coreboot.git
13 years agoFollow-up for r6025, do 0x87 twice in superio.c, too.
Uwe Hermann [Fri, 5 Nov 2010 12:44:25 +0000 (12:44 +0000)]
Follow-up for r6025, do 0x87 twice in superio.c, too.

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@6026 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1

13 years agoAccording to the description in datasheet of f71889,
Zheng Bao [Fri, 5 Nov 2010 07:59:06 +0000 (07:59 +0000)]
According to the description in datasheet of f71889,

"To enable configuration, the entry key 0x87 must be written to
the index port"

"
 -o 4e 87
 -o 4e 87 (enable configuration)
 -o 4e aa (disable configuration)
"
This piece of text appears in most of the datasheet of fintek superio.
It doesnt say it quite clear, but it seems that the 0x87 should
be written twice. I tried on f81865, which is not in the coreboot tree
yet. If the 0x87 is only written once, you can only R/W the index/data
port once. All the subsequent RW will fail. Writing twice will be ok.

Plus, in the superiotool, the function enter_conf_mode_winbond_fintek_ite_8787
also write 8787.

The fintek superio chips seem to enable the UART automatically when the
power is on. So I didnt find it failed to access.

Signed-off-by: Zheng Bao <zheng.bao@amd.com>
Acked-by: Stefan Reinauer <stepan@coresystems.de>
git-svn-id: svn://svn.coreboot.org/coreboot/trunk@6025 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1

13 years agoAdd detection support for the Fintek F81865/F81865-I.
Zheng Bao [Fri, 5 Nov 2010 00:34:12 +0000 (00:34 +0000)]
Add detection support for the Fintek F81865/F81865-I.

Signed-off-by: Zheng Bao <zheng.bao@amd.com>
Acked-by: Uwe Hermann <uwe@hermann-uwe.de>
git-svn-id: svn://svn.coreboot.org/coreboot/trunk@6024 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1

13 years agoAdd Kconfig CPU speed selection to Geode GX2 boards.
Nils Jacobs [Fri, 5 Nov 2010 00:23:11 +0000 (00:23 +0000)]
Add Kconfig CPU speed selection to Geode GX2 boards.

This is Abuild and boot tested.

Signed-off-by: Nils Jacobs <njacobs8@hetnet.nl>
Acked-by: Uwe Hermann <uwe@hermann-uwe.de>
git-svn-id: svn://svn.coreboot.org/coreboot/trunk@6023 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1

13 years agoGX2: Define the unused DIMM1 to 0xFF to make it obvious it is a bogus value.
Nils Jacobs [Fri, 5 Nov 2010 00:19:21 +0000 (00:19 +0000)]
GX2: Define the unused DIMM1 to 0xFF to make it obvious it is a bogus value.

This is Abuild and boot tested.

Signed-off-by: Nils Jacobs <njacobs8@hetnet.nl>
Acked-by: Uwe Hermann <uwe@hermann-uwe.de>
git-svn-id: svn://svn.coreboot.org/coreboot/trunk@6022 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1

13 years agoRemove banner wrapper function and unify print(k) usage.
Nils Jacobs [Fri, 5 Nov 2010 00:13:14 +0000 (00:13 +0000)]
Remove banner wrapper function and unify print(k) usage.

 - Drop banner(), use printk()s instead.

 - Uncomment a few printk()s, if a users doesn't want to see them he/she
   can lower the debug level.

 - Replace print_emerg() with printk(BIOS_EMERG) etc.

Also change 'Assymetirc' into 'Asymmetric', thanks to Idwer for spotting.

This is Abuild and boot tested.

Signed-off-by: Nils Jacobs <njacobs8@hetnet.nl>
Acked-by: Uwe Hermann <uwe@hermann-uwe.de>
git-svn-id: svn://svn.coreboot.org/coreboot/trunk@6021 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1

13 years agoFintek and Intel i3100 Super I/O cleanups.
Uwe Hermann [Fri, 5 Nov 2010 00:07:13 +0000 (00:07 +0000)]
Fintek and Intel i3100 Super I/O cleanups.

 - Drop commented out "config chip.h" and a duplicate link to a datasheet.

 - F71805F -> F71805F/FG, to mention all variants.

 - Use u8/u16/ etc. everywhere.

 - Add a missing (C) line.

 - Fix up a bunch of pnp_dev_info[] structs according to the datasheets.

 - Fintek F71889: Drop res1/PNP_IO1 from KBC, there's no 0x62/0x63 register
   pair on this Super I/O.

 - Fintek F71863FG: This Super I/O _does_ have a keyboard/mouse LDN, add the
   respective code in superio.c. Also: Add missing LDNs to f71863fg.h.

 - i3100: Add some more comments and datasheet infos.

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@6020 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1

13 years agoVarious cosmetic and coding style fixes in src/devices.
Uwe Hermann [Thu, 4 Nov 2010 23:23:47 +0000 (23:23 +0000)]
Various cosmetic and coding style fixes in src/devices.

Also:

 - Improve a few code comments, fix typos, etc.

 - Change a few more variable types to u8/u16/u32 etc.

 - Make some very long lines fit into 80chars/line.

 - Drop a huge duplicated comment, use "@see" to refer to the other one.

 - Reduce nesting level a bit by restructuring some code chunks.

 - s/Config.lb/devicetree.cb/ in a few places.

Abuild-tested.

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@6019 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1

13 years agoAdd a rom_enable() function to via/vt8231 and call it from via/epia/romstage.c
Uwe Hermann [Thu, 4 Nov 2010 18:33:42 +0000 (18:33 +0000)]
Add a rom_enable() function to via/vt8231 and call it from via/epia/romstage.c

Signed-off-by: Uwe Hermann <uwe@hermann-uwe.de>
Signed-off-by: Myles Watson <mylesgw@gmail.com>
Acked-by: Uwe Hermann <uwe@hermann-uwe.de>
git-svn-id: svn://svn.coreboot.org/coreboot/trunk@6018 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1

13 years agoAdd Fintek F71889 Super I/O support.
Alec Ari [Wed, 3 Nov 2010 21:46:41 +0000 (21:46 +0000)]
Add Fintek F71889 Super I/O support.

Untested, but should work mostly (even though some TODOs remain).

Signed-off-by: Alec Ari <neotheuser@ymail.com>
Acked-by: Uwe Hermann <uwe@hermann-uwe.de>
git-svn-id: svn://svn.coreboot.org/coreboot/trunk@6017 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1

13 years agoRemove some unused code from gx2/raminit.c.
Nils Jacobs [Wed, 3 Nov 2010 13:24:29 +0000 (13:24 +0000)]
Remove some unused code from gx2/raminit.c.

This is Abuild and boot tested.

Signed-off-by: Nils Jacobs <njacobs8@hetnet.nl>
Acked-by: Uwe Hermann <uwe@hermann-uwe.de>
git-svn-id: svn://svn.coreboot.org/coreboot/trunk@6016 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1

13 years agoClean up some more comments and white space in model_gx2/cpureginit.c.
Nils Jacobs [Wed, 3 Nov 2010 13:21:41 +0000 (13:21 +0000)]
Clean up some more comments and white space in model_gx2/cpureginit.c.

This is Abuild and boot tested.

Signed-off-by: Nils Jacobs <njacobs8@hetnet.nl>
Acked-by: Uwe Hermann <uwe@hermann-uwe.de>
git-svn-id: svn://svn.coreboot.org/coreboot/trunk@6015 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1

13 years agoClean up some comments and white space in gx2/northbridgeinit.c
Nils Jacobs [Wed, 3 Nov 2010 13:19:50 +0000 (13:19 +0000)]
Clean up some comments and white space in gx2/northbridgeinit.c
and gx2/raminit.c.

This is Abuild and boot tested.

Signed-off-by: Nils Jacobs <njacobs8@hetnet.nl>
Acked-by: Uwe Hermann <uwe@hermann-uwe.de>
git-svn-id: svn://svn.coreboot.org/coreboot/trunk@6014 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1

13 years agoNeed to clear downstream read cycle retry bit, or the bus scan will
Tobias Diedrich [Tue, 2 Nov 2010 21:24:29 +0000 (21:24 +0000)]
Need to clear downstream read cycle retry bit, or the bus scan will
hang.  Also need to set lane config to 0x00 for autonegotiation.

Signed-off-by: Tobias Diedrich <ranma+coreboot@tdiedrich.de>
Acked-by: Rudolf Marek <r.marek@assembler.cz>
git-svn-id: svn://svn.coreboot.org/coreboot/trunk@6013 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1

13 years agoThis adds pci device ids and pci_driver structs for the K8T890 CF
Tobias Diedrich [Tue, 2 Nov 2010 20:54:37 +0000 (20:54 +0000)]
This adds pci device ids and pci_driver structs for the K8T890 CF
variant. It also adds additional dev_find_device calls in k8t890_ctrl.c for
subfunctions 3 and 7.

Signed-off-by: Tobias Diedrich <ranma+coreboot@tdiedrich.de>
Acked-by: Rudolf Marek <r.marek@assembler.cz>
git-svn-id: svn://svn.coreboot.org/coreboot/trunk@6012 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1

13 years agoChange Geode GX2 to use the auto DRAM detect code from Geode LX.
Nils Jacobs [Mon, 1 Nov 2010 15:20:27 +0000 (15:20 +0000)]
Change Geode GX2 to use the auto DRAM detect code from Geode LX.

Also, change the GX2 boards to use it.

Add a processor speed setting function in human readable MHz and remove
the useless and broken PLLMSR settings (the processor speed was hardcoded
to 366MHz in pll_reset.c).

Signed-off-by: Nils Jacobs <njacobs8@hetnet.nl>
Acked-by: Uwe Hermann <uwe@hermann-uwe.de>
git-svn-id: svn://svn.coreboot.org/coreboot/trunk@6011 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1

13 years agoRemove some unused code.
Nils Jacobs [Mon, 1 Nov 2010 14:39:49 +0000 (14:39 +0000)]
Remove some unused code.

Signed-off-by: Nils Jacobs <njacobs8@hetnet.nl>
Acked-by: Uwe Hermann <uwe@hermann-uwe.de>
git-svn-id: svn://svn.coreboot.org/coreboot/trunk@6010 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1

13 years agoGX2: Clean up some white space and comments.
Nils Jacobs [Mon, 1 Nov 2010 14:36:54 +0000 (14:36 +0000)]
GX2: Clean up some white space and comments.

Also, add a copyright header to pll_reset.c.

Signed-off-by: Nils Jacobs <njacobs8@hetnet.nl>
Acked-by: Uwe Hermann <uwe@hermann-uwe.de>
git-svn-id: svn://svn.coreboot.org/coreboot/trunk@6009 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1

13 years agoGX2: Change MSR register numbers into more descriptive names.
Nils Jacobs [Mon, 1 Nov 2010 14:18:11 +0000 (14:18 +0000)]
GX2: Change MSR register numbers into more descriptive names.

Signed-off-by: Nils Jacobs <njacobs8@hetnet.nl>
Acked-by: Uwe Hermann <uwe@hermann-uwe.de>
git-svn-id: svn://svn.coreboot.org/coreboot/trunk@6008 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1

13 years agoRemove definitions of ACPI_SSDTX_NUM to 0, that's the default anyway.
Uwe Hermann [Sun, 31 Oct 2010 23:24:18 +0000 (23:24 +0000)]
Remove definitions of ACPI_SSDTX_NUM to 0, that's the default anyway.

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@6007 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1

13 years agoRemove incorrect IOAPIC lines from some mptable.c files.
Uwe Hermann [Sun, 31 Oct 2010 19:37:50 +0000 (19:37 +0000)]
Remove incorrect IOAPIC lines from some mptable.c files.

 - via/epia-n/mptable.c
 - intel/eagleheights/mptable.c (commented out anyway)
 - asus/p2b-d/mptable.c
 - asus/p2b-ds/mptable.c

Some files still incorrectly contain some smp_write_ioapic() lines from
the original mptable utility target (Supermicro P4DPE), which has one
IOAPIC in the southbridge (Intel ICH3-S), two IOAPICs contained in
the first P64H2, and two more in the second P64H2, i.e. 5 IOAPICs in total.

However, none of the boards where this chunk of code is present has
multiple IOAPICs (and even if they had, the PCI devices where those are
located would probably be different anyway), so drop the incorrect
mptable.c contents.

Also drop the lines from the mptable utility, so that future mptable.c files
don't incorrectly inherit any of this stuff.

Signed-off-by: Uwe Hermann <uwe@hermann-uwe.de>
Acked-by: Jonathan Kollasch <jakllsch@kollasch.net>
git-svn-id: svn://svn.coreboot.org/coreboot/trunk@6006 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1

13 years agoFix AMD family 10h engineering sample is reported as 'thermal test kit'.
Scott Duplichan [Sun, 31 Oct 2010 14:17:51 +0000 (14:17 +0000)]
Fix AMD family 10h engineering sample is reported as 'thermal test kit'.

Signed-off-by: Scott Duplichan <scott@notabs.org>
Acked-by: Peter Stuge <peter@stuge.se>
git-svn-id: svn://svn.coreboot.org/coreboot/trunk@6005 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1

13 years agoMptable related fixes for ASUS P2B-DS.
Uwe Hermann [Sat, 30 Oct 2010 21:27:13 +0000 (21:27 +0000)]
Mptable related fixes for ASUS P2B-DS.

 - Add "select IOAPIC" in the board's Kconfig file.

 - Set CONFIG_MAX_PHYSICAL_CPUS to 2 on the board. There are two
   CPU sockets (Slot 1) and each CPU can only have one core, multi-core CPUs
   didn't exist in that era (CONFIG_MAX_CPUS was set to 2 already).

 - Drop useless/duplicated enable_lapic() call from ASUS P2B-DS's romstage.c,
   that function is always called if either CONFIG_SMP and/or CONFIG_IOAPIC
   are set.

 - Rework ASUS P2B-DS mptable.c to fix a number of things:

   - Convert it to use mptable_write_buses() as all mptable.c files should do.

   - Fix incorrect IOAPICID (it's 0x11 for the external 82093AA IOAPIC).

   - Fix a bunch of hardcoded bus IDs, remove incorrect entries, etc.

Signed-off-by: Uwe Hermann <uwe@hermann-uwe.de>
Acked-by: Peter Stuge <peter@stuge.se>
git-svn-id: svn://svn.coreboot.org/coreboot/trunk@6004 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1

13 years agoUse common code to set PCI subsystem in mcp55.
Jonathan Kollasch [Fri, 29 Oct 2010 20:40:06 +0000 (20:40 +0000)]
Use common code to set PCI subsystem in mcp55.

Signed-off-by: Jonathan Kollasch <jakllsch@kollasch.net>
Acked-by: Uwe Hermann <uwe@hermann-uwe.de>
git-svn-id: svn://svn.coreboot.org/coreboot/trunk@6003 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1

13 years agoDeduplicate ck804 subsystem-setting functionality.
Jonathan Kollasch [Fri, 29 Oct 2010 15:56:04 +0000 (15:56 +0000)]
Deduplicate ck804 subsystem-setting functionality.

Signed-off-by: Jonathan Kollasch <jakllsch@kollasch.net>
Acked-by: Peter Stuge <peter@stuge.se>
git-svn-id: svn://svn.coreboot.org/coreboot/trunk@6002 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1

13 years agoDrop duplicate HAVE_ACPI_TABLES (trivial).
Uwe Hermann [Fri, 29 Oct 2010 14:26:27 +0000 (14:26 +0000)]
Drop duplicate HAVE_ACPI_TABLES (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@6001 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1

13 years agoThe no point in having a non-NULL ops_pci pointer when the set_subsystem operation...
Jonathan Kollasch [Thu, 28 Oct 2010 19:57:52 +0000 (19:57 +0000)]
The no point in having a non-NULL ops_pci pointer when the set_subsystem operation within is NULL anyway.

Signed-off-by: Jonathan Kollasch <jakllsch@kollasch.net>
Acked-by: Jonathan Kollasch <jakllsch@kollasch.net>
git-svn-id: svn://svn.coreboot.org/coreboot/trunk@6000 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1

13 years agoFix broken build due to missing #if CONFIG_IOAPIC.
Uwe Hermann [Thu, 28 Oct 2010 14:22:20 +0000 (14:22 +0000)]
Fix broken build due to missing #if CONFIG_IOAPIC.

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@5999 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1

13 years agoAdd IOAPIC support for Intel 82371EB and fixup SMP on ASUS P2B-D.
Uwe Hermann [Thu, 28 Oct 2010 08:19:22 +0000 (08:19 +0000)]
Add IOAPIC support for Intel 82371EB and fixup SMP on ASUS P2B-D.

 - Add enable_intel_82093aa_ioapic() which enables IOAPIC usage in the
   Intel 82371EB southbridge (sets the proper chip-select) and sets an
   IOAPIC ID.

 - We only call enable_intel_82093aa_ioapic() if a board does "select IOAPIC"
   as on 82371EB-based boards the IOAPIC is an external chip (not integrated
   in the southbridge) and it's only populated on multi-CPU boards.
   That is, we cannot unconditionally enable it, only on SMP-capable boards.

 - Due to the reason explained above, remove "select IOAPIC" from
   src/southbridge/intel/i82371eb/Kconfig, and add it to
   src/mainboard/asus/p2b-d/Kconfig.

 - Also set CONFIG_MAX_PHYSICAL_CPUS to 2 on ASUS P2B-D. There are two
   CPU sockets (Slot 1) and each CPU can only have one core, multi-core CPUs
   didn't exist in that era (CONFIG_MAX_CPUS was set to 2 already).

 - Drop useless/duplicated enable_lapic() call from ASUS P2B-D's romstage.c,
   that function is always called if either CONFIG_SMP and/or CONFIG_IOAPIC
   are set.

 - Rework ASUS P2B-D mptable.c to fix a number of things:

   - Convert it to use mptable_write_buses() as all mptable.c files should do.

   - Fix incorrect IOAPICID (it's 0x11 for the external 82093AA IOAPIC).

   - Fix a bunch of hardcoded bus IDs, remove incorrect entries, etc.

This is build-tested on ASUS P2B-D, and also boot-tested successfully there.
On Linux I now get two entries in /proc/cpuinfo (where only one appeared
before this patch), i.e. both populated CPUs are found.

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@5998 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1

13 years agoEnable CK804 AC'97 audio interface and explicitly enable NIC on A8N-E.
Jonathan Kollasch [Wed, 27 Oct 2010 20:32:49 +0000 (20:32 +0000)]
Enable CK804 AC'97 audio interface and explicitly enable NIC on A8N-E.

Signed-off-by: Jonathan Kollasch <jakllsch@kollasch.net>
Acked-by: Peter Stuge <peter@stuge.se>
git-svn-id: svn://svn.coreboot.org/coreboot/trunk@5997 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1

13 years agoCorrect an apparent copy-paste error that shows up at compile time on
Jonathan Kollasch [Wed, 27 Oct 2010 20:30:32 +0000 (20:30 +0000)]
Correct an apparent copy-paste error that shows up at compile time on
boards using ck804_early_setup.c that select CK804_USE_NIC.

Signed-off-by: Jonathan Kollasch <jakllsch@kollasch.net>
Acked-by: Peter Stuge <peter@stuge.se>
git-svn-id: svn://svn.coreboot.org/coreboot/trunk@5996 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1

13 years agoDrop referenced-yet-does-nothing static function from ms7135 romstage.
Jonathan Kollasch [Wed, 27 Oct 2010 17:41:40 +0000 (17:41 +0000)]
Drop referenced-yet-does-nothing static function from ms7135 romstage.

Signed-off-by: Jonathan Kollasch <jakllsch@kollasch.net>
Acked-by: Jonathan Kollasch <jakllsch@kollasch.net>
git-svn-id: svn://svn.coreboot.org/coreboot/trunk@5995 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1

13 years agoConvert ck804_early_smbus.c to a separately compiled unit.
Jonathan Kollasch [Wed, 27 Oct 2010 17:26:57 +0000 (17:26 +0000)]
Convert ck804_early_smbus.c to a separately compiled unit.
Additionally, make the second SMBus more accessible in romstage.

Signed-off-by: Jonathan Kollasch <jakllsch@kollasch.net>
Acked-by: Peter Stuge <peter@stuge.se>
git-svn-id: svn://svn.coreboot.org/coreboot/trunk@5994 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1

13 years agoConvert some comments to proper Doxygen syntax.
Uwe Hermann [Tue, 26 Oct 2010 22:46:43 +0000 (22:46 +0000)]
Convert some comments to proper Doxygen syntax.

Also, make them all fit in 80chars/column, fix some whitespace issues
and also some typos I noticed.

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@5993 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1

13 years agoWe need to call smp_write_lintsrc() instead of smp_write_intsrc() for
Tobias Diedrich [Tue, 26 Oct 2010 22:40:16 +0000 (22:40 +0000)]
We need to call smp_write_lintsrc() instead of smp_write_intsrc() for
local ints. This is wrong in most coreboot mptables, probably all
generated by util/mptable/mptable.c.

After fixing this now XP can boot in MPS mode on my M2V.

Signed-off-by: Tobias Diedrich <ranma+coreboot@tdiedrich.de>
Acked-by: Uwe Hermann <uwe@hermann-uwe.de>
git-svn-id: svn://svn.coreboot.org/coreboot/trunk@5992 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1

13 years agoConvert all ck804-based boards to tiny bootblock.
Jonathan Kollasch [Tue, 26 Oct 2010 16:10:20 +0000 (16:10 +0000)]
Convert all ck804-based boards to tiny bootblock.

Signed-off-by: Jonathan Kollasch <jakllsch@kollasch.net>
Acked-by: Patrick Georgi <patrick.georgi@coresystems.de>
git-svn-id: svn://svn.coreboot.org/coreboot/trunk@5991 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1

13 years agoMove bcm5785_enable_rom.c include to where it's used.
Patrick Georgi [Tue, 26 Oct 2010 15:51:57 +0000 (15:51 +0000)]
Move bcm5785_enable_rom.c include to where it's used.
Right now, it breaks the build of bootblock enabled boards
with that chipset.

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@5990 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1

13 years agoreg is only used inside the #if clause, so declare it there. trivial.
Patrick Georgi [Tue, 26 Oct 2010 15:11:45 +0000 (15:11 +0000)]
reg is only used inside the #if clause, so declare it there. trivial.

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@5989 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1

13 years agoWhen gcc 4.5.0 is used, compiling mcp55_early_setup_car.c fails. This change eliminat...
Scott Duplichan [Tue, 26 Oct 2010 05:26:01 +0000 (05:26 +0000)]
When gcc 4.5.0 is used, compiling mcp55_early_setup_car.c fails. This change eliminates the compiler warning that causes the build to fail.

Signed-off-by: Scott Duplichan <scott@notabs.org>
Acked-by: Peter Stuge <peter@stuge.se>
git-svn-id: svn://svn.coreboot.org/coreboot/trunk@5988 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1

13 years agoFactor out common mptable code to mptable_init().
Uwe Hermann [Mon, 25 Oct 2010 15:32:07 +0000 (15:32 +0000)]
Factor out common mptable code to mptable_init().

 - Drop sig[], oem[], and productid[] fields in all mptable.c files, no
   longer needed. The sig[] is always the same ("PCMP"), the oem[] is
   currently also always the same ("COREBOOT"), and productid is being
   passed into mptable_init() directly as string now.

 - LAPIC_ADDR is passed in as parameter, too. While at the moment it's
   always the same value that is passed in, the LAPIC base address could
   also be relocated theoretically, so keep it as parameter for now.

 - Fix a few productid entries, they were (partially) incorrect:

   - DK8S2 (was "DK8X", copypaste)
   - 939A785GMH (was "MAHOGANY", copypaste)
   - X6DHE-G (was "X6DHE", incomplete board name)
   - H8DME-2 (was "H8DMR", copypaste)
   - H8QME-2+ (was "H8QME", incomplete board name)
   - X6DHE-G2 (was "X6DHE", incomplete board name)
   - X6DHR-iG2 (was "X6DHR-iG", incomplete board name)
   - GA-M57SLI-S4 (was "M57SLI", incomplete board name)
   - KINO-780AM2 (was "KINO", incomplete board name)
   - DL145 G1 (was "DL145G1", small fix as per vendor website)
   - DL145 G3 (was "TREX", wrong board name)
   - DL165 G6 (was "HP DL165 G6", drop vendor)
   - S2912 (was "S2895", copypaste)
   - VT8454c (was "VIA VT8454C", drop vendor, lower-case "c")
   - EPIA-N (was "P4DPE", copypaste)
   - pc2500e (was "PC2500", incorrect name)
   - S1850 (was "S2850", copy-paste)
   - MS-7135 (was "MS7135")
   - MS-9282 (was "MS9282")
   - MS-9185 (was "MS9185")
   - MS-9652 (was "K9ND MS-9652")
   - Ultra 40 (was "ultra40")
   - E326 (was "E325", copypaste)
   - M4A785-M (was "TILAPIA", copypaste)
   - P2B-D (was "ASUS P2B-D", drop vendor)
   - P2B-DS (was "ASUS P2B-DS", drop vendor)

 - Adapt the mptable utility to use mptable_init() too.

Abuild-tested.

Signed-off-by: Uwe Hermann <uwe@hermann-uwe.de>
Acked-by: Patrick Georgi <patrick.georgi@coresystems.de>
git-svn-id: svn://svn.coreboot.org/coreboot/trunk@5987 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1

13 years agoMake ectool -i work; add missing break
Peter Stuge [Mon, 25 Oct 2010 02:12:04 +0000 (02:12 +0000)]
Make ectool -i work; add missing break

Signed-off-by: Peter Stuge <peter@stuge.se>
Acked-by: Peter Stuge <peter@stuge.se>
git-svn-id: svn://svn.coreboot.org/coreboot/trunk@5986 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1

13 years agoRunning a checked build of Windows is needed for understanding its various BIOS relat...
Scott Duplichan [Sun, 24 Oct 2010 16:22:11 +0000 (16:22 +0000)]
Running a checked build of Windows is needed for understanding its various BIOS related BSODs. Win7 checked build complains when running coreboot+seabios:

 FADT revision inconsistent with length.
     Revision:        0x1
     Length:          0xf4
     Expected Length: 0x74

Change the FADT revision from 1 to 3 to match its length and prevent the Windows checked build assert.

Signed-off-by: Scott Duplichan <scott@notabs.org>
Acked-by: Peter Stuge <peter@stuge.se>
git-svn-id: svn://svn.coreboot.org/coreboot/trunk@5985 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1

13 years agoAdd small comment about LDN 5 on F71872F/FG / F71806F/FG.
Uwe Hermann [Sun, 24 Oct 2010 14:19:09 +0000 (14:19 +0000)]
Add small comment about LDN 5 on F71872F/FG / F71806F/FG.

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@5984 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1

13 years agoFix superiotool build on non-NetBSD x86_64.
Jonathan Kollasch [Sun, 24 Oct 2010 14:18:55 +0000 (14:18 +0000)]
Fix superiotool build on non-NetBSD x86_64.

Signed-off-by: Jonathan Kollasch <jakllsch@kollasch.net>
Acked-by: Jonathan Kollasch <jakllsch@kollasch.net>
git-svn-id: svn://svn.coreboot.org/coreboot/trunk@5983 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1

13 years agoProvide for I/O space access on NetBSD.
Jonathan Kollasch [Sun, 24 Oct 2010 14:10:35 +0000 (14:10 +0000)]
Provide for I/O space access on NetBSD.

Signed-off-by: Jonathan Kollasch <jakllsch@kollasch.net>
Acked-by: Peter Stuge <peter@stuge.se>
git-svn-id: svn://svn.coreboot.org/coreboot/trunk@5982 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1

13 years agoAdd inteltool support for FreeBSD.
Idwer Vollering [Sun, 24 Oct 2010 13:50:13 +0000 (13:50 +0000)]
Add inteltool support for FreeBSD.

Signed-off-by: Idwer Vollering <vidwer@gmail.com>
Acked-by Stefan Reinauer <stepan@coreboot.org>
Acked-by: Uwe Hermann <uwe@hermann-uwe.de>
git-svn-id: svn://svn.coreboot.org/coreboot/trunk@5981 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1

13 years agoUpdate superiotool support for FreeBSD, Makefile fixes.
Idwer Vollering [Sun, 24 Oct 2010 13:42:32 +0000 (13:42 +0000)]
Update superiotool support for FreeBSD, Makefile fixes.

Signed-off-by: Idwer Vollering <vidwer@gmail.com>
Acked-by: Uwe Hermann <uwe@hermann-uwe.de>
git-svn-id: svn://svn.coreboot.org/coreboot/trunk@5980 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1

13 years agoTeach superiotool about the registers in a Fintek F71806 (and F71872).
Jonathan Kollasch [Sun, 24 Oct 2010 12:43:41 +0000 (12:43 +0000)]
Teach superiotool about the registers in a Fintek F71806 (and F71872).

Signed-off-by: Jonathan Kollasch <jakllsch@kollasch.net>
Acked-by: Peter Stuge <peter@stuge.se>
Acked-by: Uwe Hermann <uwe@hermann-uwe.de>
git-svn-id: svn://svn.coreboot.org/coreboot/trunk@5979 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1

13 years agoRevert sblk/sblink change, use sblk like the rest of the codebase does.
Uwe Hermann [Wed, 20 Oct 2010 20:21:27 +0000 (20:21 +0000)]
Revert sblk/sblink change, use sblk like the rest of the codebase does.

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@5978 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1

13 years agoNow that no boards set RAMBASE < 1M, get rid of some dead code. Trivial.
Myles Watson [Wed, 20 Oct 2010 19:23:22 +0000 (19:23 +0000)]
Now that no boards set RAMBASE < 1M, get rid of some dead code.  Trivial.

It's probably time to reconsider moving all fam10 boards to RAMBASE = 1M.

Signed-off-by: Myles Watson <mylesgw@gmail.com>
Acked-by: Myles Watson <mylesgw@gmail.com>
git-svn-id: svn://svn.coreboot.org/coreboot/trunk@5977 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1

13 years agoFor AMD family 10h processors, msr c0010058 is always programmed
Scott Duplichan [Tue, 19 Oct 2010 21:08:11 +0000 (21:08 +0000)]
For AMD family 10h processors, msr c0010058 is always programmed
for 256 buses, even if fewer are configured. This patch lets msr
c0010058 programming use the configured bus count, CONFIG_MMCONF_BUS_NUMBER.

Signed-off-by: Scott Duplichan <scott@notabs.org>
Acked-by: Myles Watson <mylesgw@gmail.com>
git-svn-id: svn://svn.coreboot.org/coreboot/trunk@5976 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1

13 years agoDrop duplicate SB_HT_CHAIN_ON_BUS0 in Kconfig for MS-7135.
Jonathan Kollasch [Tue, 19 Oct 2010 15:25:06 +0000 (15:25 +0000)]
Drop duplicate SB_HT_CHAIN_ON_BUS0 in Kconfig for MS-7135.
Trivial.

Signed-off-by: Jonathan Kollasch <jakllsch@kollasch.net>
Acked-by: Jonathan Kollasch <jakllsch@kollasch.net>
git-svn-id: svn://svn.coreboot.org/coreboot/trunk@5975 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1

13 years agoUse the correct (W83627THF, not W83627HF) superio code in MS-7135 romstage.
Jonathan Kollasch [Tue, 19 Oct 2010 15:17:18 +0000 (15:17 +0000)]
Use the correct (W83627THF, not W83627HF) superio code in MS-7135 romstage.
This is consistent with the device tree and the chip actually on the board.
Trivial.

Signed-off-by: Jonathan Kollasch <jakllsch@kollasch.net>
Acked-by: Jonathan Kollasch <jakllsch@kollasch.net>
git-svn-id: svn://svn.coreboot.org/coreboot/trunk@5974 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1

13 years agoCopy ICH4 hard_reset() for 6300ESB.
Jonathan Kollasch [Tue, 19 Oct 2010 14:02:10 +0000 (14:02 +0000)]
Copy ICH4 hard_reset() for 6300ESB.

Signed-off-by: Jonathan Kollasch <jakllsch@kollasch.net>
Acked-by: Peter Stuge <peter@stuge.se>
git-svn-id: svn://svn.coreboot.org/coreboot/trunk@5973 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1

13 years agoRemove unused variables from 6300ESB smbus_write_block().
Jonathan Kollasch [Tue, 19 Oct 2010 13:49:11 +0000 (13:49 +0000)]
Remove unused variables from 6300ESB smbus_write_block().

#ifdef DEADCODE out smbus_write_byte() and smbus_write_block() as
they are static and nothing uses them or are incompletely implemented.

Signed-off-by: Jonathan Kollasch <jakllsch@kollasch.net>
Acked-by: Peter Stuge <peter@stuge.se>
git-svn-id: svn://svn.coreboot.org/coreboot/trunk@5972 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1

13 years agoCorrect spelling of "spacing" (in comments).
Jonathan Kollasch [Tue, 19 Oct 2010 13:39:38 +0000 (13:39 +0000)]
Correct spelling of "spacing" (in comments).

Signed-off-by: Jonathan Kollasch <jakllsch@kollasch.net>
Acked-by: Peter Stuge <peter@stuge.se>
git-svn-id: svn://svn.coreboot.org/coreboot/trunk@5971 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1

13 years agoUse mptable_write_buses().
Jonathan Kollasch [Tue, 19 Oct 2010 13:11:56 +0000 (13:11 +0000)]
Use mptable_write_buses().
Remove unhelpful comment.

Signed-off-by: Jonathan Kollasch <jakllsch@kollasch.net>
Acked-by: Peter Stuge <peter@stuge.se>
git-svn-id: svn://svn.coreboot.org/coreboot/trunk@5970 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1

13 years agoModernize socket_754 Kconfig with CAR and address bits information.
Jonathan Kollasch [Tue, 19 Oct 2010 13:03:34 +0000 (13:03 +0000)]
Modernize socket_754 Kconfig with CAR and address bits information.
Also, update the board that uses this socket to match.

Signed-off-by: Jonathan Kollasch <jakllsch@kollasch.net>
Acked-by: Peter Stuge <peter@stuge.se>
git-svn-id: svn://svn.coreboot.org/coreboot/trunk@5969 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1

13 years agoRevision 5966 changed the end of line style of the 3 modified files. This change...
Scott Duplichan [Tue, 19 Oct 2010 04:58:49 +0000 (04:58 +0000)]
Revision 5966 changed the end of line style of the 3 modified files. This change restores the original end of line style.

Signed-off-by: Scott Duplichan <scott@notabs.org>
Acked-by: Scott Duplichan <scott@notabs.org>
git-svn-id: svn://svn.coreboot.org/coreboot/trunk@5968 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1

13 years agoTo reduce boot time, remove the double startup IPI and 10 ms delay from lapic_cpu_ini...
Scott Duplichan [Tue, 19 Oct 2010 04:36:42 +0000 (04:36 +0000)]
To reduce boot time, remove the double startup IPI and 10 ms delay from lapic_cpu_init.c. The change is
currently restricted to recent model AMD processors, though it could be applied to others after successful testing.

Signed-off-by: Scott Duplichan <scott@notabs.org>
Acked-by: Myles Watson <mylesgw@gmail.com>
git-svn-id: svn://svn.coreboot.org/coreboot/trunk@5967 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1

13 years agoWhen debug logging is enabled, a message such as '* AP 02 timed out:02010501'
Scott Duplichan [Tue, 19 Oct 2010 04:26:17 +0000 (04:26 +0000)]
When debug logging is enabled, a message such as '* AP 02 timed out:02010501'
is sometimes logged. The reason is that the AP first sets a completion value
such as 0x13, which is what function wait_cpu_state() is waiting for. Then a
short time later, the AP calls function init_fidvid_ap(). This function sets
a completion value of 01. When logging is off, wait_cpu_state is fast enough
to see the initial completion value for each of the APs. But with logging
enabled, one or more APs may go on to complete function init_fidvid_ap, which
sets the completion value to 01. While mostly harmless, the timeout does
increase boot time. This patch eliminates the timeout by making function
wait_cpu_state recognize 01 as an additional valid AP completion value.

Signed-off-by: Scott Duplichan <scott@notabs.org>
Acked-by: Peter Stuge <peter@stuge.se>
git-svn-id: svn://svn.coreboot.org/coreboot/trunk@5966 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1

13 years agoThis patch enables SB700 option PrefetchEnSPIFromHost in early setup.
Scott Duplichan [Mon, 18 Oct 2010 04:01:12 +0000 (04:01 +0000)]
This patch enables SB700 option PrefetchEnSPIFromHost in early setup.
It affects only systems booting from SPI flash, not those booting from
LPC flash. By default, the SB700 reads dwords from the SPI flash chip.
Setting PrefetchEnSPIFromHost causes the SB700 to read entire cache
lines from the flash chip.

Signed-off-by: Scott Duplichan <scott@notabs.org>
Acked-by: Carl-Daniel Hailfinger <c-d.hailfinger.devel.2006@gmx.net>
git-svn-id: svn://svn.coreboot.org/coreboot/trunk@5965 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1

13 years agoupdate intel microcode files.
Stefan Reinauer [Mon, 18 Oct 2010 00:21:39 +0000 (00:21 +0000)]
update intel microcode files.

Signed-off-by: Stefan Reinauer <stepan@coreboot.org>
Acked-by: Stefan Reinauer <stepan@coreboot.org>
git-svn-id: svn://svn.coreboot.org/coreboot/trunk@5964 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1

13 years agoMake update-microcodes.sh executable.
Uwe Hermann [Mon, 18 Oct 2010 00:20:40 +0000 (00:20 +0000)]
Make update-microcodes.sh executable.

This also has an additional benefit:

I was running "sh update-microcodes.sh" previously which broke with

  update-microcodes.sh: 102: Bad substitution

due to the script requiring /bin/bash instead of /bin/sh (uses bash-specific
stuff). Running "bash update-microcodes.sh" works fine.

Making the script executable in svn reduces the likelyhood of people
running the script with differing shells that may not work.

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@5963 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1

13 years agoCosmetics and coding style fixes in devices/*.
Uwe Hermann [Mon, 18 Oct 2010 00:00:57 +0000 (00:00 +0000)]
Cosmetics and coding style fixes in devices/*.

 - Whitespace and indentation fixes in various places.

 - Fix various typos.

 - Use u8, u16 etc. everywhere.

Abuild-tested.

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@5962 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1

13 years agoupdate intel microcode update script
Stefan Reinauer [Sun, 17 Oct 2010 23:55:17 +0000 (23:55 +0000)]
update intel microcode update script
Signed-off-by: Stefan Reinauer <stepan@coreboot.org>
Acked-by: Stefan Reinauer <stepan@coreboot.org>
git-svn-id: svn://svn.coreboot.org/coreboot/trunk@5961 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1

13 years agoRemoves model_65x CPUIDs from model_6xx code.
Keith Hui [Sun, 17 Oct 2010 22:07:08 +0000 (22:07 +0000)]
Removes model_65x CPUIDs from model_6xx code.
They now have their own home at cpu/intel/model_65x.

Signed-off-by: Keith Hui <buurin@gmail.com>
Acked-by: Uwe Hermann <uwe@hermann-uwe.de>
git-svn-id: svn://svn.coreboot.org/coreboot/trunk@5960 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1

13 years agoIn the RS780/RS690 ProgK8TempMmioBase() function, the dst-link field is set
Liu Tao [Sun, 17 Oct 2010 21:59:43 +0000 (21:59 +0000)]
In the RS780/RS690 ProgK8TempMmioBase() function, the dst-link field is set
to zero, so for boards with RS780 not on CPU's HT chain 0, the function will
mis-configure the MMIO dst-link routing, and the following enable_pcie_bar3()
function will hang when it visits the MMIO.

The following patch fixes the problem, and is tested on a K8 board with RS780
on HT chain 1.

Signed-off-by: Liu Tao <liutao1980@gmail.com>
Acked-by: Uwe Hermann <uwe@hermann-uwe.de>
git-svn-id: svn://svn.coreboot.org/coreboot/trunk@5959 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1

13 years agoWe currently read the CPU HT speed from HT chain 0's register.
Liu Tao [Sun, 17 Oct 2010 21:34:45 +0000 (21:34 +0000)]
We currently read the CPU HT speed from HT chain 0's register.
Fix that to read the register from the chain where the SB chip is on.

Signed-off-by: Liu Tao <liutao1980@gmail.com>
Acked-by: Myles Watson <mylesgw@gmail.com>
Acked-by: Uwe Hermann <uwe@hermann-uwe.de>
git-svn-id: svn://svn.coreboot.org/coreboot/trunk@5958 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1

13 years agoAdd more missing GPL-headers, fix inconsistencies in others.
Uwe Hermann [Sun, 17 Oct 2010 19:30:58 +0000 (19:30 +0000)]
Add more missing GPL-headers, fix inconsistencies in others.

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@5957 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1

13 years agoUse common GPL-header format in CK804 files, add missing ones (trivial).
Uwe Hermann [Sun, 17 Oct 2010 19:13:18 +0000 (19:13 +0000)]
Use common GPL-header format in CK804 files, add missing ones (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@5956 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1

13 years agoVarious Doxygen comment fixes, typo fixes, etc.
Uwe Hermann [Sun, 17 Oct 2010 19:01:48 +0000 (19:01 +0000)]
Various Doxygen comment fixes, typo fixes, etc.

 - Fix incorrect argument names for @param entries.

 - Add missing @param and @return entries, partly as TODOs.

 - s/@returns/@return/, that's a typo.

 - Small whitespace fixes while I'm at it.

 - Drop useless @brief commands, they just clutter the comments and make them
   harder to read. Doxygen has an option to always use the first sentence
   of a Doxygen-comment as @brief automatically (should be on per default).

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@5955 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1

13 years agoMove support for Deschutes Slot 1 CPUs (model_65x) into its own directory.
Keith Hui [Sat, 16 Oct 2010 08:45:31 +0000 (08:45 +0000)]
Move support for Deschutes Slot 1 CPUs (model_65x) into its own directory.

abuild-tested. I have no Deschutes CPUs to boot test this with.

Signed-off-by: Keith Hui <buurin@gmail.com>
Acked-by: Uwe Hermann <uwe@hermann-uwe.de>
git-svn-id: svn://svn.coreboot.org/coreboot/trunk@5954 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1

13 years agoDrop unused DCACHE_RAM_BASE from intel/car/cache_as_ram.inc-using sockets.
Uwe Hermann [Fri, 15 Oct 2010 07:47:51 +0000 (07:47 +0000)]
Drop unused DCACHE_RAM_BASE from intel/car/cache_as_ram.inc-using sockets.

This CAR implementation hardcodes the Cache-as-RAM base address to:

  0xd0000 - CacheSize

so the DCACHE_RAM_BASE is never actually used for this implementation
and these sockets.

Signed-off-by: Uwe Hermann <uwe@hermann-uwe.de>
Acked-by: Peter Stuge <peter@stuge.se>
git-svn-id: svn://svn.coreboot.org/coreboot/trunk@5953 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1

13 years agoCosmetics in ioapic.c (trivial, no functional changes).
Uwe Hermann [Thu, 14 Oct 2010 23:40:10 +0000 (23:40 +0000)]
Cosmetics in ioapic.c (trivial, no functional changes).

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@5952 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1

13 years agoRemove various .c #includes from Intel i810/i82801ax/i82801bx boards.
Uwe Hermann [Wed, 13 Oct 2010 23:00:41 +0000 (23:00 +0000)]
Remove various .c #includes from Intel i810/i82801ax/i82801bx boards.

This is pretty much the same mechanism as in r5929.

 - Use 'romstage-y' to turn i82801ax_early_smbus.c and i82801bx_early_smbus.c
   into distinct compilation units, and don't #include the files anymore
   in romstage.c files.

 - Ditto for northbridge/intel/i82810/raminit.c, and
   northbridge/intel/i82810/debug.c.

 - Add various header files which are now needed, drop unused includes.

 - Make functions that need to be visible non-static.

Abuild-tested.

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@5951 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1

13 years agoMove out Katmai Slot 1 CPUs (model_67x) from model_6xx to model_67x.
Keith Hui [Wed, 13 Oct 2010 17:00:42 +0000 (17:00 +0000)]
Move out Katmai Slot 1 CPUs (model_67x) from model_6xx to model_67x.

abuild-tested. Boot tested on P2B-LS.

Signed-off-by: Keith Hui <buurin@gmail.com>
Acked-by: Uwe Hermann <uwe@hermann-uwe.de>
git-svn-id: svn://svn.coreboot.org/coreboot/trunk@5950 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1

13 years agoConvert all Intel i810 boards to CAR.
Uwe Hermann [Wed, 13 Oct 2010 08:21:44 +0000 (08:21 +0000)]
Convert all Intel i810 boards to CAR.

 - Drop "select ROMCC" from the boards, as well as early_mtrr stuff.

 - Add "select CACHE_AS_RAM" to socket_PGA370/Kconfig, as well as the
   usual DCACHE_RAM_BASE and DCACHE_RAM_SIZE variables.

 - In socket_PGA370/Makefile.inc add:
   cpu_incs += $(src)/cpu/intel/car/cache_as_ram.inc

 - Other smaller related fixes.

Abuild-tested and boot-tested on MSI MS-6178.

Signed-off-by: Uwe Hermann <uwe@hermann-uwe.de>
Acked-by: Peter Stuge <peter@stuge.se>
git-svn-id: svn://svn.coreboot.org/coreboot/trunk@5949 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1

13 years agoEnable or disable the power button in Kconfig
Peter Stuge [Wed, 13 Oct 2010 06:23:02 +0000 (06:23 +0000)]
Enable or disable the power button in Kconfig

Some mainboards need to disable the power button to avoid turning off
right after being turned on, while other boards ship with a jumper over
the power button and should allow the user to configure the behavior.

This adds infrastructure in the form of four mutually exclusive options
which can be selected in a mainboard Kconfig (power button forced on/off,
and user-controllable with default on/off) and one result bool which
source code can test. (Enable the button or not.)

The options have been implemented in CS5536 code and for all mainboards
which select SOUTHBRIDGE_AMD_CS5536, but should be used also by other
chipsets where applicable. Note that if chipset code uses the result
bool ENABLE_POWER_BUTTON, then every board using that chipset must
select one out of the four control options in order to build.

All touched boards should have unchanged behavior, except
pcengines/alix1c, traverse/geos and lippert/hurricane-lx where the
power button can now be configured by the user.

Build tested for alix1c, alix2d, hurricane-lx and wyse-s50. Confirmed
to work as advertised on alix1c both with button enabled and disabled.

Includes additional traverse/geos changes from Nathan and
lippert/hurricane-lx changes from Jens to correctly use the new
feature on those boards.

Signed-off-by: Peter Stuge <peter@stuge.se>
Acked-by: Aurelien Guillaume <aurelien@iwi.me>
Acked-by: Nils Jacobs <njacobs8@hetnet.nl>
git-svn-id: svn://svn.coreboot.org/coreboot/trunk@5948 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1

13 years agoFix a stupid bug in rs780 and rs690 code.
Zheng Bao [Wed, 13 Oct 2010 05:16:48 +0000 (05:16 +0000)]
Fix a stupid bug in rs780 and rs690 code.

Signed-off-by: Zheng Bao <zheng.bao@amd.com>
Acked-by: Scott Duplichan <scott@notabs.org>
git-svn-id: svn://svn.coreboot.org/coreboot/trunk@5947 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1

13 years agoTrivial. Clean up code and add some comments.
Zheng Bao [Wed, 13 Oct 2010 02:46:59 +0000 (02:46 +0000)]
Trivial. Clean up code and add some comments.

Signed-off-by: Zheng Bao <zheng.bao@amd.com>
Acked-by: Zheng Bao <zheng.bao@amd.com>
git-svn-id: svn://svn.coreboot.org/coreboot/trunk@5946 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1

13 years agoAdd missing include of model_6bx for slot_1.
Keith Hui [Tue, 12 Oct 2010 23:22:08 +0000 (23:22 +0000)]
Add missing include of model_6bx for slot_1.

I could no longer boot my P3B-F with my Tualeron and r5938. Dies with
"unknown CPU". I believe it will happen with any Slot 1 440BX boards
that supports model_6bx CPUs.

I need to make the change below to make it work. abuild tested. Boot
tested on P2B-LS and P3B-F.

Signed-off-by: Keith Hui <buurin@gmail.com>
Acked-by: Uwe Hermann <uwe@hermann-uwe.de>
git-svn-id: svn://svn.coreboot.org/coreboot/trunk@5945 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1

13 years agoMove translate_spd_to_i82810[] from .h to .c file (trivial).
Uwe Hermann [Tue, 12 Oct 2010 21:37:03 +0000 (21:37 +0000)]
Move translate_spd_to_i82810[] from .h to .c file (trivial).

This is in preparation of further i810 fixes and switching it to CAR.

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@5944 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1

13 years agoWe define IO_APIC_ADDR in <arch/ioapic.h>, let's use it.
Uwe Hermann [Tue, 12 Oct 2010 17:34:08 +0000 (17:34 +0000)]
We define IO_APIC_ADDR in <arch/ioapic.h>, let's use it.

As both ioapic.h and acpi.h define a macro named "NMI", rename one
of them (NMI -> NMIType in acpi.h).

Abuild-tested.

Signed-off-by: Uwe Hermann <uwe@hermann-uwe.de>
Acked-by: Peter Stuge <peter@stuge.se>
git-svn-id: svn://svn.coreboot.org/coreboot/trunk@5943 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1

13 years agoReduce duplicate definition in CAR code.
Warren Turkal [Tue, 12 Oct 2010 06:13:40 +0000 (06:13 +0000)]
Reduce duplicate definition in CAR code.

Macros for the register addresses for the MTRR MSRs are already defined
in include/cpu/x86/car.h. This patch uses those macros instead of
creating a second instance of that same data.

I also added a few macros to the amd mtrr.h to make the MSR naming more
consistent.

Signed-off-by: Warren Turkal <wt@penguintechs.org>
Acked-by: Stefan Reinauer <stepan@coresystems.de>
git-svn-id: svn://svn.coreboot.org/coreboot/trunk@5942 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1

13 years agoBuild bootblock.S instead of bootblock.c.
Warren Turkal [Tue, 12 Oct 2010 06:12:00 +0000 (06:12 +0000)]
Build bootblock.S instead of bootblock.c.

The file is actually just including a bunch of assembly. The build rule
for bootblock.c even states that the file will be "assembler-with-cpp."

Signed-off-by: Warren Turkal <wt@penguintechs.org>
Acked-by: Stefan Reinauer <stepan@coresystems.de>
git-svn-id: svn://svn.coreboot.org/coreboot/trunk@5941 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1

13 years agoFix typo after r5925.
Sylvain Hitier [Mon, 11 Oct 2010 23:22:24 +0000 (23:22 +0000)]
Fix typo after r5925.

BTW, embed the always-the-same string instead of referencing it through "%s".

Do the same for i82371EB while we're at it.

Signed-off-by: Sylvain Hitier <sylvain.hitier@gmail.com>
Acked-by: Uwe Hermann <uwe@hermann-uwe.de>
git-svn-id: svn://svn.coreboot.org/coreboot/trunk@5940 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1

13 years agoBuild fixes (trivial).
Uwe Hermann [Mon, 11 Oct 2010 22:49:39 +0000 (22:49 +0000)]
Build fixes (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@5939 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1

13 years agoFirst round of ICH2/ICH2-M cleanups after split from i82801xx.
Uwe Hermann [Mon, 11 Oct 2010 21:38:49 +0000 (21:38 +0000)]
First round of ICH2/ICH2-M cleanups after split from i82801xx.

 - Drop all non-ICH2 "struct pci_driver" entries from all files.

 - Kconfig: Add missing USE_WATCHDOG_ON_BOOT.

 - Drop i82801bx_sata.c and i82801bx_usb_ehci.c, ICH2 doesn't have SATA/EHCI.

 - Simplify lots of code, getting rid of i82801xx remainders.

 - Use u8 et al (instead of uint8_t) in a few more places.

 - Use #defines from header files where possible.

 - Various other fixes and updates.

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@5938 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1

13 years agoFactor out a few commonly duplicated functions from northbridge.c.
Uwe Hermann [Mon, 11 Oct 2010 19:36:13 +0000 (19:36 +0000)]
Factor out a few commonly duplicated functions from northbridge.c.

The following functions are moved to devices/device_util.c:

 - ram_resource()

 - tolm_test()

 - find_pci_tolm()

There are only two tolm_test() / find_pci_tolm() which differ from the
defaults, one of them can easily be eliminated in a follow-up patch,
maybe even both, but for now keep it simple and only eliminate the majority.

Signed-off-by: Uwe Hermann <uwe@hermann-uwe.de>
Acked-by: Peter Stuge <peter@stuge.se>
git-svn-id: svn://svn.coreboot.org/coreboot/trunk@5937 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1

13 years agofix typo in minilzma.cc, found by Idwer.
Stefan Reinauer [Sun, 10 Oct 2010 22:05:02 +0000 (22:05 +0000)]
fix typo in minilzma.cc, found by Idwer.

Signed-off-by: Stefan Reinauer <stepan@coresystems.de>
Acked-by: Stefan Reinauer <stepan@coresystems.de>
git-svn-id: svn://svn.coreboot.org/coreboot/trunk@5936 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1

13 years agodon't include unused code, we only need the header.
Stefan Reinauer [Sun, 10 Oct 2010 21:15:53 +0000 (21:15 +0000)]
don't include unused code, we only need the header.
Acked-by: Stefan Reinauer <stepan@coresystems.de>
Signed-off-by: Stefan Reinauer <stepan@coresystems.de>
git-svn-id: svn://svn.coreboot.org/coreboot/trunk@5935 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1

13 years agofix typos found by zbao in other files.
Stefan Reinauer [Sun, 10 Oct 2010 21:15:01 +0000 (21:15 +0000)]
fix typos found by zbao in other files.

Signed-off-by: Stefan Reinauer <stepan@coresystems.de>
Acked-by: Stefan Reinauer <stepan@coresystems.de>
git-svn-id: svn://svn.coreboot.org/coreboot/trunk@5934 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1

13 years agoSame applies for SB600.
Rudolf Marek [Sun, 10 Oct 2010 20:43:00 +0000 (20:43 +0000)]
Same applies for SB600.

Following patch enables UDMA on ALL IDE devices. The current code enables it only for primary master, which causes my DVD drive to fail under windows install
and even after hard reset in linux (DMA seems lockup).

The fix should not have any influence for Linux because the IDE driver will
correctly reprogram this bit.

Signed-off-by: Rudolf Marek <r.marek@assembler.cz>
Acked-by: Uwe Hermann <uwe@hermann-uwe.de>
git-svn-id: svn://svn.coreboot.org/coreboot/trunk@5933 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1

13 years agoFollowing patch fixes the boot_switch_sata_ide logic. It can swap
Rudolf Marek [Sun, 10 Oct 2010 19:55:32 +0000 (19:55 +0000)]
Following patch fixes the boot_switch_sata_ide logic. It can swap
primary / secondary IDE channel with SATA (in IDE mode).

The bug was that setup was done in wrong device.

Signed-off-by: Rudolf Marek <r.marek@assembler.cz>
Acked-by: Uwe Hermann <uwe@hermann-uwe.de>
git-svn-id: svn://svn.coreboot.org/coreboot/trunk@5932 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1

13 years agoFollowing patch enables UDMA on ALL IDE devices. The current code enables it only...
Rudolf Marek [Sun, 10 Oct 2010 19:54:15 +0000 (19:54 +0000)]
Following patch enables UDMA on ALL IDE devices. The current code enables it only for primary master, which causes my DVD drive to fail under windows install
and even after hard reset in linux (DMA seems lockup).

The fix should not have any influence for Linux because the IDE driver will
correctly reprogram this bit.

Signed-off-by: Rudolf Marek <r.marek@assembler.cz>
Acked-by: Uwe Hermann <uwe@hermann-uwe.de>
git-svn-id: svn://svn.coreboot.org/coreboot/trunk@5931 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1

13 years agoTrivial. Spelling check.
Zheng Bao [Sun, 10 Oct 2010 15:18:53 +0000 (15:18 +0000)]
Trivial. Spelling check.

Signed-off-by: Zheng Bao <zheng.bao@amd.com>
Acked-by: Zheng Bao <zheng.bao@amd.com>
git-svn-id: svn://svn.coreboot.org/coreboot/trunk@5930 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1

13 years agoRemove various .c #includes from Intel 440BX/82371EB boards.
Uwe Hermann [Sat, 9 Oct 2010 17:00:18 +0000 (17:00 +0000)]
Remove various .c #includes from Intel 440BX/82371EB boards.

 - Use 'romstage-y' to turn i82371eb_early_pm.c and i82371eb_early_smbus.c
   into distinct compilation units, and don't #include the files anymore
   in romstage.c files.

 - Ditto for lib/debug.c, northbridge/intel/i440bx/raminit.c, and
   northbridge/intel/i440bx/debug.c.

 - Add various header files which are now needed.

 - Make functions that need to be visible non-static.

 - Drop a remaining "select ROMCC" from a 4440BX board.

Signed-off-by: Uwe Hermann <uwe@hermann-uwe.de>
Acked-by: Idwer Vollering <vidwer@gmail.com>
git-svn-id: svn://svn.coreboot.org/coreboot/trunk@5929 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1

13 years agoTrivial. Spell checking.
Zheng Bao [Sat, 9 Oct 2010 07:18:50 +0000 (07:18 +0000)]
Trivial. Spell checking.

Signed-off-by: Zheng Bao <zheng.bao@amd.com>
Acked-by: Zheng Bao <zheng.bao@amd.com>
git-svn-id: svn://svn.coreboot.org/coreboot/trunk@5928 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1

13 years agoTrivial. Spell checking.
Zheng Bao [Sat, 9 Oct 2010 02:31:10 +0000 (02:31 +0000)]
Trivial. Spell checking.

Signed-off-by: Zheng Bao <zheng.bao@amd.com>
Acked-by: Zheng Bao <zheng.bao@amd.com>
git-svn-id: svn://svn.coreboot.org/coreboot/trunk@5927 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1