harvey/sys/man/1/db

.TH DB 1
.SH NAME
db \- debugger
.SH SYNOPSIS
.B db
[
.I option ...
]
[
.I textfile
]
[
.I pid
]
.SH DESCRIPTION
.I Db
is a general purpose debugging program.
It may be used to examine files and to provide
a controlled environment for the execution
of Plan 9 programs.
.PP
A
.I textfile
is a file containing the text and initialized
data of an executable program.
A
.I memfile
is the memory image of an executing process.  It is
usually accessed via the process id
.RI ( pid )
of the process in
.BI /proc/ pid /mem\f1.
A
.I memfile
contains the text, data, and saved registers and
process state.  A
.I map
associated with each
.I textfile
or
.I memfile
supports accesses to instructions and data in the file;
see `Addresses'.
.PP
An argument consisting entirely of digits is assumed
to be a process id; otherwise, it is the name of a
.IR textfile .
When a
.I textfile
is given, the textfile map
is associated with it.
If only a
.I pid
is given, the textfile map is
associated with
.BI /proc/ pid /text\f1.
When a
.I pid
is given, the memfile map is associated with
.BI /proc/ pid /mem\f1;
otherwise it is undefined and accesses to the
.I memfile
are not permitted.
.PP
Commands to
.I db
are read from the standard input and
responses are to the standard output.
The options are
.TP
.BI -k
Use the kernel stack of process
.IR pid 
to debug the executing kernel process.
If
.I textfile
is not specified, file
.BI / $cputype /9 type
is used, where
.I type
is the second word in
.BR $terminal .
.TP
.B -w
Create
.I textfile
and
.I memfile
if they don't exist; open them for writing
as well as reading.
.TP
.BI -I path
Directory in which to look for relative path names in
.B $<
and
.B $<<
commands.
.TP
.BI -m machine
Assume instructions are for the given CPU type
(any standard architecture name, such as
.B amd64
or
.BR 386 ,
plus
.B mipsco
and
.BR sunsparc ,
which cause disassembly to the manufacturer's syntax)
instead of using the magic number to select
the CPU type.
.PP
Most
.I db
commands have the following form:
.IP
.RI [ address ]
.RB [ ,
.IR count ]
.RI [ command ]
.PP
If
.I address
is present then the current position, called `dot',
is set to
.IR address .
Initially dot
is set to 0.
Most commands are repeated
.I count
times with
dot advancing between repetitions.
The default
.I count
is 1.
.I Address
and
.I count
are expressions.
Multiple commands on one line must be separated by
.LR ; .
.SS Expressions
Expressions are evaluated as long
.IR ints .
.TP 7.2n
.B .
The value of dot.
.TP 7.2n
.B +
The value of dot
incremented by the current increment.
.TP 7.2n
.B ^
The value of dot
decremented by the current increment.
.TP 7.2n
.B \&"
The last
.I address
typed.
.TP 7.2n
.I integer
A number, in decimal radix by default.
The prefixes
.L 0
and
.L 0o
and
.L 0O
(zero oh) force interpretation
in octal radix; the prefixes
.L 0t
and
.L 0T
force interpretation in
decimal radix; the prefixes
.LR 0x ,
.LR 0X ,
and
.L #
force interpretation in
hexadecimal radix.
Thus
.LR 020 ,
.LR 0o20 ,
.LR 0t16 ,
and
.L #10 
all represent sixteen.
.TP 7.2n
.IB integer . fraction
A single-precision floating point number.
.TP 7.2n
.BI \' c\| \'
The
16-bit
value of a character.
.L \e
may be used to escape a
.LR \' .
.TP 7.2n
.BI < name
The value of
.IR name ,
which is a register name.
The register names are
those printed by the
.B $r
command.
.TP 7.2n
.I symbol
A
.I symbol
is a sequence
of upper or lower case letters, underscores or
digits, not starting with a digit.
.L \e
may be used to escape other characters.
The location of the
.I symbol
is calculated from the symbol table
in
.IR textfile .
.TP 7.2n
.IB routine . name
The address of the variable
.I name
in the specified
C routine.
Both
.I routine
and
.I name
are
.IR symbols .
If
.I name
is omitted the value is the address of the
most recently activated stack frame
corresponding to
.IR routine ;
if
.I routine
is omitted,
the active procedure
is assumed.
.TP 7.2n
.IB file : integer
The address of the instruction corresponding
to the source statement at the indicated
line number of the file.  If the source line contains
no executable statement, the address of the
instruction associated with the nearest
executable source line is returned.  Files
begin at line 1.  If multiple files of the same
name are loaded, an expression of this form resolves
to the first file encountered in the symbol table.
.TP 7.2n
.BI ( exp )
The value of the expression
.IR exp .
.LP
.I  Monadic operators
.RS
.TP 7.2n
.BI * exp
The contents of the location addressed
by
.I exp
in
.IR memfile .
.TP 7.2n
.BI @ exp
The contents of the location addressed by
.I exp
in
.IR textfile .
.TP 7.2n
.BI - exp
Integer negation.
.TP 7.2n
.BI ~ exp
Bitwise complement.
.TP 7.2n
.BI % exp
When used as an
.IR address ,
.I exp
is an offset into the segment named
.IR ublock ;
see `Addresses'.
.RE
.LP
.I "Dyadic\ operators"
are left-associative
and are less binding than monadic operators.
.RS
.TP 7.2n
.IB e1 + e2
Integer addition.
.TP 7.2n
.IB e1 - e2
Integer subtraction.
.TP 7.2n
.IB e1 * e2
Integer multiplication.
.TP 7.2n
.IB e1 % e2
Integer division.
.TP 7.2n
.IB e1 & e2
Bitwise conjunction.
.TP 7.2n
.IB e1 | e2
Bitwise disjunction.
.TP 7.2n
.IB e1 # e2
.I E1
rounded up to the next multiple of
.IR e2 .
.RE
.DT
.SS Commands
Most commands have the following syntax:
.TP .5i
.BI ? f
Locations starting at
.I address
in
.I  textfile
are printed according to the format
.IR f .
.TP
.BI / f
Locations starting at
.I address
in
.I  memfile
are printed according to the format
.IR f .
.TP
.BI = f
The value of
.I address
itself is printed according to the format
.IR f .
.PP
A
.I format
consists of one or more characters that specify a style
of printing.
Each format character may be preceded by a decimal integer
that is a repeat count for the format character.
If no format is given then the last format is used.
.PP
Most format letters fetch some data,
print it,
and advance (a local copy of) dot
by the number of bytes fetched.
The total number of bytes in a format becomes the
.IR "current increment" .
.ta 2.5n .5i
.RS
.TP
.PD 0
.B o
Print two-byte integer in octal.
.TP
.B O
Print four-byte integer in octal.
.TP
.B q
Print two-byte integer in signed octal.
.TP
.B Q
Print four-byte integer in signed octal.
.TP
.B d
Print two-byte integer in decimal.
.TP
.B D
Print four-byte integer in decimal.
.TP
.B V
Print eight-byte integer in decimal.
.TP
.B Z
Print eight-byte integer in unsigned decimal.
.TP
.B x
Print two-byte integer in hexadecimal.
.TP
.B X
Print four-byte integer in hexadecimal.
.TP
.B Y
Print eight-byte integer in hexadecimal.
.TP
.B u
Print two-byte integer in unsigned decimal.
.TP
.B U
Print four-byte integer in unsigned decimal.
.TP
.B f
Print
as a single-precision floating point number.
.TP
.B F
Print double-precision floating point.
.TP
.B b
Print the addressed byte in hexadecimal.
.TP
.B c
Print the addressed byte as an
.SM ASCII
character.
.TP
.B C
Print the addressed byte as a character.
Printable
.SM ASCII
characters
are represented normally; others
are printed in the form
.BR \exnn .
.TP
.B s
Print the addressed characters, as a
.SM UTF
string, until a zero byte
is reached.
Advance dot
by the length of the string,
including the zero terminator.
.TP
.B S
Print a string using 
the escape convention (see
.B C
above).
.TP
.B r
Print as
.SM UTF
the addressed two-byte integer (rune).
.TP
.B R
Print as
.SM UTF
the addressed two-byte integers as runes
until a zero rune is reached.
Advance dot
by the length of the string,
including the zero terminator.
.TP
.B i
Print as machine instructions.  Dot is
incremented by the size of the instruction.
.TP
.B I
As
.B i
above, but print the machine instructions in
an alternate form if possible:
.B sunsparc
and
.B mipsco
reproduce the manufacturers' syntax.
.TP
.B M
Print the addressed machine instruction in a
machine-dependent hexadecimal form.
.TP
.B a
Print the value of dot
in symbolic form.
Dot is unaffected.
.TP
.B A
Print the value of dot
in hexadecimal.
Dot is unaffected.
.TP
.B z
Print the function name, source file, and line number
corresponding to dot (textfile only). Dot is unaffected.
.TP
.B p
Print the addressed value in symbolic form.
Dot is advanced by the size of a machine address.
.TP
.B t
When preceded by an integer, tabs to the next
appropriate tab stop.
For example,
.B 8t 
moves to the next 8-space tab stop.
Dot is unaffected.
.TP
.B n
Print a newline.
Dot is unaffected.
.tr '"
.TP
.BR ' ... '
Print the enclosed string.
Dot is unaffected.
.br
.tr ''
.TP
.B ^
Dot is decremented by the current increment.
Nothing is printed.
.TP
.B +
Dot is incremented by 1.
Nothing is printed.
.TP
.B -
Dot is decremented by 1.
Nothing is printed.
.RE
.PD
.LP
Other commands include:
.TP
newline
Update dot by the current increment.
Repeat the previous command with a
.I count
of 1.
.TP
.RB [ ?/ ] l "\fI value mask\fR"
Words starting at dot
are masked with
.I mask
and compared with
.I value
until
a match is found.
If
.B l
is used,
the match is for a two-byte integer;
.B L
matches four bytes.
If no match is found then dot
is unchanged; otherwise dot
is set to the matched location.
If
.I mask
is omitted then ~0 is used.
.TP
.RB [ ?/ ] w "\fI value ...\fR"
Write the two-byte
.I value
into the addressed
location.
If the command is
.BR W ,
write four bytes.
.TP
.RB [ ?/ ] "m\fI s b e f \fP" [ ?\fR]
.br
New values for
.RI ( b,\ e,\ f )
in the segment named
.I s
are recorded.  Valid segment names are
.IR text ,
.IR data ,
or 
.IR ublock .
If less than three address expressions are given,
the remaining parameters are left unchanged.
If the list is terminated by
.L ?
or
.L /
then the file
.RI ( textfile
or
.I memfile
respectively) is used
for subsequent requests.
For example,
.L /m?
causes
.L /
to refer to
.IR textfile .
.TP
.BI > name
Dot is assigned to the variable or register named.
.TP
.B !
The rest of the line is passed to
.IR rc (1)
for execution.
.TP
.BI $ modifier
Miscellaneous commands.
The available 
.I modifiers 
are:
.RS
.TP
.PD 0
.BI < f
Read commands from the file
.IR f .
If this command is executed in a file, further commands
in the file are not seen.
If
.I f
is omitted, the current input stream is terminated.
If a
.I count
is given, and is zero, the command is ignored.
.TP
.BI << f
Similar to
.B <
except it can be used in a file of commands without
causing the file to be closed.
There is a (small) limit to the number of
.B <<
files that can be open at once.
.br
.ns
.TP
.BI > f
Append output to the file
.IR f ,
which is created if it does not exist.
If
.I f
is omitted, output is returned to the terminal.
.TP
.B ?
Print process id, the condition which caused stopping or termination,
the registers and the instruction addressed by
.BR pc .
This is the default if
.I modifier
is omitted.
.TP
.B r
Print the general registers and
the instruction addressed by
.BR pc .
Dot is set to
.BR pc .
.TP
.B R
Like
.BR $r ,
but include miscellaneous processor control registers
and floating point registers.
.TP
.B f
Print floating-point register values as
single-precision floating point numbers.
.TP
.B F
Print floating-point register values as
double-precision floating point numbers.
.TP
.B b
Print all breakpoints
and their associated counts and commands.  `B' produces the same results.
.TP
.B c
Stack backtrace.
If
.I address
is given, it specifies the address of a pair of 32-bit
values containing the
.B sp
and
.B pc
of an active process.  This allows selecting
among various contexts of a multi-threaded
process.
If
.B C
is used, the names and (long) values of all
parameters,
automatic
and static variables are printed for each active function.
If
.I count
is given, only the first
.I count
frames are printed.
.TP
.B a
Attach to the running process whose pid
is contained in
.IR address .
.TP
.B e
The names and values of all
external variables are printed.
.TP
.B w
Set the page width for output to
.I address
(default 80).
.TP
.B q
Exit from
.IR db .
.TP
.B m
Print the address maps.
.TP
.B k
Simulate kernel memory management.
.TP
.BI M machine
Set the
.I machine
type used for disassembling instructions.
.PD
.RE
.TP
.BI : modifier
Manage a subprocess.
Available modifiers are:
.RS
.TP
.PD 0
.BI h
Halt
an asynchronously running process to allow breakpointing.
Unnecessary for processes created under
.IR db ,
e.g. by
.BR :r .
.TP
.BI b c
Set breakpoint at
.IR address .
The breakpoint is executed
.IR count \-1
times before
causing a stop.
Also, if a command
.I c
is given it is executed at each
breakpoint and if it sets dot to zero
the breakpoint causes a stop.
.TP
.B d
Delete breakpoint at
.IR address .
.TP
.B r
Run
.I textfile
as a subprocess.
If
.I address
is given the
program is entered at that point; otherwise
the standard entry point is used.
.I Count
specifies how many breakpoints are to be
ignored before stopping.
Arguments to the subprocess may be supplied on the
same line as the command.
An argument starting with < or > causes the standard
input or output to be established for the command.
.TP
.BI c s
The subprocess is continued.
If
.I s
is omitted
or nonzero,
the subprocess
is sent the note that caused it to stop.
If 0
is specified,
no note is sent.
(If the stop was due to a breakpoint or single-step,
the corresponding note is elided before continuing.)
Breakpoint skipping is the same
as for
.BR r .
.TP
.BI s s
As for
.B c
except that
the subprocess is single stepped for
.I count
machine instructions.
If a note is pending,
it is received
before the first instruction is executed.
If there is no current subprocess then
.I textfile
is run
as a subprocess as for
.BR r .
In this case no note can be sent; the remainder of the line
is treated as arguments to the subprocess.
.TP
.BI S s
Identical to
.B s
except the subprocess is single stepped for
.I count
lines of C source.  In optimized code, the correspondence
between C source and the machine instructions is
approximate at best.
.TP
.BI x
The current subprocess, if any, is released by
.I db
and allowed to continue executing normally.
.TP
.B k
The current subprocess, if any, is terminated.
.TP
.BI n c
Display the pending notes for the process.
If
.I c
is specified, first delete
.I c'th
pending note.
.PD
.RE
.SS Addresses
The location in a file or memory image associated with
an address is calculated from a map
associated with the file.
Each map contains one or more quadruples
.RI ( "t, b, e, f" ),
defining a segment named
.I t
(usually, 
.IR text ,
.IR data ,
or
.IR ublock )
mapping addresses in the range
.I b
through
.I e
to the part of the file
beginning at
offset
.IR f .
The memory model of a Plan 9 process assumes
that segments are disjoint.  There
can be more than one segment of a given type (e.g., a process
may have more than one text segment) but segments
may not overlap.
An address
.I a
is translated
to a file address
by finding a segment
for which
.IR b ≤ a < e ;
the location in the file
is then
.IR address + f \- b .
.PP
Usually,
the text and initialized data of a program
are mapped by segments called 
.I text
and
.IR data .
Since a program file does not contain bss, stack or ublock data,
these data are
not mapped by the data segment.
The text segment is mapped similarly in
a normal (i.e., non-kernel)
.IR memfile .
However, the segment called 
.I data
maps memory from the beginning of the program's data space to
the base of the ublock.
This region contains the program's static data, the bss, the
heap and the stack.  A segment
called
.I ublock
maps the page containing its registers and process state.
.PP
Sometimes it is useful to define a map with a single segment
mapping the region from 0 to 0xFFFFFFFF; a map of this type
allows the entire file to be examined
without address translation.
.PP
Registers are saved at a machine-dependent offset in the ublock.
It is usually not necessary to know this offset; the
.BR $r ,
.BR $R ,
.BR $f ,
and
.BR $F
commands calculate it and display the register contents.
.PP
The
.B $m
command dumps the currently active maps.  The
.B ?m
and
.B /m
commands modify the segment parameters in the
.I textfile
and
.I memfile
maps, respectively.
.SH EXAMPLES
To set a breakpoint at the beginning of
.B write()
in extant process 27:
.IP
.EX
% db 27
:h
write:b
:c
.EE
.PP
To examine the Plan 9 kernel stack for process 27:
.IP
.EX
% db -k 27
$C
.EE
.PP
Similar, but using a kernel named
.BR test :
.IP
.EX
% db -k test 27
$C
.EE
.PP
To set a breakpoint at the entry of function
.B parse
when the local variable
.B argc
in
.B main
is equal to 1:
.IP
.EX
parse:b *main.argc-1=X
.EE
.PP
This prints the value of
.B argc-1
which as a side effect sets dot; when
.B argc
is one the breakpoint will fire.
Beware that local variables may be stored in registers; see the
BUGS section.
.PP
Debug process 127 on remote machine
.BR kremvax :
.IP
.EX
% import kremvax /proc
% db 127
$C
.EE
.SH FILES
.B /proc/*/text
.br
.B /proc/*/mem
.br
.B /proc/*/ctl
.br
.B /proc/*/note
.SH "SEE ALSO"
.IR acid (1),
.IR nm (1),
.IR proc (3)
.SH SOURCE
.B /sys/src/cmd/db
.SH DIAGNOSTICS
Exit status is null, unless the last command failed or
returned non-null status.
.SH BUGS
Examining a local variable with
.I routine.name
returns the contents of the memory allocated for the variable, but
with optimization (suppressed by the
.B -N
compiler flag) variables often reside in registers.
Also, on some architectures, the first argument is always
passed in a register.
.PP
Variables and parameters that have been
optimized away do not appear in the
symbol table, returning the error 
.IR "bad local variable"
when accessed by
.IR db .
.PP
Because of alignment incompatibilities, Motorola 68000
series machines can not be debugged remotely from a
processor of a different type.
.PP
Breakpoints should not be set on instructions scheduled
in delay slots.  When a program stops on such a breakpoint,
it is usually impossible to continue its execution.