
US7434009B2 (en) * | 2004-09-30 | 2008-10-07 | Freescale Semiconductor, Inc. | Apparatus and method for providing information to a cache module using fetch bursts |
US7257679B2 (en) | 2004-10-01 | 2007-08-14 | Advanced Micro Devices, Inc. | Sharing monitored cache lines across multiple cores |
US7313673B2 (en) * | 2005-06-16 | 2007-12-25 | International Business Machines Corporation | Fine grained multi-thread dispatch block mechanism |
US9003421B2 (en) * | 2005-11-28 | 2015-04-07 | Intel Corporation | Acceleration threads on idle OS-visible thread execution units |
KR100763200B1 (en) | 2006-02-24 | 2007-10-04 | 삼성전자주식회사 | Method and apparatus for interruptible synchronization for thread |
KR100771877B1 (en) | 2006-07-21 | 2007-11-01 | 삼성전자주식회사 | Method and apparatus for processing command set protocol for preventing from malfunction |
US20080022079A1 (en) * | 2006-07-24 | 2008-01-24 | Archer Charles J | Executing an allgather operation with an alltoallv operation in a parallel computer |
US8443341B2 (en) * | 2006-11-09 | 2013-05-14 | Rogue Wave Software, Inc. | System for and method of capturing application characteristics data from a computer system and modeling target system |
US8239657B2 (en) * | 2007-02-07 | 2012-08-07 | Qualcomm Incorporated | Address translation method and apparatus |
US8898438B2 (en) * | 2007-03-14 | 2014-11-25 | XMOS Ltd. | Processor architecture for use in scheduling threads in response to communication activity |
US7937532B2 (en) * | 2007-03-30 | 2011-05-03 | Intel Corporation | Method and apparatus for speculative prefetching in a multi-processor/multi-core message-passing machine |
JP4420055B2 (en) | 2007-04-18 | 2010-02-24 | 日本電気株式会社 | Multi-thread processor and inter-thread synchronous operation method used therefor |
US8161480B2 (en) | 2007-05-29 | 2012-04-17 | International Business Machines Corporation | Performing an allreduce operation using shared memory |
CN101681260B (en) * | 2007-06-20 | 2013-04-17 | 富士通株式会社 | Arithmetic unit |
US20090006663A1 (en) * | 2007-06-27 | 2009-01-01 | Archer Charles J | Direct Memory Access ('DMA') Engine Assisted Local Reduction |
JP2009110209A (en) * | 2007-10-29 | 2009-05-21 | Panasonic Corp | Arithmetic processing unit, processor, program conversion device, and program |
US8422402B2 (en) | 2008-04-01 | 2013-04-16 | International Business Machines Corporation | Broadcasting a message in a parallel computer |
US8281053B2 (en) | 2008-07-21 | 2012-10-02 | International Business Machines Corporation | Performing an all-to-all data exchange on a plurality of data buffers by performing swap operations |
CN101788922B (en) * | 2009-01-22 | 2013-12-25 | 国际商业机器公司 | Method and device for realizing transaction storage system based on auxiliary thread |
KR101581311B1 (en) * | 2009-03-11 | 2015-12-31 | 삼성전자주식회사 | Flash memory apparatus and method of controlling the same |
US8156275B2 (en) * | 2009-05-13 | 2012-04-10 | Apple Inc. | Power managed lock optimization |
US8364862B2 (en) * | 2009-06-11 | 2013-01-29 | Intel Corporation | Delegating a poll operation to another device |
DE112009005006T5 (en) * | 2009-06-26 | 2013-01-10 | Intel Corporation | Optimizations for an Unbound Transactional Storage System (UTM) |
EP2284693A1 (en) | 2009-08-03 | 2011-02-16 | C.R.F. Società Consortile per Azioni | Wait instruction |
US8695002B2 (en) * | 2009-10-20 | 2014-04-08 | Lantiq Deutschland Gmbh | Multi-threaded processors and multi-processor systems comprising shared resources |
US9424087B2 (en) | 2010-04-29 | 2016-08-23 | International Business Machines Corporation | Optimizing collective operations |
US8949577B2 (en) | 2010-05-28 | 2015-02-03 | International Business Machines Corporation | Performing a deterministic reduction operation in a parallel computer |
US8776081B2 (en) | 2010-09-14 | 2014-07-08 | International Business Machines Corporation | Send-side matching of data communications messages |
US8566841B2 (en) | 2010-11-10 | 2013-10-22 | International Business Machines Corporation | Processing communications events in parallel active messaging interface by awakening thread from wait state |
CN102736945B (en) * | 2011-03-31 | 2016-05-18 | 国际商业机器公司 | A kind of method and system of the Multi-instance running application |
CN102760176B (en) * | 2011-04-29 | 2015-02-11 | 无锡江南计算技术研究所 | Hardware transaction level simulation method, engine and system |
US8893083B2 (en) | 2011-08-09 | 2014-11-18 | International Business Machines Coporation | Collective operation protocol selection in a parallel computer |
US8910178B2 (en) | 2011-08-10 | 2014-12-09 | International Business Machines Corporation | Performing a global barrier operation in a parallel computer |
US8667501B2 (en) | 2011-08-10 | 2014-03-04 | International Business Machines Corporation | Performing a local barrier operation |
US8868843B2 (en) | 2011-11-30 | 2014-10-21 | Advanced Micro Devices, Inc. | Hardware filter for tracking block presence in large caches |
US9798548B2 (en) | 2011-12-21 | 2017-10-24 | Nvidia Corporation | Methods and apparatus for scheduling instructions using pre-decode data |
CN106909444B (en) | 2011-12-22 | 2021-01-12 | 英特尔公司 | Instruction processing apparatus for specifying instructions for application thread performance state and related methods |
JP5819184B2 (en) | 2011-12-28 | 2015-11-18 | 富士通株式会社 | Information processing apparatus and information processing apparatus control method |
US20150143071A1 (en) * | 2011-12-30 | 2015-05-21 | Ravi L. Sahita | Memory event notification |
US8706847B2 (en) | 2012-02-09 | 2014-04-22 | International Business Machines Corporation | Initiating a collective operation in a parallel computer |
US9495135B2 (en) | 2012-02-09 | 2016-11-15 | International Business Machines Corporation | Developing collective operations for a parallel computer |
EP2831721B1 (en) | 2012-03-30 | 2020-08-26 | Intel Corporation | Context switching mechanism for a processing core having a general purpose cpu core and a tightly coupled accelerator |
US9218288B2 (en) | 2012-06-15 | 2015-12-22 | International Business Machines Corporation | Monitoring a value in storage without repeated storage access |
US9256455B2 (en) * | 2012-11-20 | 2016-02-09 | Red Hat Isreal, Ltd. | Delivery of events from a virtual machine to host CPU using memory monitoring instructions |
US9141454B2 (en) * | 2012-12-27 | 2015-09-22 | Intel Corporation | Signaling software recoverable errors |
US10705961B2 (en) | 2013-09-27 | 2020-07-07 | Intel Corporation | Scalably mechanism to implement an instruction that monitors for writes to an address |
US9218185B2 (en) | 2014-03-27 | 2015-12-22 | International Business Machines Corporation | Multithreading capability information retrieval |
US9921848B2 (en) | 2014-03-27 | 2018-03-20 | International Business Machines Corporation | Address expansion and contraction in a multithreading computer system |
US9804846B2 (en) | 2014-03-27 | 2017-10-31 | International Business Machines Corporation | Thread context preservation in a multithreading computer system |
US9594660B2 (en) | 2014-03-27 | 2017-03-14 | International Business Machines Corporation | Multithreading computer system and program product for executing a query instruction for idle time accumulation among cores |
US10102004B2 (en) | 2014-03-27 | 2018-10-16 | International Business Machines Corporation | Hardware counters to track utilization in a multithreading computer system |
US9417876B2 (en) | 2014-03-27 | 2016-08-16 | International Business Machines Corporation | Thread context restoration in a multithreading computer system |
US9354883B2 (en) | 2014-03-27 | 2016-05-31 | International Business Machines Corporation | Dynamic enablement of multithreading |
US9552033B2 (en) * | 2014-04-22 | 2017-01-24 | Qualcomm Incorporated | Latency-based power mode units for controlling power modes of processor cores, and related methods and systems |
US9778949B2 (en) * | 2014-05-05 | 2017-10-03 | Google Inc. | Thread waiting in a multithreaded processor architecture |
GB2529899B (en) * | 2014-09-08 | 2021-06-23 | Advanced Risc Mach Ltd | Shared Resources in a Data Processing Apparatus for Executing a Plurality of Threads |
JP6227151B2 (en) * | 2014-10-03 | 2017-11-08 | インテル・コーポレーション | A scalable mechanism for executing monitoring instructions for writing to addresses |
CN107111523B (en) * | 2014-11-11 | 2020-10-20 | 瑞萨电子株式会社 | Command execution control system and command execution control method |
WO2016088220A1 (en) * | 2014-12-03 | 2016-06-09 | 株式会社日立製作所 | Computer and control method for logical processors |
CN105843592A (en) * | 2015-01-12 | 2016-08-10 | 芋头科技(杭州)有限公司 | System for implementing script operation in preset embedded system |
GB2537115B (en) * | 2015-04-02 | 2021-08-25 | Advanced Risc Mach Ltd | Event monitoring in a multi-threaded data processing apparatus |
KR102476357B1 (en) | 2015-08-06 | 2022-12-09 | 삼성전자주식회사 | Clock management unit, integrated circuit and system on chip adopting the same, and clock managing method |
US9916178B2 (en) * | 2015-09-25 | 2018-03-13 | Intel Corporation | Technologies for integrated thread scheduling |
GB2544994A (en) * | 2015-12-02 | 2017-06-07 | Swarm64 As | Data processing |
CN105677487B (en) * | 2016-01-12 | 2019-02-15 | 浪潮通用软件有限公司 | A kind of method and device controlling resource occupation |
US11023233B2 (en) | 2016-02-09 | 2021-06-01 | Intel Corporation | Methods, apparatus, and instructions for user level thread suspension |
US11061730B2 (en) * | 2016-11-18 | 2021-07-13 | Red Hat Israel, Ltd. | Efficient scheduling for hyper-threaded CPUs using memory monitoring |
CN108255587B (en) * | 2016-12-29 | 2021-08-24 | 展讯通信(上海)有限公司 | Synchronous multi-thread processor |
US10353817B2 (en) * | 2017-03-07 | 2019-07-16 | International Business Machines Corporation | Cache miss thread balancing |
TWI648620B (en) * | 2017-08-07 | 2019-01-21 | 慧榮科技股份有限公司 | Memory device and operation instruction error processing method |
KR102043538B1 (en) * | 2018-01-18 | 2019-11-11 | 주식회사 알티스트 | Computing system for monitoring process by applying partitioning |
CN109508229A (en) * | 2018-09-19 | 2019-03-22 | 安徽慧视金瞳科技有限公司 | A kind of matching process that multiple spot is drawn simultaneously |
US11068407B2 (en) | 2018-10-26 | 2021-07-20 | International Business Machines Corporation | Synchronized access to data in shared memory by protecting the load target address of a load-reserve instruction |
US10884740B2 (en) | 2018-11-08 | 2021-01-05 | International Business Machines Corporation | Synchronized access to data in shared memory by resolving conflicting accesses by co-located hardware threads |
US11119781B2 (en) | 2018-12-11 | 2021-09-14 | International Business Machines Corporation | Synchronized access to data in shared memory by protecting the load target address of a fronting load |
US11086672B2 (en) | 2019-05-07 | 2021-08-10 | International Business Machines Corporation | Low latency management of processor core wait state |
US11106608B1 (en) | 2020-06-22 | 2021-08-31 | International Business Machines Corporation | Synchronizing access to shared memory by extending protection for a target address of a store-conditional request |
CN111857591A (en) * | 2020-07-20 | 2020-10-30 | 北京百度网讯科技有限公司 | Method, apparatus, device and computer-readable storage medium for executing instructions |
CN114489792B (en) * | 2021-03-25 | 2022-10-11 | 沐曦集成电路(上海)有限公司 | Processor device and instruction execution method thereof |
US11693776B2 (en) | 2021-06-18 | 2023-07-04 | International Business Machines Corporation | Variable protection window extension for a target address of a store-conditional request |
CN116185891B (en) * | 2023-04-27 | 2023-07-21 | 珠海妙存科技有限公司 | Descriptor management method |
Citations (33)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US4491910A (en) * | 1982-02-22 | 1985-01-01 | Texas Instruments Incorporated | Microcomputer having data shift within memory |
US5357617A (en) * | 1991-11-22 | 1994-10-18 | International Business Machines Corporation | Method and apparatus for substantially concurrent multiple instruction thread processing by a single pipeline processor |
US5428757A (en) * | 1992-04-29 | 1995-06-27 | International Business Machines Corporation | Method for reducing translation look aside buffer purges in a multitasking system |
US5493660A (en) * | 1992-10-06 | 1996-02-20 | Hewlett-Packard Company | Software assisted hardware TLB miss handler |
US5524247A (en) * | 1992-01-30 | 1996-06-04 | Kabushiki Kaisha Toshiba | System for scheduling programming units to a resource based on status variables indicating a lock or lock-wait state thereof |
US5530597A (en) * | 1992-07-21 | 1996-06-25 | Advanced Micro Devices, Inc. | Apparatus and method for disabling interrupt masks in processors or the like |
US5546593A (en) * | 1992-05-18 | 1996-08-13 | Matsushita Electric Industrial Co., Ltd. | Multistream instruction processor able to reduce interlocks by having a wait state for an instruction stream |
US5584014A (en) * | 1994-12-20 | 1996-12-10 | Sun Microsystems, Inc. | Apparatus and method to preserve data in a set associative memory device |
US5584031A (en) * | 1993-11-09 | 1996-12-10 | Motorola Inc. | System and method for executing a low power delay instruction |
US5761522A (en) * | 1995-05-24 | 1998-06-02 | Fuji Xerox Co., Ltd. | Program control system programmable to selectively execute a plurality of programs |
US5835705A (en) * | 1997-03-11 | 1998-11-10 | International Business Machines Corporation | Method and system for performance per-thread monitoring in a multithreaded processor |
US5933627A (en) * | 1996-07-01 | 1999-08-03 | Sun Microsystems | Thread switch on blocked load or store using instruction thread field |
US5961639A (en) * | 1996-12-16 | 1999-10-05 | International Business Machines Corporation | Processor and method for dynamically inserting auxiliary instructions within an instruction stream during execution |
US6016542A (en) * | 1997-12-31 | 2000-01-18 | Intel Corporation | Detecting long latency pipeline stalls for thread switching |
US6341347B1 (en) * | 1999-05-11 | 2002-01-22 | Sun Microsystems, Inc. | Thread switch logic in a multiple-thread processor |
US6351808B1 (en) * | 1999-05-11 | 2002-02-26 | Sun Microsystems, Inc. | Vertically and horizontally threaded processor with multidimensional storage for storing thread data |
US6357016B1 (en) * | 1999-12-09 | 2002-03-12 | Intel Corporation | Method and apparatus for disabling a clock signal within a multithreaded processor |
US6401155B1 (en) * | 1998-12-22 | 2002-06-04 | Philips Electronics North America Corporation | Interrupt/software-controlled thread processing |
US6435905B1 (en) * | 2000-12-21 | 2002-08-20 | Hon Hai Precision Ind. Co., Ltd. | Compact electrical connector having boardlocks |
US6457082B1 (en) * | 1998-12-28 | 2002-09-24 | Compaq Information Technologies Group, L.P. | Break event generation during transitions between modes of operation in a computer system |
US6463511B2 (en) * | 2000-12-29 | 2002-10-08 | Intel Corporation | System and method for high performance execution of locked memory instructions in a system with distributed memory and a restrictive memory model |
US6463482B1 (en) * | 1998-06-22 | 2002-10-08 | Nec Corporation | Control, of conflict between MPC transfer and DMC transfer with measurement of instruction execution time |
US6493741B1 (en) * | 1999-10-01 | 2002-12-10 | Compaq Information Technologies Group, L.P. | Method and apparatus to quiesce a portion of a simultaneous multithreaded central processing unit |
US6496925B1 (en) * | 1999-12-09 | 2002-12-17 | Intel Corporation | Method and apparatus for processing an event occurrence within a multithreaded processor |
US6505229B1 (en) * | 1998-09-25 | 2003-01-07 | Intelect Communications, Inc. | Method for allowing multiple processing threads and tasks to execute on one or more processor units for embedded real-time processor systems |
US20030079094A1 (en) * | 2001-10-19 | 2003-04-24 | Ravi Rajwar | Concurrent execution of critical sections by eliding ownership of locks |
US20030126375A1 (en) * | 2001-12-31 | 2003-07-03 | Hill David L. | Coherency techniques for suspending execution of a thread until a specified memory access occurs |
US20030123186A1 (en) * | 1995-02-03 | 2003-07-03 | Masaaki Sano | Magnetic disk apparatus |
US20030126416A1 (en) * | 2001-12-31 | 2003-07-03 | Marr Deborah T. | Suspending execution of a thread in a multi-threaded processor |
US20030126379A1 (en) * | 2001-12-31 | 2003-07-03 | Shiv Kaushik | Instruction sequences for suspending execution of a thread until a specified memory access occurs |
US6625698B2 (en) * | 2000-12-28 | 2003-09-23 | Unisys Corporation | Method and apparatus for controlling memory storage locks based on cache line ownership |
US6671795B1 (en) * | 2000-01-21 | 2003-12-30 | Intel Corporation | Method and apparatus for pausing execution in a processor or the like |
US6931639B1 (en) * | 2000-08-24 | 2005-08-16 | International Business Machines Corporation | Method for implementing a variable-partitioned queue for simultaneous multithreaded processors |
Family Cites Families (13)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPS59111526A (en) | 1982-12-16 | 1984-06-27 | Fujitsu Ltd | Signal processing system |
JPH06105460B2 (en) | 1988-06-07 | 1994-12-21 | 富士通株式会社 | Multiprocessor processor switching device |
GB8817911D0 (en) | 1988-07-27 | 1988-09-01 | Int Computers Ltd | Data processing apparatus |
US4965718A (en) | 1988-09-29 | 1990-10-23 | International Business Machines Corporation | Data processing system incorporating a memory resident directive for synchronizing multiple tasks among plurality of processing elements by monitoring alternation of semaphore data |
US5546037A (en) | 1993-11-15 | 1996-08-13 | Cirrus Logic, Inc. | NAPNOP circuit for conserving power in computer systems |
JPH096633A (en) | 1995-06-07 | 1997-01-10 | Internatl Business Mach Corp <Ibm> | Method and system for operation of high-performance multiplelogical route in data-processing system |
DE69739726D1 (en) | 1996-08-27 | 2010-02-11 | Panasonic Corp | A multi-thread processor for processing multiple instruction streams independently of one another through flexible throughput control in each instruction stream |
WO1998043193A2 (en) | 1997-03-21 | 1998-10-01 | University Of Maryland | Spawn-join instruction set architecture for providing explicit multithreading |
UA55489C2 (en) * | 1997-10-07 | 2003-04-15 | Каналь+ Сосьєте Анонім | Device for processing information in a number of information flows |
US6076157A (en) | 1997-10-23 | 2000-06-13 | International Business Machines Corporation | Method and apparatus to force a thread switch in a multithreaded processor |
US6920634B1 (en) * | 1998-08-03 | 2005-07-19 | International Business Machines Corporation | Detecting and causing unsafe latent accesses to a resource in multi-threaded programs |
US6535905B1 (en) * | 1999-04-29 | 2003-03-18 | Intel Corporation | Method and apparatus for thread switching within a multithreaded processor |
US7856633B1 (en) | 2000-03-24 | 2010-12-21 | Intel Corporation | LRU cache replacement for a partitioned set associative cache |
-
2001
- 2001-12-31 US US10/039,579 patent/US7363474B2/en active Active
-
2002
- 2002-12-11 JP JP2003558691A patent/JP4601958B2/en not_active Expired - Lifetime
- 2002-12-11 WO PCT/US2002/039786 patent/WO2003058447A2/en active Application Filing
- 2002-12-11 KR KR1020047010389A patent/KR100814993B1/en active IP Right Grant
- 2002-12-11 GB GB0412867A patent/GB2400947B/en not_active Expired - Lifetime
- 2002-12-11 DE DE10297596.5T patent/DE10297596B4/en not_active Expired - Lifetime
- 2002-12-11 RU RU2004123622/09A patent/RU2308754C2/en not_active IP Right Cessation
- 2002-12-11 CN CNB028265904A patent/CN100383740C/en not_active Expired - Lifetime
- 2002-12-11 AU AU2002364557A patent/AU2002364557A1/en not_active Abandoned
- 2002-12-11 DE DE10297856.5A patent/DE10297856B4/en not_active Expired - Lifetime
- 2002-12-25 TW TW091137282A patent/TWI322959B/en not_active IP Right Cessation
- 2002-12-25 TW TW097151472A patent/TWI590153B/en not_active IP Right Cessation
-
2004
- 2004-12-07 HK HK04109671A patent/HK1066883A1/en not_active IP Right Cessation
-
2007
- 2007-08-08 US US11/891,076 patent/US20080034190A1/en not_active Abandoned
-
2008
- 2008-03-26 JP JP2008081180A patent/JP4990829B2/en not_active Expired - Lifetime
Patent Citations (33)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US4491910A (en) * | 1982-02-22 | 1985-01-01 | Texas Instruments Incorporated | Microcomputer having data shift within memory |
US5357617A (en) * | 1991-11-22 | 1994-10-18 | International Business Machines Corporation | Method and apparatus for substantially concurrent multiple instruction thread processing by a single pipeline processor |
US5524247A (en) * | 1992-01-30 | 1996-06-04 | Kabushiki Kaisha Toshiba | System for scheduling programming units to a resource based on status variables indicating a lock or lock-wait state thereof |
US5428757A (en) * | 1992-04-29 | 1995-06-27 | International Business Machines Corporation | Method for reducing translation look aside buffer purges in a multitasking system |
US5546593A (en) * | 1992-05-18 | 1996-08-13 | Matsushita Electric Industrial Co., Ltd. | Multistream instruction processor able to reduce interlocks by having a wait state for an instruction stream |
US5530597A (en) * | 1992-07-21 | 1996-06-25 | Advanced Micro Devices, Inc. | Apparatus and method for disabling interrupt masks in processors or the like |
US5493660A (en) * | 1992-10-06 | 1996-02-20 | Hewlett-Packard Company | Software assisted hardware TLB miss handler |
US5584031A (en) * | 1993-11-09 | 1996-12-10 | Motorola Inc. | System and method for executing a low power delay instruction |
US5584014A (en) * | 1994-12-20 | 1996-12-10 | Sun Microsystems, Inc. | Apparatus and method to preserve data in a set associative memory device |
US20030123186A1 (en) * | 1995-02-03 | 2003-07-03 | Masaaki Sano | Magnetic disk apparatus |
US5761522A (en) * | 1995-05-24 | 1998-06-02 | Fuji Xerox Co., Ltd. | Program control system programmable to selectively execute a plurality of programs |
US5933627A (en) * | 1996-07-01 | 1999-08-03 | Sun Microsystems | Thread switch on blocked load or store using instruction thread field |
US5961639A (en) * | 1996-12-16 | 1999-10-05 | International Business Machines Corporation | Processor and method for dynamically inserting auxiliary instructions within an instruction stream during execution |
US5835705A (en) * | 1997-03-11 | 1998-11-10 | International Business Machines Corporation | Method and system for performance per-thread monitoring in a multithreaded processor |
US6016542A (en) * | 1997-12-31 | 2000-01-18 | Intel Corporation | Detecting long latency pipeline stalls for thread switching |
US6463482B1 (en) * | 1998-06-22 | 2002-10-08 | Nec Corporation | Control, of conflict between MPC transfer and DMC transfer with measurement of instruction execution time |
US6505229B1 (en) * | 1998-09-25 | 2003-01-07 | Intelect Communications, Inc. | Method for allowing multiple processing threads and tasks to execute on one or more processor units for embedded real-time processor systems |
US6401155B1 (en) * | 1998-12-22 | 2002-06-04 | Philips Electronics North America Corporation | Interrupt/software-controlled thread processing |
US6457082B1 (en) * | 1998-12-28 | 2002-09-24 | Compaq Information Technologies Group, L.P. | Break event generation during transitions between modes of operation in a computer system |
US6351808B1 (en) * | 1999-05-11 | 2002-02-26 | Sun Microsystems, Inc. | Vertically and horizontally threaded processor with multidimensional storage for storing thread data |
US6341347B1 (en) * | 1999-05-11 | 2002-01-22 | Sun Microsystems, Inc. | Thread switch logic in a multiple-thread processor |
US6493741B1 (en) * | 1999-10-01 | 2002-12-10 | Compaq Information Technologies Group, L.P. | Method and apparatus to quiesce a portion of a simultaneous multithreaded central processing unit |
US6357016B1 (en) * | 1999-12-09 | 2002-03-12 | Intel Corporation | Method and apparatus for disabling a clock signal within a multithreaded processor |
US6496925B1 (en) * | 1999-12-09 | 2002-12-17 | Intel Corporation | Method and apparatus for processing an event occurrence within a multithreaded processor |
US6671795B1 (en) * | 2000-01-21 | 2003-12-30 | Intel Corporation | Method and apparatus for pausing execution in a processor or the like |
US6931639B1 (en) * | 2000-08-24 | 2005-08-16 | International Business Machines Corporation | Method for implementing a variable-partitioned queue for simultaneous multithreaded processors |
US6435905B1 (en) * | 2000-12-21 | 2002-08-20 | Hon Hai Precision Ind. Co., Ltd. | Compact electrical connector having boardlocks |
US6625698B2 (en) * | 2000-12-28 | 2003-09-23 | Unisys Corporation | Method and apparatus for controlling memory storage locks based on cache line ownership |
US6463511B2 (en) * | 2000-12-29 | 2002-10-08 | Intel Corporation | System and method for high performance execution of locked memory instructions in a system with distributed memory and a restrictive memory model |
US20030079094A1 (en) * | 2001-10-19 | 2003-04-24 | Ravi Rajwar | Concurrent execution of critical sections by eliding ownership of locks |
US20030126379A1 (en) * | 2001-12-31 | 2003-07-03 | Shiv Kaushik | Instruction sequences for suspending execution of a thread until a specified memory access occurs |
US20030126416A1 (en) * | 2001-12-31 | 2003-07-03 | Marr Deborah T. | Suspending execution of a thread in a multi-threaded processor |
US20030126375A1 (en) * | 2001-12-31 | 2003-07-03 | Hill David L. | Coherency techniques for suspending execution of a thread until a specified memory access occurs |
Cited By (66)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US8516484B2 (en) | 2008-02-01 | 2013-08-20 | International Business Machines Corporation | Wake-and-go mechanism for a data processing system |
US8127080B2 (en) | 2008-02-01 | 2012-02-28 | International Business Machines Corporation | Wake-and-go mechanism with system address bus transaction master |
US20090199183A1 (en) * | 2008-02-01 | 2009-08-06 | Arimilli Ravi K | Wake-and-Go Mechanism with Hardware Private Array |
US20090199028A1 (en) * | 2008-02-01 | 2009-08-06 | Arimilli Ravi K | Wake-and-Go Mechanism with Data Exclusivity |
US20090199189A1 (en) * | 2008-02-01 | 2009-08-06 | Arimilli Ravi K | Parallel Lock Spinning Using Wake-and-Go Mechanism |
US20090199029A1 (en) * | 2008-02-01 | 2009-08-06 | Arimilli Ravi K | Wake-and-Go Mechanism with Data Monitoring |
US8725992B2 (en) | 2008-02-01 | 2014-05-13 | International Business Machines Corporation | Programming language exposing idiom calls to a programming idiom accelerator |
US8788795B2 (en) | 2008-02-01 | 2014-07-22 | International Business Machines Corporation | Programming idiom accelerator to examine pre-fetched instruction streams for multiple processors |
US8452947B2 (en) | 2008-02-01 | 2013-05-28 | International Business Machines Corporation | Hardware wake-and-go mechanism and content addressable memory with instruction pre-fetch look-ahead to detect programming idioms |
US20100287341A1 (en) * | 2008-02-01 | 2010-11-11 | Arimilli Ravi K | Wake-and-Go Mechanism with System Address Bus Transaction Master |
US20100293340A1 (en) * | 2008-02-01 | 2010-11-18 | Arimilli Ravi K | Wake-and-Go Mechanism with System Bus Response |
US20110173630A1 (en) * | 2008-02-01 | 2011-07-14 | Arimilli Ravi K | Central Repository for Wake-and-Go Mechanism |
US20110173419A1 (en) * | 2008-02-01 | 2011-07-14 | Arimilli Ravi K | Look-Ahead Wake-and-Go Engine With Speculative Execution |
US20090199197A1 (en) * | 2008-02-01 | 2009-08-06 | International Business Machines Corporation | Wake-and-Go Mechanism with Dynamic Allocation in Hardware Private Array |
US20110173625A1 (en) * | 2008-02-01 | 2011-07-14 | Arimilli Ravi K | Wake-and-Go Mechanism with Prioritization of Threads |
US8640142B2 (en) | 2008-02-01 | 2014-01-28 | International Business Machines Corporation | Wake-and-go mechanism with dynamic allocation in hardware private array |
US20110173631A1 (en) * | 2008-02-01 | 2011-07-14 | Arimilli Ravi K | Wake-and-Go Mechanism for a Data Processing System |
US20110173417A1 (en) * | 2008-02-01 | 2011-07-14 | Arimilli Ravi K | Programming Idiom Accelerators |
US8640141B2 (en) | 2008-02-01 | 2014-01-28 | International Business Machines Corporation | Wake-and-go mechanism with hardware private array |
US8732683B2 (en) | 2008-02-01 | 2014-05-20 | International Business Machines Corporation | Compiler providing idiom to idiom accelerator |
US8386822B2 (en) | 2008-02-01 | 2013-02-26 | International Business Machines Corporation | Wake-and-go mechanism with data monitoring |
US8145849B2 (en) | 2008-02-01 | 2012-03-27 | International Business Machines Corporation | Wake-and-go mechanism with system bus response |
US8171476B2 (en) | 2008-02-01 | 2012-05-01 | International Business Machines Corporation | Wake-and-go mechanism with prioritization of threads |
US20090199184A1 (en) * | 2008-02-01 | 2009-08-06 | Arimilli Ravi K | Wake-and-Go Mechanism With Software Save of Thread State |
US8225120B2 (en) | 2008-02-01 | 2012-07-17 | International Business Machines Corporation | Wake-and-go mechanism with data exclusivity |
US8880853B2 (en) * | 2008-02-01 | 2014-11-04 | International Business Machines Corporation | CAM-based wake-and-go snooping engine for waking a thread put to sleep for spinning on a target address lock |
US8250396B2 (en) | 2008-02-01 | 2012-08-21 | International Business Machines Corporation | Hardware wake-and-go mechanism for a data processing system |
US8312458B2 (en) | 2008-02-01 | 2012-11-13 | International Business Machines Corporation | Central repository for wake-and-go mechanism |
US8316218B2 (en) | 2008-02-01 | 2012-11-20 | International Business Machines Corporation | Look-ahead wake-and-go engine with speculative execution |
US8341635B2 (en) | 2008-02-01 | 2012-12-25 | International Business Machines Corporation | Hardware wake-and-go mechanism with look-ahead polling |
US8612977B2 (en) | 2008-02-01 | 2013-12-17 | International Business Machines Corporation | Wake-and-go mechanism with software save of thread state |
US8886919B2 (en) | 2009-04-16 | 2014-11-11 | International Business Machines Corporation | Remote update programming idiom accelerator with allocated processor resources |
US8230201B2 (en) * | 2009-04-16 | 2012-07-24 | International Business Machines Corporation | Migrating sleeping and waking threads between wake-and-go mechanisms in a multiple processor data processing system |
US8145723B2 (en) | 2009-04-16 | 2012-03-27 | International Business Machines Corporation | Complex remote update programming idiom accelerator |
US8082315B2 (en) | 2009-04-16 | 2011-12-20 | International Business Machines Corporation | Programming idiom accelerator for remote update |
US20100268790A1 (en) * | 2009-04-16 | 2010-10-21 | International Business Machines Corporation | Complex Remote Update Programming Idiom Accelerator |
US20100269115A1 (en) * | 2009-04-16 | 2010-10-21 | International Business Machines Corporation | Managing Threads in a Wake-and-Go Engine |
US20100268915A1 (en) * | 2009-04-16 | 2010-10-21 | International Business Machines Corporation | Remote Update Programming Idiom Accelerator with Allocated Processor Resources |
US8990597B2 (en) | 2009-12-18 | 2015-03-24 | Intel Corporation | Instruction for enabling a processor wait state |
US9032232B2 (en) | 2009-12-18 | 2015-05-12 | Intel Corporation | Instruction for enabling a processor wait state |
US20110173420A1 (en) * | 2010-01-08 | 2011-07-14 | International Business Machines Corporation | Processor resume unit |
US20110173422A1 (en) * | 2010-01-08 | 2011-07-14 | International Business Machines Corporation | Pause processor hardware thread until pin |
US8447960B2 (en) | 2010-01-08 | 2013-05-21 | International Business Machines Corporation | Pausing and activating thread state upon pin assertion by external logic monitoring polling loop exit time condition |
US8898671B2 (en) | 2010-07-07 | 2014-11-25 | Panasonic Corporation | Processor that executes a plurality of threads by promoting efficiency of transfer of data that is shared with the plurality of threads |
US8713362B2 (en) | 2010-12-01 | 2014-04-29 | International Business Machines Corporation | Obviation of recovery of data store consistency for application I/O errors |
US8694821B2 (en) * | 2010-12-03 | 2014-04-08 | International Business Machines Corporation | Generation of standby images of applications |
US20120144232A1 (en) * | 2010-12-03 | 2012-06-07 | International Business Machines Corporation | Generation of Standby Images of Applications |
US10152325B2 (en) | 2012-11-28 | 2018-12-11 | Intel Corporation | Instruction and logic to provide pushing buffer copy and store functionality |
US9563425B2 (en) | 2012-11-28 | 2017-02-07 | Intel Corporation | Instruction and logic to provide pushing buffer copy and store functionality |
WO2014084919A1 (en) * | 2012-11-28 | 2014-06-05 | Intel Corporation | Instruction and logic to provide pushing buffer copy and store functionality |
EP3074860A4 (en) * | 2013-11-25 | 2017-10-11 | Marvell World Trade Ltd. | Systems and methods for loop suspension in java programming |
US9256477B2 (en) * | 2014-05-29 | 2016-02-09 | Netapp, Inc. | Lockless waterfall thread communication |
US9304702B2 (en) | 2014-05-29 | 2016-04-05 | Netapp, Inc. | System and method for parallelized performance data collection in a computing system |
US9477521B2 (en) | 2014-05-29 | 2016-10-25 | Netapp, Inc. | Method and system for scheduling repetitive tasks in O(1) |
US9575802B2 (en) | 2014-10-28 | 2017-02-21 | International Business Machines Corporation | Controlling execution of threads in a multi-threaded processor |
US9582324B2 (en) | 2014-10-28 | 2017-02-28 | International Business Machines Corporation | Controlling execution of threads in a multi-threaded processor |
US11061680B2 (en) | 2014-10-28 | 2021-07-13 | International Business Machines Corporation | Instructions controlling access to shared registers of a multi-threaded processor |
US11080064B2 (en) | 2014-10-28 | 2021-08-03 | International Business Machines Corporation | Instructions controlling access to shared registers of a multi-threaded processor |
KR20180033467A (en) * | 2015-06-09 | 2018-04-03 | 옵티멈 세미컨덕터 테크놀로지스 인코포레이티드 | Processor with advanced operating system support |
WO2016200567A1 (en) * | 2015-06-09 | 2016-12-15 | Optimum Semiconductor Technologies, Inc. | Processor with advanced operating system support |
US10908909B2 (en) | 2015-06-09 | 2021-02-02 | Optimum Semiconductor Technologies Inc. | Processor with mode support |
KR102599484B1 (en) | 2015-06-09 | 2023-11-06 | 옵티멈 세미컨덕터 테크놀로지스 인코포레이티드 | Processor with advanced operating system support |
US10275254B2 (en) | 2017-03-08 | 2019-04-30 | International Business Machines Corporation | Spin loop delay instruction |
US10365929B2 (en) | 2017-03-08 | 2019-07-30 | International Business Machines Corporation | Spin loop delay instruction |
US10656950B2 (en) | 2017-03-08 | 2020-05-19 | International Business Machines Corporation | Spin loop delay instruction |
US11231881B2 (en) | 2020-04-02 | 2022-01-25 | Dell Products L.P. | Raid data storage device multi-step command coordination system |
Also Published As
Publication number | Publication date |
---|---|
CN1688974A (en) | 2005-10-26 |
JP4990829B2 (en) | 2012-08-01 |
RU2308754C2 (en) | 2007-10-20 |
WO2003058447A3 (en) | 2004-12-16 |
TW200941340A (en) | 2009-10-01 |
GB2400947A (en) | 2004-10-27 |
JP2008165834A (en) | 2008-07-17 |
CN100383740C (en) | 2008-04-23 |
AU2002364557A1 (en) | 2003-07-24 |
GB0412867D0 (en) | 2004-07-14 |
DE10297856B4 (en) | 2014-07-17 |
TWI322959B (en) | 2010-04-01 |
JP4601958B2 (en) | 2010-12-22 |
WO2003058447A2 (en) | 2003-07-17 |
KR100814993B1 (en) | 2008-03-18 |
TWI590153B (en) | 2017-07-01 |
GB2400947B (en) | 2005-08-24 |
DE10297596B4 (en) | 2017-09-07 |
US7363474B2 (en) | 2008-04-22 |
TW200305820A (en) | 2003-11-01 |
JP2006500639A (en) | 2006-01-05 |
KR20040068993A (en) | 2004-08-02 |
HK1066883A1 (en) | 2005-04-01 |
RU2004123622A (en) | 2005-03-27 |
US20030126186A1 (en) | 2003-07-03 |
DE10297596T5 (en) | 2004-12-02 |
AU2002364557A8 (en) | 2003-07-24 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US7363474B2 (en) | Method and apparatus for suspending execution of a thread until a specified memory access occurs | |
US7127561B2 (en) | Coherency techniques for suspending execution of a thread until a specified memory access occurs | |
US7213093B2 (en) | Queued locks using monitor-memory wait | |
US20030126379A1 (en) | Instruction sequences for suspending execution of a thread until a specified memory access occurs | |
US6484254B1 (en) | Method, apparatus, and system for maintaining processor ordering by checking load addresses of unretired load instructions against snooping store addresses | |
US8539485B2 (en) | Polling using reservation mechanism | |
US9733937B2 (en) | Compare and exchange operation using sleep-wakeup mechanism | |
US8694976B2 (en) | Sleep state mechanism for virtual multithreading | |
US7366879B2 (en) | Alteration of functional unit partitioning scheme in multithreaded processor based upon thread statuses | |
US20030126416A1 (en) | Suspending execution of a thread in a multi-threaded processor | |
Becker et al. | Measuring software performance on linux | |
GB2441903A (en) | Resuming control of resources by a processor on exiting a sleep mode and disabling an associated monitor. |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
STPP | Information on status: patent application and granting procedure in general |
Free format text: RESPONSE TO NON-FINAL OFFICE ACTION ENTERED AND FORWARDED TO EXAMINER |
|
STPP | Information on status: patent application and granting procedure in general |
Free format text: FINAL REJECTION MAILED |
|
STCB | Information on status: application discontinuation |
Free format text: ABANDONED -- FAILURE TO RESPOND TO AN OFFICE ACTION |