coreboot.git
14 years agoUpdate equivalent processor revision ID to load latest microcode patches and
Marc Jones [Thu, 14 May 2009 23:42:41 +0000 (23:42 +0000)]
Update equivalent processor revision ID to load latest microcode patches and
register setting for all FAM10 processors.
This does not include new errata for FAM10 C2.

Signed-off-by: Marc Jones <marcj303@gmail.com>
Acked-by: Vincent Lim (vincent.lim@amd.com)
git-svn-id: svn://svn.coreboot.org/coreboot/trunk@4288 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1

14 years agoTrivia: remove comment
Vincent Lim [Thu, 14 May 2009 22:11:08 +0000 (22:11 +0000)]
Trivia: remove comment

Signed-off-by: Vincent Lim <vincent.lim@amd.com>
Acked-by: Vincent Lim <vincent.lim@amd.com>
git-svn-id: svn://svn.coreboot.org/coreboot/trunk@4287 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1

14 years agoFrom AMD family 10h Processor BKDG (rev. D): a platform is capable of having up to...
Vincent Lim [Thu, 14 May 2009 22:00:28 +0000 (22:00 +0000)]
From AMD family 10h Processor BKDG (rev. D): a platform is capable of having up to 8 nodes, and each nodes supports 1,2,3,4,5, or 6 cores.

Signed-off-by: Vincent Lim <vincent.lim@amd.com>
Acked-by: Ronald G. Minnich <rminnich@gmail.com>
git-svn-id: svn://svn.coreboot.org/coreboot/trunk@4286 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1

14 years ago#136: failed to boot under KVM\QEMU
Ronald G. Minnich [Thu, 14 May 2009 21:26:28 +0000 (21:26 +0000)]
#136: failed to boot under KVM\QEMU
> -------------------------------------+--------------------------------------
>   Reporter:  silicium@…             |          Owner:  somebody
>       Type:  defect                 |         Status:  new
>   Priority:  major                  |      Milestone:
>  Component:  coreboot               |        Version:  v2
>   Keywords:                         |   Dependencies:
> Patchstatus:  patch needs review     |
> -------------------------------------+--------------------------------------

Fix use of uninitialized pointers. To help in future, move
the declaration to the same scope as the use.

Signed-off-by: Ronald G. Minnich <rminnich@gmail.com>
Acked-by: Myles Watson <mylesgw@gmail.com>
git-svn-id: svn://svn.coreboot.org/coreboot/trunk@4285 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1

15 years agoThe cbfstool print command should pretty-print the type of components that are
Ward Vandewege [Thu, 14 May 2009 03:00:15 +0000 (03:00 +0000)]
The cbfstool print command should pretty-print the type of components that are
type 'deleted'.

Signed-off-by: Ward Vandewege <ward@gnu.org>
Acked-by: Peter Stuge <peter@stuge.se>
git-svn-id: svn://svn.coreboot.org/coreboot/trunk@4284 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1

15 years agoThis patch fixes a segfault when a file too large to fit is added to a rom
Ward Vandewege [Wed, 13 May 2009 20:11:04 +0000 (20:11 +0000)]
This patch fixes a segfault when a file too large to fit is added to a rom
image.

Signed-off-by: Ward Vandewege <ward@gnu.org>
Acked-by: Myles Watson <mylesgw@gmail.com>
git-svn-id: svn://svn.coreboot.org/coreboot/trunk@4283 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1

15 years agoAdd support for human-friendly component string types for the cbfstool add
Ward Vandewege [Wed, 13 May 2009 20:08:28 +0000 (20:08 +0000)]
Add support for human-friendly component string types for the cbfstool add
command.

Make use of it in config.g (Myles)

Signed-off-by: Ward Vandewege <ward@gnu.org>
Acked-by: Myles Watson <mylesgw@gmail.com>
git-svn-id: svn://svn.coreboot.org/coreboot/trunk@4282 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1

15 years agoAllow dynamic size for the {s,}elfboot bounce buffer.
Patrick Georgi [Wed, 13 May 2009 16:27:25 +0000 (16:27 +0000)]
Allow dynamic size for the {s,}elfboot bounce buffer.
Use that to fix selfboot with compressed payloads.

Signed-off-by: Patrick Georgi <patrick.georgi@coresystems.de>
Acked-by: Ronald G. Minnich <rminnich@gmail.com>
git-svn-id: svn://svn.coreboot.org/coreboot/trunk@4281 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1

15 years agoMake ACPI with low and high tables work again. The RSDP contained a
Patrick Georgi [Wed, 13 May 2009 14:39:59 +0000 (14:39 +0000)]
Make ACPI with low and high tables work again. The RSDP contained a
bogus RSDT pointer due to a wrong order of commands.

Signed-off-by: Patrick Georgi <patrick.georgi@coresystems.de>
Acked-by: Myles Watson <mylesgw@gmail.com>
git-svn-id: svn://svn.coreboot.org/coreboot/trunk@4280 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1

15 years agoRemove a shadowed variable and an unnecessary local variable in cbfstool/fs.c.
Myles Watsonmylesgw [Wed, 13 May 2009 02:48:58 +0000 (02:48 +0000)]
Remove a shadowed variable and an unnecessary local variable in cbfstool/fs.c.

It is nearly trivial.

Signed-off-by: Myles Watson<mylesgw@gmail.com>
Acked-by: Ronald G. Minnich <rminnich@gmail.com>
git-svn-id: svn://svn.coreboot.org/coreboot/trunk@4279 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1

15 years agoOops forgot small part. Set up PIRQs in mainboard Config.lb for IP1000 and RM4100...
Joseph Smith [Wed, 13 May 2009 02:47:14 +0000 (02:47 +0000)]
Oops forgot small part. Set up PIRQs in mainboard Config.lb for IP1000 and RM4100 instead of using the ones in i82801xx_lpc.c.
Signed-off-by: Joseph Smith <joe@settoplinux.org>
Acked-by: Myles Watson <mylesgw@gmail.com>
git-svn-id: svn://svn.coreboot.org/coreboot/trunk@4278 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1

15 years agoThis is the final patch that got everything working for me with the HP dl145g3.
Samuel Verstraete [Tue, 12 May 2009 15:15:07 +0000 (15:15 +0000)]
This is the final patch that got everything working for me with the HP dl145g3.
I would like to remind you that this firmware enables the hardware
virtualization on the AMD cpu's on the machine. That feature was
explicitly disabled by the factory BIOS.
Due to an error in the VGAROM no other rom loader (YABEL or X*^BIOS)
than SeaBIOS manages to load the VGA rom. The VGA ROM tries to read
config space of a device that is actually not present.
Because SeaBIOS does not support AHCI SATA it can not start the
bootable drive of the machine so i had to add filo to seabios to
manage booting:
./cbfstool coreboot.rom add-payload filo.elf img/FILO

Signed-off-by: Samuel Verstraete <samuel.verstraete@gmail.com>
Acked-by: Myles Watson <mylesgw@gmail.com>
git-svn-id: svn://svn.coreboot.org/coreboot/trunk@4277 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1

15 years agoThis fixes a rather silly bug in cbfs with filenames > 16 characters.
Ronald G. Minnich [Tue, 12 May 2009 15:06:54 +0000 (15:06 +0000)]
This fixes a rather silly bug in cbfs with filenames > 16 characters.

Tested to booting linux with qemu.
Signed-off-by: Ronald G. Minnich <rminnich@gmail.com>
Acked-by: Myles Watson<mylesgw@gmail.com>
git-svn-id: svn://svn.coreboot.org/coreboot/trunk@4276 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1

15 years agoThere's no 'svg2pdf' in Debian AFAICT, probably the same problem on
Uwe Hermann [Tue, 12 May 2009 14:24:25 +0000 (14:24 +0000)]
There's no 'svg2pdf' in Debian AFAICT, probably the same problem on
other systems too.

So, check for svg2pdf, convert, and inkscape and use the first one that is
found to convert the SVG files to PDF.

Signed-off-by: Uwe Hermann <uwe@hermann-uwe.de>
Acked-by: Uwe Hermann <uwe@hermann-uwe.de>
git-svn-id: svn://svn.coreboot.org/coreboot/trunk@4275 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1

15 years agoI guess a couple of function calls pushed these boards over the 0x1700 edge on
Myles Watson [Tue, 12 May 2009 14:14:54 +0000 (14:14 +0000)]
I guess a couple of function calls pushed these boards over the 0x1700 edge on
the build server.

Add Config-abuild.lb to fix s2892 & s2891.

Signed-off-by: Myles Watson <mylesgw@gmail.com>
Acked-by: Myles Watson <mylesgw@gmail.com>
git-svn-id: svn://svn.coreboot.org/coreboot/trunk@4274 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1

15 years agoTrivially copy Config.lb to Config-abuild.lb to fix asus/m2v-mx_se.
Myles Watson [Tue, 12 May 2009 14:03:12 +0000 (14:03 +0000)]
Trivially copy Config.lb to Config-abuild.lb to fix asus/m2v-mx_se.

Signed-off-by: Myles Watson <mylesgw@gmail.com>
Acked-by: Myles Watson <mylesgw@gmail.com>
git-svn-id: svn://svn.coreboot.org/coreboot/trunk@4273 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1

15 years agoFix pdflatex build issues (trivial).
Uwe Hermann [Tue, 12 May 2009 14:01:14 +0000 (14:01 +0000)]
Fix pdflatex build issues (trivial).

Signed-off-by: Uwe Hermann <uwe@hermann-uwe.de>
Acked-by: Uwe Hermann <uwe@hermann-uwe.de>
git-svn-id: svn://svn.coreboot.org/coreboot/trunk@4272 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1

15 years agoUse the debugging functions to print out the tree and resources.
Myles Watson [Tue, 12 May 2009 13:43:34 +0000 (13:43 +0000)]
Use the debugging functions to print out the tree and resources.

Signed-off-by: Myles Watson <mylesgw@gmail.com>
Acked-by: Carl-Daniel Hailfinger <c-d.hailfinger.devel.2006@gmx.net>
git-svn-id: svn://svn.coreboot.org/coreboot/trunk@4271 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1

15 years agoBring v3-style debug output to v2. Fix a minor typo.
Myles Watson [Mon, 11 May 2009 22:45:35 +0000 (22:45 +0000)]
Bring v3-style debug output to v2.  Fix a minor typo.

Signed-off-by: Myles Watson <mylesgw@gmail.com>
Acked-by: Ronald G. Minnich <rminnich@gmail.com>
git-svn-id: svn://svn.coreboot.org/coreboot/trunk@4270 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1

15 years agoThis patch adds high table support to qemu. It was already added to
Myles Watson [Mon, 11 May 2009 22:44:14 +0000 (22:44 +0000)]
This patch adds high table support to qemu.  It was already added to
src/northbridge/intel/i440bx/ but not to
src/cpu/emulation/qemu-x86/northbridge.c

It also adds a driver for the ISA device that is found when using
0.9.1  If you look in a log without this patch you won't find the RTC
init lines.

Signed-off-by: Myles Watson <mylesgw@gmail.com>
Acked-by: Ronald G. Minnich <rminnich@gmail.com>
git-svn-id: svn://svn.coreboot.org/coreboot/trunk@4269 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1

15 years agoTrivial white space fixes so that the next patches are easier to read.
Myles Watson [Mon, 11 May 2009 22:24:53 +0000 (22:24 +0000)]
Trivial white space fixes so that the next patches are easier to read.

Signed-off-by: Myles Watson <mylesgw@gmail.com>
Acked-by: Myles Watson <mylesgw@gmail.com>
git-svn-id: svn://svn.coreboot.org/coreboot/trunk@4268 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1

15 years agoWe should separate the it8718f_24mhz_clkin like the way IT8712 does.
Zheng Bao [Mon, 11 May 2009 13:45:11 +0000 (13:45 +0000)]
We should separate the it8718f_24mhz_clkin like the way IT8712 does.

Signed-off-by: Zheng Bao <zheng.bao@amd.com>
Acked-by; Ronald G. Minnich <rminnich@gmail.com>
Acked-by: Uwe Hermann <uwe@hermann-uwe.de>
git-svn-id: svn://svn.coreboot.org/coreboot/trunk@4267 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1

15 years agoFix manual build on the Kontron board (trivial).
Uwe Hermann [Mon, 11 May 2009 01:44:54 +0000 (01:44 +0000)]
Fix manual build on the Kontron board (trivial).

A manual build was yielding section overlaps, so increase
ROM_IMAGE_SIZE to the same size the Config-abuild.lb is using.

Build-tested by me using a manual build.

Signed-off-by: Uwe Hermann <uwe@hermann-uwe.de>
Acked-by: Uwe Hermann <uwe@hermann-uwe.de>
git-svn-id: svn://svn.coreboot.org/coreboot/trunk@4266 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1

15 years agoFollowing patch fixes the XIP computation issue. I removed the normal image
Rudolf Marek [Sun, 10 May 2009 20:35:18 +0000 (20:35 +0000)]
Following patch fixes the XIP computation issue. I removed the normal image
because it was not working anyway (it was hardcoded) and because it allows me to
fix the XIP base to something sane (and use generic computation and approach)

This board is bit tricky because until now it required the VGA BIOS on the flash
start. XIP will work with 64KB aligned base, therefore the VGA ROM image must be
aligned too to 64KB.

Signed-off-by: Rudolf Marek <r.marek@assembler.cz>
Acked-by: Ronald G. Minnich <rminnich@gmail.com>
git-svn-id: svn://svn.coreboot.org/coreboot/trunk@4265 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1

15 years agoI would have liked to get an ack, but the error this corrects is pretty
Ronald G. Minnich [Sat, 9 May 2009 17:14:58 +0000 (17:14 +0000)]
I would have liked to get an ack, but the error this corrects is pretty
critical, since unless it is fixed this tool creates empty tables that cause
coreboot to (in some cases, e.g. on qemu) triple fault and die.

For the record, an empty option_table is not allowed. The table must,
at least, have 3 32-bit entries in this order:
type -- should be 200, 0r 0xc8, i.e. 0xc8, 0, 0, 0
size of table in LE order, 4 bytes
size of header in LE order, which is always 12,0,0,0

Signed-off-by: Ronald G. Minnich <rminnich@gmail.com>
Acked-by: Ronald G. Minnich <rminnich@gmail.com>
git-svn-id: svn://svn.coreboot.org/coreboot/trunk@4264 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1

15 years agoTrivial clean up of print usage and parameter checking.
Myles Watson [Fri, 8 May 2009 20:07:00 +0000 (20:07 +0000)]
Trivial clean up of print usage and parameter checking.

Signed-off-by: Myles Watson <mylesgw@gmail.com>
Acked-by: Myles Watson <mylesgw@gmail.com>
git-svn-id: svn://svn.coreboot.org/coreboot/trunk@4263 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1

15 years agoAdd -Werror to help us keep the code clean.
Myles Watson [Fri, 8 May 2009 19:39:15 +0000 (19:39 +0000)]
Add -Werror to help us keep the code clean.
Change sizes from unsigned int to int.
Clean up some usage and parameter checking.

Signed-off-by: Myles Watson <mylesgw@gmail.com>
Acked-by: Ronald G. Minnich <rminnich@gmail.com>
git-svn-id: svn://svn.coreboot.org/coreboot/trunk@4262 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1

15 years agoI have made a very simple mod to cbfstool that is compatible with the
Ronald G. Minnich [Fri, 8 May 2009 19:23:00 +0000 (19:23 +0000)]
I have made a very simple mod to cbfstool that is compatible with the
src/lib/ code in coreboot. I.e. the tool changes but the coreboot code
does not.

Currently, as cbfstool manages the ROM, there are files and empty
space. To allocate files, the code does, first, a walk of the headers
and, if that fails, does a brute-force search of the rest of the
space.

We all agree that the brute-force search has lots of problems from a
performance and correctness standpoint.

I've made a slight change. Instead of an "empty space" area with no
valid headers, I've made a header for the empty space.

So cbfs creation looks like this:
- set up the boot block
- create a file, of type CBFS_COMPONENT_NULL, that contains the empty
space. CBFS_COMPONENT_NULL was already defined in cbfs.h

Here's an example:

[rminnich@xcpu2 cbfstool]$ ./cbfstool testcbfs create 1048576 2048
(cbfstool) E: Unable to open (null): Bad address
[rminnich@xcpu2 cbfstool]$ ./cbfstool testcbfs print
testcbfs: 1024 kB, bootblocksize 2048, romsize 1048576, offset 0x0
Alignment: 16 bytes

Name                           Offset     Type         Size
                              0x0        0xffffffff   1046456

So how do we create a new file?

It's easy: walk the files and find a file of type CBFS_COMPONENT_NULL,
which is as large
or larger than the file you are trying to create. Then you use that file.
- if the file is the same size as the NULL file, then it's easy: take it
- if the file is smaller than the NULL file, you split the NULL file
into two parts.

note that this works in the base case: the base case is that the whole
storage is CBFS_COMPONENT_NULL.

Here's an example of adding a file.
[rminnich@xcpu2 cbfstool]$ ./cbfstool testcbfs add-stage testfixed t
[rminnich@xcpu2 cbfstool]$ ./cbfstool testcbfs print
testcbfs: 1024 kB, bootblocksize 2048, romsize 1048576, offset 0x0
Alignment: 16 bytes

Name                           Offset     Type         Size
t                              0x0        stage        23176
                              0x5ab0     0xffffffff   1023240

Note that the NULL split and got smaller. But the entire ROM is still
contained by the two files. To walk this entire rom will require two
FLASH accesses.

Add another file:
[rminnich@xcpu2 cbfstool]$ ./cbfstool testcbfs add-stage testfixed tt
[rminnich@xcpu2 cbfstool]$ ./cbfstool testcbfs print
testcbfs: 1024 kB, bootblocksize 2048, romsize 1048576, offset 0x0
Alignment: 16 bytes

Name                           Offset     Type         Size
t                              0x0        stage        23176
tt                             0x5ab0     stage        23176
                              0xb560     0xffffffff   1000024
[rminnich@xcpu2 cbfstool]$

So, taking current ROMs as an example, I can reduce FLASH accesses for
cbfs from (potentially) thousands to (typically) less than 10.

Index: fs.c
Changes for readability and cleanliness. Move common blobs of code to functions.
New function: rom_alloc,which allocates files by finding NULL files and using/splitting.
Other changes as needed to support this usage.
Index: util.c
Creating a cbfs archive now requires creation of a NULL file covering the file system space.
Index: cbfs.h
Add a DELETED file type with value 0. Any file can be marked deleted by zero its type; this is a
FLASH-friendly definition for all known FLASH types.

Signed-off-by: Ronald G. Minnich <rminnich@gmail.com>
I think it is a step in the right direction.  Could you add the
function prototype to cbfstool.h?

Acked-by: Myles Watson <mylesgw@gmail.com>
(I added the prototype)

git-svn-id: svn://svn.coreboot.org/coreboot/trunk@4261 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1

15 years agoTrivial fixup IRQS on IP1000 and RM4100.
Joseph Smith [Fri, 8 May 2009 00:45:47 +0000 (00:45 +0000)]
Trivial fixup IRQS on IP1000 and RM4100.
Signed-off-by: Joseph Smith <joe@settoplinux.org>
Acked-by: Joseph Smith <joe@settoplinux.org>
git-svn-id: svn://svn.coreboot.org/coreboot/trunk@4260 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1

15 years agoSet up PIRQs in mainboard Config.lb for IP1000 and RM4100 instead of using the ones...
Joseph Smith [Fri, 8 May 2009 00:24:24 +0000 (00:24 +0000)]
Set up PIRQs in mainboard Config.lb for IP1000 and RM4100 instead of using the ones in i82801xx_lpc.c.
Signed-off-by: Joseph Smith <joe@settoplinux.org>
Acked-by: Myles Watson <mylesgw@gmail.com>
git-svn-id: svn://svn.coreboot.org/coreboot/trunk@4259 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1

15 years agoDisable the AC97 modem via the ICH4 LPC disable function register early in the boot...
Joseph Smith [Fri, 8 May 2009 00:19:13 +0000 (00:19 +0000)]
Disable the AC97 modem via the ICH4 LPC disable function register early in the boot process.
Signed-off-by: Joseph Smith <joe@settoplinux.org>
Acked-by: Myles Watson <mylesgw@gmail.com>
git-svn-id: svn://svn.coreboot.org/coreboot/trunk@4258 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1

15 years agoTrivial checksum fixup for irq tables on IP1000 and RM4100.
Joseph Smith [Thu, 7 May 2009 05:47:05 +0000 (05:47 +0000)]
Trivial checksum fixup for irq tables on IP1000 and RM4100.
Signed-off-by: Joseph Smith <joe@settoplinux.org>
Acked-by: Joseph Smith <joe@settoplinux.org>
git-svn-id: svn://svn.coreboot.org/coreboot/trunk@4257 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1

15 years agoFix my last commit. I looked at the wrong dead laptop.
Uwe Hermann [Thu, 7 May 2009 00:21:02 +0000 (00:21 +0000)]
Fix my last commit. I looked at the wrong dead laptop.

Signed-off-by: Uwe Hermann <uwe@hermann-uwe.de>
Acked-by: Uwe Hermann <uwe@hermann-uwe.de>
git-svn-id: svn://svn.coreboot.org/coreboot/trunk@4256 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1

15 years agoSupport for detecting the SMSC FDC37N869 (trivial).
Uwe Hermann [Thu, 7 May 2009 00:03:18 +0000 (00:03 +0000)]
Support for detecting the SMSC FDC37N869 (trivial).

No datasheet available, chip identified by probing and looking at the PCB.

Signed-off-by: Uwe Hermann <uwe@hermann-uwe.de>
Acked-by: Uwe Hermann <uwe@hermann-uwe.de>
git-svn-id: svn://svn.coreboot.org/coreboot/trunk@4255 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1

15 years agoFix a bug introduced in the copy_and_run refactoring.
Patrick Georgi [Tue, 5 May 2009 14:06:20 +0000 (14:06 +0000)]
Fix a bug introduced in the copy_and_run refactoring.
The new code always decompressed to dst (as it should)
and then jumped to _iseg, when it should jump to dst.

With dst != _iseg this breaks (coreboot_apc)

Signed-off-by: Patrick Georgi <patrick.georgi@coresystems.de>
Acked-by: Ward Vandewege <ward@gnu.org>
git-svn-id: svn://svn.coreboot.org/coreboot/trunk@4254 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1

15 years agoThis patch removes these warnings:
Myles Watson [Mon, 4 May 2009 20:27:09 +0000 (20:27 +0000)]
This patch removes these warnings:
Makefile:435: warning: overriding commands for target `src/lib/memset.o'

And replaces these debug messages:
partobj dir 0 parent <__main__.partobj instance at 0x7f1e846a7ab8>
part pci_domain
with:
partobj dir 0 parent northbridge_amd_amdk8_root_complex_dev2 part pci_domain

Signed-off-by: Myles Watson <mylesgw@gmail.com>
Acked-by: Ronald G. Minnich <rminnich@gmail.com>
git-svn-id: svn://svn.coreboot.org/coreboot/trunk@4253 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1

15 years agoThe rev 4099 broke ECC boards, they need to have tidy the ECC tags. Myles reverted...
Rudolf Marek [Mon, 4 May 2009 19:26:43 +0000 (19:26 +0000)]
The rev 4099 broke ECC boards, they need to have tidy the ECC tags. Myles reverted this change.
I think we can return the 4099 back under HAVE_ACPI_RESUME define to make everyone happy (and booting ;).

Signed-off-by: Rudolf Marek <r.marek@assembler.cz>
Acked-by: Rudolf Marek <r.marek@assembler.cz>
git-svn-id: svn://svn.coreboot.org/coreboot/trunk@4252 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1

15 years agoAssign PIRQs in mainboard Config.lb or use the default ones listed in i82801xx_lpc.c.
Joseph Smith [Sat, 2 May 2009 21:30:57 +0000 (21:30 +0000)]
Assign PIRQs in mainboard Config.lb or use the default ones listed in i82801xx_lpc.c.
Signed-off-by: Joseph Smith <joe@settoplinux.org>
Acked-by: Ronald G. Minnich <rminnich@gmail.com>
git-svn-id: svn://svn.coreboot.org/coreboot/trunk@4251 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1

15 years agoRun dos2unix on all files:
Stefan Reinauer [Sat, 2 May 2009 12:42:30 +0000 (12:42 +0000)]
Run dos2unix on all files:

find . -type f| grep -v svn | xargs dos2unix

Signed-off-by: Stefan Reinauer <stepan@coresystems.de>
Acked-by: Stefan Reinauer <stepan@coresystems.de>
git-svn-id: svn://svn.coreboot.org/coreboot/trunk@4250 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1

15 years agoTrivial, update email address.
Joseph Smith [Sat, 2 May 2009 00:59:03 +0000 (00:59 +0000)]
Trivial, update email address.
Signed-off-by: Joseph Smith <joe@settoplinux.org>
Acked-by: Joseph Smith <joe@settoplinux.org>
git-svn-id: svn://svn.coreboot.org/coreboot/trunk@4249 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1

15 years agoTrivial fix up to the GPIO's connected to the IP1000 and RM4100, only set ones that...
Joseph Smith [Sat, 2 May 2009 00:50:58 +0000 (00:50 +0000)]
Trivial fix up to the GPIO's connected to the IP1000 and RM4100, only set ones that are actually connected to something.
Signed-off-by: Joseph Smith <joe@settoplinux.org>
Acked-by: Joseph Smith <joe@settoplinux.org>
git-svn-id: svn://svn.coreboot.org/coreboot/trunk@4248 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1

15 years agoTrivial removal of svn:executable property from vga.c.
Peter Stuge [Fri, 1 May 2009 14:33:19 +0000 (14:33 +0000)]
Trivial removal of svn:executable property from vga.c.

Signed-off-by: Peter Stuge <peter@stuge.se>
Acked-by: Peter Stuge <peter@stuge.se>
git-svn-id: svn://svn.coreboot.org/coreboot/trunk@4247 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1

15 years agoThis is a patch to use another IRQ besides IRQ12 to fix conflicts with i8042 - PS...
Joseph Smith [Fri, 1 May 2009 04:53:58 +0000 (04:53 +0000)]
This is a patch to use another IRQ besides IRQ12 to fix conflicts with i8042 - PS/2 Mouse.
Signed-off-by: Joseph Smith <joe@settoplinux.org>
Acked-by: Myles Watson <mylesgw@gmail.com>
git-svn-id: svn://svn.coreboot.org/coreboot/trunk@4246 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1

15 years agoThis patch changes the default rom emulator to VM86.
Joseph Smith [Fri, 1 May 2009 04:44:36 +0000 (04:44 +0000)]
This patch changes the default rom emulator to VM86.
Signed-off-by: Joseph Smith <joe@settoplinux.org>
Acked-by: Myles Watson <mylesgw@gmail.com>
git-svn-id: svn://svn.coreboot.org/coreboot/trunk@4245 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1

15 years agoThis patch allows a custom vga driver that will give the flexibility to run code...
Joseph Smith [Fri, 1 May 2009 04:37:13 +0000 (04:37 +0000)]
This patch allows a custom vga driver that will give the flexibility to run code after vga is initialized for tv-out.
Signed-off-by: Joseph Smith <joe@settoplinux.org>
Acked-by: Myles Watson <mylesgw@gmail.com>
git-svn-id: svn://svn.coreboot.org/coreboot/trunk@4244 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1

15 years agoTrivial patch to make #defines generated from cmos.layout have unique names. Kills...
Myles Watson [Fri, 1 May 2009 04:36:38 +0000 (04:36 +0000)]
Trivial patch to make #defines generated from cmos.layout have unique names.  Kills a few more compiler warnings.

Signed-off-by: Myles Watson <mylesgw@gmail.com>
Acked-by: Myles Watson <mylesgw@gmail.com>
git-svn-id: svn://svn.coreboot.org/coreboot/trunk@4243 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1

15 years agoTrivial cosmetic fixes to IP1000 and RM4100 irq_tables.c Signed-off-by: Joseph Smith...
Joseph Smith [Fri, 1 May 2009 03:04:20 +0000 (03:04 +0000)]
Trivial cosmetic fixes to IP1000 and RM4100 irq_tables.c Signed-off-by: Joseph Smith <joe@settoplinux.org> Acked-by: Joseph Smith <joe@settoplinux.org>

git-svn-id: svn://svn.coreboot.org/coreboot/trunk@4242 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1

15 years agoThe semantics of the changed code are pretty simple and looking at
Myles Watson [Fri, 1 May 2009 02:24:40 +0000 (02:24 +0000)]
The semantics of the changed code are pretty simple and looking at
other parts of the same file shows that it is a common construct.

Remove the shadowed variable.

Signed-off-by: Myles Watson <mylesgw@gmail.com>
Acked-by: Peter Stuge <peter@stuge.se>
git-svn-id: svn://svn.coreboot.org/coreboot/trunk@4241 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1

15 years agoRemove warnings from compilation of the s2892 with and without CBFS.
Myles Watson [Thu, 30 Apr 2009 22:45:41 +0000 (22:45 +0000)]
Remove warnings from compilation of the s2892 with and without CBFS.

I didn't try to remove "defined but not used" warnings because there are too
many ifdefs to be sure I wouldn't break something.

For shadowed variable declarations I renamed the inner-most variable.

The one in src/pc80/keyboard.c might need help.  I didn't change the
functionality but it looks like a bug.

I boot tested it on s2892 and abuild tested it.

Signed-off-by: Myles Watson <mylesgw@gmail.com>
Acked-by: Ronald G. Minnich <rminnich@gmail.com>
git-svn-id: svn://svn.coreboot.org/coreboot/trunk@4240 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1

15 years agoSome driver fixes for libpayload:
Stefan Reinauer [Thu, 30 Apr 2009 16:46:12 +0000 (16:46 +0000)]
Some driver fixes for libpayload:
- fix minor bug in serial driver.
- latest USB stack fixes
- fix dead store in options.c

Signed-off-by: Stefan Reinauer <stepan@coresystems.de>
Acked-by: Patrick Georgi <patrick.georgi@coresystems.de>
git-svn-id: svn://svn.coreboot.org/coreboot/trunk@4239 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1

15 years agoAdd high tables support to all northbridges.
Stefan Reinauer [Thu, 30 Apr 2009 13:58:42 +0000 (13:58 +0000)]
Add high tables support to all northbridges.

Signed-off-by: Stefan Reinauer <stepan@coresystems.de>
Acked-by: Patrick Georgi <patrick.georgi@coresystems.de>
git-svn-id: svn://svn.coreboot.org/coreboot/trunk@4238 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1

15 years agoTrivial fix for tyan/s2735: a newly used option wasn't defined as "used"
Patrick Georgi [Thu, 30 Apr 2009 12:53:00 +0000 (12:53 +0000)]
Trivial fix for tyan/s2735: a newly used option wasn't defined as "used"

Signed-off-by: Patrick Georgi <patrick.georgi@coresystems.de>
Acked-by: Patrick Georgi <patrick.georgi@coresystems.de>
git-svn-id: svn://svn.coreboot.org/coreboot/trunk@4237 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1

15 years agoAdd "printk" support to all CAR targets
Stefan Reinauer [Thu, 30 Apr 2009 10:16:39 +0000 (10:16 +0000)]
Add "printk" support to all CAR targets

Signed-off-by: Stefan Reinauer <stepan@coresystems.de>
Acked-by: Ronald G. Minnich <rminnich@gmail.com>
git-svn-id: svn://svn.coreboot.org/coreboot/trunk@4236 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1

15 years ago* Use latest version of intel microcodes from their Linux drivers page for
Stefan Reinauer [Thu, 30 Apr 2009 10:14:22 +0000 (10:14 +0000)]
* Use latest version of intel microcodes from their Linux drivers page for
  models 6ex and 6fx (core and core2 solo and duo). Also, use the names suggested
  by Intel for the microcode files instead our short version of it. This allows to
  create new microcode patches with a simple set of scripts.
* some minor cpu setup fixes for c and p states

Signed-off-by: Stefan Reinauer <stepan@coresystems.de>
Acked-by: Ronald G. Minnich <rminnich@gmail.com>
git-svn-id: svn://svn.coreboot.org/coreboot/trunk@4235 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1

15 years agoAnd add the new file I forgot to "svn add" in the last commit.
Patrick Georgi [Thu, 30 Apr 2009 07:23:15 +0000 (07:23 +0000)]
And add the new file I forgot to "svn add" in the last commit.

Signed-off-by: Patrick Georgi <patrick.georgi@coresystems.de>
Acked-by: Patrick Georgi <patrick.georgi@coresystems.de>
git-svn-id: svn://svn.coreboot.org/coreboot/trunk@4234 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1

15 years agoRefactor copy_and_run so that it uses a single code base instead of
Patrick Georgi [Thu, 30 Apr 2009 07:07:22 +0000 (07:07 +0000)]
Refactor copy_and_run so that it uses a single code base instead of
3 (with one of them way too much assembler code).

On the way, I had to make some changes to the way the code is built,
which is an effort I want to expand over time.
Right now, large portions of the in-ROM part of coreboot is compiled as
a single file, with lots of .c files including other .c files.
That has its justification for pre-raminit code, but it also affects
lots of post-raminit code (memcpy doesn't really make sense before
raminit, or at least CAR)

The coreboot_apc code (AMD boards) gained some .c includes because I
don't know that part of the code enough to really rework it and only
have limited possibilities to test it. The includes should give an
identical situation for this part of the code.

This change was posted as set of 6 patches to the list, but they
were mostly split for review purposes, hence commit them all at once.
They can still be backed up using the patch files, if necessary.

Signed-off-by: Patrick Georgi <patrick.georgi@coresystems.de>
Acked-by: Ronald G. Minnich <rminnich@gmail.com>
git-svn-id: svn://svn.coreboot.org/coreboot/trunk@4233 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1

15 years agoincrease image size
Stefan Reinauer [Wed, 29 Apr 2009 20:34:41 +0000 (20:34 +0000)]
increase image size

Signed-off-by: Stefan Reinauer <stepan@coresystems.de>
Acked-by: Stefan Reinauer <stepan@coresystems.de>
git-svn-id: svn://svn.coreboot.org/coreboot/trunk@4232 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1

15 years agoworking memalign version for libpayload. This fixes problems with the USB stack
Stefan Reinauer [Wed, 29 Apr 2009 19:11:18 +0000 (19:11 +0000)]
working memalign version for libpayload. This fixes problems with the USB stack
in FILO.

Signed-off-by: Stefan Reinauer <stepan@coresystems.de>
Acked-by: Patrick Georgi <patrick.georgi@coresystems.de>
git-svn-id: svn://svn.coreboot.org/coreboot/trunk@4231 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1

15 years agoMostly cosmetical changes.
Stefan Reinauer [Wed, 29 Apr 2009 19:10:10 +0000 (19:10 +0000)]
Mostly cosmetical changes.
- #if 0 some incomplete, non-working code instead of failing half way through
  the function
- Don't read a NULL pointer in wclrtoeol

Signed-off-by: Stefan Reinauer <stepan@coresystems.de>
Acked-by: Patrick Georgi <patrick.georgi@coresystems.de>
git-svn-id: svn://svn.coreboot.org/coreboot/trunk@4230 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1

15 years agodon't initialize "in" as it is initialized again one line below.
Stefan Reinauer [Wed, 29 Apr 2009 19:09:19 +0000 (19:09 +0000)]
don't initialize "in" as it is initialized again one line below.

Signed-off-by: Stefan Reinauer <stepan@coresystems.de>
Acked-by: Patrick Georgi <patrick.georgi@coresystems.de>
git-svn-id: svn://svn.coreboot.org/coreboot/trunk@4229 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1

15 years agogetopt for libpayload. in case someone wants to pass parameters to payloads and
Stefan Reinauer [Wed, 29 Apr 2009 19:08:29 +0000 (19:08 +0000)]
getopt for libpayload. in case someone wants to pass parameters to payloads and
parse them.

Signed-off-by: Stefan Reinauer <stepan@coresystems.de>
Acked-by: Patrick Georgi <patrick.georgi@coresystems.de>
git-svn-id: svn://svn.coreboot.org/coreboot/trunk@4228 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1

15 years agodd the family10h Rev C0-C2 support to coreboot.
Vincent Lim vincent.lim [Tue, 28 Apr 2009 16:36:16 +0000 (16:36 +0000)]
dd the family10h Rev C0-C2 support to coreboot.

Signed-off-by: Vincent Lim vincent.lim@amd.com
Acked-by: Patrick Georgi <patrick.georgi@coresystems.de>
git-svn-id: svn://svn.coreboot.org/coreboot/trunk@4227 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1

15 years agofix warning: no return statement in function returning non-void
Stefan Reinauer [Tue, 28 Apr 2009 14:49:50 +0000 (14:49 +0000)]
fix warning: no return statement in function returning non-void

Signed-off-by: Stefan Reinauer <stepan@coresystems.de>
Acked-by: Stefan Reinauer <stepan@coresystems.de>
git-svn-id: svn://svn.coreboot.org/coreboot/trunk@4226 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1

15 years agoOnly add ACPI tables if ACPI is enabled for the board.
Patrick Georgi [Tue, 28 Apr 2009 14:49:21 +0000 (14:49 +0000)]
Only add ACPI tables if ACPI is enabled for the board.
Trivial fix to make abuild happy.

Signed-off-by: Patrick Georgi <patrick.georgi@coresystems.de>
Acked-by: Patrick Georgi <patrick.georgi@coresystems.de>
git-svn-id: svn://svn.coreboot.org/coreboot/trunk@4225 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1

15 years agoEnable HAVE_HIGH_TABLES by default for northbridges with
Patrick Georgi [Tue, 28 Apr 2009 14:18:16 +0000 (14:18 +0000)]
Enable HAVE_HIGH_TABLES by default for northbridges with
support for it.
The related mainboards don't need to activate it
themselves anymore.

Signed-off-by: Patrick Georgi <patrick.georgi@coresystems.de>
Acked-by: Stefan Reinauer <stepan@coresystems.de>
Acked-by: Myles Watson <mylesgw@gmail.com>
git-svn-id: svn://svn.coreboot.org/coreboot/trunk@4224 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1

15 years agoadd_mainboard_resources is necessary for some boards (eg. kontron), but
Patrick Georgi [Tue, 28 Apr 2009 12:57:25 +0000 (12:57 +0000)]
add_mainboard_resources is necessary for some boards (eg. kontron), but
this generic code could be added to the caller of
add_mainboard_resources (wrapped in HAVE_HIGH_TABLES, of course).
That way, boards that really need it (for other things) can use this
function, while others don't have to do anything to use
HAVE_HIGH_TABLES.

Signed-off-by: Patrick Georgi <patrick.georgi@coresystems.de>
Acked-by: Stefan Reinauer <stepan@coresystems.de>
Acked-by: Peter Stuge <peter@stuge.se>
git-svn-id: svn://svn.coreboot.org/coreboot/trunk@4223 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1

15 years agoAdd high table support to via vt8454c.
Patrick Georgi [Mon, 27 Apr 2009 20:19:48 +0000 (20:19 +0000)]
Add high table support to via vt8454c.

Signed-off-by: Patrick Georgi <patrick.georgi@coresystems.de>
Acked-by: Myles Watson <mylesgw@gmail.com>
git-svn-id: svn://svn.coreboot.org/coreboot/trunk@4222 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1

15 years agoAdd high tables support for Supermicro H8DME.
Ward Vandewege [Mon, 27 Apr 2009 20:19:06 +0000 (20:19 +0000)]
Add high tables support for Supermicro H8DME.

Signed-off-by: Ward Vandewege <ward@gnu.org>
Acked-by: Myles Watson <mylesgw@gmail.com>
git-svn-id: svn://svn.coreboot.org/coreboot/trunk@4221 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1

15 years agoSpecial handling for MP table in low memory is only necessary if there are
Ward Vandewege [Mon, 27 Apr 2009 20:00:29 +0000 (20:00 +0000)]
Special handling for MP table in low memory is only necessary if there are
tables in low memory.

This removes a hang when HAVE_LOW_TABLES=0 and HAVE_HIGH_TABLES=1. With this
patch I can boot all the way to a payload. Tested on a Supermicro H8DME.

Many thanks to Patrick Georgi for figuring this out.

Signed-off-by: Ward Vandewege <ward@gnu.org>
Acked-by: Myles Watson <mylesgw@gmail.com>
git-svn-id: svn://svn.coreboot.org/coreboot/trunk@4220 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1

15 years agoMigrate via/vt8454c to CBFS.
Patrick Georgi [Mon, 27 Apr 2009 18:31:09 +0000 (18:31 +0000)]
Migrate via/vt8454c to CBFS.

Signed-off-by: Patrick Georgi <patrick.georgi@coresystems.de>
Acked-by: Myles Watson <mylesgw@gmail.com>
git-svn-id: svn://svn.coreboot.org/coreboot/trunk@4219 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1

15 years agoCreate a valid stack pointer after leaving CAR, so function calls don't
Patrick Georgi [Mon, 27 Apr 2009 18:26:43 +0000 (18:26 +0000)]
Create a valid stack pointer after leaving CAR, so function calls don't
reset the machine in the small window between CAR and coreboot_ram.

Signed-off-by: Patrick Georgi <patrick.georgi@coresystems.de>
Acked-by: Myles Watson <mylesgw@gmail.com>
git-svn-id: svn://svn.coreboot.org/coreboot/trunk@4218 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1

15 years agoRevert 4099 patch that causes an ECC error. Memory has to be written while ECC
Myles Watson [Mon, 27 Apr 2009 16:08:26 +0000 (16:08 +0000)]
Revert 4099 patch that causes an ECC error.  Memory has to be written while ECC
error checking is disabled.  The purpose of the patch was to preserve memory
used by ACPI resume code.  One possible solution is to read that memory and
write it back while ECC error-checking is disabled.

Signed-off-by: Myles Watson <mylesgw@gmail.com>
Acked-by: Myles Watson <mylesgw@gmail.com>
git-svn-id: svn://svn.coreboot.org/coreboot/trunk@4217 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1

15 years agoChange Flashrom's URL to svn://coreboot.org/flashrom/trunk.
Patrick Georgi [Sun, 26 Apr 2009 19:50:53 +0000 (19:50 +0000)]
Change Flashrom's URL to svn://coreboot.org/flashrom/trunk.

Signed-off-by: Patrick Georgi <patrick.georgi@coresystems.de>
Acked-by: Patrick Georgi <patrick.georgi@coresystems.de>
git-svn-id: svn://svn.coreboot.org/coreboot/trunk@4216 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1

15 years agoFlashrom is now moved over to its own repository.
Patrick Georgi [Sun, 26 Apr 2009 18:17:40 +0000 (18:17 +0000)]
Flashrom is now moved over to its own repository.
Add a note to the coreboot-v2 version of the tree that
contains the new location.

Signed-off-by: Patrick Georgi <patrick.georgi@coresystems.de>
Acked-by: Patrick Georgi <patrick.georgi@coresystems.de>
git-svn-id: svn://svn.coreboot.org/coreboot/trunk@4215 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1

15 years agoUse pci_rom, as we have it. Thanks Myles!
Patrick Georgi [Sun, 26 Apr 2009 07:20:45 +0000 (07:20 +0000)]
Use pci_rom, as we have it. Thanks Myles!
trivial patch.

Signed-off-by: Patrick Georgi <patrick.georgi@coresystems.de>
Acked-by: Patrick Georgi <patrick.georgi@coresystems.de>
git-svn-id: svn://svn.coreboot.org/coreboot/trunk@4214 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1

15 years agoTrivial: allow "," in filenames
Patrick Georgi [Sun, 26 Apr 2009 07:15:19 +0000 (07:15 +0000)]
Trivial: allow "," in filenames

Signed-off-by: Patrick Georgi <patrick.georgi@coresystems.de>
Acked-by: Patrick Georgi <patrick.georgi@coresystems.de>
git-svn-id: svn://svn.coreboot.org/coreboot/trunk@4213 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1

15 years agoEnable CBFS for qemu and kontron. Both are builds-and-runs
Patrick Georgi [Sat, 25 Apr 2009 22:15:29 +0000 (22:15 +0000)]
Enable CBFS for qemu and kontron. Both are builds-and-runs
tested, incl. optionrom-in-cbfs for kontron, and compressed payloads
for both.

Signed-off-by: Patrick Georgi <patrick.georgi@coresystems.de>
Acked-by: Myles Watson <mylesgw@gmail.com>
git-svn-id: svn://svn.coreboot.org/coreboot/trunk@4212 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1

15 years agoThe flashrom makefile wants to redirect both stdout and stderr to
Stephan Guilloux [Sat, 25 Apr 2009 22:07:28 +0000 (22:07 +0000)]
The flashrom makefile wants to redirect both stdout and stderr to
/dev/null for one compile test.
The old variant of using &>/dev/null works on bash and zsh, but not on
dash and tcsh. dash and tcsh interpret it as "background command and
truncate /dev/null" which is not what we want. >& works on tcsh and
bash, but it is not POSIX compliant.
Since make uses /bin/sh and /bin/sh has to be POSIX compliant, we can
use the POSIX variant of stderr and stdout redirection.

>/dev/null 2>&1
is POSIX compliant. This is specified in SuSv3, Shell Command Language,
sections 2.7.2 and 2.7.6.

Signed-off-by: Stephan Guilloux <stephan.guilloux@free.fr>
Signed-off-by: Carl-Daniel Hailfinger <c-d.hailfinger.devel.2006@gmx.net>
Acked-by: Ward Vandewege <ward@gnu.org>
git-svn-id: svn://svn.coreboot.org/coreboot/trunk@4211 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1

15 years agoMake the CBFS file lookup skip file data instead of brute-forcing
Patrick Georgi [Sat, 25 Apr 2009 14:39:12 +0000 (14:39 +0000)]
Make the CBFS file lookup skip file data instead of brute-forcing
its way through it, looking for magic numbers.
For one, it should speed up file access, esp. with many entries,
but it also helps against false positives (eg. seabios, which
contains the magic number for its own CBFS support, which _might_
just be aligned properly)

Also avoid infinite loops and give up searching for new files for
invalid magic numbers.

Signed-off-by: Patrick Georgi <patrick.georgi@coresystems.de>
Acked-by: Stefan Reinauer <stepan@coresystems.de>
git-svn-id: svn://svn.coreboot.org/coreboot/trunk@4210 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1

15 years agoThe master cbfs record was located at the end of the flash and overwrote
Myles Watson [Sat, 25 Apr 2009 12:39:04 +0000 (12:39 +0000)]
The master cbfs record was located at the end of the flash and overwrote
anything that was there.  For ck804 or mcp55-based machines that was the
romstrap.

The fix is simple:
1. Put the master cbfs record above the bootblock instead of on it.

Signed-off-by: Myles Watson <mylesgw@gmail.com>
Acked-by: Peter Stuge <peter@stuge.se>
Acked-by: Ronald G. Minnich <rminnich@gmail.com>
git-svn-id: svn://svn.coreboot.org/coreboot/trunk@4209 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1

15 years agoEnable cbfs payload compression (the "l" flag) if payloads are
Patrick Georgi [Sat, 25 Apr 2009 07:33:25 +0000 (07:33 +0000)]
Enable cbfs payload compression (the "l" flag) if payloads are
supposed to be compressed (with lzma only, as cbfstool lacks
nrv2b compression support for now)

Signed-off-by: Patrick Georgi <patrick.georgi@coresystems.de>
Acked-by: Ronald G. Minnich <rminnich@gmail.com>
git-svn-id: svn://svn.coreboot.org/coreboot/trunk@4208 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1

15 years agoMove decompression further down the code flow, so that - where
Patrick Georgi [Sat, 25 Apr 2009 07:32:24 +0000 (07:32 +0000)]
Move decompression further down the code flow, so that - where
possible - code is decompressed directly to the right place
(instead of copying around, as before).

The downside of this approach is that it's not possible (without API
changes to the decompressors) to put partial segments into bounce
buffers. So if a segment collides with coreboot _and_ is compressed,
it's bounced entirely.
But, as this only brings back the copy we already had before, the new
worst case is better than the average before.

It also fixes handling of compressed segments.

Signed-off-by: Patrick Georgi <patrick.georgi@coresystems.de>
Acked-by: Ronald G. Minnich <rminnich@gmail.com>
git-svn-id: svn://svn.coreboot.org/coreboot/trunk@4207 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1

15 years agoRemove the inclusion of lzma.c (which really contains code)
Patrick Georgi [Fri, 24 Apr 2009 16:44:34 +0000 (16:44 +0000)]
Remove the inclusion of lzma.c (which really contains code)
from another (rom_stream.c and others), instead linking it like any
source file should be linked.

The same should (and will) be done with nrv2b.c, but that has some
deeper implications as various CAR implementations include that
directly, and thus requires more care.

It fixes an issue with the cbfs code.

Signed-off-by: Patrick Georgi <patrick.georgi@coresystems.de>
Acked-by: Ronald G. Minnich <rminnich@gmail.com>
git-svn-id: svn://svn.coreboot.org/coreboot/trunk@4206 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1

15 years agoMAX may already be defined. Also, fix smaller cosmetics (trivial).
Uwe Hermann [Fri, 24 Apr 2009 16:17:41 +0000 (16:17 +0000)]
MAX may already be defined. Also, fix smaller cosmetics (trivial).

Signed-off-by: Uwe Hermann <uwe@hermann-uwe.de>
Acked-by: Uwe Hermann <uwe@hermann-uwe.de>
git-svn-id: svn://svn.coreboot.org/coreboot/trunk@4205 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1

15 years agoAnother v3-style #ifdef in v2 to kill
Patrick Georgi [Fri, 24 Apr 2009 10:23:56 +0000 (10:23 +0000)]
Another v3-style #ifdef in v2 to kill

Signed-off-by: Patrick Georgi <patrick.georgi@coresystems.de>
Acked-by: Peter Stuge <peter@stuge.se>
git-svn-id: svn://svn.coreboot.org/coreboot/trunk@4204 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1

15 years agoThese are some really horrible bugs that got through.
Ronald G. Minnich [Fri, 24 Apr 2009 06:32:29 +0000 (06:32 +0000)]
These are some really horrible bugs that got through.

(and, for the record: no more #ifdef in coreboot. We're not going to
have this happen again. If we do have it in v2, let's remove it.)

Signed-off-by: Ronald G. Minnich <rminnich@gmail.com>
Acked-by: Myles Watson <mylesgw@gmail.com>
git-svn-id: svn://svn.coreboot.org/coreboot/trunk@4203 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1

15 years agoChange the behaviour of the ACPI generating code so it only
Patrick Georgi [Fri, 24 Apr 2009 06:28:48 +0000 (06:28 +0000)]
Change the behaviour of the ACPI generating code so it only
writes at most one full ACPI table.
In the cases where both HAVE_LOW_TABLES and HAVE_HIGH_TABLES
are enabled, the table is written to high memory, and an RSDP
is written to the low memory that points to the high mem one.
All other cases work exactly the same way as before.

Signed-off-by: Patrick Georgi <patrick.georgi@coresystems.de>
Acked-by: Myles Watson <mylesgw@gmail.com>
git-svn-id: svn://svn.coreboot.org/coreboot/trunk@4202 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1

15 years agoRemove duplicate code.
Patrick Georgi [Fri, 24 Apr 2009 06:27:31 +0000 (06:27 +0000)]
Remove duplicate code.

Signed-off-by: Patrick Georgi <patrick.georgi@coresystems.de>
Acked-by: Myles Watson <mylesgw@gmail.com>
git-svn-id: svn://svn.coreboot.org/coreboot/trunk@4201 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1

15 years agoflashrom: Support MX25L3235D
Stephan Guilloux [Thu, 23 Apr 2009 22:51:56 +0000 (22:51 +0000)]
flashrom: Support MX25L3235D

Signed-off-by: Stephan Guilloux <stephan.guilloux@free.fr>
Acked-by: Peter Stuge <peter@stuge.se>
git-svn-id: svn://svn.coreboot.org/coreboot/trunk@4200 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1

15 years agoAdd 'install' target for ectool (trivial).
Uwe Hermann [Thu, 23 Apr 2009 22:22:47 +0000 (22:22 +0000)]
Add 'install' target for ectool (trivial).

Signed-off-by: Uwe Hermann <uwe@hermann-uwe.de>
Acked-by: Uwe Hermann <uwe@hermann-uwe.de>
git-svn-id: svn://svn.coreboot.org/coreboot/trunk@4199 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1

15 years agoFix an uninitialized variable. If it didn't end up being zero it sometimes
Myles Watson [Thu, 23 Apr 2009 18:46:32 +0000 (18:46 +0000)]
Fix an uninitialized variable.  If it didn't end up being zero it sometimes
caused a seg fault, sometimes executed somewhere else.  Also add an error if
the algorithm is unknown.

Signed-off-by: Myles Watson <mylesgw@gmail.com>
Acked-by: Ward Vandewege <ward@gnu.org>
git-svn-id: svn://svn.coreboot.org/coreboot/trunk@4198 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1

15 years agoThis patch cleans up Makefile generation. It removes the
Myles Watson [Thu, 23 Apr 2009 17:01:37 +0000 (17:01 +0000)]
This patch cleans up Makefile generation.  It removes the
coreboot.romfs file since CBFS will eventually be the standard.

Signed-off-by: Myles Watson <mylesgw@gmail.com>
Acked-by: Patrick Georgi <patrick.georgi@coresystems.de>
git-svn-id: svn://svn.coreboot.org/coreboot/trunk@4197 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1

15 years agoDon't duplicate option description in README, the manpage already has
Uwe Hermann [Thu, 23 Apr 2009 14:57:55 +0000 (14:57 +0000)]
Don't duplicate option description in README, the manpage already has
that info. Also, additional small cosmetic fix.

Signed-off-by: Uwe Hermann <uwe@hermann-uwe.de>
Acked-by: Uwe Hermann <uwe@hermann-uwe.de>
git-svn-id: svn://svn.coreboot.org/coreboot/trunk@4196 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1

15 years agoAdd VERSION to hp/dl145_g3. Trivial.
Myles Watson [Thu, 23 Apr 2009 13:41:12 +0000 (13:41 +0000)]
Add VERSION to hp/dl145_g3. Trivial.

Signed-off-by: Myles Watson <mylesgw@gmail.com>
Acked-by: Myles Watson <mylesgw@gmail.com>
git-svn-id: svn://svn.coreboot.org/coreboot/trunk@4195 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1

15 years agoThis patch hooks up git mirrored svn revisions and adds some
Carl-Daniel Hailfinger [Thu, 23 Apr 2009 13:05:45 +0000 (13:05 +0000)]
This patch hooks up git mirrored svn revisions and adds some
error checking to the svnrevision call.

If a .svn directory exists in the top level directory and the svnversion
utility is available, we use svnversion.
Otherwise, if a .git directory exists in the top level directory and the
git utility is available, we use git log.

Signed-off-by: Carl-Daniel Hailfinger <c-d.hailfinger.devel.2006@gmx.net>
Acked-by: Myles Watson <mylesgw@gmail.com>
git-svn-id: svn://svn.coreboot.org/coreboot/trunk@4194 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1

15 years agoThis continues the doco attempt.
Ronald G. Minnich [Thu, 23 Apr 2009 03:59:33 +0000 (03:59 +0000)]
This continues the doco attempt.

This also mentions some ideas on the new booting setup for v2.

The latest changes will remove all need for people to do math.

With Peter's corrections as well.

Signed-off-by: Ronald G. Minnich <rminnich@gmail.com>
Acked-by: Marc Jones <marcj303@gmail.com>
git-svn-id: svn://svn.coreboot.org/coreboot/trunk@4193 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1

15 years agoInstead of just
Stefan Reinauer [Wed, 22 Apr 2009 23:39:19 +0000 (23:39 +0000)]
Instead of just

  coreboot-v2 $ util/abuild/abuild -t kontron/986lcd-m $PWD

you can now also say

  coreboot-v2 $ util/abuild/abuild -t kontron/986lcd-m/Config-myconf.lb $PWD

and instead of using Config-abuild.lb or creating a temporary Config-abuild.lb,
abuild will use the existing Config-myconf.lb to build your image.

Signed-off-by: Stefan Reinauer <stepan@coresystems.de>
Acked-by: Stefan Reinauer <stepan@coresystems.de>
git-svn-id: svn://svn.coreboot.org/coreboot/trunk@4192 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1

15 years agoSometimes when we debug the code, we need to know which version we are
Zheng Bao [Wed, 22 Apr 2009 23:27:25 +0000 (23:27 +0000)]
Sometimes when we debug the code, we need to know which version we are
working on.
Add the svn revision to the coreboot version string.

Signed-off-by: Zheng Bao <zheng.bao@amd.com>
Acked-by: Myles Watson <mylesgw@gmail.com>
Acked-by: Carl-Daniel Hailfinger <c-d.hailfinger.devel.2006@gmx.net>
git-svn-id: svn://svn.coreboot.org/coreboot/trunk@4191 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1

15 years agomini fix to reliably compile inteltool on darwin, and on Linux both on x86/x86_64.
Stefan Reinauer [Wed, 22 Apr 2009 23:17:44 +0000 (23:17 +0000)]
mini fix to reliably compile inteltool on darwin, and on Linux both on x86/x86_64.
Signed-off-by: Stefan Reinauer <stepan@coresystems.de>
Acked-by: Stefan Reinauer <stepan@coresystems.de>
git-svn-id: svn://svn.coreboot.org/coreboot/trunk@4190 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1

15 years agomakes the smi handler a little bit less verbose
Stefan Reinauer [Wed, 22 Apr 2009 22:55:15 +0000 (22:55 +0000)]
makes the smi handler a little bit less verbose

Signed-off-by: Stefan Reinauer <stepan@coresystems.de>
Acked-by: Stefan Reinauer <stepan@coresystems.de>
git-svn-id: svn://svn.coreboot.org/coreboot/trunk@4189 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1