binutils-gdb/sim/ppc
Michael Meissner a9ddbbf5a1 12/19 release from Andrew that fixes Netbsd break problem 1995-12-19 16:41:06 +00:00
..
.Sanitize Changes from Andrew 1995-12-15 20:20:13 +00:00
COPYING
COPYING.LIB
ChangeLog Work with new and old BFDs; Do not inline model externs functions by default 1995-12-16 15:02:48 +00:00
Makefile.in Fix warnings and dependency 1995-11-14 09:52:29 +00:00
README.psim 12/19 release from Andrew that fixes Netbsd break problem 1995-12-19 16:41:06 +00:00
basics.h Do not include sysdep.h, bfd no longer provides it 1995-09-08 13:59:49 +00:00
bits.c New changes from Andrew 1995-10-31 18:29:37 +00:00
bits.h
cap.c Changes from Andrew 1995-12-15 20:20:13 +00:00
cap.h Changes from Andrew 1995-12-15 20:20:13 +00:00
config.in Build again on SunOS; Make MODEL_INLINE default to 1 if inlining, not 2; Use __attribute__((const)) 1995-11-27 22:54:40 +00:00
configure Let autoconf determine if host is big endian or not 1995-11-30 11:13:08 +00:00
configure.in 12/19 release from Andrew that fixes Netbsd break problem 1995-12-19 16:41:06 +00:00
corefile-n.h Changes from Andrew 1995-12-15 20:20:13 +00:00
corefile.c first stage in function unit support; add new switches & latest code from andrew 1995-11-08 18:57:06 +00:00
corefile.h Changes from Andrew 1995-12-15 20:20:13 +00:00
cpu.c Add determining when we do not have enough writeback slots; Do not do model specific handling if not printing out the information 1995-11-28 18:47:07 +00:00
cpu.h Build again on SunOS; Make MODEL_INLINE default to 1 if inlining, not 2; Use __attribute__((const)) 1995-11-27 22:54:40 +00:00
debug.c more functional unit changes 1995-11-16 19:02:52 +00:00
debug.h Fix warnings to everything can be compiled with -Wall; Redo model specific changes once again to speed things up 1995-11-24 16:44:37 +00:00
device.c Changes from Andrew 1995-12-15 20:20:13 +00:00
device.h Fix warnings when using full inlining. 1995-12-15 21:06:22 +00:00
device_table.c Changes from Andrew 1995-12-15 20:20:13 +00:00
dgen.c Use autoconf correctly; provide more stats with -I 1995-11-02 20:21:35 +00:00
double.c
dp-bit.c
emul_bugapi.c Changes from Andrew 1995-12-15 20:20:13 +00:00
emul_bugapi.h Changes from Andrew 1995-12-15 20:20:13 +00:00
emul_chirp.c Changes from Andrew 1995-12-15 20:20:13 +00:00
emul_chirp.h Changes from Andrew 1995-12-15 20:20:13 +00:00
emul_generic.c first stage in function unit support; add new switches & latest code from andrew 1995-11-08 18:57:06 +00:00
emul_netbsd.c Changes from Andrew 1995-12-15 20:20:13 +00:00
emul_netbsd.h New changes from Andrew 1995-10-31 18:29:37 +00:00
events.c
events.h
filter_filename.c first stage in function unit support; add new switches & latest code from andrew 1995-11-08 18:57:06 +00:00
filter_filename.h Lots of changes 1995-11-01 19:32:38 +00:00
idecode_branch.h
idecode_fields.h
igen.c Sort instruction names; Add igen -R option; count # of CRs that mtcrf moved 1995-11-25 01:35:14 +00:00
inline.c Add option to make options inline 1995-11-15 23:01:47 +00:00
inline.h Add option to make options inline 1995-11-15 23:01:47 +00:00
interrupts.c Changes from Andrew 1995-12-15 20:20:13 +00:00
interrupts.h Changes from Andrew 1995-12-15 20:20:13 +00:00
lf.c Use autoconf correctly; provide more stats with -I 1995-11-02 20:21:35 +00:00
lf.h Check lf_printf for type correctness 1995-11-25 06:08:56 +00:00
main.c Add determining when we do not have enough writeback slots; Do not do model specific handling if not printing out the information 1995-11-28 18:47:07 +00:00
misc.c Andrew's latest changes & print all instruction counts if -I 1995-11-02 14:27:18 +00:00
misc.h Lots of changes 1995-11-01 19:32:38 +00:00
mon.c Add determining when we do not have enough writeback slots; Do not do model specific handling if not printing out the information 1995-11-28 18:47:07 +00:00
mon.h More model specific changes 1995-11-15 22:53:59 +00:00
options.c Add option to make options inline 1995-11-15 23:01:47 +00:00
options.h Add option to make options inline 1995-11-15 23:01:47 +00:00
os_emul.c New changes from Andrew 1995-10-31 18:29:37 +00:00
os_emul.h New changes from Andrew 1995-10-31 18:29:37 +00:00
ppc-cache-rules Fix warnings to everything can be compiled with -Wall; Redo model specific changes once again to speed things up 1995-11-24 16:44:37 +00:00
ppc-instructions Add determining when we do not have enough writeback slots; Do not do model specific handling if not printing out the information 1995-11-28 18:47:07 +00:00
ppc-opcode-complex New changes from Andrew 1995-10-31 18:29:37 +00:00
ppc-opcode-simple New changes from Andrew 1995-10-31 18:29:37 +00:00
ppc-opcode-stupid Andrew's latest changes & print all instruction counts if -I 1995-11-02 14:27:18 +00:00
ppc-opcode-test-1 Changes from Andrew 1995-12-15 20:20:13 +00:00
ppc-opcode-test-2 Changes from Andrew 1995-12-15 20:20:13 +00:00
ppc.mt
psim.c Add determining when we do not have enough writeback slots; Do not do model specific handling if not printing out the information 1995-11-28 18:47:07 +00:00
psim.h Add support to count the number of instructions issued. 1995-10-02 18:19:17 +00:00
registers.c Use autoconf correctly; provide more stats with -I 1995-11-02 20:21:35 +00:00
sim-endian-n.h first stage in function unit support; add new switches & latest code from andrew 1995-11-08 18:57:06 +00:00
sim-endian.c first stage in function unit support; add new switches & latest code from andrew 1995-11-08 18:57:06 +00:00
sim-endian.h Let autoconf determine if host is big endian or not 1995-11-30 11:13:08 +00:00
sim_callbacks.h
sim_calls.c Add determining when we do not have enough writeback slots; Do not do model specific handling if not printing out the information 1995-11-28 18:47:07 +00:00
std-config.h Changes from Andrew 1995-12-15 20:20:13 +00:00
table.c more functional unit changes 1995-11-16 19:02:52 +00:00
table.h New changes from Andrew 1995-10-31 18:29:37 +00:00
vm.c Changes from Andrew 1995-12-15 20:20:13 +00:00
vm.h
vm_n.h first stage in function unit support; add new switches & latest code from andrew 1995-11-08 18:57:06 +00:00
words.h

README.psim

		PSIM - model of a PowerPC platform

    Copyright (C) 1994-1995, Andrew Cagney <cagney@highland.com.au>.

    This program is free software; you can redistribute it and/or modify
    it under the terms of the GNU General Public License as published by
    the Free Software Foundation; either version 2 of the License, or
    (at your option) any later version.

    This program is distributed in the hope that it will be useful,
    but WITHOUT ANY WARRANTY; without even the implied warranty of
    MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
    GNU General Public License for more details.
 
    You should have received a copy of the GNU General Public License
    along with this program; if not, write to the Free Software
    Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA.


This directory contains the program PSIM that implements a model of a
PowerPC platform.  PSIM can either be built standalone or as part of
the debugger GDB.


What is PSIM?

	PSIM is an ANSI C program that models a PowerPC platform.
	The platform it implements can vary from:

		o	A user program environment (UEA) complete
			with emulated system calls

	to

		o	A hardware platform with several processors
			interacting with each other and modeled hardware.


What motivated PSIM?

	As an idea, psim was first discussed seriously during mid
	1994.  At that time its main objectives were:


		o	good performance

			Many simulators loose out by only providing
			a binary interface to the internals.  This
			inteface eventually becomming a bottle neck
			in the simulators performance.

			It was intended that PSIM would avoid this
			problem by giving the user access to the
			full source code.

			Further, by exploiting the power of modern
			compilers it was hoped that PSIM would achieve
			good performance with out having to compromize
			its internal design.


		o	practical portability

			Rather than try to be portable to every
			C compiler on every platform, it was decided
			that PSIM would restrict its self to suporting
			ANSI compilers that included the extension
			of a long long type.

			GCC is one such compiler, consequenly PSIM
			should be portable to any machine running GCC.


		o	flexability in its design

			PSIM should allow the user to select the
			features required and customize the build
			accordingly.  By having the source code,
			the compler is able to eliminate any un
			used features of the simulator.

			After all, let the compiler do the work.


		o	SMP

			A model that allowed the simulation of
			SMP platforms with out the large overhead
			often encountered with such models.


	PSIM achieves each of these objectives.


Who would be interested in PSIM?

	o	the curious

		Using psim, gdb, gcc and binutils the curious
		user can contruct an environment that allows
		them to play with PowerPC user programs with out
		the need for real hardware.


	o	the analyst

		PSIM includes many (contributed) monitoring
		features which (unlike many other simulators)
		do not come with a great penalty in performance.

		Thus the performance analyst is able to use
		this simulator to model the inpact of changes
		to the system they are analysing.  Be that
		system a compiler or real hardware platform.

		If PSIM doesn't monitor a components of interest,
		the source code is freely available, and hence
		there is no hinderance to changing things
		to meet a specific analysts needs.


	o	the serious SW developer

		PSIM models all three levels of the PowerPC
		Architecture: UEA, VEA and OEA.  Further,
		the internal design is such that PSIM can
		be extended to suport additional
		development requirements.  Such requirements
		might include (for the UEA) a new Operating
		System emulation through to (for the OEA)
		a model of a different hardware platform.


What features does PSIM have?


	Monitoring and modeling

		PSIM includes (thanks to Michael Meissner)
		a detailed model of the various PowerPC
		implementations schedulers.


	SMP
		
		The PowerPC ISA defines SMP synchronizing instructions
		this simulator models a limited subset of their
		behavor.  Consequently, if you limit code to the
		use the modeled behavour, PSIM can be used to
		model SMP PowerPC platforms.

		People intending to use this system should study
		the code implementing the lwarx etc instructions.
		
	ENDIAN SUPORT

		PSIM implements all of Big-endian, little-endian
		and PowerPC little endian (XOR endian).

	ISA models

		PSIM includes a model of UEA, VEA and OEA.  This
		inclues the time base registers (VEA) and HTAB
		and BATS (OEA).

		In addition, a preliminary model of the 64 bit
		PowerPC architecture is included.

	Hardware

		PSIM's internals are based around the concept
		of a Device Tree.  This tree intentionaly
		resembles that of the Device Tree found in
		OpenBoot firmware.  Psim is flexable enough
		to allow the user to fully configure the
		actual hardware model from a device tree
		specification taken from a file.

		PSIM also contains several built in device
		trees.
		
	Emulation

		PSIM is able (UEA) to emulate UNIX calls
		based on NetBSD abi through to (preliminary)
		the ROM rom calls found in common firmware
		(OpenBoot and BUGAPI).

	floating point

		Preliminary suport for floating point is included.
		Real kernels don't need floating point.


Can PSIM model a CHRP a.k.a. PowerPC Platform machine?

	No. but that is now one of its main objectives. (Did you notice
	it was written PowerPC Platform instead of PowerPC platform?).


How do I build PSIM?

	To build PSIM you will need the following:

	gdb-4.15.tar.gz		From your favorite GNU ftp site.
				I've also tested psim-951016 with
				gdb-4.15.1.


	ftp://ftp.ci.com.au/pub/clayton/README.pim

				This file.

	ftp://ftp.ci.com.au/pub/clayton/gdb-4.15+psim-951016.diff.gz

				This contains a few minor patches to
				gdb-4.15 so that will include psim
				when it is built.

	ftp://ftp.ci.com.au/pub/clayton/gdb-4.15+psim-951016.tar.gz

				This contains the psim files propper.

	ftp://ftp.ci.com.au/pub/clayton/psim-test-951016.tar.gz

				(Optional) A scattering of pre-compiled
				programs that run under the simulator.

	gcc			Again available from your favorite
				GNU ftp site.

	patch			Sun's patch behaves a little wierd
				and doesn't appear to like creating
				empty files.


	Since PSIM is still being developed, from time to time, to meet
	a specific analysts needsfurther psim snap shots are occasionally
	made available.  These snapshots may or may not work with GDB-4.15.
	Several of the more recent snapshots are:

	ftp://ftp.ci.com.au/pub/clayton/psim-951218.tar.gz

				Hopefully merges in Michael stuff
				with mine,  adds multiple emulations
				(OpenBoot and NetBSD),  revamps
				inline stuff, rearanges devices so
				that phandls and ihandles can be
				implemented.

	ftp://ftp.ci.com.au/pub/clayton/psim-951203.tar.gz

				A good snapshot

				This includes extensions from Michael
				Meissner that add monitoring of the
				PowerPC's register and bus architectures.


	ftp://ftp.ci.com.au/pub/clayton/psim-test-951218.tar.gz

				Prebuilt test programs for PSIM.
				Includes examples of UEA, VEA and
				OEA code.
				Requires gcc-2.7.2 and binutils-2.6
				to rebuild.


	Procedure:

	0.	A starting point

		$ ls -1
		gdb-4.15+psim-951016.diff.gz
		gdb-4.15+psim-951016.tar.gz
		gdb-4.15.tar.gz
		psim-test-951016.tar.gz


	1.	Unpack gdb

		$ gunzip < gdb-4.15.tar.gz | tar xf -


	2.	Change to the gdb directory, apply the psim patches and unpack
		the psim files.

		$ cd gdb-4.15

		$ gunzip < ../gdb-4.15+psim-951016.diff.gz | more
		$ gunzip < ../gdb-4.15+psim-951016.diff.gz | patch -p1

		$ gunzip < ../gdb-4.15+psim-951016.tar.gz | tar tvf -
		$ gunzip < ../gdb-4.15+psim-951016.tar.gz | tar xvf -


	3.	Configure gdb

		$ more gdb/README

		then something like (I assume SH):

		$ CC=gcc ./configure --target=powerpc-unknown-eabisim

		eabisim is needed as by default (because PSIM needs GCC) the
		simulator is not built.


	4.	Build

		$ make CC=gcc

		alternativly, if you are short on disk space or just want the
		simulator built:

		$ ( cd libiberty && make CC=gcc )
		$ ( cd bfd && make CC=gcc )
		$ ( cd sim/ppc && make CC=gcc )


	5.	Install

		$ make CC=gcc install

		or just

		$ cp gdb/gdb ~/bin/powerpc-unknown-eabisim-gdb
		$ cp sim/ppc/run ~/bin/powerpc-unknown-eabisim-run


How do I use the simulator?


	(I assume that you've unpacked the psim-test archive).

	1.	As a standalone program

		Print out the users environment:

		$ powerpc-unknown-eabisim-run envp

		Print out the arguments:

		$ powerpc-unknown-eabisim-run argv a b c

		Check that sbrk works:

		$ powerpc-unknown-eabisim-run break


	2.	Example of running GDB:	

		The main thing to note is that before you can run the simulator
		you must enable it.  The example below illustrates this:

		$ powerpc-unknown-eabisim-gdb envp
		(gdb) target sim
		(gdb) load
		(gdb) break main
		(gdb) run
		.
		.
		.


Where do I send bugs or report problems?

	There is a mailing list (subscribe through majordomo@ci.com.au) (that
	is almost never used) at:

	powerpc-psim@ci.com.au

	If I get the ftp archive updated I post a note to that mailing list.
	In addition your welcome to send bugs or problems either to me or to
	that e-mail list.


Are there any known problems?

	See the ChangeLog file looking for lines taged with the word FIXME.

	COREFILE.C: The implementation of corefile.c (defined by
	corefile.h) isn't the best.  It is intended to be functionaly
	correct rather than fast.  One option being considered
	is to add a data cache to reduce the overhead of the most
	common case of data read/writes.

	HTAB (page) code for OEA model untested.  Some of the vm code
	instructions unimplemented.

	Lacks PowerOpen (a.k.a. XCOFF a.k.a. AIX) and NT startups.  The
	PowerOpen worked until I added the ELF one.

	Missing VEA system calls.

	Missing or commented out instructions.

	64bit target untested.

	64bit host broken.  For instance use of scanf "%x", &long long.

	Event code for pending events from within signal handlers not
	finished/tested.

	Better and more devices.

	PORTABILITY (Notes taken from Michael Meissner): Heavy use of the ##
	operator - fix using the clasic X/**/Y hack; Use of the signed
	keyword.  In particular, signed char has no analogue in classic C
	(though most implementations of classic C use signed chars); Use of
	long long which restricts the target compiler to be GCC.


Who helped?

	Thanks go to the following who each helped in some way.

	Allen Briggs, Bett Koch, David Edelsohn, Gordon Irlam,
	Michael Meissner, Bob Mercier, Richard Perini,
	Richard Stallman, Mitchele Walker