Apparatus for Delaying the Output of Data Onto a System Bus

The present application relates in general to data processing systems, and in particular, to a circuit for allowing a processor to communicate data to a bus device operating at a slower speed.

High performance microprocessors ("CPUs") run at higher internal clock speeds than are used on external buses utilized to couple the processor to other bus devices, such as ASICs (application specific integrated circuits), external memory devices, I/O (input/output) devices, etc.

In typical implementations, data is launched onto the external bus by the processor on the active edge of the external clock (system or bus clock speed). Slower external bus devices may miss this data because internal delay from their clock pins to their internal latches can be longer than the delay from their data pins to the internal latch data pins. This is often true because incoming data to a bus device may only be sent to a few places within the device, while clock signals are distributed throughout the chip containing the bus device.

Thus, the problem is that incoming data to a slower bus device needs to be held valid long enough for a clock signal to arrive at that data stage within the bus device. This discrepancy has traditionally been handled by specifying a hold time between the clock edge and the data change edge, so that the clock delays in the slower external portions do not cause them to miss data launched on the bus by the processor. This hold time is often implemented with buffers for the data to take into account the inherent buffering provided to the clock signal within the external bus device.

The problem with such a solution is that as microprocessors are designed to be faster and faster, it becomes increasingly difficult to design a delay that is a substantial portion of the total clock time and is consistent across process variations. When board designers of external bus devices are requesting hold times on the order of three nanoseconds and microprocessors are being designed with gate delays in the 150picosecond range, designing a delay line that is an order of magnitude as long as a nominal gate delay is problematic. Furthermore, as the design of processors involves the "tweaking" of the internal speed of the processors to faster and faster speeds, these imbedded delays within the bus devices eventually become insufficient. Moreover, these imbedded delays within the clock devices may cost more silicon to implement.

Thus, what is needed in the art is a process independent delay that provides the requisite hold time for the data sent to a slower external bus device from a processor.

The foregoing need is satisfied by the present invention, which implements a delay that is a ratio of an internal clock speed within the processor to the external clock speed on the external bus coupling the processor to external bus devices. This is accomplished by latching the data to be launched onto the external bus with a latch circuit that is timed with a clock signal that is faster than the external bus clock. This provides a delay in the launch of the data.

One advantage of the present invention is that it does not require a delay line that may be a difficult portion to design and requires more area (silicon) on a chip.

Another advantage of the present invention is that the hold time is based on a clock instead of a process dependent delay. This means that the hold time specification is the same for a fast external device as a slow external device, allowing speed sorting of devices without respecifying the hold time.

Yet another advantage of the present invention is that it allows for a design that can be ported to a faster technology without redesigning the bus interface to maintain the hold time.

The foregoing has outlined rather broadly the features and technical advantages of the present invention in order that the detailed description of the invention that follows may be better understood. Additional features and advantages of the invention will be described hereinafter which form the subject of the claims of the invention.

For a more complete understanding of the present disclosure, and the advantages thereof, reference is now made to the following descriptions taken in conjunction with the accompanying drawings, in which:

FIG. 1 illustrates a data processing system configured in accordance with the present disclosure;

FIG. 2 illustrates a more detailed diagram of a processor configured in accordance with the present disclosure; and

FIG. 3 illustrates a timing diagram illustrating an embodiment of the present disclosure.

In the following description, numerous specific details are set forth such as specific word or byte lengths, etc. to provide a thorough understanding of the present disclosure. However, it will be obvious to those skilled in the art that the present disclosure may be practiced without such specific details. In other instances, well-known circuits have been shown in block diagram form in order not to obscure the disclosure in unnecessary detail. For the most part, details concerning timing considerations and the like have been omitted inasmuch as such details are not necessary to obtain a complete understanding of the present disclosure and are within the skills of persons of ordinary skill in the relevant art.

Refer now to the drawings wherein depicted elements are not necessarily shown to scale and wherein like or similar elements are designated by the same reference numeral through the several views.

A representative hardware environment for practicing the present invention is depicted in FIG. 1, which illustrates a typical hardware configuration of workstation 113 having central processing unit 110, such as a conventional microprocessor, and a number of other units interconnected via system bus 112. Workstation 113 includes random access memory (RAM) 114, read only memory (ROM) 116, and input/output (I/O) adapter 118 for connecting peripheral devices such as disk units 120 and tape drives 140 to bus 112, user interface adapter 122 for connecting keyboard 124, mouse 126, speaker 128, microphone 132, and/or other user interface devices such as a touch screen device (not shown) to bus 112, communication adapter 134 for connecting the workstation to a data processing network, ASIC 139, and display adapter 136 for connecting bus 112 to display device 138.

Referring next to FIG. 2, there is illustrated a more detailed diagram of CPU (processor) 110. CPU 110 includes PLL 201, divider circuit (DIV) 202, and latches 203 and 204, which are further described below. CPU 110 may include other circuitry (not shown for clarity) typically found in a microprocessor. A result of processing within this logic circuitry produces DATA that CPU 110 then desires to launch onto bus 112, which is also referred to as the system bus or external bus. This DATA is intended to be sent to one or more bus devices, such as I/O devices, memory devices, ASIC 139, etc. CPU 110 may be implemented on a single integrated circuit.

CPU 110 includes phase lock loop ("PLL") 201, which is well-known in the art and common to microprocessors, for receiving an input frequency (from, for example, a system clock device (not shown)) and producing output frequencies that are multiples of the input frequency. These multiple frequencies may be 2.times., 3.times., 4.times., etc. of the input frequency, and are utilized as the internal clock(s) within CPU 110.

The output of PLL 201, which is a multiple of the system clock, is divided by divider circuit 202 to produce the System Clock. The Delay Clock indicated in FIG. 2 is equal to one of the internal clock frequencies within CPU 110 and is thus faster than the System Clock.

The System Clock is provided to the clock input of latch circuit 203, while the Delay Clock signal is provided to the clock input of latch circuit 204.

In typical designs, DATA produced within the internal circuitry of CPU 110 to be launched onto bus 112 is latched by latch 203 coupled to the System Clock. An additional latch 204 is implemented that receives the output of latch 203 and is clocked by the Delay Clock signal. The result of this configuration is that the launching of the DATA is delayed, wherein the delay is a ratio of the Delay Clock signal to the System Clock signal. For example, if CPU 110 is running with an internal clock (Delay Clock) of 66 MHz and an external or System Clock of 33 MHz, the launch of the DATA is then delayed by 1/2 of the internal clock, providing a 7.5 nanosecond hold time. At different internal/external clock speeds, a different number of 1/2 clock delays can be used. This is illustrated by the timing diagram in FIG. 3, wherein the Delay Clock signal is twice as fast as the System Clock signal. Line A corresponds to label A in FIG. 2 while line B corresponds to label B in FIG. 2.

Latch 204 can be designed so that the DATA is launched onto external bus 112 on a rising or falling edge of the Delay Clock signal. In this example, the DATA is launched 1/2 cycle after it would have originally been launched from latch 203 if latch 204 was not present. This delay in the launch of the DATA provides the requisite delay for the external bus devices and is process independent.

It should be understood that various changes, substitutions and alterations can be made herein without departing from the spirit and scope of the invention as defined by the appended claims.