lsbpos and msbpos have incorrect behaviour when given 0.
lsbpos(0) returns 8, and msbpos(0) hangs. The latter is
because the check i >= 0 is always true for an unsigned
integer, causing it to loop indefinitely (this was flagged
by Coverity).
0 doesn't have a lsb or msb position, so we change both
functions to return -1 in this case to indicate an error.
The code already guards against calling these functions
with 0, but we make this more explicit to prevent errors
in the future.
Found-by: Coverity Scan, CID 1347356, 1347386
Signed-off-by: Jacob Garber <jgarber1@ualberta.ca>
Change-Id: Ic5be50846cc545dcd48593e5ed3fd6068a6104cb
Reviewed-on: https://review.coreboot.org/c/coreboot/+/32054
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
Reviewed-by: Patrick Georgi <pgeorgi@google.com>
The SkipExtGfxScan option is defined in the device tree, but doesn`t
update the value in the UPD. It uses the default value - 0. This
means that the FSP will scan all external graphics devices, in spite
of the configuration in devicetree.cb for a specific board.
Patch updates SkipExtGfxScan options in UPD from devicetree.cb.
This change affects all boards with skl/kbl processor.
Change-Id: Ie88a41bdf31f7c3e88df6c70c82a1cbf866372c4
Signed-off-by: Maxim Polyakov <max.senia.poliak@gmail.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/32170
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Nico Huber <nico.h@gmx.de>
Set new tcc_offset value to 10 degree C. This configures the Thermal
Control Circuit (TCC) activation value to 90 degree C. It prevents
any abrupt thermal shutdown while running heavy workload. This helps
to take early thermal throttling action when CPU temperature goes
above 90 degree C.
Change-Id: Ica77264782b4a3f3e72e73e1b8cb8b2e464fb033
Signed-off-by: Sumeet Pawnikar <sumeet.r.pawnikar@intel.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/32181
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Furquan Shaikh <furquan@google.com>
The effect of pointer aliasing on writes is that any data on CPU
registers that has been resolved from (non-const and non-volatile)
memory objects has to be discarded and resolved. In other words, the
compiler assumes that a pointer that does not have an absolute value
at build-time, and is of type 'void *' or 'char *', may write over
any memory object.
Using a unique datatype for MMIO writes makes the pointer to _not_
qualify for pointer aliasing with any other objects in memory. This
avoid constantly resolving the PCI MMCONF address, which is a derived
value from a 'struct device *'.
Change-Id: Id112aa5e729ffd8015bb806786bdee38783b7ea9
Signed-off-by: Kyösti Mälkki <kyosti.malkki@gmail.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/31752
Reviewed-by: Aaron Durbin <adurbin@chromium.org>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
This patch adds support PCIe Gen 3 with 8GT/s link speed for PEG x16
slot. All parameters for FSP are set during initialization in
romstage. Now there is no need to additionally configure the FSP
before building the ROM image.
Tested on Intel Core i5-6600 processor with the following devices:
- LP11000e Fibre Channel HBA (Gen2 x8);
- PEX8734 PCIe Fabric/Switch (Gen3 x16);
- NVIDIA GeForce GTX 1060 GPU (Gen3 x16).
GPU works with an nouveau and proprietary driver under Ubuntu 18.04.2
(4.15.0-46-generic GNU/Linux kernel). Discrete graphic card is used
as primary device for display output. Dynamic switching is not yet
supported.
Tianocore (edk2-stable201811-216-g51be9d0) is used as the payload.
Change-Id: Ia4f29df47d76de5069fe53120434cc7c2ab6f044
Signed-off-by: Maxim Polyakov <max.senia.poliak@gmail.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/31948
Reviewed-by: Nico Huber <nico.h@gmx.de>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
FSP options list (for each PEG port):
- PegXEnable,
- PegXMaxLinkWidth,
- PegXMaxLinkSpeed,
- PegXPowerDownUnusedLanes,
- PegXGen3EqPh2Enable,
- PegXGen3EqPh3Method.
Add PegMaxLinkWidth to chip.h. This option overrides the number of
active lines from the devicetree.cb for each enabled PEG port (for
example for boards that use x4 instead of x16 lines in PEG0). If the
PegMaxLinkWidth is not defined, the port uses the maximum possible
number of lines.
To enable or disable the corresponding PEG root port you need to add
to the devicetree.cb:
device pci 01.0 on end # enable PEG0 root port
device pci 01.1 off end # do not configure PEG1
If PEG port is not defined in the devicetree, it will be disabled in
FSP.
It has been tested on ASRock H110M-DVS motherboard (Skylake i5-6600
CPU).
Change-Id: I23708f7060edf08739adf61fe61a419329907563
Signed-off-by: Maxim Polyakov <max.senia.poliak@gmail.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/32045
Reviewed-by: Nico Huber <nico.h@gmx.de>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
The InternalGfx option in devicetree.cb is not used to enable iGPU.
The patch removes this option from chip.h and mb/*/devicetree.cb
files for all boards with skl/kbl processor.
Change-Id: I41ecca3fdfb1d4b20ee634a13263ff481dcf440e
Signed-off-by: Maxim Polyakov <max.senia.poliak@gmail.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/32171
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Nico Huber <nico.h@gmx.de>
This patch includes the following changes:
1. Sets FSP options in romstage_fsp20.c to select primary GPU.
List of options:
- InternalGfx,
- PrimaryDisplay.
2. iGPU will be initialized if the corresponding PCI device is defined
in the device tree as:
device pci 02.0 on end
In this case, it is not necessary to set the InternalGfx option to
enable this device
3. Primary_iGFX is used as the default value for all skl/kbl boards
(since the PrimaryDisplay option isn`t defined in the devicetree.cb)
Change-Id: Ie3f9362676105e41c69139a094dbb9e8b865689f
Signed-off-by: Maxim Polyakov <max.senia.poliak@gmail.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/32044
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Nico Huber <nico.h@gmx.de>
One of many steps to compile with -Wconversion, as unsigned int and int
aren't the same thing.
BUG=b:111443775
BRANCH=none
TEST=make junit.xml shows fewer warnings with -Wconversion enabled
Change-Id: I9673ca70da32a1e5117b27fa89167e03379af9c1
Signed-off-by: Patrick Georgi <pgeorgi@google.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/32183
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Patrick Rudolph <siro@das-labor.org>
RTC time contains invalid values on system without RTC battery.
Handle 'invalid' the same way as 'cmos_invalid'. This will reset CMOS date
when calling function enables 'invalid'.
BUG=N/A
TEST=Portwell PQ-M107 booting Linux Embedded
Change-Id: I5eae57d00f328400a8b03c28b7ecdbbc71522206
Signed-off-by: Frans Hendriks <fhendriks@eltan.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/29329
Reviewed-by: Nico Huber <nico.h@gmx.de>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
I just got hit by a double-evaluation bug again, it's time to attempt
to fix this once more. Unfortunately there are several issues that don't
make this easy:
- bitfield variables don't support typeof()
- local macro variables that shadow others trigger -Werror=shadow
- sign warnings with integer literal and unsigned var in typeof-MIN()
- ({ statement expressions }) can not be used outside functions
- romcc doesn't support any of the fancy GCC/clang extensions
This patch tries to address all of them as far as possible with macro
magic. We don't have the technology to solve the bitfield and
non-function context issues yet (__builtin_choose_expr() still throws a
"no statement expression outside a function" error if it's only in the
branch that's not chosen, unfortunately), so we'll have to provide
alternative macros for use in those cases (and we'll avoid making
__ALIGN_MASK() double-evaluation safe for now, since it would be
annoying to do that there and having an alignment mask with side
effects seems very unlikely). romcc can continue using unsafe versions
since we're hopefully not writing a lot of new code for it. Sign
warnings can be avoided in literal/variable comparisons by always using
the type of the variable there. Shadowing is avoided by picking very
explicit local variable names and using a special __COUNTER__ solution
for MIN() and MAX() (the only ones of these you're likely to nest).
Also add DIV_ROUND_UP() to libpayload since it's a generally quite
useful thing to have.
Change-Id: Iea35156c9aa9f6f2c7b8f00991418b746f44315d
Signed-off-by: Julius Werner <jwerner@chromium.org>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/32027
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
Reviewed-by: Patrick Georgi <pgeorgi@google.com>
This change moves the I2C/SPI devices and configs which do not apply
to all variants to override tree. Currently, there are just two
variants. However, as we prepare to add more variants, these devices
need to be moved out of the base devicetree.
BUG=b:129728235
TEST=Verified that I2C/SPI devices are present in static.c for hatch
and hatch_whl.
Change-Id: I9426f6bf5f8514de5f1889e22e57105749fd92de
Signed-off-by: Furquan Shaikh <furquan@google.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/32138
Reviewed-by: Rizwan Qureshi <rizwan.qureshi@intel.com>
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
Reviewed-by: Tim Wawrzynczak <twawrzynczak@chromium.org>
Reviewed-by: Shelley Chen <shchen@google.com>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
The patches added to `make` require that we use automake & aclocal
to rebuild the configuration, but version 1.15 of autotools is
expected. After debian sid updated to autotools 1.16, the tools can't
be located.
We'll just pretend to have version 1.15 with symbolic links. This
doesn't seem to be a good solution but gets the job done.
Change-Id: I9f616b96e728106e7adf321325caa06808e064c2
Signed-off-by: Martin Roth <martinr@coreboot.org>
Signed-off-by: Nico Huber <nico.h@gmx.de>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/28544
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
Reviewed-by: Patrick Georgi <pgeorgi@google.com>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
In S3 the PCH is driving the DEVSLP signal low, assuming that the SATA device
is already powered off. However on hatch the SATA power is still enabled. And,
since DEVSLP is low, this causes the SATA device to not enter low power state.
The fix here is to set the pad config to be reset on PLTRST assertion which
will cause the pin to be high impedance state and will be pulled up by the
SATA device.
BUG=b:126611255
BRANCH=None
TEST=Make sure that S3 and S0ix is working fine on hatch.
And also make sure that DEVSLP is pulled high in S3.
Change-Id: Ifb6a71a72244522c8dd8d48e9b9f8dc6feef8981
Signed-off-by: Rizwan Qureshi <rizwan.qureshi@intel.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/32136
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Furquan Shaikh <furquan@google.com>
Serial IRQ was configured in quiet mode, but not enabled.
Enable serial IRQ and use 'enum seriirq_mode' as a devicetree
option.
Function sc_enable_serial_irqs() is added to enabled serial IRQs.
enable_serirq_quiet_mode() is renamed to
sc_set_serial_irqs_mode(). This function use the 'serirq_mode' to
set the mode. The call to this function is moved from finalize to init
having serial IRQs enable in early stage.
Serial IRQs must be enabled in continuous mode for at least one frame
before switching into quiet mode.
BUG=N/A
TEST=Portwell PQ7-M107
Change-Id: I7844cad69dc0563fa6109d779d0afb7c2edd7245
Signed-off-by: Frans Hendriks <fhendriks@eltan.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/29398
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Michał Żygowski <michal.zygowski@3mdeb.com>
There's an issue with the newest toolchain that is blowing the bootblock
size on Smaug when compiling for chromeos. Increasing the bootblock
size by 2KB will take care of the issue for a while.
Signed-off-by: Martin Roth <martinroth@chromium.org>
Change-Id: I58f7f1cedc8fc5b4c4287f5a120ed76140e1f7a9
Reviewed-on: https://review.coreboot.org/c/coreboot/+/32163
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Julius Werner <jwerner@chromium.org>
Reviewed-by: HAOUAS Elyes <ehaouas@noos.fr>
Adding Kohaku as a variant of hatch.
BUG=b:129706980
BRANCH=NONE
TEST=./util/abuild/abuild -p none -t google/hatch -x -a
make sure HATCH_KOHAKU is built as well.
Change-Id: I5b451f421f6d353005e6b73eac180dcec2e8b0c0
Signed-off-by: Shelley Chen <shchen@google.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/32161
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Furquan Shaikh <furquan@google.com>
