Merely document that the file is based on the pci_ids.h file
from the Linux kernel (trivial).
We assume all code from Linux to be licensed under the terms of the
GPLv2, if no explicit license header says something else.
Signed-off-by: Uwe Hermann <uwe@hermann-uwe.de>
Acked-by: Uwe Hermann <uwe@hermann-uwe.de>
git-svn-id: svn://coreboot.org/repository/LinuxBIOSv3@248 f3766cd6-281f-0410-b1cd-43a5c92072e9
header, and I'm too lazy to do the research for such simple functions.
First, drop all functions, as we don't really use all of them. Only
reimplement from scratch the following ones: strnlen(), strlen(), strcmp().
No fancy, fast, or small implementations, but I think we don't care
(they're only called once or twice anyway). But, at least all functions
are well-documented now.
Signed-off-by: Uwe Hermann <uwe@hermann-uwe.de>
Acked-by: Stefan Reinauer <stepan@coresystems.de>
git-svn-id: svn://coreboot.org/repository/LinuxBIOSv3@247 f3766cd6-281f-0410-b1cd-43a5c92072e9
Don't put our own license header in the file, just leave it as is (with
a note where it was taken from).
Signed-off-by: Uwe Hermann <uwe@hermann-uwe.de>
Acked-by: Stefan Reinauer <stepan@coresystems.de>
git-svn-id: svn://coreboot.org/repository/LinuxBIOSv3@246 f3766cd6-281f-0410-b1cd-43a5c92072e9
This tree shows the new model. It demonstrates the constructor array
in use, for devices that are and are not specified in the dts. It
introduces a new generic structure, device_id, analogous to
device_path, which can describe all the types of device IDs we have.
It shows a way to set up arrays of structs, in the dts, for the
constructors, so we avoid ldscript hacks.
Signed-off-by: Ronald G. Minnich <rminnich@gmail.com>
Acked-by: Stefan Reinauer <stepan@coresystems.de>
git-svn-id: svn://coreboot.org/repository/LinuxBIOSv3@233 f3766cd6-281f-0410-b1cd-43a5c92072e9
Make udelay() use 'unsigned int' instead of 'int'.
Fix interface of log2(), the current implementation takes an
'unsigned int' and returns an 'int'.
Purely cosmetic fix in the license header (consistency reasons).
Signed-off-by: Uwe Hermann <uwe@hermann-uwe.de>
Acked-by: Stefan Reinauer <stepan@coresystems.de>
git-svn-id: svn://coreboot.org/repository/LinuxBIOSv3@231 f3766cd6-281f-0410-b1cd-43a5c92072e9
Update the file to the latest version from the Linux kernel (trivial).
Signed-off-by: Uwe Hermann <uwe@hermann-uwe.de>
Acked-by: Uwe Hermann <uwe@hermann-uwe.de>
git-svn-id: svn://coreboot.org/repository/LinuxBIOSv3@229 f3766cd6-281f-0410-b1cd-43a5c92072e9
(e81e557a8f1e22fab1b1bdc649ac0d24017ddce2).
We stripped all PCI vendor IDs and device IDs, as we will maintain
our own short list for LinuxBIOS (we don't need that many entries).
Signed-off-by: Uwe Hermann <uwe@hermann-uwe.de>
Acked-by: Stefan Reinauer <stepan@coresystems.de>
git-svn-id: svn://coreboot.org/repository/LinuxBIOSv3@213 f3766cd6-281f-0410-b1cd-43a5c92072e9
kexec-tools 1.101 (the current version).
Document the origin of the files in HACKING and in the files themselves.
Signed-off-by: Uwe Hermann <uwe@hermann-uwe.de>
Acked-by: Stefan Reinauer <stepan@coresystems.de>
git-svn-id: svn://coreboot.org/repository/LinuxBIOSv3@203 f3766cd6-281f-0410-b1cd-43a5c92072e9
Comitting IP checksum code, and its usage in linuxbios table creation.
Mods to the dts, so that the device ops for the domain are set to the
proper structure. This change is important. It gets rid of the obscure,
confusing use of the enable_dev function to pick the right ops for a
device. It makes the ops initilization very clear at the top level, in
the dts. This has been tested and works, linux boots on Bochs under
this version.
Signed-off-by: Ronald G. Minnich <rminnich@gmail.com>
Acked-by: Stefan Reinauer <stepan@coresystems.de>
git-svn-id: svn://coreboot.org/repository/LinuxBIOSv3@191 f3766cd6-281f-0410-b1cd-43a5c92072e9
code (This goes hand in hand, as some parts of serial were hardcoded
to one architecture until now)
* include/uart8250.h: add TTYSx defines that are used in
multiple places. formerly part of arch/x86/serial.c.
Drop init_uart8250 (unused)
* lib/uart8250.c: drop arch/x86/config.h usage
Drop init_uart8250 (unused)
* arch/powerpc/Kconfig, arch/x86/Kconfig: add CONFIG_ARCH to
contain the directory name under LinuxBIOSv3/arch/
* arch/x86/console.c: drop some dead code. Drop hardcoded config.h values.
use generic uart8250.h header
* arch/x86/cachemain.c: Drop hardcoded config.h values. Still use hardcoded
ROM size for now. (To be changed later)
* arch/x86/config.h: dropped, no longer needed
* arch/x86/serial.c: factor out generically used defines to uart8250.h
* Makefile: use mainboard architecture instead of target architecture to choose
include path.
Read .xcompile if available (configure replacement).
Create build.h with compile time and version.
* util/xcompile/xcompile: new file. Search for supported cross compilers
linkers, assemblers (and potentially supported compiler flags etc)
This is a very slick configure replacement.
* util/dtc/Makefile: fix Makefile for cross compilation
Signed-off-by: Stefan Reinauer <stepan@coresystems.de>
Signed-off-by: Uwe Hermann <uwe@hermann-uwe.de>
Acked-by: Ronald G. Minnich <rminnich@gmail.com>
git-svn-id: svn://coreboot.org/repository/LinuxBIOSv3@190 f3766cd6-281f-0410-b1cd-43a5c92072e9
- fix ifdef in x86/arch/pci_ops.h
- fix function prototypes to match declarations in include/device/pci.h
Signed-off-by: Stefan Reinauer <stepan@coresystems.de>
Acked-by: Ronald G. Minnich <rminnich@gmail.com>
git-svn-id: svn://coreboot.org/repository/LinuxBIOSv3@189 f3766cd6-281f-0410-b1cd-43a5c92072e9
This patch includes the LinuxBIOS-related modifications.
We do not remove any (probably unused) parts of the file anymore.
Sure, the file is bigger now, but a lot easier to diff against
the upstream source, which is more important.
Signed-off-by: Uwe Hermann <uwe@hermann-uwe.de>
Acked-by: Ronald G. Minnich <rminnich@gmail.com>
git-svn-id: svn://coreboot.org/repository/LinuxBIOSv3@187 f3766cd6-281f-0410-b1cd-43a5c92072e9
Signed-off-by: Stefan Reinauer <stepan@coresystems.de>
Acked-by: Ronald G. Minnich <rminnich@gmail.com>
git-svn-id: svn://coreboot.org/repository/LinuxBIOSv3@186 f3766cd6-281f-0410-b1cd-43a5c92072e9
revised BSD license (without advertising clause).
Add the BSD license text to the files, in addition to the GPL header.
Signed-off-by: Uwe Hermann <uwe@hermann-uwe.de>
Acked-by: Stefan Reinauer <stepan@coresystems.de>
git-svn-id: svn://coreboot.org/repository/LinuxBIOSv3@176 f3766cd6-281f-0410-b1cd-43a5c92072e9
Signed-off-by: Ronald G. Minnich <rminnich@gmail.com>
Acked-by: Ronald G. Minnich <rminnich@gmail.com>
Acked-by: Stefan Reinauer <stepan@coresystems.de>
git-svn-id: svn://coreboot.org/repository/LinuxBIOSv3@172 f3766cd6-281f-0410-b1cd-43a5c92072e9
Phase 1: done
Phase 2: early setup ...
Phase 2: done
Phase 3: Enumerating buses...
qemu-x86 enable_dev done
dev_phase3_scan: scanning Root Device
scan_static_bus for root(Root Device)
cpus: Unknown device path type: 0
cpus() enabled
i440bxemulation_enable_dev:
i440bxemulation_enable_dev: DONE
northbridge_intel_i440bxemulation() enabled
northbridge_intel_i440bxemulation() scanning...
dev_phase3_scan: scanning
pci_scan_bus start
PCI: pci_scan_bus for bus 00
PCI: scan devfn 0x0 to 0xff
PCI: devfn 0x0
PCI: pci_scan_bus pci_scan_get_dev returns dev <NULL>
Change dts compiler to emit a new struct member, dtsname, for devices,
so we can get actual useful names for things.
Several mods and printks added.
printk(BIOS_SPEW
gives no output for reasons I don't understand.
Next in line is bringing back v2 support for pci, but not doing it the
way v2 does it.
note the cpus() printk above. cpus don't have a valid path yet.
We still need to work out the dts syntax for systems with multiple links
(opteron)
Signed-off-by: Ronald G. Minnich <rminnich@gmail.com>
Acked-by: Ronald G. Minnich <rminnich@gmail.com>
Acked-by: Stefan Reinauer <stepan@coresystems.de>
git-svn-id: svn://coreboot.org/repository/LinuxBIOSv3@163 f3766cd6-281f-0410-b1cd-43a5c92072e9
Also added comments.
A big change is in the dts. In OFW trees, the hierarchy seems to be:
/root/cpu/northbridge
south
other pci
note that the north is in the hierarchy under the south. This hierarchy
makes no sense on systems with a shared frontside bus, or at least I
don't see how it can. In those systems, it's easier to think about
the CPUs AND northbridges as children of the front side bus.
in LinuxBIOS, it has always been this:
/root/cpu/whatever
/root/northbridge/
south
other pci
There have been many discussions over how it ought to be, for 8 years
now, and we've always come back to how LB does it. So I have changed the
dts for qemu for now to match LB's way of doing things. Note that the
new system is flexible enough that, on K8, we CAN do things as above:
/cpu@0/amd8knorthbridge/etc.
/cpu@1/amd8knorthbridge/etc.
But on qemu, for now, the root is the mainboard, and the CPU and
northbridge are siblings.
I've added some informational printks, cleaned up pci_ops, and done
other things so that it builds and all seems to work -- until it hangs
hard in enumeration in i440bx ...
Signed-off-by: Ronald G. Minnich <rminnich@gmail.com>
Acked-by: Ronald G. Minnich <rminnich@gmail.com>
Acked-by: Stefan Reinauer <stepan@coresystems.de>
git-svn-id: svn://coreboot.org/repository/LinuxBIOSv3@160 f3766cd6-281f-0410-b1cd-43a5c92072e9
Some minor cosmetic fixes (trivial).
Signed-off-by: Uwe Hermann <uwe@hermann-uwe.de>
Acked-by: Uwe Hermann <uwe@hermann-uwe.de>
git-svn-id: svn://coreboot.org/repository/LinuxBIOSv3@155 f3766cd6-281f-0410-b1cd-43a5c92072e9
* exchange [u]intXX_t with [us]XX types as we agreed upon.
* drop unneeded btext.h for now
* fix license header in device/pci_ops.c
* fix printk in device/pci_ops.c
* add signed types to arch/types.h
* fix include/elf.h (remove glibc-ism)
* add appropriate copyright notice to lib/delay.c
* add pciconf.h header
Signed-off-by: Stefan Reinauer <stepan@coresystems.de>
Acked-by: Stefan Reinauer <stepan@coresystems.de>
Acked-by: Ronald G. Minnich <rminnich@gmail.com>
git-svn-id: svn://coreboot.org/repository/LinuxBIOSv3@144 f3766cd6-281f-0410-b1cd-43a5c92072e9
renamed the phase3 etc. to stuff like phase3_scan, so you can get a
rought idea what it is. The names mean more.
adding pci_device and, at the same time, showing how we can get rid of
the really ugly stuff that crept in. note you can specify ops in the
dts, which avoids the need for hideous stuff like this:
static void enable_dev(struct device *dev)
{
/* Set the operations if it is a special bus type */
if (dev->path.type == DEVICE_PATH_PCI_DOMAIN) {
dev->ops = &pci_domain_ops;
pci_set_method(dev);
}
else if (dev->path.type == DEVICE_PATH_APIC_CLUSTER) {
dev->ops = &cpu_bus_ops;
}
}
So that foolishness is gone.
added delay functions.
Note that we have include/lib.h, and define all the functions in there,
instead of in lots of fiddly includes.
Brought back the enable op, once I understood it; renamed it to
something that makes sense.
I'll be on a plane soon, will continue to work, but at least you can see
what's going on here.
Signed-off-by: Ronald G. Minnich <rminnich@gmail.com>
Acked-by: Ronald G. Minnich <rminnich@gmail.com>
Acked-by: Stefan Reinauer <stepan@coresystems.de>
git-svn-id: svn://coreboot.org/repository/LinuxBIOSv3@139 f3766cd6-281f-0410-b1cd-43a5c92072e9
Remaining to be done:
PIRQ table
MP table
ACPI table
relocate the GDT reload to a sensible place -- or figure out if we even
need to bother; we're back to execute in place and the GDT is in flash
rom ==> we can leave it until we boot the payload I think.
At this point, however, we've got the basic bits to sensibly create
tables.
This has been tested and works on bochs.
It is working to a point, but still failing on qemu for reasons we don't
quite understand.
Elfboot
Found ELF candidate at offset 0
New segment addr 0x100000 size 0x21310 offset 0xc0 filesize 0x7348
(cleaned up) New segment addr 0x100000 size 0x7348 offset 0xc0
set 00100000 to 0 for 0 bytes
Copy to 00100000 from fffc3f24 for 29512 bytes
New segment addr 0x121320 size 0x48 offset 0x7420 filesize 0x48
(cleaned up) New segment addr 0x121320 size 0x48 offset 0x7420
set 00121320 to 0 for 0 bytes
Copy to 00121320 from fffcb284 for 72 bytes
Dropping non PT_LOAD segment
Dropping non PT_LOAD segment
Jumping to boot code at 0x1047c0
FILO version 0.5 (rminnich@q.ccstar.lanl.gov) Sun Feb 25 10:19:16 MST
2007
collect_sys_info: boot eax = 0xfe
collect_sys_info: boot ebx = 0xffffd4ca
collect_sys_info: boot arg = 0x1047c0
collect_linuxbios_info: NOT Searching for LinuxBIOS tables...
Can't get memory map from firmware. Using hardcoded default.
collect_sys_info: 0000000000000000-00000000000a0000
collect_sys_info: 0000000000100000-0000000002000000
collect_sys_info: RAM 32 MB
relocate: Current location: 0x1000c7-0x12142e
relocate: Relocating to 0x1fdec90-0x1fffff7... ok
Press <Enter> for default boot, or <Esc> for boot prompt...
boot: hdc1:/phase1 root=/dev/hdc1 console=ttyS0,115200
malloc_check: invalid head->prev_size: 0x0
Signed-off-by: Ronald G. Minnich <rminnich@gmail.com>
Acked-by: Ronald G. Minnich <rminnich@gmail.com>
Acked-by: Stefan Reinauer <stepan@coresystems.de>
git-svn-id: svn://coreboot.org/repository/LinuxBIOSv3@133 f3766cd6-281f-0410-b1cd-43a5c92072e9
resource .c functions -- sigh -- this will be next.
fix types, fix some usage, but we do not yet generate pirqi, mptables,
or acpi. This will be next.
This code will be called from phase6 in stage 2, and probably
attached to a cpu. Or, it will be inline in the stage2.c code, since
the need for tables is generic.
I'm also going to move the cryptic usage of lgdt from here to stage 2,
phase 2, of the cpu code. The device model improvements are making it
easier to think about when things happen, we are finding.
I'm going to get lunch and go see some of brussels. It's a miserable day
but ...
Signed-off-by: Ronald G. Minnich <rminnich@gmail.com>
Acked-by: Ronald G. Minnich <rminnich@gmail.com>
Acked-by: Stefan Reinauer <stepan@coresystems.de>
git-svn-id: svn://coreboot.org/repository/LinuxBIOSv3@130 f3766cd6-281f-0410-b1cd-43a5c92072e9
We have merged the plethora of include files into one. A given linuxbios
target architecture must support all the functions described therein.
All the structs etc. in include/tables.h are known to be
architecture-independent.
We hope this new layout is easier to folow than the old one.
Todo: Remove the LGDT code from tables writing (how did THAT get in
there ;-) and put it somewhere sane; add OFW table support. We are going
to need some nice OFW table code.
Also, the license headers should be correct in this commit.
Signed-off-by: Ronald G. Minnich <rminnich@gmail.com>
Acked-by: Ronald G. Minnich <rminnich@gmail.com>
Acked-by: Stefan Reinauer <stepan@coresystems.de>
git-svn-id: svn://coreboot.org/repository/LinuxBIOSv3@128 f3766cd6-281f-0410-b1cd-43a5c92072e9
understood. This file has been tested and works with FILO 0.5.
Signed-off-by: Ronald G. Minnich <rminnich@gmail.com>
Acked-by: Ronald G. Minnich <rminnich@gmail.com>
Acked-by: Stefan Reinauer <stepan@coresystems.de>
git-svn-id: svn://coreboot.org/repository/LinuxBIOSv3@123 f3766cd6-281f-0410-b1cd-43a5c92072e9
Signed-off-by: Stefan Reinauer <stepan@coresystems.de>
Acked-by: Ronald G. Minnich <rminnich@gmail.com>
git-svn-id: svn://coreboot.org/repository/LinuxBIOSv3@118 f3766cd6-281f-0410-b1cd-43a5c92072e9
Signed-off-by: Stefan Reinauer <stepan@coresystems.de>
Acked-by: Ronald G. Minnich <rminnich@gmail.com>
git-svn-id: svn://coreboot.org/repository/LinuxBIOSv3@111 f3766cd6-281f-0410-b1cd-43a5c92072e9
We're trying to avoid the bounce buffer mess, which is really complex,
by running elfboot out of the boot block. This code includes a really
dumb allocator in elfboot which may or may not work. It basically
allocates off the elfboot() stack.
This builds, but elfboot is not tested. That's next.
Signed-off-by: Ronald G. Minnich <rminnich@gmail.com>
Acked-by: Ronald G. Minnich <rminnich@gmail.com>
Acked-by: Stefan Reinauer <stepan@coresystems.de>
git-svn-id: svn://coreboot.org/repository/LinuxBIOSv3@102 f3766cd6-281f-0410-b1cd-43a5c92072e9
Signed-off-by: Ronald G. Minnich <rminnich@gmail.com>
Acked-by: Ronald G. Minnich <rminnich@gmail.com>
Acked-by: Stefan Reinauer <stepan@coresystems.de>
git-svn-id: svn://coreboot.org/repository/LinuxBIOSv3@101 f3766cd6-281f-0410-b1cd-43a5c92072e9
Some updates to the docs, but these will change more.
add memcmp. Crude implementation, but it really does not matter.
add compute_ip_checksum.c
mods to elfboot. Thanks to our new design, BOUNCE BUFFERS ARE DEAD. Yee
ha!
updates to makefile
Signed-off-by: Ronald G. Minnich <rminnich@gmail.com>
Acked-by: Ronald G. Minnich <rminnich@gmail.com>
Acked-by: Stefan Reinauer <stepan@coresystems.de>
git-svn-id: svn://coreboot.org/repository/LinuxBIOSv3@96 f3766cd6-281f-0410-b1cd-43a5c92072e9
* add generic lib.h header for lib/ prototypes
* fix some prototypes
* put back return value parsing of stage2
Signed-off-by: Stefan Reinauer <stepan@coresystems.de>
Acked-by: Ronald G. Minnich <rminnich@gmail.com>
git-svn-id: svn://coreboot.org/repository/LinuxBIOSv3@91 f3766cd6-281f-0410-b1cd-43a5c92072e9
Signed-off-by: Ronald G. Minnich <rminnich@gmail.com>
Acked-by: Ronald G. Minnich <rminnich@gmail.com>
Acked-by: Stefan Reinauer <stepan@coresystems.de>
git-svn-id: svn://coreboot.org/repository/LinuxBIOSv3@90 f3766cd6-281f-0410-b1cd-43a5c92072e9
directories in include, since having all those directories has obscured
understanding.
Add elfboot.c to lib. Same argument, it is not in boot, it is in lib,
since fewer directories is better.
Signed-off-by: Ronald G. Minnich <rminnich@gmail.com>
Acked-by: Ronald G. Minnich <rminnich@gmail.com>
Acked-by: Stefan Reinauer <stepan@coresystems.de>
git-svn-id: svn://coreboot.org/repository/LinuxBIOSv3@89 f3766cd6-281f-0410-b1cd-43a5c92072e9
Fix up dts to set up ops struct member. Fix dts for qemu mainboard.
We are getting past stage2 now, it is time for elfboot.
Signed-off-by: Ronald G. Minnich <rminnich@gmail.com>
Acked-by: Ronald G. Minnich <rminnich@gmail.com>
Acked-by: Stefan Reinauer <stepan@coresystems.de>
git-svn-id: svn://coreboot.org/repository/LinuxBIOSv3@88 f3766cd6-281f-0410-b1cd-43a5c92072e9
* rename magic device callbacks to phases
* drop obsolete printk_* macros
* add some prototypes to lar.h
Signed-off-by: Ronald G. Minnich <rminnich@gmail.com>
Acked-by: Stefan Reinauer <stepan@coresystems.de>
git-svn-id: svn://coreboot.org/repository/LinuxBIOSv3@74 f3766cd6-281f-0410-b1cd-43a5c92072e9
Use standard LinuxBIOS license header (trivial, no semantic changes).
Signed-off-by: Uwe Hermann <uwe@hermann-uwe.de>
Acked-by: Uwe Hermann <uwe@hermann-uwe.de>
git-svn-id: svn://coreboot.org/repository/LinuxBIOSv3@63 f3766cd6-281f-0410-b1cd-43a5c92072e9
accordance to the newboot document:
* reset vector (16 bytes)
* vpd (240bytes)
* boot block (8k - 256b)
* lar archive (256-8 k)
The boot block is kind of simple, still. It enables pmode, car, and
starts looking for an initram module in the lar archive.
Note: This doesnt do much at the moment,
as gas seems to produce buggy code in init.S.
Take this as a suggestion of how it might work and please provide
patches fixing it and bringing it into shape.
Signed-off-by: Stefan Reinauer <stepan@coresystems.de>
Acked-by: Ronald G. Minnich <rminnich@gmail.com>
git-svn-id: svn://coreboot.org/repository/LinuxBIOSv3@62 f3766cd6-281f-0410-b1cd-43a5c92072e9
* fix arch/io.h to use consistent types
* add compression code and start integration into Kconfig
* update to newer version of Kconfig, and rename some occurences
of "Linux" to "LinuxBIOS"
* set up Make framework to create linuxbios.rom
Signed-off-by: Stefan Reinauer <stepan@coresystems.de>
Acked-by: Ronald G Minnich <rminnich@lanl.gov>
git-svn-id: svn://coreboot.org/repository/LinuxBIOSv3@55 f3766cd6-281f-0410-b1cd-43a5c92072e9
These are all trivial changes.
Signed-off-by: Stefan Reinauer <stepan@coresystems.de>
Acked-by: Stefan Reinauer <stepan@coresystems.de>
git-svn-id: svn://coreboot.org/repository/LinuxBIOSv3@50 f3766cd6-281f-0410-b1cd-43a5c92072e9
Signed-off-by: Segher Boessenkool <segher@kernel.crashing.org>
Acked-by: Peter Stuge <peter@stuge.se>
Acked-by: Ronald G Minnich <rminnich@gmail.com>
Acked-by: Stefan Reinauer <stepan@coresystems.de>
git-svn-id: svn://coreboot.org/repository/LinuxBIOSv3@47 f3766cd6-281f-0410-b1cd-43a5c92072e9
