Patent Application
20160043706
1. Field of the Invention
Embodiments of the present invention relate generally to computer systems and, more specifically, to a low power flip-flop element with gated clock.
2. Description of the Related Art
In computer systems, there is widespread utilization of flip-flop elements. A flip-flop element, also known as a bi-stable multivibrator, is a circuit that has two stable states and may be used to store data. A periodic signal, known as a clock, is applied to one input of the flip-flop element in order to transfer the state of a second, data input signal to the output on the rising edge of the clock.
A typical flip-flop element is configured in a master-slave arrangement. An input latch, know as the master latch, couples the input data to a second latch, known as the slave latch, on the rising edge of the clock signal. The slave latch couples the data from the master latch to the output on the falling edge of the clock signal. The master latch and slave latch each generate multiple clock inversions to implement the data capture functionality.
A number of P-channel and N-channel metal-oxide-semiconductor (MOS) field effect transistors (FETs) are configured in a circuit arrangement to construct the flip-flop element. Within the circuit arrangement, some of the FETs are coupled in pairs, consisting of one PFET and one NFET, to form inverters. Inverters consume power during the switching intervals, and power consumption increases as the frequency of operation increases. Further, all FETs dissipate power due to charging and discharging of junction capacitances whenever a varying signal is applied.
During operation, in a conventional flip-flop element, clocking action is applied to a number of the FET junctions and inverters within the flip-flop element on each clock cycle. For example, if the data input is high, the flip-flop element clocks the high input level to the output. If, on the next cycle, the data input is low, the flip-flop element will clock the low input level to the output. Alternatively, if the data input has remained high, the flip-flop element will clock the high level input level to the output on the next cycle even though the output already had been set high on the prior cycle. Thus, all clocking action results in numerous junction transitions regardless of whether the input changes.
Each clocking action in each individual FET causes dissipation, and the sum of these dissipations equals the dissipation of each individual flip-flop element. The sum of the dissipations of all flip-flop elements represents a portion of the system usage that may exceed twenty percent of the total power consumption.
One drawback to the above approach is that clocking of flip-flop elements incurs power dissipation on each clock cycle regardless of the state of the input data. In a computer system, over a long period, the number of ones and zeroes may be expected to be generally equivalent. Computer data carries information, but a simple alternating sequence of ones and zeroes, such as a clock signal, carries no information. Consequently, consecutive bit sequences during which the logic state remains unchanged must occur in the data signal. In such sequences, it is necessary to capture the first bit only, as the output is unchanged for subsequent bits. Thus, the power consumption due to internal clocking of flip-flop elements is generally avoidable and represents excessive usage.
As the foregoing illustrates, what is needed in the art is a more effective technique for reducing the power dissipated by a flip-flop element.
One embodiment of the present invention sets forth a circuit element configured to perform a data capture operation, including a first latch element configured to receive a first data signal that has a first logic state, invert the first logic state to generate a first inverted logic state, receive a first clock signal, and invert the first clock signal to generate a first inverted clock signal. The circuit element further includes a first logic element coupled to the first latch element and configured to gate the first inverted clock signal with the first inverted logic state to generate a second clock signal, and output the second clock signal to the first latch element, where the first latch element, in response to the second clock signal, inverts the first data signal to generate a first inverted data signal.
One advantage of the disclosed approach is that gating the clock inversion according to the state of the input data substantially reduces the number of transitions that the constituent FETs and inverters undergo, which results in less overall power consumption. Further, combining functionality of constituent FETs reduces the number of constituent FETs and, hence, reduces the number of overall transitions in the circuit, which further reduces power consumption.
So that the manner in which the above recited features of the present invention can be understood in detail, a more particular description of the invention, briefly summarized above, may be had by reference to embodiments, some of which are illustrated in the appended drawings. It is to be noted, however, that the appended drawings illustrate only typical embodiments of this invention and are therefore not to be considered limiting of its scope, for the invention may admit to other equally effective embodiments.
In the following description, numerous specific details are set forth to provide a more thorough understanding of the present invention. However, it will be apparent to one of skill in the art that the present invention may be practiced without one or more of these specific details.
In operation, I/O bridge 107 is configured to receive user input information from input devices 108, such as a keyboard or a mouse, and forward the input information to CPU 102 for processing via communication path 106 and memory bridge 105. Switch 116 is configured to provide connections between I/O bridge 107 and other components of the computer system 100, such as a network adapter 118 and various add-in cards 120 and 121.
As also shown, I/O bridge 107 is coupled to a system disk 114 that may be configured to store content and applications and data for use by CPU 102 and parallel processing subsystem 112. As a general matter, system disk 114 provides non-volatile storage for applications and data and may include fixed or removable hard disk drives, flash memory devices, and CD-ROM (compact disc read-only-memory), DVD-ROM (digital versatile disc-ROM), Blu-ray, HD-DVD (high definition DVD), or other magnetic, optical, or solid state storage devices. Finally, although not explicitly shown, other components, such as universal serial bus or other port connections, compact disc drives, digital versatile disc drives, film recording devices, and the like, may be connected to I/O bridge 107 as well.
In various embodiments, memory bridge 105 may be a Northbridge chip, and I/O bridge 107 may be a Southbridge chip. In addition, communication paths 106 and 113, as well as other communication paths within computer system 100, may be implemented using any technically suitable protocols, including, without limitation, AGP (Accelerated Graphics Port), HyperTransport, or any other bus or point-to-point communication protocol known in the art.
In some embodiments, parallel processing subsystem 112 comprises a graphics subsystem that delivers pixels to a display device 110 that may be any conventional cathode ray tube, liquid crystal display, light-emitting diode display, or the like. In such embodiments, the parallel processing subsystem 112 incorporates circuitry optimized for graphics and video processing, including, for example, video output circuitry. As described in greater detail below in
In various embodiments, parallel processing subsystem 112 may be integrated with one or more of the other elements of
It will be appreciated that the system shown herein is illustrative and that variations and modifications are possible. The connection topology, including the number and arrangement of bridges, the number of CPUs 102, and the number of parallel processing subsystems 112, may be modified as desired. For example, in some embodiments, system memory 104 could be connected to CPU 102 directly rather than through memory bridge 105, and other devices would communicate with system memory 104 via memory bridge 105 and CPU 102. In other alternative topologies, parallel processing subsystem 112 may be connected to I/O bridge 107 or directly to CPU 102, rather than to memory bridge 105. In still other embodiments, I/O bridge 107 and memory bridge 105 may be integrated into a single chip instead of existing as one or more discrete devices. Lastly, in certain embodiments, one or more components shown in
In some embodiments, PPU 202 comprises a graphics processing unit (GPU) that may be configured to implement a graphics rendering pipeline to perform various operations related to generating pixel data based on graphics data supplied by CPU 102 and/or system memory 104. When processing graphics data, PP memory 204 can be used as graphics memory that stores one or more conventional frame buffers and, if needed, one or more other render targets as well. Among other things, PP memory 204 may be used to store and update pixel data and deliver final pixel data or display frames to display device 110 for display. In some embodiments, PPU 202 also may be configured for general-purpose processing and compute operations.
In operation, CPU 102 is the master processor of computer system 100, controlling and coordinating operations of other system components. In particular, CPU 102 issues commands that control the operation of PPU 202. In some embodiments, CPU 102 writes a stream of commands for PPU 202 to a data structure (not explicitly shown in either
As also shown, PPU 202 includes an I/O (input/output) unit 205 that communicates with the rest of computer system 100 via the communication path 113 and memory bridge 105. I/O unit 205 generates packets (or other signals) for transmission on communication path 113 and also receives all incoming packets (or other signals) from communication path 113, directing the incoming packets to appropriate components of PPU 202. For example, commands related to processing tasks may be directed to a host interface 206, while commands related to memory operations (e.g., reading from or writing to PP memory 204) may be directed to a crossbar unit 210. Host interface 206 reads each pushbuffer and transmits the command stream stored in the pushbuffer to a front end 212.
As mentioned above in conjunction with
In operation, front end 212 transmits processing tasks received from host interface 206 to a work distribution unit (not shown) within task/work unit 207. The work distribution unit receives pointers to processing tasks that are encoded as task metadata (TMD) and stored in memory. The pointers to TMDs are included in a command stream that is stored as a pushbuffer and received by the front end unit 212 from the host interface 206. Processing tasks that may be encoded as TMDs include indices associated with the data to be processed as well as state parameters and commands that define how the data is to be processed. For example, the state parameters and commands could define the program to be executed on the data. The task/work unit 207 receives tasks from the front end 212 and ensures that GPCs 208 are configured to a valid state before the processing task specified by each one of the TMDs is initiated. A priority may be specified for each TMD that is used to schedule the execution of the processing task. Processing tasks also may be received from the processing cluster array 230. Optionally, the TMD may include a parameter that controls whether the TMD is added to the head or the tail of a list of processing tasks (or to a list of pointers to the processing tasks), thereby providing another level of control over execution priority.
PPU 202 advantageously implements a highly parallel processing architecture based on a processing cluster array 230 that includes a set of C general processing clusters (GPCs) 208, where C≧1. Each GPC 208 is capable of executing a large number (e.g., hundreds or thousands) of threads concurrently, where each thread is an instance of a program. In various applications, different GPCs 208 may be allocated for processing different types of programs or for performing different types of computations. The allocation of GPCs 208 may vary depending on the workload arising for each type of program or computation.
Memory interface 214 includes a set of D of partition units 215, where D≧1. Each partition unit 215 is coupled to one or more dynamic random access memories (DRAMs) 220 residing within PPM memory 204. In one embodiment, the number of partition units 215 equals the number of DRAMs 220, and each partition unit 215 is coupled to a different DRAM 220. In other embodiments, the number of partition units 215 may be different than the number of DRAMs 220. Persons of ordinary skill in the art will appreciate that a DRAM 220 may be replaced with any other technically suitable storage device. In operation, various render targets, such as texture maps and frame buffers, may be stored across DRAMs 220, allowing partition units 215 to write portions of each render target in parallel to efficiently use the available bandwidth of PP memory 204.
A given GPC 208 may process data to be written to any of the DRAMs 220 within PP memory 204. Crossbar unit 210 is configured to route the output of each GPC 208 to the input of any partition unit 215 or to any other GPC 208 for further processing. GPCs 208 communicate with memory interface 214 via crossbar unit 210 to read from or write to various DRAMs 220. In one embodiment, crossbar unit 210 has a connection to I/O unit 205, in addition to a connection to PP memory 204 via memory interface 214, thereby enabling the processing cores within the different GPCs 208 to communicate with system memory 104 or other memory not local to PPU 202. In the embodiment of
Again, GPCs 208 can be programmed to execute processing tasks relating to a wide variety of applications, including, without limitation, linear and nonlinear data transforms, filtering of video and/or audio data, modeling operations (e.g., applying laws of physics to determine position, velocity and other attributes of objects), image rendering operations (e.g., tessellation shader, vertex shader, geometry shader, and/or pixel/fragment shader programs), general compute operations, etc. In operation, PPU 202 is configured to transfer data from system memory 104 and/or PP memory 204 to one or more on-chip memory units, process the data, and write result data back to system memory 104 and/or PP memory 204. The result data may then be accessed by other system components, including CPU 102, another PPU 202 within parallel processing subsystem 112, or another parallel processing subsystem 112 within computer system 100.
As noted above, any number of PPUs 202 may be included in a parallel processing subsystem 112. For example, multiple PPUs 202 may be provided on a single add-in card, or multiple add-in cards may be connected to communication path 113, or one or more of PPUs 202 may be integrated into a bridge chip. PPUs 202 in a multi-PPU system may be identical to or different from one another. For example, different PPUs 202 might have different numbers of processing cores and/or different amounts of PP memory 204. In implementations where multiple PPUs 202 are present, those PPUs may be operated in parallel to process data at a higher throughput than is possible with a single PPU 202. Systems incorporating one or more PPUs 202 may be implemented in a variety of configurations and form factors, including, without limitation, desktops, laptops, handheld personal computers or other handheld devices, servers, workstations, game consoles, embedded systems, and the like.
Operation of GPC 208 is controlled via a pipeline manager 305 that distributes processing tasks received from a work distribution unit (not shown) within task/work unit 207 to one or more streaming multiprocessors (SMs) 310. Pipeline manager 305 may also be configured to control a work distribution crossbar 330 by specifying destinations for processed data output by SMs 310.
In one embodiment, GPC 208 includes a set of M of SMs 310, where M≧1. Also, each SM 310 includes a set of functional execution units (not shown), such as execution units and load-store units. Processing operations specific to any of the functional execution units may be pipelined, which enables a new instruction to be issued for execution before a previous instruction has completed execution. Any combination of functional execution units within a given SM 310 may be provided. In various embodiments, the functional execution units may be configured to support a variety of different operations including integer and floating point arithmetic (e.g., addition and multiplication), comparison operations, Boolean operations (AND, OR, XOR), bit-shifting, and computation of various algebraic functions (e.g., planar interpolation and trigonometric, exponential, and logarithmic functions, etc.). Advantageously, the same functional execution unit can be configured to perform different operations.
In operation, each SM 310 is configured to process one or more thread groups. As used herein, a “thread group” or “warp” refers to a group of threads concurrently executing the same program on different input data, with one thread of the group being assigned to a different execution unit within an SM 310. A thread group may include fewer threads than the number of execution units within the SM 310, in which case some of the execution may be idle during cycles when that thread group is being processed. A thread group may also include more threads than the number of execution units within the SM 310, in which case processing may occur over consecutive clock cycles. Since each SM 310 can support up to G thread groups concurrently, it follows that up to G*M thread groups can be executing in GPC 208 at any given time.
Additionally, a plurality of related thread groups may be active (in different phases of execution) at the same time within an SM 310. This collection of thread groups is referred to herein as a “cooperative thread array” (“CTA”) or “thread array.” The size of a particular CTA is equal to m*k, where k is the number of concurrently executing threads in a thread group, which is typically an integer multiple of the number of execution units within the SM 310, and m is the number of thread groups simultaneously active within the SM 310.
Although not shown in
Each GPC 208 may have an associated memory management unit (MMU) 320 that is configured to map virtual addresses into physical addresses. In various embodiments, MMU 320 may reside either within GPC 208 or within the memory interface 214. The MMU 320 includes a set of page table entries (PTEs) used to map a virtual address to a physical address of a tile or memory page and optionally a cache line index. The MMU 320 may include address translation lookaside buffers (TLB) or caches that may reside within SMs 310, within one or more L1 caches, or within GPC 208.
In graphics and compute applications, GPC 208 may be configured such that each SM 310 is coupled to a texture unit 315 for performing texture mapping operations, such as determining texture sample positions, reading texture data, and filtering texture data.
In operation, each SM 310 transmits a processed task to work distribution crossbar 330 in order to provide the processed task to another GPC 208 for further processing or to store the processed task in an L2 cache (not shown), parallel processing memory 204, or system memory 104 via crossbar unit 210. In addition, a pre-raster operations (preROP) unit 325 is configured to receive data from SM 310, direct data to one or more raster operations (ROP) units within partition units 215, perform optimizations for color blending, organize pixel color data, and perform address translations.
It will be appreciated that the core architecture described herein is illustrative and that variations and modifications are possible. Among other things, any number of processing units, such as SMs 310, texture units 315, or preROP units 325, may be included within GPC 208. Further, as described above in conjunction with
As shown, flip-flop element 402 includes master latch 404, slave latch 406, and NAND gate 408. Flip-flop element 402 couples clock 412 to M-Clk 420, the clock input of master latch 404, to S-Clk 430, the clock input of slave latch 406, and to one input of NAND gate 408.
Flip-flop element 402, further, couples data in 410 to M-D 418, the data input port of master latch 404. Master latch 404 produces the state of data in 410 at the output port, M-Q 426, on the rising edge of clock 412. Master latch 404 couples the state of data in 410 to S-D 428, the data input port of slave latch 406. Master latch 404, then, couples the inversion of the state of data in 410, at M-Q-not 424, to the second input of NAND gate 408. When M-Q-not 424 is low, the output of NAND gate 408 is gated high.
NAND gate 408 produces an inversion of clock 412 only when m-q-not 424 is high. Thus, gated-clock-not 414 is an inversion of clock 412 that is gated according to the state of data in 410. Gated-clock-not 414 is, then, coupled to M-Clk-not 422, the inverted clock input of master latch 404, and to S-Clk-not 432, the inverted clock input of slave latch 406. Slave latch 406 forwards the state of the S-D 428 input to the S-Q 434 output to produce latched data 416.
In this manner, all clock inversion transitions within master latch 404 and slave latch 406 are controlled by gated-clock-not 414. When gated-clock-not 414 is at a high logic level, the high input to NAND gate 408 enables NAND gate 408 to produce the inversion of clock 412 at the output as gated-clock-not 414. All junctions coupled to gated-clock-not 414, then, transition between the high and low level with each cycle of clock 412. However, when gated-clock-not 414 is at a low logic level, the low input to NAND gate 408 forces the output of NAND gate 408 to a steady high state, and all transitions of gated-clock-not 414 cease. All junctions coupled to gated-clock-not 414 will, then, be in an idle state. Thus, the number of transitions and the consequent power penalty associated with the inverted clock may be reduced, as described above, by generally the proportionality of the two logic states. Consequently, a substantial reduction in system power usage may be realized.
Flip-flop element 402 couples data in 410 to the input stack of master latch 404. PFET 508 and NFET 510 constitute an inverter that is gated by clock 412, applied to PFET 506, and by gated-clock-not 414, applied to NFET 512. The input stack of master latch 404 couples inverted data to the keeper of master latch 404 as M-Q-not 424. Gated-clock-not 414 controls PFET 514, which gates the common 504 connection to PFET 516 and NFET 518. Similarly, clock 412 controls FET 520, which gates the supply voltage 502 connection to PFET 516 and NFET 518. Thus, master latch 404 retains the inversion of the state of data in 410.
Master latch 404 couples captured data inversion, M-Q-not 424, to one input of NAND gate 408. Flip-flop element 402 couples clock 412 to the second input of NAND gate 408. NAND gate 408, then, produces gated-clock-not 414 which is the gated inversion of clock 412.
Master latch 404, further, couples captured data, M-Q-26 to the data input of slave latch 406, NFET 534. Gated-clock-not 414 controls PFET 532, which gates the supply voltage 502 connection to NFET 534. Similarly, clock 412 controls NFET 536, which gates the common 504 connection to NFET 534.
The circuit topology of the input stack of slave latch 406 consists of a single FET in the inverter portion. The function of an upper FET of the inverter portion, analogous to PFET 508 in master latch 404, would be redundant with the implementation of the inversion clock via gated-clock-not 414. Further, in the keeper of slave latch 406, PFET 506, common to master latch 404 input stack, performs the functionality of the upper clocking FET, analogous to PFET 514 in master latch 404. Thus, slave latch 704 eliminates two FETs compared to conventional designs.
Eliminating two FETs reduces power usage in the flip-flop element. Further, the gating of the clock inversion achieved by NAND gate 408 allow substantial reduction in the number of junction transitions occurring on each clock cycle. Thus, the staticized flip-flop element affords substantial reduction in the power consumption.
The inclusion of PFET 806 transforms the functionality of the logic gate 804 into a two-input and-or-invert, (AOI22). Data in 410 couples to inverter 808, producing the inversion of data in 410 to drive PFET 806. Consequently, PFET 806 enables toggling of gated-clock-not 414 by the switching action of PFET 530 only when data in 410 is high.
Thus, PFET 806 prevents a series of zeroes at data in 410 from repeatedly precharging gated-clock-not 414. Reducing the number of transitions of gated-clock-not 414 further reduces the power usage of staticized master-slave flip-flop element 802.
In sum, a flip-flop element is configured to gate the clock inversions within a master-slave flip-flop element. The flip-flop element reduces the number of circuit elements within the flip-flop element by collapsing elements with common functionality into a single circuit element. Further, by making the actions of judiciously selected circuit elements conditional upon the state of the input data, the flip-flop element circuit reduces the number of internal transitions. In this manner, by reducing the number of circuit elements as well as the number of transitions, the flip-flop element achieves substantial reduction in overall system power consumption, resulting in a more efficient system.
One advantage of the flip-flop elements disclosed herein is that gating the clock inversion according to the state of the input data reduces the number of transitions that internal components incur. Combining elements with common functionality further reduces component count and transitions within the flip-flop element. Reducing the number of FETs and inverters within the flip-flop element and the number of internal transitions results in a substantial reduction in flip-flop element power usage.
The invention has been described above with reference to specific embodiments. Persons of ordinary skill in the art, however, will understand that various modifications and changes may be made thereto without departing from the broader spirit and scope of the invention as set forth in the appended claims. The foregoing description and drawings are, accordingly, to be regarded in an illustrative rather than a restrictive sense.
Therefore, the scope of embodiments of the present invention is set forth in the claims that follow.
