Classifications

• • G—PHYSICS
  • G06—COMPUTING OR CALCULATING; COUNTING
  • G06F—ELECTRIC DIGITAL DATA PROCESSING
  • G06F1/00—Details not covered by groups G06F3/00 - G06F13/00 and G06F21/00
  • G06F1/04—Generating or distributing clock signals or signals derived directly therefrom
• • G—PHYSICS
  • G11—INFORMATION STORAGE
  • G11C—STATIC STORES
  • G11C11/00—Digital stores characterised by the use of particular electric or magnetic storage elements; Storage elements therefor
  • G11C11/21—Digital stores characterised by the use of particular electric or magnetic storage elements; Storage elements therefor using electric elements
  • G11C11/34—Digital stores characterised by the use of particular electric or magnetic storage elements; Storage elements therefor using electric elements using semiconductor devices
  • G11C11/40—Digital stores characterised by the use of particular electric or magnetic storage elements; Storage elements therefor using electric elements using semiconductor devices using transistors
  • G11C11/41—Digital stores characterised by the use of particular electric or magnetic storage elements; Storage elements therefor using electric elements using semiconductor devices using transistors forming static cells with positive feedback, i.e. cells not needing refreshing or charge regeneration, e.g. bistable multivibrator or Schmitt trigger
  • G11C11/413—Auxiliary circuits, e.g. for addressing, decoding, driving, writing, sensing, timing or power reduction
  • G11C11/417—Auxiliary circuits, e.g. for addressing, decoding, driving, writing, sensing, timing or power reduction for memory cells of the field-effect type
• • G—PHYSICS
  • G11—INFORMATION STORAGE
  • G11C—STATIC STORES
  • G11C11/00—Digital stores characterised by the use of particular electric or magnetic storage elements; Storage elements therefor
  • G11C11/21—Digital stores characterised by the use of particular electric or magnetic storage elements; Storage elements therefor using electric elements
  • G11C11/34—Digital stores characterised by the use of particular electric or magnetic storage elements; Storage elements therefor using electric elements using semiconductor devices
  • G11C11/40—Digital stores characterised by the use of particular electric or magnetic storage elements; Storage elements therefor using electric elements using semiconductor devices using transistors
  • G11C11/41—Digital stores characterised by the use of particular electric or magnetic storage elements; Storage elements therefor using electric elements using semiconductor devices using transistors forming static cells with positive feedback, i.e. cells not needing refreshing or charge regeneration, e.g. bistable multivibrator or Schmitt trigger
  • G11C11/413—Auxiliary circuits, e.g. for addressing, decoding, driving, writing, sensing, timing or power reduction
  • G11C11/417—Auxiliary circuits, e.g. for addressing, decoding, driving, writing, sensing, timing or power reduction for memory cells of the field-effect type
  • G11C11/418—Address circuits
• • G—PHYSICS
  • G11—INFORMATION STORAGE
  • G11C—STATIC STORES
  • G11C11/00—Digital stores characterised by the use of particular electric or magnetic storage elements; Storage elements therefor
  • G11C11/21—Digital stores characterised by the use of particular electric or magnetic storage elements; Storage elements therefor using electric elements
  • G11C11/34—Digital stores characterised by the use of particular electric or magnetic storage elements; Storage elements therefor using electric elements using semiconductor devices
  • G11C11/40—Digital stores characterised by the use of particular electric or magnetic storage elements; Storage elements therefor using electric elements using semiconductor devices using transistors
  • G11C11/41—Digital stores characterised by the use of particular electric or magnetic storage elements; Storage elements therefor using electric elements using semiconductor devices using transistors forming static cells with positive feedback, i.e. cells not needing refreshing or charge regeneration, e.g. bistable multivibrator or Schmitt trigger
  • G11C11/413—Auxiliary circuits, e.g. for addressing, decoding, driving, writing, sensing, timing or power reduction
  • G11C11/417—Auxiliary circuits, e.g. for addressing, decoding, driving, writing, sensing, timing or power reduction for memory cells of the field-effect type
  • G11C11/419—Read-write [R-W] circuits
• • G—PHYSICS
  • G11—INFORMATION STORAGE
  • G11C—STATIC STORES
  • G11C7/00—Arrangements for writing information into, or reading information out from, a digital store
  • G11C7/22—Read-write [R-W] timing or clocking circuits; Read-write [R-W] control signal generators or management 
  • G11C7/222—Clock generating, synchronizing or distributing circuits within memory device

Definitions

• Disclosed embodiments generally relate to pulse clock generation. More particularly, exemplary embodiments are directed to generating pulse clocks with programmable edges and pulse widths configured for varying requirements of different memory access operations.
• T-SRAM 6 Transistor Static Random Access Memory
• 6T-SRAM 6 Transistor Static Random Access Memory
• a common solution involves the use of multiple voltage domains, such that memory cells may be operated at a relatively high voltage level while other on-chip logic may be operated at lower voltages.
• level shifters are commonly used to convert low voltage to high voltage and vice versa as needed.
• level shifters introduce latency, which may be undesirable in timing critical paths.
• pulse clocks are often utilized to overcome the effects of device variations (e.g. due to process variations). These pulse clocks may be subjected to conflicting needs for different memory access operations. For example, during a memory read operation it may be desirable for the rising edge of the pulse clock to arrive as fast as possible at memory cells of the memory array being accessed, in order to enable quick read access. Moreover, during the memory read operation it may be desirable for the pulse width of the pulse clock to be narrow, in order to reduce power consumption by disallowing full voltage swings on bit lines.
• memory write operations are usually not timing critical, and thus can tolerate higher arrival delays in the rising edge of the pulse clock.
• memory write operations may benefit from delays in the pulse clock so as to allow sufficient set-up time for the data written.
• the pulse width of the pulse clock may be wider to allow for sufficient write time in order to guarantee successful completion of the write operations, especially at lower supply voltage levels.
• Conventional pulse clock implementations are not well suited to meet these conflicting needs for read and write operations on memory cells.
• Exemplary embodiments of the invention are directed to systems and methods for generating pulse clocks with programmable edges and pulse widths configured for varying requirements of different memory access operations.
• an exemplary embodiment is directed to a method for generating a pulse clock comprising: configuring a programmable rising edge delay of the pulse clock; configuring a programmable pulse width of the pulse clock; configuring a level shifter for shifting a voltage level of the pulse clock; and generating the pulse clock in accordance with the programmed rising edge delay, programmed pulse width, and shifted voltage level.
• Another exemplary embodiment is directed to a pulse clock generation circuit comprising: a selective delay logic to provide a programmable rising edge delay of the pulse clock; a selective pulse width widening logic to provide a programmable pulse width of the pulse clock; a built-in level shifter configured to shift a voltage level of the pulse clock; and logic configured to generate the pulse clock in accordance with the programmed rising edge delay, programmed pulse width, and shifted voltage level.
• Yet another exemplary embodiment is directed to a pulse clock generation system for generating a pulse clock, the pulse clock generation system comprising: a selective delay means for providing a programmable rising edge delay of the pulse clock; a selective pulse width widening means for providing a programmable pulse width of the pulse clock; a built-in level shifting means for shifting a voltage level of the pulse clock; and means for generating the pulse clock in accordance with the programmed rising edge delay, programmed pulse width, and shifted voltage level.
• Yet another exemplary embodiment is directed to non-transitory computer-readable storage medium comprising code, which, when executed by a processor, causes the processor to perform operations for programming a pulse clock generator, the non- transitory computer-readable storage medium comprising: code for configuring a programmable rising edge delay of the pulse clock; code for configuring a programmable pulse width of the pulse clock; code for configuring a level shifter for shifting a voltage level of the pulse clock; and code for generating the pulse clock in accordance with the programmed rising edge delay, programmed pulse width, and shifted voltage level.
• FIG. 1 illustrates a schematic representation of self-timed pulse clock generation circuit
• FIG. 2 illustrates a transistor- level schematic view of pulse clock generation logic 112 of the self-timed pulse clock generation circuit 100 of FIG. 1.
• FIG. 3 illustrates a timing diagram comprising representations of pulse clocks pertaining to self-timed pulse clock generation circuit 100.
• FIG. 4 illustrates a conventional 6T SRAM cell 400 to which a pulse clock generated according to exemplary embodiments may be applied.
• FIG. 5 is a flow chart representing an exemplary method of generating a pulse clock according to exemplary embodiments.
• FIG. 6 illustrates an exemplary wireless communication system 600 in which an embodiment of the disclosure may be advantageously employed.
• Exemplary embodiments are directed to self-timed pulse clock generation circuits configured for programmable pulse width and rising edge delays in order to meet conflicting requirements of read and write operations. Further, exemplary embodiments may include a built-in level shifter configured to convert a raw clock in a low voltage domain to pulse clocks for read and write operations in high voltage domain without introducing additional delays from a separate voltage level shifter, while also preventing glitches and charge sharing problems which may occur in conventional implementations .
• raw clock 122 may be input to self-timed pulse clock generation circuit 100, and may serve as a base clock for generating customized clock pulses for read and write operations.
• Raw clock 122 may be generated by known techniques such as by the use of piezoelectric crystals, etc.
• Raw clock 122 may be made available at a low voltage level or may be derived from a low voltage domain.
• Raw clock 122 may first enter selective delay logic 102, wherein raw clock 122 may be delayed selectively for write operations.
• Programmable delay logic 104, read/write control 126 and selector 106 may be configured as shown for selectively delaying raw clock 122.
• read/write control 126 if a read operation is indicated by read/write control 126, then raw clock 122 may be selected by selector 106 to form the selectively delayed output clock clk_dl 116.
• programmable delay logic 104 may delay raw clock 122 by a desired amount and generate a delayed clock 124, which may then be selected by selector 106 to form the selectively delayed output clock clk_dl 116.
• Programmable delay logic 104 may be implemented by known techniques, such as, by controlling a number of inverters or buffer elements to be coupled in series. Effective sizes of these inverters or buffer elements may also be controlled to adjust the delay amount. Skilled persons will recognize suitable implementations for programmable delay logic 104.
• Pulse clock generation logic 112 may ultimately output pulse clock 114 which meets above-noted requirements for read and write operations in exemplary embodiments.
• Pulse clock generation logic 112 may include logic for voltage level shifting, such that output pulse clock 114 is made available at a high voltage level without introducing unwanted latency from a separate voltage level shifter. Further, pulse clock generation logic 112 may be configured to eliminate glitches in output pulse clock 114.
• Feedback path 118 from output pulse clock 114 back to pulse clock generation logic 112 through reset 121 may include read delay mimic logic 108 and selective pulse width widening logic 110 as shown. Feedback path 118 may provide a self -resetting and thus self-timing functionality to pulse clock generation logic 112. Pulse clock generation logic 112, read delay mimic logic 108, and selective pulse width widening logic 110 will now be described in further detail.
• pulse clock generation logic 112 which includes a transistor-level schematic.
• Enable 130 and power-up reset 128 are input to clock generation logic 112 from an external source (not shown) to enable or reset the output pulse clock 114.
• An exemplary operation of pulse clock generation logic 112 will now be described. Initially, power-up reset 128 is driven high, thus causing the output of inverter i6 to be driven low and triggering pull-up transistor p3 to pull up lat_node 142 and thereby driving output pulse clock 114 low via inverter i5. It will be observed that pull-up transistor p3 is in the high voltage domain, as indicated in FIG. 2. Thus, output pulse clock 114 is derived in the high voltage domain.
• power-up reset 128 is driven low and enable 130 is driven high.
• the selectively delayed output clock, clk_dl 116 traverses through inverters il, i2, and i3, and thus the output of i3 presents an inverted and delayed pulse following clk_dl.
• both inputs to nandl, i.e. the output of i3 and enable 130 are high, thus causing the output of nandl to be low.
• pull-up transistors p5 and p6 will be partially turned off, because the source voltages of pull-up transistors p5 and p6 are in the high voltage domain and their gate voltages are in the low voltage domain; and all four pull-down transistors nO, nl, n2, and n4 will be turned on at this instance. Therefore, both nodes int_node 144 and lat_node 142 will be pulled down and inverter i5 will drive output pulse clock 114 high.
• pull-down transistor n3 will be turned on, which will provide additional current source to speed up the pull down process of lat_node 142 and int_node 144 .
• output pulse clock 114 is already in the high voltage domain. Accordingly, the rising edge of raw clock 122 in the low voltage domain creates a rising edge of output pulse clock 114 in the high voltage domain without requiring additional voltage level shifting logic. While pull-up transistors p5 and p6 being turned off halfway may incur minor delay and power consumption costs, these costs may be offset by the additional pull-down path provided by pull-down transistors n3 and n4.
• lat_node 142 will remain in the low state and pulse clock 114 will remain in the high state due to the latch created by inverter i4 and the combination of pull-up transistors p4, p6 and pull-down transistors n3, n4, until a reset mechanism returns lat_node 142 to high state, thus driving output pulse clock 114 low.
• Reset 121 which will be described in further detail below, can control the falling edge of output pulse clock 114. It will be recognized that controlling the time period between the rising edge of output pulse clock 114 and the falling edge of output pulse clock 114 will determine the width of the clock pulse derived at output pulse clock 114.
• selective pulse width widening logic 110 may be employed to selectively expand the pulse width between rising and falling edges of output pulse clock 114. Similar to selective delay logic 102, selective pulse width widening logic 110 may employ multiplexor 107 to choose between the input signal without added delays and a delayed version of the input signal, based on read/write control 126. The timing of read/write control 126 may be appropriately adjusted by buffer 127 to account for delays in reaching selective pulse width widening block 110 after passing through selective delay block 102 which was previously discussed. Accordingly, depending on whether the operation is a read or a write, the delay path through selective pulse width widening logic may be controlled, such that the pulse width for write operations is made wider.
• output 120 of read delay mimic logic 108 may be selectively delayed to cause the width of the output of selective pulse width widening logic 110, reset 121 to change. This will control the time it takes for reset 121 to rise from a low to high level and then fall back to the low level.
• reset 121 falls, a next rising edge of output pulse clock 114 may be generated by repeating the above process. The next falling edge may be suitably delayed for read operations by read delay mimic logic 108.
• read delay mimic logic 108 is configured to ultimately delay the falling edge of output pulse clock 114, thus controlling the pulse width.
• the rising edge of output pulse clock 114 is controlled by selective delay logic 102.
• feedback path 118 from output pulse clock 114 to reset 121 will now be described. As shown in FIG. 1, this path comprises read delay mimic logic 108 and selective pulse width widening logic 110.
• Read delay mimic logic 108 may be included to control the delay in the falling edge of output pulse clock 114 from the rising edge of output pulse clock 114 in order to account for memory array access delays encountered during a read operation.
• read delay mimic logic 108 may be configured to provide a delay corresponding to the delay which may be incurred in accessing a memory cell in the memory array.
• Read delay mimic logic may comprise a programmable delay element to match an expected read access delay. Thus, the rising edge of output pulse clock 114 will be delayed by read delay mimic logic 108 to form the rising edge of output 120.
• selective pulse width widening logic 110 will select the input without any delay added and the rising edge of output 120 will essentially appear as the rising edge of reset 121, which is inverted by inverter i7 in pulse clock generation logic 112.
• the inverter formed by pull-up transistor pO and pull-down transistor nO will reset lat_node 142 to a high state.
• the feedback loop including read mimic delay logic 108 of FIG. 1 will essentially ensure that the falling edge of output pulse clock 114 is made available at a delayed time in the high voltage domain in the case of read operations.
• node clk_d2 140 will fall after the rising edge of selectively delayed output clock clk_dl 116 traverses inverters il, i2, i3, and nandl and norl. This will cause pulldown transistor n2 to be turned off, and also pull-up transistor p6 to be turned on. This will cause lat_node 142 to rise, thus creating the falling edge of output pulse clock 114. At this point in time, pull-down transistor nl may either be turned on or off, without affecting the latching function of lat_node 142.
• selective pulse width widening logic 110 will selectively introduce an additional delay in the feedback path before reset 121 may be asserted and thus allowing lat_node 142 to rise and create the falling edge of output pulse clock 114. In this manner, write operations will have a wider pulse width. As previously described, a wider pulse width may facilitate write operations by providing additional time for completion of write operations (which as will be recalled, are usually not timing critical).
• pull-down transistors nl and n2 are a self-timed process.
• the time period for which pull-down transistors nl and n2 are turned on may be determined by the delay through inverters il, i2, i3, and gates nandl and norl.
• pull-down transistors nl and n2 will be turned off before reset 121 is de- asserted. Accordingly, in cases where read/write operations require multiple cycles for their completion, a single pulse of enable 130 may be provided during the entire operation, thus ensuring that only a single pulse of output pulse clock 114 is generated for the duration of the entire operation.
• the rising edge of output pulse clock 114 may be delayed from raw clock 122 by a controllable amount for read operations and write operations.
• the latch mechanism including inverter i4 and the combination of pull-up transistors p4, p6 and pull-down transistors n3, n4 will hold output pulse clock 114 at a high state until a falling edge is imposed after a controllable time period for read and write operations by asserting signal reset 121.
• pulse clock generation logic 112 is also configured to avoid glitches and undesirable charge sharing which may occur, for example, between lat_node 142 and int_node 144.
• reset 121 changes between high and low state
• lat_node 142 may remain at a high state due to the latch mechanism, but int_node 144 may fall to a low state, thus causing a large voltage differential to occur between drain and source nodes of pull-down transistor nO. This may lead to unwanted charge sharing problems.
• int_node 144 may be pulled up to high state using the pull-up transistors p5, p6, pi and p2.
• both lat_node 142 and int_node 144 will be in the high state, and a large differential voltage will no longer affect pull-down transistor nO.
• Timing diagram 302 illustrates raw clock 122 at a low voltage level.
• output pulse clock 114 generated for a read operation is at a high voltage level and comprises a rising edge delay 304d; and as shown in timing diagram 306, output pulse clock 114 generated for a write operation is at a high voltage level and comprises a rising edge delay 306d.
• expected read array delay is accounted for (e.g. through read delay mimic logic 108 of Fig.
• rising edge delay 304d may be made as fast as required to meet the needs of a fast read operation. In comparison, rising edge delay 306d may not need to be fast and thus may be slowed down as required. Further, compared to timing diagram 304, timing diagram 306 illustrates pulse width 306w for a write operation to be wider than pulse width 304w for a read operation, in order to satisfy wider pulse width requirements of write operations as described above.
• output pulse clock 114 is used by logic components well known in the art, such as a word line decoder, generally designated by logic 402 in FIG. 4.
• the operation of 6T SRAM cell 400 is well known.
• Transistors Ml, M2, M3, and M4 form a cross- coupled inverter storage element, while access transistors M5 and M6 control read/write operations for the storage element.
• Word line WL controls the gates of access transistors M5 and M6, while complementary bit lines BL and BLB are used to select the particular 6T SRAM cell 400 for write operations and to read out the stored value for read operations.
• output pulse clock 114 may be applied to word line WL in order to meet specific pulse clock requirements for read and write operations, as shown, for example, in FIG. 3.
• delays introduced by selective delay logic 102 may delay the rising edge of output pulse clock 114 during write operations, which again allows additional set up time for bit lines BL and BLB, and moreover may reduce power because accompanying buffers and drivers may be downsized. Further, the wider pulse width created by selective pulse width widening logic 110 may facilitate write operations even at lower voltage levels, which may lead to improved yield.
• integrating the voltage level shifting mechanism in pulse clock generation logic 112 may lead to lower latency and improved reuse of functional blocks, as compared to conventional implementations which have level shifters separated from pulse clock generation logic.
• the feedback path to built-in level shifter via pull-down transistors n3 and n4 may speed up the rising edge of output pulse clock 114 especially when the voltage difference between the low voltage level of raw clock 122 and the high voltage level of output pulse clock 114 is large.
• the latch mechanism in pulse clock generation logic 112 may additionally facilitate holding the logic state of output pulse clock 114 stable.
• an embodiment can include a method for generating a pulse clock (e.g. output pulse clock 114 of FIG. 1) comprising: configuring a programmable rising edge delay (e.g. rising edge delay 304d and 306d as illustrated in timing diagrams 304 and 306 of FIG. 3) of the pulse clock - Block 502; configuring a programmable pulse width (e.g. width 304w and 306w as illustrated in timing diagrams 304 and 306 of FIG. 3) of the pulse clock - Block 504; configuring a level shifter (e.g.
• a software module may reside in RAM memory, flash memory, ROM memory, EPROM memory, EEPROM memory, registers, hard disk, a removable disk, a CD-ROM, or any other form of storage medium known in the art.
• An exemplary storage medium is coupled to the processor such that the processor can read information from, and write information to, the storage medium. In the alternative, the storage medium may be integral to the processor.
• FIG. 6 a block diagram of a particular illustrative embodiment of a wireless device that includes a multi-core processor configured according to exemplary embodiments is depicted and generally designated 600.
• the device 600 includes a digital signal processor (DSP) 664, which may include self-timed pulse clock generation circuit 100 of FIG. 1.
• DSP 664 may be coupled to memory 632 as shown, wherein the generated pulse clock from self-timed pulse clock generation circuit 100 may be used for memory accesses on memory 632.
• FIG. 6 also shows an external device such as display controller 626 that is coupled to DSP 664 and to display 628.
• An external device such as coder/decoder (CODEC) 634 (e.g., an audio and/or voice CODEC) can also be coupled to DSP 664.
• CODEC coder/decoder
• Other components/external devices, such as wireless controller 640 (which may include a modem) are also illustrated.
• Speaker 636 and microphone 638 can be coupled to CODEC 634.
• FIG. 6 also indicates that wireless controller 640 can be coupled to wireless antenna 642.
• DSP 664, display controller 626, memory 632, CODEC 634, and wireless controller 640 are included in a system-in-package or system-on-chip device 622.
• input device 630 and power supply 644 are coupled to the system-on-chip device 622.
• display 628, input device 630, speaker 636, microphone 638, wireless antenna 642, and power supply 644 are external to the system-on-chip device 622.
• each of display 628, input device 630, speaker 636, microphone 638, wireless antenna 642, and power supply 644 can be coupled to a component of the system-on-chip device 622, such as an interface or a controller.
• FIG. 6 depicts a wireless communications device
• DSP 664 and memory 632 may also be integrated into a set-top box, a music player, a video player, an entertainment unit, a navigation device, a personal digital assistant (PDA), a fixed location data unit, or a computer.
• a processor e.g., DSP 664 may also be integrated into such a device.
• an embodiment of the invention can include a computer readable media embodying a method for generating a pulse clock with a built-in level shifter and a programmable rising edge and programmable pulse width. Accordingly, the invention is not limited to illustrated examples and any means for performing the functionality described herein are included in embodiments of the invention.

Landscapes

• Engineering & Computer Science (AREA)
• Microelectronics & Electronic Packaging (AREA)
• Computer Hardware Design (AREA)
• Theoretical Computer Science (AREA)
• Physics & Mathematics (AREA)
• General Engineering & Computer Science (AREA)
• General Physics & Mathematics (AREA)
• Pulse Circuits (AREA)

Priority Applications (4)

Applications Claiming Priority (2)

Publications (1)

Family

ID=48083683

Family Applications (1)

Country Status (6)

Families Citing this family (17)

Citations (2)

Family Cites Families (11)

• 2012
  • 2012-03-29 US US13/433,891 patent/US8638153B2/en not_active Expired - Fee Related
• 2013
  • 2013-03-28 JP JP2015503586A patent/JP6242852B2/ja not_active Expired - Fee Related
  • 2013-03-28 WO PCT/US2013/034414 patent/WO2013149040A1/en not_active Ceased
  • 2013-03-28 EP EP13715579.2A patent/EP2831694B1/en not_active Not-in-force
  • 2013-03-28 KR KR1020147029964A patent/KR20140139595A/ko not_active Withdrawn
  • 2013-03-28 CN CN201380016930.XA patent/CN104204992B/zh not_active Expired - Fee Related

Patent Citations (2)

Non-Patent Citations (1)

Also Published As

Similar Documents

Legal Events