- Add SPD_LPDDR4X_200b_1R_16Gb_16Row_DDP_4267.spd.hex, initially
used for the SKhynix H9HCNNNBKMMLXR-NEE part with DRAM ID #2
- Add SPD_LPDDR4X_200b_2R_64Gb_ODP_4267.spd.hex, initially
used for the SKhynix H9HCNNNFAMMLXR-NEE part with DRAM ID #3
- Add SPD_LPDDR4X_200b_2R_32Gb_QDP_4267.spd.hex, initially
used for the Micron MT53E1G32D2NP-046 WT:A part with DRAM ID #4
BUG=b:147857288
TEST=none
Change-Id: I60d8bb05a4d6d3608adc7de69efc8623d1ca610d
Signed-off-by: Nick Vaccaro <nvaccaro@google.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/41126
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
Reviewed-by: Duncan Laurie <dlaurie@chromium.org>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Mainboard specific dock-init mechanism introduced
https://review.coreboot.org/c/coreboot/+/36093 works on most boards,
but https://ticket.coreboot.org/issues/256 shows that some boards
(e.g. x201 and t410) need communication with h8 EC to enable or
disable dock, (in dock_connect() and dock_disconnect() respectively)
so they must be done after the h8 EC is brought up, which is not
garanteed in the above mainboard specific dock-init mechanism.
This time, a hook function h8_mb_init() will be called at the end of
h8_enable(). (in place of the ancient h8_mainboard_init_dock() removed
in CB:36093) Its default implementation is a weak empty function, but
could be overrided with a strong one for boards needing to perform
actions which should be done after h8 EC is brought up.
This should fix the regression detected in
https://ticket.coreboot.org/issues/256
Change-Id: I3674fbfeab2ea2cd2a4453a8e77521157d553388
Signed-off-by: Bill XIE <persmule@hardenedlinux.org>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/39708
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Patrick Rudolph <siro@das-labor.org>
This change selects SOC_AMD_COMMON_BLOCK_HAS_ESPI which enables
the capability for using eSPI on Picasso.
Additionally, it also calls espi_setup() and espi_configure_decodes()
if mainboard enables use of eSPI and skips LPC decodes in that case.
BUG=b:153675913,b:154445472
Change-Id: I4876f1bff4305a23e8ccc48a2d0d3b64cdc9703d
Signed-off-by: Furquan Shaikh <furquan@google.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/41075
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Aaron Durbin <adurbin@chromium.org>
Reviewed-by: Raul Rangel <rrangel@chromium.org>
This change adds a helper function espi_setup() which allows SoCs to
configure connection to slave. Most of the configuration is dependent
upon mainboard settings in espi_config done as part of the device
tree. The general flow for setup involves the following steps:
1. Set initial configuration (lowest operating frequency and single mode).
2. Perform in-band reset and set initial configuration since the
settings would be lost by the reset.
3. Read slave capabilities.
4. Set slave configuration based on mainboard settings.
5. Perform eSPI host controller configuration to match the slave
configuration and set polarities for VW interrupts.
6. Perform VW channel setup and deassert PLTRST#.
7. Perform peripheral channel setup.
8. Perform OOB channel setup.
9. Perform flash channel setup.
10. Enable subtractive decoding if requested by mainboard.
BUG=b:153675913
Signed-off-by: Furquan Shaikh <furquan@google.com>
Change-Id: I872ec09cd92e9bb53f22e38d2773f3491355279e
Reviewed-on: https://review.coreboot.org/c/coreboot/+/41272
Reviewed-by: Aaron Durbin <adurbin@chromium.org>
Reviewed-by: Raul Rangel <rrangel@chromium.org>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Memory SPD files for each variant are now stored in the variant's
mb/google/volteer/variants/<variant_name>/spd directory instead
of storing them in mb/google/volteer/spd.
This change moves SPDs to where they are needed and changes the
makefile to look for them in their new locations.
BUG=b:156126658
TEST="emerge-volteer coreboot chromeos-bootimage", flash and boot
a proto2 SKU4 to the kernel.
Change-Id: I759c979027477a2a4c5489a6b12278799488d6e7
Signed-off-by: Nick Vaccaro <nvaccaro@google.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/41184
Reviewed-by: Duncan Laurie <dlaurie@chromium.org>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
To change frequency, the SOC PLL team suggests procedure below:
First, we need to enable the intermediate clock and
switch the ca53 clock source to the intermediate clock.
Second, disable the armpll_ll clock output.
Third, raise armpll_ll frequency and enable the clock output.
The last, switch the ca53 clock source back to armpll_ll and
disable the intermediate clock.
BUG=b:154451241
BRANCH=jacuzzi
TEST=Boots correctly on Jacuzzi.
Signed-off-by: Weiyi Lu <weiyi.lu@mediatek.com>
Change-Id: Ib9556ba340da272fb62588f45851c93373cfa919
Reviewed-on: https://review.coreboot.org/c/coreboot/+/41077
Reviewed-by: Hung-Te Lin <hungte@chromium.org>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
This change updates espi_debug.c to use switch case instead of if-else
for operating frequency and i/o mode prints. This is done to address
the review comments received on CB:41254.
Signed-off-by: Furquan Shaikh <furquan@google.com>
Change-Id: I4f323b79f030818e2daa983d4f17ddf7a3192171
Reviewed-on: https://review.coreboot.org/c/coreboot/+/41346
Reviewed-by: Felix Held <felix-coreboot@felixheld.de>
Reviewed-by: Angel Pons <th3fanbus@gmail.com>
Reviewed-by: Aaron Durbin <adurbin@chromium.org>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
CB:41194 got rid of "this file is part of" lines. However, there are
some changes that landed right around the same time including those
lines. This change uses the following command to drop the lines from
new files:
sed -i -e '/file is part of /d' $(git grep "file is part of " |egrep ":( */\*.*\*/\$|#|;#|-- | *\* )" | cut -d: -f1 |grep -v crossgcc |grep -v gcov | grep -v /elf.h |grep -v nvramtool)
Signed-off-by: Furquan Shaikh <furquan@google.com>
Change-Id: Ic3c1d717416f6b7e946f84748e2b260552c06a1b
Reviewed-on: https://review.coreboot.org/c/coreboot/+/41342
Reviewed-by: Patrick Georgi <pgeorgi@google.com>
Reviewed-by: Raul Rangel <rrangel@chromium.org>
Reviewed-by: Felix Held <felix-coreboot@felixheld.de>
Reviewed-by: Angel Pons <th3fanbus@gmail.com>
Reviewed-by: HAOUAS Elyes <ehaouas@noos.fr>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Immediately following FSP-S, update the data fabric routing
registers to make the region between HPET and LAPIC as non-posted.
If AGESA is modified to do this, we can delete data_fabric_util.c. If
AGESA is modified to not program the registers, then we can simplify
data_fabric_set_mmio_np().
BUG=b:147042464, b:156296146
TEST=boot trembyle
Change-Id: Idbafaac158f5a4c533d2d88db79bb4d6244e5355
Signed-off-by: Marshall Dawson <marshalldawson3rd@gmail.com>
Signed-off-by: Raul E Rangel <rrangel@chromium.org>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/41268
Reviewed-by: Furquan Shaikh <furquan@google.com>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
These are used to setup the data fabric.
Definitions came from 55570-B1 Rev 3.14 - PPR for AMD Family 17h Model 18h
BUG=b:147042464
Signed-off-by: Raul E Rangel <rrangel@chromium.org>
Change-Id: Ib51f6e2fd304da9948d6625608af71f25b974854
Reviewed-on: https://review.coreboot.org/c/coreboot/+/41266
Reviewed-by: Furquan Shaikh <furquan@google.com>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Family 17h devices are designed with a new internal architecture,
frequently referred to as the data fabric. Although designed to
behave somewhat like the older integrated northbridge designs,
the D18Fx definitions are completely new.
The previous northbridge.c was copied from stoneyridge which is
completely different.
Change-Id: Id70cbda99657249179fb8cf5e461dd6a37ec9153
Signed-off-by: Raul E Rangel <rrangel@chromium.org>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/41265
Reviewed-by: Furquan Shaikh <furquan@google.com>
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
Reviewed-by: Aaron Durbin <adurbin@chromium.org>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
This is a device specific register, not a northbridge register.
BUG=b:147042464
Signed-off-by: Raul E Rangel <rrangel@chromium.org>
Change-Id: I97b63571e336f541dcb274e4c8c608f6fc59ff42
Reviewed-on: https://review.coreboot.org/c/coreboot/+/41263
Reviewed-by: Furquan Shaikh <furquan@google.com>
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
Reviewed-by: Aaron Durbin <adurbin@chromium.org>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
When any USB image disk is connected to the DUT through
HUAWEI/APPLE Dongle, press Ctrl + u on the dev screen,
it cannot boot from USB.
We found the SS hub cannot be enumerated. So disable xHCI
compliance mode.
BRANCH=octopus
BUG=b:155347573
TEST=Confirm successful boot from USB
Change-Id: Iea4a3df156da0627336f7d6c1e03837b6cf0e7f2
Signed-off-by: tong.lin <tong.lin@bitland.corp-partner.google.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/40905
Reviewed-by: Marco Chen <marcochen@google.com>
Reviewed-by: Karthik Ramasubramanian <kramasub@google.com>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
This change adds a helper function lpc_early_init() which does the
following things:
1. Enables LPC controller
2. Disables any LPC decodes (These can be set up later by SoC or
mainboard as required).
3. Sets SPI base so that MMIO base for SPI and eSPI controllers is
initialized.
BUG=b:153675913
Signed-off-by: Furquan Shaikh <furquan@google.com>
Change-Id: I016f29339466c3fee92fe9b62a13d72297c29b8e
Reviewed-on: https://review.coreboot.org/c/coreboot/+/41257
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Aaron Durbin <adurbin@chromium.org>
Reviewed-by: Raul Rangel <rrangel@chromium.org>
Picasso has an LPC and eSPI bridge on the same PCI DEVFN. They can both
be active at the same time. This adds a way to specify which devices
belong on which bus.
i.e.,
device pci 14.3 on # - D14F3 bridge
device espi 0 on
chip ec/google/chromeec
device pnp 0c09.0 on end
end
end
device lpc 0 on
end
end
BUG=b:154445472
TEST=Built trembyle and saw static.c contained the espi bus.
Signed-off-by: Raul E Rangel <rrangel@chromium.org>
Change-Id: I0c2f40813c05680f72e5f30cbb13617e8f994841
Reviewed-on: https://review.coreboot.org/c/coreboot/+/41099
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Aaron Durbin <adurbin@chromium.org>
The default setting of the root port ASPM configuration can be
overridden from the device tree by using a non zero value.
BUG=N/A
TEST=tested on facebook monolith
Change-Id: I85c545d5eacb10f43b94228f1caf1163028645e0
Signed-off-by: Wim Vervoorn <wvervoorn@eltan.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/41171
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Frans Hendriks <fhendriks@eltan.com>
Current Interrupt setting use 2nd parameters as device function number.
- Correct as interrupt pin number according to _PRT package format.
{Address, pin, Source, Source index}
- Use irq number directly rather than irq definition as its number
is not for PCI device.
The issue found while enabling GBE and GBE interrupt is not working
without this change.
Reference
- ACPI spec 6.2.13 _PRT
- FSP reference code:
https://github.com/otcshare/CCG-TGL-Generic-SiC/blob/TGL.3163.01/
ClientOneSiliconPkg/IpBlock/Itss/LibraryPrivate/PeiItssPolicyLib/
PeiItssPolicyLibVer2.c
- BIOS reference code:
https://github.com/otcshare/CCG-TGL-Generic-Full/blob/master/
TigerLakeBoardPkg/Acpi/AcpiTables/Dsdt/PciTree.asl
TEST=boot to OS with GBE enabled and check GBE interrupt
Signed-off-by: Wonkyu Kim <wonkyu.kim@intel.com>
Change-Id: I8084b30c668c155ebabbee90b5f70054813b328e
Reviewed-on: https://review.coreboot.org/c/coreboot/+/41153
Reviewed-by: Duncan Laurie <dlaurie@chromium.org>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Add functionality to use process call cycle. It can be used to
write/read data to/from e.g. EEPROM attached to SMBus Controller
via I2C.
Tested on:
* C246
Change-Id: Ifdac6cf70a4ce744601f5d152a83d2125ea88360
Signed-off-by: Christian Walter <christian.walter@9elements.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/39875
Reviewed-by: Patrick Rudolph <siro@das-labor.org>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
pci_domain_set_resources is duplicated in all the SOCs. This change
promotes the duplicated function.
Picasso was adding it again in the northbridge patch. I decided to
promote the function instead of duplicating it.
BUG=b:147042464
TEST=Build and boot trembyle.
Signed-off-by: Raul E Rangel <rrangel@chromium.org>
Change-Id: Iba9661ac2c3a1803783d5aa32404143c9144aea5
Reviewed-on: https://review.coreboot.org/c/coreboot/+/41041
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Aaron Durbin <adurbin@chromium.org>
IedSize and EnableC6Dram are removed in JSL FSP v2114 so
remove them from 'fsp_params.c'.
BUG=155054804
BRANCH=None
TEST=Build and boot JSLRVP
Change-Id: I47bd3f87bdb59625098c0d734695f02d738f8bbd
Signed-off-by: Ronak Kanabar <ronak.kanabar@intel.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/41239
Reviewed-by: Maulik V Vaghela <maulik.v.vaghela@intel.com>
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
This patch control SATA related UPDs based on the devicetree
configuration as per each board's requirement.
BUG=b:155595624
BRANCH=None
TEST=Build, boot JSLRVP, Verified UPD values from FSP log
Change-Id: I4f7e7508b8cd483508293ee3e7b760574d8f025f
Signed-off-by: Ronak Kanabar <ronak.kanabar@intel.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/41029
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Karthik Ramasubramanian <kramasub@google.com>
Reviewed-by: Aamir Bohra <aamir.bohra@intel.com>
Reviewed-by: V Sowmya <v.sowmya@intel.com>
Reviewed-by: Maulik V Vaghela <maulik.v.vaghela@intel.com>
This change provides a helper function espi_update_static_bar() that
informs the eSPI common driver about the static BAR to use for eSPI
controller instead of reading the SPIBASE. This is required to support
the case of verstage running on PSP.
BUG=b:153675913
Signed-off-by: Furquan Shaikh <furquan@google.com>
Change-Id: I1f11bb2e29ea0acd71ba6984e42573cfe914e5d7
Reviewed-on: https://review.coreboot.org/c/coreboot/+/41256
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Aaron Durbin <adurbin@chromium.org>
Reviewed-by: Raul Rangel <rrangel@chromium.org>
FSP provides the UPD's for SATA and DMI power optimization.
In this patch we are adding the soc's config support to set
those power optimization bits in FSP. By default those
optimizations are enabled. To disable those we need to set
the DmiPwrOptimizeDisable and SataPwrOptimizeDisable to 1
in devicetree.
BUG=b:151162424
BRANCH=None
TEST=Build and boot volteer and TGL RVP.
Change-Id: Iefc5e7e48d69dccae43dc595dff2f824e53f5749
Signed-off-by: Shaunak Saha <shaunak.saha@intel.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/40005
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Angel Pons <th3fanbus@gmail.com>
This change adds the following helper functions for eSPI decode:
1. espi_open_io_window() - Open generic IO window decoded by eSPI
2. espi_open_mmio_window() - Open generic MMIO window decoded by eSPI
3. espi_configure_decodes() - Configures standard and generic I/O
windows using the espi configuration provided by mainboard in device tree.
BUG=b:153675913,b:154445472
Change-Id: Idb49ef0477280eb46ecad65131d4cd7357618941
Signed-off-by: Furquan Shaikh <furquan@google.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/41073
Reviewed-by: Aaron Durbin <adurbin@chromium.org>
Reviewed-by: Raul Rangel <rrangel@chromium.org>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
This change adds eSPI register definitions for I/O and MMIO decode
using eSPI on AMD SoCs. Additionally, it also adds a macro to define
the offset of ESPI MMIO base from SPI MMIO base.
BUG=b:153675913
Change-Id: Ifb70ae0c63cc823334a1d851faf4dda6d1c1fc1a
Signed-off-by: Furquan Shaikh <furquan@google.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/41072
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Raul Rangel <rrangel@chromium.org>
Reviewed-by: Aaron Durbin <adurbin@chromium.org>
This change adds helper functions that can be used to check support
for different slave capabilities.
BUG=b:153675913
Signed-off-by: Furquan Shaikh <furquan@google.com>
Change-Id: Ic66b06f9efcafd0eda4c6029fa67489de76bbed4
Reviewed-on: https://review.coreboot.org/c/coreboot/+/41253
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Raul Rangel <rrangel@chromium.org>
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
This change sets LPC_IO_PORT_DECODE_ENABLE to 0 as part of
lpc_disable_decodes() to ensure that the I/O port decodes are also disabled.
Signed-off-by: Furquan Shaikh <furquan@google.com>
Change-Id: I1474f561997f2ee1231bd0fcaab4d4d4e98ff923
Reviewed-on: https://review.coreboot.org/c/coreboot/+/41251
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Raul Rangel <rrangel@chromium.org>
Reviewed-by: Aaron Durbin <adurbin@chromium.org>
This change switches to using the common block SPI driver for
performing early SPI initialization and for re-configuring SPI speed
and mode after FSP-S has run.
Signed-off-by: Furquan Shaikh <furquan@google.com>
Change-Id: Ia3186ce59b66c2f44522a94fa52659b4942649b1
Reviewed-on: https://review.coreboot.org/c/coreboot/+/41250
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Aaron Durbin <adurbin@chromium.org>
Reviewed-by: Raul Rangel <rrangel@chromium.org>
This change adds support for using common SoC configuration by adding
soc_amd_common_config to soc_amd_picasso_config and helper function to
return pointer to the structure to amd common block code.
Change-Id: I8bd4eac3b19c9ded2d9a3e95ac077f014730f9d1
Signed-off-by: Furquan Shaikh <furquan@google.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/41249
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Raul Rangel <rrangel@chromium.org>
Reviewed-by: Aaron Durbin <adurbin@chromium.org>
This change adds support for following SPI configuration functions to
common block SPI driver and exposes them to be used by SoC:
1. fch_spi_early_init(): Sets up SPI ROM base, enables SPI ROM,
enables prefetching, disables 4dw burst mode and sets SPI speed and mode.
2. fch_spi_config_modes(): This allows SoC to configure SPI speed and
mode. It uses SPI settings from soc_amd_common_config to configure the
speed and mode.
These functions expect SoC to include soc_amd_common_config in SoC
chip config and mainboard to configure these settings in device tree.
Signed-off-by: Furquan Shaikh <furquan@google.com>
Change-Id: Ia4f231bab69e8450005dd6abe7a8e014d5eb7261
Reviewed-on: https://review.coreboot.org/c/coreboot/+/41248
Reviewed-by: Aaron Durbin <adurbin@chromium.org>
Reviewed-by: Raul Rangel <rrangel@chromium.org>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
This change adds a Kconfig option to request allocation of prefetch
memory for hotplug devices above the 4G boundary. In order to
select this option by default and still allow users to disable this if
required, another option is added to request allocation of prefetch
memory below 4G boundary which defaults to n but can be overriden
by mainboards.
Without this change, if the number of pciexp bridges supporting
hot-plug is more than 4 or if the reserved prefetch memory size for
hot-plug cases was increased, then the resource allocator would fail
to satisfy the resource requirement below 4G boundary.
BUG=b:149186922
TEST=Enabled resource allocation above 4G for prefetch memory on volteer
and verified that it gets allocated above 4G boundary.
Signed-off-by: Furquan Shaikh <furquan@google.com>
Change-Id: I061d935eef9fcda352230b03b5cf14e467924e50
Reviewed-on: https://review.coreboot.org/c/coreboot/+/39489
Reviewed-by: Aaron Durbin <adurbin@chromium.org>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
This change updates the resource limit for PCI domain to allow
resource allocation above 4G boundary. The resource limit is set to
the highest physical address for the CPU.
BUG=b:149186922
Signed-off-by: Furquan Shaikh <furquan@google.com>
Change-Id: Idfcc9a390d309886ee2b7880b29502c740e6578e
Reviewed-on: https://review.coreboot.org/c/coreboot/+/39488
Reviewed-by: Aaron Durbin <adurbin@chromium.org>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
This change adds support for allocating resources above the 4G
boundary by making use of memranges for resource windows enabled in
the previous CL.
It adds a new resource flag IORESOURCE_ABOVE_4G which is used in the
following ways:
a) Downstream device resources can set this flag to indicate that they
would like to have their resource allocation above the 4G
boundary. These semantics will have to be enabled in the drivers
managing the devices. It can also be extended to be enabled via
devicetree. This flag is automatically propagated by the resource
allocator from downstream devices to the upstream bridges in pass
1. It is done to ensure that the resource allocator has a global view
of downstream requirements during pass 2 at domain level.
b) Bridges have a single resource window for each of mem and prefmem
resource types. Thus, if any downstream resource of the bridge
requests allocation above 4G boundary, all the other downstream
resources of the same type under the bridge will be allocated above 4G
boundary.
c) During pass 2, resource allocator at domain level splits
IORESOURCE_MEM into two different memory ranges -- one for the window
below 4G and other above 4G. Resource allocation happens separately
for each of these windows.
d) At the bridge level, there is no extra logic required since the
resource will live entirely above or below the 4G boundary. Hence, all
downstream devices of any bridge will fall within the window allocated
to the bridge resource. To handle this case separately from that of
domain, initializing of memranges for a bridge is done differently
than the domain.
Limitation:
Resources of a given type at the bridge or downstream devices
cannot live both above and below 4G boundary. Thus, if a bridge has
some downstream resources requesting allocation for a given type above
4G boundary and other resources of the same type requesting allocation
below 4G boundary, then all these resources of the same type get
allocated above 4G boundary.
BUG=b:149186922
TEST=Verified that resources get allocated above the 4G boundary
correctly on volteer.
Signed-off-by: Furquan Shaikh <furquan@google.com>
Change-Id: I7fb2a75cc280a307300d29ddabaebfc49175548f
Reviewed-on: https://review.coreboot.org/c/coreboot/+/39487
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Aaron Durbin <adurbin@chromium.org>
This change updates the resource allocator in coreboot to allow using
multiple ranges for resource allocation rather than restricting
available window to a single base/limit pair. This is done in
preparation to allow 64-bit resource allocation.
Following changes are made as part of this:
a) Resource allocator still makes 2 passes at the entire tree. The
first pass is to gather the resource requirements of each device
under each domain. It walks recursively in DFS fashion to gather the
requirements of the leaf devices and propagates this back up to the
downstream bridges of the domain. Domain is special in the sense that
it has fixed resource ranges. Hence, the resource requirements from
the downstream devices have no effect on the domain resource
windows. This results in domain resource limits being unmodified after
the first pass.
b) Once the requirements for all the devices under the domain are
gathered, resource allocator walks a second time to allocate resources
to downstream devices as per the requirements. Here, instead of
maintaining a single window for allocating resources, it creates a
list of memranges starting with the resource window at domain and then
applying constraints to create holes for any fixed resources. This
ensures that there is no overlap with fixed resources under the
domain.
c) Domain does not differentiate between mem and prefmem. Since they
are allocated space from the same resource window at the domain level,
it considers all resource requests from downstream devices of the
domain independent of the prefetch type.
d) Once resource allocation is done at the domain level, resource
allocator walks down the downstream bridges and continues the same
process until it reaches the leaves. Bridges have separate windows for
mem and prefmem. Hence, unlike domain, the resource allocator at
bridge level ensures that downstream requirements are satisfied by
taking prefetch type into consideration.
e) This whole 2-pass process is performed for every domain in the
system under the assumption that domains do not have overlapping
address spaces.
Noticeable differences from previous resource allocator:
a) Changes in print logs observed due to flows being slightly
different.
b) Base, limit and size of domain resources are no longer updated
based on downstream requirements.
c) Memranges are used instead of a single base/limit pair for
determining resource allocation.
d) Previously, if a resource request did not fit in the available
base/limit window, then the resource would be allocated over DRAM or
any other address space defeating the principle of "no overlap". With
this change, any time a resource cannot fit in the available ranges,
it complains and ensures that the resource is effectively disabled by
setting base same as the limit.
e) Resource allocator no longer looks at multiple links to determine
the right bus for a resource. None of the current boards have multiple
buses under any downstream device of the domain. The only device with
multiple links seems to be the cpu cluster device for some AMD
platforms.
BUG=b:149186922
TEST=Verified that resource allocation looks correct based on
addresses assigned on Volteer.
Signed-off-by: Furquan Shaikh <furquan@google.com>
Change-Id: Ia1f089877c62e119c6a994a10809c9cc0050ec9a
Reviewed-on: https://review.coreboot.org/c/coreboot/+/39486
Reviewed-by: Aaron Durbin <adurbin@chromium.org>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Nick Vaccaro