coreboot.git
2 years agolibpayload: head.S: Avoid clearing BSS (and heap) again master
Julius Werner [Sat, 6 Aug 2016 04:27:47 +0000 (21:27 -0700)]
libpayload: head.S: Avoid clearing BSS (and heap) again

3 out of 4 architectures currently zero out the payload BSS in early
assembly code, which is pointless since the code loading the payload has
already done that (with a more efficient memset). ARM64 has never had
any code like this and can run just fine without it. This also defeats
the new optimization of moving the heap out of the BSS, since all three
implementations assume that everything between _edata and _end is BSS.
We should just take this out.

Change-Id: I45cd2dabd94da43ff0f77e990f11c877cee6cda1
Signed-off-by: Julius Werner <jwerner@chromium.org>
Reviewed-on: https://review.coreboot.org/16091
Tested-by: build bot (Jenkins)
Reviewed-by: Aaron Durbin <adurbin@chromium.org>
Reviewed-by: Stefan Reinauer <stefan.reinauer@coreboot.org>
2 years agolibpayload: cbfs: Fix minor memory leak in some edge cases
Julius Werner [Fri, 12 Aug 2016 19:50:42 +0000 (12:50 -0700)]
libpayload: cbfs: Fix minor memory leak in some edge cases

cbfs_get_handle() allocates memory for a handle and doesn't free it if
it errors out later, leaving the memory permanently leaked. Fix.

Change-Id: Ide198105ce3ad6237672ff152b4490c768909564
Reported-by: Coverity
Signed-off-by: Julius Werner <jwerner@chromium.org>
Reviewed-on: https://review.coreboot.org/16207
Reviewed-by: Aaron Durbin <adurbin@chromium.org>
Tested-by: build bot (Jenkins)
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
2 years agoDocumentation: add start of documentation of the build system
Patrick Georgi [Thu, 11 Aug 2016 08:50:28 +0000 (10:50 +0200)]
Documentation: add start of documentation of the build system

Change-Id: Ic4a4b4d71852bfe0b1fc52373e88d0a53b145844
Signed-off-by: Patrick Georgi <pgeorgi@chromium.org>
Reviewed-on: https://review.coreboot.org/16150
Tested-by: build bot (Jenkins)
Reviewed-by: Stefan Reinauer <stefan.reinauer@coreboot.org>
2 years agochromeec/acpi: add Tablet event and EC ACPI MEM
jiazi Yang [Thu, 28 Jul 2016 09:15:01 +0000 (05:15 -0400)]
chromeec/acpi: add Tablet event and EC ACPI MEM

Switch DPTF table when TABLET/NOTEBOOK mode changes
1. EC send EC_HOST_EVENT_MODE_CHANGE(29/0x1D) when mode changes
2. Host read current "physical mode" from EC ERAM

BUG=chrome-os-partner:53526
BRANCH=master
TEST=build glados

Change-Id: I836d2b9d1a24c455c4b8d4b85f7edc19259d2f71
Signed-off-by: Martin Roth <martinroth@chromium.org>
Original-Commit-Id: 9506c4c07e0f713c9a22a0231bc4255f6876783f
Original-Change-Id: I5a3363ff9c958decb5aed1c85fc2a1ef6670931d
Original-Signed-off-by: jiazi Yang <Tomato_Yang@asus.com>
Original-Signed-off-by: Shasha Zhao <Sarah_Zhao@asus.com>
Original-Signed-off-by: jiazi Yang <Tomato_Yang@asus.com>
Original-Reviewed-on: https://chromium-review.googlesource.com/365991
Original-Reviewed-by: Aaron Durbin <adurbin@chromium.org>
Reviewed-on: https://review.coreboot.org/16151
Tested-by: build bot (Jenkins)
Reviewed-by: Patrick Georgi <pgeorgi@google.com>
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
2 years agodrivers/intel/fsp2_0: Fix FSP reset path
Lee Leahy [Fri, 12 Aug 2016 00:40:01 +0000 (17:40 -0700)]
drivers/intel/fsp2_0: Fix FSP reset path

Don't verify HOB list pointer or HOBs when FSP returns a reset request.

BRANCH=none
BUG=chrome-os-partner:56159
TEST=Build and run on Galileo Gen2.

Change-Id: I6382f5ff92092623955806ebff340608c4ee156a
Signed-off-by: Lee Leahy <Leroy.P.Leahy@intel.com>
Reviewed-on: https://review.coreboot.org/16162
Tested-by: build bot (Jenkins)
Reviewed-by: Aaron Durbin <adurbin@chromium.org>
2 years agosoc/apollolake: enable access to RTC NVRAM
Ravi Sarawadi [Wed, 10 Aug 2016 18:33:56 +0000 (11:33 -0700)]
soc/apollolake: enable access to RTC NVRAM

FSP unconditionally locks parts of the NVRAM in the RTC.
This change will enable coreboot to update the locking policy
and be able to unlock the region

BUG=chrome-os-partner:55944
TEST=Check 'crossystem dev_boot_usb=1'

Change-Id: I70fd2bafa6ff9eb9cdf284b9780e4b90dee0f4ce
Signed-off-by: Ravi Sarawadi <ravishankar.sarawadi@intel.com>
Reviewed-on: https://review.coreboot.org/16144
Tested-by: build bot (Jenkins)
Reviewed-by: Furquan Shaikh <furquan@google.com>
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
Reviewed-by: Giri P Mudusuru <giri.p.mudusuru@intel.com>
2 years agoreef: Update chromeos.fmd
Furquan Shaikh [Wed, 10 Aug 2016 20:25:21 +0000 (13:25 -0700)]
reef: Update chromeos.fmd

1. Get rid of LBP2 partition
2. Shrink RO size
3. Increase RW-A and RW-B sizes
4. Increase RW_MRC_CACHE size

CQ-DEPEND=CL:366793
BUG=chrome-os-partner:52127, chrome-os-partner:55699,
chrome-os-partner:55778
BRANCH=None
TEST=Compiles successfully. Boots to OS.

Change-Id: Iad41d8cc7697e6d73f1aa2c699b0e8559349b77e
Signed-off-by: Furquan Shaikh <furquan@google.com>
Reviewed-on: https://review.coreboot.org/16145
Tested-by: build bot (Jenkins)
Reviewed-by: Aaron Durbin <adurbin@chromium.org>
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
2 years agosoc/qualcomm/ipq40xx: Reduce the delay in I2C.
Kan Yan [Thu, 30 Jun 2016 20:46:56 +0000 (13:46 -0700)]
soc/qualcomm/ipq40xx: Reduce the delay in I2C.

3ms delay was found in testing to be sufficient for
qup_i2c_write_fifo_flush(), but 1 additional ms was added to give
additional headroom.
Change the Delay from 10ms to 4ms.

BUG=b:28942403
TEST=Boot up Gale board and the TPM functions normally.
BRANCH=None

Change-Id: I6821e2a101cc44e11d74eb6a6215aa9b848ae8c6
Signed-off-by: Martin Roth <martinroth@chromium.org>
Original-Commit-Id: d93520fab15c5695ea18db21d0f3b24a108f204d
Original-Change-Id: I202f5b8a1ef62bb039c56ba5a25b48b205cf4a67
Original-Signed-off-by: Kan Yan <kyan@google.com>
Original-Reviewed-on: https://chromium-review.googlesource.com/357961
Original-Reviewed-by: Suresh Rajashekara <sureshraj@chromium.org>
Original-Reviewed-by: SARAVANAKUMAR SUDALAI <ssudalai@qti.qualcomm.com>
Reviewed-on: https://review.coreboot.org/16126
Tested-by: build bot (Jenkins)
Reviewed-by: Patrick Georgi <pgeorgi@google.com>
2 years agorockchip/rk3399: Add code to neuter Type-C PHY for firmware USB
Julius Werner [Thu, 4 Aug 2016 02:18:39 +0000 (19:18 -0700)]
rockchip/rk3399: Add code to neuter Type-C PHY for firmware USB

The Rockchip RK3399 integrates a USB Type-C PHY in charge of things like
SuperSpeed line muxing for rotated cable orientations in the SoC. While
fancy, this is very complicated and we don't want to implement support
for the whole thing in firmware. The USB Type-C standard has
intentionally been designed in a way that the USB 2.0 (HighSpeed) lines
always "just work" in any orientation (by just shorting different pins
in the connector together) so that simple use cases like ours can get
basic USB functionality without much hassle.

However, a semi-configured Type-C PHY can confuse USB 3.0 capable
devices into thinking we're actually supporting SuperSpeed, and fail at
that rather than establishing a reliable HighSpeed connection. This
patch sets enough bits in the Type-C PHY to electrically isolate the
SuperSpeed lines from the connector so that the connected device isn't
going to get any fancy ideas and reliably falls back to USB 2.0.

Also clean up the rest of the USB code while we're at it: avoid writing
a few bits that are already in the right state from their reset values
anyway, or reading values whose content we already know for this SoC.
Rename the USB controllers to the name actually used in the Rockchip
documentation (USB OTGx) rather than the name blindly copied from
Exynos code (USB DRDx).

BRANCH=None
BUG=chrome-os-partner:54621
TEST=Plug a USB 3.0 Patriot Memory stick into both ports in all
orientations, observe how it gets reliably detected now (safe for some
known hardware issues on my board).

Change-Id: Ifce6bcddd69f2e8f2e2a2f48faf65551e084da1e
Signed-off-by: Martin Roth <martinroth@chromium.org>
Original-Commit-Id: c526906f998bf66067d3addb8b3d3a126c188b1e
Original-Change-Id: Ie80a201a58764c4d851fe4a5098a5acfc4bcebdf
Original-Signed-off-by: Julius Werner <jwerner@chromium.org>
Original-Reviewed-on: https://chromium-review.googlesource.com/366160
Original-Reviewed-by: liangfeng wu <wulf@rock-chips.com>
Original-Reviewed-by: Shelley Chen <shchen@chromium.org>
Original-Reviewed-by: <515506667@qq.com>
Reviewed-on: https://review.coreboot.org/16125
Tested-by: build bot (Jenkins)
Reviewed-by: Patrick Georgi <pgeorgi@google.com>
2 years agoarch/riscv: Fix the page table setup code
Jonathan Neuschäfer [Tue, 9 Aug 2016 00:07:12 +0000 (02:07 +0200)]
arch/riscv: Fix the page table setup code

In particular:

- Fix the condition of the loop that fills the mid-level page table
- Adhere to the format of sptbr

Change-Id: I575093445edfdf5a8f54b0f8622ff0e89f77ccec
Signed-off-by: Jonathan Neuschäfer <j.neuschaefer@gmx.net>
Reviewed-on: https://review.coreboot.org/16120
Tested-by: build bot (Jenkins)
Reviewed-by: Ronald G. Minnich <rminnich@gmail.com>
2 years agoarch/riscv: Update encoding.h and dependent files
Jonathan Neuschäfer [Tue, 9 Aug 2016 00:07:10 +0000 (02:07 +0200)]
arch/riscv: Update encoding.h and dependent files

I copied it from commit e10d2def7d of spike and made sure the copyright
header is still there.

Change-Id: Ie8b56cd2f4855b97d36a112a195866f4ff0feec5
Signed-off-by: Jonathan Neuschäfer <j.neuschaefer@gmx.net>
Reviewed-on: https://review.coreboot.org/15832
Tested-by: build bot (Jenkins)
Reviewed-by: Stefan Reinauer <stefan.reinauer@coreboot.org>
2 years agolenovo/x60: add info message if dock is present
Alexander Couzens [Wed, 10 Aug 2016 11:17:18 +0000 (13:17 +0200)]
lenovo/x60: add info message if dock is present

Change-Id: I5a6d41f815f65719780499fa18c131311a9dc8f7
Signed-off-by: Alexander Couzens <lynxis@fe80.eu>
Reviewed-on: https://review.coreboot.org/16136
Tested-by: build bot (Jenkins)
Reviewed-by: Kyösti Mälkki <kyosti.malkki@gmail.com>
2 years agolenovo/x60: add GPIOs initialisation before dock check
Antonello Dettori [Wed, 10 Aug 2016 12:32:07 +0000 (14:32 +0200)]
lenovo/x60: add GPIOs initialisation before dock check

Add GPIOs initialisation before dock check.

Needed in order to properly detect the presence or absence of the lenovo
dock.
Previously the check always reported the dock as connected and currently
it always reports it as disconnected since the GPIOs are not properly
initialised during the check.

Tested and confirmed working.

Change-Id: I7fbf8c2262a1eb5dee9cbe5e23bf44f7f8181009
Signed-off-by: Antonello Dettori <dev@dettori.io>
Reviewed-on: https://review.coreboot.org/16139
Tested-by: build bot (Jenkins)
Reviewed-by: Alexander Couzens <lynxis@fe80.eu>
Reviewed-by: Kyösti Mälkki <kyosti.malkki@gmail.com>
2 years agovendorcode/intel/fsp: Add fsp 2.0 header files for skylake and kabylake
Rizwan Qureshi [Wed, 3 Aug 2016 06:53:18 +0000 (12:23 +0530)]
vendorcode/intel/fsp: Add fsp 2.0 header files for skylake and kabylake

Add FSP 2.0 header files, these files are common for Skylake
and Kabylake, name the folder as skykabylake to signify the same.

Change-Id: I71b43a59c9a9b0adf1ee48285e4a72e24a13df2d
Signed-off-by: Rizwan Qureshi <rizwan.qureshi@intel.com>
Reviewed-on: https://review.coreboot.org/16050
Tested-by: build bot (Jenkins)
Reviewed-by: Aaron Durbin <adurbin@chromium.org>
Reviewed-by: Giri P Mudusuru <giri.p.mudusuru@intel.com>
2 years agobuild system: Print the content of all regions we add files to
Patrick Georgi [Wed, 10 Aug 2016 15:13:00 +0000 (17:13 +0200)]
build system: Print the content of all regions we add files to

That's more useful than just COREBOOT for more complex scenarios

Change-Id: I93cd686d698799a3331ca2ea487cd6efb304caa0
Signed-off-by: Patrick Georgi <pgeorgi@chromium.org>
Reviewed-on: https://review.coreboot.org/16143
Tested-by: build bot (Jenkins)
Reviewed-by: Aaron Durbin <adurbin@chromium.org>
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
2 years agobuild system: drop -cbfstool-opts variable support
Patrick Georgi [Wed, 10 Aug 2016 15:06:12 +0000 (17:06 +0200)]
build system: drop -cbfstool-opts variable support

It was a band-aid that isn't required any more.

Change-Id: Ib1793ae8fe25eecf9bd5ab8e5feef0d9380b43c2
Signed-off-by: Patrick Georgi <pgeorgi@chromium.org>
Reviewed-on: https://review.coreboot.org/16142
Tested-by: build bot (Jenkins)
Reviewed-by: Aaron Durbin <adurbin@chromium.org>
2 years agointel/fsp1_1: Use new per-region position override
Patrick Georgi [Wed, 10 Aug 2016 15:04:43 +0000 (17:04 +0200)]
intel/fsp1_1: Use new per-region position override

It cooperates better with the file sorting heuristic.

Change-Id: I1c071243720352970dd2c4c2afed12451f91dcaa
Signed-off-by: Patrick Georgi <pgeorgi@chromium.org>
Reviewed-on: https://review.coreboot.org/16141
Tested-by: build bot (Jenkins)
Reviewed-by: Aaron Durbin <adurbin@chromium.org>
2 years agobuild system: allow overriding file position and alignment per region
Patrick Georgi [Wed, 10 Aug 2016 15:02:58 +0000 (17:02 +0200)]
build system: allow overriding file position and alignment per region

To override fallback/foo's position or alignment in region BAR, use
fallback/foo-BAR-{position,align} = 0x1234

Like for the global settings, specifying both isn't allowed
because that's rather pointless.

Change-Id: I94f41ebc9f35108267265df4164f23b70e3d0bf6
Signed-off-by: Patrick Georgi <pgeorgi@chromium.org>
Reviewed-on: https://review.coreboot.org/16140
Tested-by: build bot (Jenkins)
Reviewed-by: Aaron Durbin <adurbin@chromium.org>
2 years agobuild system: remove early stage cbfs-file sorting
Patrick Georgi [Wed, 10 Aug 2016 11:29:03 +0000 (13:29 +0200)]
build system: remove early stage cbfs-file sorting

They're now sorted later in the process after the per-region file lists
are determined.

Change-Id: I0bba381d09dc4b99e2fe5cae16ff7ffcb5b3aa82
Signed-off-by: Patrick Georgi <pgeorgi@chromium.org>
Reviewed-on: https://review.coreboot.org/16138
Tested-by: build bot (Jenkins)
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
Reviewed-by: Aaron Durbin <adurbin@chromium.org>
2 years agobuild system: order per-region files to optimize placement success
Patrick Georgi [Wed, 10 Aug 2016 11:25:02 +0000 (13:25 +0200)]
build system: order per-region files to optimize placement success

Make sure that files with a fixed position are placed first (whose order
doesn't matter: either they collide or they don't), then all aligned
files (where we just hope that the right thing happens) and finally the
files with no further requirements (again, hope).
It's still a pretty good heuristic given a typical coreboot image.

The global sorting that happens earlier in the build flow will be
removed in the future to make room for per-region requirements.

Change-Id: I269c00b2ece262c95d310b76a6651c9574badb58
Signed-off-by: Patrick Georgi <pgeorgi@chromium.org>
Reviewed-on: https://review.coreboot.org/16137
Tested-by: build bot (Jenkins)
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
Reviewed-by: Aaron Durbin <adurbin@chromium.org>
2 years agointel/amenia: Add MAINBOARD_FAMILY for amenia
Bora Guvendik [Tue, 9 Aug 2016 22:31:30 +0000 (15:31 -0700)]
intel/amenia: Add MAINBOARD_FAMILY for amenia

BUG=chrome-os-partner:51844
TEST=Boot to chrome

Change-Id: I66178cc75872f14941434081d9650a569a084d04
Signed-off-by: Bora Guvendik <bora.guvendik@intel.com>
Reviewed-on: https://review.coreboot.org/16135
Tested-by: build bot (Jenkins)
Reviewed-by: Martin Roth <martinroth@google.com>
2 years agointel/amenia: set default value for BOOT_MEDIA_SPI_BUS
Bora Guvendik [Tue, 9 Aug 2016 21:13:36 +0000 (14:13 -0700)]
intel/amenia: set default value for BOOT_MEDIA_SPI_BUS

BUG=chrome-os-partner:51844
TEST=Boot to chrome

Change-Id: I60d411482812d98cb8dd11d66b0fc96ea9bae895
Signed-off-by: Bora Guvendik <bora.guvendik@intel.com>
Reviewed-on: https://review.coreboot.org/16134
Tested-by: build bot (Jenkins)
Reviewed-by: Martin Roth <martinroth@google.com>
2 years agointel/amenia: Select UART_FOR_CONSOLE for amenia
Bora Guvendik [Tue, 9 Aug 2016 20:44:02 +0000 (13:44 -0700)]
intel/amenia: Select UART_FOR_CONSOLE for amenia

Set default value for UART port

BUG=chrome-os-partner:51844
TEST=Boot to chrome and check console

Change-Id: I5e76066e0ff531303595dcd5a99f2f8db379e89b
Signed-off-by: Bora Guvendik <bora.guvendik@intel.com>
Reviewed-on: https://review.coreboot.org/16133
Tested-by: build bot (Jenkins)
Reviewed-by: Martin Roth <martinroth@google.com>
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
2 years agointel/amenia: Update flash size to 16MB
Bora Guvendik [Fri, 5 Aug 2016 21:18:20 +0000 (14:18 -0700)]
intel/amenia: Update flash size to 16MB

Update flash image size to 16MB and update image layout
in flashmap descriptor file.

BUG=chrome-os-partner:51844
TEST=Boot to chrome

Change-Id: Ibdfb2949d06aedc38ddcef1078c2d14abcfa2dac
Signed-off-by: Bora Guvendik <bora.guvendik@intel.com>
Reviewed-on: https://review.coreboot.org/16083
Tested-by: build bot (Jenkins)
Reviewed-by: Martin Roth <martinroth@google.com>
2 years agosoc/intel/common: Add support for serial console based ACPI debug
Subrata Banik [Fri, 5 Aug 2016 12:55:55 +0000 (18:25 +0530)]
soc/intel/common: Add support for serial console based ACPI debug

This patch enables serial debug functionality for ASL code based on
UART type(legacy/LPSS).

From Skylake onwards all Intel platform uses LPSS based UART for serial
console hence provide option to redirect ASL log over LPSS UART.

Example:
Name (OBJ, 0x12)
APRT (OBJ)
APRT ("CORE BOOT")

Output:
0x12
CORE BOOT

BUG=none
BRANCH=none
TEST=Built and boot kunimitsu to ensure to be able to get ASL console log.

Change-Id: I18c65654b8eb1ac27af1f283d413376fd79d47db
Signed-off-by: Subrata Banik <subrata.banik@intel.com>
Reviewed-on: https://review.coreboot.org/16070
Tested-by: build bot (Jenkins)
Reviewed-by: Aaron Durbin <adurbin@chromium.org>
2 years agofsp_Broadwell_DE: Do not set IRQ3 and IRQ4 to level
Kevin Paul Herbert [Thu, 4 Aug 2016 21:35:30 +0000 (14:35 -0700)]
fsp_Broadwell_DE: Do not set IRQ3 and IRQ4 to level

When booting Linux as a coreboot payload, serial access does not work
properly. This is because the setup code erroneously sets IRQ3 and
IRQ4 to level. The UART on Broadwell is 8250/16550 compatible, thus
ISA and edge-triggered.

This change is not necessary on the non-FSP version of Broadwell support.
The non-FSP version does not set these IRQ overrides.

Fix verified booting Linux 4.6.0-rc2 on Intel Camelback Mountain CRB,
using Intel FSP 1.0.

Change-Id: I17b466676e7f4891c3e75ce6208e1580c9eaf742
Signed-off-by: Kevin Paul Herbert <kevin@trippers.org>
Reviewed-on: https://review.coreboot.org/16065
Tested-by: build bot (Jenkins)
Reviewed-by: Patrick Georgi <pgeorgi@google.com>
2 years agobuildgcc: printf no-color before quiting
Zheng Bao [Fri, 5 Aug 2016 05:41:51 +0000 (13:41 +0800)]
buildgcc: printf no-color before quiting

On some kind of terms (shell in emacs), the color-ctrl
letters don't work. The backspaces can not delete
correct number of letters. So we don't print color-ctrl
letters in loop.

Change-Id: I1f1729095e8968a9344ed9f1f278f7c78f7110e9
Signed-off-by: Zheng Bao <fishbaozi@gmail.com>
Reviewed-on: https://review.coreboot.org/16066
Tested-by: build bot (Jenkins)
Reviewed-by: Stefan Reinauer <stefan.reinauer@coreboot.org>
Reviewed-by: Martin Roth <martinroth@google.com>
2 years agorockchip/common: Set weekday to unknown in rtc_get()
Martin Roth [Tue, 26 Jul 2016 15:03:25 +0000 (09:03 -0600)]
rockchip/common: Set weekday to unknown in rtc_get()

Prior to this patch, time->wday was not being initialized in rtc_get(),
but was still being used by rtc_display() to print a day.

Set to -1 which gets printed as "unknown ".

Fixes coverity issue 1357459 - Uninitialized scalar variable

Change-Id: Idecb7968f854df997b58a342e1a06a879f299394
Signed-off-by: Martin Roth <martinroth@google.com>
Reviewed-on: https://review.coreboot.org/15899
Tested-by: build bot (Jenkins)
Reviewed-by: Patrick Georgi <pgeorgi@google.com>
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
Reviewed-by: Julius Werner <jwerner@chromium.org>
2 years agosoc/intel/quark: Switch to using serial routines for FSP
Lee Leahy [Sat, 6 Aug 2016 16:51:35 +0000 (09:51 -0700)]
soc/intel/quark: Switch to using serial routines for FSP

Switch from passing FSP the serial port address to passing FSP the
serial port output routine.  This enables coreboot to use any UART in
the system and also log the FSP output.

TEST=Build and run on Galileo Gen2

Change-Id: I67d820ea0360a3188480455dd2595be7f2debd5c
Signed-off-by: Lee Leahy <leroy.p.leahy@intel.com>
Reviewed-on: https://review.coreboot.org/16105
Tested-by: build bot (Jenkins)
Reviewed-by: Martin Roth <martinroth@google.com>
2 years agodrivers/intel/fsp2_0: Add fsp_write_line function
Lee Leahy [Tue, 9 Aug 2016 16:04:45 +0000 (09:04 -0700)]
drivers/intel/fsp2_0: Add fsp_write_line function

Add fsp_write_line function which may be called by FSP to output debug
serial data to the console.

TEST=Build and run on Galileo Gen2

Change-Id: If7bfcea1af82209dcdc5a9f9f2d9334842c1595e
Signed-off-by: Lee Leahy <leroy.p.leahy@intel.com>
Reviewed-on: https://review.coreboot.org/16129
Tested-by: build bot (Jenkins)
Reviewed-by: Aaron Durbin <adurbin@chromium.org>
2 years agodrivers/intel/fsp1_1: Add fsp_write_line function
Lee Leahy [Tue, 9 Aug 2016 16:02:13 +0000 (09:02 -0700)]
drivers/intel/fsp1_1: Add fsp_write_line function

Add fsp_write_line function which may be called by FSP to output debug
serial data to the console.

TEST=Build and run on Galileo Gen2

Change-Id: Ib01aef448798e47ac613b38eb20bf25537b9221f
Signed-off-by: Lee Leahy <leroy.p.leahy@intel.com>
Reviewed-on: https://review.coreboot.org/16128
Tested-by: build bot (Jenkins)
Reviewed-by: Martin Roth <martinroth@google.com>
2 years agoconsole: Add write line routine
Lee Leahy [Tue, 9 Aug 2016 15:59:30 +0000 (08:59 -0700)]
console: Add write line routine

Add write line routine which is called indirectly by FSP.

TEST=Build and run on Galileo Gen2.

Change-Id: Idefb6e9ebe5a2b614055dabddc1882bfa3bba673
Signed-off-by: Lee Leahy <leroy.p.leahy@intel.com>
Reviewed-on: https://review.coreboot.org/16127
Tested-by: build bot (Jenkins)
Reviewed-by: Martin Roth <martinroth@google.com>
2 years agobuild system: change addition order of files to go by region
Patrick Georgi [Tue, 9 Aug 2016 18:19:19 +0000 (20:19 +0200)]
build system: change addition order of files to go by region

Instead of adding each file in all requested regions, sort by region,
then by file.

This is in preparation of per-region file options
(eg. position, alignment)

Change-Id: Ide09a1c8840279380294a059bbd5d2f9f0cba780
Signed-off-by: Patrick Georgi <pgeorgi@chromium.org>
Reviewed-on: https://review.coreboot.org/16130
Tested-by: build bot (Jenkins)
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
Reviewed-by: Martin Roth <martinroth@google.com>
2 years agoutil/cbfstool: Initialize elf_writer pointer to avoid crash
Jonathan Neuschäfer [Tue, 9 Aug 2016 13:01:58 +0000 (15:01 +0200)]
util/cbfstool: Initialize elf_writer pointer to avoid crash

If some error happens in cbfs_payload_make_elf, the code jumps to "out",
and elf_writer_destroy(ew) is called. This may happen before an elf
writer is allocated.
To avoid accessing an uninitialized pointer, initialize ew to NULL;
elf_writer_destroy will perform no action in this case.

Change-Id: I5f1f9c4d37f2bdeaaeeca7a15720c7b4c963d953
Reported-By: Coverity Scan (1361475)
Signed-off-by: Jonathan Neuschäfer <j.neuschaefer@gmx.net>
Reviewed-on: https://review.coreboot.org/16124
Tested-by: build bot (Jenkins)
Reviewed-by: Aaron Durbin <adurbin@chromium.org>
2 years agoMakefiles: Use $(MAINBOARD_DIR) instead of $(CONFIG_MAINBOARD_DIR)
Martin Roth [Mon, 8 Aug 2016 22:14:55 +0000 (16:14 -0600)]
Makefiles: Use $(MAINBOARD_DIR) instead of $(CONFIG_MAINBOARD_DIR)

The variable MAINBOARD_DIR already has the quotes stripped off.

Change-Id: Ib434ce92bdbc49180fb3f713b26d65ba4cf8c441
Signed-off-by: Martin Roth <martinroth@chromium.org>
Reviewed-on: https://review.coreboot.org/16117
Tested-by: build bot (Jenkins)
Reviewed-by: Lee Leahy <leroy.p.leahy@intel.com>
Reviewed-by: Patrick Georgi <pgeorgi@google.com>
2 years agoMakefile.inc: Strip CONFIG_DEVICETREE quotes at top of makefile
Martin Roth [Mon, 8 Aug 2016 21:12:11 +0000 (15:12 -0600)]
Makefile.inc: Strip CONFIG_DEVICETREE quotes at top of makefile

Minor change - Instead of stripping the quotes from CONFIG_DEVICETREE
inline, add it to the location where we normalize all the other Kconfig
variables.

Change-Id: Idbc58179c7b45160afef7d7e44f9b3b334f8c4a7
Signed-off-by: Martin Roth <martinroth@chromium.org>
Reviewed-on: https://review.coreboot.org/16116
Tested-by: build bot (Jenkins)
Reviewed-by: Patrick Georgi <pgeorgi@google.com>
2 years agogoogle/reef: Add mainboard handler function for gpio SMI
Shaunak Saha [Mon, 25 Jul 2016 05:54:13 +0000 (22:54 -0700)]
google/reef: Add mainboard handler function for gpio SMI

This patch adds mainboard_smi_gpi_handler which handles the
SMI event. This can happen in situations like lidclose and
system goes to shutdown.

BUG=chrome-os-partner:54977
TEST=When system is in firmware mode executing the command
     lidclose from ec console shuts down the system.

Change-Id: I8ff6001e48dcbbd4cee5097e759352d8fea6189b
Signed-off-by: Shaunak Saha <shaunak.saha@intel.com>
Reviewed-on: https://review.coreboot.org/15834
Tested-by: build bot (Jenkins)
Reviewed-by: Aaron Durbin <adurbin@chromium.org>
2 years agosoc/apollolake: add GPIO SMI support
Shaunak Saha [Mon, 25 Jul 2016 03:50:12 +0000 (20:50 -0700)]
soc/apollolake: add GPIO SMI support

GPIOs which trigger SMIs set the GPIO_SMI_STS status bits in SMI_STS
register. This patch also sets the SMI_EN bit in enable register for
each community based on GPIOROUTSMI bit in gpio pad. When SMI on a
gpio happens status needs to be gathered on gpio number which is done
by reading the GPI_SMI_STS and GPI_SMI_EN registers.

BUG=chrome-os-partner:54977
TEST=When system is in firmware mode executing the command
     lidclose from ec console shuts down the system.

Change-Id: Id89a526106d1989c2bd3416ab81913e6cf743d17
Signed-off-by: Shaunak Saha <shaunak.saha@intel.com>
Reviewed-on: https://review.coreboot.org/15833
Reviewed-by: Aaron Durbin <adurbin@chromium.org>
Tested-by: build bot (Jenkins)
2 years agoelog: Ensure eventlog will always get initialized when configured in
Julius Werner [Mon, 8 Aug 2016 23:18:28 +0000 (16:18 -0700)]
elog: Ensure eventlog will always get initialized when configured in

Commit 0d9cd92e (chromeos: Clean up elog handling) removed the
individual elog_init() calls from mainboards that did them and automated
adding certain events through the boot state machine. Unfortunately,
the new code would sometimes not log any specific event at all, and
thereby also never call elog_init() (through elog_add_event()) which
adds the "System boot" event.

We can assume that any board that configures the eventlog at all
actually wants to use it, so let's just add another call to elog_init()
to the boot state machine so we can ensure it gets called at least once.

BRANCH=None
BUG=chrome-os-partner:56001
TEST=Booted Kevin, confirmed that eventlog code runs again.

Change-Id: Ibe7bfc94b3e3d11ba881399a39f9915991c89d8c
Signed-off-by: Julius Werner <jwerner@chromium.org>
Reviewed-on: https://review.coreboot.org/16118
Reviewed-by: Aaron Durbin <adurbin@chromium.org>
Tested-by: build bot (Jenkins)
2 years agodrivers/elog: provide return status for all operations
Aaron Durbin [Sat, 6 Aug 2016 15:02:37 +0000 (10:02 -0500)]
drivers/elog: provide return status for all operations

Instead of relying on global state to determine if an error
occurred provide the ability to know if an add or shrink
operation is successful. Now the call chains report the
error back up the stack and out to the callers.

BUG=chrome-os-partner:55932

Change-Id: Id4ed4d93e331f1bf16e038df69ef067446d00102
Signed-off-by: Aaron Durbin <adurbin@chromium.org>
Reviewed-on: https://review.coreboot.org/16104
Tested-by: build bot (Jenkins)
Reviewed-by: Furquan Shaikh <furquan@google.com>
2 years agodrivers/elog: clean up SMBIOS related code
Aaron Durbin [Sat, 6 Aug 2016 02:45:54 +0000 (21:45 -0500)]
drivers/elog: clean up SMBIOS related code

Don't conditionally compile parts of the code. The unused pieces
get culled by the linker, and the #if's just clutter things up.

BUG=chrome-os-partner:55932

Change-Id: Ic18b2deb0cfef7167c05f0a641eae2f4cdc848ee
Signed-off-by: Aaron Durbin <adurbin@chromium.org>
Reviewed-on: https://review.coreboot.org/16102
Tested-by: build bot (Jenkins)
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
Reviewed-by: Furquan Shaikh <furquan@google.com>
2 years agodrivers/elog: consolidate checks in elog_find_flash()
Aaron Durbin [Fri, 5 Aug 2016 20:41:37 +0000 (15:41 -0500)]
drivers/elog: consolidate checks in elog_find_flash()

There were checks against global variables trying to determine
failing cases of elog_find_flash(). Instead move the checks
into elog_find_flash() and return value indicating failure.
A minimum 4KiB check was added to ensure the eventlog is at
least that size which makes the heuristic checks cleaner.

BUG=chrome-os-partner:55932

Change-Id: I4d9d13148555e05d4f217a10f995831a0e437fc3
Signed-off-by: Aaron Durbin <adurbin@chromium.org>
Reviewed-on: https://review.coreboot.org/16101
Tested-by: build bot (Jenkins)
Reviewed-by: Martin Roth <martinroth@google.com>
2 years agogoogle/gru: Fix rk3399-gru write protect
Douglas Anderson [Thu, 4 Aug 2016 16:27:40 +0000 (09:27 -0700)]
google/gru: Fix rk3399-gru write protect

The write protect GPIO is active high, not active low.

After fixing I can see this after removing the write-protect screw:
$ crossystem | grep wpsw_boot
wpsw_boot              = 0

Putting the screw in shows:
$ crossystem | grep wpsw_boot
wpsw_boot              = 1

Caution: this CL contains explicit material.  It explicitly sets the
pullup on the WP GPIO even though that's the boot default.

BRANCH=None
BUG=chrome-os-partner:55933
TEST=See desc.

Change-Id: I23e17e3bbbe7dcd83e81814de46117491e61baaa
Signed-off-by: Martin Roth <martinroth@chromium.org>
Original-Commit-Id: e6969f4be42c00c6e88bbb14929cf0454462ad21
Original-Change-Id: Ie65db9cf182b0a0a05ae412f86904df6b239e0f4
Original-Signed-off-by: Douglas Anderson <dianders@chromium.org>
Original-Reviewed-on: https://chromium-review.googlesource.com/366131
Original-Tested-by: Brian Norris <briannorris@chromium.org>
Original-Reviewed-by: Julius Werner <jwerner@chromium.org>
Reviewed-on: https://review.coreboot.org/16115
Tested-by: build bot (Jenkins)
Reviewed-by: Furquan Shaikh <furquan@google.com>
2 years agodrivers/elog: remove unnecessary global state
Aaron Durbin [Fri, 5 Aug 2016 20:24:33 +0000 (15:24 -0500)]
drivers/elog: remove unnecessary global state

There were 3 variables indicating the state of the event log
region. However, there's no need to keep track of those
individually. The only thing required is to know is if
elog_scan_flash() failed. There's no other tracking required
beyond that.

BUG=chrome-os-partner:55932

Change-Id: I88ad32091d3c37966a2ac6272f8ad95bcc8c4270
Signed-off-by: Aaron Durbin <adurbin@chromium.org>
Reviewed-on: https://review.coreboot.org/16100
Tested-by: build bot (Jenkins)
Reviewed-by: Martin Roth <martinroth@google.com>
2 years agodrivers/elog: sync events to non-volatile storage last
Aaron Durbin [Fri, 5 Aug 2016 19:33:53 +0000 (14:33 -0500)]
drivers/elog: sync events to non-volatile storage last

There were multiple paths where writes and erases of the flash
were being done. Instead provide a single place for synchronizing
the non-volatile storage from the mirrored event log. This
synchronization point resides as the very last thing done when
adding an event to the log. The shrinking check happens before
committing the event to non-volatile storage so there's no need
to attempt a shrink in elog_init() because any previous events
committed already honored the full threshold.

BUG=chrome-os-partner:55932

Change-Id: Iaec9480eb3116fdc2f823c25d028a4cfb65a6eaf
Signed-off-by: Aaron Durbin <adurbin@chromium.org>
Reviewed-on: https://review.coreboot.org/16099
Tested-by: build bot (Jenkins)
Reviewed-by: Furquan Shaikh <furquan@google.com>
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
2 years agoflashmap: Allocate at least one entry in kv_pair_new()
Stefan Reinauer [Thu, 21 Apr 2016 05:13:37 +0000 (22:13 -0700)]
flashmap: Allocate at least one entry in kv_pair_new()

Change-Id: I971fa85ed977884d050790560a5a8f2ce955eb7c
Signed-off-by: Stefan Reinauer <stefan.reinauer@coreboot.org>
Reviewed-on: https://review.coreboot.org/14444
Tested-by: build bot (Jenkins)
Reviewed-by: Patrick Georgi <pgeorgi@google.com>
2 years agopayloads: add Kconfig option for bayou
Antonello Dettori [Mon, 25 Jul 2016 12:23:46 +0000 (14:23 +0200)]
payloads: add Kconfig option for bayou

Add an option to add bayou as the primary payload.

Change-Id: I8c0164344537b82870198b13ef6fdf20e7d095ef
Signed-off-by: Antonello Dettori <dev@dettori.io>
Reviewed-on: https://review.coreboot.org/15954
Tested-by: build bot (Jenkins)
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
Reviewed-by: Omar Pakker
Reviewed-by: Patrick Georgi <pgeorgi@google.com>
2 years agogigbyte/ga-g41m-es2l: add cmos.layout and cmos.default
Arthur Heymans [Sat, 18 Jun 2016 21:26:46 +0000 (23:26 +0200)]
gigbyte/ga-g41m-es2l: add cmos.layout and cmos.default

This adds a cmos.layout and a cmos.default to ga-g41m-es2l.
This allows to set things like baud_rate, debug_level, etc.
from cmos.

Change-Id: I25df7a1f3a0ce486b96cfe05bda628f604b0baec
Signed-off-by: Arthur Heymans <arthur@aheymans.xyz>
Reviewed-on: https://review.coreboot.org/15493
Tested-by: build bot (Jenkins)
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
Reviewed-by: Patrick Georgi <pgeorgi@google.com>
2 years agox4x: make preallocated IGD memory a cmos option
Arthur Heymans [Sat, 18 Jun 2016 19:52:30 +0000 (21:52 +0200)]
x4x: make preallocated IGD memory a cmos option

This allows to set the preallocated memory for the IGD on x4x
using a cmos option.
If no cmos option is found a default value of 64M is used.

TESTED most options on ga-g41m-es2l with 2G dimm in one slot and 2x2G.
352M also works in contrast with gm45 where it is known to cause issues
with certain ram combinations.

Change-Id: I9051d080be82f6dfab37d353252e29b2ed1fca7f
Signed-off-by: Arthur Heymans <arthur@aheymans.xyz>
Reviewed-on: https://review.coreboot.org/15492
Tested-by: build bot (Jenkins)
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
Reviewed-by: Patrick Georgi <pgeorgi@google.com>
2 years agox4x: add non documented vram sizes
Arthur Heymans [Sat, 18 Jun 2016 19:08:58 +0000 (21:08 +0200)]
x4x: add non documented vram sizes

The Intel documtentation, "Intel ® 4 Series Chipset Family"
mentions the possibility of 1, 4, 8 and 16M of preallocated
memory for the IGD, but does not document this.

This allows to set those undocumented values.
TESTED on ga-g41m-es2l with 2G dimm in one slot and 2x2G.

Change-Id: I92beb8d78907d4514a5aaf69248dd607dcf227c0
Signed-off-by: Arthur Heymans <arthur@aheymans.xyz>
Reviewed-on: https://review.coreboot.org/15491
Tested-by: build bot (Jenkins)
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
Reviewed-by: Patrick Georgi <pgeorgi@google.com>
2 years agovendorcode/amd: Remove dead code
Patrick Georgi [Mon, 8 Aug 2016 09:43:43 +0000 (11:43 +0200)]
vendorcode/amd: Remove dead code

Change-Id: I6b21822d60d379cb8cd21b69c714a437bb7977ce
Signed-off-by: Patrick Georgi <pgeorgi@chromium.org>
Found-by: Coverity Scan #1254643 and others
Reviewed-on: https://review.coreboot.org/16112
Tested-by: build bot (Jenkins)
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
Reviewed-by: Kyösti Mälkki <kyosti.malkki@gmail.com>
2 years agosuperio/*: Relocate Kconfig to chip folder.
Omar Pakker [Fri, 29 Jul 2016 21:31:45 +0000 (23:31 +0200)]
superio/*: Relocate Kconfig to chip folder.

This moves the Kconfig from the Super I/O manufacturer folder
to the chip folder instead.
This makes new chip commits self-contained unit as
edits to the central Kconfig file are no longer required.

Change-Id: I7aee07919f2ae9204850c669e0ed3cb17d4de8cd
Signed-off-by: Omar Pakker <omarpakker+coreboot@gmail.com>
Reviewed-on: https://review.coreboot.org/15973
Tested-by: build bot (Jenkins)
Reviewed-by: Idwer Vollering <vidwer@gmail.com>
2 years agobuildgcc: Use upstream patch for aarch64 build issue
Paul Kocialkowski [Sun, 7 Aug 2016 09:42:01 +0000 (11:42 +0200)]
buildgcc: Use upstream patch for aarch64 build issue

Upstream proposed and merged a patch fixing the ARM Trusted Firmware
build issue that occurs with recent version sof binutils. This includes
this patch instead of the previous one.

See binutils commit 7ea12e5c3ad54da440c08f32da09534e63e515ca:
"Fix the generation of alignment frags in code sections for AArch64."

The issue was reported at:
https://sourceware.org/bugzilla/show_bug.cgi?id=20364

Change-Id: I16a8043d3562107b8e84e93d3f3d768d26dac7e4
Signed-off-by: Paul Kocialkowski <contact@paulk.fr>
Reviewed-on: https://review.coreboot.org/16110
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
Tested-by: build bot (Jenkins)
Reviewed-by: Patrick Georgi <pgeorgi@google.com>
2 years agobuild system: drop commented out code
Patrick Georgi [Fri, 5 Aug 2016 11:10:39 +0000 (13:10 +0200)]
build system: drop commented out code

Change-Id: I7f36a317f0a7cf4634246a255be79e6bcb2b2442
Signed-off-by: Patrick Georgi <pgeorgi@chromium.org>
Found-by: Daisuke Nojiri <dnojiri@chromium.org>
Reviewed-on: https://review.coreboot.org/16067
Tested-by: build bot (Jenkins)
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
Reviewed-by: Kyösti Mälkki <kyosti.malkki@gmail.com>
2 years agosoc/intel/quark: Remove TODO message from FspUpdVpd.h
Lee Leahy [Mon, 8 Aug 2016 16:33:28 +0000 (09:33 -0700)]
soc/intel/quark: Remove TODO message from FspUpdVpd.h

Remove the TODO message from FspUpdVpd.h

TEST=Build and run on Galileo Gen2

Change-Id: Icd565c6062ef59b1e4a68310bb6f9ed62fb014af
Signed-off-by: Lee Leahy <leroy.p.leahy@intel.com>
Reviewed-on: https://review.coreboot.org/16114
Tested-by: build bot (Jenkins)
Reviewed-by: Martin Roth <martinroth@google.com>
Reviewed-by: Giri P Mudusuru <giri.p.mudusuru@intel.com>
2 years agosoc/nvidia/tegra210: remove unused spi boot device support
Aaron Durbin [Sat, 6 Aug 2016 18:50:21 +0000 (13:50 -0500)]
soc/nvidia/tegra210: remove unused spi boot device support

All mainboards utilizing the t210 SoC use the CBFS spi wrapper
for the boot device support. Therefore, remove the unutilized
spi boot device.

BUG=chrome-os-partner:55932

Change-Id: Id49ca6e5bf353bba8c03e62f5a9a873ad1ce7081
Signed-off-by: Aaron Durbin <adurbin@chromium.org>
Reviewed-on: https://review.coreboot.org/16109
Tested-by: build bot (Jenkins)
Reviewed-by: Furquan Shaikh <furquan@google.com>
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
2 years agosoc/nvidia/tegra132: remove tegra132 support
Aaron Durbin [Sat, 6 Aug 2016 18:42:37 +0000 (13:42 -0500)]
soc/nvidia/tegra132: remove tegra132 support

As no more mainboards are utilizing this SoC support code remove
it. It can be resurrected if ever needed.

BUG=chrome-os-partner:55932

Change-Id: Ic3caf6e6c9b62d012679b996abaa525c8bf679a9
Signed-off-by: Aaron Durbin <adurbin@chromium.org>
Reviewed-on: https://review.coreboot.org/16108
Tested-by: build bot (Jenkins)
Reviewed-by: Furquan Shaikh <furquan@google.com>
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
2 years agomainboard/google/rush_ryu: remove rush_ryu mainboard
Aaron Durbin [Sat, 6 Aug 2016 18:41:51 +0000 (13:41 -0500)]
mainboard/google/rush_ryu: remove rush_ryu mainboard

The rush_ryu board was a development platform that never made it into
a product. Remove it as it's not available to anyone.

BUG=chrome-os-partner:55932

Change-Id: Ia3836ff8cade3009730543177a66736ae197572b
Signed-off-by: Aaron Durbin <adurbin@chromium.org>
Reviewed-on: https://review.coreboot.org/16107
Tested-by: build bot (Jenkins)
Reviewed-by: Furquan Shaikh <furquan@google.com>
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
2 years agomainboard/google/rush: remove rush mainboard
Aaron Durbin [Sat, 6 Aug 2016 18:40:11 +0000 (13:40 -0500)]
mainboard/google/rush: remove rush mainboard

The rush board was a development platform that never made it into
a product. Remove it as it's not available to anyone.

BUG=chrome-os-partner:55932

Change-Id: I0f77bb791491509da7bd9cf25050e01c2f734a2f
Signed-off-by: Aaron Durbin <adurbin@chromium.org>
Reviewed-on: https://review.coreboot.org/16106
Tested-by: build bot (Jenkins)
Reviewed-by: Furquan Shaikh <furquan@google.com>
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
2 years agogoogle/gru: Update board/RAM ID ADC values
Julius Werner [Mon, 1 Aug 2016 21:33:26 +0000 (14:33 -0700)]
google/gru: Update board/RAM ID ADC values

Looks like our hardware guys have decided to change some voltage ranges
in the Gru/Kevin ADC IDs since we last wrote a table. This patch updates
it to the latest values from the Spreadsheet of Truth. Also adds further
values up to rev15.

BRANCH=none
BUG=none
TEST=none

Change-Id: I1aa093ca3abe952afd658eb7da01b325f798eaa0
Signed-off-by: Martin Roth <martinroth@chromium.org>
Original-Commit-Id: e42b4685c91f01ce1cff61638b17042be9d575fd
Original-Change-Id: I646fd03dc385df1a8f0af8cb85ff3128cc31f8d8
Original-Signed-off-by: Julius Werner <jwerner@chromium.org>
Original-Reviewed-on: https://chromium-review.googlesource.com/365111
Original-Tested-by: Douglas Anderson <dianders@chromium.org>
Original-Reviewed-by: Douglas Anderson <dianders@chromium.org>
Reviewed-on: https://review.coreboot.org/16053
Tested-by: build bot (Jenkins)
Reviewed-by: Furquan Shaikh <furquan@google.com>
2 years agorockchip/rk3399: sdram: correct read obs and set DQS driver register
Lin Huang [Tue, 26 Jul 2016 02:29:45 +0000 (10:29 +0800)]
rockchip/rk3399: sdram: correct read obs and set DQS driver register

We were using the wrong register when reading the obs value and setting
the DQS driver.  This did not affect LPDDR3 performance, but still needs
to be fixed.

BUG=none
BRANCH=none
TEST=boot from kevin

Change-Id: I144f575e27fba11872a8c5463ab1e2986f385ede
Signed-off-by: Martin Roth <martinroth@chromium.org>
Original-Commit-Id: 98221e6b03fc09cbf62af29a270e7a8aa8dfb986
Original-Change-Id: Ie179f9a2955c5712951d40b3ada9c14a51c09c8d
Original-Signed-off-by: Lin Huang <hl@rock-chips.com>
Original-Reviewed-on: https://chromium-review.googlesource.com/363170
Original-Reviewed-by: Julius Werner <jwerner@chromium.org>
Reviewed-on: https://review.coreboot.org/16052
Tested-by: build bot (Jenkins)
Reviewed-by: Furquan Shaikh <furquan@google.com>
2 years agosoc/qualcomm/ipq40xx: Use block mode for I2C
Varadarajan Narayanan [Tue, 17 May 2016 05:48:30 +0000 (11:18 +0530)]
soc/qualcomm/ipq40xx: Use block mode for I2C

In FIFO mode, the I2C driver was not able to fetch
more than 32 bytes of data from the TPM device. Switch to
block mode to be able to read more data.

BUG=chrome-os-partner:51096
TEST=TPM commands succeed
BRANCH=None

Change-Id: Ib52a1b03667f61a08ce048d38407a5b60abf660d
Signed-off-by: Martin Roth <martinroth@chromium.org>
Original-Commit-Id: fbcd40dc67d796d3e31675bd35321282667fe9fa
Original-Change-Id: I765b76f9d7743f6d387470de594fb6eee99e08ca
Original-Signed-off-by: Varadarajan Narayanan <varada@codeaurora.org>
Original-Reviewed-on: https://chromium-review.googlesource.com/357960
Original-Commit-Ready: Kan Yan <kyan@google.com>
Original-Tested-by: Kan Yan <kyan@google.com>
Original-Reviewed-by: Kan Yan <kyan@google.com>
Reviewed-on: https://review.coreboot.org/16051
Tested-by: build bot (Jenkins)
Reviewed-by: Furquan Shaikh <furquan@google.com>
2 years agodrivers/elog: treat offsets relative to start of mirror
Aaron Durbin [Thu, 4 Aug 2016 20:59:35 +0000 (15:59 -0500)]
drivers/elog: treat offsets relative to start of mirror

Instead of treating offsets relative to after the header make
the offsets relative to the in-memory mirror buffer. This
simplifies the logic in that all offsets are treated the same.
It also allows one to remove a global variable.

BUG=chrome-os-partner:55932

Change-Id: I42491e05755d414562b02b6f9ae47f5c357d2f8a
Signed-off-by: Aaron Durbin <adurbin@chromium.org>
Reviewed-on: https://review.coreboot.org/16098
Tested-by: build bot (Jenkins)
Reviewed-by: Furquan Shaikh <furquan@google.com>
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
2 years agodrivers/elog: use region_device for mirroring into ram
Aaron Durbin [Thu, 4 Aug 2016 19:25:59 +0000 (14:25 -0500)]
drivers/elog: use region_device for mirroring into ram

A region_device can be used to represent the in-memory mirror
of the event log. The region_device infrastructure has builtin
bounds checking so there's no need to duplicate that. In addition,
it allows for removing much of the math juggling for the buffer
size, etc.

BUG=chrome-os-partner:55932

Change-Id: Ic7fe9466019640b449257c5905ed919ac522bb58
Signed-off-by: Aaron Durbin <adurbin@chromium.org>
Reviewed-on: https://review.coreboot.org/16097
Tested-by: build bot (Jenkins)
Reviewed-by: Furquan Shaikh <furquan@google.com>
2 years agodrivers/elog: use offsets for checking cleared buffers
Aaron Durbin [Thu, 4 Aug 2016 15:21:06 +0000 (10:21 -0500)]
drivers/elog: use offsets for checking cleared buffers

There's only 2 users of checking if the event buffer is cleared
to the EOL value. Each were passing pointers of the in-memory
mirror while also doing calculations for the size to check. Since
the in-memory mirror is one big buffer the only thing required
to know is the offset to start checking from. The check is always
done through the end of the buffer.

BUG=chrome-os-partner:55932

Change-Id: Icd4a7edc74407d6578fc93e9eb533abd3aa17277
Signed-off-by: Aaron Durbin <adurbin@chromium.org>
Reviewed-on: https://review.coreboot.org/16096
Tested-by: build bot (Jenkins)
Reviewed-by: Furquan Shaikh <furquan@google.com>
2 years agosconfig: Reformat C code
Martin Roth [Sat, 6 Aug 2016 00:32:18 +0000 (18:32 -0600)]
sconfig: Reformat C code

Change-Id: Idfd1bd8240413026b992ae1382a57bccf9d8ddb5
Signed-off-by: Martin Roth <martinroth@google.com>
Reviewed-on: https://review.coreboot.org/16082
Tested-by: build bot (Jenkins)
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
Reviewed-by: Patrick Georgi <pgeorgi@google.com>
2 years agosconfig: Remove mainboard chip.h support
Martin Roth [Sat, 6 Aug 2016 00:15:06 +0000 (18:15 -0600)]
sconfig: Remove mainboard chip.h support

The mainboard chip.h files were (mostly) removed long ago.

Change-Id: I1d5a9381945427c96868fa17756e6ecabb1048b2
Signed-off-by: Martin Roth <martinroth@google.com>
Reviewed-on: https://review.coreboot.org/16080
Tested-by: build bot (Jenkins)
Reviewed-by: Kyösti Mälkki <kyosti.malkki@gmail.com>
Reviewed-by: Patrick Georgi <pgeorgi@google.com>
2 years agosconfig: Remove bootblock and Kconfig modes
Martin Roth [Fri, 5 Aug 2016 23:40:39 +0000 (17:40 -0600)]
sconfig: Remove bootblock and Kconfig modes

The command line parameters for these modes haven't worked in two
years and nobody noticed.  They're obviously not getting used, so
remove them.

TEST=Generate static.c before and after the change, verify they're
identical.

Change-Id: I1d746fb53a2f232155f663f4debc447d53d4cf6b
Signed-off-by: Martin Roth <martinroth@google.com>
Reviewed-on: https://review.coreboot.org/16079
Tested-by: build bot (Jenkins)
Reviewed-by: Patrick Georgi <pgeorgi@google.com>
2 years agosconfig: Update command line parameters
Martin Roth [Fri, 5 Aug 2016 21:46:56 +0000 (15:46 -0600)]
sconfig: Update command line parameters

Instead of having directories and file names hardcoded, pass in the full
path and filename of both the input and output files.

In the makefile, create variables for these values, and use them in
places that previously had the names and paths written out.

Change-Id: Icb6f536547ce3193980ec5d60c786a29755c2813
Signed-off-by: Martin Roth <martinroth@google.com>
Reviewed-on: https://review.coreboot.org/16078
Tested-by: build bot (Jenkins)
Reviewed-by: Aaron Durbin <adurbin@chromium.org>
Reviewed-by: Patrick Georgi <pgeorgi@google.com>
2 years agosconfig: pass in devicetree filename
Martin Roth [Tue, 24 Nov 2015 19:34:16 +0000 (12:34 -0700)]
sconfig: pass in devicetree filename

Instead of forcing the hardcoded 'devicetree.cb' filename under the
mainboard directory, this allows mainboards to select a filename for
the devicetree file.

This allows mainboard variants that need to use different devicetree
files to live under the same directory.

Change-Id: I761e676ba5d5f70d1fb86656b528f63db169fcef
Signed-off-by: Martin Roth <martinroth@google.com>
Reviewed-on: https://review.coreboot.org/12529
Tested-by: build bot (Jenkins)
Reviewed-by: Patrick Georgi <pgeorgi@google.com>
2 years agodrivers/elog: perform writes in terms of offsets
Aaron Durbin [Thu, 4 Aug 2016 03:21:29 +0000 (22:21 -0500)]
drivers/elog: perform writes in terms of offsets

Instead of taking pointers and back-calculating the
proper offset perform writes in terms of the offsets
within the elog region in flash.

BUG=chrome-os-partner:55932

Change-Id: I5fd65423f5a6e03825c788bc36417f509b58f64d
Signed-off-by: Aaron Durbin <adurbin@chromium.org>
Reviewed-on: https://review.coreboot.org/16095
Reviewed-by: Furquan Shaikh <furquan@google.com>
Tested-by: build bot (Jenkins)
2 years agolibpayload: lzma: Allocate scratchpad on the heap
Julius Werner [Sat, 6 Aug 2016 03:43:47 +0000 (20:43 -0700)]
libpayload: lzma: Allocate scratchpad on the heap

Allocating a 15980-byte scratchpad on the stack when your default stack
size is set to 16KB is really not a great idea. We're regularly
overflowing into the end of our heap when using LZMA in libpayload, and
just happen not to notice it because the heap rarely gets filled up all
the way. Of course, since we always *have* a heap in libpayload, the
much saner solution is to just use it directly to allocate the
scratchpad rather than accidentally grow backwards into it anyway.

Change-Id: Ibe4f02057a32bd156a126302178fa6fcab637d2c
Signed-off-by: Julius Werner <jwerner@chromium.org>
Reviewed-on: https://review.coreboot.org/16089
Tested-by: build bot (Jenkins)
Reviewed-by: Aaron Durbin <adurbin@chromium.org>
2 years agosupermicro/h8scm: Remove last unused chip.h file
Martin Roth [Sat, 6 Aug 2016 00:16:07 +0000 (18:16 -0600)]
supermicro/h8scm: Remove last unused chip.h file

Change-Id: I84d61c8ade6e42e314a31e1155b4d5628b16199a
Signed-off-by: Martin Roth <martinroth@google.com>
Reviewed-on: https://review.coreboot.org/16081
Tested-by: build bot (Jenkins)
Reviewed-by: Kyösti Mälkki <kyosti.malkki@gmail.com>
2 years agodrivers/elog: remove parameters from elog_flash_erase()
Aaron Durbin [Thu, 4 Aug 2016 02:57:15 +0000 (21:57 -0500)]
drivers/elog: remove parameters from elog_flash_erase()

The elog_flash_erase() was only called to erase the entire
elog region in flash. Therefore, drop the parameters and
perform the full erase.

BUG=chrome-os-partner:55932

Change-Id: I6590347ae60d407bc0df141e9196eb70532f8585
Signed-off-by: Aaron Durbin <adurbin@chromium.org>
Reviewed-on: https://review.coreboot.org/16094
Tested-by: build bot (Jenkins)
Reviewed-by: Furquan Shaikh <furquan@google.com>
2 years agodrivers/elog: remove unnecessary check in elog_shrink()
Aaron Durbin [Thu, 4 Aug 2016 04:01:50 +0000 (23:01 -0500)]
drivers/elog: remove unnecessary check in elog_shrink()

There was a check against the next event offset against
the shrink size in elog_shrink(). However, all calls
to elog_shrink() were conditionalized on the next
event offset exceeding the full threshold. The shrink
size is set to the minimum of the full threshold and
a percentage of the elog region size. Therefore, it's
impossible for the next event offset to be less than
the shrink size because full threshold is always greater
than or equal to the shrink size.

BUG=chrome-os-partner:55932

Change-Id: Ie6ff106f1c53c15aa36a82223a235a7ac97fd8c7
Signed-off-by: Aaron Durbin <adurbin@chromium.org>
Reviewed-on: https://review.coreboot.org/16093
Tested-by: build bot (Jenkins)
Reviewed-by: Furquan Shaikh <furquan@google.com>
2 years agodrivers/elog: use event region size when adding a clear event
Aaron Durbin [Wed, 3 Aug 2016 22:32:47 +0000 (17:32 -0500)]
drivers/elog: use event region size when adding a clear event

For the elog shrink case we log the number of bytes shrunk
from the event log. However, when clearing the log the
size recorded was the entire region size including the header
as well as the event region space. To be more consistent
mark the clearing event with the number of bytes actually
cleared out (excluding the header size).

BUG=chrome-os-partner:55932

Change-Id: I7c33da97bd29a90bfe975b1c6f148f181016f13f
Signed-off-by: Aaron Durbin <adurbin@chromium.org>
Reviewed-on: https://review.coreboot.org/16092
Tested-by: build bot (Jenkins)
Reviewed-by: Furquan Shaikh <furquan@google.com>
2 years agodrivers/elog: remove unused function
Aaron Durbin [Thu, 4 Aug 2016 05:34:57 +0000 (00:34 -0500)]
drivers/elog: remove unused function

get_rom_size() is no longer used. Remove it.

BUG=chrome-os-partner:55932

Change-Id: Id9fa8f67b67ee355243a5c763cfafa0ce76e9b2b
Signed-off-by: Aaron Durbin <adurbin@chromium.org>
Reviewed-on: https://review.coreboot.org/16088
Tested-by: build bot (Jenkins)
Reviewed-by: Furquan Shaikh <furquan@google.com>
2 years agosoc/intel/common: fix gsmi handler
Aaron Durbin [Sat, 6 Aug 2016 02:26:53 +0000 (21:26 -0500)]
soc/intel/common: fix gsmi handler

The gsmi_exec() expects the parameter to be a pointer
to the 32-bit register storage of the SMI save state.
The previous code was passing a pointer with the value
obtained from the saved-state -- not a pointer to the
storage of the register value. This bug causes gsmi
to not log events because it's interrogating the
parameter buffer itself as if it were a pointer.

BUG=chrome-os-partner:55932

Change-Id: I37981424f1414edad1456b31cad1b99020d57db6
Signed-off-by: Aaron Durbin <adurbin@chromium.org>
Reviewed-on: https://review.coreboot.org/16087
Tested-by: build bot (Jenkins)
Reviewed-by: Duncan Laurie <dlaurie@chromium.org>
Reviewed-by: Furquan Shaikh <furquan@google.com>
2 years agochromeos chipsets: select RTC usage
Aaron Durbin [Sat, 6 Aug 2016 02:23:37 +0000 (21:23 -0500)]
chromeos chipsets: select RTC usage

Since RTC is now a Kconfig ensure RTC is selected on the
x86 chipsets which are in Chrome OS devices. This allows
the eventlog to have proper timestamps instead of all
zeros.

BUG=chrome-os-partner:55993

Change-Id: I24ae7d9b3bf43a5791d4dc04aae018ce17fda72b
Signed-off-by: Aaron Durbin <adurbin@chromium.org>
Reviewed-on: https://review.coreboot.org/16086
Reviewed-by: Duncan Laurie <dlaurie@chromium.org>
Tested-by: build bot (Jenkins)
Reviewed-by: Furquan Shaikh <furquan@google.com>
2 years agovendorcode/google/chromeos: remove unnecessary includes in elog.c
Aaron Durbin [Thu, 4 Aug 2016 05:31:18 +0000 (00:31 -0500)]
vendorcode/google/chromeos: remove unnecessary includes in elog.c

The elog.c file had stale includes no longer needed. Remove them.

BUG=chrome-os-partner:55932

Change-Id: I891a57d08281c3c56e9d35489d6dea6c47eaa27b
Signed-off-by: Aaron Durbin <adurbin@chromium.org>
Reviewed-on: https://review.coreboot.org/16085
Tested-by: build bot (Jenkins)
Reviewed-by: Duncan Laurie <dlaurie@chromium.org>
Reviewed-by: Furquan Shaikh <furquan@google.com>
2 years agocommonlib/region: make buffer argument const for writeat
Aaron Durbin [Thu, 4 Aug 2016 19:33:58 +0000 (14:33 -0500)]
commonlib/region: make buffer argument const for writeat

The buffer for writeat() should be marked as const as
the contents won't be manipulated within the call.

BUG=chrome-os-partner:55932

Change-Id: I968570c1cf80f918a07b97af625a56f11b5889c1
Signed-off-by: Aaron Durbin <adurbin@chromium.org>
Reviewed-on: https://review.coreboot.org/16084
Tested-by: build bot (Jenkins)
Reviewed-by: Duncan Laurie <dlaurie@chromium.org>
Reviewed-by: Furquan Shaikh <furquan@google.com>
2 years agosoc/intel/skylake: Clean up SoC ASL code.
Barnali Sarkar [Thu, 4 Aug 2016 12:16:55 +0000 (17:46 +0530)]
soc/intel/skylake: Clean up SoC ASL code.

List of changes done here in this patch

1. Remove CARD definition from EMMC and SD Card Controller in scs.asl
since _RMV method does not get evaluated while setting up removable
attribute in sysfs in kernel.
"cat /sys/block/mmcblk1/removable" this command always returns 0.

This CARD Device includes _ADR which follows SDIO Bus format. But,
SD/EMMC sits on PCI Bus.
Hence this CARD Device specific _ADR code is also not needed.

2. Remove Base Address for ACPI debug output memory buffer in
systemagent.asl as it is not getting used throughout the code.

BUG=none
BRANCH=none
TEST=Build and boot kunimitsu

Change-Id: I29effaffdafcc21e26445ec3c54aedecdbc50274
Signed-off-by: Barnali Sarkar <barnali.sarkar@intel.com>
Reviewed-on: https://review.coreboot.org/16068
Tested-by: build bot (Jenkins)
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
Reviewed-by: Duncan Laurie <dlaurie@chromium.org>
Reviewed-by: Subrata Banik <subrata.banik@intel.com>
2 years agoskylake/devicetree: Add PIRQ Routing programming
Barnali Sarkar [Wed, 3 Aug 2016 06:45:22 +0000 (12:15 +0530)]
skylake/devicetree: Add PIRQ Routing programming

Program PIRQ Routing with correct values, as done by FSP, and also in
'soc/intel/skylake/romstage/pch.c' file. If not done, these values get
overridden by "0" during PxRC -> PIRQ programming in ramstage, in
'soc/intel/skylake/lpc.c' file pch_pirq_init()function.

BUG=none
BRANCH=none
TEST=Build and boot kunimitsu

Change-Id: Ibeb9a64824a71c253e45d6a1c6088abd737cf046
Signed-off-by: Barnali Sarkar <barnali.sarkar@intel.com>
Reviewed-on: https://review.coreboot.org/16044
Tested-by: build bot (Jenkins)
Reviewed-by: Aaron Durbin <adurbin@chromium.org>
Reviewed-by: Subrata Banik <subrata.banik@intel.com>
2 years agosoc/intel/skylake: Cleanup patch for Skylake SoC
Barnali Sarkar [Tue, 2 Aug 2016 12:19:56 +0000 (17:49 +0530)]
soc/intel/skylake: Cleanup patch for Skylake SoC

Here is the list of items of code cleanup
1. Define TCO registers in smbus.h and not in pmc.h (as per EDS).
2. Include smbus.h wherever these TCO register defines were used.
3. Remove duplication of define in gpio_defs.h.
4. Remove unnecessary console.h include from memmap.h as no prints done.
5. Remove unnecessary comment from pch.c.

BUG=none
BRANCH=none
TEST=Built and boot kunimitsu.

Change-Id: Ibe6d2537ddde3c1c7f8ea5ada1bfaa9be79c0e3b
Signed-off-by: Barnali Sarkar <barnali.sarkar@intel.com>
Reviewed-on: https://review.coreboot.org/16027
Tested-by: build bot (Jenkins)
Reviewed-by: Rizwan Qureshi <rizwan.qureshi@intel.com>
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
Reviewed-by: Aaron Durbin <adurbin@chromium.org>
Reviewed-by: Subrata Banik <subrata.banik@intel.com>
2 years agogoogle/reef: Configure SDIO D1 to enable SCS Power Gating
Jagadish Krishnamoorthy [Thu, 4 Aug 2016 17:17:22 +0000 (10:17 -0700)]
google/reef: Configure SDIO D1 to enable SCS Power Gating

SDIO D1 pin needs to be configured as Native mode to
enable SCS Power Gating.

BUG=chrome-os-partner:54251
TEST=Verify SCS Power Gating

Change-Id: Ic33b26443203217678e11d195eb965a7e628ad82
Signed-off-by: Jagadish Krishnamoorthy <jagadish.krishnamoorthy@intel.com>
Reviewed-on: https://review.coreboot.org/16062
Tested-by: build bot (Jenkins)
Reviewed-by: Aaron Durbin <adurbin@chromium.org>
2 years agogoogle/chromeec: Enable/Disable ccache with config variable
Martin Roth [Tue, 2 Aug 2016 21:32:51 +0000 (15:32 -0600)]
google/chromeec: Enable/Disable ccache with config variable

If the CONFIG_CCACHE variable is NOT set, define the CCACHE variable as
blank on the Chrome EC make command line.  This will overrride and
disable the CCACHE variable in the Chrome EC makefile.

Change-Id: Idb1da06941084cea104d77748820971edf151f7b
Signed-off-by: Martin Roth <martinroth@google.com>
Reviewed-on: https://review.coreboot.org/16035
Tested-by: build bot (Jenkins)
Reviewed-by: Patrick Georgi <pgeorgi@google.com>
2 years agolibpayload: arm64: Fix MMU range overlap check
Julius Werner [Fri, 5 Aug 2016 17:37:52 +0000 (10:37 -0700)]
libpayload: arm64: Fix MMU range overlap check

The ARM64 MMU code maintains a list of used ranges, to avoid mapping the
DMA buffer over the coreboot tables and things like that. Unfortunately,
the overlap with ranges in that list is checked with

 (start1 >= start2 && start1 <= end2) || (end1 >= start2 && end1 <= end2)

which is not a full overlap check and misses the case where the second
region is completely contained within the first. This patch replaces
that code with a properly vetted primitive from Stack Overflow.

BRANCH=none
BUG=chrome-os-partner:54416
TEST=Observe how Kevin recovery screen now gets drawn at 10x the speed.

Change-Id: I7e2706426762794e160d743bbfc40da1e26eee12
Signed-off-by: Julius Werner <jwerner@chromium.org>
Reviewed-on: https://review.coreboot.org/16075
Tested-by: build bot (Jenkins)
Reviewed-by: Aaron Durbin <adurbin@chromium.org>
2 years agovendorcode/amd/pi/Lib: Fix reporting fatal error
Patrick Georgi [Wed, 3 Aug 2016 21:10:36 +0000 (23:10 +0200)]
vendorcode/amd/pi/Lib: Fix reporting fatal error

Instead of writing the first word of 6 "post code structs" where only
one exists (leading to 0xDEAD and 5 garbage words), write the correct
set.

Change-Id: Ifdfa53a970dda33dc9dc8c05788875077c001ecf
Signed-off-by: Patrick Georgi <pgeorgi@chromium.org>
Found-by: Coverity Scan #1361054, #1361055, #1361056
Reviewed-on: https://review.coreboot.org/16058
Tested-by: build bot (Jenkins)
Reviewed-by: Kyösti Mälkki <kyosti.malkki@gmail.com>
2 years agocbfstool/cbfs_image: Fix resource leak for tbuff
Furquan Shaikh [Fri, 5 Aug 2016 15:32:23 +0000 (08:32 -0700)]
cbfstool/cbfs_image: Fix resource leak for tbuff

Change-Id: I1f4626e1bda92af38e7967d7e05a4c7143942cf6
Signed-off-by: Furquan Shaikh <furquan@google.com>
Reviewed-on: https://review.coreboot.org/16074
Tested-by: build bot (Jenkins)
Reviewed-by: Aaron Durbin <adurbin@chromium.org>
2 years agocbfstool/cbfs_image: Check for return value of buffer_create
Furquan Shaikh [Fri, 5 Aug 2016 15:27:18 +0000 (08:27 -0700)]
cbfstool/cbfs_image: Check for return value of buffer_create

Free any buffers if required.

Change-Id: Iccd435dba51275d875a5fdb5649cdcd0541fd84c
Signed-off-by: Furquan Shaikh <furquan@google.com>
Found-by: Coverity Scan # 1361254
Reviewed-on: https://review.coreboot.org/16073
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
Tested-by: build bot (Jenkins)
Reviewed-by: Aaron Durbin <adurbin@chromium.org>
2 years agocbfstool/cbfs_image: Fix resource leak for segs
Furquan Shaikh [Fri, 5 Aug 2016 15:20:37 +0000 (08:20 -0700)]
cbfstool/cbfs_image: Fix resource leak for segs

Free segs whenever returning from cbfs_payload_make_elf()

Change-Id: I0dd722dd488723cecffe1f5621244bb0344056a6
Signed-off-by: Furquan Shaikh <furquan@google.com>
Found-by: Coverity Scan # 1361268
Reviewed-on: https://review.coreboot.org/16072
Tested-by: build bot (Jenkins)
Reviewed-by: Antonello Dettori <dev@dettori.io>
Reviewed-by: Aaron Durbin <adurbin@chromium.org>
2 years agocbfstool/elfheaders: Make elf_writer_destroy NULL-safe
Furquan Shaikh [Fri, 5 Aug 2016 19:04:55 +0000 (12:04 -0700)]
cbfstool/elfheaders: Make elf_writer_destroy NULL-safe

This relieves caller from having to check if the parameter being passed
in is NULL.

Change-Id: I3ea935c12d46c6fb5534e0f2077232b9e25240f1
Signed-off-by: Furquan Shaikh <furquan@google.com>
Reviewed-on: https://review.coreboot.org/16076
Reviewed-by: Aaron Durbin <adurbin@chromium.org>
Tested-by: build bot (Jenkins)
2 years agocbfstool/cbfs_image: Initialze empty_sz to 0
Furquan Shaikh [Fri, 5 Aug 2016 15:12:31 +0000 (08:12 -0700)]
cbfstool/cbfs_image: Initialze empty_sz to 0

Change-Id: I8b9cfe56b5893ba11047fcc1a6727e7e12a15772
Signed-off-by: Furquan Shaikh <furquan@google.com>
Found-by: Coverity Scan # 1361276
Reviewed-on: https://review.coreboot.org/16071
Reviewed-by: Aaron Durbin <adurbin@chromium.org>
Tested-by: build bot (Jenkins)
2 years agoUse VBOOT_SOURCE instead of hardcoding vboot path
Paul Kocialkowski [Sun, 24 Jul 2016 10:14:38 +0000 (12:14 +0200)]
Use VBOOT_SOURCE instead of hardcoding vboot path

This replaces all occurrences of a hardcoded vboot path to the
VBOOT_SOURCE variable, that may be overridden from the command line,
witch fallback to the source from 3rdparty.

Change-Id: Ia57d498d38719cc71e17060b76b0162c4ab363ed
Signed-off-by: Paul Kocialkowski <contact@paulk.fr>
Reviewed-on: https://review.coreboot.org/15825
Tested-by: build bot (Jenkins)
Reviewed-by: Patrick Georgi <pgeorgi@google.com>
Reviewed-by: Julius Werner <jwerner@chromium.org>
2 years agosoc/intel/skylake: Add Kabylake device Ids
Rizwan Qureshi [Wed, 3 Aug 2016 13:46:18 +0000 (19:16 +0530)]
soc/intel/skylake: Add Kabylake device Ids

Adding kabylake device ids for chip inits.
Skylake and Kabylak do not differ much, the intention
is to support both SoCs in the same code base.

Change-Id: I9ff4c6ca08fe681798001ce81cca2c085ce32325
Signed-off-by: Rizwan Qureshi <rizwan.qureshi@intel.com>
Reviewed-on: https://review.coreboot.org/16049
Tested-by: build bot (Jenkins)
Reviewed-by: Duncan Laurie <dlaurie@chromium.org>
2 years agogoogle/reef: Enable I2C2 for use in bootblock
Duncan Laurie [Thu, 4 Aug 2016 00:15:02 +0000 (17:15 -0700)]
google/reef: Enable I2C2 for use in bootblock

Enable I2C bus 2 for early init so it can be used by vboot for TPM
communication for verifying the memory init code.

BUG=chrome-os-partner:53336
BRANCH=none
TEST=build and boot on reef

Change-Id: Id4940ab01d8ccf288ab0a7a9a2f19867ed464e8d
Signed-off-by: Duncan Laurie <dlaurie@chromium.org>
Reviewed-on: https://review.coreboot.org/16059
Tested-by: build bot (Jenkins)
Reviewed-by: Aaron Durbin <adurbin@chromium.org>
2 years agoacpi: Generate object for coreboot table region
Duncan Laurie [Thu, 26 May 2016 19:47:05 +0000 (12:47 -0700)]
acpi: Generate object for coreboot table region

Generate an object to describe the coreboot table region in ACPI
with the HID "CORE0000" so it can be used by kernel drivers.

To keep track of the "CORE" HID usage add them to an enum and add
a function to generate the HID in AML:  Name (_HID, "CORExxxx")

BUG=chromium:589817
BRANCH=none
TEST=build and boot on chell, dump SSDT to verify contents:

Device (CTBL)
{
    Name (_HID, "CORE0000")  // _HID: Hardware ID
    Name (_UID, Zero)  // _UID: Unique ID
    Method (_STA, 0, NotSerialized)  // _STA: Status
    {
        Return (0x0F)
    }
    Name (_CRS, ResourceTemplate ()  // _CRS: Current Resource Settings
    {
        Memory32Fixed (ReadOnly,
            0x7AB84000,         // Address Base
            0x00008000,         // Address Length
            )
    })
}

Change-Id: I2c681c1fee02d52b8df2e72f6f6f0b76fa9592fb
Signed-off-by: Duncan Laurie <dlaurie@chromium.org>
Reviewed-on: https://review.coreboot.org/16056
Tested-by: build bot (Jenkins)
Reviewed-by: Aaron Durbin <adurbin@chromium.org>
Reviewed-by: Patrick Georgi <pgeorgi@google.com>
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
2 years agodrivers/intel/fsp1_1: only set a base address for FSP in COREBOOT CBFS
Aaron Durbin [Fri, 6 May 2016 21:06:36 +0000 (16:06 -0500)]
drivers/intel/fsp1_1: only set a base address for FSP in COREBOOT CBFS

The -b FSP_LOC argument to cbfstool is only valid for the COREBOOT
CBFS. Don't pass that value for all other CBFS regions.

Change-Id: Ib5321e7a7dbee8d26eb558933c8ce3fea50b11fe
Signed-off-by: Aaron Durbin <adurbin@chromium.org>
Reviewed-on: https://review.coreboot.org/14641
Tested-by: build bot (Jenkins)
Reviewed-by: Patrick Georgi <pgeorgi@google.com>
2 years agodrivers/intel/fsp2_0: Ensure EC is in right mode before memory init
Furquan Shaikh [Tue, 2 Aug 2016 21:04:59 +0000 (14:04 -0700)]
drivers/intel/fsp2_0: Ensure EC is in right mode before memory init

If EC_GOOGLE_CHROMEEC is enabled, ensure that the EC is in correct mode
before running memory init. This saves additional memory training
required in recovery path because of reboot later in ramstage.

BUG=chrome-os-partner:54245

Change-Id: Ic71c054afdcd0001cea95563fe513783b56f3e60
Signed-off-by: Furquan Shaikh <furquan@google.com>
Reviewed-on: https://review.coreboot.org/16034
Tested-by: build bot (Jenkins)
Reviewed-by: Aaron Durbin <adurbin@chromium.org>
2 years agogoogle/reef: Correct SD card pins config
Chiranjeevi Rapolu [Wed, 3 Aug 2016 21:09:33 +0000 (14:09 -0700)]
google/reef: Correct SD card pins config

SD CLK and CLK_FB needs to be pulled down by 20K.
SD CD_N is active LOW, needs to be pulled up by 20K
SD WP pin is not connected for uSD cards, enable writes
by default by pulling low by 20K.

BUG=chrome-os-partner:54866
BRANCH=None
TEST=Test with uSD cards.

Change-Id: Ia4bbd966ffb21e276dfc31a74f4ea54718900d66
Signed-off-by: Chiranjeevi Rapolu <chiranjeevi.rapolu@intel.com>
Reviewed-on: https://review.coreboot.org/16057
Tested-by: build bot (Jenkins)
Reviewed-by: Aaron Durbin <adurbin@chromium.org>