Trailing-Edge
-
PDP-10 Archives
-
LCG_Integration_Tools_Clearinghouse_T20_v7_30Apr86
-
tools/decnet10/mlib09.mac
There are 4 other files named mlib09.mac in the archive. Click here to see a list.
TITLE MACLIB -- MACRO SUBROUTINE LIBRARY
SUBTTL SECTION 9. -- NUMERICAL DATA CONVERSION
;
;+
; =================================================================
;
; MAPC DECSYSTEM-10 MACRO SUBROUTINE LIBRARY
;
; Developed by R. W. Stamerjohn, DEC-10 Systems Group
;
; These routines and all related documentation were developed at
; Monsanto Agricultural Products Company, St. Louis, Mo. 63167.
;
; =================================================================
;
; Use the following DECsystem-10 Monitor command to include
; the MACLIB subroutine library as part of a user program:
;
; .EX YOURS,REL:MACLIB/SEARCH
;
; If you do not include the /SEARCH switch, you will load
; the entire library as part of your program.
;
;
;-
;
COMMENT %
;+
List of Routines in This Section:
--------------------------------
C16T36 Format-Driven PDP-11 to DEC-10 Binary Conversion
C36T16 Format-Driven DEC-10 to PDP-11 Binary Conversion
C32T36 Format-Driven VAX-11 to DEC-10 Binary Conversion
C36T32 Format-Driven DEC-10 to VAX-11 Binary Conversion
ASCV16/32 ASCII Conversion
C8CV16/32 C*8 Complex Conversion (Single-Precision)
I1CV16/32 I*1 Signed Integer Conversion (8-bit)
I2CV16/32 I*2 Signed Integer Conversion (16-bit)
I4CV16/32 I*4 Signed Integer Conversion (32-bit)
IFCV16/32 I*F Signed Integer Conversion (32-bit)
L1CV16/32 L*1 Logical Conversion (8-bit)
L2CV16/32 L*2 Logical Conversion (16-bit)
L4CV16/32 L*4 Logical Conversion (32-bit)
M1CV16/32 M*1 Unsigned Integer Conversion (8-bit)
M2CV16/32 M*2 Unsigned Integer Conversion (16-bit)
M4CV16/32 M*4 Unsigned Integer Conversion (32-bit)
MFCV16/32 M*F Unsigned Integer Conversion (32-bit)
R4CV16/32 R*4 Floating Conversion (Single-Precision)
R8CV16/32 R*8 Floating Conversion (Double-Precision)
;-
Module Revision History
------------------------
END COMMENT %
PRGEND
TITLE C16T36
SUBTTL Format-Driven PDP-11 to DEC-10 Conversion
;
; Author: R. W. Stamerjohn, MAPC DEC-10 Systems Group
; Written: 05-Apr-83
;
;+
; Purpose: This subroutine provides format-driven conversion
; of PDP-11 format numerical data to DECsystem-10
; notation.
;
; Usage: CALL C16T36(array1,size1,array2,size2,format,cnt,error)
;
; Arguments: array1 = Array containing PDP-11 format data
; size1 = Size of "array1" in PDP-11 bytes
; array2 = Array to receive results of connversion
; size2 = Size of "array2" in DECsystem-10 words
; format = Alphanumeric format string
; cnt = Character count of "format" (-1=ASCIZ)
; error = Integer error return variable
;
;-
; Register Definition:
;
T0=0 ;Function return (low)
T1=1 ;Function return (high)
TP=2 ;Temporary AC
A1=3 ;Current array1 address
A2=4 ;Current array2 address
S1=5 ;Size (bytes) of array 1
S2=6 ;Size (words) of array 2
IX=7 ;Index to conversion tables
RP=10 ;Repeat count
FT=11 ;Format string address
FC=12 ;Formate string character count
AX=13 ;Array byte index
EA=15 ;Error return variable address
A=16 ;Fortran argument block pointer
P=17 ;Push-down stack pointer
ENTRY C16T36
C16T36: PUSHJ P,ARGS..## ;Setup accumulators
POPJ P, ;Exit if no return variable
PXCV.1: PUSHJ P,FORM..## ;Get next formatted argument
POPJ P, ;Exit if done or error
CAIN IX,ASCIDX ;Was type "AS"
JRST PXCVAS ;Yes, convert ASCII
CAIN IX,SKIIDX ;Was type "SI"
JRST PXCVSI ;Yes, skip input bytes
CAIN IX,SKOIDX ;Was type "SO"
JRST PXCVSO ;Yes, skip output words
PXCV.2: MOVEI TP,BINDSP ;Get binary dispatch table
ADD TP,IX ;For dispatch address
MOVEI A,ARGLST ;Get function argument list
PUSHJ P,@(TP) ;Call conversion function
HRRZ TP,SIZTBL(IX) ;Get number of arguments to store
SOJL S2,PXCVE3 ;Jump if no space
MOVEM T0,(A2) ;Store return value
ADDI A2,1 ;Bump array address
SOJE TP,PXCV.3 ;Jump if no second argument
SOJL S2,PXCVE3 ;Jump if no space
MOVEM T1,(A2) ;Store return value
ADDI A2,1 ;Bump array address
PXCV.3: HLRZ TP,SIZTBL(IX) ;Get size of converted field
SUB S1,TP ;Get bytes remaining in field
JUMPL S1,PXCVE2 ;Jump if no more
ADD AX,TP ;Bump array index
SOJG RP,PXCV.2 ;Loop if repeated field
JRST PXCV.1 ;Convert next format
PXCVSI: SUB S1,RP ;Get bytes remaining in field
JUMPL S1,PXCVE2 ;Jump if no more
ADD AX,RP ;Bump array index
JRST PXCV.1 ;Convert next format
PXCVSO: SUB S2,RP ;Get words remaining in field
JUMPL S2,PXCVE3 ;Jump if no more space
ADD A2,RP ;Skip space in output array
JRST PXCV.1 ;Convert next format
PXCVAS: SOJL S2,PXCVE3 ;Jump if no more space
SETZM (A2) ;Zero memory location
MOVEI IX,1 ;Initialize character position
PXAS.1: MOVEI TP,ASCV16## ;Set function routine
MOVEI A,ASCLST ;Set argument list
PUSHJ P,(TP) ;Call function routine
ADDM T0,(A2) ;Store converted character
SOJL S1,PXCVE2 ;Jump if no more bytes
AOS AX ;Bump byte position
AOS IX ;Bump character position
SOJLE RP,PXAS.2 ;Get next format when done
CAIG IX,5 ;Are we at next word
JRST PXAS.1 ;No, convert next character
AOJA A2,PXCVAS ;Begin new word
PXAS.2: AOJA A2,PXCV.1 ;Advance to next word
PXCVE3: MOVEI TP,3 ;Flag array2 overflow
MOVEM TP,(EA) ;Return error
POPJ P, ;Return to caller
PXCVE2: MOVEI TP,2 ;Flag array1 overflow
MOVEM TP,(EA) ;Return error
POPJ P, ;Return to caller
-2,,0
ARGLST: 20,,A1
00,,AX
-4,,0
ASCLST: 20,,A1
00,,AX
00,,00
00,,IX
BINDSP: I1CV16## ;I*1 conversion
I2CV16## ;I*2 conversion
I4CV16## ;I*4 conversion
IFCV16## ;I*F conversion
L1CV16## ;L*1 conversion
L2CV16## ;L*2 conversion
L4CV16## ;L*4 conversion
C8CV16## ;C*8 conversion
R4CV16## ;R*4 conversion
R8CV16## ;R*8 conversion
M1CV16## ;M*1 conversion
M2CV16## ;M*2 conversion
M4CV16## ;M*4 conversion
MFCV16## ;M*F conversion
SIZTBL: 1,,1 ;I*1 conversion
2,,1 ;I*2 conversion
4,,1 ;I*4 conversion
4,,1 ;I*F conversion
1,,1 ;L*1 conversion
2,,1 ;L*2 conversion
4,,1 ;L*4 conversion
10,,2 ;C*8 conversion
4,,1 ;R*4 conversion
10,,2 ;R*8 conversion
1,,1 ;M*1 conversion
2,,1 ;M*2 conversion
4,,1 ;M*4 conversion
4,,1 ;M*F conversion
ASCIDX=.-SIZTBL
SKIIDX=.-SIZTBL+1
SKOIDX=.-SIZTBL+2
PRGEND
TITLE C36T16
SUBTTL Format-Driven DEC-10 to PDP-11 Conversion
;
; Author: R. W. Stamerjohn, MAPC DEC-10 Systems Group
; Written: 05-Apr-83
;
;+
; Purpose: This subroutine provides format-driven conversion
; of DEC-10 format numerical data to PDP-11 notation.
;
; Usage: CALL C36T16(array1,size1,array2,size2,format,cnt,error)
;
; Arguments: array1 = Array containing DECsystem-10 format data
; size1 = Size of "array1" in DECsystem-10 words
; array2 = Array to receive results of connversion
; size2 = Size of "array2" in PDP-11 bytes
; format = Alphanumeric format string
; cnt = Character count of "format" (-1=ASCIZ)
; error = Integer error return variable
;
;-
; Register Definition:
;
T0=0 ;Function return (low)
T1=1 ;Function return (high)
TP=2 ;Temporary AC
A1=3 ;Current array1 address
A2=4 ;Current array2 address
S1=5 ;Size (words) of array 1
S2=6 ;Size (bytes) of array 2
IX=7 ;Index to conversion tables
RP=10 ;Repeat count
FT=11 ;Format string address
FC=12 ;Formate string character count
AX=13 ;Array byte index
EA=15 ;Error return variable address
A=16 ;Fortran argument block pointer
P=17 ;Push-down stack pointer
ENTRY C36T16
C36T16: PUSHJ P,ARGS..## ;Setup accumulators
POPJ P, ;Exit if no return variable
ADD A2,[POINT 8,,7] ;Form byte pointer to output array
PXFT.1: PUSHJ P,FORM..## ;Get next formatted argument
POPJ P, ;Exit if done or error
CAIN IX,ASCIDX ;Was type "AS"
JRST PXFTAS ;Yes, convert ASCII
CAIN IX,SKIIDX ;Was type "SI"
JRST PXFTSI ;Yes, skip input words
CAIN IX,SKOIDX ;Was type "SO"
JRST PXFTSO ;Yes, skip output bytes
PXFT.2: HRRZ TP,SIZTBL(IX) ;Get number of DECsystem-10 words
SOJL S1,PXFTE2 ;Jump if no more left
MOVE T0,(A1) ;Get next DECsystem-10 word
AOS A1 ;Bump DECsystem-10 pointer
SOJE TP,PXFT.3 ;Jump if only one word needed
SOJL S1,PXFTE2 ;Jump if no more left
MOVE T1,(A1) ;Get next DECsystem-10 word
AOS A1 ;Bump DECsystem-10 pointer
PXFT.3: MOVEI TP,4 ;Preset conversion error
PUSHJ P,@FTMDSP(IX) ;Jump to conversion function
PUSH P,T1 ;Store any high value
HLRE TP,SIZTBL(IX) ;Get bytes to store
JUMPG TP,PXFT.4 ;If positive, no swap
LDB T1,[POINT 16,T0,35] ;Get low word
LSH T0,-20 ;Shift high word down
DPB T1,[POINT 16,T0,19] ;Store low word as high
MOVN TP,TP ;Make count positive
PXFT.4: CAIN TP,10 ;is this a double word
MOVEI TP,4 ;Yes, work on only one now
SUB S2,TP ;Is there enough room?
JUMPL S2,[POP P,T0
JRST PXFTE3] ;No, declare error
PUSHJ P,PXFTDB ;Go deposit word
POP P,T0 ;Restore high value
HLRZ TP,SIZTBL(IX) ;Get bytes to store
CAIE TP,-10 ;Skip if high value
JRST PXFT.5 ;Check repeat count
LDB T1,[POINT 16,T0,35] ;Get low word
LSH T0,-20 ;Shift high word down
DPB T1,[POINT 16,T0,19] ;Store low word as high
MOVEI TP,4 ;Work on one word
SUB S2,TP ;Is there enough room?
JUMPL S2,PXFTE3 ;No, declare error
PUSHJ P,PXFTDB ;Go deposit word
PXFT.5: SOJG RP,PXFT.2 ;Loop on repeat count
JRST PXFT.1 ;Otherwise, get next argument
PXFTSI: SUB S1,RP ;Get words remaining if field
JUMPL S1,PXFTE3 ;Jump if no more
ADD A1,RP ;Skip pass words in array
JRST PXFT.1 ;Convert next format
PXFTSO: SUB S2,RP ;Get bytes remaining in field
JUMPL S2,PXFTE3 ;Jump if no more
ADD AX,RP ;Update byte count
JRST PXFT.1 ;Convert next format
PXFTAS: SUB S2,RP ;Is there enough room?
JUMPL S2,PXFTE3 ;Declare error if not
MOVE T0,RP ;Get repeat count
IDIVI T0,5 ;Get number of DECsystem-10 words
SKIPE T1 ;Skip if no remainder
AOS T0 ;Bump count to next word
SUB S1,T0 ;Is there enough room
JUMPL S1,PXFTE2 ;Declare error if not
ADD T0,A1 ;Form address of next word
PUSH P,T0 ;Save for later
ADD A1,[POINT 7,,] ;Form point to characters
PXAS.1: ILDB T0,A1 ;Get next character
MOVEI TP,1 ;Store one byte
PUSHJ P,PXFTDB ;Deposit character
SOJG RP,PXAS.1 ;Loop till done
POP P,A1 ;Get next DEcsystem-10 word
JRST PXFT.1 ;Get next format
PXFTDB: MOVE T1,AX ;Copy current index
ADJBP T1,A2 ;Make pointer to deposit
DPB T0,T1 ;Deposit byte
LSH T0,-10 ;Shift down next byte
AOS AX ;Bump byte index
SOJG TP,PXFTDB ;Loop till finished
POPJ P, ;Return to caller
PXFTE3: MOVEI TP,3 ;Flag array2 overflow
MOVEM TP,(EA) ;Return error
POPJ P, ;Return to caller
PXFTE2: MOVEI TP,2 ;Flag array1 overflow
MOVEM TP,(EA) ;Return error
POPJ P, ;Return to caller
PXFTIF:
PXFTI4:
PXFTI2:
PXFTI1: MOVN T1,T0 ;Copy negative of value
SKIPL T0 ;Skip if value negative
PXFTMF:
PXFTM4:
PXFTM2:
PXFTM1: MOVE T1,T0 ;Get positive value
TDNE T1,MSKTBL(IX) ;Is value too large
MOVEM TP,(EA) ;Yes, set conversion error
POPJ P, ;Return value
PXFTL4:
PXFTL2:
PXFTL1: AND T0,[400000,,000000] ;Clear all but sign bit
TLZE T0,400000 ;Test size and zero
SETO T0, ;Set to -1
POPJ P, ;Return value
PXFTC8: PUSH P,T1 ;Store imaginary value
PUSHJ P,PXFTR4 ;Convert real part
POP P,T1 ;Get imaginary value
PUSH P,T0 ;Store converted real
MOVE T0,T1 ;Copy imaginary
PUSHJ P,PXFTR4 ;Convert imaginary part
MOVE T1,T0 ;Copy converted imaginary
POP P,T0 ;Get converted real
POPJ P, ;Return value
PXFTR4: JUMPE T0,PXR4.1 ;Just return if zero
PUSH P,T0 ;Save value
SKIPG T0 ;Skip if positive
MOVN T0,T0 ;Make value positive
LDB TP,[POINT 8,T0,8] ;Get mantissa
ASH T0,-3 ;Shift down exponent
AND T0,[000037,,777777] ;Clear extraneous bits
DPB TP,[POINT 8,T0,12] ;Deposit mantissa
POP P,TP ;Get value
SKIPG TP ;Skip if positive
TLO T0,20000 ;Turn on sign bit
PXR4.1: POPJ P, ;Return value
PXFTR8: SKIPN T0 ;Skip if non-zero
JUMPE T1,PXR8.1 ;Just return if zero
PUSH P,T0 ;Save value
SKIPG T0 ;Skip if positive
DMOVN T0,T0 ;Make value positive
LDB TP,[POINT 8,T0,8] ;Get exponent
ASHC T0,-3 ;Shift down mantissa
AND T0,[000037,,777777] ;Clear extraneous bits
DPB TP,[POINT 8,T0,12] ;Deposit exponent
ASH T1,-3 ;Shift down mantissa
AND T1,[037777,,777777] ;Clear extraneous bits
POP P,TP ;Get back value
SKIPG TP ;Skip if positive
TLO T0,20000 ;Turn on sign bit
PXR8.1: POPJ P, ;Return value
FTMDSP: PXFTI1 ;I*1 conversion
PXFTI2 ;I*2 conversion
PXFTI4 ;I*4 conversion
PXFTIF ;I*F conversion
PXFTL1 ;L*1 conversion
PXFTL2 ;L*2 conversion
PXFTL4 ;L*4 conversion
PXFTC8 ;C*8 conversion
PXFTR4 ;R*4 conversion
PXFTR8 ;R*8 conversion
PXFTM1 ;M*1 conversion
PXFTM2 ;M*2 conversion
PXFTM4 ;M*4 conversion
PXFTMF ;M*F conversion
MSKTBL: 777777,,777400 ;I*1 mask
777777,,600000 ;I*2 mask
740000,,000000 ;I*4 mask
740000,,000000 ;I*F mask
777777,,777777 ;L*1 mask
777777,,777777 ;L*2 mask
777777,,777777 ;L*4 mask
777777,,777777 ;C*8 mask
777777,,777777 ;R*4 mask
777777,,777777 ;R*8 mask
777777,,777400 ;M*1 mask
777777,,600000 ;M*2 mask
740000,,000000 ;M*4 mask
740000,,000000 ;M*F mask
SIZTBL: 1,,1 ;I*1 conversion
2,,1 ;I*2 conversion
4,,1 ;I*4 conversion
-4,,1 ;I*F conversion
1,,1 ;L*1 conversion
2,,1 ;L*2 conversion
4,,1 ;L*4 conversion
-10,,2 ;C*8 conversion
-4,,1 ;R*4 conversion
-10,,2 ;R*8 conversion
1,,1 ;M*1 conversion
2,,1 ;M*2 conversion
4,,1 ;M*4 conversion
-4,,1 ;M*F conversion
ASCIDX=.-SIZTBL
SKIIDX=.-SIZTBL+1
SKOIDX=.-SIZTBL+2
PRGEND
TITLE C32T36
SUBTTL Format-Driven VAX-11 to DEC-10 Conversion
;
; Author: R. W. Stamerjohn, MAPC DEC-10 Systems Group
; Written: 05-Apr-83
;
;+
; Purpose: This subroutine provides format-driven conversion
; of VAX-11 format numerical data to DECsystem-10
; notation.
;
; Usage: CALL C32T36(array1,size1,array2,size2,format,cnt,error)
;
; Arguments: array1 = Array containing VAX-11 format data
; size1 = Size of "array1" in VAX-11 bytes
; array2 = Array to receive results of connversion
; size2 = Size of "array2" in DECsystem-10 words
; format = Alphanumeric format string
; cnt = Character count of "format" (-1=ASCIZ)
; error = Integer error return variable
;
;-
; Register Definition:
;
T0=0 ;Function return (low)
T1=1 ;Function return (high)
TP=2 ;Temporary AC
A1=3 ;Current array1 address
A2=4 ;Current array2 address
S1=5 ;Size (bytes) of array 1
S2=6 ;Size (words) of array 2
IX=7 ;Index to conversion tables
RP=10 ;Repeat count
FT=11 ;Format string address
FC=12 ;Formate string character count
AX=13 ;Array byte index
EA=15 ;Error return variable address
A=16 ;Fortran argument block pointer
P=17 ;Push-down stack pointer
ENTRY C32T36
C32T36: PUSHJ P,ARGS..## ;Setup accumulators
POPJ P, ;Exit if no return variable
PXCV.1: PUSHJ P,FORM..## ;Get next formatted argument
POPJ P, ;Exit if done or error
CAIN IX,ASCIDX ;Was type "AS"
JRST PXCVAS ;Yes, convert ASCII
CAIN IX,SKIIDX ;Was type "SI"
JRST PXCVSI ;Yes, skip input bytes
CAIN IX,SKOIDX ;Was type "SO"
JRST PXCVSO ;Yes, skip output words
PXCV.2: MOVEI TP,BINDSP ;Get binary dispatch table
ADD TP,IX ;For dispatch address
MOVEI A,ARGLST ;Get function argument list
PUSHJ P,@(TP) ;Call conversion function
HRRZ TP,SIZTBL(IX) ;Get number of arguments to store
SOJL S2,PXCVE3 ;Jump if no space
MOVEM T0,(A2) ;Store return value
ADDI A2,1 ;Bump array address
SOJE TP,PXCV.3 ;Jump if no second argument
SOJL S2,PXCVE3 ;Jump if no space
MOVEM T1,(A2) ;Store return value
ADDI A2,1 ;Bump array address
PXCV.3: HLRZ TP,SIZTBL(IX) ;Get size of converted field
SUB S1,TP ;Get bytes remaining in field
JUMPL S1,PXCVE2 ;Jump if no more
ADD AX,TP ;Bump array index
SOJG RP,PXCV.2 ;Loop if repeated field
JRST PXCV.1 ;Convert next format
PXCVSI: SUB S1,RP ;Get bytes remaining in field
JUMPL S1,PXCVE2 ;Jump if no more
ADD AX,RP ;Bump array index
JRST PXCV.1 ;Convert next format
PXCVSO: SUB S2,RP ;Get words remaining in field
JUMPL S2,PXCVE3 ;Jump if no more space
ADD A2,RP ;Skip space in output array
JRST PXCV.1 ;Convert next format
PXCVAS: SOJL S2,PXCVE3 ;Jump if no more space
SETZM (A2) ;Zero memory location
MOVEI IX,1 ;Initialize character position
PXAS.1: MOVEI TP,ASCV16## ;Set function routine
MOVEI A,ASCLST ;Set argument list
PUSHJ P,(TP) ;Call function routine
ADDM T0,(A2) ;Store converted character
SOJL S1,PXCVE2 ;Jump if no more bytes
AOS AX ;Bump byte position
AOS IX ;Bump character position
SOJLE RP,PXAS.2 ;Get next format when done
CAIG IX,5 ;Are we at next word
JRST PXAS.1 ;No, convert next character
AOJA A2,PXCVAS ;Begin new word
PXAS.2: AOJA A2,PXCV.1 ;Advance to next word
PXCVE3: MOVEI TP,3 ;Flag array2 overflow
MOVEM TP,(EA) ;Return error
POPJ P, ;Return to caller
PXCVE2: MOVEI TP,2 ;Flag array1 overflow
MOVEM TP,(EA) ;Return error
POPJ P, ;Return to caller
-2,,0
ARGLST: 20,,A1
00,,AX
-4,,0
ASCLST: 20,,A1
00,,AX
00,,00
00,,IX
BINDSP: I1CV32## ;I*1 conversion
I2CV32## ;I*2 conversion
I4CV32## ;I*4 conversion
IFCV32## ;I*F conversion
L1CV32## ;L*1 conversion
L2CV32## ;L*2 conversion
L4CV32## ;L*4 conversion
C8CV32## ;C*8 conversion
R4CV32## ;R*4 conversion
R8CV32## ;R*8 conversion
M1CV32## ;M*1 conversion
M2CV32## ;M*2 conversion
M4CV32## ;M*4 conversion
MFCV32## ;M*F conversion
SIZTBL: 1,,1 ;I*1 conversion
2,,1 ;I*2 conversion
4,,1 ;I*4 conversion
4,,1 ;I*F conversion
1,,1 ;L*1 conversion
2,,1 ;L*2 conversion
4,,1 ;L*4 conversion
10,,2 ;C*8 conversion
4,,1 ;R*4 conversion
10,,2 ;R*8 conversion
1,,1 ;M*1 conversion
2,,1 ;M*2 conversion
4,,1 ;M*4 conversion
4,,1 ;M*F conversion
ASCIDX=.-SIZTBL
SKIIDX=.-SIZTBL+1
SKOIDX=.-SIZTBL+2
PRGEND
TITLE C36T32
SUBTTL Format-Driven DEC-10 to VAX-11 Conversion
;
; Author: R. W. Stamerjohn, MAPC DEC-10 Systems Group
; Written: 05-Apr-83
;
;+
; Purpose: This subroutine provides format-driven conversion
; of DEC-10 format numerical data to VAX-11 notation.
;
; Usage: CALL C36T16(array1,size1,array2,size2,format,cnt,error)
;
; Arguments: array1 = Array containing DECsystem-10 format data
; size1 = Size of "array1" in DECsystem-10 words
; array2 = Array to receive results of connversion
; size2 = Size of "array2" in VAX-11 bytes
; format = Alphanumeric format string
; cnt = Character count of "format" (-1=ASCIZ)
; error = Integer error return variable
;
;-
; Register Definition:
;
T0=0 ;Function return (low)
T1=1 ;Function return (high)
TP=2 ;Temporary AC
A1=3 ;Current array1 address
A2=4 ;Current array2 address
S1=5 ;Size (words) of array 1
S2=6 ;Size (bytes) of array 2
IX=7 ;Index to conversion tables
RP=10 ;Repeat count
FT=11 ;Format string address
FC=12 ;Formate string character count
AX=13 ;Array byte index
EA=15 ;Error return variable address
A=16 ;Fortran argument block pointer
P=17 ;Push-down stack pointer
ENTRY C36T32
C36T32: PUSHJ P,ARGS..## ;Setup accumulators
POPJ P, ;Exit if no return variable
ADD A2,[POINT 8,,7] ;Form byte pointer to output array
PXFT.1: PUSHJ P,FORM..## ;Get next formatted argument
POPJ P, ;Exit if done or error
CAIN IX,ASCIDX ;Was type "AS"
JRST PXFTAS ;Yes, convert ASCII
CAIN IX,SKIIDX ;Was type "SI"
JRST PXFTSI ;Yes, skip input words
CAIN IX,SKOIDX ;Was type "SO"
JRST PXFTSO ;Yes, skip output bytes
PXFT.2: HRRZ TP,SIZTBL(IX) ;Get number of DECsystem-10 words
SOJL S1,PXFTE2 ;Jump if no more left
MOVE T0,(A1) ;Get next DECsystem-10 word
AOS A1 ;Bump DECsystem-10 pointer
SOJE TP,PXFT.3 ;Jump if only one word needed
SOJL S1,PXFTE2 ;Jump if no more left
MOVE T1,(A1) ;Get next DECsystem-10 word
AOS A1 ;Bump DECsystem-10 pointer
PXFT.3: MOVEI TP,4 ;Preset conversion error
PUSHJ P,@FTMDSP(IX) ;Jump to conversion function
PUSH P,T1 ;Store any high value
HLRE TP,SIZTBL(IX) ;Get bytes to store
JUMPG TP,PXFT.4 ;If positive, no swap
LDB T1,[POINT 16,T0,35] ;Get low word
LSH T0,-20 ;Shift high word down
DPB T1,[POINT 16,T0,19] ;Store low word as high
MOVN TP,TP ;Make count positive
PXFT.4: CAIN TP,10 ;is this a double word
MOVEI TP,4 ;Yes, work on only one now
SUB S2,TP ;Is there enough room?
JUMPL S2,PXFTE3 ;No, declare error
PUSHJ P,PXFTDB ;Go deposit word
POP P,T0 ;Restore high value
HLRZ TP,SIZTBL(IX) ;Get bytes to store
CAIE TP,-10 ;Skip if high value
JRST PXFT.5 ;Check repeat count
LDB T1,[POINT 16,T0,35] ;Get low word
LSH T0,-20 ;Shift high word down
DPB T1,[POINT 16,T0,19] ;Store low word as high
MOVEI TP,4 ;Work on one word
SUB S2,TP ;Is there enough room?
JUMPL S2,PXFTE3 ;No, declare error
PUSHJ P,PXFTDB ;Go deposit word
PXFT.5: SOJG RP,PXFT.2 ;Loop on repeat count
JRST PXFT.1 ;Otherwise, get next argument
PXFTSI: SUB S1,RP ;Get words remaining if field
JUMPL S1,PXFTE3 ;Jump if no more
ADD A1,RP ;Skip pass words in array
JRST PXFT.1 ;Convert next format
PXFTSO: SUB S2,RP ;Get bytes remaining in field
JUMPL S2,PXFTE3 ;Jump if no more
ADD AX,RP ;Update byte count
JRST PXFT.1 ;Convert next format
PXFTAS: SUB S2,RP ;Is there enough room?
JUMPL S2,PXFTE3 ;Declare error if not
MOVE T0,RP ;Get repeat count
IDIVI T0,5 ;Get number of DECsystem-10 words
SKIPE T1 ;Skip if no remainder
AOS T0 ;Bump count to next word
SUB S1,T0 ;Is there enough room
JUMPL S1,PXFTE2 ;Declare error if not
ADD T0,A1 ;Form address of next word
PUSH P,T0 ;Save for later
ADD A1,[POINT 7,,] ;Form point to characters
PXAS.1: ILDB T0,A1 ;Get next character
MOVEI TP,1 ;Store one byte
PUSHJ P,PXFTDB ;Deposit character
SOJG RP,PXAS.1 ;Loop till done
POP P,A1 ;Get next DEcsystem-10 word
JRST PXFT.1 ;Get next format
PXFTDB: MOVE T1,AX ;Copy current index
ADJBP T1,A2 ;Make pointer to deposit
DPB T0,T1 ;Deposit byte
LSH T0,-10 ;Shift down next byte
AOS AX ;Bump byte index
SOJG TP,PXFTDB ;Loop till finished
POPJ P, ;Return to caller
PXFTE3: MOVEI TP,3 ;Flag array2 overflow
MOVEM TP,(EA) ;Return error
POPJ P, ;Return to caller
PXFTE2: MOVEI TP,2 ;Flag array1 overflow
MOVEM TP,(EA) ;Return error
POPJ P, ;Return to caller
PXFTIF:
PXFTI4:
PXFTI2:
PXFTI1: MOVN T1,T0 ;Copy negative of value
SKIPL T0 ;Skip if value negative
PXFTMF:
PXFTM4:
PXFTM2:
PXFTM1: MOVE T1,T0 ;Get positive value
TDNE T1,MSKTBL(IX) ;Is value too large
MOVEM TP,(EA) ;Yes, set conversion error
POPJ P, ;Return value
PXFTL4:
PXFTL2:
PXFTL1: AND T0,[400000,,000000] ;Clear all but sign bit
TLZE T0,400000 ;Test size and zero
SETO T0, ;Set to -1
POPJ P, ;Return value
PXFTC8: PUSH P,T1 ;Store imaginary value
PUSHJ P,PXFTR4 ;Convert real part
POP P,T1 ;Get imaginary value
PUSH P,T0 ;Store converted real
MOVE T0,T1 ;Copy imaginary
PUSHJ P,PXFTR4 ;Convert imaginary part
MOVE T1,T0 ;Copy converted imaginary
POP P,T0 ;Get converted real
POPJ P, ;Return value
PXFTR4: JUMPE T0,PXR4.1 ;Just return if zero
PUSH P,T0 ;Save value
SKIPG T0 ;Skip if positive
MOVN T0,T0 ;Make value positive
LDB TP,[POINT 8,T0,8] ;Get mantissa
ASH T0,-3 ;Shift down exponent
AND T0,[000037,,777777] ;Clear extraneous bits
DPB TP,[POINT 8,T0,12] ;Deposit mantissa
POP P,TP ;Get value
SKIPG TP ;Skip if positive
TLO T0,20000 ;Turn on sign bit
PXR4.1: POPJ P, ;Return value
PXFTR8: SKIPN T0 ;Skip if non-zero
JUMPE T1,PXR8.1 ;Just return if zero
PUSH P,T0 ;Save value
SKIPG T0 ;Skip if positive
DMOVN T0,T0 ;Make value positive
LDB TP,[POINT 8,T0,8] ;Get exponent
ASHC T0,-3 ;Shift down mantissa
AND T0,[000037,,777777] ;Clear extraneous bits
DPB TP,[POINT 8,T0,12] ;Deposit exponent
ASH T1,-3 ;Shift down mantissa
AND T1,[037777,,777777] ;Clear extraneous bits
POP P,TP ;Get back value
SKIPG TP ;Skip if positive
TLO T0,20000 ;Turn on sign bit
PXR8.1: POPJ P, ;Return value
FTMDSP: PXFTI1 ;I*1 conversion
PXFTI2 ;I*2 conversion
PXFTI4 ;I*4 conversion
PXFTIF ;I*F conversion
PXFTL1 ;L*1 conversion
PXFTL2 ;L*2 conversion
PXFTL4 ;L*4 conversion
PXFTC8 ;C*8 conversion
PXFTR4 ;R*4 conversion
PXFTR8 ;R*8 conversion
PXFTM1 ;M*1 conversion
PXFTM2 ;M*2 conversion
PXFTM4 ;M*4 conversion
PXFTMF ;M*F conversion
MSKTBL: 777777,,777400 ;I*1 mask
777777,,600000 ;I*2 mask
740000,,000000 ;I*4 mask
740000,,000000 ;I*F mask
777777,,777777 ;L*1 mask
777777,,777777 ;L*2 mask
777777,,777777 ;L*4 mask
777777,,777777 ;C*8 mask
777777,,777777 ;R*4 mask
777777,,777777 ;R*8 mask
777777,,777400 ;M*1 mask
777777,,600000 ;M*2 mask
740000,,000000 ;M*4 mask
740000,,000000 ;M*F mask
SIZTBL: 1,,1 ;I*1 conversion
2,,1 ;I*2 conversion
4,,1 ;I*4 conversion
-4,,1 ;I*F conversion
1,,1 ;L*1 conversion
2,,1 ;L*2 conversion
4,,1 ;L*4 conversion
-10,,2 ;C*8 conversion
-4,,1 ;R*4 conversion
-10,,2 ;R*8 conversion
1,,1 ;M*1 conversion
2,,1 ;M*2 conversion
4,,1 ;M*4 conversion
-4,,1 ;M*F conversion
ASCIDX=.-SIZTBL
SKIIDX=.-SIZTBL+1
SKOIDX=.-SIZTBL+2
PRGEND
TITLE ARGS..
SUBTTL Setup Format-Driven Accumulators
;
; Author: R. W. Stamerjohn, MAPC DEC-10 Systems Group
; Written: 05-Apr-83
;
;+
; Purpose: This subroutine sets up the accumulators for the
; format-driven subroutines.
;
; Usage: PUSHJ P,ARGS..
; <error return>
; <normal return>
;
; Arguments: A = Fortran argument list
;
;-
; Register Definition:
;
T0=0 ;Temporary AC
T1=1 ;Temporary AC
TP=2 ;Temporary AC
A1=3 ;Current array1 address
A2=4 ;Current array2 address
S1=5 ;Size of array 1
S2=6 ;Size of array 2
FT=11 ;Format string address
FC=12 ;Format string character count
AX=13 ;Array byte index
EA=15 ;Error return variable address
A=16 ;Fortran argument block pointer
P=17 ;Push-down stack pointer
ENTRY ARGS..
ARGS..: MOVS TP,-1(A) ;Get number of arguments
CAIE TP,-7 ;Where seven arguments specified?
POPJ P, ;No, take error exit
MOVEI A1,@0(A) ;Load address of array1
MOVE S1,@1(A) ;Load size of array1
MOVEI A2,@2(A) ;Load address of array2
MOVE S2,@3(A) ;Load size of array2
MOVEI T1,4(A) ;Load address of format string
MOVE FC,@5(A) ;Load character count
MOVEI EA,@6(A) ;Load error return address
SETZ AX, ;Zero array byte index
SETZM (EA) ;Zero error return
PUSHJ P,CKSA..## ;Get type for format string
MOVE FT,T1 ;Copy byte pointer
JUMPGE FC,ARGS.2 ;If counted string, no need to count
MOVE FC,TP ;Assume CKSA count is correct
CAIN T0,15 ;Is this F77 string?
JRST ARGS.2 ;Yes, jump
PUSH P,FT ;Save pointer
SETZ FC, ;Zero the character count
ARGS.1: ILDB TP,FT ;Load next format character
SKIPE TP ;Skip if end of format
AOJA FC,ARGS.1 ;Count character and loop
POP P,FT ;Restore format array pointer
ARGS.2: AOS (P) ;Bump return address
POPJ P, ;Return to caller
PRGEND
TITLE FORM..
SUBTTL Get Next Format-Driven Field
;
; Author: R. W. Stamerjohn, MAPC DEC-10 Systems Group
; Written: 05-Apr-83
;
;+
; Purpose: This subroutine gets the next format argument.
;
; Usage: PUSHJ P,FORM..
; <error return>
; <normal return>
;
; Arguments: See C16T36/C36T16
;
;-
; Register Definition:
;
T1=1 ;Function return
TP=2 ;Temporary AC
IX=7 ;Index to conversion tables
RP=10 ;Repeat count
FT=11 ;Format string address
FC=12 ;Formate string character count
EA=15 ;Error return variable address
A=16 ;Fortran argument block pointer
P=17 ;Push-down stack pointer
ENTRY FORM..
FORM..: JUMPLE FC,FORM.X ;Exit if no more
SETZ RP, ;Zero repeat count
FORM.1: SOJL FC,FORM.E ;Count character and jump on error
ILDB TP,FT ;Get next format character
CAIL TP,"0" ;Skip if not number
CAILE TP,"9" ;Skip if number
JRST FORM.2 ;Go convert format code
IMULI RP,^D10 ;Multiple current repeat count
SUBI TP,"0" ;Get decimal digit
ADD RP,TP ;Add to total repeat count
JRST FORM.1 ;Continue repeat count loop
FORM.2: SKIPN RP ;Skip if count non-zero
MOVEI RP,1 ;Set default count to one
SOJL FC,FORM.E ;Count character and jump on error
ILDB T1,FT ;Get next character
LSH TP,7 ;Shift first character
ADD TP,T1 ;Form conversion character
MOVE IX,[-KEYSIZ,,KEYVAL] ;Get index to name table
FORM.3: CAMN TP,(IX) ;Do names match?
JRST FORM.4 ;Jump if yes
AOBJN IX,FORM.3 ;Loop till done
JRST FORM.E ;Not found, declare error
FORM.4: HRRZ IX,IX ;Zero high part of index
SUBI IX,KEYVAL ;Form conversion index
FORM.5: SOJLE FC,FORM.6 ;Count char and jump if empty
ILDB TP,FT ;Get next character
CAIN TP,"," ;Skip if not comma
JRST FORM.6 ;Comma, take skip return
CAIN TP," " ;Skip if not blank
JRST FORM.5 ;Blank, loop on next char
JRST FORM.E ;Error, neither comma nor blank
FORM.6: AOS (P) ;Take skip return
POPJ P, ;Return to caller
FORM.E: MOVEI TP,1 ;Declare format error
MOVEM TP,(EA) ;[2]Return error code
FORM.X: POPJ P, ;Take non-skip return
KEYVAL: "I1" ;I*1 conversion
"I2" ;I*2 conversion
"I4" ;I*4 conversion
"IF" ;I*F conversion
"L1" ;L*1 conversion
"L2" ;L*2 conversion
"L4" ;L*4 conversion
"C8" ;C*8 conversion
"R4" ;R*4 conversion
"R8" ;R*8 conversion
"M1" ;M*1 conversion
"M2" ;M*2 conversion
"M4" ;M*4 conversion
"MF" ;M*F conversion
"AS" ;AS conversion
"SI" ;SI conversion
"SO" ;SO conversion
KEYSIZ=.-KEYVAL
PRGEND
TITLE I1CV??
SUBTTL I*1 Signed Integer Conversion (8-bit)
;
; Author: R. W. Stamerjohn, MAPC DEC-10 Systems Group
; Written: 05-Apr-83
;
;+
; Purpose: Functions to convert one 8-bit signed integer to
; DECsystem-10 signed integer.
;
; Usage: ivar = I1CV16 (array,ptr)
; ivar = I1CV32 (array,ptr)
;
; Arguments: array = Array containing byte-oriented data
; ptr = Starting byte number within array to convert
;
;-
; Register Definition:
;
T0=0
A=16
P=17
ENTRY I1CV16
ENTRY I1CV32
I1CV16:
I1CV32: PUSHJ P,M1CV16## ;Get numerical argument in T0
AND T0,[000000,,000377] ;Clear sign extension
TRNE T0,200 ;Is sign of I*1 on?
IOR T0,[777777,,777400] ;Yes, set sign extension
POPJ P, ;Return
PRGEND
TITLE I2CV??
SUBTTL I*2 Signed Integer Conversion (16-bit)
;
; Author: R. W. Stamerjohn, MAPC DEC-10 Systems Group
; Written: 05-Apr-83
;
;+
; Purpose: Functions to convert one 16-bit signed integer to
; DECsystem-10 signed integer.
;
; Usage: ivar = I2CV16 (array,ptr)
; ivar = I2CV32 (array,ptr)
;
; Arguments: array = Array containing byte-oriented data
; ptr = Starting byte number within array to convert
;
; Note, the byte ordering in 'array' for I2CV??
; is assumed to be 0,1 where 0 is the least
; significant byte of the 16-bit integer.
;
;-
; Register Definition:
;
T0=0
A=16
P=17
ENTRY I2CV16
ENTRY I2CV32
I2CV16:
I2CV32: PUSHJ P,M2CV16## ;Get numerical argument in T0
AND T0,[000000,,177777] ;Clear sign extension
TRNE T0,100000 ;Is sign of I*1 on?
IOR T0,[777777,,600000] ;Yes, set sign extension
POPJ P, ;Return
PRGEND
TITLE I4CV??
SUBTTL I*4 Signed Integer Conversion (32-bit)
;
; Author: R. W. Stamerjohn, MAPC DEC-10 Systems Group
; Written: 05-Apr-83
;
;+
; Purpose: Functions to convert one 32-bit signed integer,
; ordered (low,high), to DECsystem-10 signed integer.
;
; Usage: ivar = I4CV16 (array,ptr)
; ivar = I4CV32 (array,ptr)
;
; Arguments: array = Array containing byte-oriented data
; ptr = Starting byte number within array to convert
;
; Note, the byte ordering in 'array' for I4CV??
; is assumed to be 0,1,2,3 where 0 is the least
; significant byte of the 32-bit integer.
;
;-
; Register Definition:
;
T0=0
A=16
P=17
ENTRY I4CV16
ENTRY I4CV32
I4CV16:
I4CV32: PUSHJ P,M4CV16## ;Get numerical argument in T0
AND T0,[037777,,777777] ;Clear sign extension
TLNE T0,20000 ;Is sign of I*4 on?
IOR T0,[740000,,000000] ;Yes, set sign extension
POPJ P, ;Return
PRGEND
TITLE IFCV??
SUBTTL I*F Signed Integer Conversion (32-bit)
;
; Author: R. W. Stamerjohn, MAPC DEC-10 Systems Group
; Written: 05-Apr-83
;
;+
; Purpose: Functions to convert one 32-bit signed integer,
; ordered (high,low), to DECsystem-10 signed integer.
;
; Usage: ivar = IFCV16 (array,ptr)
; ivar = IFCV32 (array,ptr)
;
; Arguments: array = Array containing byte-oriented data
; ptr = Starting byte number within array to convert
;
; Note, the byte ordering in 'array' for IFCV??
; is assumed to be 2,3,0,1 where 0 is the least
; byte.
;
;-
; Register Definition:
;
T0=0
A=16
P=17
ENTRY IFCV16
ENTRY IFCV32
IFCV16:
IFCV32: PUSHJ P,MFCV16## ;Get numerical argument in T0
AND T0,[037777,,777777] ;Clear sign extension
TLNE T0,20000 ;Is sign of I*4 on?
IOR T0,[740000,,000000] ;Yes, set sign extension
POPJ P, ;Return
PRGEND
TITLE L1CV??
SUBTTL L*1 Logical Conversion (8-bit)
;
; Author: R. W. Stamerjohn, MAPC DEC-10 Systems Group
; Written: 05-Apr-83
;
;+
; Purpose: Functions to convert one 8-bit VAX/PDP-11 logical
; value to DECsystem-10 logical value.
;
; Usage: lvar = L1CV16 (array,ptr)
; lvar = L1CV32 (array,ptr)
;
; Arguments: array = Array containing byte-oriented data
; ptr = Starting byte number within array to convert
;
;-
; Register Definition:
;
T0=0
A=16
P=17
ENTRY L1CV16
ENTRY L1CV32
L1CV16: PUSHJ P,M1CV16## ;Get numerical argument in T0
AND T0,[000000,,000200] ;Clear all but sign
TRZE T0,200 ;Is sign of L*1 on?
SETO T0, ;Yes, set return to -1
POPJ P, ;Return
L1CV32: PUSHJ P,M1CV16## ;Get numerical argument in T0
AND T0,[000000,,000001] ;Clear all but lsb
TRZE T0,1 ;Is lsb of L*1 set?
SETO T0, ;Yes, set return to -1
POPJ P, ;Return
PRGEND
TITLE L2CV??
SUBTTL L*2 Logical Conversion (16-bit)
;
; Author: R. W. Stamerjohn, MAPC DEC-10 Systems Group
; Written: 05-Apr-83
;
;+
; Purpose: Functions to convert one 16-bit VAX/PDP-11
; logical value to DECsystem-10 logical value.
;
; Usage: lvar = L2CV16 (array,ptr)
; lvar = L2CV32 (array,ptr)
;
; Arguments: array = Array containing byte-oriented data
; ptr = Starting byte number within array to convert
;
; Note, the byte ordering in 'array' for L2CV??
; is assumed to be 0,1 where 0 is the least
; significant byte of the 16-bit logical.
;
;-
; Register Definition:
;
T0=0
A=16
P=17
ENTRY L2CV16
ENTRY L2CV32
L2CV16: PUSHJ P,M2CV16## ;Get numerical argument in T0
AND T0,[000000,,100000] ;Clear all but sign
TRZE T0,100000 ;Is sign of L*2 on?
SETO T0, ;Yes, set return to -1
POPJ P, ;Return
L2CV32: PUSHJ P,M2CV16## ;Get numerical argument in T0
AND T0,[000000,,000001] ;Clear all but lsb
TRZE T0,1 ;Is lsb set?
SETO T0, ;Yes, set return to -1
POPJ P, ;Return
PRGEND
TITLE L4CV??
SUBTTL L*4 Logical Conversion (32-bit)
;
; Author: R. W. Stamerjohn, MAPC DEC-10 Systems Group
; Written: 05-Apr-83
;
;+
; Purpose: Functions to convert one 32-bit VAX/PDP-11 logical
; value to DECsystem-10 logical value.
;
; Usage: lvar = L4CV16 (array,ptr)
; lvar = L4CV32 (array,ptr)
;
; Arguments: array = Array containing byte-oriented data
; ptr = Starting byte number within array to convert
;
; Note, the byte ordering in 'array' for L4CV??
; is assumed to be 0,1,2,3 where 0 is the least
; significant byte of the 32-bit logical.
;
;-
; Register Definition:
;
T0=0
A=16
P=17
ENTRY L4CV16
ENTRY L4CV32
L4CV16: PUSHJ P,M4CV16## ;Get numerical argument in T0
AND T0,[020000,,000000] ;Clear all but sign
TLZE T0,20000 ;Is sign of L*4 on?
SETO T0, ;Yes, set return to -1
POPJ P, ;Return
L4CV32: PUSHJ P,M4CV16## ;Get numerical argument in T0
AND T0,[000000,,000001] ;Clear all but lsb
TRZE T0,1 ;Is lsb of L*4 on?
SETO T0, ;Yes, set return to -1
POPJ P, ;Return
PRGEND
TITLE C8CV??
SUBTTL C*8 Complex Conversion (Single-Precision)
;
; Author: R. W. Stamerjohn, MAPC DEC-10 Systems Group
; Written: 05-Apr-83
;
;+
; Purpose: Functions to convert two 32-bit VAX/PDP-11 real values
; to DECsystem-10 complex value.
;
; Usage: cvar = C8CV16 (array,ptr)
; cvar = C8CV32 (array,ptr)
;
; Arguments: array = Array containing byte-oriented data
; ptr = Starting byte number within array to convert
;
; Note, the byte ordering in 'array' for C8CV??
; is assumed to be 0,1,..7 where 0 is the first
; byte of the two 32-bit floating PDP-11 numbers.
;
;-
; Register Definition:
;
T0=0
T1=1
P0=2
P1=3
A=16
P=17
ENTRY C8CV16
ENTRY C8CV32
C8CV16:
C8CV32: PUSH P,P0 ;Save a worker
PUSH P,P1 ;Save a worker
MOVEI P0,@0(A) ;Get array address
MOVE P1,@1(A) ;Get byte index
ADDI P1,4 ;Get imaginary part
MOVEI A,ARGLST ;Get fortran arguments
PUSHJ P,R4CV16## ;Convert to DECsystem-10 real
PUSH P,T0 ;Save imaginary part
SUBI P1,4 ;Get real part
PUSHJ P,R4CV16## ;Convert to DECsystem-10 real
POP P,T1 ;Get imaginary part
POP P,P1 ;Restore worker
POP P,P0 ;Restore worker
POPJ P, ;Return
-2,,0
ARGLST: 20,,P0
00,,P1
PRGEND
TITLE R4CV??
SUBTTL R*4 Real Conversion (Single-Precision)
;
; Author: R. W. Stamerjohn, MAPC DEC-10 Systems Group
; Written: 05-Apr-83
;
;+
; Purpose: Function to convert 32-bit VAX/PDP-11 floating value
; to DECsystem-10 real value.
;
; Usage: rvar = R4CV16 (array,ptr)
; Usage: rvar = R4CV32 (array,ptr)
;
; Arguments: array = Array containing byte-oriented data
; ptr = Starting byte number within array to convert
;
; Note, the byte ordering in 'array' for R4CV??
; is assumed to be 0,1,2,3 where 0 is the first
; byte of the 32-bit VAX/PDP-11 floating PDP-11 number.
;
;-
; Register Definition:
;
T0=0
T1=1
A=16
P=17
ENTRY R4CV16
ENTRY R4CV32
R4CV16:
R4CV32: PUSHJ P,MFCV16## ;Get numerical argument in T0
JUMPE T0,R4CV.1 ;Just return if zero
PUSH P,T0 ;Save PDP-11 value
LDB T1,[POINT 8,T0,12] ;Get PDP-11 exponent
AND T0,[000037,,777777] ;Clear all but PDP-11 mantissa
IOR T0,[000040,,000000] ;Set PDP-11 hidden bit
ASH T0,3 ;Shift into DECsystem-10 mantissa
DPB T1,[POINT 8,T0,8] ;Store DECsystem-10 exponent
POP P,T1 ;Restore PDP-11 value
TLNE T1,20000 ;Test if value was negative?
MOVN T0,T0 ;Yes, set DECsystem-10 negative
R4CV.1: POPJ P, ;Return
PRGEND
TITLE R8CV??
SUBTTL R*8 Real Conversion (Double-Precision)
;
; Author: R. W. Stamerjohn, MAPC DEC-10 Systems Group
; Written: 05-Apr-83
;
;+
; Purpose: Functions to convert 64-bit VAX/PDP-11 floating value
; to DECsystem-10 floating value.
;
; Usage: rvar = R8CV16 (array,ptr)
; Usage: rvar = R8CV32 (array,ptr)
;
; Arguments: array = Array containing byte-oriented data
; ptr = Starting byte number within array to convert
;
; Note, the byte ordering in 'array' for R8CV??
; is assumed to be 0,1,..7 where 0 is the first
; byte of the 64-bit floating VAX/PDP-11 number.
;
;-
; Register Definition:
;
T0=0
T1=1
P0=2
P1=3
A=16
P=17
ENTRY R8CV16
ENTRY R8CV32
R8CV16:
R8CV32: PUSH P,P0 ;Save a worker
PUSH P,P1 ;Save a worker
MOVEI P0,@0(A) ;Get array address
MOVE P1,@1(A) ;Get byte index
ADDI P1,4 ;Get low part
MOVEI A,ARGLST ;Get fortran arguments
PUSHJ P,MFCV16## ;Get low part in T0
PUSH P,T0 ;Save imaginary part
SUBI P1,4 ;Get high part
PUSHJ P,MFCV16## ;Get numerical argument in T0,T1
POP P,T1 ;Now have number in T0,T1
SKIPN T1 ;Skip if non-zero
JUMPE T0,R8CV.1 ;Just return if zero
PUSH P,T0 ;Save PDP-11 value
LDB P1,[POINT 8,T0,12] ;Get PDP-11 exponent
AND T0,[000037,,777777] ;Clear all but PDP-11 mantissa
IOR T0,[000040,,000000] ;Set PDP-11 hidden bit
ASH T1,3 ;Shift low-order mantissa
ASHC T0,3 ;Shift combined mantissa
DPB P1,[POINT 8,T0,8] ;Store DECsystem-10 exponent
POP P,P1 ;Restore PDP-11 value
TLNE P1,20000 ;Test if value was negative?
DMOVN T0,T0 ;Yes, set DECsystem-10 negative
POP P,P0 ;Restore worker
POP P,P1 ;Restore worker
R8CV.1: POPJ P, ;Return
-2,,0
ARGLST: 20,,P0
00,,P1
PRGEND
TITLE ASCV??
SUBTTL ASCII Character Conversion
;
; Author: R. W. Stamerjohn, MAPC DEC-10 Systems Group
; Written: 05-Apr-83
;
;+
; Purpose: Function to convert one VAX/PDP-11 character from
; byte array to DECsystem-10 character.
;
; Usage: ivar = ASCV16 (array,ptr,possrc,postgt)
;
; Arguments: array = Array containing byte-oriented data
; ptr = Byte number within array to convert
; possrc = Position of source charactare (ignore)
; postgt = Position (1-5) to store character
;
; The possrc value is always one for the VAX/PDP-11
; and is ignored by this routine. The postgt argument
; is position in the DEC-10 word to store the character.
;
;-
; Register Definition:
;
T0=0
T1=1
A=16
P=17
ENTRY ASCV16
ENTRY ASCV32
ASCV16:
ASCV32: PUSHJ P,M1CV16## ;Get numerical argument in T0
MOVE T1,@3(A) ;Get position of character
IMUL T1,[-7] ;Get negative bits/character
ADDI T1,^D36 ;Get number shifts needed
LSH T0,(T1) ;Shift character to position
POPJ P, ;Return
PRGEND
TITLE M1CV??
SUBTTL M*1 Unsigned Integer Conversion (8-bit)
;
; Author: R. W. Stamerjohn, MAPC DEC-10 Systems Group
; Written: 05-Apr-83
;
;+
; Purpose: Functions to convert one VAX/PDP-11 byte from
; byte array to DECsystem-10 integer.
;
; No sign extension performed.
;
; Usage: ivar = M1CV16 (array,ptr)
; ivar = M1CV32 (array,ptr)
;
; Arguments: array = Array containing byte-oriented data
; ptr = Byte number within array to convert
;
;-
; Register Definition:
;
T0=0
T1=1
A=16
P=17
ENTRY M1CV16
ENTRY M1CV32
M1CV16:
M1CV32: MOVE T0,@1(A) ;Get byte index
MOVEI T1,1 ;Set byte count
PUSHJ P,BYTE..## ;Input binary bytes
POPJ P, ;Return
PRGEND
TITLE M2CV??
SUBTTL M*2 Unsigned Integer Conversion (16-bit)
;
; Author: R. W. Stamerjohn, MAPC DEC-10 Systems Group
; Written: 05-Apr-83
;
;+
; Purpose: Function to convert two VAX/PDP-11 bytes from
; byte array to DECsystem-10 integer.
;
; No sign extension performed.
;
; Usage: ivar = M2CV16 (array,ptr)
; ivar = M2CV32 (array,ptr)
;
; Arguments: array = Array containing byte-oriented data
; ptr = Starting byte number within array to convert
;
; Note, the byte ordering in 'array' for M2CV??
; is assumed to be 0 and 1 where 0 is the first
; 8-bit byte of the set.
;
; These same bytes will appear in 'ivar' as:
;
; ------------------------------------
; |xx|xxxxxxx|xxxxxxx| 1 | 0 |
; ------------------------------------
;-
; Register Definition:
;
T0=0
T1=1
A=16
P=17
ENTRY M2CV16
ENTRY M2CV32
M2CV16:
M2CV32: MOVE T0,@1(A) ;Get byte index
MOVEI T1,2 ;Set byte count
PUSHJ P,BYTE..## ;Input binary bytes
POPJ P, ;Return
PRGEND
TITLE M4CV??
SUBTTL M*4 Unsigned Integer Conversion (32-bit)
;
; Author: R. W. Stamerjohn, MAPC DEC-10 Systems Group
; Written: 05-Apr-83
;
;+
; Purpose: Function to convert four VAX/PDP-11 bytes from
; byte array to DECsystem-10 integer. Assume order
; is (low,high).
;
; No sign extension performed.
;
; Usage: ivar = M4CV16 (array,ptr)
; ivar = M4CV32 (array,ptr)
;
; Arguments: array = Array containing byte-oriented data
; ptr = Starting byte number within array to convert
;
; Note, the byte ordering in 'array' for M4CV??
; is assumed to be 0,1,2,3 where 0 is the first
; 8-bit byte of the set.
;
; These same bytes will appear in 'ivar' as:
;
; ------------------------------------
; |xx| 3 | 2 | 1 | 0 |
; ------------------------------------
;-
; Register Definition:
;
T0=0
T1=1
A=16
P=17
ENTRY M4CV16
ENTRY M4CV32
M4CV16:
M4CV32: MOVE T0,@1(A) ;Get byte index
MOVEI T1,4 ;Set byte count
PUSHJ P,BYTE..## ;Input binary bytes
POPJ P, ;Return
PRGEND
TITLE MFCV??
SUBTTL M*F Unsigned Integer Conversion (32-bit)
;
; Author: R. W. Stamerjohn, MAPC DEC-10 Systems Group
; Written: 05-Apr-83
;
;+
; Purpose: Function to convert four VAX/PDP-11 bytes from
; byte array to DECsystem-10 integer. Assume order
; is (high,low).
;
; No sign extension performed.
;
; Usage: ivar = MFCV16 (array,ptr)
; Usage: ivar = MFCV32 (array,ptr)
;
; Arguments: array = Array containing byte-oriented data
; ptr = Starting byte number within array to convert
;
; Note, the byte ordering in 'array' for MFCV??
; is assumed to be 0,1,2,3 where 0 is the first
; 8-bit byte of the set.
;
; These same bytes will appear in 'ivar' as:
;
; ------------------------------------
; |xx| 1 | 0 | 3 | 2 |
; ------------------------------------
;
;-
; Register Definition:
;
T0=0
T1=1
A=16
P=17
ENTRY MFCV16
ENTRY MFCV32
MFCV16:
MFCV32: MOVE T0,@1(A) ;Get byte index
MOVEI T1,4 ;Set byte count
PUSHJ P,BYTE..## ;Input binary bytes
MOVE T1,T0 ;Copy return value
ANDI T0,177777 ;Clear all but low word
LSH T0,20 ;Shift lower to upper
LSH T1,-20 ;Shift upper to lower
ADD T0,T1 ;Form result
POPJ P, ;Return
PRGEND
TITLE BYTE..
SUBTTL Get Array Mode Bytes
;
; Author: R. W. Stamerjohn, MAPC DEC-10 Systems Group
; Written: 05-Apr-83
;
;+
; Purpose: Get bytes from Array Mode array.
;
; Usage: PUSHJ P,BYTE..
;
; Arguments: T0 = Index into array
; T1 = Number of bytes
; 0(A) = Address of array
;
;-
; Register Definition:
;
T0=0
T1=1
P1=2
P2=3
P3=4
P4=5
A=16
P=17
ENTRY BYTE..
BYTE..: PUSH P,P1 ;Save a worker
PUSH P,P2 ;Save a worker
PUSH P,P3 ;Save a worker
PUSH P,P4 ;Save a worker
MOVE P1,T0 ;Copy byte index
SETZ T0, ;Zero final result
MOVEI P2,@0(A) ;Get address of array
ADD P2,[POINT 8,,7] ;Form point to first byte
MOVN P3,T1 ;Get negative of count
HRLZ P3,P3 ;Put in upper half
BYTE.1: MOVE P4,P1 ;Get current index
ADJBP P4,P2 ;Form pointer to byte
LDB T1,P4 ;Get desired byte
HRRZ P4,P3 ;Get current position
IMULI P4,10 ;Get bits to shift
LSH T1,(P4) ;Shift byte to correct position
ADD T0,T1 ;Add to final result
AOS P1 ;Bump index
AOBJN P3,BYTE.1 ;Add one and loop
POP P,P4 ;Restore worker
POP P,P3 ;Restore worker
POP P,P2 ;Restore worker
POP P,P1 ;Restore worker
POPJ P, ;Return
END