Enforcer.doc
Enforcer()
FindHit()
LawBreaker()
Move4K()
SegTracker()
Enforcer
NAME
Enforcer V37 - An advanced version of Enforcer - Requires V37
SYNOPSIS
Enforcer - A tool to watch for illegal memory accesses
FUNCTION
Enforcer will use the MMU in the advanced 680x0 processors
to set up MMU tables to watch for illegal accesses to memory
such as the low-page and non-existent pages.
To use, run Enforcer (plus any options you may wish)
If you wish to detach, just use RUN >NIL: INPUTS
The options for Enforcer are as follows:
QUIET/S - This tells Enforcer to not complain about any invalid
access and to just build MMU tables for cache setting
reasons -- mainly used in conjunction with an
Amiga BridgeBoard in a 68030 environment so that
the system can run with the data cache turned on.
In this case,
RUN >NIL: Enforcer QUIET
should be placed into the startup-sequence right
after SetPatch.
TINY/S - This tells Enforcer to output a minimal hit. The
output is basically the first line of the Enforcer
hit. (see below)
SMALL/S - This tells Enforcer to output the hit line, the
USP: line, and the Name: line. (This means that
no register or stack display will be output)
SHOWPC/S - This tells Enforcer to also output the two lines
that contain the memory area around the PC where
the hit happened. Useful for disassembly.
This option will not do anything if QUIET, SMALL or
TINY output modes are selected.
STACKLINES/K/N - This lets you pick the number of lines of stack
backtrace to display. The default is 2. If set
to 0, no stack backtrace will be displayed. There
is NO ENFORCED LIMIT on the number of lines.
STACKCHECK/S - This option tells Enforcer that you wish all of
the long words displayed in the stack to be checked
against the global seglists via SegTracker.
This will tell you what seglist various return
addresses are on the stack. If you are not
displaying stack information in the Enforcer hit
then STACKCHECK will have nothing to check.
If you are displaying stack information, then
each long word will be check and only those which
are in one of the tracked seglists will be
displayed in a SegTracker line.
The output will show the PC address first and
then work its way back on the stack such that you
can read it from bottom up as the order of calling
or from top down as the stack-frame backtrace.
AREGCHECK/S - This option tells Enforcer that you wish all of
the values in the Address Registers checked via
SegTracker, much like STACKCHECK.
DREGCHECK/S - This option tells Enforcer that you wish all of
the values in the Data Registers checked via
SegTracker, much like STACKCHECK.
DATESTAMP/S - This makes Enforcer output a date and time with each
hit. Due to the nature of the way Enforcer must
work, the time can not be read during the Enforcer
hit itself so the time output will be the last time
value the main Enforcer task set up. Enforcer will
update this value every second as to try to not
use any real CPU time. The time displayed in the
hit will thus be exact.
(Assuming the system clock is correct.)
The date is output before anything from the hit
other than the optional introduction string.
DEADLY/S - This makes Enforcer be a bit more nasty. Normally,
when an illegal read happens, Enforcer returns 0
as the result of this read. With this option,
Enforcer will return $ABADFEED as the read data.
This option can make programs with Enforcer hits
cause even more hits.
FSPACE/S - This option will make the special $00F00000 address
space available for writing to. This is useful for
those people with $00F00000 boards. Mainly Commodore
internal development work -- should only be used
in that enviroment.
NORELOADKICK/S - In A3000 SuperKickstart machines, Enforcer would
force the Kickstart to be reloaded on a reboot.
This option will disable this feature of Enforcer.
Note that on some systems, reboots may not be
reliable in this case. Also note that if software
tries to reboot the machine and it does not use the
2.x ColdReboot() function, the system will not
correctly reboot. (It may, in fact, hang)
VERBOSE/S - This option will make Enforcer display information
as to the mapping of the I/O boards and other
technical information. This information maybe useful
in specialized debugging.
LED/K/N - This option lets you specify the speed at which
the LED will be toggled for each Enforcer hit.
The default is 1 (which is like it always was)
Setting it to 0 will make Enforcer not touch
the LED. Using a larger value will make the
flash take longer (such that it can be noticed
when doing I/O models other than the default
serial output) The time that the flash will
take is a bit more than 1.3 microseconds times
the number. So 1000 will be a bit more than
1.3 milliseconds. (Or 1000000 is a bit more than
1.3 seconds.)
PARALLEL/S - This option will make Enforcer use the parallel port
hardware rather than the serial port for output.
RAWIO/S - This option will make Enforcer stuff the hit report
(special IO) into an internal buffer and then from the main
Enforcer process output the results via the
RawPutChar() EXEC debugging LVO. Since the output
happens on the Enforcer task it is possible for a
hit that ends in a system crash to not be able to
be reported. This option is here such that tools
which can redirect debugging output can redirect
the Enforcer output too.
FILE/K - This option will make Enforcer output the hit report
(special IO) but to a file insted of sending it to the hardware
directly or using the RAWIO LVO. A good example of
such a file is CON:0/0/640/100/HIT/AUTO/WAIT.
Another thing that can be done is to have a program
sit on a named pipe and have Enforcer output to it.
This program can then do whatever it feels like with
the Enforcer hits. (Such as decode them, etc.)
*NOTE* It is not a good idea to have Enforcer hits
go to a file on a disk as if the system crashes
during/after the Enforcer hit, the disk may
become corrupt.
STDIO/S - This option will make Enforcer output the hit report
(special IO) to STDOUT. This option only works from the CLI as it
requires STDOUT. It is best used with redirection or
pipes.
BUFFERSIZE/K/N - This lets you set Enforcer's internal output buffer
for the special I/O options. This option is only
valid with the RAWIO, FILE, or STDIO options.
The minimum setting is 8000. The default is 8000.
Having the right amount of buffer is rather
important for the special I/O modes. The reason
is due to the fact that no operating system calls
can be made from a bus error. Thus, in the
special I/O mode, Enforcer must store the output
in this buffer and, via some special magic,
wake up the Enforcer task to read the buffer and
write it out as needed. However, if a task is
in Forbid() or Disable() when the Enforcer hit
happens, the Enforcer task will not be able to
output the results of the hit. This buffer lets
a number of hits happen even if the Enforcer task
was unable to do the I/O. If the number of
hits that happen before the I/O was able to
run gets too large, the last few hits will either
be cut off completely or contain only partial
information.
INTRO/K - This optional introduction string will be output
at the start of every Enforcer hit. For example:
INTRO="*NBad Program!" The default is no string.
PRIORITY/K/N - This lets you set Enforcer's I/O task priority.
The default for this priority is 99. In some
special cases, you may wish to adjust this.
It is, however, recommended that if you are using
one of the special I/O options (RAWIO, FILE, or
STDIO) that you keep the priority rather high.
If the priority you supply is outside of the
valid task priority range (-127 to 127) Enforcer
will use the default priority.
NOALERTPATCH/S - This option disables the patching of the EXEC
Alert() function. Normally Enforcer will patch
this function to provide information as to what
called Alert() and to prevent the Enforcer hits
that a call to Alert() would cause.
ON/S - Mainly for completeness. If not specified, it
is assumed you want to turn ON Enforcer.
QUIT=OFF/S - Tells Enforcer to turn off. Enforcer can also be
stopped by sending a CTRL-C to its process.
RESULTS
When running, a set of MMU tables that map addresses that are not
in the system's address map as invalid such that any access to them
will cause an access fault. Enforcer will then display this fact
and generate a diagnostic message as to what the illegal access
was. The first memory page (the one starting at location 0) is
also marked as invalid as many programming errors cause invalid
access to these addresses. These addresses are completely off
limits to applications.
When an access violation happens, a report such as the following
is output.
03-Apr-93 21:26:18
WORD-WRITE to 00000000 data=4444 PC: 07895CA4
USP: 078D692C SR: 0000 SW: 0729 (U0)(-)(-) TCB: 078A2690
Data: DDDD0000 DDDD1111 DDDD2222 DDDD3333 DDDD4444 DDDD5555 DDDD6666 DDDD7777
Addr: AAAA0000 AAAA1111 AAAA2222 AAAA3333 AAAA4444 AAAA5555 07800804 --------
Stck: 00000000 07848E1C 00009C40 078A30B4 BBBBBBBB BBBBBBBB BBBBBBBB BBBBBBBB
Stck: BBBBBBBB BBBBBBBB BBBBBBBB BBBBBBBB BBBBBBBB 078E9048 00011DA8 DEADBEEF
----> 07895CA4 - "lawbreaker" Hunk 0000 Offset 0000007C
PC-8: AAAA1111 247CAAAA 2222267C AAAA3333 287CAAAA 44442A7C AAAA5555 31C40000
PC *: 522E0127 201433FC 400000DF F09A522E 012611C7 00CE4EAE FF7642B8 0324532E
Name: "New_Shell" CLI: "lawbreaker" Hunk 0000 Offset 0000007C
LONG-READ from AAAA4444 PC: 07895CA8
USP: 078D692C SR: 0015 SW: 0749 (U0)(F)(-) TCB: 078A2690
Data: DDDD0000 DDDD1111 DDDD2222 DDDD3333 DDDD4444 DDDD5555 DDDD6666 DDDD7777
Addr: AAAA0000 AAAA1111 AAAA2222 AAAA3333 AAAA4444 AAAA5555 07800804 --------
Stck: 00000000 07848E1C 00009C40 078A30B4 BBBBBBBB BBBBBBBB BBBBBBBB BBBBBBBB
Stck: BBBBBBBB BBBBBBBB BBBBBBBB BBBBBBBB BBBBBBBB 078E9048 00011DA8 DEADBEEF
----> 07895CA8 - "lawbreaker" Hunk 0000 Offset 00000080
PC-8: 247CAAAA 2222267C AAAA3333 287CAAAA 44442A7C AAAA5555 31C40000 522E0127
PC *: 201433FC 400000DF F09A522E 012611C7 00CE4EAE FF7642B8 0324532E 01266C08
Name: "New_Shell" CLI: "lawbreaker" Hunk 0000 Offset 00000080
25-Jul-93 17:15:06
Alert !! Alert 35000000 TCB: 07642F70 USP: 07657C10
Data: 00000000 DDDD1111 DDDD2222 DDDD3333 0763852A DDDD5555 DDDD6666 35000000
Addr: AAAA0000 AAAA1111 AAAA2222 AAAA3333 AAAA4444 0763852A 07400810 --------
Stck: 076385A0 00000000 0752EE9A 00002800 07643994 00000000 0762F710 076305F0
----> 076385A0 - "lawbreaker" Hunk 0000 Offset 00000098
Here is a breakdown of what these reports are saying:
In the first report, the first line is the date stamp.
The first line of each report describes the access violation
and where it happened from. In the case of a WRITE, the data
that was being written will be displayed as well. If an instruction
mode access caused the fault, there will be an (INST) in the line.
The first line may also contain the BUS ERROR message. This will
be displayed when an address that is valid in the system lists
causes a physical bus fault during the access. This usually
will happen with plug-in cards or when a hardware problem causes
some form of system fault. Watch out, if this does show up, your
system may be unstable and/or unreliable.
The second line (starts USP:) displays the USER stack pointer (USP),
the status register (SR:), the special status word (SW:). It then
displays the supervisor/user state and the interrupt level. This
will be from (U0) to (U7) or (S0) to (S7) (S=Supervisor) Next
is the forbid state (F=forbid, -=not) and the disable state (D or -)
of the task that was running when the access fault took place.
Finally, the task control block address is displayed (TCB:)
The next two lines contain the data and address register dumps from
when the access fault happened. Note that A7 is not listed here.
It is the stack pointer and is listed as USP: in the line above.
Then come the lines of stack backtrace. These lines show the
data on the stack. If the stack is in invalid memory, Enforcer will
display a message to that fact.
If SegTracker was installed before Enforcer, the "---->" lines
will display in which seglist the given addresses are in based on the
global tracking that SegTracker does. (See docs on SegTracker)
If no seglist match is found, no lines will be displayed.
One line will be displayed for each of the stack longwords asked
for (see the STACKCHECK option) and one line for the PC address of
the Enforcer hit. (The PC line is always checked for is SegTracker
is installed.) The lines are in order: hit, first stack find,
second stack find, etc. This is useful for tracking down who
called the routine that caused the Enforcer hit.
Next, optionally, comes the data around the program counter when the
access fault happened. The first line (PC-8:) is the 8 long-words
before the program counter. The second line starts at the program
counter and goes for 8 long words.
The last line displays the name of the task that was running when
the access fault took place. If the task was a CLI, it will display
the name of the CLI command that was running. If the access fault
was found to have happened within the seglist of a loaded program,
the segment number and the offset from the start of the segment will
be displayed. (Note that this works for any LoadSeg()'ed process)
Note that the name will display as "Processor Interrupt Level x"
if the access happened in an interrupt.
The other output that could happen is when a program or the OS
calls the EXEC Alert function. Enforcer catches these calls
and will display the alert information as seen above. (With the
data and time if needed)
WARNING
Enforcer is for software testing. In this role it is vital.
Software that causes Enforcer hits may well not be able to run on
newer hardware. (Enforcer hits of high addresses on systems not
running Enforcer but with a 68040 will most likely crash the system)
Future systems and hardware will make this even more important. The
system can NOT survive software that causes Enforcer hits.
However, Enforcer is NOT a system protector. As a side effect, it
may well keep a system from crashing when Enforcer hits happen, but
it may just as well make the software crash earlier. Enforcer is
mainly a development and testing tool.
Enforcer causes no ill effects with correctly working software.
If a program fails to work while Enforcer is active, you should
contact the developer of that program.
NOTES
This is Enforcer V37. Bryce Nesbitt came up with the original
"Enforcer" that has been instrumental to the improvement in the
quality of software on the Amiga. The Amiga users and developers
owe him a great deal for this. Thank you Bryce! Enforcer V37,
however, is a greatly enhanced and more advanced tool.
Enforcer V37 came about due to a number of needs. These included
the need for more output options and better performance. It also
marks the removal of all kludges that were in the older versions.
Also, some future plans required some of these changes...
In addition, the complete redesign was needed in order to also
support the 68040. The internal design of Enforcer is now set up
such that CPU/MMU specific code can be cleanly accessed from the
general house keeping aspect of the code. The MMU bus error
handling is, however, 100% CPU specific.
Since AbsExecBase is in low memory, reads of this address are slower
with Enforcer running. Caching AbsExecBase locally is highly
recommended since it is in CHIP memory and on systems with FAST
memory, it will be faster to access the local cached value. (In
addition to the performance increase when running Enforcer) Note
that doing many reads of location 4 will hurt interrupt performance.
When the Amiga produces an ALERT, EXEC places some magic numbers
into some special locations in low memory. The exact pattern
changes between versions of the operating system.
Enforcer will patch the EXEC function ColdReboot() in an attempt to
"get out of the way" when someone tries to reboot the system.
Enforcer will clean up as much as possible the MMU tables and then
call the original LVO. When Enforcer is asked to quit, it will
check to make sure it can remove itself from this LVO. If it can
not, it will not quit at that time. If run from the shell, it will
display a message saying that it tried but could not exit. Enforcer
will continue to be active and you can try later to deactivate it.
Enforcer will also patch the EXEC function Alert() in an attempt to
provide better tracking of other events in the system. It is also
patched such that dead-end alerts will correctly reset the system
and be displayed. With this patch in place, the normal alerts will
not be seen but will be replaced by the Enforcer output shown
above. See LawBreaker for a more complete example of this.
68020 NOTES
The 68020 does not have a built-in MMU but has a co-processor
feature that lets an external MMU be connected. Enforcer MMU code
is designed for use with 68851 MMU. This is the some-what 68030
compatible MMU by Motorola. Enforcer uses the same code for both
the 68030 and the 68020/68851. For this reason, 68020/68851 users
should see the 68030 NOTES section.
68030 NOTES
The 68030 uses cycle/instruction continuation and will
supply the data on reads and ignore writes during an access
fault rather than let the real bus cycle happen. This means
that on a fault caused by MMU tables, no bus cycle to the
fault address will be generated. (For those of you with analyzers)
In some cases, the 68030 will have advanced the Program Counter
past the instruction by the time the access fault happens.
This is usually only on WRITE faults. For this reason, the PC
may either point at the instruction that caused the fault or
just after the instruction that caused the fault. (Which could
mean that it is pointing to the middle of the instruction
that caused the fault.)
Note that there is a processor called 68EC030. This processor
has a disabled or defective MMU. However, it may function well
enough for Enforcer to think it has a fully functional MMU and
thus Enforcer will attempt to run. However, even if it looks like
the MMU is functioning, it is not fully operational and thus may
cause strange system activity and even crashes. Do not assume
that Enforcer is safe to use on 68EC030 systems.
68040 NOTES
Enforcer, on the 68040, *requires* that the 68040.library be
installed and it requires an MMU 68040 CPU. The 68EC040 does not
have a MMU. The 68LC040 does have an MMU and is supported. Enforcer
will work best in a system with the 68040.library 37.10 or better
but it does know how to deal with systems that do not have that
version.
Due to the design of the 68040, Enforcer is required to do a number
of things differently. For example, the MMU page size can only be
either 8K or 4K. This means that to protect the low 1K of memory,
Enforcer will end up having to mark the first 4K of memory as
invalid and emulate the access to the 3K of that memory that is
valid. For this reason Enforcer moves a number of possible
structures from the first 4K of memory to higher addresses. This
means that the system will continue to run at a reasonable speed.
The first time Enforcer is run it may need to allocate memory for
these structures that it will move. Enforcer can never return this
memory to the system.
In addition to the fact that the 68040 MMU table size is different,
the address fault handling is also different. Namely, the 68040 can
only rerun the cycle and not continue it like the 68030. This means
that on a 68040, the page must be made available first and then made
unavailable. Make this work, Enforcer will switch the instruction
that caused the error into trace mode and let it run with a special
MMU setup. When the trace exception comes in, the MMU is set up
back to the way it was. Enforcer does its best to keep debuggers
working. Note, however, that the interrupt level during a trace of
a READ will end up being set to 7. This is to prevent interrupts
from changing the order of trace/MMU table execution. The level
will be restored to the original state before continuing. Since T0
mode tracing is also supported, there are also some changes in the
way it operates. T0 mode tracing is defined, on the 68040, to cause
a trace whenever the instruction pipeline needed to be reloaded.
While on the 68020/030 processors this was normally only for the
branch instructions, in the 68040 this includes a large number of
other instructions. (Including NOP!) Anyway, if an Enforcer hit
happens while in T0 tracing mode, the trace will happen even on
instructions that normally would not cause a T0 mode trace. Since
this may actually help in debugging and because it was not possible
to do anything else, this method of operation is deemed acceptable.
Another issue with the 68040 is that WRITE faults happen *after* the
instruction has executed. (Except for MOVEM) In fact, it is common
for the 68040 to execute one or more extra instructions before the
WRITE fault is executed. This design makes the 68040 much faster,
but it also makes the Program Counter value that Enforcer can report
for the fault much less likely to be pointing to the instruction
that caused it. The worst cases are sequences such as a write fault
followed by a branch instruction. In these cases, the branch is
usually already executed before the write fault happens and thus the
PC will be pointing to the target of the branch. There is nothing
that can be done within Enforcer to help out here. You will just
need to be aware of this and deal with it as best as possible.
Along with the above issue, is the fact that since a write fault may
be delayed, a read fault may happen before the write fault shows up.
Internally, enforcer does not do special processing for these and
they will not show up. Since another hit was happening anyway, it
is felt that it is best to just not report the hit. Along the same
lines, the hit generated from a MOVEM instruction may only show as a
single hit rather than 1 for each register moved.
On the Amiga, MOVE16 is not supported 100%. Causing an Enforcer hit
with a MOVE16 will cause major problems and maybe cause Enforcer or
your task to lock. Since MOVE16 is not supported, this is not a
major issue. Just watch out if you are using this 68040
instruction. (Also, watch out for the 68040 CPU bug with MOVE16)
The functions CachePreDMA(), CachePostDMA(), and CacheControl() are
patched when the 68040 MMU is turned on by Enforcer. These
functions are patched such the issues with DMA and the 68040
COPYBACK data caches are addressed. The 68040.library normally
deals with this, however since Enforcer turns on the MMU, the method
of dealing with it in the 68040.library will not work. For this
reason, Enforcer will patch these and implement the required fix for
when the MMU is on. When Enforcer is asked to exit, it will check
if it can remove itself from these functions. If it can not, it
will ignore the request to exit. If Enforcer was run from the CLI,
it will print a message saying that it can not exit when the attempt
is made.
WRITING DEBUGGERS
If you wish to make a debugger that works with Enforcer to help
pinpoint Enforcer hits in the application and not cause Enforcer
hits itself, here are some simple tips and a bit of code.
DEBUGGERS: TRAPPING A HIT
To trap a hit requires a number of things to work.
First, the debugger itself must never cause an Enforcer hit.
For help on that, see the "DEBUGGERS: NOT CAUSING A HIT"
Second, the debugger must be global. That is, you must be
able to deal with a task getting a hit that is not the task
under test. There are a number of simple ways to deal with
this, and I will leave this up to the debugger writer.
(One method will be shown below)
Third, the debugger must start *AFTER* Enforcer starts.
If it is started before Enforcer, the hits will not be
trapped. (Note that this is not a problem)
A very important point: The code needs to be fast for
the special case of location 4. This is shown in the
code below. It is very important that this be fast.
Note that it is much prefered that debuggers use the
method described below for trapping hits. It should
be much more supportable this way as any of the tricky
work that may need to be done in the hit processing
will be handled by Enforcer itself. If you wish the
hit decoded, you can capture the Enforcer output via a
pipe or some other method (such as RAWIO) or you can
leave that issue up to the user.
Now, given the above, the following bits of code can be
used to get the debugger to switch into single-step mode
at the point of the Enforcer hit. You can also set some
data value here to tell your debugger about this.
;
; The following code is inserted into the bus error vector.
; Make sure you follow the VBR to find the vector.
; Store the old vector in the address OldVector
; Make sure you already have the single-step trap vector
; installed before you install this. Note that any extra
; code you add in the comment area *MUST NOT* cause a bus
; fault of any kind, including reading of location 4.
;
; This is the 68020 and 68030 version...
;
EnforcerHit: ds.l 1 ; Some private flag
MyTask: ds.l 1 ; Task under test
MyExecBase: ds.l 1 ; The local copy
OldVector: ds.l 1 ; One long word
NewVector: cmp.l #4,$10(sp) ; 68020 and 68030
beq.s TraceSkip ; If AbsExecBase, OK
;
; Now, if you wish to only trap a specific task,
; do the check at this point. For example, a
; simple single-task debugger would do something
; like this:
move.l a0,-(sp) ; Save this...
move.l MyExecBase(pc),a0 ; Get ExecBase...
move.l ThisTask(a0),a0 ; Get ThisTask
cmp.l MyTask(pc),a0 ; Are they the same?
move.l (sp)+,a0 ; Restore A0 (no flags)
bne.s TraceSkip ; If not my task, skip
;
bset.b #7,(sp) ; Set trace bit...
; If you have any other data to set, do it now...
; Set as setting the EnforcerHit bit in your data...
addq.l #1,EnforcerHit : Count the hit...
;
TraceSkip: move.l OldVector(pc),-(sp) ; Ready to return
rts
;
; This is the 68040 version...
;
NewVector040: cmp.l #4,$14(sp) ; 68040
beq.s TraceSkip040 ; If AbsExecBase, OK
;
; Now, if you wish to only trap a specific task,
; do the check at this point. For example, a
; simple single-task debugger would do something
; like this:
move.l a0,-(sp) ; Save this...
move.l MyExecBase(pc),a0 ; Get ExecBase...
move.l ThisTask(a0),a0 ; Get ThisTask
cmp.l MyTask(pc),a0 ; Are they the same?
move.l (sp)+,a0 ; Restore A0 (no flags)
bne.s TraceSkip ; If not my task, skip
;
bset.b #7,(sp) ; Set trace bit...
; If you have any other data to set, do it now...
; Set as setting the EnforcerHit bit in your data...
addq.l #1,EnforcerHit : Count the hit...
;
TraceSkip040: move.l OldVector(pc),-(sp) ; Ready to return
rts
DEBUGGERS: NOT CAUSING A HIT
In order not to cause Enforcer hits, you can do a number
of things. The easiest is to test the address with the TypeOfMem()
EXEC function. If TypeOfMem() returns 0, the address is not
in the memory lists. However, this does not mean it is not a
valid address in all cases. (ROM, chip registers, I/O boards)
For those cases, you can build a "valid memory access table"
much like Enforcer does. Here is the code from Enforcer for
the base memory tables:
/*
* Mark_Address(mmu,start address,length,type)
*/
/*
* Special case the first page of CHIP RAM
*/
mmu=Mark_Address(mmu,0,0x1000,INVALID | NONCACHEABLE);
/*
* Map in the free memory
*/
Forbid();
mem=(struct MemHeader *)SysBase->MemList.lh_Head;
while (mem->mh_Node.ln_Succ)
{
mmu=Mark_Address(mmu,
(ULONG)(mem->mh_Lower),
(ULONG)(mem->mh_Upper)-(ULONG)(mem->mh_Lower),
((MEMF_CHIP & TypeOfMem(mem->mh_Lower)) ?
(NONCACHEABLE | VALID) : (CACHEABLE | VALID)));
mem=(struct MemHeader *)(mem->mh_Node.ln_Succ);
}
Permit();
/*
* Map in the autoconfig boards
*/
if (ExpansionBase=OpenLibrary("expansion.library",0))
{
struct ConfigDev *cd=NULL;
while (cd=FindConfigDev(cd,-1L,-1L))
{
/* Skip memory boards... */
if (!(cd->cd_Rom.er_Type & ERTF_MEMLIST))
{
mmu=Mark_Address(mmu,
(ULONG)(cd->cd_BoardAddr),
cd->cd_BoardSize,
VALID | NONCACHEABLE);
}
}
CloseLibrary(ExpansionBase);
}
/*
* Now for the control areas...
*/
mmu=Mark_Address(mmu,0x00BC0000,0x00040000,VALID | NONCACHEABLE);
mmu=Mark_Address(mmu,0x00D80000,0x00080000,VALID | NONCACHEABLE);
/*
* and the ROM...
*/
mmu=Mark_Address(mmu,
0x00F80000,
0x00080000,
VALID | CACHEABLE | WRITEPROTECT);
/*
* If the credit card resource, make the addresses valid...
*/
if (OpenResource("card.resource"))
{
mmu=Mark_Address(mmu,0x00600000,0x00440002,VALID | NONCACHEABLE);
}
/*
* If CD-based Amiga (CDTV, A570, etc.)
*/
if (FindResident("cdstrap"))
{
mmu=Mark_Address(mmu,0x00E00000,0x00080000,VALID | NONCACHEABLE);
mmu=Mark_Address(mmu,0x00B80000,0x00040000,VALID | NONCACHEABLE);
}
/*
* Check for ReKick/ZKick/KickIt
*/
if ((((ULONG)(SysBase->LibNode.lib_Node.ln_Name)) >> 16) == 0x20)
{
mmu=Mark_Address(mmu,
0x00200000,
0x00080000,
VALID | CACHEABLE | WRITEPROTECT);
}
SEE ALSO
"A master's secrets are only as good as the
master's ability to explain them to others." - Michael Sinz
BUGS
None?
FindHit
NAME
FindHit - A tool that can locate the source file and line number
that a SegTracker report happened at.
SYNOPSIS
FindHit will read the executable file and if there is debugging
information in it, will try to locate the source file and line
number that correspond to the Enforcer hit HUNK/OFFSET.
FUNCTION
FindHit uses the Lattice/SAS/MetaScope standard 'LINE'
debug hunk to locate the closest line to the hunk/offset given.
Note that this can only happen if the executable has the
LINE debugging turned on. (The LawBreaker program has this
such that you can test this yourself.)
In SAS/C 6.x, you need to compile with DEBUG=LINE or better
and do not use the link option of NODEBUG.
In SAS/C 5.x, you need to compile with -d1 or better.
Note that FindHit works with the old SAS/C 5.x 'SRC '
debugging information too. This is required for -d2 or
higher debugging support. However, I do not have 'SRC ' hunk
documentation and thus FindHit may be very specific to the
SAS/C 5.x version of this hunk.
In DICE (2.07 registered being the one I tried) the -d1
debug switch also supports the 'LINE' debug hunk and
works with FindHit.
In HX68 and CAPE, you need to add the DEBUG directive to
the assembly code program. (See LawBreaker source)
For other languages, or other versions of the above, please
see the documentation that comes with the language.
INPUTS
FILE/A - The executable file, with debugging information.
OFFSETS/A/M - The HEX offset (with or without leading $)
If a hunk number other than the default
is needed, it is expressed as hunk:offset.
The default hunk is that of the last argument
or hunk 0 if no hunk number has been given.
For example: 12 $22 $3:12 22 4:$12 32 $0:$32
will find information for:
hunk $0, offset $12
hunk $0, offset $22
hunk $3, offset $12
hunk $3, offset $22
hunk $4, offset $12
hunk $4, offset $32
hunk $0, offset $32
EXAMPLE
FindHit FooBar $0342 $1:4F2 3:$1A 2C
badcode.c : Line 184
No line number information for Hunk $1, Offset $4F2
badcode2.c : Line 12
See the Enforcer documentation about issues dealing with the
exact location of the Enforcer hit. The line given may
not be exactly where the hit happened.
The way I use this is to always have line debugging turned on
when I compile. This does not change the quality of the code
and takes only a small amount of extra disk space. However,
what I do is to link the program twice: Once to a file called
program.ld which contains all of the debugging information.
Then, I link program.ld to program, stripping debug information.
The command line for SLINK or BLINK is as follows:
BLINK program.ld TO program NODEBUG
I keep both of these on hand; with program being the one I
distribute and use. When a hit happens, I can just use program.ld
with FindHit to get the line number and source file that it happened
in. This way you can distribute your software without the debugging
information and still be able to use FindHit on the actual code.
(After all, that link command does nothing but strip symbol and
debug hunks)
NOTES
Note that this program does nothing when run from the Workbench
and thus does not have an icon.
SEE ALSO
"Quantum Physics: The Dreams that Stuff is made of." - Michael Sinz
BUGS
LawBreaker
NAME
LawBreaker - A quicky test of Enforcer
SYNOPSIS
This is a quick test of Enforcer and its reporting abilities.
FUNCTION
This program is used to make sure that Enforcer is correctly
installed and operating. LawBreaker works from either the CLI
or Workbench. It will try to read and write certain memory
areas that will cause an Enforcer hit or four.
LawBreaker will also do an Alert to show how Enforcer reports
an Alert.
Note that the LawBreaker executable has debugging information
in it (standard LINE format debug hunk) such that you can
try the FindHit program to find the line that causes the hit.
INPUTS
Just run it...
RESULTS
When running Enforcer, you will see some output from Enforcer.
Output on a 68030 machine would look something like this:
25-Jul-93 17:15:04
WORD-WRITE to 00000000 data=0000 PC: 0763857C
USP: 07657C14 SR: 0004 SW: 04C1 (U0)(-)(-) TCB: 07642F70
Data: DDDD0000 DDDD1111 DDDD2222 DDDD3333 0763852A DDDD5555 DDDD6666 DDDD7777
Addr: AAAA0000 AAAA1111 AAAA2222 AAAA3333 AAAA4444 0763852A 07400810 --------
Stck: 00000000 0752EE9A 00002800 07643994 00000000 076786D8 000208B0 2EAC80EE
Stck: 487AFD12 486C82C4 4EBA3D50 4EBAEA28 4FEF0014 52ACE2E4 204D43EC 88BC203C
----> 0763857C - "lawbreaker" Hunk 0000 Offset 00000074
Name: "Shell" CLI: "LawBreaker" Hunk 0000 Offset 00000074
25-Jul-93 17:15:04
LONG-READ from AAAA4444 PC: 07638580
USP: 07657C14 SR: 0015 SW: 0501 (U0)(F)(-) TCB: 07642F70
Data: DDDD0000 DDDD1111 DDDD2222 DDDD3333 0763852A DDDD5555 DDDD6666 DDDD7777
Addr: AAAA0000 AAAA1111 AAAA2222 AAAA3333 AAAA4444 0763852A 07400810 --------
Stck: 00000000 0752EE9A 00002800 07643994 00000000 076786D8 000208B0 2EAC80EE
Stck: 487AFD12 486C82C4 4EBA3D50 4EBAEA28 4FEF0014 52ACE2E4 204D43EC 88BC203C
----> 07638580 - "lawbreaker" Hunk 0000 Offset 00000078
Name: "Shell" CLI: "LawBreaker" Hunk 0000 Offset 00000078
25-Jul-93 17:15:04
BYTE-WRITE to 00000101 data=11 PC: 0763858A
USP: 07657C14 SR: 0010 SW: 04A1 (U0)(F)(D) TCB: 07642F70
Data: 00000000 DDDD1111 DDDD2222 DDDD3333 0763852A DDDD5555 DDDD6666 DDDD7777
Addr: AAAA0000 AAAA1111 AAAA2222 AAAA3333 AAAA4444 0763852A 07400810 --------
Stck: 00000000 0752EE9A 00002800 07643994 00000000 076786D8 000208B0 2EAC80EE
Stck: 487AFD12 486C82C4 4EBA3D50 4EBAEA28 4FEF0014 52ACE2E4 204D43EC 88BC203C
----> 0763858A - "lawbreaker" Hunk 0000 Offset 00000082
Name: "Shell" CLI: "LawBreaker" Hunk 0000 Offset 00000082
25-Jul-93 17:15:04
LONG-WRITE to 00000102 data=00000000 PC: 07638592
USP: 07657C14 SR: 0014 SW: 0481 (U0)(-)(D) TCB: 07642F70
Data: 00000000 DDDD1111 DDDD2222 DDDD3333 0763852A DDDD5555 DDDD6666 DDDD7777
Addr: AAAA0000 AAAA1111 AAAA2222 AAAA3333 AAAA4444 0763852A 07400810 --------
Stck: 00000000 0752EE9A 00002800 07643994 00000000 076786D8 000208B0 2EAC80EE
Stck: 487AFD12 486C82C4 4EBA3D50 4EBAEA28 4FEF0014 52ACE2E4 204D43EC 88BC203C
----> 07638592 - "lawbreaker" Hunk 0000 Offset 0000008A
Name: "Shell" CLI: "LawBreaker" Hunk 0000 Offset 0000008A
25-Jul-93 17:15:06
Alert !! Alert 35000000 TCB: 07642F70 USP: 07657C10
Data: 00000000 DDDD1111 DDDD2222 DDDD3333 0763852A DDDD5555 DDDD6666 35000000
Addr: AAAA0000 AAAA1111 AAAA2222 AAAA3333 AAAA4444 0763852A 07400810 --------
Stck: 076385A0 00000000 0752EE9A 00002800 07643994 00000000 0762F710 076305F0
----> 076385A0 - "lawbreaker" Hunk 0000 Offset 00000098
Now, using FindHit, you would type:
FindHit LawBreaker 0:82
and it will tell you the source file name and the line number
where the hit happened. See the FindHit documentation.
NOTES
If enforcer is not running, the program should not cause the
system to crash. It will, however, write to certain areas
of low memory. Also, it will cause read access of some
addresses that may not exist. This may cause bus faults.
SEE ALSO
"Quantum Physics: The Dreams that Stuff is made of." - Michael Sinz
BUGS
There are 4 known Enforcer hits in this code and 1 alert, however,
they will not be fixed. ;^)
Move4K
NAME
Move4K - Moves as much out of the lower 4K of RAM as possible
SYNOPSIS
On 68040 systems, as much of the lower 4K of CHIP RAM as possible
is removed from system use.
FUNCTION
On 68040 systems the MMU page sizes are 4K and 8K. Enforcer
uses the 4K page size. Since watching for hits of low memory
is a vital part of Enforcer, this means that the first 4K
of RAM will be marked invalid. On current systems, only
the first 1K of RAM is invalid and thus 3K of RAM in that
first 4K will end up needing to be emulated in Enforcer.
In order to reduce the overhead that this causes (and the
major performance loss) this program will try to move as much
from that first 4K as possible and make any of the free
memory within the first 4K inaccessible.
Enforcer itself also has this logic, but it may be useful
to be able to run this program as the first program in
the Startup-Sequence (*AFTER* SetPatch) to try to limit
the number of things that may use the lower 4K of RAM.
INPUTS
Just run it... Can be run from CLI or Workbench
RESULTS
Any available memory in the lower 4K of CHIP RAM is removed
plus a special graphics buffer is moved if it needs to be.
After running this program you may have a bit less CHIP RAM
than before. You can run this program as many times as you
wish since it only moves things if it needs to.
NOTES
This program will do nothing on systems without a 68040.
It does not, however, check for the MMU and thus it will
move the lower 4K even if the CPU is not able to run Enforcer.
V39 of the operating system already does have the lowest
MMU page empty and thus this program will effectively do
nothing under V39.
SEE ALSO
"Eloquence is vehement simplicity"
BUGS
None.
SegTracker
NAME
SegTracker - A global SegList tracking utility
SYNOPSIS
A global tracking utility for disk loaded files including
libraries and devices. If placed in the startup-sequence
right after SetPatch, it will track all disk loaded segments
(other than those loaded by SetPatch)
FUNCTION
SegTracker will patch the DOS LoadSeg(), NewLoadSeg(), and UnLoadSeg()
functions in order to track the SegLists that are loaded.
SegTracker keeps these seglist stored in a "safe" manner and
even handles programs which SegList split.
The first time the program is run, it installs the patches
and semaphore. After that point, it just finds the semaphore
and uses it.
By using SegTracker, it will be possible to better identify
where Enforcer hits come from when dealing with libraries
and devices. Basically, it is a system-global Hunk-o-matic.
External programs can then pass in an address to SegTracker
either via the command line or via the given function pointer
in the SegTracker semaphore and get back results as to what
hunk and offset the address is at.
To work with the function directly, you need to find the
the semaphore of "SegTracker" using FindSemaphore().
The structure found will be the following:
struct SegSem
{
struct SignalSemaphore seg_Semaphore;
SegTrack *seg_Find;
};
The function pointer points to a routine that takes an address
and two pointers to long words for returning the Segment number
and Offset within the segment. The function returns the name
of the file loaded. Note that you must call this function
while in Forbid() and then copy the name as the seglist may
be UnLoadSeg'ed at any moment and the name string will then
no longer be in memory.
typedef char (* __asm SegTrack(register __a0 ULONG Address,
register __a1 ULONG *SegNum,
register __a2 ULONG *Offset));
The above is for use in C code function pointer prototype
in SAS/C 5 and 6.
INPUTS
SHOW/S - Shows all of the segments being tracked.
DUMP/S - Displays all of the segment elements being tracked.
FIND/M - Find the hex (in $xxxxx format) address in
the tracked segments. Multiple addresses
can be given.
Options are not available from Workbench as they require
the CLI. However, you can run SegTracker from Workbench
to install it.
EXAMPLE USAGE
/*
* A simple program that will "find" given addresses in the SegLists
* This program has been compiled with SAS/C 6.2 without errors or
* warnings.
*
* Compiler options:
* DATA=FARONLY PARAMETERS=REGISTER NOSTACKCHECK
* NOMULTIPLEINCLUDES STRINGMERGE STRUCTUREEQUIVALENCE
* MULTIPLECHARACTERCONSTANTS DEBUG=LINE NOVERSION
* OPTIMIZE OPTIMIZERINLOCAL NOICONS
*
* Linker options:
* FindSeg.o TO FindSeg SMALLCODE SMALLDATA NODEBUG LIB LIB:sc.lib
*/
#include
#include
#include
#include
#include
#include
#include
#include
#include
#include
#include
#include
#include "FindSeg_rev.h"
#define EXECBASE (*(struct ExecBase **)4)
typedef char (* __asm SegTrack(register __a0 ULONG,
register __a1 ULONG *,
register __a2 ULONG *));
struct SegSem
{
struct SignalSemaphore seg_Semaphore;
SegTrack *seg_Find;
};
#define SEG_SEM "SegTracker"
#define TEMPLATE "FIND/M" VERSTAG
#define OPT_FIND 0
#define OPT_COUNT 1
ULONG cmd(void)
{
struct ExecBase *SysBase;
struct Library *DOSBase;
struct RDArgs *rdargs;
ULONG rc=RETURN_FAIL;
struct SegSem *segSem;
char **hex;
LONG opts[OPT_COUNT];
SysBase = EXECBASE;
if (DOSBase = OpenLibrary("dos.library",37))
{
memset((char *)opts, 0, sizeof(opts));
if (!(rdargs = ReadArgs(TEMPLATE, opts, NULL)))
{
PrintFault(IoErr(),NULL);
}
else if (CheckSignal(SIGBREAKF_CTRL_C))
{
PrintFault(ERROR_BREAK,NULL);
}
else if (segSem=(struct SegSem *)FindSemaphore(SEG_SEM))
{
rc=RETURN_OK;
if (opts[OPT_FIND])
{
for (hex=(char **)opts[OPT_FIND];(*hex);hex++)
{
char *p;
ULONG val;
ULONG tmp[4];
ULONG c;
val=0;
p=*hex;
if (*p=='$') p++; /* Support $hex */
while (*p)
{
c=(ULONG)*p;
if ((c>='a') && (c<='f')) c-=32;
c-='0';
if (c>9)
{
c-=7;
if (c<10) c=16;
}
if (c<16)
{
val=(val << 4) + c;
p++;
}
else
{
val=0;
p=&p[strlen(p)];
}
}
/*
* Ok, we need to do this within Forbid()
* as segments can unload at ANY time, including
* during AllocMem(), so we use a stack buffer...
*
*/
Forbid();
if (p=(*segSem->seg_Find)(tmp[0]=val,&tmp[2],&tmp[3]))
{
char Buffer[200];
stccpy(Buffer,p,200);
tmp[1]=(ULONG)Buffer;
VPrintf("$%08lx - %s : Hunk %ld, Offset $%08lx",tmp);
/*
* Now get the SegList address by passing the
* same pointer for both hunk & offset. Note
* that this is only in the newer SegTrackers
* To test if this worked, check if the result
* of this call is either a hunk or an offset.
*/
(*segSem->seg_Find)(val,&tmp[0],&tmp[0]);
/*
* This "kludge" is for compatibility reasons
* Check if result is the same as either the hunk
* or the offset. If so, do not print it...
*/
if ((tmp[0]!=tmp[2]) && (tmp[0]!=tmp[3]))
{
VPrintf(", SegList $%08lx",tmp);
}
PutStr("\n");
}
else VPrintf("$%08lx - Not found\n",tmp);
Permit();
}
}
}
else PutStr("Could not find SegTracker semaphore.\n");
if (rdargs) FreeArgs(rdargs);
CloseLibrary(DOSBase);
}
else if (DOSBase=OpenLibrary("dos.library",0))
{
Write(Output(),"Requires Kickstart 2.04 (37.175) or later.\n",43);
CloseLibrary(DOSBase);
}
return(rc);
}
NOTES
The earlier this command is run, the better off it will be in
tracking disk loaded segments. Under debug usage, you may
wish to run the command right *AFTER* SetPatch.
Some things may not call UnLoadSeg() to free their seglists.
There is no way SegTracker can follow a seglist that is not
unloaded via the dos.library call to UnLoadSeg(). For this
reason, SegTracker adds new LoadSeg() segments to the top
of its list. This way, if any old segments are still on
the list but have been unloaded via some other method
they will not clash with newer segments during the find operation.
Note that the resident list is one such place where
UnLoadSeg() is not called to free the seglist. Thus,
if something is made resident and then later unloaded
it will still be listed as tracked by SegTracker.
In order to support a new feature in CPR, the SegTracker function
got a "kludge" added to it. If a segment is found, you can then
call the function again with the same address but with having
both pointers point to the same longword of storage. By doing
this, the function will now return (in that longword) the
SegList pointer (CPTR not BPTR) of the file that contains
the address. The reason this method was used was such that
it was compatible with older SegTracker versions. In older
versions you would not get the result you wanted but you would
also not crash. See the example above for more details on how
to use this feature. The SegTracker FIND option has been
expanded to include this information.
Due to the fact that I am working on a design of a new set of
debugging tools (Enforcer/SegTracker/etc) I do not wish to
expand the current SegTracker model in too many ways.
SEE ALSO
"Quantum Physics: The Dreams that Stuff is made of." - Michael Sinz
BUGS
Converted on 22 Apr 2000 with RexxDoesAmigaGuide2HTML 2.1 by Michael Ranner.