Processor Basics 107
is avoided. If this automatic disabling had been avoided, then the processor would continue branching to
the same ISR repeatedly as long as the interrupting signal is active. To allow further branching for the same
interrupt, it is customary to enable the interrupt at the last stage of execution of its ISR (before executing
the RETURN instruction). Note that before execution of this enabling instruction, the ISR must ensure that
the original interrupting signal is no longer existent.
Masking also prevents the processor from reacting to any interrupt. However, in this case the pro-
cessor may have knowledge of the existence of any pending interrupting signal without any obligation
for branching to its service routine. Some processors offer a non-maskable interrupt ( NMI ), which
may never be disabled or masked. These non-maskable interrupts are necessary to react against some
extreme emergency condition, e.g., power failure (or a virus attack!). At this stage the reader must note
the difference between disabling an interrupt and masking an interrupt. In the rst case, the processor
would never know about the existence of any eventual interrupting signal, while in the second case it
can trace it. However, in either cases, no automatic branching to the ISR would be allowed.
Whatever we have discussed so far, related to processor basics, is very abstract in nature. If we study
any particular processor, we shall encounter many speci c details pertaining to that processor. In the fol-
lowing sections, we shall discuss about three processors, namely Intel 8085, Intel 8086 and Intel 8051.
Subsequently, in the advanced processor section, we shall discuss about Intel 80386 and Pentium 4.
5.7 INTEL 8085 MICROPROCESSOR
Intel 8085 was introduced around 1976 as an improvement of its preceeder, Intel 8080. This 8-bit processor
(8085) contained 6,500 transistors within a 40 pin dual in-line package (DIP ) and operated with a single
power supply of +5 volts. It worked with a maximum clock speed of 5 MHz having 8 data and 16 address
lines, capable of addressing 64 kB of memory locations (bytes) directly. Designed around Princeton archi-
tecture, 8085 does not distinguish between program and data memory. Interfaced with two more devices,
8155 and 8755, it formed a minimum working system. The IC 8755 contained 2K EPROM and two 8-bit
I/O ports. The other device, IC 8155, contained 256-bytes of RAM, three ports and a 14-bit Timer.
5.7.1 Pins and External Signals
Figure 5.12 (a) presents the pin diagram of 8085 with all of its signal nomenclature. From the designer’s
view point, the device may be taken as in Figure 5.12 (b). To make 8085 functional, minimum input
required would be Vcc, Vss, X1, X2, RESET IN, HOLD, READY and TRAP. X1 and X2 are input pins
for an external quartz crystal. Its prescribed frequency is 6.144 MHz, which is divided by 2 internally
by 8085 to generate its system clock of 3.072 MHz frequency. The RESET IN input must be low during
power on to generate a system reset. HOLD input, generally used for a DMA request, must be grounded
along with TRAP input – the non-maskable interrupt. READY input must be pulled high to make 8085
functional.
Lower 8 address signals of 8085 (A0 – A7) are multiplexed with its data bus (D0 – D7), forming the
multiplexed lower address data bus, AD0 – AD7. The falling edge of the address latch enable (ALE)
signal may be used to de-multiplex the lower address signals (A0 – A7), using an external octal latch,
e.g., 74373, if it is required by the system ( Figure 5.13 ).
5.7.2 Internal Architecture
Intel 8085 belonged to the group of earlier generation of microprocessor, when pipeline architecture
was not in vogue. Pipeline architecture was rst introduced by Intel in its rst 16-bit processor 8086,
M05_GHOS1557_01_SE_C05.indd 107M05_GHOS1557_01_SE_C05.indd 107 4/29/11 5:06 PM4/29/11 5:06 PM