HT36M4
Functional Description
Execution Flow
The system clock for the HT36M4 is derived from either
a crystal or an RC oscillator. The oscillator frequency di-
vided by 2 is the system clock for the MCU
(f
OSC
=2´f
SYS
) and it is internally divided into four
non-overlapping clocks. One instruction cycle consists
of four system clock cycles.
Instruction fetching and execution are pipelined in such
a way that a fetch takes one instruction cycle while de-
coding and execution takes the next instruction cycle.
However, the pipelining scheme causes each instruc-
tion to effectively execute in one cycle. If an instruction
changes the program counter, two cycles are required
to complete the instruction.
Program Counter
-
PC
The 13-bit program counter (PC) controls the sequence
in which the instructions stored in program ROM are ex-
ecuted and its contents specify a maximum of 8192 ad-
dresses for each bank.
After accessing a program memory word to fetch an in-
struction code, the contents of the program counter are
incremented by one. The program counter then points to
the memory word containing the next instruction code.
When executing a jump instruction, conditional skip ex-
ecution, loading PCL register, subroutine call, initial re-
set, internal interrupt, external interrupt or return from
subroutine, the PC manipulates the program transfer by
loading the address corresponding to each instruction.
The conditional skip is activated by instruction. Once the
condition is met, the next instruction, fetched during the
current instruction execution, is discarded and a dummy
cycle replaces it to retrieve the proper instruction. Other-
wise proceed with the next instruction.
The lower byte of the program counter (PCL) is a read-
able and writeable register (06H). Moving data into the
PCL performs a short jump. The destination will be
within 256 locations.
Once a control transfer takes place, an additional
dummy cycle is required.
T 1
T 2
T 3
T 4
T 1
T 2
T 3
T 4
S y s te m
C lo c k o f M C U
(f
S Y S
= f
O S C
/2 )
P C
T 1
T 2
T 3
T 4
P C
P C + 1
P C + 2
F e tc h IN S T (P C )
E x e c u te IN S T (P C -1 )
F e tc h IN S T (P C + 1 )
E x e c u te IN S T (P C )
F e tc h IN S T (P C + 2 )
E x e c u te IN S T (P C + 1 )
Execution Flow
Program Counter
*15~*13
000
000
000
*12
0
0
0
*11
0
0
0
*10
0
0
0
*9
0
0
0
*8
0
0
0
*7
0
0
0
*6
0
0
0
*5
0
0
0
*4
0
0
0
*3
0
1
1
*2
0
0
1
*1
0
0
0
*0
0
0
0
Mode
Initial Reset
Timer/Event Counter 0
Overflow
Timer/Event Counter 1
Overflow
Skip
Loading PCL
Jump, Call Branch
Return From Subroutine
Program Counter+2
PF2~
PF0
PF2~
PF0
S15~
S13
*12
*11
*10
*9
#9
S9
*8
#8
S8
@7
#7
S7
@6
#6
S6
@5
#5
S5
@4
#4
S4
@3
#3
S3
@2
#2
S2
@1
#1
S1
@0
#0
S0
#12 #11 #10
S12 S11 S10
Program Counter
Note:
PF2~PF0: Bits of Bank Register
*12~*0: Bits of Program Counter
@7~@0: Bits of PCL
Rev. 1.10
5
March 14, 2007
#12~#0: Bits of Instruction Code
S15~S0: Bits of Stack Register
HOLTEK [ HOLTEK SEMICONDUCTOR INC ]
