Patent Application
20230106087
Chiplets are an emerging technique for integrating various processing functionalities. Generally, a chiplet system is made up of discrete modules (each a “chiplet”) that are integrated on an interposer, and in many examples interconnected as desired through one or more established networks, to provide a system with the desired functionality. The interposer and included chiplets can be packaged together to facilitate interconnection with other components of a larger system. Each chiplet can include one or more individual integrated circuits (ICs), or “chips”, potentially in combination with discrete circuit components, and commonly coupled to a respective substrate to facilitate attachment to the interposer. Most or all chiplets in a system will be individually configured for communication through the one or more established networks.
The configuration of chiplets as individual modules of a system is distinct from such a system being implemented on single chips that contain distinct device blocks (e.g., intellectual property (IP) blocks) on one substrate (e.g., single die), such as a system-on-a-chip (SoC), or multiple discrete packaged devices integrated on a printed circuit board (PCB). In general, chiplets provide better performance (e.g., lower power consumption, reduced latency, etc.) than discrete packaged devices, and chiplets provide greater production benefits than single die chips. These production benefits can include higher yields or reduced development costs and time.
Chiplet systems can include, for example, one or more application (or processor) chiplets and one or more support chiplets. Here, the distinction between application and support chiplets is simply a reference to the likely design scenarios for the chiplet system. Thus, for example, a synthetic vision chiplet system can include, by way of example only, an application chiplet to produce the synthetic vision output along with support chiplets, such as a memory controller chiplet, a sensor interface chiplet, or a communication chiplet. In a typical use case, the synthetic vision designer can design the application chiplet and source the support chiplets from other parties. Thus, the design expenditure (e.g., in terms of time or complexity) is reduced because by avoiding the design and production of functionality embodied in the support chiplets. Chiplets also support the tight integration of IP blocks that can otherwise be difficult, such as those manufactured using different processing technologies or using different feature sizes (or utilizing different contact technologies or spacings). Thus, multiple IC’s or IC assemblies, with different physical, electrical, or communication characteristics can be assembled in a modular manner to provide an assembly providing desired functionalities. Chiplet systems can also facilitate adaptation to suit needs of different larger systems into which the chiplet system will be incorporated. In an example, IC’s or other assemblies can be optimized for the power, speed, or heat generation for a specific function—as can happen with sensors—can be integrated with other devices more easily than attempting to do so on a single die. Additionally, by reducing the overall size of the die, the yield for chiplets tends to be higher than that of more complex, single die devices.
The disclosure will be understood more fully from the detailed description given below and from the accompanying drawings of various embodiments of the disclosure. The drawings, however, should not be taken to limit the disclosure to the specific embodiments, but are for explanation and understanding only.
A typical barrel-multithreaded processor has a fixed pipeline length based on the number of hardware threads. Typically, at each clock, the register file is switched to a new thread for a given pipeline stage. This implies that the minimum time between thread instructions executions (e.g., between adjacent instructions in the thread) is equal to the number of supported hardware threads. Thus, the delay until a thread can execute a next instruction is fixed, independent of the instruction operation. This approach is used in many barrel-multithreaded processors to keep processor core control hardware simple. However, as noted above, this is inefficient on a lightly loaded processor.
To address the lightly-loaded barrel-multithreaded processor issue, a variable length pipeline can be used in the processor. Here, an instruction for a thread can complete early, performing a writeback to a register as soon as possible and without waiting for the complete processor pipeline to complete. This is accomplished by tracking the thread identifier (ID) for the thread when the instruction is in the pipeline and re-inserting the thread ID into a thread ready-to-run queue when the register writeback is completed. Once the thread ID is back in the thread ready-to-run queue, the thread can be rescheduled, potentially enabling much faster rescheduling of a thread than is possible in traditional barrel-multithreaded processor implementations.
An issue that can arise with the variable pipeline length is a register writeback conflict between threads on a single clock cycle. In the traditional barrel-multithreading implementation, such a conflict will not occur because the writeback cycle is at the same stage of the pipeline across threads. Because threads start the pipeline at different clock cycles, only one thread attempts to perform the writeback on any given cycle. However, with the variable length pipeline, a conflict can occur.
To address the register writeback conflict, instruction register writebacks are checked to determine whether or not there is a conflict. If there is no conflict, the instruction can writeback to the register as soon as it is able. Thus, on a move instruction completing at stage two of the pipeline, the instruction can writeback to the register at the clock cycle when stage two is complete. However, if there is a conflict, a delay (e.g., in clock cycles) is tracked with the instruction. Thus, the move instruction can conflict with an earlier instruction with a longer (e.g., later stage) completion. Here, the move instruction can be given a delay, such as a single clock cycle. When the move instruction completes, the delay is consulted and the move instruction is prevented from performing the writeback. Here, the move instruction is held in the pipeline until the delay expires and the writeback can complete. At this time, the thread can be rescheduled.
By providing a variable length pipeline, a significant weakness of barrel-multithreaded processors is addressed. Although barrel-multithreaded processors are generally used when many threads of execution are expected, many programs start with a single thread executing, only later changing to use more threads. Thus, for many programs, the barrel-multithreaded processor can be inefficient until sufficient parallelism (e.g., hardware threads) exist when the pipeline has a fixed length. By introducing the variable-length pipeline, this problem is ameliorated. Additional details and examples are provided below.
The application chiplet 125 is illustrated as including a network-on-chip (NOC) 130 to support a chiplet network 155 for inter-chiplet communications. In example embodiments NOC 130 can be included on the application chiplet 125. In an example, NOC 130 can be defined in response to selected support chiplets (e.g., chiplets 135, 140, and 150) thus enabling a designer to select an appropriate number or chiplet network connections or switches for the NOC 130. In an example, the NOC 130 can be located on a separate chiplet, or even within the interposer 120. In examples as discussed herein, the NOC 130 implements a chiplet protocol interface (CPI) network.
The CPI is a packet-based network that supports virtual channels to enable a flexible and high-speed interaction between chiplets. CPI enables bridging from intra-chiplet networks to the chiplet network 155. For example, the Advanced extensible Interface (AXI) is a widely used specification to design intra-chip communications. AXI specifications, however, cover a great variety of physical design options, such as the number of physical channels, signal timing, power, etc. Within a single chip, these options are generally selected to meet design goals, such as power consumption, speed, etc. However, to achieve the flexibility of the chiplet system, an adapter, such as CPI, is used to interface between the various AXI design options that can be implemented in the various chiplets. By enabling a physical channel to virtual channel mapping and encapsulating time-based signaling with a packetized protocol, CPI bridges intra-chiplet networks across the chiplet network 155.
CPI can use a variety of different physical layers to transmit packets. The physical layer can include simple conductive connections, or can include drivers to increase the voltage, or otherwise facilitate transmitting the signals over longer distances. An example of one such physical layer can include the Advanced Interface Bus (AIB), which in various examples, can be implemented in the interposer 120. AIB transmits and receives data using source synchronous data transfers with a forwarded clock. Packets are transferred across the AIB at single data rate (SDR) or dual data rate (DDR) with respect to the transmitted clock. Various channel widths are supported by AIB. AIB channel widths are in multiples of 20 bits when operated in SDR mode (20, 40, 60, ...), and multiples of 40 bits for DDR mode: (40, 80, 120,...). The AIB channel width includes both transmit and receive signals. The channel can be configured to have a symmetrical number of transmit (TX) and receive (RX) input/outputs (I/Os), or have a non-symmetrical number of transmitters and receivers (e.g., either all transmitters or all receivers). The channel can act as an AIB principal or subordinate depending on which chiplet provides the principal clock. AIB I/O cells support three clocking modes: asynchronous (i.e. non-clocked), SDR, and DDR. In various examples, the non-clocked mode is used for clocks and some control signals. The SDR mode can use dedicated SDR only I/O cells, or dual use SDR/DDR I/O cells.
In an example, CPI packet protocols (e.g., point-to-point or routable) can use symmetrical receive and transmit I/O cells within an AIB channel. The CPI streaming protocol allows more flexible use of the AIB I/O cells. In an example, an AIB channel for streaming mode can configure the I/O cells as all TX, all RX, or half TX and half RX. CPI packet protocols can use an AIB channel in either SDR or DDR operation modes. In an example, the AIB channel is configured in increments of 80 I/O cells (i.e. 40 TX and 40 RX) for SDR mode and 40 I/O cells for DDR mode. The CPI streaming protocol can use an AIB channel in either SDR or DDR operation modes. Here, in an example, the AIB channel is in increments of 40 I/O cells for both SDR and DDR modes. In an example, each AIB channel is assigned a unique interface identifier. The identifier is used during CPI reset and initialization to determine paired AIB channels across adjacent chiplets. In an example, the interface identifier is a 20-bit value comprising a seven-bit chiplet identifier, a seven-bit column identifier, and a six-bit link identifier. The AIB physical layer transmits the interface identifier using an AIB out-of-band shift register. The 20-bit interface identifier is transferred in both directions across an AIB interface using bits 32-51 of the shift registers.
AIB defines a stacked set of AIB channels as an AIB channel column. An AIB channel column has some number of AIB channels, plus an auxiliary channel. The auxiliary channel contains signals used for AIB initialization. All AIB channels (other than the auxiliary channel) within a column are of the same configuration (e.g., all TX, all RX, or half TX and half RX, as well as having the same number of data I/O signals). In an example, AIB channels are numbered in continuous increasing order starting with the AIB channel adjacent to the AUX channel. The AIB channel adjacent to the AUX is defined to be AIB channel zero.
Generally, CPI interfaces on individual chiplets can include serialization-deserialization (SERDES) hardware. SERDES interconnects work well for scenarios in which high-speed signaling with low signal count are desirable. SERDES, however, can result in additional power consumption and longer latencies for multiplexing and demultiplexing, error detection or correction (e.g., using block level cyclic redundancy checking (CRC)), link-level retry, or forward error correction. However, when low latency or energy consumption is a primary concern for ultra-short reach, chiplet-to-chiplet interconnects, a parallel interface with clock rates that allow data transfer with minimal latency can be utilized. CPI includes elements to minimize both latency and energy consumption in these ultra-short reach chiplet interconnects.
For flow control, CPI employs a credit-based technique. A recipient, such as the application chiplet 125, provides a sender, such as the memory controller chiplet 140, with credits that represent available buffers. In an example, a CPI recipient includes a buffer for each virtual channel for a given time-unit of transmission. Thus, if the CPI recipient supports five messages in time and a single virtual channel, the recipient has five buffers arranged in five rows (e.g., one row for each unit time). If four virtual channels are supported, then the recipient has twenty buffers arranged in five rows. Each buffer holds the payload of one CPI packet.
When the sender transmits to the recipient, the sender decrements the available credits based on the transmission. Once all credits for the recipient are consumed, the sender stops sending packets to the recipient. This ensures that the recipient always has an available buffer to store the transmission.
As the recipient processes received packets and frees buffers, the recipient communicates the available buffer space back to the sender. This credit return can then be used by the sender allow transmitting of additional information.
Also illustrated is a chiplet mesh network 160 that uses a direct, chiplet-to-chiplet technique without the need for the NOC 130. The chiplet mesh network 160 can be implemented in CPI, or another chiplet-to-chiplet protocol. The chiplet mesh network 160 generally enables a pipeline of chiplets where one chiplet serves as the interface to the pipeline while other chiplets in the pipeline interface only with themselves.
Additionally, dedicated device interfaces, such as one or more industry standard memory interfaces 145 (such as, for example, synchronous memory interfaces, such as DDR5, DDR 6), can also be used to interconnect chiplets. Connection of a chiplet system or individual chiplets to external devices (such as a larger system can be through a desired interface (for example, a PCIE interface). Such as external interface can be implemented, in an example, through a host interface chiplet 135, which in the depicted example, provides a PCIE interface external to chiplet system 110. Such dedicated interfaces 145 are generally employed when a convention or standard in the industry has converged on such an interface. The illustrated example of a Double Data Rate (DDR) interface 145 connecting the memory controller chiplet 140 to a dynamic random access memory (DRAM) memory device 150 is just such an industry convention.
Of the variety of possible support chiplets, the memory controller chiplet 140 is likely present in the chiplet system 110 due to the near omnipresent use of storage for computer processing as well as sophisticated state-of-the-art for memory devices. Thus, using memory device chiplets 150 and memory controller chiplets 140 produced by others gives chiplet system designers access to robust products by sophisticated producers. Generally, the memory controller chiplet 140 provides a memory device specific interface to read, write, or erase data. Often, the memory controller chiplet 140 can provide additional features, such as error detection, error correction, maintenance operations, or atomic operator execution. For some types of memory, maintenance operations tend to be specific to the memory device 150, such as garbage collection in NAND flash or storage class memories, temperature adjustments (e.g., cross temperature management) in NAND flash memories. In an example, the maintenance operations can include logical-to-physical (L2P) mapping or management to provide a level of indirection between the physical and logical representation of data. In other types of memory, for example DRAM, some memory operations, such as refresh can be controlled by a host processor or of a memory controller at some times, and at other times controlled by the DRAM memory device, or by logic associated with one or more DRAM devices, such as an interface chip (in an example, a buffer).
Atomic operators are a data manipulation that, for example, can be performed by the memory controller chiplet 140. In other chiplet systems, the atomic operators can be performed by other chiplets. For example, an atomic operator of “increment” can be specified in a command by the application chiplet 125, the command including a memory address and possibly an increment value. Upon receiving the command, the memory controller chiplet 140 retrieves a number from the specified memory address, increments the number by the amount specified in the command, and stores the result. Upon a successful completion, the memory controller chiplet 140 provides an indication of the commands success to the application chiplet 125. Atomic operators avoid transmitting the data across the chiplet network 160, resulting in lower latency execution of such commands.
Atomic operators can be classified as built-in atomics or programmable (e.g., custom) atomics. Built-in atomics are a finite set of operations that are immutably implemented in hardware. Programmable atomics are small programs that can execute on a programmable atomic unit (PAU) (e.g., a custom atomic unit (CAU)) of the memory controller chiplet 140.
The memory device chiplet 150 can be, or include any combination of, volatile memory devices or non-volatile memories. Examples of volatile memory devices include, but are not limited to, random access memory (RAM)—such as DRAM) synchronous DRAM (SDRAM), graphics double data rate type 6 SDRAM (GDDR6 SDRAM), among others. Examples of non-volatile memory devices include, but are not limited to, negative-and-(NAND)-type flash memory, storage class memory (e.g., phase-change memory or memristor based technologies), ferroelectric RAM (FeRAM), among others. The illustrated example includes the memory device 150 as a chiplet, however, the memory device 150 can reside elsewhere, such as in a different package on the peripheral board 105. For many applications, multiple memory device chiplets can be provided. In an example, these memory device chiplets can each implement one or multiple storage technologies. In an example, a memory chiplet can include, multiple stacked memory die of different technologies, for example one or more static random access memory (SRAM) devices stacked or otherwise in communication with one or more dynamic random access memory (DRAM) devices. Memory controller 140 can also serve to coordinate operations between multiple memory chiplets in chiplet system 110; for example, to utilize one or more memory chiplets in one or more levels of cache storage, and to use one or more additional memory chiplets as main memory. Chiplet system 110 can also include multiple memory controllers 140, as can be used to provide memory control functionality for separate processors, sensors, networks, etc. A chiplet architecture, such as chiplet system 110 offers advantages in allowing adaptation to different memory storage technologies; and different memory interfaces, through updated chiplet configurations, without requiring redesign of the remainder of the system structure.
The off-die memory controller 220 is directly coupled to the off-die memory 275 (e.g., via a bus or other communication connection) to provide write operations and read operations to and from the one or more off-die memory, such as off-die memory 275 and off-die memory 280. In the depicted example, the off-die memory controller 220 is also coupled for output to the atomic and merge unit 250, and for input to the cache controller 215 (e.g., a memory side cache controller).
In the example configuration, cache controller 215 is directly coupled to the cache 210, and can be coupled to the network communication interface 225 for input (such as incoming read or write requests), and coupled for output to the off-die memory controller 220.
The network communication interface 225 includes a packet decoder 230, network input queues 235, a packet encoder 240, and network output queues 245 to support a packet-based chiplet network 285, such as CPI. The chiplet network 285 can provide packet routing between and among processors, memory controllers, hybrid threading processors, configurable processing circuits, or communication interfaces. In such a packet-based communication system, each packet typically includes destination and source addressing, along with any data payload or instruction. In an example, the chiplet network 285 can be implemented as a collection of crossbar switches having a folded Clos configuration, or a mesh network providing for additional connections, depending upon the configuration.
In various examples, the chiplet network 285 can be part of an asynchronous switching fabric. Here, a data packet can be routed along any of various paths, such that the arrival of any selected data packet at an addressed destination can occur at any of multiple different times, depending upon the routing. Additionally, chiplet network 285 can be implemented at least in part as a synchronous communication network, such as a synchronous mesh communication network. Both configurations of communication networks are contemplated for use for examples in accordance with the present disclosure.
The memory controller chiplet 205 can receive a packet having, for example, a source address, a read request, and a physical address. In response, the off-die memory controller 220 or the cache controller 215 will read the data from the specified physical address (which can be in the off-die memory 275 or in the cache 210), and assemble a response packet to the source address containing the requested data. Similarly, the memory controller chiplet 205 can receive a packet having a source address, a write request, and a physical address. In response, the memory controller chiplet 205 will write the data to the specified physical address (which can be in the cache 210 or in the off-die memories 275 or 280), and assemble a response packet to the source address containing an acknowledgement that the data was stored to a memory.
Thus, the memory controller chiplet 205 can receive read and write requests via the chiplet network 285 and process the requests using the cache controller 215 interfacing with the cache 210, if possible. If the request cannot be handled by the cache controller 215, the off-die memory controller 220 handles the request by communication with the off-die memories 275 or 280, the atomic and merge unit 250, or both. As noted above, one or more levels of cache can also be implemented in off-die memories 275 or 280; and in some such examples can be accessed directly by cache controller 215. Data read by the off-die memory controller 220 can be cached in the cache 210 by the cache controller 215 for later use.
The atomic and merge unit 250 are coupled to receive (as input) the output of the off-die memory controller 220, and to provide output to the cache 210, the network communication interface 225, or directly to the chiplet network 285. The memory hazard unit 260, write merge unit 255 and the built-in (e.g., predetermined) atomic unit 265 can each be implemented as state machines with other combinational logic circuitry (such as adders, shifters, comparators, AND gates, OR gates, XOR gates, or any suitable combination thereof) or other logic circuitry. These components can also include one or more registers or buffers to store operand or other data. The PAU 270 can be implemented as one or more processor cores or control circuitry, and various state machines with other combinational logic circuitry or other logic circuitry, and can also include one or more registers, buffers, or memories to store addresses, executable instructions, operand and other data, or can be implemented as a processor.
The write merge unit 255 receives read data and request data, and merges the request data and read data to create a single unit having the read data and the source address to be used in the response or return data packet). The write merge unit 255 provides the merged data to the write port of the cache 210 (or, equivalently, to the cache controller 215 to write to the cache 210). Optionally, the write merge unit 255 provides the merged data to the network communication interface 225 to encode and prepare a response or return data packet for transmission on the chiplet network 285.
When the request data is for a built-in atomic operator, the built-in atomic unit 265 receives the request and reads data, either from the write merge unit 255 or directly from the off-die memory controller 220. The atomic operator is performed, and using the write merge unit 255, the resulting data is written to the cache 210, or provided to the network communication interface 225 to encode and prepare a response or return data packet for transmission on the chiplet network 285.
The built-in atomic unit 265 handles predefined atomic operators such as fetch-and-increment or compare-and-swap. In an example, these operations perform a simple read-modify-write operation to a single memory location of 32-bytes or less in size. Atomic memory operations are initiated from a request packet transmitted over the chiplet network 285. The request packet has a physical address, atomic operator type, operand size, and optionally up to 32-bytes of data. The atomic operator performs the read-modify-write to a cache memory line of the cache 210, filling the cache memory if necessary. The atomic operator response can be a simple completion response, or a response with up to 32-bytes of data. Example atomic memory operators include fetch-and-AND, fetch-and-OR, fetch-and-XOR, fetch-and-add, fetch-and-subtract, fetch-and-increment, fetch-anddecrement, fetch-and-minimum, fetch-and-maximum, fetch-and-swap, and compare-and-swap. In various example embodiments, 32-bit and 64-bit operations are supported, along with operations on 16 or 32 bytes of data. Methods disclosed herein are also compatible with hardware supporting larger or smaller operations and more or less data.
Built-in atomic operators can also involve requests for a “standard” atomic operator on the requested data, such as comparatively simple, single cycle, integer atomics—such as fetch-and-increment or compare-and-swap—which will occur with the same throughput as a regular memory read or write operation not involving an atomic operator. For these operations, the cache controller 215 can generally reserve a cache line in the cache 210 by setting a hazard bit (in hardware), so that the cache line cannot be read by another process while it is in transition. The data is obtained from either the off-die memory 275 or the cache 210, and is provided to the built-in atomic unit 265 to perform the requested atomic operator. Following the atomic operator, in addition to providing the resulting data to the packet encoder 240 to encode outgoing data packets for transmission on the chiplet network 285, the built-in atomic unit 265 provides the resulting data to the write merge unit 255, which will also write the resulting data to the cache 210. Following the writing of the resulting data to the cache 210, any corresponding hazard bit which was set will be cleared by the memory hazard unit 260.
The PAU 270 enables high performance (high throughput and low latency) for programmable atomic operators (also referred to as “custom atomic transactions” or “custom atomic operators”), comparable to the performance of built-in atomic operators. Rather than executing multiple memory accesses, in response to an atomic operator request designating a programmable atomic operator and a memory address, circuitry in the memory controller chiplet 205 transfers the atomic operator request to PAU 270 and sets a hazard bit stored in a memory hazard register corresponding to the memory address of the memory line used in the atomic operator, to ensure that no other operation (read, write, or atomic) is performed on that memory line, which hazard bit is then cleared upon completion of the atomic operator. Additional, direct data paths provided for the PAU 270 executing the programmable atomic operators allow for additional write operations without any limitations imposed by the bandwidth of the communication networks and without increasing any congestion of the communication networks.
The PAU 270 includes a multi-threaded processor, for example, such as a RISC-V ISA based multi-threaded processor, having one or more processor cores, and further having an extended instruction set for executing programmable atomic operators. When provided with the extended instruction set for executing programmable atomic operators, the PAU 270 can be embodied as one or more hybrid threading processors. In some example embodiments, the PAU 270 provides barrel-style, round-robin instantaneous thread switching to maintain a high instruction-per-clock rate.
Programmable atomic operators can be performed by the PAU 270 involving requests for a programmable atomic operator on the requested data. A user can prepare programming code to provide such programmable atomic operators. For example, the programmable atomic operators can be comparatively simple, multi-cycle operations such as floating-point addition, or comparatively complex, multi-instruction operations such as a Bloom filter insert. The programmable atomic operators can be the same as or different than the predetermined atomic operators, insofar as they are defined by the user rather than a system vendor. For these operations, the cache controller 215 can reserve a cache line in the cache 210, by setting a hazard bit (in hardware), so that cache line cannot be read by another process while it is in transition. The data is obtained from either the cache 210 or the off-die memories 275 or 280, and is provided to the PAU 270 to perform the requested programmable atomic operator. Following the atomic operator, the PAU 270 will provide the resulting data to the network communication interface 225 to directly encode outgoing data packets having the resulting data for transmission on the chiplet network 285. In addition, the PAU 270 will provide the resulting data to the cache controller 215, which will also write the resulting data to the cache 210. Following the writing of the resulting data to the cache 210, any corresponding hazard bit which was set will be cleared by the cache control circuit 215.
In selected examples, the approach taken for programmable atomic operators is to provide multiple, generic, custom atomic request types that can be sent through the chiplet network 285 to the memory controller chiplet 205 from an originating source such as a processor or other system component. The cache controllers 215 or off-die memory controller 220 identify the request as a custom atomic and forward the request to the PAU 270. In a representative embodiment, the PAU 270: (1) is a programmable processing element capable of efficiently performing a user defined atomic operator; (2) can perform load and stores to memory, arithmetic and logical operations and control flow decisions; and (3) leverages the RISC-V ISA with a set of new, specialized instructions to facilitate interacting with such controllers 215, 220 to atomically perform the user-defined operation. In desirable examples, the RISC-V ISA contains a full set of instructions that support high level language operators and data types. The PAU 270 can leverage the RISC-V ISA, but will commonly support a more limited set of instructions and limited register file size to reduce the die size of the unit when included within the memory controller chiplet 205.
As mentioned above, prior to the writing of the read data to the cache 210, the set hazard bit for the reserved cache line is to be cleared, by the memory hazard clear unit 260. Accordingly, when the request and read data is received by the write merge unit 255, a reset or clear signal can be transmitted by the memory hazard clear unit 260 to the cache 210 to reset the set memory hazard bit for the reserved cache line. Also, resetting this hazard bit will also release a pending read or write request involving the designated (or reserved) cache line, providing the pending read or write request to an inbound request multiplexer for selection and processing.
In an example, the processor 305 is a pipelined such that multiple stages of different instructions are executed together per clock cycle. The processor 305 is also a barrel-multithreaded processor, with circuitry to switch between different register files (e.g., sets of registers containing current processing state) upon each clock cycle of the processor 305. This enables efficient context switching between currently executing threads. In an example, the processor 305 supports eight threads, resulting in eight register files. In an example, some or all of the register files are not integrated into the processor 305, but rather reside in the local memory 310 (registers 320). This reduces circuit complexity in the processor 305 by eliminating the traditional flip-flops used for these registers 320.
The local memory 310 can also house a cache 330 and instructions 325 for atomic operators. The atomic instructions 325 comprise sets of instructions to support the various application-loaded atomic operators. When an atomic operator is requested-by the application chiplet 125, for example-a set of instructions corresponding to the atomic operator are executed by the processor 305. In an example, the atomic instructions 325 are partitioned to establish the sets of instructions. In this example, the specific programmable atomic operator being requested by a requesting process can identify the programmable atomic operator by the partition number. The partition number can be established when the programmable atomic operator is registered with (e.g., loaded onto) the PAU 300. Additional metadata for the programmable atomic instructions 325 can also be stored in the local memory 310, such as the partition tables.
Atomic operators manipulate the cache 330, which is generally synchronized (e.g., flushed) when a thread for an atomic operator completes. Thus, aside from initial loading from the external memory, such as the off-die memory 275 or 280, latency is reduced for most memory operations during execution of a programmable atomic operator thread.
The processor 305 is configured to have a variable length pipeline to address inefficiency issues arising from situations when the processor 305 is lightly loaded, such as when fewer than all available threads are used. To accomplish this, the processor 305 is configured to enable instruction completion at a stage in the pipeline prior to the end of the pipeline. Thus, assuming that the pipeline is eight stages long, but an instruction can finish in two stages (e.g., a register move instruction) or in six stages (e.g., a double precision multiplication), then the instructions can complete at stages two and six respectively. Completion, in this context, indicates that the instruction is ready to writeback to a register. As noted above, and in the discussion with respect to
In an example, to determine the completion time for the instruction, the processor 305 is configured to determine that the completion time is less than all possible pipeline stages in the pipeline. This condition coincides with the fixed-completion-time instruction discussed below with respect to
The processor 305 is configured to detect a conflict between the instruction and a different instruction based on the completion time. Here, the different instruction is in the pipeline and the conflict occurs when the completion time equals a second completion time for the different instruction. Thus, the detection ascertains that, upon a bottle-neck for completing the instruction such as a single cycle for a register writeback, both the current instruction and the different instruction will arrive at the bottle neck at the same moment.
To track conflicts, the processor 305 can include a scoreboard (e.g., a register writeback scoreboard). The scoreboard is a structure in which instruction completion can be marked. Thus, for example, when clock cycles are used to determine completion times, the scoreboard has a field for a given future clock cycle that is marked (e.g., with a logical one) when an instruction in the pipeline (e.g., the different instruction) is expected to complete upon that cycle. Other structures could be used to represent that the time will be in use, such as a shifting structure (e.g., register) encoded such that each bit represents a number of cycles in the future, the structure shifted down by one at every clock cycle.
The processor 305 is configured to detect the conflict by looking up a clock cycle upon which the completion time falls in the scoreboard to get a result. This result is then examined by the processor 305 to detect that the different instruction will perform a register writeback at the clock cycle. Thus, there is a conflict between the instruction and the different instruction.
The processor 305 is configured to calculate a difference between the completion time and an unconflicted completion time. In an example, the processor 305 uses the scoreboard to look forward to future cycles when there is no conflict (e.g., an unconflicted completion time). This difference is a delay in completion, which if enforced, will ensure that the instruction can writeback to the register. If there is no conflict, the difference will be zero. In an example, the processor 305 is configured to write an indication that the instruction will complete at a second clock cycle in the scoreboard as part of inserting the instruction into the pipeline. Here, the second clock cycle corresponds to the unconflicted completion time. Thus, the scoreboard is updated by the processor 305 with the completion time plus the difference to enable conflict detection on future instructions.
The processor 305 is configured to delay completion of the instruction by the difference. Thus, if the instruction could complete at stage two of the pipeline, but conflicts with the different instruction and a difference of one is given to resolve the conflict, the processor 305 will perform the register writeback for the instruction on the clock cycle following the completion of stage two in the pipeline.
In an example, to support the delay, the processor 305 can include a write delay structure (e.g., write delay circuitry or writeback delay component 445 of
As the thread instructions can complete at different times (e.g., clock cycles, pipeline stages, etc.) a flexible technique to reschedule them can include a buffer of thread IDs that are presently in the pipeline (e.g., a pipelined-thread-ID buffer). The processor 305 can be configured to place a thread ID for an instruction into the pipelined-thread-ID buffer as part of inserting the instruction into the pipeline. Then, when the instruction is complete, the processor 305 is configured to move the thread ID from the pipelined-thread-ID buffer to a thread-ready-to-run queue. From here, each thread ID can be extracted, in order, to schedule the next instruction.
The instruction writeback slot component 425 (e.g., instruction writeback circuitry) examines the instruction and determines at what stage in the pipeline 150 the instruction would complete. In this context, instruction completion means a point when the instruction will attempt to write back to the register file 435. For example, a move instruction (e.g., moving data from one register to another) can be quite quick and complete at stage two of the pipeline 450, or in two clock cycles from entering the pipeline. In contrast, a double precision addition instruction can complete at stage four of the pipeline 450 or take four clock cycles. These types of instructions have a known, or fixed, completion within a stage of the pipeline 450, and thus are here called fixed-completion-time instructions. Other instructions, such as requests to off-die memory or to a component outside of the processor (e.g., a co-processor, cryptography component, or other IP block) can have a completion time beyond the pipeline stages, which can also be variable, and thus are called variable-completion-time instructions.
If the instruction writeback slot 425 determines that the instruction is a variable-completion-time instruction, a form of thread self-scheduling can be used. Here, the processor will hand over the timing of the re-scheduling of the thread, to a degree, to the external or long-running component. To accomplish this, the thread ID travels with the instruction. Thus, for example, as a memory request instruction enters the pipeline 450 to process the memory request, the thread ID for the instruction accompanies the memory request. At this point, the thread ID is no longer in the thread-ready-to-run queue 405, nor in any other illustrated structure. Accordingly, the thread will not be rescheduled. However, when the memory request is complete, the response from memory is received by the processor and stored in buffer 465. A separate buffer 470 can be used for another IP block (e.g., coprocessor) response, of the buffer 465 can be used for all variable-completion-time instruction responses. The multiplexer 460 determines when the threads corresponding to the responses in the buffers 465 or 470 will be placed back into the thread ready-to-run queue for rescheduling.
If the instruction is a fixed-completion-time instruction, the instruction writeback slot component 425 checks the register writeback scoreboard 430 to determine whether any other register writeback could be attempted at the same clock cycle as the current instruction. For example, if, two clock cycles ago, an instruction can writeback in four clock cycles, and the current instruction can writeback in two clock cycles, then both instructions will be able to writeback on the same clock cycle. This potential collision is tracked in the register writeback scoreboard 430. The register writeback scoreboard 430 tracks which clock cycles do not have a potential writeback. Thus, when the 4-cycle instruction was processed by the instruction writeback slot component 425, the fourth clock cycle was marked to indicate a writeback by the 4-cycle component. Then, two cycles later, when the 2-cycle instruction is processed, the clock cycle when the 2-cycle instruction could finish is marked by the 4-cycle instruction, indicating a register writeback conflict.
When the instruction writeback slot component 425 detects a register writeback conflict, the instruction writeback slot component 425 resolves the conflict by computing a delay for the conflicting instruction based on unconflicted cycles from the register writeback scoreboard 430. Thus, if the cycle following the conflicted cycle is free, then the 2-cycle instruction can be given a 1-cycle writeback delay. If there is no conflict, then the instruction is given a 0-cycle writeback delay.
Once the register writeback delay is determined by the instruction writeback slot component 425, the instruction writeback slot component 425 writes the register writeback delay to the write delay component 445. An example implementation of the writeback delay component 445 can be a set of shift-registers, one register for each stage in the pipeline 450, or for a subset of stages in the pipeline 450 that could prompt a register writeback. As illustrated, the arrows leading from the pipeline 450 and the writeback delay component correspond, such that the topmost arrow of each corresponds to a writeback of stage two of the pipeline 450, the second topmost arrow to stage four, and the last arrow to stage six.
Upon each clock cycle, the shift registers of the writeback delay component 445 decrement the values stored therein. Thus, if the 2-cycle instruction had a three cycle delay, a numeric three is stored in the shift-register corresponding to the second stage of the pipeline 450. After the first cycle, the value in the register would be a two, then a one upon another cycle, and finally a zero upon the third cycle.
When the instructions are ready to writeback from the pipeline 450, the stage at which the writeback specifies which register in the writeback delay component 445 is consulted to determine whether the writeback will occur. For example, the 4-cycle instruction has a zero in the writeback delay component 445 when the fourth stage of the pipeline 450 completes, and so can writeback to the register file 435 immediately. However, the register in the writeback delay component 445 that corresponds to the 2-cycle instruction does not have a zero, and so the write back will not occur for this instruction. Rather, the thread for the 2-cycle instruction will be held at stage two of the pipeline 450. At the next cycle, the writeback delay component 445 register for stage two will decrement to zero and the writeback to the register file 435 of the 2-cycle instruction can proceed.
Instructions with later completion times (e.g., later pipeline stages) have priority over. Thus, on a given clock cycle, if an instruction at stage six and stage four are prepared to writeback to the register file 435, then the stage six instruction will perform the writeback. This ensures that the maximum writeback delay does not put the writeback beyond the pipeline length, which is the typical delay imposed upon other barrel-multithreaded processors. When the delay for an instruction is short than the pipeline length, then fewer clock cycles are wasted in a lightly loaded processor because shorter-cycle instructions can complete in fewer cycles than it takes to complete every stage in the pipeline 450.
Once the writeback to the register file 435 is complete, the thread can be rescheduled. This is accomplished by the pipelined thread ID component 440. The pipelined thread ID component 440 contains thread IDs of instructions moving through the pipeline 450. When an instruction is able to perform a writeback, the thread ID of that instruction is removed from the pipelined thread ID component 440 and delivered to the multiplexer 460 for reinsertion into the thread ready-to-run queue 405. In an example, the multiplexer selects an available thread ID from the pipelined thread ID component 440 before any other source (e.g., a memory response). Thus, the buffers 465 or 470 maintains the thread IDs from the variable-completion-time instructions until there is no input from the pipelined thread ID component 440.
In an example, the method is performed by a processor (e.g., the processor 305) in a PAU (e.g., PAU 300 or PAU 270) in a memory controller (e.g., memory controller 140 or memory controller 205). In an example, the memory controller is a chiplet (e.g., memory controller 140). In an example, the memory controller chiplet is integrated into a chiplet system (e.g., chiplet system 110).
At operation 505, a completion time for the instruction is determined prior to insertion in a pipeline of the barrel-multithreaded processor. In an example, the completion time is a number of clock cycles. In an example, determining the completion time for the instruction includes determining that the completion time is less than all possible pipeline stages in the pipeline.
At operation 510, a conflict between the instruction and a different instruction is detected based on the completion time. Here, the different instruction is in the pipeline and the conflict occurs when the completion time equals a second completion time for the different instruction.
In an example, detecting the conflict includes looking up a clock cycle upon which the completion time falls in a register writeback scoreboard to get a result. Then, the result is examined to detect that the result indicates the different instruction will perform a register writeback at the clock cycle. In an example, the operations of the method 500 are expanded to include writing an indication that the instruction will complete at a second clock cycle in the register writeback scoreboard as part of inserting the instruction into the pipeline. Here, the second clock cycle corresponds to the unconflicted completion time.
At operation 515, a difference between the completion time and an unconflicted completion time is calculated. In an example, is a second number of clock cycles when the completion time is measured in clock cycles. In an example, the maximum delay (e.g., completion time plus the difference) is eight clock cycles.
At operation 520, completion of the instruction is delayed by the difference. In an example, delaying completion of the instruction by the difference includes detecting that the instruction is ready to perform a register writeback at a stage in the pipeline and performing a read in a write delay data structure for a record that corresponds to the stage in the pipeline. If a determination that a value of the record is not zero, then the instruction is prevented from performing the register writeback. In an example, the operations of the method 500 can be expanded to include writing the difference into the record as the value as part of inserting the instruction into the pipeline. In an example, the operations of the method 500 can be expanded to include decrementing the value with each clock cycle.
In an example, the operations of the method 500 can be expanded to include moving a thread ID for the instruction from a pipelined-thread-ID buffer into a thread-ready-to-run queue in response to completion of the instruction. In an example, the thread ID is inserted into the pipelined-thread-ID buffer as part of inserting the instruction into the pipeline.
In alternative embodiments, the machine 600 can operate as a standalone device or can be connected (e.g., networked) to other machines. In a networked deployment, the machine 600 can operate in the capacity of a server machine, a client machine, or both in server-client network environments. In an example, the machine 600 can act as a peer machine in peer-to-peer (P2P) (or other distributed) network environment. The machine 600 can be a personal computer (PC), a tablet PC, a set-top box (STB), a personal digital assistant (PDA), a mobile telephone, a web appliance, a network router, switch or bridge, or any machine capable of executing instructions (sequential or otherwise) that specify actions to be taken by that machine. Further, while only a single machine is illustrated, the term “machine” shall also be taken to include any collection of machines that individually or jointly execute a set (or multiple sets) of instructions to perform any one or more of the methodologies discussed herein, such as cloud computing, software as a service (SaaS), other computer cluster configurations.
The machine (e.g., computer system) 600 can include a hardware processor 602 (e.g., a central processing unit (CPU), a graphics processing unit (GPU), a hardware processor core, or any combination thereof), a main memory 604, a static memory (e.g., memory or storage for firmware, microcode, a basicinput-output (BIOS), unified extensible firmware interface (UEFI), etc.) 606, and mass storage 608 (e.g., hard drives, tape drives, flash storage, or other block devices) some or all of which can communicate with each other via an interlink (e.g., bus) 630. The machine 600 can further include a display unit 610, an alphanumeric input device 612 (e.g., a keyboard), and a user interface (UI) navigation device 614 (e.g., a mouse). In an example, the display unit 610, input device 612 and UI navigation device 614 can be a touch screen display. The machine 600 can additionally include a storage device (e.g., drive unit) 608, a signal generation device 618 (e.g., a speaker), a network interface device 620, and one or more sensors 616, such as a global positioning system (GPS) sensor, compass, accelerometer, or other sensor. The machine 600 can include an output controller 628, such as a serial (e.g., universal serial bus (USB), parallel, or other wired or wireless (e.g., infrared (IR), near field communication (NFC), etc.) connection to communicate or control one or more peripheral devices (e.g., a printer, card reader, etc.).
Registers of the processor 602, the main memory 604, the static memory 606, or the mass storage 608 can be, or include, a machine readable medium 622 on which is stored one or more sets of data structures or instructions 624 (e.g., software) embodying or utilized by any one or more of the techniques or functions described herein. The instructions 624 can also reside, completely or at least partially, within any of registers of the processor 602, the main memory 604, the static memory 606, or the mass storage 608 during execution thereof by the machine 600. In an example, one or any combination of the hardware processor 602, the main memory 604, the static memory 606, or the mass storage 608 can constitute the machine readable media 622. While the machine readable medium 622 is illustrated as a single medium, the term “machine readable medium” can include a single medium or multiple media (e.g., a centralized or distributed database, or associated caches and servers) configured to store the one or more instructions 624.
The term “machine readable medium” can include any medium that is capable of storing, encoding, or carrying instructions for execution by the machine 600 and that cause the machine 600 to perform any one or more of the techniques of the present disclosure, or that is capable of storing, encoding or carrying data structures used by or associated with such instructions. Non-limiting machine-readable medium examples can include solid-state memories, optical media, magnetic media, and signals (e.g., radio frequency signals, other photon-based signals, sound signals, etc.). In an example, a non-transitory machine-readable medium comprises a machine-readable medium with a plurality of particles having invariant (e.g., rest) mass, and thus are compositions of matter. Accordingly, non-transitory machine-readable media are machine readable media that do not include transitory propagating signals. Specific examples of non-transitory machine readable media can include: non-volatile memory, such as semiconductor memory devices (e.g., electrically programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM)) and flash memory devices; magnetic disks, such as internal hard disks and removable disks; magneto-optical disks; and CD-ROM and DVD-ROM disks.
In an example, information stored or otherwise provided on the machine readable medium 622 can be representative of the instructions 624, such as instructions 624 themselves or a format from which the instructions 624 can be derived. This format from which the instructions 624 can be derived can include source code, encoded instructions (e.g., in compressed or encrypted form), packaged instructions (e.g., split into multiple packages), or the like. The information representative of the instructions 624 in the machine readable medium 622 can be processed by processing circuitry into the instructions to implement any of the operations discussed herein. For example, deriving the instructions 624 from the information (e.g., processing by the processing circuitry) can include: compiling (e.g., from source code, object code, etc.), interpreting, loading, organizing (e.g., dynamically or statically linking), encoding, decoding, encrypting, unencrypting, packaging, unpackaging, or otherwise manipulating the information into the instructions 624.
In an example, the derivation of the instructions 624 can include assembly, compilation, or interpretation of the information (e.g., by the processing circuitry) to create the instructions 624 from some intermediate or preprocessed format provided by the machine readable medium 622. The information, when provided in multiple parts, can be combined, unpacked, and modified to create the instructions 624. For example, the information can be in multiple compressed source code packages (or object code, or binary executable code, etc.) on one or several remote servers. The source code packages can be encrypted when in transit over a network and decrypted, uncompressed, assembled (e.g., linked) if necessary, and compiled or interpreted (e.g., into a library, stand-alone executable etc.) at a local machine, and executed by the local machine.
The instructions 624 can be further transmitted or received over a communications network 626 using a transmission medium via the network interface device 620 utilizing any one of a number of transfer protocols (e.g., frame relay, internet protocol (IP), transmission control protocol (TCP), user datagram protocol (UDP), hypertext transfer protocol (HTTP), etc.). Example communication networks can include a local area network (LAN), a wide area network (WAN), a packet data network (e.g., the Internet), mobile telephone networks (e.g., cellular networks), plain old telephone (POTS) networks, and wireless data networks (e.g., Institute of Electrical and Electronics Engineers (IEEE) 802.11 family of standards known as Wi-Fi®, IEEE 802.16 family of standards known as WiMax®), IEEE 802.15.4 family of standards, peer-to-peer (P2P) networks, among others. In an example, the network interface device 620 can include one or more physical jacks (e.g., Ethernet, coaxial, or phone jacks) or one or more antennas to connect to the communications network 626. In an example, the network interface device 620 can include a plurality of antennas to wirelessly communicate using at least one of single-input multiple-output (SIMO), multiple-input multiple-output (MIMO), or multiple-input single-output (MISO) techniques. The term “transmission medium” shall be taken to include any intangible medium that is capable of storing, encoding or carrying instructions for execution by the machine 600, and includes digital or analog communications signals or other intangible medium to facilitate communication of such software. A transmission medium is a machine readable medium. To better illustrate the methods and apparatuses described herein, a non-limiting set of Example embodiments are set forth below as numerically identified Examples.
Example 1 is a processor comprising: a pipeline; an instruction writeback circuitry configured to: determine a completion time for an instruction prior to insertion into the pipeline; detect a conflict between the instruction and a different instruction based on the completion time, the different instruction being in the pipeline, the conflict detected when the completion time equals a second completion time for the different instruction; and calculating a difference between the completion time and an unconflicted completion time; and write delay circuitry configured to delay completion of the instruction by the difference.
In Example 2, the subject matter of Example 1, wherein the completion time is a number of clock cycles.
In Example 3, the subject matter of Example 2, wherein the difference is a second number of clock cycles.
In Example 4, the subject matter of Example 3, wherein, to determine the completion time for the instruction, the instruction writeback circuitry is configured to determine that the completion time is less than all possible pipeline stages in the pipeline.
In Example 5, the subject matter of Example 4, wherein the maximum delay is eight clock cycles.
In Example 6, the subject matter of any of Examples 1-5, wherein, to detect the conflict, the instruction writeback circuitry is configured to: look up a clock cycle upon which the completion time falls in a register writeback scoreboard to get a result; and detect that the result indicates the different instruction will perform a register writeback at the clock cycle.
In Example 7, the subject matter of Example 6, wherein the instruction writeback circuitry is configured to: write an indication that the instruction will complete at a second clock cycle in the register writeback scoreboard as part of inserting the instruction into the pipeline, the second clock cycle corresponding to the unconflicted completion time.
In Example 8, the subject matter of any of Examples 1-7, wherein, to delay completion of the instruction by the difference, the pipeline is configured to: detect that the instruction is ready to perform a register writeback at a stage in the pipeline; perform a read in a write delay data structure of the write delay circuitry for a record that corresponds to the stage in the pipeline; determine that a value of the record is not zero; and prevent the instruction from performing the register writeback.
In Example 9, the subject matter of Example 8, wherein the instruction writeback circuitry is configured to: write the difference into the record as the value as part of inserting the instruction into the pipeline.
In Example 10, the subject matter of any of Examples 8-9, wherein the write delay circuitry is configured to: decrement the value with each clock cycle.
In Example 11, the subject matter of any of Examples 1-10, wherein the pipeline is configured to: move a thread identifier (ID) for the instruction from a pipelined-thread-ID buffer into a thread-ready-to-run queue in response to completion of the instruction, the thread ID being inserted into the pipelined-thread-ID buffer as part of inserting the instruction into the pipeline.
In Example 12, the subject matter of any of Examples 1-11, wherein the processor is included in a programmable atomic unit in a memory controller.
In Example 13, the subject matter of Example 12, wherein the memory controller is a chiplet in a chiplet system.
Example 14 is a method comprising: determining a completion time for an instruction prior to insertion into a pipeline of a processor; detecting a conflict between the instruction and a different instruction based on the completion time, the different instruction being in the pipeline, the conflict detected when the completion time equals a second completion time for the different instruction; calculating a difference between the completion time and an unconflicted completion time; and delaying completion of the instruction by the difference.
In Example 15, the subject matter of Example 14, wherein the completion time is a number of clock cycles.
In Example 16, the subject matter of Example 15, wherein the difference is a second number of clock cycles.
In Example 17, the subject matter of Example 16, wherein determining the completion time for the instruction includes determining that the completion time is less than all possible pipeline stages in the pipeline.
In Example 18, the subject matter of Example 17, wherein the maximum delay is eight clock cycles.
In Example 19, the subject matter of any of Examples 14-18, wherein detecting the conflict includes: looking up a clock cycle upon which the completion time falls in a register writeback scoreboard to get a result; and detecting that the result indicates the different instruction will perform a register writeback at the clock cycle.
In Example 20, the subject matter of Example 19, comprising: writing an indication that the instruction will complete at a second clock cycle in the register writeback scoreboard as part of inserting the instruction into the pipeline, the second clock cycle corresponding to the unconflicted completion time.
In Example 21, the subject matter of any of Examples 14-20, wherein delaying completion of the instruction by the difference includes: detecting that the instruction is ready to perform a register writeback at a stage in the pipeline; performing a read in a write delay data structure for a record that corresponds to the stage in the pipeline; determining that a value of the record is not zero; and preventing the instruction from performing the register writeback.
In Example 22, the subject matter of Example 21, comprising: writing the difference into the record as the value as part of inserting the instruction into the pipeline.
In Example 23, the subject matter of any of Examples 21-22, comprising: decrementing the value with each clock cycle.
In Example 24, the subject matter of any of Examples 14-23, comprising: moving a thread identifier (ID) for the instruction from a pipelined-thread-ID buffer into a thread-ready-to-run queue in response to completion of the instruction, the thread ID being inserted into the pipelined-thread-ID buffer as part of inserting the instruction into the pipeline.
In Example 25, the subject matter of any of Examples 14-24, wherein the processor is included in a programmable atomic unit in a memory controller.
In Example 26, the subject matter of Example 25, wherein the memory controller is a chiplet in a chiplet system.
Example 27 is a machine-readable medium including instructions that, when executed by circuitry of a processor, cause the processor to perform operations comprising: determining a completion time for an instruction prior to insertion into a pipeline of the processor; detecting a conflict between the instruction and a different instruction based on the completion time, the different instruction being in the pipeline, the conflict detected when the completion time equals a second completion time for the different instruction; calculating a difference between the completion time and an unconflicted completion time; and delaying completion of the instruction by the difference.
In Example 28, the subject matter of Example 27, wherein the completion time is a number of clock cycles.
In Example 29, the subject matter of Example 28, wherein the difference is a second number of clock cycles.
In Example 30, the subject matter of Example 29, wherein determining the completion time for the instruction includes determining that the completion time is less than all possible pipeline stages in the pipeline.
In Example 31, the subject matter of Example 30, wherein the maximum delay is eight clock cycles.
In Example 32, the subject matter of any of Examples 27-31, wherein detecting the conflict includes: looking up a clock cycle upon which the completion time falls in a register writeback scoreboard to get a result; and detecting that the result indicates the different instruction will perform a register writeback at the clock cycle.
In Example 33, the subject matter of Example 32, wherein the operations comprise: writing an indication that the instruction will complete at a second clock cycle in the register writeback scoreboard as part of inserting the instruction into the pipeline, the second clock cycle corresponding to the unconflicted completion time.
In Example 34, the subject matter of any of Examples 27-33, wherein delaying completion of the instruction by the difference includes: detecting that the instruction is ready to perform a register writeback at a stage in the pipeline; performing a read in a write delay data structure for a record that corresponds to the stage in the pipeline; determining that a value of the record is not zero; and preventing the instruction from performing the register writeback.
In Example 35, the subject matter of Example 34, wherein the operations comprise: writing the difference into the record as the value as part of inserting the instruction into the pipeline.
In Example 36, the subject matter of any of Examples 34-35, wherein the operations comprise: decrementing the value with each clock cycle.
In Example 37, the subject matter of any of Examples 27-36, wherein the operations comprise: moving a thread identifier (ID) for the instruction from a pipelined-thread-ID buffer into a thread-ready-to-run queue in response to completion of the instruction, the thread ID being inserted into the pipelined-thread-ID buffer as part of inserting the instruction into the pipeline.
In Example 38, the subject matter of any of Examples 27-37, wherein the processor is included in a programmable atomic unit in a memory controller.
In Example 39, the subject matter of Example 38, wherein the memory controller is a chiplet in a chiplet system.
Example 40 is a system comprising: means for determining a completion time for an instruction prior to insertion into a pipeline of a processor; means for detecting a conflict between the instruction and a different instruction based on the completion time, the different instruction being in the pipeline, the conflict detected when the completion time equals a second completion time for the different instruction; means for calculating a difference between the completion time and an unconflicted completion time; and means for delaying completion of the instruction by the difference.
In Example 41, the subject matter of Example 40, wherein the completion time is a number of clock cycles.
In Example 42, the subject matter of Example 41, wherein the difference is a second number of clock cycles.
In Example 43, the subject matter of Example 42, wherein the means for determining the completion time for the instruction include means for determining that the completion time is less than all possible pipeline stages in the pipeline.
In Example 44, the subject matter of Example 43, wherein the maximum delay is eight clock cycles.
In Example 45, the subject matter of any of Examples 40-44, wherein the means for detecting the conflict include: means for looking up a clock cycle upon which the completion time falls in a register writeback scoreboard to get a result; and means for detecting that the result indicates the different instruction will perform a register writeback at the clock cycle.
In Example 46, the subject matter of Example 45, comprising: means for writing an indication that the instruction will complete at a second clock cycle in the register writeback scoreboard as part of inserting the instruction into the pipeline, the second clock cycle corresponding to the unconflicted completion time.
In Example 47, the subject matter of any of Examples 40-46, wherein the means for delaying completion of the instruction by the difference include: means for detecting that the instruction is ready to perform a register writeback at a stage in the pipeline; means for performing a read in a write delay data structure for a record that corresponds to the stage in the pipeline; means for determining that a value of the record is not zero; and means for preventing the instruction from performing the register writeback.
In Example 48, the subject matter of Example 47, comprising: means for writing the difference into the record as the value as part of inserting the instruction into the pipeline.
In Example 49, the subject matter of any of Examples 47-48, comprising: means for decrementing the value with each clock cycle.
In Example 50, the subject matter of any of Examples 40-49, comprising: means for moving a thread identifier (ID) for the instruction from a pipelined-thread-ID buffer into a thread-ready-to-run queue in response to completion of the instruction, the thread ID being inserted into the pipelined-thread-ID buffer as part of inserting the instruction into the pipeline.
In Example 51, the subject matter of any of Examples 40-50, wherein the processor is included in a programmable atomic unit in a memory controller.
In Example 52, the subject matter of Example 51, wherein the memory controller is a chiplet in a chiplet system.
Example 53 is at least one machine-readable medium including instructions that, when executed by processing circuitry, cause the processing circuitry to perform operations to implement of any of Examples 1-52.
Example 54 is an apparatus comprising means to implement of any of Examples 1-52.
Example 55 is a system to implement of any of Examples 1-52.
Example 56 is a method to implement of any of Examples 1-52.
The above detailed description includes references to the accompanying drawings, which form a part of the detailed description. The drawings show, by way of illustration, specific embodiments in which the invention can be practiced. These embodiments are also referred to herein as “examples”. Such examples can include elements in addition to those shown or described. However, the present inventors also contemplate examples in which only those elements shown or described are provided. Moreover, the present inventors also contemplate examples using any combination or permutation of those elements shown or described (or one or more aspects thereof), either with respect to a particular example (or one or more aspects thereof), or with respect to other examples (or one or more aspects thereof) shown or described herein.
In this document, the terms “a” or “an” are used, as is common in patent documents, to include one or more than one, independent of any other instances or usages of “at least one” or “one or more.” In this document, the term “or” is used to refer to a nonexclusive or, such that “A or B” can include “A but not B,” “B but not A,” and “A and B,” unless otherwise indicated. In the appended claims, the terms “including” and “in which” are used as the plain-English equivalents of the respective terms “comprising” and “wherein”. Also, in the following claims, the terms “including” and “comprising” are open-ended, that is, a system, device, article, or process that includes elements in addition to those listed after such a term in a claim are still deemed to fall within the scope of that claim. Moreover, in the following claims, the terms “first,” “second,” and “third,” etc. are used merely as labels, and are not intended to impose numerical requirements on their objects.
The above description is intended to be illustrative, and not restrictive. For example, the above-described examples (or one or more aspects thereof) can be used in combination with each other. Other embodiments can be used, such as by one of ordinary skill in the art upon reviewing the above description. It is submitted with the understanding that it will not be used to interpret or limit the scope or meaning of the claims. Also, in the above Detailed Description, various features can be grouped together to streamline the disclosure. This should not be interpreted as intending that an unclaimed disclosed feature is essential to any claim. Rather, inventive subject matter can lie in less than all features of a particular disclosed embodiment. Thus, the following claims are hereby incorporated into the Detailed Description, with each claim standing on its own as a separate embodiment, and it is contemplated that such embodiments can be combined with each other in various combinations or permutations. The scope of the invention should be determined with reference to the appended claims, along with the full scope of equivalents to which such claims are entitled.
This application is a continuation of U.S. Application Serial No. 17/074,722, filed Oct. 20, 2020, which is incorporated herein by reference in its entirety.
| Number | Date | Country | |
|---|---|---|---|
| Parent | 17074722 | Oct 2020 | US |
| Child | 18079164 | US |
