![[BACK]](/cvsweb/icons/back.gif){kind=icon}
Return to adb.1 CVS log ![[TXT]](/cvsweb/icons/text.gif){kind=icon}
![[DIR]](/cvsweb/icons/dir.gif){kind=icon}
Up to [Research Unix] / researchv10dc / man / adm / man1|
Return to adb.1 CVS log | Up to [Research Unix] / researchv10dc / man / adm / man1 |
1.1 root 1: .TH ADB 1
2: .CT 1 debug_tune
3: .ds TW \v'.25m'\s+2~\s-2\v'-.25m'
4: .ds ST \v'.25m'*\v'-.25m'
5: .ds IM \v'.1m'=\v'-.1m'\s-2\h'-.1m'>\h'.1m'\s+2
6: .ds LE \(<=
7: .ds LT \s-2<\s+2
8: .ds GT \s-2>\s+2
9: .SH NAME
10: adb \- debugger
11: .SH SYNOPSIS
12: .B adb
13: [
14: .I option ...
15: ]
16: [
17: .I objfil
18: [
19: .I corfil
20: ]
21: ]
22: .SH DESCRIPTION
23: .I Adb
24: is a general purpose debugging program.
25: It may be used to examine files and to provide
26: a controlled environment for the execution
27: of UNIX programs.
28: .PP
29: .I Objfil
30: is normally an executable program file, preferably
31: containing a symbol table;
32: if not then the
33: symbolic features of
34: .I adb
35: cannot be used although the file can still
36: be examined.
37: The default for
38: .I objfil
39: is
40: .LR a.out .
41: .I Corfil
42: is assumed to be a core image file produced after
43: executing
44: .IR objfil ;
45: the default for
46: .I corfil
47: is
48: .LR core .
49: .PP
50: Requests to
51: .I adb
52: are read from the standard input and
53: responses are to the standard output.
54: Quit signals are ignored; interrupts
55: cause return to the next
56: .I adb
57: command.
58: The options are
59: .TP
60: .B -w
61: Create
62: .I objfil
63: and
64: .I corfil
65: if they don't exist; open them for writing
66: as well as reading.
67: .TP
68: .BI -I path
69: Directory in which to look for relative pathnames in
70: .B $<
71: and
72: .B $<<
73: commands.
74: .PP
75: In general requests to
76: .I adb
77: have the following form.
78: Multiple requests on one line must be separated by
79: .LR ; .
80: .IP
81: .RI [ address ]
82: .RB [ ,
83: .IR count ]
84: .RI [ command ]
85: .PP
86: If
87: .I address
88: is present then the current position, called `dot',
89: is set to
90: .IR address .
91: Initially dot
92: is set to 0.
93: In general commands are repeated
94: .I count
95: times.
96: Dot advances between repetitions.
97: The default
98: .I count
99: is 1.
100: .I Address
101: and
102: .I count
103: are expressions.
104: .PP
105: Some formats,
106: data sizes,
107: and command details have
108: different behavior
109: on different systems.
110: See the
111: .SM "MACHINE DEPENDENCIES"
112: attachment for details.
113: .SS Expressions
114: Expressions are computed with sufficient precision
115: to address the largest possible file;
116: generally this means a long integer.
117: On the VAX,
118: expressions are 32 bits;
119: on the Cray,
120: 64 bits.
121: .TP 7.2n
122: .B .
123: The value of dot.
124: .TP 7.2n
125: .B +
126: The value of dot
127: incremented by the current increment.
128: .TP 7.2n
129: .B ^
130: The value of dot
131: decremented by the current increment.
132: .TP 7.2n
133: .B \&"
134: The last
135: .I address
136: typed.
137: .TP 7.2n
138: .I integer
139: A number
140: in the
141: .IR "default radix" ;
142: see the
143: .B $d
144: command.
145: Regardless of the default,
146: the prefixes
147: .L 0o
148: and
149: .L 0O
150: (zero oh) force interpretation
151: in octal radix; the prefixes
152: .L 0t
153: and
154: .L 0T
155: force interpretation in
156: decimal radix; the prefixes
157: .LR 0x ,
158: .LR 0X ,
159: and
160: .L #
161: force interpretation in
162: hexadecimal radix.
163: Thus
164: .LR 0o20 ,
165: .LR 0t16 ,
166: and
167: .L #10
168: all represent sixteen.
169: .TP 7.2n
170: .IB integer . fraction
171: A floating point number.
172: .TP 7.2n
173: .BI \' cccc\| \'
174: The
175: .SM ASCII
176: value of one or more characters.
177: .L \e
178: may be used to escape a
179: .LR \' .
180: .TP 7.2n
181: .BI < name
182: The value of
183: .IR name ,
184: which is either a variable name or a register name.
185: .I Adb
186: maintains a number of variables
187: named by single letters or digits.
188: The register names are
189: those printed by the
190: .B $r
191: command.
192: .TP 7.2n
193: .I symbol
194: A
195: .I symbol
196: is a sequence
197: of upper or lower case letters, underscores or
198: digits, not starting with a digit.
199: .L \e
200: may be used to escape other characters.
201: The value of the
202: .I symbol
203: is taken from the symbol table
204: in
205: .IR objfil .
206: .TP 7.2n
207: .IB routine . name
208: The address of the variable
209: .I name
210: in the specified
211: C routine.
212: Both
213: .I routine
214: and
215: .I name
216: are
217: .IR symbols .
218: If
219: .I name
220: is omitted the value is the address of the
221: most recently activated C stack frame
222: corresponding to
223: .IR routine ;
224: if
225: .I routine
226: is omitted,
227: the active procedure
228: is assumed.
229: .TP 7.2n
230: .BI ( exp )
231: The value of the expression
232: .IB exp .
233: .LP
234: .I Monadic operators
235: .TP 7.2n
236: .BI * exp
237: The contents of the location addressed
238: by
239: .I exp
240: in
241: .IR corfil .
242: .TP 7.2n
243: .BI @ exp
244: The contents of the location addressed by
245: .I exp
246: in
247: .IR objfil .
248: .TP 7.2n
249: .BI - exp
250: Integer negation.
251: .TP 7.2n
252: .BI ~ exp
253: Bitwise complement.
254: .TP 7.2n
255: .BI % exp
256: If
257: .I exp
258: is used as an address,
259: it is in register space;
260: see `Addresses'.
261: .LP
262: .I "Dyadic\ operators"
263: are left associative
264: and are less binding than monadic operators.
265: .TP 7.2n
266: .IB e1 + e2
267: Integer addition.
268: .TP 7.2n
269: .IB e1 - e2
270: Integer subtraction.
271: .TP 7.2n
272: .IB e1 * e2
273: Integer multiplication.
274: .TP 7.2n
275: .IB e1 % e2
276: Integer division.
277: .TP 7.2n
278: .IB e1 & e2
279: Bitwise conjunction.
280: .TP 7.2n
281: .IB e1 | e2
282: Bitwise disjunction.
283: .TP 7.2n
284: .IB e1 # e2
285: .I E1
286: rounded up to the next multiple of
287: .IR e2 .
288: .DT
289: .SS Commands
290: Most commands consist of a verb followed by a modifier or list
291: of modifiers.
292: The following verbs are available.
293: (The commands
294: .L ?
295: and
296: .L /
297: may be followed by
298: .LR * ;
299: see `Addresses'
300: for further details.)
301: .TP .5i
302: .BI ? f
303: Locations starting at
304: .I address
305: in
306: .I objfil
307: are printed according to the format
308: .IR f .
309: .TP
310: .BI / f
311: Locations starting at
312: .I address
313: in
314: .I corfil
315: are printed according to the format
316: .I f.
317: .TP
318: .BI = f
319: The value of
320: .I address
321: itself is printed in the
322: styles indicated by the format
323: .IR f .
324: (For
325: .B i
326: format
327: .L ?
328: is printed for the parts of the instruction that reference
329: subsequent words.)
330: .PP
331: A
332: .I format
333: consists of one or more characters that specify a style
334: of printing.
335: Each format character may be preceded by a decimal integer
336: that is a repeat count for the format character.
337: If no format is given then the last format is used.
338: .PP
339: Most format letters fetch some data,
340: print it,
341: and advance (a local copy of) dot
342: by the number of bytes fetched.
343: The total number of bytes in a format becomes the
344: .I current increment.
345: `Long integers' are full words,
346: the size of an expression item:
347: e.g.\&
348: 4 bytes on the VAX,
349: 8 bytes on the Cray.
350: `Short integers'
351: are some useful shorter size:
352: 2 byte short integers on the VAX,
353: 2 byte parcels on the Cray.
354: .ta 2.5n .5i
355: .RS
356: .TP
357: .PD 0
358: .B r
359: Print short integer in the current default radix.
360: .TP
361: .PD 0
362: .B R
363: Print long integer in the current default radix.
364: .TP
365: .PD 0
366: .B o
367: Print short integer in octal.
368: .TP
369: .B O
370: Print long integer in octal.
371: .TP
372: .B q
373: Print short in signed octal.
374: .TP
375: .B Q
376: Print long in signed octal.
377: .TP
378: .B d
379: Print short in decimal.
380: .TP
381: .B D
382: Print long in decimal.
383: .TP
384: .B x
385: Print short in hexadecimal.
386: .TP
387: .B X
388: Print long in hexadecimal.
389: .TP
390: .B u
391: Print short in unsigned decimal.
392: .TP
393: .B U
394: Print long in unsigned decimal.
395: .TP
396: .B f
397: Print
398: as a floating point number.
399: .TP
400: .B F
401: Print double-precision floating point.
402: .TP
403: .B b
404: Print the addressed byte in octal.
405: .TP
406: .B c
407: Print the addressed character.
408: .TP
409: .B C
410: Print the addressed character.
411: Control characters
412: are printed in the form
413: .BI ^ X
414: and the delete character is printed as
415: .LR ^? .
416: .TP
417: .B s
418: Print the addressed characters until a zero character
419: is reached.
420: Advance dot
421: by the length of the string,
422: including the zero terminator.
423: .TP
424: .B S
425: Print a string using
426: the
427: .BI ^ X
428: escape convention (see
429: .B C
430: above).
431: .TP
432: .B Y
433: Print a long integer in date format (see
434: .IR ctime (3)).
435: .TP
436: .B i
437: Print as machine instructions.
438: This style of printing causes variables
439: 0, (1, ...)
440: to be set to the offset parts
441: of the first (second, ...)
442: operand of the instruction.
443: .TP
444: .B a
445: Print the value of dot
446: in symbolic form.
447: Dot is unaffected.
448: .TP
449: .B p
450: Print the addressed value in symbolic form.
451: Dot is advanced by the size of a machine address
452: (4 bytes on the VAX,
453: 8 bytes on the Cray).
454: .TP
455: .B t
456: When preceded by an integer tabs to the next
457: appropriate tab stop.
458: For example,
459: .B 8t
460: moves to the next 8-space tab stop.
461: Dot is unaffected.
462: .TP
463: .B n
464: Print a newline.
465: Dot is unaffected.
466: .tr '"
467: .TP
468: .BR ' ... '
469: Print the enclosed string.
470: Dot is unaffected.
471: .br
472: .tr ''
473: .TP
474: .B ^
475: Dot is decremented by the current increment.
476: Nothing is printed.
477: .TP
478: .B +
479: Dot is incremented by 1.
480: Nothing is printed.
481: .TP
482: .B -
483: Dot is decremented by 1.
484: Nothing is printed.
485: .RE
486: .PD
487: .TP
488: newline
489: Update dot by the current increment.
490: Repeat the previous command with a
491: .I count
492: of 1.
493: .TP
494: .RB [ ?/ ] l "\fI value mask\fR"
495: Words starting at dot
496: are masked with
497: .I mask
498: and compared with
499: .I value
500: until
501: a match is found.
502: If
503: .B l
504: is used,
505: the match is for a short integer;
506: .B L
507: matches longs.
508: If no match is found then dot
509: is unchanged; otherwise dot
510: is set to the matched location.
511: If
512: .I mask
513: is omitted then \-1 is used.
514: .TP
515: .RB [ ?/ ] w "\fI value ...\fR"
516: Write the short
517: .I value
518: into the addressed
519: location.
520: If the command is
521: .BR W ,
522: write a long.
523: Option
524: .B -w
525: must be in effect.
526: .TP
527: .RB [ ?/ ] "m\fI b e f \fP" [ ?\fR]
528: .br
529: New values for
530: .RI ( b,\ e,\ f )
531: in the first map entry
532: are recorded.
533: If less than three expressions are given then
534: the remaining map parameters are left unchanged.
535: The address type (instruction or data)
536: is unchanged in any case.
537: If the
538: .L ?
539: or
540: .L /
541: is followed by
542: .L *
543: then
544: the second segment
545: of the mapping is changed.
546: If the list is terminated by
547: .L ?
548: or
549: .L /
550: then the file
551: .RI ( objfil
552: or
553: .I corfil
554: respectively) is used
555: for subsequent requests.
556: For example,
557: .L /m?
558: will cause
559: .L /
560: to refer to
561: .IR objfil .
562: .TP
563: .BI > name
564: Dot is assigned to the variable or register named.
565: .TP
566: .B !
567: A shell is called to read the
568: rest of the line following `!'.
569: .TP
570: .BI $ modifier
571: Miscellaneous commands.
572: The available
573: .I modifiers
574: are:
575: .RS
576: .TP
577: .PD 0
578: .BI < f
579: Read commands from the file
580: .IR f .
581: If
582: .I f
583: cannot be found, try
584: .BI /usr/lib/adb/ f.
585: If this command is executed in a file, further commands
586: in the file are not seen.
587: If
588: .I f
589: is omitted, the current input stream is terminated.
590: If a
591: .I count
592: is given, and is zero, the command will be ignored.
593: The value of the count will be placed in variable
594: .B 9
595: before the first command in
596: .I f
597: is executed.
598: .TP
599: .BI << f
600: Similar to
601: .B <
602: except it can be used in a file of commands without
603: causing the file to be closed.
604: Variable
605: .B 9
606: is saved during the execution of this command, and restored
607: when it completes.
608: There is a (small) limit to the number of
609: .B <<
610: files that can be open at once.
611: .br
612: .ns
613: .TP
614: .BI > f
615: Append output to the file
616: .IR f ,
617: which is created if it does not exist.
618: If
619: .I f
620: is omitted, output is returned to the terminal.
621: .TP
622: .B ?
623: Print process id, the signal which caused stopping or termination,
624: as well as the registers.
625: This is the default if
626: .I modifier
627: is omitted.
628: .TP
629: .B r
630: Print the general registers and
631: the instruction addressed by
632: .BR pc .
633: Dot is set to
634: .BR pc .
635: .TP
636: .B R
637: Like
638: .BR $r ,
639: but include miscellaneous registers
640: such as the kernel stack pointer.
641: .TP
642: .B b
643: Print all breakpoints
644: and their associated counts and commands.
645: .TP
646: .B c
647: C stack backtrace.
648: If
649: .I address
650: is given then it is taken as the
651: address of the current frame;
652: otherwise,
653: the current C frame pointer
654: is used.
655: If
656: .B C
657: is used then the names and (long) values of all
658: parameters,
659: automatic
660: and static variables are printed for each active function.
661: If
662: .I count
663: is given then only the first
664: .I count
665: frames are printed.
666: .TP
667: .B a
668: Set the maximum number of arguments
669: printed by
670: .B $c
671: or
672: .B $C
673: to
674: .IR address .
675: The default is 20.
676: .TP
677: .B d
678: Set the default radix to
679: .I address
680: and report the new value.
681: .I Address
682: is interpreted in the (old) current radix;
683: .L 10$d
684: never changes the default radix.
685: To make decimal the default radix, use
686: .LR 0t10$d .
687: A radix of zero (the initial default) is a special case;
688: input with a leading zero is octal,
689: that with a leading sharp-sign
690: .L #
691: is hexadecimal,
692: other numbers are decimal.
693: When the default radix is zero,
694: the default output radix is
695: appropriate to the machine:
696: hexadecimal on the VAX,
697: octal on the Cray.
698: .TP
699: .B e
700: The names and values of all
701: external variables are printed.
702: .TP
703: .B w
704: Set the page width for output to
705: .I address
706: (default 80).
707: .TP
708: .B s
709: Set the limit for symbol matches to
710: .I address
711: (default 255).
712: .TP
713: .B q
714: Exit from
715: .IR adb .
716: .TP
717: .B v
718: Print all non zero variables in the current radix.
719: .TP
720: .B m
721: Print the address maps.
722: .TP
723: .B k
724: Simulate kernel memory management.
725: .TP
726: .B p
727: Simulate per-process memory management.
728: .IP
729: .B $k
730: and
731: .B $p
732: are used for system debugging.
733: Their details vary with machine and operating system.
734: .PD
735: .RE
736: .TP
737: .BI : modifier
738: Manage a subprocess.
739: Available modifiers are:
740: .RS
741: .TP
742: .PD 0
743: .BI b c
744: Set breakpoint at
745: .IR address .
746: The breakpoint is executed
747: .IR count \-1
748: times before
749: causing a stop.
750: Each time the breakpoint is encountered
751: the command
752: .I c
753: is executed.
754: If this command is omitted or sets dot
755: to zero
756: then the breakpoint causes a stop.
757: .TP
758: .B d
759: Delete breakpoint at
760: .IR address .
761: .TP
762: .B r
763: Run
764: .I objfil
765: as a subprocess.
766: If
767: .I address
768: is given explicitly then the
769: program is entered at this point; otherwise
770: the program is entered at its standard entry point.
771: .I count
772: specifies how many breakpoints are to be
773: ignored before stopping.
774: Arguments to the subprocess may be supplied on the
775: same line as the command.
776: An argument starting with < or > causes the standard
777: input or output to be established for the command.
778: All signals are enabled on entry to the subprocess.
779: .TP
780: .BI c s
781: The subprocess is continued.
782: If
783: .I s
784: is omitted
785: or nonzero,
786: the subprocess
787: is sent the signal that caused it to stop;
788: if 0
789: is specified,
790: no signal is sent.
791: Breakpoints
792: and single-stepping
793: don't count as signals.
794: Breakpoint skipping is the same
795: as for
796: .BR r .
797: .TP
798: .BI s s
799: As for
800: .B c
801: except that
802: the subprocess is single stepped
803: .I count
804: times.
805: If a signal is sent,
806: it is received
807: before the first instruction is executed.
808: If there is no current subprocess then
809: .I objfil
810: is run
811: as a subprocess as for
812: .BR r .
813: In this case no signal can be sent; the remainder of the line
814: is treated as arguments to the subprocess.
815: .TP
816: .B k
817: The current subprocess, if any, is terminated.
818: .PD
819: .RE
820: .SS Variables
821: .I Adb
822: provides a number of variables.
823: Named variables are set initially by
824: .I adb
825: but are not used subsequently.
826: Numbered variables are reserved for communication
827: as follows.
828: .TP
829: .BR 0 ", " 1 ", ..."
830: The offset parts of the first, second, ...
831: operands of the last instruction printed.
832: Meaningless if the operand was a register.
833: .br
834: .ns
835: .TP
836: .B 9
837: The count on the last
838: .B $<
839: or
840: .B $<<
841: command.
842: .PP
843: On entry the following are set
844: from the system header in the
845: .IR corfil .
846: If
847: .I corfil
848: does not appear to be a
849: core image then
850: these values are set from
851: .IR objfil .
852: .TP
853: .B b
854: The base address of the data segment.
855: .PD 0
856: .TP
857: .B d
858: The data segment size.
859: .TP
860: .B e
861: The entry point.
862: .TP
863: .B m
864: The `magic' number
865: .RI ( a.out (5)).
866: .TP
867: .B s
868: The stack segment size.
869: .TP
870: .B t
871: The text segment size.
872: .PD
873: .SS Addresses
874: The address in a file associated with
875: a written address is determined by a mapping
876: associated with that file.
877: Each mapping is represented by one or more quadruples
878: .RI ( "t, b, e, f" ),
879: each mapping addresses of type
880: .I t
881: (instruction,
882: data,
883: user block)
884: in the range
885: .I b
886: through
887: .I e
888: to the part of the file
889: beginning at
890: address
891: .IR f .
892: An address
893: .I a
894: of type
895: .I t
896: is mapped
897: to a file address
898: by finding a quadruple
899: of type
900: .IR t ,
901: for which
902: .IR b \*(LE a < e ;
903: the file address
904: is
905: .IR address + f \(mi b .
906: As a special case,
907: if an instruction space address is not found,
908: a second search is made
909: for the same address in data space.
910: .PP
911: Typically,
912: the text segment of a program
913: is mapped as instruction space,
914: the data and bss segments
915: as data space.
916: If
917: .I objfil
918: is an
919: .IR a.out,
920: or if
921: .I corfil
922: is a core image
923: or process file,
924: maps are set accordingly.
925: Otherwise,
926: a single `data space'
927: map is set up,
928: with
929: .I b
930: and
931: .I f
932: set to zero,
933: and
934: .I e
935: set to a huge number;
936: thus the entire file can be examined
937: without address translation.
938: .PP
939: The
940: .B ?
941: and
942: .B /
943: commands attempt to examine
944: instruction and data space
945: respectively.
946: .B ?*
947: tries for data space
948: (in
949: .IR objfil );
950: .B /*
951: accesses instruction space
952: (in
953: .IR corfil ).
954: .PP
955: Registers in
956: process and core images
957: are a special case;
958: they live in a special `register' address space.
959: The contents of register 0
960: are located at address
961: .BR %0 ;
962: register 1
963: at
964: .BR %4
965: (if registers are 4 bytes long);
966: and so on.
967: .B %
968: addresses
969: are mapped to the registers
970: for the `current frame,'
971: set by local variable references,
972: and reset to the outermost frame
973: (the `real' registers)
974: whenever a process runs
975: or a stack trace is requested.
976: .PP
977: Simulated memory management
978: translations
979: (the
980: .B $k
981: and
982: .B $p
983: commands)
984: are done before the mapping described above.
985: .SH FILES
986: .F a.out
987: .br
988: .F core
989: .br
990: .F /usr/lib/adb/*
991: parameter files
992: .br
993: .F /proc/*
994: .SH SEE\ ALSO
995: .IR cin (1),
996: .IR pi (9.1),
997: .IR nm (1),
998: .IR proc (4),
999: .IR a.out (5),
1000: .IR bigcore (1)
1001: .br
1002: J. F. Maranzano and S. R. Bourne,
1003: `A Tutorial Introduction to ADB' in
1004: Bell Laboratories,
1005: .I UNIX Programmer's Manual,
1006: Volume\ 2,
1007: Holt, Rinehart and Winston (1984)
1008: .SH DIAGNOSTICS
1009: `Adb' when there is no current command or format.
1010: Exit status is 0, unless last command failed or
1011: returned nonzero status.
1012: .SH BUGS
1013: Either the explanation
1014: or the implementation
1015: of register variables
1016: is too complex and arcane.
1017: .SH MACHINE DEPENDENCIES
1018: .SS PDP-11
1019: Short integers (printed by
1020: .B r
1021: format)
1022: are 2 bytes;
1023: long integers
1024: (printed by
1025: .B R
1026: format)
1027: are 4 bytes.
1028: Addresses printed by
1029: .B a
1030: format are 2 bytes.
1031: .PP
1032: Register variables match the hardware in the
1033: obvious way:
1034: .B r0
1035: is at address
1036: .BR %0 ,
1037: .B r1
1038: at
1039: .BR %2 ,
1040: and so on.
1041: .PP
1042: The default output radix
1043: is octal.
1044: .PP
1045: .B $k
1046: and
1047: .B $p
1048: are unimplemented.
1049: .SS VAX
1050: Short integers are 2 bytes,
1051: long integers are 4 bytes,
1052: addresses are 4 bytes.
1053: .PP
1054: Register variables match the hardware in the
1055: obvious way: r0
1056: is at address
1057: .BR %0 ,
1058: r1 at
1059: .BR %4 ,
1060: and so on.
1061: .PP
1062: The default output radix
1063: is hexadecimal.
1064: .PP
1065: .B $k
1066: sets the system base register pointer to
1067: .IR address .
1068: System space addresses are thereafter
1069: mapped according to the system page table
1070: starting at that physical address.
1071: An
1072: .I address
1073: of zero
1074: turns off mapping.
1075: .PP
1076: .B $p
1077: sets the process control block pointer to
1078: .IR address ;
1079: user space addresses are thereafter
1080: translated according to the user page tables
1081: described by the PCB.
1082: Kernel mapping must already be in effect.
1083: .I Address
1084: may be a physical address
1085: (that of the PCB)
1086: or the system space virtual address
1087: of a page table entry
1088: pointing to the PCB
1089: (the number stored in
1090: .IR p_addr ).
1091: If
1092: .I address
1093: is zero,
1094: user mapping is turned off;
1095: addresses less than
1096: 0x80000000
1097: will be treated as physical addresses.
1098: .PP
1099: The command
1100: .L "$<crash"
1101: will initialize registers
1102: and mapping from a kernel crash dump.
1103: .SS Cray
1104: Short integers are 2 bytes;
1105: long integers are 8 bytes.
1106: Addresses are 8 bytes.
1107: .PP
1108: Registers are funny,
1109: and yet to be described.
1110: .PP
1111: The default output radix is octal.
1112: .PP
1113: .B $k
1114: and
1115: .B $p
1116: are unimplemented.
This archive runs on limited infrastructure. Preserving old code on modern bandwidth. Automated agents are requested to crawl responsibly.