The 'dram density' is a misnomer because the memory initialization
code treats that input parameter as a per rank density. Therefore,
update the variables to further clarify how it's actually being
used.
BUG=chrome-os-partner:55446
Change-Id: Ie4c944f35b531812205ac0bb1c70f39ac401495e
Signed-off-by: Aaron Durbin <adurbin@chromium.org>
Reviewed-on: https://review.coreboot.org/15773
Tested-by: build bot (Jenkins)
Reviewed-by: Furquan Shaikh <furquan@google.com>
Reviewed-by: Duncan Laurie <dlaurie@chromium.org>
The 16Gb devices use two ranks per channel within the DRAM module.
However, the density settings are really on a per rank basis so
indicate dual rank with a device density of 8Gb.
BUG=chrome-os-partner:55446
Change-Id: Ib5dba6f9ed248750d68b726996c71def9b75961e
Signed-off-by: Aaron Durbin <adurbin@chromium.org>
Reviewed-on: https://review.coreboot.org/15772
Reviewed-by: Duncan Laurie <dlaurie@chromium.org>
Tested-by: build bot (Jenkins)
Reviewed-by: Furquan Shaikh <furquan@google.com>
Despite the UPD comments the Chx_RankEnable fields are a bit
mask which indicates which ranks are enabled for physical
channel. Add the ability to set the rank mask correctly for
dual rank LPDDR4 modules.
BUG=chrome-os-partner:55446
Change-Id: I9dbed7bb6a4b512e57f6b4481180932a7cce91ff
Signed-off-by: Aaron Durbin <adurbin@chromium.org>
Reviewed-on: https://review.coreboot.org/15771
Tested-by: build bot (Jenkins)
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
Reviewed-by: Furquan Shaikh <furquan@google.com>
Reviewed-by: Andrey Petrov <andrey.petrov@intel.com>
Reviewed-by: Duncan Laurie <dlaurie@chromium.org>
The reset requests are handled in the FSP 2.0 wrapper, but
the current code doesn't check any non-successful return
values. Provide parity with the memory init path which die()s
under those circumstances.
Change-Id: I9df61323f742b4e94294321e3ca3ab58a68ca4dd
Signed-off-by: Aaron Durbin <adurbin@chromium.org>
Reviewed-on: https://review.coreboot.org/15766
Tested-by: build bot (Jenkins)
Reviewed-by: Andrey Petrov <andrey.petrov@intel.com>
Reviewed-by: Furquan Shaikh <furquan@google.com>
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
Not all are matched, but this makes it easier to backport
MTRR changes from haswell.
Change-Id: Ida5943b1469fc0089a31ff3b18131fb82b0941c6
Signed-off-by: Kyösti Mälkki <kyosti.malkki@gmail.com>
Reviewed-on: https://review.coreboot.org/15760
Tested-by: build bot (Jenkins)
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
Reviewed-by: Aaron Durbin <adurbin@chromium.org>
These guards have been removed starting with model_206ax.
Change-Id: Id63034ec4080e37eee2c120aa1f1ef604db5b203
Signed-off-by: Kyösti Mälkki <kyosti.malkki@gmail.com>
Reviewed-on: https://review.coreboot.org/15758
Tested-by: build bot (Jenkins)
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
Reviewed-by: Patrick Georgi <pgeorgi@google.com>
Since the socket layer is implemented with this CPU model, there
could potentially be multiple CPU models included. There can be
only one cache_as_ram include, so select it directly within
the socket directory.
Change-Id: Ia52bb152276eddfd1fb33ddb7f5d153ab8e8163c
Signed-off-by: Kyösti Mälkki <kyosti.malkki@gmail.com>
Reviewed-on: https://review.coreboot.org/15757
Tested-by: build bot (Jenkins)
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
Reviewed-by: Aaron Durbin <adurbin@chromium.org>
The EVT board uses an active high power control signal while
the previous board used an active low signal. Update the tables
to reflect the differences.
BUG=chrome-os-partner:55470
Change-Id: I198c0e4e019fcffe2cf748d382351ac965a81077
Signed-off-by: Aaron Durbin <adurbin@chromium.org>
Reviewed-on: https://review.coreboot.org/15763
Tested-by: build bot (Jenkins)
Reviewed-by: Furquan Shaikh <furquan@google.com>
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
I mistakenly assumed the order of the bits matched how one
would assign values as they wrote them msb .. lsb. However, the
gpio lib doesn't do that. Correct the order so that values are
read out correctly.
BUG=chrome-os-partner:54949t
Change-Id: I5304dfe2ba6f8eb073acab3377327167573ec2cc
Signed-off-by: Aaron Durbin <adurbin@chromium.org>
Reviewed-on: https://review.coreboot.org/15753
Reviewed-by: Duncan Laurie <dlaurie@chromium.org>
Reviewed-by: Furquan Shaikh <furquan@google.com>
Tested-by: build bot (Jenkins)
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
Zero-filling memory below 1 MiB resets car_migrated variable so
any CAR GLOBALs are not addressed correctly for the remaining
time in romstage. Also there is no actual need to do this as
ramstage loader handles BSS.
This fixes regression with commit 70cd54310 that broke fam10 boards
with romstage spinlocks enabled.
Change-Id: I7418821997a980ae5b818bd57e8a1b6507a543af
Signed-off-by: Kyösti Mälkki <kyosti.malkki@gmail.com>
Reviewed-on: https://review.coreboot.org/15754
Tested-by: build bot (Jenkins)
Reviewed-by: Aaron Durbin <adurbin@chromium.org>
Tested-by: Raptor Engineering Automated Test Stand <noreply@raptorengineeringinc.com>
When no CFLAGS are explicitly provided to it, the GMP configure script
will figure out the best optimization flags to use on its own. In
particular, it will setup the march, mfpu and mtune flags based on
hardware detection.
However, when CFLAGS are provided, they are used as-is and such
detection doesn't happen. When the march, mfpu and mtune flags are not
provided (which happens when GMP wasn't built already), not only will
related optimizations be disabled, but some code might not build because
of missing support. This happens with NEON instructions on ARMv7 hosts.
Thus, it is better not to set CFLAGS and leave it up to the GMP
configure script to get them right and still reuse those later.
Change-Id: I6ffcbac1298523d1b8ddf29a8bca1b00298828a7
Signed-off-by: Paul Kocialkowski <contact@paulk.fr>
Reviewed-on: https://review.coreboot.org/15452
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
Tested-by: build bot (Jenkins)
Reviewed-by: Nico Huber <nico.h@gmx.de>
Read timestamps from the last boot sequence and display the information
as if using cbmem -t.
Tested on QEMU with a SeaBIOS payload.
Change-Id: I44f1f6d6e4ef5458aca555c8a7d32cc8aae46502
Signed-off-by: Antonello Dettori <dev@dettori.io>
Reviewed-on: https://review.coreboot.org/15600
Tested-by: build bot (Jenkins)
Reviewed-by: Martin Roth <martinroth@google.com>
Split the additional time stamps concerning depthcharge from
the cbmem utility sourcecode and move them into
commonlib/timestamp_serialized.h header.
Change-Id: Ic23c3bc12eac246336b2ba7c7c39eb2673897d5a
Signed-off-by: Antonello Dettori <dev@dettori.io>
Reviewed-on: https://review.coreboot.org/15725
Tested-by: build bot (Jenkins)
Reviewed-by: Martin Roth <martinroth@google.com>
EVT has a wake signal for track pad which is routed to GP_15.
BUG=chrome-os-partner:54960
Change-Id: I9a73a3dc74e3bbed63509a3c076ec17a6559da55
Signed-off-by: Furquan Shaikh <furquan@google.com>
Reviewed-on: https://review.coreboot.org/15723
Tested-by: build bot (Jenkins)
Reviewed-by: Aaron Durbin <adurbin@chromium.org>
The tpm2_marshal_command() function returns a negative value on error,
so we must use a signed type for the return value.
This was found by the coverity scan:
https://scan.coverity.com/projects/coreboot?tab=overview
CID:1357675
CID:1357676
Change-Id: I56d2ce7d52b9b70e43378c13c66b55ac2948f218
Signed-off-by: Duncan Laurie <dlaurie@google.com>
Found-by: Coverity Scan
Reviewed-on: https://review.coreboot.org/15717
Tested-by: build bot (Jenkins)
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
Reviewed-by: Vadim Bendebury <vbendeb@chromium.org>
Add missing break to LEG_GPIO_REGS case to return the correct value for
legacy GPIO reads. Fixes coverity issue CID 1357460.
Found by Coverity, Fixes:
* CID 1357460 (#1 of 1): Unused value (UNUSED_VALUE)
returned_value: Assigning value from reg_legacy_gpio_read(step->reg)
to value here, but that stored value is overwritten before it can be
used.
value_overwrite: Overwriting previous write to value with value from
reg_pcie_afe_read(step->reg).
TEST=Build and run on Galileo Gen2.
Change-Id: I6c52e8801a32f510ac94276fe0c097850cbfde57
Signed-off-by: Lee Leahy <leroy.p.leahy@intel.com>
Reviewed-on: https://review.coreboot.org/15732
Tested-by: build bot (Jenkins)
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
Reviewed-by: Martin Roth <martinroth@google.com>
The 'speed' variable isn't being used after refactoring.
Change-Id: Id27a920c61b2bba18d391a7bfefe570235402dec
Signed-off-by: Martin Roth <martinroth@chromium.org>
Reviewed-on: https://review.coreboot.org/15749
Tested-by: build bot (Jenkins)
Reviewed-by: Jonathan Neuschäfer <j.neuschaefer@gmx.net>
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
FSP 2.0 spec only defines 2 reset request (COLD, WARM) exit codes. The
rest 6 codes are platform-specific and may vary. Modify helper function
so that only basic resets are handled and let SoC deal with the rest.
Change-Id: Ib2f446e0449301407b135933a2088bcffc3ac32a
Signed-off-by: Andrey Petrov <andrey.petrov@intel.com>
Reviewed-on: https://review.coreboot.org/15730
Reviewed-by: Aaron Durbin <adurbin@chromium.org>
Tested-by: build bot (Jenkins)
At first boot CSE spends long time preparing media for use. As result
it may not be able to deal with a CPU reset. Add reset_prepare()
callback that polls CSE readiness.
BUG=chrome-os-partner:55055
TEST=build with release version of fsp, reboot, observe polling for
CSE, then proper reboot happening
Change-Id: I639ef900b97132f1a7f269bb864d70009df9fdfe
Signed-off-by: Andrey Petrov <andrey.petrov@intel.com>
Reviewed-on: https://review.coreboot.org/15721
Tested-by: build bot (Jenkins)
Reviewed-by: Aaron Durbin <adurbin@chromium.org>
Some Intel SoC may need preparation before reset can be properly
handled. Add callback that chip/soc code can implement.
BUG=chrome-os-partner:55055
Change-Id: I45857838e1a306dbcb9ed262b55e7db88a8944e5
Signed-off-by: Andrey Petrov <andrey.petrov@intel.com>
Reviewed-on: https://review.coreboot.org/15720
Tested-by: build bot (Jenkins)
Reviewed-by: Aaron Durbin <adurbin@chromium.org>
Add functions to read Host Firmware Status register and a helper
function to determine if CSE is ready.
BUG=chrome-os-partner:55055
TEST=none
Change-Id: If511a51c04f7e59427d7952fa67b61060e2be404
Signed-off-by: Andrey Petrov <andrey.petrov@intel.com>
Reviewed-on: https://review.coreboot.org/15713
Tested-by: build bot (Jenkins)
Reviewed-by: Aaron Durbin <adurbin@chromium.org>
Ensure that the stack provided to FSPM doesn't overlap the current
program which is loading the FSPM component. If there is a conflict
that's an error since it could cause the current program to crash.
BUG=chrome-os-partner:52679
Change-Id: Ifff465266e5bb3cb3cf9b616d322a46199f802c7
Signed-off-by: Aaron Durbin <adurbin@chromium.org>
Reviewed-on: https://review.coreboot.org/15746
Tested-by: build bot (Jenkins)
Reviewed-by: Furquan Shaikh <furquan@google.com>
Reviewed-by: Andrey Petrov <andrey.petrov@intel.com>
If the system is in recovery mode force a full retrain.
BUG=chrome-os-partner:52679
Change-Id: I4e87685600880d815fe3198b820a10aa269baf37
Signed-off-by: Aaron Durbin <adurbin@chromium.org>
Reviewed-on: https://review.coreboot.org/15745
Tested-by: build bot (Jenkins)
Reviewed-by: Furquan Shaikh <furquan@google.com>
Reviewed-by: Andrey Petrov <andrey.petrov@intel.com>
Utilizing the FSP revision while saving the memory training data is
important because it means when the FSP is updated the memory training
is redone. The previous implementation was just using '0' as a revision.
Because of that behavior a retrain would not have been done on an FSP
upgrade.
BUG=chrome-os-partner:52679
Change-Id: I1430bd78c770a840d2deff2476f47150c02cf27d
Signed-off-by: Aaron Durbin <adurbin@chromium.org>
Reviewed-on: https://review.coreboot.org/15744
Tested-by: build bot (Jenkins)
Reviewed-by: Furquan Shaikh <furquan@google.com>
Reviewed-by: Andrey Petrov <andrey.petrov@intel.com>
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
The FSPS component loading was just loading to any memory address
listed in the header. That could be anywhere in the address space
including ramstage itself -- let alone corrupting the OS memory on
S3 resume. Remedy this by loading and relocating FSPS into cbmem.
The UEFI 2.4 header files include path are selected to provide the
types necessary for FSP relocation.
BUG=chrome-os-partner:52679
Change-Id: Iaba103190731fc229566a3b0231cf967522040db
Signed-off-by: Aaron Durbin <adurbin@chromium.org>
Reviewed-on: https://review.coreboot.org/15742
Tested-by: build bot (Jenkins)
Reviewed-by: Furquan Shaikh <furquan@google.com>
Reviewed-by: Andrey Petrov <andrey.petrov@intel.com>
Reviewed-by: John Zhao <john.zhao@intel.com>
The previously implementation for loading the FSPM component didn't
handle platforms which expects FSPM to be XIP. For the non-XIP case,
romstage's address space wasn't fully being checked for overlaps.
Lastly, fixup the API as the range_entry isn't needed any longer.
This API change requires a apollolake to be updated as well.
BUG=chrome-os-partner:52679
Change-Id: I24d0c7d123d12f15a8477e1025bf0901e2d702e7
Signed-off-by: Aaron Durbin <adurbin@chromium.org>
Reviewed-on: https://review.coreboot.org/15741
Tested-by: build bot (Jenkins)
Reviewed-by: Furquan Shaikh <furquan@google.com>
The current FSP component loading mechanism doesn't handle all the
requirements actually needed. Two things need to be added:
1. XIP support for MemoryInit component
2. Relocating SiliconInit component to not corrupt OS memory.
In order to accommodate those requirements the validation
and header initialization needs to be a separate function.
Therefore, provide fsp_validate_component() to help achieve those
requirements.
BUG=chrome-os-partner:52679
Change-Id: I53525498b250033f3187c05db248e07b00cc934d
Signed-off-by: Aaron Durbin <adurbin@chromium.org>
Reviewed-on: https://review.coreboot.org/15740
Tested-by: build bot (Jenkins)
Reviewed-by: Furquan Shaikh <furquan@google.com>
Instead of performing the same tasks in the chipset code move
the common sequences into the FSP 2.0 driver. This handles the
S3 paths as well as saving and restoring the memory data. The
chipset code can always override the settings if needed.
BUG=chrome-os-partner:52679
Change-Id: I098bf95139a0360f028a50aa50d16d264bede386
Signed-off-by: Aaron Durbin <adurbin@chromium.org>
Reviewed-on: https://review.coreboot.org/15739
Tested-by: build bot (Jenkins)
Reviewed-by: Furquan Shaikh <furquan@google.com>
The amount of reserved memory just below the DRAM limit in
32-bit space is defined in the FSP 2.0 specification within
the FSPM_ARCH_UPD structure. There's no need to make the
chipset code set the same value as needed for coreboot.
The chipset code can always change the value if it needs
after the common setting being applied.
Remove the call in soc/intel/apollolake as it's no longer
needed.
BUG=chrome-os-partner:52679
Change-Id: I69a1fee7a7b53c109afd8ee0f03cb8506584d571
Signed-off-by: Aaron Durbin <adurbin@chromium.org>
Reviewed-on: https://review.coreboot.org/15738
Reviewed-by: Furquan Shaikh <furquan@google.com>
Tested-by: build bot (Jenkins)
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
Reviewed-by: Andrey Petrov <andrey.petrov@intel.com>
The gcc compiler treats sizeof(void) == 1. Therefore requesting
a 1 byte reservation in cbmem and writing a pointer into the
buffer returned is wrong. Fix the size of the request to be
32-bits because FSP 2.0 is in 32-bit space by definition. Also,
since the access to the field happens across stage boundaries
it's important to ensure fixed widths are used in case a later
stage has a different pointer bit width.
BUG=chrome-os-partner:52679
Change-Id: Ib4efc7d5369d44a995318aac6c4a7cfdc73e4a8c
Signed-off-by: Aaron Durbin <adurbin@chromium.org>
Reviewed-on: https://review.coreboot.org/15737
Reviewed-by: Furquan Shaikh <furquan@google.com>
Tested-by: build bot (Jenkins)
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
The cbmem string for 'AFTER CAR' didn't have the proper spacing
so when that entry is added to cbmem it results in a misaligned
log entry with the others.
Change-Id: If940e85b7dc5fb8372d7e2845270dadad67ab3a0
Signed-off-by: Aaron Durbin <adurbin@chromium.org>
Reviewed-on: https://review.coreboot.org/15735
Reviewed-by: Furquan Shaikh <furquan@google.com>
Tested-by: build bot (Jenkins)
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
The Chrome OS options that will be shipped on this platform were
being set in the chromium repo with an external config file. Set
the options in the mainboard Kconfig file so there's no discrepancy
as to what will be used.
Change-Id: I05f0d1245611c16f54273728519a08e6edff3429
Signed-off-by: Aaron Durbin <adurbin@chromium.org>
Reviewed-on: https://review.coreboot.org/15733
Reviewed-by: Furquan Shaikh <furquan@google.com>
Tested-by: build bot (Jenkins)
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
Fix issue where zero-sized BIOS region could cause bitshift
for '-1' which is an unspecified behavior.
Change-Id: Icb62bf413a1a0d293657503ef21fe97b5f9a5484
Signed-off-by: Andrey Petrov <andrey.petrov@intel.com>
Reviewed-on: https://review.coreboot.org/15727
Tested-by: build bot (Jenkins)
Reviewed-by: Aaron Durbin <adurbin@chromium.org>
Fix and use the failsafe CAS detection logic rather than
recalulating the values from raw SPDs.
Tested on GA-G41M-ES2L with 2x2GB DDR2-800 DIMMs
(which worked before and still work)
Change-Id: I6af0f1705d099f7bcbff8c9baa94a68dae689e01
Signed-off-by: Damien Zammit <damien@zamaudio.com>
Reviewed-on: https://review.coreboot.org/15726
Tested-by: build bot (Jenkins)
Reviewed-by: Kyösti Mälkki <kyosti.malkki@gmail.com>
Reviewed-by: HAOUAS Elyes <ehaouas@noos.fr>
This function is unused since coreboot starts payloads in machine mode,
and it uses the obsolete eret instruction.
Change-Id: I98d7d0de5a3959821c21a0ba4319efb610fdefde
Signed-off-by: Jonathan Neuschäfer <j.neuschaefer@gmx.net>
Reviewed-on: https://review.coreboot.org/15729
Tested-by: build bot (Jenkins)
Reviewed-by: Ronald G. Minnich <rminnich@gmail.com>
Using the opcode directly is necessary for the transition to the GCC
6.1.0 based toolchain, because the old toolchain only supports eret and
the new toolchain only supports mret.
Change-Id: I17e14d4793ae5259f7ce3ce0211cbb27305506cc
Signed-off-by: Jonathan Neuschäfer <j.neuschaefer@gmx.net>
Reviewed-on: https://review.coreboot.org/15290
Tested-by: build bot (Jenkins)
Reviewed-by: Ronald G. Minnich <rminnich@gmail.com>
In order to save power in S3, we remove reset gpio setting in kernel.
We still need to initialize touchscreen ic.
Do it by pulling low reset gpio for 500us and then pulling high
in firmware.
BRANCH=none
BUG=chrome-os-partner:55170
TEST=build on elm.
Change-Id: Idbe0175a1fc1fa0b05e81706194c79d52c6101f6
Signed-off-by: Martin Roth <martinroth@chromium.org>
Original-Commit-Id: f40cc9a22c2551c2c9455cb8b60f36353602bca6
Original-Change-Id: If2ac815c4fd5c5ae15443348a49eb31449b724b1
Original-Signed-off-by: YH Huang <yh.huang@mediatek.com>
Original-Reviewed-on: https://chromium-review.googlesource.com/360312
Original-Reviewed-by: Julius Werner <jwerner@chromium.org>
Original-Reviewed-by: Yidi Lin <yidi.lin@mediatek.com>
Original-Reviewed-by: Johnny Chuang <johnny.chuang@emc.com.tw>
Reviewed-on: https://review.coreboot.org/15719
Reviewed-by: Furquan Shaikh <furquan@google.com>
Tested-by: build bot (Jenkins)
Sathyanarayana Nujella