The replacement function confirms CBMEM TOC is wiped clean on power
cycles and resets. It also introduces compatibility interface to ease
up transition to DYNAMIC_CBMEM.
Change-Id: Ic5445c5bff4aff22a43821f3064f2df458b9f250
Signed-off-by: Kyösti Mälkki <kyosti.malkki@gmail.com>
Reviewed-on: http://review.coreboot.org/4668
Reviewed-by: Aaron Durbin <adurbin@google.com>
Tested-by: build bot (Jenkins)
Now it boots up to message "Could not find payload".
Change-Id: I07ddca7046492f7e0dec15a8ea00c2870b09ee67
Signed-off-by: Vladimir Serbinenko <phcoder@gmail.com>
Reviewed-on: http://review.coreboot.org/4754
Tested-by: build bot (Jenkins)
Reviewed-by: Alexandru Gagniuc <mr.nuke.me@gmail.com>
Per our discussions with Gerd, qemu will now always do native graphics
on coreboot. The VGA BIOS capability is not needed and will no longer
be supported. Attempts to build without native graphics will result in
an error.
This code builds for both x86 emulation targets. I'm hitting an issue
testing that is unrelated to coreboot; if someone can test, that
would be helpful. Be sure to start qemu with -vga std.
We also add a test for the PCI BAR being zero and return silently if it
is.
Change-Id: I66188f61e1bac7ad93c989cc10f3e0b55140e148
Signed-off-by: Ronald G. Minnich <rminnich@google.com>
Reviewed-on: http://review.coreboot.org/4258
Tested-by: build bot (Jenkins)
Reviewed-by: Vladimir Serbinenko <phcoder@gmail.com>
Reviewed-by: Gerd Hoffmann <kraxel@redhat.com>
Qemu makes the guest uuid (qemu -uuid $uuid) available
to the guest via fw_cfg. Other smbios fields can be
configured in qemu using the -smbios command line
switch (check the qemu manpage for details).
This patch adds coreboot support for this, so the
values provided by qemu will actually show up in the
smbios table.
Change-Id: Ifd9ae0d02749af4e7070a65eadbd1a9585a8a8e6
Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
Reviewed-on: http://review.coreboot.org/4086
Tested-by: build bot (Jenkins)
Reviewed-by: Patrick Georgi <patrick@georgi-clan.de>
Reviewed-by: Ronald G. Minnich <rminnich@gmail.com>
Starting with release 1.7 qemu provides acpi tables via fw_cfg. Main
advantage is that new (virtual) hardware which needs acpi support
JustWorks[tm] without having to patch & update the firmware (seabios,
coreboot, ...) accordingly.
So if we find acpi tables in fw_cfg try loading them, otherwise fallback
to the builtin acpi tables.
Change-Id: I792232829b870ff6ed8414a3007e0af17f6c4223
Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
Reviewed-on: http://review.coreboot.org/4040
Tested-by: build bot (Jenkins)
qemu 1.7+ provides a fw_cfg file named "etc/e820" with e820-like entries
for reservations and ram regions. Use it for ram detection if present,
otherwise fallback to the traditional cmos method.
Change-Id: Icac6c99d2a053e59dfdd28e48d1ceb3d56a61bdc
Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
Reviewed-on: http://review.coreboot.org/4030
Tested-by: build bot (Jenkins)
Qemu can provide files using the firmware config interface.
This is used to pass config options, virtual machine config
info and option roms into the guest.
This patch adds support for reading the file index and loading
files from qemu.
Change-Id: I57d4a734527c4117239f355121cf1fb8a390ab0d
Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
Reviewed-on: http://review.coreboot.org/4029
Tested-by: build bot (Jenkins)
It is expected this will always be a casted get_top_of_ram() call
on x86, no reason to do that under chipset.
Change-Id: I3a49abe13ca44bf4ca1e26d1b3baf954bc5a29b7
Signed-off-by: Kyösti Mälkki <kyosti.malkki@gmail.com>
Reviewed-on: http://review.coreboot.org/3972
Tested-by: build bot (Jenkins)
Reviewed-by: Aaron Durbin <adurbin@google.com>
This reverts commit de1fe7f655.
While things appeared to work, there were actually invalid references
to CAR storage after CAR was torn down on boards without
EARLY_CBMEM_INIT. It was discussed use of CAR_GLOBAL should be
restricted to boards that handle CAR migration properly.
Change-Id: I9969d2ea79c334a7f95a0dbb7c78065720e6ccae
Signed-off-by: Kyösti Mälkki <kyosti.malkki@gmail.com>
Reviewed-on: http://review.coreboot.org/3968
Tested-by: build bot (Jenkins)
Reviewed-by: Aaron Durbin <adurbin@google.com>
If romstage does not make cbmem_initialize() call, linker should
optimize the code for CAR migration away.
This simplifies design of CBMEM console by a considerable amount.
As console buffer is now migrated within cbmem_initialize() call there
is no longer need for cbmemc_reinit() call made at end of romstage.
Change-Id: I8675ecaafb641fa02675e9ba3f374caa8e240f1d
Signed-off-by: Kyösti Mälkki <kyosti.malkki@gmail.com>
Reviewed-on: http://review.coreboot.org/3916
Tested-by: build bot (Jenkins)
Reviewed-by: Aaron Durbin <adurbin@google.com>
Assume EARLY_CBMEM_INIT=y everywhere and remove option from Kconfig.
If romstage does not make the cbmem_initialize() call, features like
COLLECT_TIMESTAMPS and early CBMEM_CONSOLE will execute during
romstage, but that data will get lost as no CAR migration is
executed.
Change-Id: I5615645ed0f5fd78fbc372cf5c3da71a3134dd85
Signed-off-by: Kyösti Mälkki <kyosti.malkki@gmail.com>
Reviewed-on: http://review.coreboot.org/3917
Tested-by: build bot (Jenkins)
Reviewed-by: Aaron Durbin <adurbin@google.com>
Without this coreboot may (depends on the amount of memory) place the
pci bars below 0xb0000000, then the linux kernel goes move them around
so they are inside the window declared in the acpi tables.
This breaks vesafb as the vga framebuffer gets moved after vgabios
initialization. It's also not exactly nice to expect the OS fix our
mess ;)
Change-Id: If6b50ea863958eea71b567ccb7a06c6a28076111
Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
Reviewed-on: http://review.coreboot.org/3927
Tested-by: build bot (Jenkins)
Reviewed-by: Ronald G. Minnich <rminnich@gmail.com>
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
Use the new helper function set_top_of_ram() to remove remaining
uses of high_tables_base and _size under northbridge/.
Change-Id: I6b0d9615002ed2aff578c5811d7bd43dd2594453
Signed-off-by: Kyösti Mälkki <kyosti.malkki@gmail.com>
Reviewed-on: http://review.coreboot.org/3561
Tested-by: build bot (Jenkins)
Reviewed-by: Stefan Reinauer <stefan.reinauer@coreboot.org>
Reviewed-by: Aaron Durbin <adurbin@google.com>
The slightly hackish ioapic ressource reservation is needed for i440fx
emulation only, for q35 the ich9 southbridge driver handles this just
fine.
[ Side note: The i440fx chipset emulated by qemu is pimped up with alot
of stuff which never existed on real hardware, which leads
to tweaks like this one. ]
Change-Id: I06bf54cbc247ccf17aa9063fb7dee9def323c605
Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
Reviewed-on: http://review.coreboot.org/3850
Tested-by: build bot (Jenkins)
Reviewed-by: Patrick Georgi <patrick@georgi-clan.de>
QEMU has a bunch of non-standard virtual devices on various I/O ports.
Allocate resources for them so the coreboot resource management knows
those ports are used.
Change-Id: I51a85967cf2dcd634b0c883210bb52c0c34c8283
Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
Reviewed-on: http://review.coreboot.org/3851
Tested-by: build bot (Jenkins)
Reviewed-by: Ronald G. Minnich <rminnich@gmail.com>
Change all PCI configuration accesses to MMIO in qemu-q35
emulation
To enable MMIO style access, add (move) explicit PCI IO config write
in the bootblock. As there is no northbridge/x/x/bootblock.c
file, a mainboard/x/x/bootblock.c file is added for this purpose.
Change-Id: I979efb3d9b2f359a9ccbd1b4f6c05f83bab43007
Signed-off-by: Kyösti Mälkki <kyosti.malkki@gmail.com>
Reviewed-on: http://review.coreboot.org/3599
Tested-by: build bot (Jenkins)
Reviewed-by: Ronald G. Minnich <rminnich@gmail.com>
PCI bus operations are static through the ramstage, and should be
initialized from the very beginning. For all the replaced instances,
there is no MMCONF_SUPPORT nor MMCONF_SUPPORT_DEFAULT selected for
the northbridge, so these continue to use PCI IO config access.
Change-Id: I658abd4a02aa70ad4c9273568eb5560c6e572fb1
Signed-off-by: Kyösti Mälkki <kyosti.malkki@gmail.com>
Reviewed-on: http://review.coreboot.org/3607
Tested-by: build bot (Jenkins)
Reviewed-by: Stefan Reinauer <stefan.reinauer@coreboot.org>
It might be that you want an early console in romstage before RAM is up, but
you can't or don't want to support the console all the way back in the
bootblock. By making the console in those two different environments
configurable seperately that becomes possible.
On the 5250 console output as early as the bootblock works, but on the 5420 it
only starts working in the ROM stage after clocks have been initialized.
Change-Id: I68ae3fcb4d828fa8a328a30001c23c81a4423bb8
Signed-off-by: Gabe Black <gabeblack@chromium.org>
Reviewed-on: http://review.coreboot.org/3671
Tested-by: build bot (Jenkins)
Reviewed-by: Stefan Reinauer <stefan.reinauer@coreboot.org>
Not all ARM systems need "BL1", and the layout of BL* and bootblock may be
different (ex, Exynos 5250 may use a new BL1 with variable length checksum
header).
To support that better, define the real base address (and ROM offset) of boot
block, and then we can post-processing ROM image file by filling data / checksum
and any other information.
Change-Id: I0e3105e52500b6b457371ad33a9aa546acf28928
Signed-off-by: Hung-Te Lin <hungte@chromium.org>
Signed-off-by: Gabe Black <gabeblack@chromium.org>
Reviewed-on: http://review.coreboot.org/3664
Tested-by: build bot (Jenkins)
Reviewed-by: Stefan Reinauer <stefan.reinauer@coreboot.org>
When starting the Exynos5250 port, a lot of unneeded u-boot code
was imported. This is an attempt to get rid of a lot of unneeded
code before the port is used as a basis for further ARM ports.
There is a lot more that can be done, including cleaning up the
5250's Kconfig file.
Change-Id: I2d88676c436eea4b21bcb62f40018af9fabb3016
Signed-off-by: Stefan Reinauer <reinauer@google.com>
Signed-off-by: Gabe Black <gabeblack@chromium.org>
Reviewed-on: http://review.coreboot.org/3642
Tested-by: build bot (Jenkins)
Reviewed-by: Stefan Reinauer <stefan.reinauer@coreboot.org>
In ram stage, all code flow should be tied to the resource allocator.
Stuff that has to happen before everything else goes into the mainboard
enable function in mainboard.c. This patch empties the main() wrapper
around hardwaremain.c, allowing to get rid of this special case in the
ARM port.
Change-Id: Ide91a23f1043b64acf64471f180a2297f0f40d97
Signed-off-by: Stefan Reinauer <reinauer@google.com>
Signed-off-by: Gabe Black <gabeblack@chromium.org>
Reviewed-on: http://review.coreboot.org/3638
Tested-by: build bot (Jenkins)
Reviewed-by: Stefan Reinauer <stefan.reinauer@coreboot.org>
In commit Rudolf Marek discovered, that it is not uniformly written. As
»ASL names are not case-sensitive and will be converted to upper case.« [2]
this change does not have any functional change.
The following command was used to create this patch.
$ git grep -l 'package()' src/mainboard | xargs sed -i 's,package(),Package(),'
[1] http://review.coreboot.org/#/c/3318/
[2] http://www.acpi.info/spec40a.htm
(18.2.1 ASL Names)
Change-Id: I1784dbc50936a1ef9d4376209a3c324ef1fb85cf
Signed-off-by: Paul Menzel <paulepanter@users.sourceforge.net>
Reviewed-on: http://review.coreboot.org/3516
Tested-by: build bot (Jenkins)
Reviewed-by: Marc Jones <marc.jones@se-eng.com>
This commit was tested on qemu with and without CONFIG_DYNAMIC_CBMEM
by running cmbmem -c once booted. The qemu command that was used was:
qemu-system-i386 -bios ./build/coreboot.rom -serial stdio -hda ../virt/parabola.img
Note that using CONFIG_RELOCATABLE_RAMSTAGE make it fails like that:
Loading image.
CBFS: Decompressing stage fallback/coreboot_ram @ 0x3ffbefc0 (184400 bytes)
Loading module at 3ffbf000 with entry 3ffbf000. filesize: 0x18db8 memsize: 0x2c050
Processing 1703 relocs with adjust value of 0x3ffbe000
FATAL: Essential component is missing.
However without CONFIG_RELOCATABLE_RAMSTAGE set it boots fine.
Change-Id: I633a8c3832eee4e8bed244940fdc370b98dd26f0
Signed-off-by: Denis 'GNUtoo' Carikli <GNUtoo@no-log.org>
Reviewed-on: http://review.coreboot.org/3504
Reviewed-by: Aaron Durbin <adurbin@google.com>
Tested-by: build bot (Jenkins)
So the pci allocation code knows where memory is and doesn't
try map pci devices there. We also don't have to check for
overlaps between pci hole and memory then.
Change-Id: I5eaea0e4d21210719685860fa1f16ca7b2137cde
Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
Reviewed-on: http://review.coreboot.org/3491
Tested-by: build bot (Jenkins)
Reviewed-by: Ronald G. Minnich <rminnich@gmail.com>
Prepare tree for adding q35 support:
Move emulation/qemu-x86 to emulation/qemu-i440fx.
Rename some stuff to include 'i440fx'.
Change-Id: Ib8c58175c5734cfcda1b22404ef52c09d38f0462
Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
Reviewed-on: http://review.coreboot.org/3429
Tested-by: build bot (Jenkins)
Reviewed-by: Ronald G. Minnich <rminnich@gmail.com>
First copy over from SeaBIOS git repo, then adapt for coreboot:
Disable cpu/pci hotplug bits. Disable dynamic pci window.
Both depend on stuff in the SSDT tables created by SeaBIOS.
Bits are left in, but deactivated via #if 0, so it's easier
to see the differences when diffing the coreboot tables with
the SeaBIOS tables.
Adapt dsdt DefinitionBlock.
Enable acpi table generation in acpi_tables.c.
With this patch linux boots successfully with ACPI enabled.
It's not bug-free though. Missing cpu detection leads to
funky messages like this one:
weird, boot CPU (#0) not listed by the BIOS.
and SMP most likely wouldn't work either.
Change-Id: Ic3803a6f1ef6d54c11cc4ca3844d3032a374ae6b
Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
Reviewed-on: http://review.coreboot.org/3342
Reviewed-by: Stefan Reinauer <stefan.reinauer@coreboot.org>
Tested-by: build bot (Jenkins)
Add boot cpu to the device tree. Figure the number of CPUs installed
(using the qemu firmware config interface) and add cpu devices for them,
so they show up in all generated BIOS tables correctly. This gets SMP
going.
Change-Id: I0e99f98942d8ca90150b27fc13c1c7e926a1a644
Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
Reviewed-on: http://review.coreboot.org/3345
Tested-by: build bot (Jenkins)
Reviewed-by: Stefan Reinauer <stefan.reinauer@coreboot.org>
qemu has a special device to pass configuration information
from qemu to the firmware. This patch adds initial support
the interface, namely some infrastructure, detection code and
a function to query the number of CPUs.
Change-Id: I43ff5f4fbf12334a91422aa38f514a82a1d5219e
Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
Reviewed-on: http://review.coreboot.org/3343
Tested-by: build bot (Jenkins)
Reviewed-by: Stefan Reinauer <stefan.reinauer@coreboot.org>
Do the setup for all PCI slots, not only the third.
Also remove the bogus message, as slot 3 may carry
any device, not only NICs.
This makes IRQ setup simliar to SeaBIOS.
SeaBIOS assignments (with patch for logging added,
and a bunch of pci devices for testing purposes):
PCI IRQ [piix]: bdf=00:01.3 pin=1 line=10
PCI IRQ [piix]: bdf=00:03.0 pin=1 line=11
PCI IRQ [piix]: bdf=00:04.0 pin=1 line=11
PCI IRQ [piix]: bdf=00:05.0 pin=1 line=10
PCI IRQ [piix]: bdf=00:06.0 pin=1 line=10
PCI IRQ [piix]: bdf=00:1d.0 pin=1 line=10
PCI IRQ [piix]: bdf=00:1d.1 pin=2 line=10
PCI IRQ [piix]: bdf=00:1d.2 pin=3 line=11
PCI IRQ [piix]: bdf=00:1d.7 pin=4 line=11
Coreboot assignments without this patch:
Assigning IRQ 11 to 0:3.0
Coreboot assignments with this patch:
Assigning IRQ 10 to 0:1.3
Assigning IRQ 11 to 0:3.0
Assigning IRQ 11 to 0:4.0
Assigning IRQ 10 to 0:5.0
Assigning IRQ 10 to 0:6.0
Assigning IRQ 10 to 0:1d.0
Assigning IRQ 10 to 0:1d.1
Assigning IRQ 11 to 0:1d.2
Assigning IRQ 11 to 0:1d.7
Change-Id: Ie96be39185f2f1cbde3c9fc50e29faff59c28493
Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
Reviewed-on: http://review.coreboot.org/3334
Tested-by: build bot (Jenkins)
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
Reviewed-by: Patrick Georgi <patrick@georgi-clan.de>
