The PCI_INTR table is an Index/Data pair of I/O ports
0xC00 and 0xC01. This table is responsible for physically
routing IRQs to the PIC and IOAPIC. The settings given
in this table are chipset and mainboard dependent, so the
table values will reside in the mainboard.c file. This
allows for a system to uniquely set its IRQ routing.
The function to write the PCI_INTR table resides in
cimx_util.c because the indices into the table have
the same definitions for all SBx00 FCH chipsets.
The next piece is a function that will read the PCI_INTR
table and program the INT_LINE and INT_PIN registers in
PCI config space appropriately. This function will read
a devices' INT_PIN register, which is always hardcoded to
a value if it uses hardware interrupts. It then uses this
value, along with the device and function numbers to
determine an index into the PCI_INTR table. It will read
the table and program the corresponding value into the PCI
config space register 0x3C, INT_LINE. Finally, it will set
this IRQ number to LEVEL_TRIGGERED on the PIC because it is
a PCI device interrupt and the must be level triggered.
For example, the SB800 USB EHCI device 0:18.2 has an INT_PIN
value hardcoded to 2. This corresponds to PIN B. On the
Persimmon mainboard, I want the USB device to use IRQ 11. I
will program the PCI_INTR table at index 0x31 (this USB device
index) to 11. This function will then read the INT_PIN register,
read the PCI_INTR table, and then program the INT_LINE register
with the value it read. It will then set the IRQ on the PIC to
LEVEL_TRIGGERED by writing a 1 to I/O port 0x4D1 at bit position 4.
Also, the SB700 has slightly different register definitions than
the newer SB800 and SB900 so it needs its own set of #defines for
the pci_intr registers.
Only the Persimmon mainboard is adapted to this change as an
example for other mainboards.
Change-Id: I6de858289a17fa1e1abacf6328ea5099be74b1d6
Signed-off-by: Mike Loptien <mike.loptien@se-eng.com>
Reviewed-on: http://review.coreboot.org/5877
Tested-by: build bot (Jenkins)
Reviewed-by: Kyösti Mälkki <kyosti.malkki@gmail.com>
Reviewed-by: Edward O'Callaghan <eocallaghan@alterapraxis.com>
Without this change the IOAPIC memory window would collide
with PCI config space. This was tested on the hp/abm board.
Change-Id: I5dd53463961f75bab80a41dc7beff8d0434b24ae
Signed-off-by: Dave Frodin <dave.frodin@se-eng.com>
Reviewed-on: http://review.coreboot.org/5946
Tested-by: build bot (Jenkins)
Reviewed-by: Kyösti Mälkki <kyosti.malkki@gmail.com>
We should configure i8254/i8259 down in to the southbridge rather than
romstage of every AGESA/CIMx board much like Intel boards do.
Change-Id: Id7c4f0baa0819d52aef9b0ee03c20d0fa16b9352
Signed-off-by: Edward O'Callaghan <eocallaghan@alterapraxis.com>
Reviewed-on: http://review.coreboot.org/5669
Tested-by: build bot (Jenkins)
Reviewed-by: Kyösti Mälkki <kyosti.malkki@gmail.com>
The ACPI IO ports, and the respective SMI (for HAVE_SMI_HANDLER), were
initialized when the FADT table was written. This works well on a cold
boot, but the ACPI ports are not initialized on S3 resume, as ACPI
tables are not written. This will not work on S3 resume if the default
ports are not what we set them, or if AGESA sets them to some other
value.
To solve this, move the port configuration to southbridge chip init.
Change-Id: Ib4043f0fa5e20f08d320acd12ce84d4d789cd035
Signed-off-by: Alexandru Gagniuc <mr.nuke.me@gmail.com>
Reviewed-on: http://review.coreboot.org/5559
Tested-by: build bot (Jenkins)
Reviewed-by: Aaron Durbin <adurbin@gmail.com>
The power button was declared by hudson's ASL as \_SB.PCI0.PWRB, and
always had the wake source declared as GPE3. This is not the correct
wake source for all boards. On some laptops declaring a wake source is
not needed, as the wake mechanism is handled by the EC.
Move the declaration of the power button to mainboard ASL files, and
scope it as \_SB.PWRB . This also makes the naming consistent with the
examples in the ACPI spec. The wake source for the PWRB of HP Pavilion
M6 1035dx is removed, as it is incorrect.
Change-Id: I9c76566025e7f200c0376673f6c6ea299afa4a5d
Signed-off-by: Alexandru Gagniuc <mr.nuke.me@gmail.com>
Reviewed-on: http://review.coreboot.org/5546
Tested-by: build bot (Jenkins)
Reviewed-by: Aaron Durbin <adurbin@gmail.com>
The ACPI IO ports were defined twice, and used inconsistently. Only
keep one of the definitions for consistency.
Change-Id: If5744f9375fdaa97ceb9ba03dca8aa825eecf159
Signed-off-by: Alexandru Gagniuc <mr.nuke.me@gmail.com>
Reviewed-on: http://review.coreboot.org/5558
Tested-by: build bot (Jenkins)
Reviewed-by: Kyösti Mälkki <kyosti.malkki@gmail.com>
The SPI controller driver used numerical offsets to access SPI
registers, making it unreadable without the datasheet. Use less magic
and more #defines to improve readability.
Change-Id: I8a1f11645cfce027e5df7a41a98c70249695889e
Signed-off-by: Alexandru Gagniuc <mr.nuke.me@gmail.com>
Reviewed-on: http://review.coreboot.org/5557
Tested-by: build bot (Jenkins)
Reviewed-by: Kyösti Mälkki <kyosti.malkki@gmail.com>
Following boards use cimx/sb700:
amd/dinar
supermicro/h8qgi
supermicro/h8scm
tyan/s8226
Only amd/dinar had APIC_ID_OFFSET defined, thus all had 0x0.
There was a nonsense preprocessor directive (MAX_CPUS * MAX_PHYSICAL_CPUS >= 1).
Except for tyan, (MAX_CPUS * MAX_PHYSICAL_CPUS) % 256 == 0.
Together with documented 4-bit restriction for APIC ID field, this APIC ID
programming matches with MP tables and ACPI tables.
I believe this would also fix cases of cimx/sb700 with MAX_CPUS<16, which
we do not have in the tree.
Change-Id: If8d65e95788ba02fc8d331a7af03a4d0d8cf5c69
Signed-off-by: Kyösti Mälkki <kyosti.malkki@gmail.com>
Reviewed-on: http://review.coreboot.org/5539
Tested-by: build bot (Jenkins)
Reviewed-by: Alexandru Gagniuc <mr.nuke.me@gmail.com>
Match the definition of NODE_PCI() with get_node_pci(), so romstage
and ramstage agree of the PCI BDFs for nodes.
Note that all board have CONFIG_CDB = 0x18 and the maximum for
nodes = 8, so we always have (CONFIG_CDB + x) < 32.
Change-Id: I676ee53a65ef5b1243df2c5889577dd987c8fc9c
Signed-off-by: Kyösti Mälkki <kyosti.malkki@gmail.com>
Reviewed-on: http://review.coreboot.org/5536
Tested-by: build bot (Jenkins)
Reviewed-by: Alexandru Gagniuc <mr.nuke.me@gmail.com>
Refactor hudson_enable_gevent_smi() to allow configuring the interrupt
mode and trigger level. Move the utilities which are useful in SMM to
a separate file that is included in both ramstage and SMM. This is
useful for SMI handlers which need to enable or disable GEVENT SMIs
on-the-fly. A follow-up patch makes use of this infrastructure.
Change-Id: Ifa4c300c00c178b18d7280690cfc4b8367c669b8
Signed-off-by: Alexandru Gagniuc <mr.nuke.me@gmail.com>
Reviewed-on: http://review.coreboot.org/170
Tested-by: build bot (Jenkins)
Reviewed-by: Aaron Durbin <adurbin@gmail.com>
This function isn't used on hudson, and seems to be copy-paste from
older southbridges. It is used in sb700 to enable or disable certain
PCI devices. On hudson, these configuration bits are moved to the PM
space.
Change-Id: I9b967a2d0a5dddc8341204dadeed90460251915c
Signed-off-by: Alexandru Gagniuc <mr.nuke.me@gmail.com>
Reviewed-on: http://review.coreboot.org/5513
Tested-by: build bot (Jenkins)
Reviewed-by: Aaron Durbin <adurbin@gmail.com>
This enables the ACPI SMI command port in the FADT table, and sets up
the hardware accordingly. If we have SMI enabled, then we don't set
the SCI_EN bit at boot, causing the OS to send the ACPI_ENABLE
command, as required by the ACPI spec. This gives us a chance to hook
into the mainboard_smi_apmc() handler.
Change-Id: Ib4c63d55b3132578dcae48bfe2092d4ea35821dd
Signed-off-by: Alexandru Gagniuc <mr.nuke.me@gmail.com>
Reviewed-on: http://review.coreboot.org/5511
Tested-by: build bot (Jenkins)
Reviewed-by: Aaron Durbin <adurbin@gmail.com>
This sets up the infrastructure to handle SMIs generated by the Hudson
southbridge. An API for interfacing to mainboard handlers is not
defined at this point. A few functions are defined to allow mainboard
code to enable SMIs from GEVENT pins. These are the only functions which
I expect to be needed anytime in the foreseeable future.
SMIs are always acknowledged and cleared, as not clearing an SMI will
cause us to re-enter the SMI, effectively bricking the machine if a
southbridge-generated SMI without a handler occurs.
Change-Id: Ibceb21ac5423eb134d3eb7d24800280b183f7619
Signed-off-by: Alexandru Gagniuc <mr.nuke.me@gmail.com>
Reviewed-on: http://review.coreboot.org/5494
Reviewed-by: Aaron Durbin <adurbin@gmail.com>
Tested-by: build bot (Jenkins)
The MMIO region is set up by AGESA very early on, so we can use it to
access the PM register space in ramstage. 16-bit accessors are also
provided to simplify some setup tasks. 16-bit accesses are not
possible via PIO.
The pm2_iowrite/read accessors are removed, as they are not used.
Change-Id: Ie7967b5086eb004525c39721338c6495aedc8165
Signed-off-by: Alexandru Gagniuc <mr.nuke.me@gmail.com>
Reviewed-on: http://review.coreboot.org/5503
Tested-by: build bot (Jenkins)
Reviewed-by: Aaron Durbin <adurbin@gmail.com>
Just like in commit
* 1d87dac hp/pavilion_m6_1035dx: Sanitize #includes
Include AGESA headers specifying the path relative to AGESA_ROOT. The
path is specified relative to AGESA_ROOT as opposed to src/ since this
code may include headers from different AGESA families, depending on
the board.
Change-Id: Ide38cc34e207a8b617d1d319fd9c17a785f55833
Signed-off-by: Alexandru Gagniuc <mr.nuke.me@gmail.com>
Reviewed-on: http://review.coreboot.org/5423
Tested-by: build bot (Jenkins)
Reviewed-by: Edward O'Callaghan <eocallaghan@alterapraxis.com>
Reviewed-by: Idwer Vollering <vidwer@gmail.com>
Not all boards which use the AMD Hudson southbridge have IDE. However,
the southbridge's asl included an 'ide.asl' file which had to be
present in $(mainboard_dir)/acpi.
Address this issue by removing the inclusion of 'ide.asl' from the
southbridge 'fch.asl' and remove 'ide.asl' from Hudson boards, none
of which have IDE.
If future hudosn board will come with IDE, the device can be declared
in the PCIO scope of dsdt.asl, right below the inclusion of 'fch.asl'.
Change-Id: Ie2efb7ebf8f5b527e26d7aaaeafbd3053a9a6b28
Signed-off-by: Alexandru Gagniuc <mr.nuke.me@gmail.com>
Reviewed-on: http://review.coreboot.org/5459
Tested-by: build bot (Jenkins)
Reviewed-by: Edward O'Callaghan <eocallaghan@alterapraxis.com>
Reviewed-by: Patrick Georgi <patrick@georgi-clan.de>
Follow along hudson, cut out "SLP_TYP type was 0" excessively filling
the buffer. We could make this conditional on non-zero?
Change-Id: Iffd4c146b2ac4f57dbc3a011a683c92b6e132e39
Signed-off-by: Edward O'Callaghan <eocallaghan@alterapraxis.com>
Reviewed-on: http://review.coreboot.org/5495
Tested-by: build bot (Jenkins)
Reviewed-by: Kyösti Mälkki <kyosti.malkki@gmail.com>
Not all boards which use the AMD cimx/sb800 southbridge have IDE.
However, the southbridge's asl included an 'ide.asl' file which had to
be present in $(mainboard_dir)/acpi.
Address this issue by including ide.asl only in boards which have IDE,
and remove it from all other cimx/sb800 boards.
Change-Id: I57fcb4db9f85234b05ae1705ef81a576c478cee6
Signed-off-by: Edward O'Callaghan <eocallaghan@alterapraxis.com>
Reviewed-on: http://review.coreboot.org/5460
Tested-by: build bot (Jenkins)
Reviewed-by: Patrick Georgi <patrick@georgi-clan.de>
Removing `-Wno-unused-but-set-variable` from `CFLAGS` results in the error
below, when building for example the HP DL145 GL1.
CC southbridge/amd/amd8111/acpi.ramstage.o
src/southbridge/amd/amd8111/acpi.c: In function 'acpi_init':
src/southbridge/amd/amd8111/acpi.c💯11: error: variable 'dword' set but not used [-Werror=unused-but-set-variable]
Removing the variable `dword` fixes this error.
The read is left in the code, as I do not know if it has an effect or
not.
Change-Id: I9957cef3a996c5974c275423c9de63ccf230974e
Signed-off-by: Paul Menzel <paulepanter@users.sourceforge.net>
Reviewed-on: http://review.coreboot.org/5315
Tested-by: build bot (Jenkins)
Reviewed-by: Patrick Georgi <patrick@georgi-clan.de>
The previous SBxxx generations were setting up LPC bridge based
on the PNP resources. Implement it also for AGESA Hudson.
The AGESA itself opens one big region DFLT_SIO_PME_BASE_ADDRESS
(512 bytes). Make the code smart enough to detect already used
region and if any resource fits into AGESA defined region, do nothing.
Change-Id: I718d034bc4c778697a7bd0506d4550c8f5a43159
Signed-off-by: Rudolf Marek <r.marek@assembler.cz>
Reviewed-on: http://review.coreboot.org/4497
Tested-by: build bot (Jenkins)
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
Reviewed-by: Ronald G. Minnich <rminnich@gmail.com>
It is inappropriate for chipset code to be redefining
types -- especially NULL to a non-pointer type. There's
only one non-straight forward change. A condition
being checked was '!ptr_type == NULL' (0 as int). That
check is actually 'ptr_type != NULL'.
Change-Id: Iab5733e5a573baba6fec94e0c955ba4fad72c836
Signed-off-by: Aaron Durbin <adurbin@chromium.org>
Reviewed-on: http://review.coreboot.org/5088
Tested-by: build bot (Jenkins)
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
Reviewed-by: Patrick Georgi <patrick@georgi-clan.de>
Also relocate and split header files, there is some interest
for EHCI debug support without PCI.
Change-Id: Ibe91730eb72dfe0634fb38bdd184043495e2fb08
Signed-off-by: Kyösti Mälkki <kyosti.malkki@gmail.com>
Reviewed-on: http://review.coreboot.org/5129
Tested-by: build bot (Jenkins)
Reviewed-by: Aaron Durbin <adurbin@google.com>
Broken with/since commit d1cb0eec.
Original intention was to set the frequency for 'Fast Read' command
in bits 15..14, and enable 'Fast Read' command.
Modified register contains SPI frequency for 'Normal Read' command
in bits 13..12. Default for this is 11b for 16.5 MHz. Existing code
unintentionally clears these bits, increasing SPI frequency to 66MHz
for 'Normal Read' command.
This is above specifications for many common SPI flash components
and also makes flashrom older than 0.9.7-r1750 to operate unreliably
on read/write/erase for these platforms.
Change-Id: I30109e2a0410c0bb0bdc968ea71787396b32e761
Signed-off-by: Kyösti Mälkki <kyosti.malkki@gmail.com>
Reviewed-on: http://review.coreboot.org/5089
Tested-by: build bot (Jenkins)
Reviewed-by: Kevin O'Connor <kevin@koconnor.net>
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
Reviewed-by: Marc Jones <marc.jones@se-eng.com>
imc_reg_init: init fan control related registers.
enable_imc_thermal_zone: AGESA does not enable thermal zone. We enable
it here.
Change-Id: I93c729982d78b6d2c7c20bcb1a3e27a7dd0eba91
Signed-off-by: WANG Siyuan <SiYuan.Wang@amd.com>
Signed-off-by: WANG Siyuan <wangsiyuanbuaa@gmail.com>
Reviewed-on: http://review.coreboot.org/4300
Tested-by: build bot (Jenkins)
Reviewed-by: Rudolf Marek <r.marek@assembler.cz>
Show POST codes on a PCI device: implement hudson_pci_port80().
Remove the comments that use pci_locate_device():
using the code found in the comment seems to break booting.
This shares much code with sb600/sb700/sb800,
however the deduplication work needs to be discusses somewhere else
than in this review board.
Tested on an Asus F2A85-M.
The contribution is (C) by Rudolf Marek.
Change-Id: I54fb1dcb0614452c775ed70d867ab44ff263a61a
Author: Rudolf Marek <r.marek@assembler.cz>
Signed-off-by: Idwer Vollering <vidwer@gmail.com>
Reviewed-on: http://review.coreboot.org/4559
Tested-by: build bot (Jenkins)
Reviewed-by: Rudolf Marek <r.marek@assembler.cz>
Options for selecting the USB port and controller for usbdebug
were unintentionally hidden with commit 8232bc2c on AGESA platforms
using cimx/sb700 or cimx/sb800.
Change-Id: Ibacc81a580519fe7fa86f08374046625327340b4
Signed-off-by: Kyösti Mälkki <kyosti.malkki@gmail.com>
Reviewed-on: http://review.coreboot.org/4607
Tested-by: build bot (Jenkins)
Reviewed-by: Aaron Durbin <adurbin@google.com>
AMD fam10 raminit cannot be built without RAMINIT_SYSINFO, this
is not a true option but copy-paste remainder from AMD K8.
Change-Id: Id8edc112f3bacebd1732304ac9ee6e77cc6263b7
Signed-off-by: Kyösti Mälkki <kyosti.malkki@gmail.com>
Reviewed-on: http://review.coreboot.org/4581
Tested-by: build bot (Jenkins)
Reviewed-by: Patrick Georgi <patrick@georgi-clan.de>
Reviewed-by: Alexandru Gagniuc <mr.nuke.me@gmail.com>
The main purpose of option rom is to supply int* handlers.
But supplying those is outside of coreboot scope and if someone needs those
they should run SeaBIOS anyway which runs the option roms wonderfully.
Running VGA oprom is kept because they're needed to init graphics.
This patch still keeps the options to include the option roms to make them
available to SeaBIOS.
Change-Id: I646334cf88094d3bf8f527779a68a07e0b4b93ec
Signed-off-by: Vladimir Serbinenko <phcoder@gmail.com>
Reviewed-on: http://review.coreboot.org/4545
Tested-by: build bot (Jenkins)
Reviewed-by: Patrick Georgi <patrick@georgi-clan.de>
Reviewed-by: Kevin O'Connor <kevin@koconnor.net>
Do not directly check the return value of get_option, but instead compare
the returned value against a CB_CMOS_ error code, or against CB_SUCCESS.
Change-Id: I2fa7761d13ebb5e9b4606076991a43f18ae370ad
Signed-off-by: Alexandru Gagniuc <mr.nuke.me@gmail.com>
Reviewed-on: http://review.coreboot.org/4266
Tested-by: build bot (Jenkins)
Reviewed-by: Stefan Reinauer <stefan.reinauer@coreboot.org>
Besides the AGESA static settings, the settings in mainboard/buildOpt.c also
change the final configuration. We need to make sure the settings in FchParam
in resume stage are the same as they were in cold boot stage, otherwise the
board can not wake up more than once.
Tested on AMD/Olive Hill, AMD/Parmer and ASRock/imb-a180.
(USB keyboard doesn't work when board wakes up. It is not introduced by this
patch. It needs more debugging.)
Change-Id: I5a5e5502080e358ffc3577dc6a40bb762844d998
Signed-off-by: Zheng Bao <zheng.bao@amd.com>
Signed-off-by: Zheng Bao <fishbaozi@gmail.com>
Reviewed-on: http://review.coreboot.org/3932
Tested-by: build bot (Jenkins)
Reviewed-by: Rudolf Marek <r.marek@assembler.cz>
The main usbdebug file lib/usbdebug.c was removed from romstage
build with commit f8bf5a10 but the chipset-specific parts were not,
leading to unresolved symbol errors for AMD platforms.
Add a silent Kconfig variable USBDEBUG_IN_ROMSTAGE for convenient
use of this feature.
Change-Id: I0cd3fccf2612cf08497aa5c3750c89bf43ff69be
Signed-off-by: Kyösti Mälkki <kyosti.malkki@gmail.com>
Reviewed-on: http://review.coreboot.org/3983
Tested-by: build bot (Jenkins)
Reviewed-by: Aaron Durbin <adurbin@google.com>
Chipsets sb700 and sb800/hudson have more than one USB EHCI controller,
implement the selection logic using already existing Kconfig option.
Change-Id: I9e0df1669d73863c95c36a3a7fee40d58f6f097e
Signed-off-by: Kyösti Mälkki <kyosti.malkki@gmail.com>
Reviewed-on: http://review.coreboot.org/3928
Tested-by: build bot (Jenkins)
Reviewed-by: Stefan Reinauer <stefan.reinauer@coreboot.org>
Northbridge code includes these headers, so they all need to
have the same name to allow different combinations of northbridge
and southbridge. This changes the sb900 names to match sb700 &
sb800, and points agesa/family12 and amd/torpedo to the new file
names.
Change-Id: I7a654ce9ae591a636a56177f64fb8cb953b4b04f
Signed-off-by: Corey Osgood <corey.osgood@gmail.com>
Reviewed-on: http://review.coreboot.org/3825
Tested-by: build bot (Jenkins)
Reviewed-by: Patrick Georgi <patrick@georgi-clan.de>
AMD northbridges have a complex way to resolve top_of_ram.
Once it is resolved, it is stored in NVRAM to be used on resume.
TODO: Redesign these get_top_of_ram() functions from scratch.
Change-Id: I3cceb7e9b8b07620dacf138e99f98dc818c65341
Signed-off-by: Kyösti Mälkki <kyosti.malkki@gmail.com>
Reviewed-on: http://review.coreboot.org/3557
Tested-by: build bot (Jenkins)
Reviewed-by: Aaron Durbin <adurbin@google.com>
Nowadays, chipsets or boards do not only have one USB port with the
capabilities of a debug port but several ones. Some of these ports are
easier accessible than others, so making them configurable is also necessary.
This change adds infrastructure to switch between EHCI controllers,
but does not implement it for any chipset.
Change-Id: I079643870104fbc64091a54e1bfd56ad24422c9f
Signed-off-by: Kyösti Mälkki <kyosti.malkki@gmail.com>
Reviewed-on: http://review.coreboot.org/3438
Tested-by: build bot (Jenkins)
Reviewed-by: Aaron Durbin <adurbin@google.com>
