System and method for controlling timing of output signals

Abstract

The timing of output signals can be controlled by coupling a digital signal through a signal distribution tree having a plurality of branches extending from an input node to respective clock inputs of a plurality of latches. A phase interpolator is included in a signal path common to all of the branches, and a respective delay line is included in each of the branches. Each of the latches couples a signal applied to its data input to an output terminal responsive to a transition of the digital signal applied to its clock input. The delay lines are adjusted so that the latches are simultaneously clocked. The delay of the phase interpolator is adjusted so that the signals are coupled to the output terminals of the latches with a predetermined timing relationship relative to signals coupled to output terminals of a second signal distribution tree.

Description

TECHNICAL FIELD

This invention relates to transmitting digital signals from electronic devices such as testers, memory devices, etc., and, more particularly, to a system and method for precisely controlling the timing at which a digital signals are output from the electronic devices.

BACKGROUND OF THE INVENTION

Digital signals are received and transmitted by many types of digital electronic devices at ever increasing rates. For example, the rate at which command, address and write data signals are applied to memory devices, such as dynamic random access memory (“DRAM”) devices, continues to increase, as does the rate at which read data signals are output from memory devices. As the rate at which such signals are transmitted continues to increase, it has become more difficult to ensure that the signals are transmitted at the proper time and in synchronism with each other. For example, typical clock trees 10, 14 are shown in FIG. 1. The clock tree 10 couples a first internal clock signal ICLK₁ through a series of symmetrically connected buffers 18, which may be two-transistor inverters, to the clock inputs of a plurality of latches 20_{0, 1, . . . N}. Each of the latches 20_{0, 1, . . . N}, receives at its data input a respective data bit D_{0, 1, . . . N}, and outputs a respective data signal DQ_{0, 1, . . . N}, to a respective DQ terminal responsive to the rising edge of the clock signal applied to the its clock input. Insofar as all of the latches 20_{0, 1, . . . N} are driven through the same clock tree 10, the DQ terminals are considered to be in the same “pin group.”

The other clock tree 14 receives a second internal clock signal ICLK₂, and couples the ICLK₂ signal through buffers 22 to the clock inputs of respective latches 24a,b. The data inputs of the latches 24a,b are coupled to a logic “1” level. The latch 24a therefore outputs a high data strobe signal to the DQS terminal responsive to the rising edge of the ICLK₂ signal. This high at the output of the latch 24a also resets the latch 24b. The latch 24b receives the ICLK₂ signal through an odd number of buffers 22 so that it outputs a high complementary data strobe signal DQS* responsive to the falling edge of the ICLK₂ signal. The high at the output of the latch 24b also resets the latch 24a. The DQS and DQS* signals are considered to be in the same pin group, which is different from the DQ signal pin group.

The clock trees 10, 14 shown in FIG. 1 are typical of those used in, for example, memory controllers to output write data signals to memory devices, memory devices to output read data signals to a memory controller, or testing systems to output digital signals to devices under test. The data strobe signals DQS, DQS* are typically used in source synchronous applications to strobe data signals transmitted from the latches 20 at a receiving device. For example, in double data rate (“DDR”) memory devices, the rising edge of the DQS signal is used by a serializing device to transmit a first set of data signals from the latches 20, and the rising edge of the DQS* signal is used to transmit a second set of data signals from the latches 20. At a receiving device, the rising edge of the DQS signal is used to latch the first set of data signals from the latches 20, and the rising edge of the DQS* signal is used to latch the second set of data signals from the latches 20. In such memory devices, the DQ signals can be considered one pin group, the address signals can be considered another pin set, and the command signals can be considered still another pin set, insofar as they are each triggered by an internal clock signal coupled through different clock trees.

As the speed at which data signals are transmitted continues to increase, it has become more difficult to ensure that the DQ signals are all transmitted at the same time, and that the DQS and DQS* signals have the proper timing relative to the DQ signals. With further reference to FIG. 1, one problem with ensuring that the DQ signals are all transmitted at the same time results from unequal lengths in the signal path from the node to which the ICLK₁ signal is applied to the clock inputs of the latches 20. The unequal path lengths can cause the ICLK₁ signal to be applied to the latches 20 at different times, thereby causing the latches 20 to output the DQ signals at different times.

It can also be seen from FIG. 1 that the number of buffers 18 through which the ICLK₁ signal propagates is different from the number of buffers 22 through which the ICLK₂ signal propagates. This difference can cause the transitions of the DQS and DQS* signals to occur before or after the DQ signals output by the latches 20 are valid. The propagation time differences can be compensated for to some extent by adding delay in the signal path of the ICLK₂ signal, such as by adding addition buffers. However, adding buffers that are to be used only for increasing delay takes up valuable space on a semiconductor die. Also, the propagation delays through the buffers 18, 22 generally can vary with process variations, supply voltage fluctuations, and temperature changes. Therefore, if the propagation delays of the ICLK₁ and ICLK₂ signals are equalized for one set of conditions, the propagation delays may no longer be equal for different processing runs of a device, for different supply voltages and/or for different temperatures. Alternatively, a smaller number of buffers 18 could be used in the clock tree 10, and each buffer 18 could be coupled to a larger number of latches 20. However, the buffers 18 would then be loaded to a greater extent than the loading of the buffers 22. As a result, the ICLK₁ signal coupled through the heavily loaded buffers 18 would be delayed to a greater extent than the delay of the ICLK₂ signal coupled through the lightly loaded buffers 22. As a result, the transitions of the DQS and DQS* signals might not occur at a time the DQ signals are valid.

While the number of buffers 22 through which the ICLK₂ signal propagates is different from the number of buffers 18 through which the ICLK₁ signal propagates, the number of buffers 18 through which the ICLK₁ signal propagates to reach each of the latches 20 is the same for all branches of the clock tree 10. Therefore, the timing at which the ICLK₁ signal reaches each of the latches 20 will theoretically be the same despite process, voltage and temperature variations. However, the lengths of the conductors through which the ICLK₁ signal must propagate to reach each of the latches 20 will generally not be the same. Furthermore, it is generally not possible to compensate for these different propagation times by, for example, inserting additional buffers in the signal path because the propagation times of the buffers, but not the propagation time of conductors, will generally vary with process, voltage and temperature variations.

The manner in which the propagation delay of the buffers 18, 22 vary with, for example, temperature is shown in the graph of FIG. 2, which also shows the relatively constant conductor or wire propagation delay. As shown in FIG. 2, the total propagation delay is the sum of the buffer or other semiconductor element delay and the wire delays. The slope and magnitude of the total propagation delay curve will vary with the relative contribution of the semiconductor element delays and the wire delays. In general, the delay curve will be steeper if the semiconductor element delays are a higher percentage of the total delay, and it will be shallower if the wire delays are a higher percentage of the total delay. The variation in both the slope and magnitude of the total propagation delay depending on the absolute and relative delay of the semiconductor element delays and the wire delays makes it very difficult to control the output times of digital signals both within each pin group and between different pin groups.

There is therefore a need for a system and method for ensuring that digital signals are transmitted from electronic devices, such as memory devices, memory controllers, testing systems and the like, with precisely controlled timing.

SUMMARY OF THE INVENTION

A system for controlling the timing at which a signal is transmitted includes a first signal distribution tree having a plurality of branches. A first digital input signal is applied to an input node of the tree and is coupled through the respective branches to a plurality of respective first output nodes. The first signal distribution tree further includes a phase interpolator through which the digital input signal is coupled between the input node and each of the first output nodes. At least one branch through which the input signal is coupled to at least one of the first output nodes includes a delay line coupled in series with the phase interpolator. A second signal distribution tree also has a plurality of branches. A second digital input signal is applied to an input node of the second signal distribution tree, and is coupled through the respective branches to a plurality of respective second output nodes. The second signal distribution tree also includes a delay line in at least one of its branches from the input node to a respective one of the second output nodes. The delay lines in the first signal distribution tree may be adjusted so that the input signal is coupled through the first signal distribution tree from the input node to each of the first output nodes at substantially the same time. Similarly, the delay lines in the second signal distribution tree may be adjusted so that the input signal is coupled through the second signal distribution tree from the input node to each of the second output nodes at substantially the same time. The delay of the phase interpolator may be adjusted to provide a predetermined timing relationship between the coupling of the input signal to the first output nodes and the coupling of the input signal to the second output nodes.

The first and second clock distribution trees may be used in a memory controller or memory device to control the timing of data signals coupled between the memory controller and memory device relative to each other as well as relative to the timing of a data strobe signal coupled between the memory controller and memory device.

The first and second clock distribution trees may also be used in an integrated circuit tester to apply test signals to an integrated circuit having a range of timing relationships relative to a strobe signal that is also applied to the integrated circuit being tested. The tester then receives results signals from the integrated circuit being tested, which indicate the performance of the integrated circuit to various timing relationships in the range.

BRIEF DESCRIPTION OF THE DRAWINGS

FIG. 1 is a logic diagram showing a pair of clock trees conventionally used to output digital signals in a variety of electronic devices.

FIG. 2 is a graph showing the relationship between propagation delay and temperature in the clock trees of FIG. 1.

FIG. 3 is a logic diagram showing a clock tree according to one example of the invention for controlling the timing of DQ signals in one pin group relative to each other and to signals in a DQS pin group.

FIG. 4 is a clock signal path according to another example of the invention that can be used to generate a delay that varies in a selected manner as a function of process, supply voltage and temperature variations.

FIG. 5 is a graph showing two examples of delay functions that can be achieved using the clock signal path of FIG. 4.

FIG. 6 is a block diagram of a testing system according to one example of the invention being used to test a memory device.

FIG. 7 is a logic diagram of one example of a write data signal generator that may be used in the testing system of FIG. 6.

FIG. 8 is a block diagram of a computer system having a memory controller and a memory device, both of which use a data signal generator according to one example of the invention.

DETAILED DESCRIPTION

A system and method for controlling the output times of digital signals according to one example of the invention is shown in FIG. 3. An internal clock signal ICLK is coupled through a first clock tree 40 to the clock input of a large number of latches 44, although only one latch 44 and one branch of the tree 40 is shown in FIG. 3 in the interests of clarity. In the example of FIG. 3, the latches 44 each output a respective bit of a write data signal DQ. The tree 40 is considered to be heavily loaded in that the ICLK signal is coupled to the large number of latches 44 through a large number of branches, and hence a large number of buffers 48.

The ICLK signal is also coupled through a second clock tree 50 to a pair of latches, although only one latch 52 is shown for purposes of clarity. The second clock tree generates complementary DQS and DQS* signals. The tree 50 is considered to be lightly loaded because the ICLK signal is coupled to the latches 52 through only two branches, each containing a single buffer 56. As a result of the different loadings of the clock trees 40, 50, as well as differences in the number of buffers and lengths of conductors in the trees 40, 50, it is difficult to control the timing of the DQ signals relative to each other and to the DQS signal. According to one example of the invention, this difficulty is largely overcome by offsetting the phase of the ICLK signal using a respective phase interpolator 60, 62 in each of the trees 40, 50. As is well-known in the art, a phase interpolator is able to interpolate between the phase of two input signals by a precisely controlled amount. The phase interpolators 60, 62 each receive the ICLK signal and its complement ICLK*. The degree of precision of the delay of a signal generated by a phase interpolator depends on the precision of the input signal frequency. The phase interpolators 60, 62 can, for example, interpolate between the phases of the ICLK and ICLK* signals in 100 increments. If the ICLK and ICLK* signals have a frequency of 5 mHz, the phase offset of the signals output from the phase interpolators 60, 62 can then be adjusted in 1 nanosecond increments. The degree of precision of the delay depends on the precision of the input signal frequency, and is thus not adversely affected by process, supply voltage or temperature variations.

The outputs of the interpolators 60, 62 are coupled through branches of the respective clock trees 40, 50, each of which contains a buffer, although only one buffer 64, 66 is shown in each branch in FIG. 3. At least some of the branches in the clock trees 40, 50 contain respective delay lines 68, 70. As is well known in the art, the delay provided by the delay lines 68, 70 varies with process, supply voltage, and temperature variations. The delay provided by a delay line in one branch of a tree can therefore track semiconductor element delays in another branch of the same or a different tree.

In operation, the phase interpolators 64, 66 are adjusted to maintain a predetermined timing relationship between the signals in the DQ signal pin group and the signal in the DQS pin group. The phase interpolators 60, 62 primarily compensate for differences in the wire delays in the clock trees 40, 50. Like the wire delays, the delays of the phase interpolators 60, 62 are relatively insensitive to process, supply voltage or temperature variations. The delay lines 68, 70 primarily compensate for differences in the delays in each branch of the clock trees 40, 50 so that the DQ signals all transition at the same time, and both of the DQS signals transition at the same time. The delay lines 68, 70 are sensitive to process, supply voltage or temperature variations, and they roughly therefore track propagation delay variations of the buffers 48, 56, 64, 66 and other semiconductor circuit elements resulting from those same factors.

Another advantage to placing a delay line in series with a phase interpolator in the clock trees is that it is possible to control the slope and magnitude of the delay as a function of processing, supply voltage, and temperature variations. For example, with reference to FIG. 4, an ICLK signal is applied to the series combination of a phase interpolator 74 and a delay line 76, each of which provides a controllable delay. As mentioned above, the phase interpolator 74 is relatively insensitive to process, voltage and temperature variations. On the other hand, the delay line 76 is sensitive to process, voltage and temperature variations. In a first example shown in FIG. 5, the delay of the delay line 76 (and any semiconductor circuit element in series with the delay line 76) is 5 ns at a relatively low temperature and doubles with temperature to 10 ns. In this example, the combined delay of the wire delay and phase interpolator 74 is a constant 30 ns. The total delay therefore starts at 35 ns and increases with temperature to 40 ns. In a second example shown in FIG. 5 as a solid line, the delay of the delay line 76 is 30 ns at a relatively low temperature and again doubles with temperature to 60 ns. In this example, the delay of the wire delay and phase interpolator 74 is a constant 5 ns. The total delay therefore starts at 35 ns and increases with temperature to 65 ns. Therefore, the slope of the total delay in the second example is substantially greater than the slope of the total delay in the first example. By combining the phase interpolator 74 in series with the delay line 76 in this manner, a delay vs. temperature relationship having virtually any magnitude and slope can be created.

The system and method for controlling the timing at which digital signals are output can also be used to test the timing margins of digital circuits, such as memory devices. For example, two memory device timing parameters that are normally tested are the maximum data set-up time, which is abbreviated as t_DS, and the minimum data hold time, which is abbreviated as t_DH. As mentioned above, in source synchronous data transfers, write data signals DQ are transmitted in synchronism with a data strobe signal DQS. The maximum time needed for the write data signals DQ to become valid after the transition of DQS, i.e., the data set up time t_DS, is normally specified for a memory device. Similarly, the minimum time that the write data signals DQ must remain valid after the transition of DQS, i.e., the data hold time t_DH, is also normally specified for a memory device.

The time between t_DS and t_DH is the data valid period. The length of the data valid period may be excessively reduced by any increase in the set-up time beyond the specified maximum set-up time t_DS or any decrease of the data hold time from the specified minimum data hold time t_DH. As the length of the data hold period gets smaller, it becomes more difficult for the memory device to position transitions of the DQS signal in the data valid period. It is therefore important to determine the data set-up and data hold times of a memory device being tested to ensure that a sufficient data valid period can be achieved.

A memory device can be tested to determine the values of t_DS and T_DH, as well as other timing parameters, by varying the timing relationship between the write data signals DQ and the data strobe signal DQS, and determining which relationships allow the DQ signals to be written to the memory device. For example, with reference to FIG. 6, a testing system 80 is coupled to a device under test, which, in this example, is a memory device 84, such as a DRAM device. The testing system 80 generates and provides to the memory device 84 memory commands, memory addresses, and write data, and it receives read data from the memory device 84. The testing system 80 also generates and provides to the memory device 84 a data strobe DQS signal. The testing system 80 also includes circuitry 86 for providing write data signals with precisely controlled timing. Also included in the testing system 80 is extensive circuitry of conventional design, which is not shown and will not be explained for purposes of brevity and clarity.

One example of the write data signal generating circuitry 86 is shown in FIG. 7. The circuitry 86 includes a phase interpolator 90, which receives and delays an ICLK signal by a selected delay amount. The delayed ICLK signal is then distributed through a clock tree represented by a buffer 92 to four delay lines 94, 96, 98, 100. The delay provided by each of the delay lines 94-100 can be precisely controlled. The outputs of the delay lines 94-100 are applied through respective buffers 106, 108, 100, 112 to the clock inputs of respective latches 120, 122, 124, 126, each of which receive a respective data signal at their data input. The latches 120-126 output four write data signals DQ. The delays of the phase interpolator 90 and the delay lines 94-100 are controlled by a control circuitry 128.

In operation, the delay lines 94-100 are adjusted so that the DQ signals are output from all of the latches 120-126 at the same time. The phase interpolator 90 is then adjusted to vary the delay time of the phase interpolator 90 over a range of delay values. The timing of a data strobe signal DQS (FIG. 6) is maintained constant so that the timing relationship between the DQ signals and the DQS signal is varied. As each delay value, the ability of a memory device to capture the DQ signals is tested, such as by conducting a read after each attempted write. The timing margins of the memory devices, such as the maximum set-up time t_DS and the data hold time t_DH, can then be determined.

As mentioned above, the system and method for controlling the output times of digital signals shown in FIG. 3 can be used to output write data signals from a memory controller and to output read data signals from a memory device, such as a DRAM device. With reference to FIG. 8, a computer system 200 is shown that can take advantage of various examples of the present invention. The computer system 200 includes a processor 202 for performing various functions, such as performing specific calculations or tasks. In addition, the computer system 200 includes one or more input devices 204, such as a keyboard or a mouse, coupled to the processor 202 through a memory controller 206 and a processor bus 208 to allow an operator to interface with the computer system 200. Typically, the computer system 200 also includes one or more output devices 210 coupled to the processor 202, such output devices typically being a printer or a video terminal. One or more data storage devices 212 are also typically coupled to the processor 202 through the memory controller 206 to store data or retrieve data from external storage media (not shown). Examples of typical data storage devices 212 include hard and floppy disks, tape cassettes, and compact disk read-only memories (CD-ROMs).

The computer system 200 also includes a DRAM device 220 that is coupled to the memory controller 206 through a control bus 222, an address bus 224 and a data bus 230. The memory controller 206 includes a write data output circuit 234 similar to the circuit shown in FIG. 3. The write data output circuit 234 is operable to apply write data signals and at least one write signal to the data bus 230 at precisely controlled times. Similarly, the DRAM device 220 also includes a read data output circuit 238 that is operable to apply read data signals and at least one read strobe signal to the data bus 230 at precisely controlled times. For this reason, the DRAM device 220 and memory controller 206 are able to operate at very high speeds without the need to design either the memory controller 206 or the DRAM device 220 with precisely controlled signal propagation times. A configuration register 240 in the memory controller 206 and a mode register 244 in the DRAM device 220 may be programmed to select the delays of the phase interpolators and delay lines used in the write data output circuit 234 and the read data output circuit 238, respectively. The computer system 200 may also include a cache memory 248 coupled to the processor 202 through the processor bus 208 to provide for the rapid storage and reading of data and/or instructions, as is well known in the art.

Although the present invention has been described with reference to the disclosed examples, persons skilled in the art will recognize that changes may be made in form and detail without departing from the spirit and scope of the invention. For example, although clock trees using various examples of the present invention generally have a phase interpolator in each tree and a delay line in each branch, it will be understood that it may be possible to omit a phase interpolator in one or more clock trees and use a phase interpolator in another tree to match the timing in the clock tree with the omitted phase interpolator. Similarly, it may be possible to omit a delay line in one or more branch of a clock tree and use a delay line in another branch to match the timing in the branch with the omitted delay line. Such modifications are well within the skill of those ordinarily skilled in the art. Accordingly, the invention is not limited except as by the appended claims.

Claims

1. An apparatus, comprising: a plurality of latches;a signal distribution tree including a plurality of branches and configured to provide a signal to the plurality of latches over the plurality of branches; anda phase interpolator coupled to a latch of the plurality of latches through a branch of the plurality of branches, the branch of the plurality of branches including a delay line configured to provide a delay to the signal.

2. The apparatus of claim 1, wherein the delay is a first delay and the delay line is a first delay line, and wherein a second branch of the plurality of branches includes a second delay line configured to provide a second delay to the signal.

3. The apparatus of claim 1, wherein the delay is based, at least in part, on a process variation, a voltage variation, a temperature variation, or a combination thereof.

4. The apparatus of claim 2, wherein the second delay line is further configured to track a delay in another branch of the plurality of branches.

5. The apparatus of claim 1, wherein the plurality of latches comprises a first plurality of latches, the plurality of branches comprises a first plurality of branches, and the signal distribution tree comprises a first signal distribution tree, the apparatus further comprising: a second plurality of latches; anda second signal distribution tree having a second plurality of branches and configured to provide the signal to the second plurality of latches over the second plurality of branches.

6. The apparatus of claim 5, wherein the first plurality of latches comprises a first number of latches and the second plurality of latches comprises a second number of latches, the first number different than the second number.

7. The apparatus of claim 5, wherein the phase interpolator comprises a first phase interpolator and wherein a branch of the second plurality of branches includes a second phase interpolator.

8. The apparatus of claim 7, wherein the delay line comprises a first delay line and wherein the second phase interpolator includes a second delay line.

9. The apparatus of claim 5, wherein the second signal distribution tree is configured to provide complementary clock signals.

10. The apparatus of claim 1, wherein the branch of the plurality of branches comprises a buffer.

11. The apparatus of claim 1, further comprising: a control circuit coupled to the phase interpolator and configured to control the delay line.

12. The apparatus of claim 1, wherein the phase interpolator is coupled to at least one other latch of the plurality of latches through at least one other branch of the plurality of branches.

13. The apparatus of claim 12, wherein each branch of the plurality of branches includes a respective delay line.

14. The apparatus of claim 1, wherein the apparatus comprises a write data signal generating circuitry.

15. The apparatus of claim 1, wherein the apparatus comprises a write data output circuit.

16. The apparatus of claim 1, wherein the apparatus comprises a read data output circuit.

17. The apparatus of claim 1, wherein the apparatus comprises a memory device.

18. The apparatus of claim 1, wherein the apparatus comprises a testing system.

19. The apparatus of claim 1, wherein the apparatus comprises a memory controller.

20. The apparatus of claim 1, wherein the apparatus comprises a computer system.

21. An apparatus comprising: a number of latches; anda clock tree configured to couple a signal to the number of latches, wherein a phase interpolator is coupled through a branch of the clock tree containing a delay line to a latch of the number of latches.

22. The apparatus of claim 21, wherein the number of latches comprises a number of first latches and the clock tree comprises a first clock tree, the apparatus further comprising a second clock tree configured to couple the signal to a number of second latches.

23. The apparatus of claim 22, wherein a second phase interpolator is coupled through a branch of the second clock tree to a latch of the number of second latches.

24. The apparatus of claim 23, wherein the branch of the second clock tree contains a second delay line.

25. The apparatus of claim 23, wherein the branch of the first clock tree contains a first buffer and the branch of the second clock tree contains a second buffer.

26. The apparatus of claim 22, wherein the number of first latches is different than the number of second latches.

27. The apparatus of claim 22, wherein each of the number of first latches is configured to output a respective bit of a data signal.

28. The apparatus of claim 27, wherein the second clock tree is configured to generate complementary strobe signals.

29. The apparatus of claim 21, wherein a phase interpolator being coupled through a branch of the clock tree to a latch of the number of latches comprises an output of the phase interpolator being coupled through a number of branches of the clock tree to the number of latches.

30. The apparatus of claim 29, wherein each of the number of branches contains a respective delay line.

31. The apparatus of claim 21, further comprising control circuitry configured to control a delay of the phase interpolator.

32. The apparatus of claim 31, wherein the control circuitry is further configured to control the delay of the delay line.

33. A method, comprising: providing a signal to a signal distribution tree including a phase interpolator and a delay line;applying a first delay to the signal with a phase interpolator to provide a first delayed signal;applying a second delay to the first delayed signal with a delay line to provide a second delayed signal; andproviding the second delayed signal to a latch over a branch of the signal distribution tree.

34. The method of claim 33, wherein the signal comprises a first signal and applying a first delay to the signal with a phase interpolator comprises maintaining a timing relationship between the first signal and a second signal.

35. The method of claim 33, wherein the signal distribution tree comprises a first signal distribution tree, the method further comprising: providing the signal to a second signal distribution tree.

36. The method of claim 35, wherein the phase interpolator comprises a first phase interpolator and applying a first delay to the signal with a phase interpolator comprises applying the first delay to the signal with the first phase interpolator in the first signal distribution tree, the method further comprising applying a third delay to the signal with a second phase interpolator in the second signal distribution tree.

37. The method of claim 33, wherein applying a first delay to the signal with a phase interpolator comprises applying a first delay to the signal, the first delay based, at least in part, on a propagation delay.

38. The method of claim 33, further comprising: providing a bit of data from the latch.

39. The method of claim 33, further comprising: buffering the signal.