summaryrefslogtreecommitdiff
path: root/Documentation/RFC
diff options
context:
space:
mode:
authorStefan Reinauer <stefan.reinauer@coreboot.org>2015-05-05 13:10:04 -0700
committerStefan Reinauer <stefan.reinauer@coreboot.org>2015-05-06 19:09:47 +0200
commit4d8b843d3740cc90c98f181d49f0d4a67cb0a1b7 (patch)
tree19306425b170975b5317a63a84b0e6f4b94b5f40 /Documentation/RFC
parent29ed46caccd5cea8401c5d133895fa3a9d6f5030 (diff)
downloadcoreboot-4d8b843d3740cc90c98f181d49f0d4a67cb0a1b7.tar.xz
Rename documentation -> Documentation
In order to be closer to the Linux kernel source tree structure, rename documentation to Documentation. Change-Id: I8690f666638ef352d201bd3c3dc1923b0d24cb12 Signed-off-by: Stefan Reinauer <stefan.reinauer@coreboot.org> Reviewed-on: http://review.coreboot.org/10110 Tested-by: build bot (Jenkins) Reviewed-by: Patrick Georgi <pgeorgi@google.com>
Diffstat (limited to 'Documentation/RFC')
-rw-r--r--Documentation/RFC/chip.tex266
-rw-r--r--Documentation/RFC/config.tex291
2 files changed, 557 insertions, 0 deletions
diff --git a/Documentation/RFC/chip.tex b/Documentation/RFC/chip.tex
new file mode 100644
index 0000000000..5e366b8461
--- /dev/null
+++ b/Documentation/RFC/chip.tex
@@ -0,0 +1,266 @@
+ RFC for the chip specification architecture
+
+\begin{abstract}
+At the end of this document is the original message that motivated the
+change.
+\end{abstract}
+
+\section{Scope}
+This document defines how LinuxBIOS programmers can specify chips that
+are used, specified, and initalized. The current scope is for superio
+chips, but the architecture should allow for specification of other chips such
+as southbridges. Multiple chips of same or different type are supported.
+
+\section{Goals}
+The goals of the new chip architecture are these:
+\begin{itemize}
+\item seperate implementation details from specification in the Config file
+(translation: no more C code in Config files)
+\item make the specification easier for people to use and understand
+\item remove private details of a given chip to the chip file as much
+as possible
+\item allow unique register-set-specifiers for each chip
+\end{itemize}
+
+\section{Specification in the Config file}
+The specification looks like this:
+\begin{verbatim}
+chip <name> [path=<path>] ["<configuration>"]
+\end{verbatim}
+The name is in the standard LinuxBIOS form of type/vendor/name, e.g.
+"southbridge/intel/piix4e" or "superio/ite/it8671f". The class of the
+chip is derived from the first pathname component of the name, and the chip
+configuration is derived from the following components.
+
+The path defines the access mechanism to the chip.
+It is optional. If present, it overrides the default path to the chip.
+
+The configuration defines chip-specific configuration details, and is also
+optional. Note that an empty configuration will leave the chip with
+no enabled resources. This may be desirable in some cases.
+
+\section{Results of specifying a chip}
+
+When one or more chips are specified, the data about the chips
+is saved until the entire file is parsed. At this point, the config tool
+creates a file in the build directory called chip.c This file contains
+a common struct containing information about
+each individual chip and an array of pointers to these structures.
+
+For each chip, there are two structures. The structures contain control
+information for the chip, and register initialization information. The
+names of the structures are derived by ``flattening'' the chip name,
+as in the current linuxbios. For example, superio/ite/xyz uses
+two structs, one called superio_ite_xyz_control and one called
+superio_ite_xyz_init. The control struct is initialized from the
+chip name and path information, and has a pointer to the
+config struct. The config struct is initialized from the quote string
+
+\begin{verbatim}
+From rminnich@lanl.gov Fri May 16 10:34:13 2003
+Date: Tue, 13 May 2003 08:11:46 -0600 (MDT)
+From: ron minnich <rminnich@lanl.gov>
+To: linuxbios@clustermatic.org
+Subject: RFC:new superio proposal
+
+Abstract:
+ The superio architecture for linuxbios has worked for the last 2
+years but is being stretched to the limit by the changes in superio chips.
+The architecture depended on superio resources being relatively constant
+between chips, but this assumption no longer holds. In this document we
+propose several alternatives and solicit comments.
+
+Overview:
+The superio architecture in linuxbios was developed over time, and
+modified as circumstances required. In the beginning it was relatively
+simple and assumed only one superio per mainboard. The latest version
+allows an arbitrary number of superios per mainboard, and allows complete
+specification of the superio base I/O address along with the specification
+of reasonable default valures for both the base I/O address and the
+superio parameters such as serial enable, baud rate, and so on.
+
+Specification of superio control parameters is done by a configuration
+line such as:
+
+nsuperio sis/950 com1={1} floppy=1 lpt=1
+
+This fragment sets the superio type to sis/950; sets com1, floppy, and lpt
+to enabled; and leaves the defaults to com1 (baud rate, etc.) to the
+default values.
+
+While it is not obvious, these configuration parameters are fragments of a
+C initializer. The initializers are used to build a statically initialized
+structure of this type:
+
+struct superio {
+ struct superio_control *super; // the ops for the device.
+ unsigned int port; // if non-zero, overrides the default port
+ // com ports. This is not done as an array (yet).
+ // We think it's easier to set up from python if it is not an
+ // array.
+ struct com_ports com1, com2, com3, com4;
+ // DMA, if it exists.
+ struct lpt_ports lpt1, lpt2;
+ /* flags for each device type. Unsigned int. */
+ // low order bit ALWAYS means enable. Next bit means to enable
+ // LPT is in transition, so we leave this here for the moment.
+ // The winbond chips really stretched the way this works.
+ // so many functions!
+ unsigned int ide, floppy, lpt;
+ unsigned int keyboard, cir, game;
+ unsigned int gpio1, gpio2, gpio3;
+ unsigned int acpi,hwmonitor;
+};
+
+These structures are, in turn, created and statically initialized by a
+config-tool-generated structure that defines all the superios. This file
+is called nsuperio.c, is created for each mainboard you build, only
+appears in the build directory, and looks like this:
+
+===
+extern struct superio_control superio_winbond_w83627hf_control;
+
+struct superio superio_winbond_w83627hf= {
+ &superio_winbond_w83627hf_control,
+ .com1={1}, .com2={1}, .floppy=1, .lpt=1, .keyboard=1, .hwmonitor=1};
+
+struct superio *all_superio[] = {&superio_winbond_w83627hf,
+};
+
+unsigned long nsuperio = 1;
+===
+
+This example shows a board with one superio (nsuperio). The superio
+consists of a winbond w83627hf, with com1, com2, floppy, lpt, keyboard,
+and hwmonitor enabled. Note that this structure also allows for
+over-riding the default superio base, although that capability is rarely
+used.
+
+The control structure is used to define how to access the superio for
+purposes of control. It looks like this:
+===
+struct superio_control {
+ void (*pre_pci_init)(struct superio *s);
+ void (*init)(struct superio *s);
+ void (*finishup)(struct superio *s);
+ unsigned int defaultport; /* the defaultport. Can be overridden
+ * by commands in config
+ */
+ // This is the print name for debugging
+ char *name;
+};
+===
+
+There are three methods for stages of hardwaremain. First is pre_pci_init
+(for chips like the acer southbridge that require you to enable some
+resources BEFORE pci scan); init, called during the 'middle' phase of
+hardwaremain; and finishup, called before the payload is loaded.
+
+This approach was inspired by and borrows heavily on the Plan 9 kernel
+configuration tools.
+
+The problem:
+
+When the first version of the superio structure came out it was much
+smaller. It has grown and in the limit this structure is the union of all
+possibly superio chips. Obviously, in the long term, this is not
+practical: we can not anticipate all possible superio chips for all time.
+
+The common PC BIOS solution to this type of problem is to continue with
+binary structures but add version numbers to them, so that all code that
+uses a given structure has to check the version number. Personally, I find
+this grotesque and would rather not work this way.
+
+Using textual strings for configuration is something I find far more
+attractive. Plan 9 has shown that this approach has no real limits and
+suffices for configuration tasks. The Linux kernel does more limited use
+of strings for configuration, but still depends on them. Strings are
+easier to read and work with than binary structures, and more important, a
+lot easier to deal with when things start going wrong.
+
+The proposed solution:
+
+What follows are three possible ideas for specifying superio resources and
+their settings.
+
+A common part of the new idea is to eliminate the common superio
+structure, due to the many variations in chips, and make it invisible
+outside a given superio source file -- the superio structure is now
+private to a given superio. Thus, sis/950/superio.c would contain its own
+superio structure definitions, and also might contain more than once
+instance of these structures (consider a board with 2 sis 950 chips).
+
+The control structure would change as follows:
+struct superio_control {
+ int (*create)(struct superio *s);
+ void (*pre_pci_init)(struct superio *s);
+ void (*init)(struct superio *s);
+ void (*finishup)(struct superio *s);
+ unsigned int defaultport; /* the defaultport. Can be overridden
+ * by commands in config
+ */
+ // This is the print name for debugging
+ char *name;
+};
+
+I.e. we add a new function for creating the superio.
+
+Communication of superio settings from linuxbios to the superio would be
+via textual strings. The superio structure becomes this:
+
+struct superio {
+ struct superio_control *super; // the ops for the device.
+ unsigned int port; // if non-zero, overrides the default port
+ struct configuration *config;
+};
+
+
+So now the question becomes, what is the configuration structure?
+There are several choices. The simplest, from my point of view, are
+keyword-value pairs:
+struct configuration {
+ const char *keyword;
+ const char *value;
+};
+
+These get filled in by the config tool as before. The linuxbios libary can
+then provide a generic parsing function for the superios to use.
+
+The remaining question is how should the superio command look in
+freebios2?
+
+superio sis/950 "com1=115200,8n1 lpt=1 com2=9600"
+
+or
+
+superio sis/950 "com1baud=115200 lpt=1 com1chars=8n1"
+
+or
+
+superio sis/950 ((com1 115200 8n1) (lpt 1))
+
+So, my questions:
+
+1. Does this new scheme look workable. If not, what needs to change?
+2. What should the 'struct configuration' be? does keyword/value work?
+3. what should the superio command look like?
+
+Comments welcome.
+
+I'd like to adopt this "RFC" approach for freebios2 as much as we can.
+There was a lot of give-and-take in the early days of linuxbios about
+structure and it proved useful. There's a lot that will start happening in
+freebios2 now, and we need to try to make sure it will work for everyone.
+
+Those of you who are doing mainboards, please look at freebios2 and see
+how it looks for you. There's a lot of good work that has been done (not
+by me so far, thanks Eric and Stefan), and more that needs to be done.
+Consider trying out romcc as an "assembly code killer". See how it fits
+together and if you can work with it or need changes. Bring comments back
+to this list.
+
+thanks
+
+ron
+
+\end{verbatim}
diff --git a/Documentation/RFC/config.tex b/Documentation/RFC/config.tex
new file mode 100644
index 0000000000..6d6c433025
--- /dev/null
+++ b/Documentation/RFC/config.tex
@@ -0,0 +1,291 @@
+ New config language for LinuxBIOS
+
+\begin{abstract}
+We describe the new configuration language for LinuxBIOS.
+\end{abstract}
+
+\section{Scope}
+This document defines the new configuration language for LinuxBIOS.
+
+\section{Goals}
+The goals of the new language are these:
+\begin{itemize}
+\item Simplified Makefiles so people can see what is set
+\item Move from the regular-expression-based language to something
+a bit more comprehensible and flexible
+\item make the specification easier for people to use and understand
+\item allow unique register-set-specifiers for each chip
+\item allow generic register-set-specifiers for each chip
+\item generate static initialization code, as needed, for the
+specifiers.
+\end{itemize}
+
+\section{Language}
+Here is the new language. It is very similar to the old one, differing
+in only a few respects. It borrows heavily from Greg Watson's suggestions.
+
+I am presenting it in a pseudo-BNF in the hopes it will be easier. Things
+in '' are keywords; things in ``'' are strings in the actual text.
+\begin{verbatim}
+#exprs are composed of factor or factor + factor etc.
+expr ::= factor ( ``+'' factor | ``-'' factor | )*
+#factors are term or term * term or term / term or ...
+factor ::= term ( ``*'' term | ``/'' term | ... )*
+#
+unary-op ::= ``!'' ID
+# term is a number, hexnumber, ID, unary-op, or a full-blown expression
+term ::= NUM | XNUM | ID | unary-op | ``(`` expr ``)''
+
+# Option command. Can be an expression or quote-string.
+# Options are used in the config tool itself (in expressions and 'if')
+# and are also passed to the C compiler when building linuxbios.
+# It is an error to have two option commands in a file.
+# It is an error to have an option command after the ID has been used
+# in an expression (i.e. 'set after used' is an error)
+option ::= 'option' ID '=' (``value'' | term)
+
+# Default command. The ID is set to this value if no option command
+# is scanned.
+# Multiple defaults for an ID will produce warning, but not errors.
+# It is OK to scan a default command after use of an ID.
+# Options always over-ride defaults.
+default ::= 'default' ID '=' (``value'' | term)
+
+# the mainboard, southbridge, northbridge commands
+# cause sourcing of Config.lb files as in the old config tool
+# as parts are sourced, a device tree is built. The structure
+# of the tree is determined by the structure of the components
+# as they are specified. To attach a superio to a southbridge, for
+# example, one would do this:
+# southbridge acer/5432
+# superio nsc/123
+# end
+# end
+# the tool generates static initializers for this hierarchy.
+
+# add C code to the current component (motherboard, etc. )
+# to initialise the component-INDEPENDENT structure members
+init ::= 'init' ``CODE''
+
+# add C code to the current component (motherboard, etc. )
+# to initialise the component-DEPENDENT structure members
+register ::= 'register' ``CODE''
+
+
+# mainboard command
+# statements in this block will set variables controlling the mainboard,
+# and will also place components (northbridge etc.) in the device tree
+# under this mainboard
+mainboard ::= 'mainboard' PATH (statements)* 'end'
+
+# standard linuxbios commands
+southbridge ::= 'southbridge' PATH (statemnts)* 'end'
+northbridge ::= 'northbridge' PATH (statemnts)* 'end'
+superio ::= 'superio PATH (statemnts)* 'end'
+cpu ::= 'cpu' PATH (statemnts)* 'end'
+arch ::= 'arch' PATH (statemnts)* 'end'
+
+# files for building linuxbios
+# include a file in crt0.S
+mainboardinit ::= 'mainboardinit' PATH
+
+# object file
+object ::= 'object' PATH
+# driver objects are just built into the image in a different way
+driver ::= 'driver' PATH
+
+# Use the Config.lb file in the PATH
+dir ::= 'dir' PATH
+
+# add a file to the set of ldscript files
+ldscript ::= 'ldscript' PATH
+
+# dependencies or actions for the makerule command
+dep ::= 'dep' ``dependency-string''
+act ::= 'act' ``actions''
+depsacts ::= (dep | act)*
+# set up a makerule
+#
+makerule ::= 'makerule' PATH depsacts
+
+#defines for use in makefiles only
+# note usable in the config tool, not passed to cc
+makedefine ::= 'makedefine' ``RAWTEXT''
+
+# add an action to an existing make rule
+addaction ::= 'addaction' PATH ``ACTION''
+
+# statements
+statement ::=
+ option
+ | default
+ | cpu
+ | arch
+ | northbridge
+ | southbridge
+ | superio
+ | object
+ | driver
+ | mainboardinit
+ | makerule
+ | makedefine
+ | addaction
+ | init
+ | register
+ | iif
+ | dir
+ | ldscript
+
+statements ::= (statement)*
+
+# target directory specification
+target ::= 'target' PATH
+
+# and the whole thing
+board ::= target (option)* mainboard
+
+\end{verbatim}
+
+\subsubsection{Command definitions}
+\subsubsubsection{option}
+\subsubsubsection{default}
+\subsubsubsection{cpu}
+\subsubsubsection{arch}
+\subsubsubsection{northbridge}
+\subsubsubsection{southbridge}
+\subsubsubsection{superio}
+\subsubsubsection{object}
+\subsubsubsection{driver}
+\subsubsubsection{mainboardinit}
+\subsubsubsection{makerule}
+\subsubsubsection{makedefine}
+\subsubsubsection{addaction}
+\subsubsubsection{init}
+\subsubsubsection{register}
+\subsubsubsection{iif}
+\subsubsubsection{dir}
+\subsubsubsection{ldscript}
+
+
+A sample file:
+
+\begin{verbatim}
+target x
+
+# over-ride the default rom size in the mainboard file
+option CONFIG_ROM_SIZE=1024*1024
+mainboard amd/solo
+end
+
+\end{verbatim}
+
+Sample mainboard file
+\begin{verbatim}
+#
+###
+### Set all of the defaults for an x86 architecture
+###
+arch i386 end
+cpu k8 end
+#
+option CONFIG_DEBUG=1
+default CONFIG_USE_FALLBACK_IMAGE=1
+option A=(1+2)
+option B=0xa
+#
+###
+### Build our 16 bit and 32 bit linuxBIOS entry code
+###
+mainboardinit cpu/i386/entry16.inc
+mainboardinit cpu/i386/entry32.inc
+ldscript cpu/i386/entry16.lds
+ldscript cpu/i386/entry32.lds
+#
+###
+### Build our reset vector (This is where linuxBIOS is entered)
+###
+if CONFIG_USE_FALLBACK_IMAGE
+ mainboardinit cpu/i386/reset16.inc
+ ldscript cpu/i386/reset16.lds
+else
+ mainboardinit cpu/i386/reset32.inc
+ ldscript cpu/i386/reset32.lds
+end
+.
+.
+.
+if CONFIG_USE_FALLBACK_IMAGE mainboardinit arch/i386/lib/noop_failover.inc end
+#
+###
+### Romcc output
+###
+#makerule ./failover.E dep "$(CONFIG_MAINBOARD)/failover.c" act "$(CPP) -I$(TOP)/src $(CPPFLAGS) $(CONFIG_MAINBOARD)/failover.c > ./failever.E"
+#makerule ./failover.inc dep "./romcc ./failover.E" act "./romcc -O ./failover.E > failover.inc"
+#mainboardinit ./failover.inc
+makerule ./auto.E dep "$(CONFIG_MAINBOARD)/auto.c" act "$(CPP) -I$(TOP)/src -$(ROMCCPPFLAGS) $(CPPFLAGS) $(CONFIG_MAINBOARD)/auto.c > ./auto.E"
+makerule ./auto.inc dep "./romcc ./auto.E" act "./romcc -O ./auto.E > auto.inc"
+mainboardinit ./auto.inc
+#
+###
+### Include the secondary Configuration files
+###
+northbridge amd/amdk8
+end
+southbridge amd/amd8111
+end
+#mainboardinit arch/i386/smp/secondary.inc
+superio nsc/pc87360
+ register "com1={1} com2={0} floppy=1 lpt=1 keyboard=1"
+end
+dir /pc80
+##dir /src/superio/winbond/w83627hf
+cpu p5 end
+cpu p6 end
+cpu k7 end
+cpu k8 end
+#
+###
+### Build the objects we have code for in this directory.
+###
+##object mainboard.o
+driver mainboard.o
+object static_devices.o
+if CONFIG_HAVE_MP_TABLE object mptable.o end
+if CONFIG_HAVE_PIRQ_TABLE object irq_tables.o end
+### Location of the DIMM EEPROMS on the SMBUS
+### This is fixed into a narrow range by the DIMM package standard.
+###
+option SMBUS_MEM_DEVICE_START=(0xa << 3)
+option SMBUS_MEM_DEVICE_END=(SMBUS_MEM_DEVICE_START +1)
+option SMBUS_MEM_DEVICE_INC=1
+#
+### The linuxBIOS bootloader.
+###
+option CONFIG_PAYLOAD_SIZE = (CONFIG_ROM_SECTION_SIZE - CONFIG_ROM_IMAGE_SIZE)
+option CONFIG_ROM_PAYLOAD_START = (0xffffffff - CONFIG_ROM_SIZE + CONFIG_ROM_SECTION_OFFSET + 1)
+#
+
+\end{verbatim}
+
+I've found the output of the new tool to be easier to
+handle. Makefile.settings looks like this, for example:
+\begin{verbatim}
+TOP:=/home/rminnich/src/yapps2/freebios2
+TARGET_DIR:=x
+export CONFIG_MAINBOARD:=/home/rminnich/src/yapps2/freebios2/src/mainboard/amd/solo
+export CONFIG_ARCH:=i386
+export CONFIG_RAMBASE:=0x4000
+export CONFIG_ROM_IMAGE_SIZE:=65535
+export CONFIG_PAYLOAD_SIZE:=131073
+export CONFIG_MAX_CPUS:=1
+export CONFIG_HEAP_SIZE:=8192
+export CONFIG_STACK_SIZE:=8192
+export CONFIG_MEMORY_HOLE:=0
+export COREBOOT_VERSION:=1.1.0
+export CC:=$(CONFIG_CROSS_COMPILE)gcc
+
+\end{verbatim}
+
+In other words, instead of expressions, we see the values. It's easier to
+deal with.
+