Only the BSP CPU was able to start its hyper-threading CPU siblings.
When an AP CPU attempts this it calls start_cpu() within start_cpu(),
deadlocking the system with start_cpu_lock.
At the time intel_sibling_init() is run, the BSP CPU is still
walking the cpu_bus linked list in lapic_cpu_init: start_other_cpus().
A sibling CPU appended at the end of this list will get started.
Also fail compile with #error if SERIAL_CPU_INIT==0, as microcode
updates on hyper-threading sibling CPUs must be serialized.
Tested with HT-enabled P4 Xeons on dual-socket604 platform.
Change-Id: I0053f58f49ed604605ce0a55e826d3e1afdc90b6
Signed-off-by: Kyösti Mälkki <kyosti.malkki@gmail.com>
Reviewed-on: http://review.coreboot.org/775
Reviewed-by: Ronald G. Minnich <rminnich@gmail.com>
Tested-by: build bot (Jenkins)
Reviewed-by: Rudolf Marek <r.marek@assembler.cz>
Renamed CONFIG_ROMBASE to ROMSTAGE_BASE and removed it from Kconfig.
Removed no-op calculation in ldscript.
Change-Id: I53d39b60f07db76c8537b3133e59360687b9d4a7
Signed-off-by: Kyösti Mälkki <kyosti.malkki@gmail.com>
Reviewed-on: http://review.coreboot.org/802
Tested-by: build bot (Jenkins)
Reviewed-by: Patrick Georgi <patrick@georgi-clan.de>
Reviewed-by: Rudolf Marek <r.marek@assembler.cz>
There was some corner case where commit-msg failed. Update to
latest upstream version.
Change-Id: I822d6c3f64728de7356401465e00575ac5af8196
Signed-off-by: Patrick Georgi <patrick@georgi-clan.de>
Reviewed-on: http://review.coreboot.org/798
Tested-by: build bot (Jenkins)
Reviewed-by: Bernhard Urban <lewurm@gmail.com>
Reviewed-by: Rudolf Marek <r.marek@assembler.cz>
This is developer's testtool. Output from a "rotate ones" -style
pattern helps figure out how DIMM addresses are encoded or routed
on a certain mainboard.
Scattered test should cover every data and address lines on the memory
bus, but is probably limited to the first bank of first DIMM.
Change-Id: I533a7a873bcc434f99e7faed9dc9337d9ab64196
Signed-off-by: Kyösti Mälkki <kyosti.malkki@gmail.com>
Signed-off-by: Patrick Georgi <patrick@georgi-clan.de>
[pg: rebase]
Reviewed-on: http://review.coreboot.org/294
Tested-by: build bot (Jenkins)
Reviewed-by: Rudolf Marek <r.marek@assembler.cz>
Changing CMOS value for power-on-after-power-fail was only honored
after reboot, which is counter intuitive (set from "enable" to "disable",
power-off, replug device -> device turns on; and similar cases).
Change-Id: If1d88c1c34c3333b636ed3ec1e1fb9bea394e615
Signed-off-by: Patrick Georgi <patrick.georgi@secunet.com>
Reviewed-on: http://review.coreboot.org/444
Tested-by: build bot (Jenkins)
Reviewed-by: Stefan Reinauer <stefan.reinauer@coreboot.org>
We used a hard coded value for some reason. Don't do that, but use CMOS
instead.
Change-Id: Ib83aa07a3e55bed075150354a060317ebc9d5ba7
Signed-off-by: Patrick Georgi <patrick.georgi@secunet.com>
Reviewed-on: http://review.coreboot.org/443
Tested-by: build bot (Jenkins)
Reviewed-by: Stefan Reinauer <stefan.reinauer@coreboot.org>
We use the L modifier in a non-standard way (for
long long instead of long double, which we have no
business with).
clang complains, to reduce its use, to make
emulation/qemu-x86 happier.
Long term, we should consider eliminating public uses
of 'L' (but internal use in vtxprintf to denote
long long is fine)
Change-Id: If9a17d9ae9925cdc8736445e7d5eedc59c7028c6
Signed-off-by: Patrick Georgi <patrick@georgi-clan.de>
Reviewed-on: http://review.coreboot.org/781
Tested-by: build bot (Jenkins)
Reviewed-by: Stefan Reinauer <stefan.reinauer@coreboot.org>
The previous commit was incomplete and missed setting the entire
alloc area.
There are also additional problems with gcc optimizations of the
pointer math. The "auto" casting by gcc wouldn't return warnings,
but it was causing the optimization to be incorrect. We are now
very explicit in the casting in the pointer math.
Change-Id: I020808c8d1dda544fe862b9efb0e5345eeab5aab
Signed-off-by: Marc Jones <marc.jones@se-eng.com>
Reviewed-on: http://review.coreboot.org/804
Tested-by: build bot (Jenkins)
Reviewed-by: Peter Stuge <peter@stuge.se>
Provide a way to redefine the names of normal and fallback via CBFS.
This way updates can use some more expressive naming scheme (numbers,
dates, version numbers) and replace the coreboot-stages file to
point to the new version (with the current version as new "old").
If coreboot-stages doesn't exist, the default behaviour remains to
use "normal" and "fallback".
Change-Id: I77c134d79ed95831ad5098b7663c15e95d3b5a2a
Signed-off-by: Patrick Georgi <patrick@georgi-clan.de>
Reviewed-on: http://review.coreboot.org/589
Tested-by: build bot (Jenkins)
Reviewed-by: Marc Jones <marcj303@gmail.com>
It was not obvious which CAR was compiled in. Also build would fail
if a socket included two models with both having an include for CAR.
Change-Id: I000c2e24807c3d99347a43d120333c13fbf91af4
Signed-off-by: Kyösti Mälkki <kyosti.malkki@gmail.com>
Reviewed-on: http://review.coreboot.org/626
Tested-by: build bot (Jenkins)
Reviewed-by: Patrick Georgi <patrick@georgi-clan.de>
The Fam15 northbridge.c had hardcoded the CBMEM size. It should use
the one in cbmem.h instead.
Change-Id: I8a00e05884bdb1d1a4a012433b0adfbb9eb22983
Signed-off-by: Stefan Reinauer <reinauer@google.com>
Reviewed-on: http://review.coreboot.org/796
Tested-by: build bot (Jenkins)
Reviewed-by: Marc Jones <marcj303@gmail.com>
The Fam12 northbridge.c had hardcoded the CBMEM size. It should use
the one in cbmem.h instead.
Change-Id: I1eca18e21fa59ae32e802d8452e42e8b7a3575cf
Signed-off-by: Stefan Reinauer <reinauer@google.com>
Reviewed-on: http://review.coreboot.org/795
Tested-by: build bot (Jenkins)
Reviewed-by: Marc Jones <marcj303@gmail.com>
The Fam10 northbridge.c had hardcoded the CBMEM size. It should use
the one in cbmem.h instead.
Change-Id: Id6c4128d8f5f6a417f83daa3a39b2bfc8e810f8a
Signed-off-by: Stefan Reinauer <reinauer@google.com>
Reviewed-on: http://review.coreboot.org/794
Tested-by: build bot (Jenkins)
Reviewed-by: Marc Jones <marcj303@gmail.com>
Also any CPU_AMD_AGESA_FAMILYxx selects CPU_AMD_AGESA, so remove
the explicit selects from the mainboards.
Change-Id: I4d71726bccd446b0f4db4e26448b5c91e406a641
Signed-off-by: Kyösti Mälkki <kyosti.malkki@gmail.com>
Reviewed-on: http://review.coreboot.org/792
Tested-by: build bot (Jenkins)
Reviewed-by: Patrick Georgi <patrick@georgi-clan.de>
Kconfig leaked XIP_ROM_SIZE to other platforms and also
defined obsolete option XIP_ROM_BASE.
Alias AMD_AGESA as NORTHBRIDGE_AMD_AGESA.
Break the circular dependency with family15 Kconfig.
Change-Id: Ic7891012220e1bef758a5a39002b66971d5206e3
Signed-off-by: Kyösti Mälkki <kyosti.malkki@gmail.com>
Reviewed-on: http://review.coreboot.org/773
Tested-by: build bot (Jenkins)
Reviewed-by: Patrick Georgi <patrick@georgi-clan.de>
So set their XIP configuration to ROM_SIZE.
Change-Id: I6c1abccec3b1d7389c85df55343ff0fc68a61eec
Signed-off-by: Patrick Georgi <patrick@georgi-clan.de>
Reviewed-on: http://review.coreboot.org/797
Tested-by: build bot (Jenkins)
Reviewed-by: Kyösti Mälkki <kyosti.malkki@gmail.com>
clang is more picky on that.
Change-Id: Iaa8472beb6e275c39037d11e1a72dbb80d46424b
Signed-off-by: Patrick Georgi <patrick@georgi-clan.de>
Reviewed-on: http://review.coreboot.org/779
Tested-by: build bot (Jenkins)
Reviewed-by: Stefan Reinauer <stefan.reinauer@coreboot.org>
clang doesn't know about the side effect, so we have to tell it
that it's okay not to care about the result.
Change-Id: Ib11890bff6779e36cf09c178d224695ea16a8ae8
Signed-off-by: Patrick Georgi <patrick@georgi-clan.de>
Reviewed-on: http://review.coreboot.org/783
Tested-by: build bot (Jenkins)
Reviewed-by: Stefan Reinauer <stefan.reinauer@coreboot.org>
clang is much more trigger happy than gcc on those.
Change-Id: Ie7c219de3cc26675692eab7361a4ad551f1c65a7
Signed-off-by: Patrick Georgi <patrick@georgi-clan.de>
Reviewed-on: http://review.coreboot.org/786
Tested-by: build bot (Jenkins)
Reviewed-by: Stefan Reinauer <stefan.reinauer@coreboot.org>
Use NORTHBRIDGE_INTEL_I945 to select the driver directory for build.
Use _SUBTYPE_945GC and _SUBTYPE_945GM to define at compile-time
which model of I945 the driver is built for.
Change-Id: I11b1e0998d0fc28f8946bad4f0989036a9b18af4
Signed-off-by: Kyösti Mälkki <kyosti.malkki@gmail.com>
Reviewed-on: http://review.coreboot.org/684
Tested-by: build bot (Jenkins)
Reviewed-by: Patrick Georgi <patrick@georgi-clan.de>
The original comment says it's a Via C3 and not Epia requirement
to deliver IOAPIC interrupts on APIC serial bus.
Change-Id: I73c55755e0ec1ac5756b4ee7ccdfc8eb93184e4f
Signed-off-by: Kyösti Mälkki <kyosti.malkki@gmail.com>
Signed-off-by: Patrick Georgi <patrick@georgi-clan.de>
Reviewed-on: http://review.coreboot.org/435
Tested-by: build bot (Jenkins)
Use separate Kconfig option to select a driver directory for
build and the specific type of southbridge to support.
Change-Id: I9482d4ea0f0234b9b7ff38144e45022ab95cf3f3
Signed-off-by: Kyösti Mälkki <kyosti.malkki@gmail.com>
Reviewed-on: http://review.coreboot.org/685
Tested-by: build bot (Jenkins)
Reviewed-by: Patrick Georgi <patrick@georgi-clan.de>
In a case of CS & 0x0fff != 0x0, lidt memory operand does not point
to nullidt, this can raise an exception and shutdown the CPU.
When an AP CPU receives 8-bit Start-Up IPI vector yzH, it starts
execute at physical address 000yz000H. Seems this translates to
either yz00:0000 or y000:z000 (CS:IP), depending of the CPU model.
With the change entry16.inc is relocatable as the commentary suggests
and can be used as ap_sipi_vector on SMP systems.
Change-Id: I885a2888179700ba6e2b11d4f2d6a64ddea4c2dc
Signed-off-by: Kyösti Mälkki <kyosti.malkki@gmail.com>
Reviewed-on: http://review.coreboot.org/707
Tested-by: build bot (Jenkins)
Reviewed-by: Idwer Vollering <vidwer@gmail.com>
Reviewed-by: Ronald G. Minnich <rminnich@gmail.com>
Reviewed-by: Patrick Georgi <patrick@georgi-clan.de>
The Fam14 northbridge.c had hardcoded the cbmem size. It should use
in cbmem.h instead.
Change-Id: I910329fc98a4cf04dc81ef66f3aa05a1916f5b1d
Signed-off-by: Marc Jones <marc.jones@se-eng.com>
Reviewed-on: http://review.coreboot.org/790
Reviewed-by: Stefan Reinauer <stefan.reinauer@coreboot.org>
Tested-by: build bot (Jenkins)
Since cbfs_core.h provides a macro that uses ntohl, make sure ntohl is available by
including byteorder.h
Change-Id: I9ab8cb51bd680e861b28d5130d09547bb9ab3b1f
Signed-off-by: Gabe Black <gabeblack@google.com>
Reviewed-on: http://review.coreboot.org/709
Tested-by: build bot (Jenkins)
Reviewed-by: Peter Stuge <peter@stuge.se>
Fixes the warnings generated in the torpedo mainboard build by AGESA.
Removing broken tests.
Change-Id: Ib444fa2bf4dd94cadb4ce33040eb5650d1c0325b
Signed-off-by: Martin L Roth <martin@se-eng.com>
Reviewed-on: http://review.coreboot.org/667
Tested-by: build bot (Jenkins)
Reviewed-by: Stefan Reinauer <stefan.reinauer@coreboot.org>
This commit includes the changes to enable the HDMI on Union
Station. The changes switch the output from the display port
to the HDMI.
Change-Id: I4e15ff6db7d056f156791ff1406d4bae35ff2767
Signed-off-by: Frank Vibrans <frank.vibrans@se-eng.com>
Reviewed-on: http://review.coreboot.org/788
Tested-by: build bot (Jenkins)
Reviewed-by: Peter Stuge <peter@stuge.se>
Because the Union Station platform doesn't have an SIO chip,
this commit removes the Fintek SIO support.
Change-Id: Idba4222ce136821dee2530a72d1630eb5ad613a2
Signed-off-by: Frank Vibrans <frank.vibrans@se-eng.com>
Reviewed-on: http://review.coreboot.org/787
Reviewed-by: Peter Stuge <peter@stuge.se>
Tested-by: build bot (Jenkins)
Change-Id: I69ee67c35113d98e034bdccf5d00e8452d3d9bd2
Signed-off-by: Patrick Georgi <patrick@georgi-clan.de>
Reviewed-on: http://review.coreboot.org/778
Tested-by: build bot (Jenkins)
Reviewed-by: Rudolf Marek <r.marek@assembler.cz>
Makefile.inc uses $( ) syntax on the shell. That's isn't as universal
as one would like.
Change-Id: I9a8fd511eef7fefc1458d5bae2cd7ef5475b7392
Signed-off-by: Patrick Georgi <patrick@georgi-clan.de>
Reviewed-on: http://review.coreboot.org/777
Tested-by: build bot (Jenkins)
Reviewed-by: Bernhard Urban <lewurm@gmail.com>
gcc4.6.2 was optimizing the libpayload alloc() function and failing to
reload a pointer after the memory had been manipulated by a pointer in
the inlined function setup(). Change the pointer type to volatile
and now pass it to the setup() function. Also clean up the
declaration so that it isn't cast a bunch times in the function.
Change-Id: I1637bd7bd5d9cf82ac88925cbfe76d319aa3cd82
Signed-off-by: Marc Jones <marc.jones@se-eng.com>
Reviewed-on: http://review.coreboot.org/705
Tested-by: build bot (Jenkins)
Reviewed-by: Patrick Georgi <patrick@georgi-clan.de>
This way u-boot won't try to use a UART that isn't plugged in.
Change-Id: I9a3a0d074dd03add8afbd4dad836c4c6a05abe6f
Signed-off-by: Gabe Black <gabeblack@google.com>
Reviewed-on: http://review.coreboot.org/729
Reviewed-by: Patrick Georgi <patrick@georgi-clan.de>
Tested-by: build bot (Jenkins)
no dumping yet
Change-Id: I4e687ca816c8d6d1c95255b0abf6a19513e23f86
Signed-off-by: Stefan Reinauer <reinauer@google.com>
Reviewed-on: http://review.coreboot.org/734
Reviewed-by: Ronald G. Minnich <rminnich@gmail.com>
Tested-by: build bot (Jenkins)
Packing a device tree into the coreboot table can easily make
the table exceed the current limit of 8KB. However, right now
there is no error handling in place to catch that case.
Increase the maximum memory usable for all tables from 64KB to
128KB and increase the maximum coreboot table size from 8KB
to 32KB.
Change-Id: I2025bf070d0adb276c1cd610aa8402b50bdf2525
Signed-off-by: Stefan Reinauer <reinauer@google.com>
Reviewed-on: http://review.coreboot.org/704
Tested-by: build bot (Jenkins)
Reviewed-by: Ronald G. Minnich <rminnich@gmail.com>
The code when reporting the coreboot table size did not account
for the last added table record. This change fixes the problem.
. rebuild coreboot, program it on the target, restart it
. look for 'Wrote coreboot table at:' in the console log
. observe the adequate table size reported
$ grep 'Wrote coreboot table:' /tmp/cb.log
Wrote coreboot table at: 00000500, 0x10 bytes, checksum c06f
Wrote coreboot table at: 7f6fc000, 0x1a73 bytes, checksum 3e45
$
Change-Id: Ic55501a4ae06fab2bcda9aea58e362325f2edccf
Signed-off-by: Vadim Bendebury <vbendeb@chromium.org>
Reviewed-on: http://review.coreboot.org/703
Tested-by: build bot (Jenkins)
Reviewed-by: Ronald G. Minnich <rminnich@gmail.com>
CONFIG_ variables are used inconsistently within the file
src/arch/x86/boot/coreboot_table.c. #ifdef will do the wrong
thing if the option is disabled. #if (CONFIG_FOO == 1) is
not needed.
Change-Id: Ifcac6ceac5fb34b931281beae500023597b3533b
Signed-off-by: Stefan Reinauer <reinauer@google.com>
Reviewed-on: http://review.coreboot.org/701
Reviewed-by: Ronald G. Minnich <rminnich@gmail.com>
Tested-by: build bot (Jenkins)
If you build in parallel, option_table.h will occasionally not be there yet
and the build will fail.
Change-Id: I828956ab2e05c48d20c2f7c55616cc8fa19e1227
Signed-off-by: Stefan Reinauer <reinauer@google.com>
Reviewed-on: http://review.coreboot.org/698
Tested-by: build bot (Jenkins)
Reviewed-by: Patrick Georgi <patrick@georgi-clan.de>
I am not sure if the sub bus being 0 is a problem, or if the assumption
there has to be at least one non empty link is just wrong. It certainly
does not hurt to add a small consistency check in either case.
Change-Id: I098446deef96a8baae26a7ca1ddd96e626a06dc5
Signed-off-by: Stefan Reinauer <reinauer@google.com>
Reviewed-on: http://review.coreboot.org/693
Tested-by: build bot (Jenkins)
Reviewed-by: Ronald G. Minnich <rminnich@gmail.com>
... and only pull in early init code if the OXPCIe is used for console.
Change-Id: I01feca3b9e8376a75c17554ba1bd200d523dff8d
Signed-off-by: Stefan Reinauer <reinauer@google.com>
Reviewed-on: http://review.coreboot.org/692
Tested-by: build bot (Jenkins)
Reviewed-by: Ronald G. Minnich <rminnich@gmail.com>
Because it's included everywhere anyways.
Change-Id: I99a9e6edac08df57c50ef3a706fdbd395cad0abc
Signed-off-by: Stefan Reinauer <reinauer@google.com>
Reviewed-on: http://review.coreboot.org/691
Tested-by: build bot (Jenkins)
Reviewed-by: Ronald G. Minnich <rminnich@gmail.com>
It has a smaller footprint than the already supported MPEX2S952
Change-Id: Ie36b67f9628882d516ca34ff164f0e8918955a5b
Signed-off-by: Stefan Reinauer <reinauer@google.com>
Tested-by: Stefan Reinauer <reinauer@google.com>
Reviewed-by: Duncan Laurie <dlaurie@google.com>
Reviewed-on: http://review.coreboot.org/690
Tested-by: build bot (Jenkins)
Reviewed-by: Ronald G. Minnich <rminnich@gmail.com>
Right now coreboot only executes VGA Option ROMs. However, this is not
good enough. For security reasons we want to execute only Option ROMs
stored in our r/o CBFS.
This patch adds a new option to disable execution of arbitrary Option
ROMs.
Also fix the capitalization of Option ROM in src/devices/Kconfig
Change-Id: I485291c06ec5cd1f875357401831fe32ccfc5f2f
Signed-off-by: Stefan Reinauer <reinauer@google.com>
Reviewed-on: http://review.coreboot.org/730
Tested-by: build bot (Jenkins)
Reviewed-by: Ronald G. Minnich <rminnich@gmail.com>
Reviewed-by: Mathias Krause <minipli@googlemail.com>
Kyösti Mälkki