Add explicit CBMEM_STAGE_CACHE option. Rename
CACHE_RELOCATED_RAMSTAGE_OUTSIDE_CBMEM to TSEG_STAGE_CACHE.
Platforms with SMM_TSEG=y always need to implement
stage_cache_external_region(). It is allowed to return with a
region of size 0 to effectively disable the cache.
There are no provisions in Kconfig to degrade from
TSEG_STAGE_CACHE to CBMEM_STAGE_CACHE.
As a security measure CBMEM_STAGE_CACHE default is changed to
disabled. AGESA platforms without TSEG will experience slower
S3 resume speed unless they explicitly select the option.
Change-Id: Ibbdc701ea85b5a3208ca4e98c428b05b6d4e5340
Signed-off-by: Kyösti Mälkki <kyosti.malkki@gmail.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/34664
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Furquan Shaikh <furquan@google.com>
No need to limit these declarations to FSP. Both
PARALLEL_MP_INIT smm_relocate() and TSEG_STAGE_CACHE
can be built on top of this.
Change-Id: I7b0b9b8c8bee03aabe251c50c47dc42f6596e169
Signed-off-by: Kyösti Mälkki <kyosti.malkki@gmail.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/34701
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Arthur Heymans <arthur@aheymans.xyz>
This implementation adds support to provide list of LPSS controllers
for a canonlake and icelake platforms. It implements strong function
of get_soc_lpss_controllers defined under intel common block lpss
driver.
Change-Id: I36c87e2324caf8ed3e4bb3e3dc6f5d4edf3e8d46
Signed-off-by: Aamir Bohra <aamir.bohra@intel.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/34136
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Paul Fagerburg <pfagerburg@chromium.org>
Reviewed-by: Furquan Shaikh <furquan@google.com>
Reviewed-by: Tim Wawrzynczak <twawrzynczak@chromium.org>
This implementation revises the UART PCI device reference in common
UART driver. The SOC functions have been aligned to provide the UART
PCI device reference using pcidev_path_on_root.
The uart_get_device() return type is changed, and files in which
it gets used are updated.
Change-Id: Ie0fe5991f3b0b9c596c3de9472e98e4091d7dd87
Signed-off-by: Aamir Bohra <aamir.bohra@intel.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/34582
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Paul Fagerburg <pfagerburg@chromium.org>
Reviewed-by: Tim Wawrzynczak <twawrzynczak@chromium.org>
Reviewed-by: Nico Huber <nico.h@gmx.de>
Clear the GPI Interrupt Status & Enable registers to prevent any
interrupt storms due to GPI.
BUG=b:138282962
TEST=Ensure that the Interrupt status & enable registers are reset
during the boot-up when the system is brought out of G3, S5 & S3. Ensure
that the system boots fine to ChromeOS.
Change-Id: I2185355d0095601e0778b6bf47ae137cc53e4051
Signed-off-by: David Wu <david_wu@quanta.corp-partner.google.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/34624
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Karthik Ramasubramanian <kramasub@google.com>
Reviewed-by: Tim Wawrzynczak <twawrzynczak@chromium.org>
Reviewed-by: Paul Fagerburg <pfagerburg@chromium.org>
There were two hooks in the boot state machine which dumped the ME
status to the debug UART, which is unnecessary. Removed the hook
for the BS_OS_RESUME_CHECK state, leaving just BS_PAYLOAD_LOAD, which
is called before FspNotifyEndOfFirmware, as required.
BUG=b:138463532
BRANCH=none
TEST=Boot up, check cbmem to ensure the ME status messages are
only printed one time.
Change-Id: I86bc6e33de4096f33023730ffabb25715c985de0
Signed-off-by: Tim Wawrzynczak <twawrzynczak@chromium.org>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/34616
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Paul Fagerburg <pfagerburg@chromium.org>
There will be a possible out of bounds array access if
power_limit_1_time == ARRAY_SIZE(power_limit_time_sec_to_msr), so
prevent that in the index check. This issue was fixed for other cpus in
commit 5cfef13f8d (cpu/intel: Fix out-of-bounds read due to off-by-one
in condition). Based on the discussion for that commit, also remove the
magic constant 28 in favour of the index of the last array element.
Change-Id: Ic3f8735b23a368f8a9395757bd52c2c40088afa1
Signed-off-by: Jacob Garber <jgarber1@ualberta.ca>
Found-by: Coverity CID 1229673
Reviewed-on: https://review.coreboot.org/c/coreboot/+/34498
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Patrick Rudolph <siro@das-labor.org>
Prevent implicit cast to pointers.
The compiler doesn't warn about the conversion from
integer to pointer without a cast, because SA_DEV_ROOT
is literally '0' and there seems to be an exception
for that conversion.
Change-Id: I64fc156e3b9f578414ad03a00edb7cf3e33205c1
Signed-off-by: Kyösti Mälkki <kyosti.malkki@gmail.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/34544
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Nico Huber <nico.h@gmx.de>
The "internal PME" wake-up source could be from integrated LAN,
HD audio/audio DSP, SATA, XHCI, CNVi, or an ME maskable host wake.
chromium:1680839 adds USB port details to the wake-up when the
XHCI causes the wake-up. Expand the logging for wake-up details to
identify and log the other wake-up sources with more details. Note that
wake on Integrated LAN (GbE), SATA, and ME Maskable Host Wake are not
in use on Hatch, so these will not be tested.
BUG=b:128936450
BRANCH=none
TEST=``FW_NAME=hatch emerge-hatch chromeos-ec depthcharge vboot_reference
libpayload coreboot-private-files intel-cmlfsp coreboot-private-files-hatch
coreboot chromeos-bootimage``
Ensure /build/hatch/firmware/image-hatch.serial.bin has been built.
Program image-hatch.serial.bin into the DUT using flashrom.
Switch the DUT to the console (Ctrl-Alt-F2, or use the AP console via
servo).
XHCI USB 2.0
* Plug a USB keyboard into a USB-A port
* ``powerd_dbus_suspend``
* Verify low power mode by issuing the ``powerinfo`` command on the EC
console (via servo). Expect to see ``power state 4 = S0ix``.
* Press a key on the USB keyboard
* ``mosys eventlog list`` shows:
12 | 2019-06-26 14:52:23 | S0ix Enter
13 | 2019-06-26 14:53:07 | S0ix Exit
14 | 2019-06-26 14:53:07 | Wake Source | PME - XHCI (USB 2.0 port) | 3
15 | 2019-06-26 14:53:07 | Wake Source | GPE # | 109
CNVi (connected to Wi-Fi):
* Enable wake on disconnect via ``iw phy0 wowlan enable disconnect``
* Set up a hotspot on an Android phone
* Connect the Chromebook to th hotspot
* ``powerd_dbus_suspend``
* Verify low power mode by issuing the ``powerinfo`` command on the EC
console (via servo). Expect to see ``power state 4 = S0ix``.
* Turn off the hotspot on the phone
* ``mosys eventlog list`` shows:
8 | 2019-07-11 10:58:17 | S0ix Enter
9 | 2019-07-11 10:59:17 | S0ix Exit
10 | 2019-07-11 10:59:17 | Wake Source | PME - WIFI | 0
11 | 2019-07-11 10:59:17 | Wake Source | GPE # | 109
XHCI USB 3.0
* TBD
HD Audio
* TBD
Change-Id: I2c71f6a56b4e1658a7427f67fa78af773b97ec7f
Signed-off-by: Paul Fagerburg <pfagerburg@chromium.org>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/34289
Reviewed-by: Karthik Ramasubramanian <kramasub@google.com>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
We do not want to disguise somewhat complex function
calls as simple macros.
Change-Id: I298f7f9a1c6a64cfba454e919eeaedc7bb2d4801
Signed-off-by: Kyösti Mälkki <kyosti.malkki@gmail.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/34411
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Furquan Shaikh <furquan@google.com>
Since PCH_DEVFN_PMC device is a PCI device that may be
hidden from enumeration, use SA_DEVFN_ROOT instead to
locate the SOC configuration.
Change-Id: I4b5195827fb32ec1dbd0bd6c9e243f4f9a4775ca
Signed-off-by: Kyösti Mälkki <kyosti.malkki@gmail.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/34327
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Duncan Laurie <dlaurie@chromium.org>
Reviewed-by: Furquan Shaikh <furquan@google.com>
It feels appropriate to define SoC specific XHCI USB info in SoC
specific XHCI source file and an API to get that information instead of
defining it in elog source file. This will help in other situations
where the information is required.
BUG=None
BRANCH=None
TEST=Boot to ChromeOS.
Change-Id: Ie63a29a7096bfcaab87baaae947b786ab2345ed1
Signed-off-by: Karthikeyan Ramasubramanian <kramasub@google.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/34290
Reviewed-by: Paul Fagerburg <pfagerburg@chromium.org>
Reviewed-by: Furquan Shaikh <furquan@google.com>
Reviewed-by: Tim Wawrzynczak <twawrzynczak@chromium.org>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
- Add CPU, MCH & IGD IDs for new Coffeelake SKUs
- Add PCH, LPC, SPI IDs for CNP-H PCH CM246 & C246
- Make some minor alignments & naming corrections to align with the rest
TEST= build, boot to both Linux & windows OS on CFL H & S platforms
and verified all the device Id's in serial console logs.
Signed-off-by: Lean Sheng Tan <lean.sheng.tan@intel.com>
Change-Id: I343b11ea8d9c33eb189d7478511a473b145f4ab4
Reviewed-on: https://review.coreboot.org/c/coreboot/+/34157
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Boon Tiong Teo <boon.tiong.teo@intel.com>
Reviewed-by: Nico Huber <nico.h@gmx.de>
Fix regression with commit
903b40a soc/intel: Replace uses of dev_find_slot()
Platforms where FSP hides PCI devices before enumeration
may halt with error message 'PCI: dev is NULL!'.
The workaround here is to print an error message revealing
the faulty source code function and revert to old behaviour
of dev_find_slot().
Change-Id: I5eab3e7f1993b686103eaa257aacda379dc259fa
Signed-off-by: Kyösti Mälkki <kyosti.malkki@gmail.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/34285
Reviewed-by: Matt DeVillier <matt.devillier@gmail.com>
Reviewed-by: Maxim Polyakov <max.senia.poliak@gmail.com>
Reviewed-by: Angel Pons <th3fanbus@gmail.com>
Reviewed-by: Christian Walter <christian.walter@9elements.com>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
The FSP_CAR option has additional configuration options whose default
values result in boot failures. Since default values should always boot,
default to the open-source CAR NEM Enhanced implementation instead. This
also allows us to get rid of an unnecessary vendor-specific special case.
Change-Id: I30b1808f91701c07dce6f1de08c213150e8a675a
Signed-off-by: Angel Pons <th3fanbus@gmail.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/34287
Reviewed-by: Aaron Durbin <adurbin@chromium.org>
Reviewed-by: Subrata Banik <subrata.banik@intel.com>
Reviewed-by: Lean Sheng Tan <lean.sheng.tan@intel.com>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
We got rid of the dangerous reconfiguration of arbitrary pads in
coreboot, but FSP still overrode that. Make sure that it doesn't
enable a UART for debug output when it isn't configured in core-
boot.
This, again, shows how dangerous it is to leave any FSP UPD at
its binary default.
Change-Id: I7280a80f71ddddbe78352eb696e6f5844d2df0b2
Signed-off-by: Nico Huber <nico.huber@secunet.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/34167
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
Reviewed-by: Angel Pons <th3fanbus@gmail.com>
The GPID method returns the PCR Port ID of the given GPIO community.
The CGPM method alters the given GPIO community's PM bits, given in
Arg1.
Change-Id: I098ee08573eb4f8a45d9b5ae84f2d85ce525c9b8
Signed-off-by: Tim Wawrzynczak <twawrzynczak@chromium.org>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/34179
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Subrata Banik <subrata.banik@intel.com>
Reviewed-by: Furquan Shaikh <furquan@google.com>
Remove the per-platform empty stubs, builds would
just fail as there is no equivalent conditional for
the smmrelocate.c file.
Change-Id: Ie11f307b7bc5415bfdba6a2c66aed01b70d9f0e0
Signed-off-by: Kyösti Mälkki <kyosti.malkki@gmail.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/34150
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Furquan Shaikh <furquan@google.com>
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
This is only a qualifier between TSEG and ASEG.
Change-Id: I8051df92d9014e3574f6e7d5b6f1d6677fe77c82
Signed-off-by: Kyösti Mälkki <kyosti.malkki@gmail.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/34135
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Furquan Shaikh <furquan@google.com>
This change intentionally removes the definition of PCH_DEV_PMC from
ramstage to avoid silent errors. This device gets hidden from PCI bus
in FSP-S and hence dropped from the root bus by the resource
allocator. In order to avoid incorrect references to the device, avoid
defining it in ramstage where it known to return NULL.
BUG=b:136861224
Change-Id: I4f69470ec80c7127a2b604ed2b1f794f5a63e126
Signed-off-by: Furquan Shaikh <furquan@google.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/34120
Reviewed-by: Subrata Banik <subrata.banik@intel.com>
Reviewed-by: Kyösti Mälkki <kyosti.malkki@gmail.com>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
PMC device gets hidden from PCI bus after FSP-S call. Thus, it gets
removed from the root bus as leftover unused device. With change
903b40a8a4 ("soc/intel: Replace uses of dev_find_slot()"), all uses
of dev_find_slot() were replaced by pcidev_path_on_root() which relies
on scanning of root bus to find the requested device. Since PMC device
is removed from the root bus, pcidev_path_on_root() returns NULL for
it thus resulting in configuration being skipped for the PMC
ultimately resulting in S3 failures.
Since the PCH_DEV_PMC was just used to get to chip config, this change
replaces the use of PCH_DEV_PMC with SA_DEV_ROOT.
BUG=b:136861224
TEST=Verified that S3 works fine on hatch.
Change-Id: Ie5ade00ac2aca697608f1bdea9764b71c26e2112
Signed-off-by: Furquan Shaikh <furquan@google.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/34116
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Kyösti Mälkki <kyosti.malkki@gmail.com>
Reviewed-by: Subrata Banik <subrata.banik@intel.com>
outb accepts a value followed by a port
Change-Id: I6fe3961b4f8cb2454e3b2564c3eae6af06c9e69d
Signed-off-by: Jeremy Soller <jeremy@system76.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/33940
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
Reviewed-by: Lance Zhao <lance.zhao@gmail.com>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Some of the values used for GPIO_CFG and MISCCFG were not correct,
causing GPEs to not work correctly. This adjusts them according to the
values found in the original ACPI tables for the System76 Gazelle.
Unfortunately, the Intel documentation[1] mentioned below is
also incorrect. I have mentioned this to Intel already. The source
for the Intel CoffeeLake FSP also confirms these new numbers.
This was tested on a System76 Gazelle (gaze14). The EC uses GPP_K3 for
its GPE and GPP_K6 is used for the lid switch GPE. Both function
correctly after applying this change.
[1] Intel Document #572235:
Intel ® 300 Series Chipset Families
Platform Controller Hub
External Design Specification (EDS) - Volume 2 of 2
Change-Id: I4ecc9552468037598ef5d4e10122d660dcbfe71d
Signed-off-by: Jeremy Soller <jeremy@system76.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/33941
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Nico Huber <nico.h@gmx.de>
Reviewed-by: Felix Held <felix-coreboot@felixheld.de>
To call dev_find_slot(0, xx) in romstage can produce
invalid results since PCI bus enumeration has not
been progressed yet.
Replace this with method that relies on bus topology
that walks the root bus only.
Change-Id: I2883610059bb9fa860bba01179e7d5c58cae00e5
Signed-off-by: Kyösti Mälkki <kyosti.malkki@gmail.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/33996
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Arthur Heymans <arthur@aheymans.xyz>
With VBOOT=y && VBOOT_MEASURED_BOOT=y message
digest will be allocated from the stack and
1 KiB reserve used with the recent platforms
was no longer sufficient.
The comment of LZMA scratchpad consuming stack
was obsolete for postcar, so these can be reduced
to same 4 KiB.
Change-Id: Iba1fb5bfad6946f316feac2d8c998a782142a56a
Signed-off-by: Mario Scheithauer <mario.scheithauer@siemens.com>
Signed-off-by: Kyösti Mälkki <kyosti.malkki@gmail.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/33775
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Aaron Durbin <adurbin@chromium.org>
Reviewed-by: Werner Zeh <werner.zeh@siemens.com>
Enhance elog wake source information with more details about which USB port
resulted in a wake from S3 or S0ix.
BUG=b:123429132
BRANCH=none
TEST=``FW_NAME=hatch emerge-hatch chromeos-ec depthcharge vboot_reference
libpayload coreboot-private-files intel-cmlfsp coreboot-private-files-hatch
coreboot chromeos-bootimage``
Ensure /build/hatch/firmware/image-hatch.serial.bin has been built.
Plug a keyboard into a USB port on the DUT.
Switch the DUT to the console (Ctrl-Alt-F2, or use the AP console via
servo).
On the console, run ``powerd_dbus_suspend``.
Wait for the DUT to enter low power mode.
Verify low power mode by issuing the ``powerinfo`` command on the EC
console (via servo). Expect to see ``power state 4 = S0ix``.
Press a key on the USB keyboard.
The DUT wakes up.
On the console, run ``mosys eventlog list`` and look for the wake source.
156 | 2019-06-26 09:46:07 | S0ix Enter
157 | 2019-06-26 12:14:05 | S0ix Exit
158 | 2019-06-26 12:14:05 | Wake Source | Internal PME | 0
159 | 2019-06-26 12:14:05 | Wake Source | GPE # | 109
Program image-hatch.serial.bin into the DUT using flashrom.
Repeat the ``powerd_dbus_suspend``, ``powerinfo``, ``mosys eventlog list``
sequence.
12 | 2019-06-26 14:52:23 | S0ix Enter
13 | 2019-06-26 14:53:07 | S0ix Exit
14 | 2019-06-26 14:53:07 | Wake Source | PME - XHCI (USB 2.0 port) | 3
15 | 2019-06-26 14:53:07 | Wake Source | GPE # | 109
Change-Id: Ie9ef870e219733dea9806c766f5351db25689b32
Signed-off-by: Paul Fagerburg <pfagerburg@chromium.org>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/33843
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Tim Wawrzynczak <twawrzynczak@chromium.org>
FSP sets the use of the 8254 timer via the Enable8254ClockGating
UPD, which defaults to enabled, overriding what is set by coreboot.
Per the FSP integration guide, this UPD needs to be disabled when
a legacy OS is booted (ie, when SeaBIOS is used as the payload).
Add a Kconfig option to set the UPD properly based on payload
selection, and remove the existing coreboot code in lpc.c since
it is either ineffective or being overridden by FSP.
Test: build/boot out-of-tree WHL board with both SeaBIOS and
Tianocore, ensure 8254 timer usage set correctly for each.
Signed-off-by: Matt DeVillier <matt.devillier@puri.sm>
Change-Id: I0e888bf754cb72093f14fc02f39bddcd6d288203
Reviewed-on: https://review.coreboot.org/c/coreboot/+/33512
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Nico Huber <nico.h@gmx.de>
