When using LPAE, the address space is split to 2MB blocks. This change makes
the space reserved for DMA consistent with the block size.
TEST=Booted nyan with and without LPAE. Built nyan_big.
BUG=None
BRANCH=None
Signed-off-by: Daisuke Nojiri <dnojiri@chromium.org>
Tested-by: Daisuke Nojiri <dnojiri@chromium.org>
Original-Change-Id: I75c77484f6ca9f23b583ef651956d0265a9b4474
Original-Reviewed-on: https://chromium-review.googlesource.com/188571
Original-Reviewed-by: Julius Werner <jwerner@chromium.org>
Original-Commit-Queue: Daisuke Nojiri <dnojiri@chromium.org>
Original-Tested-by: Daisuke Nojiri <dnojiri@chromium.org>
(cherry picked from commit 16a40a48c2e3fc131a348d5e7d377d26f4b20aaf)
Signed-off-by: Marc Jones <marc.jones@se-eng.com>
Change-Id: Ib79c9491dc504d28f811bbf0d91cffd292f5eb86
Reviewed-on: http://review.coreboot.org/7413
Tested-by: build bot (Jenkins)
Reviewed-by: David Hendricks <dhendrix@chromium.org>
Add a missing "~" so that we mask off just OSC_XOFS field and not the
rest of the register.
BUG=chrome-os-partner:26326
TEST=XHCI sometimes works after LP0.
BRANCH=none
Original-Change-Id: I2df2387dbad6920d36aa2ae5e6cd91e9ec42fa08
Original-Signed-off-by: Andrew Bresticker <abrestic@chromium.org>
Original-Reviewed-on: https://chromium-review.googlesource.com/188897
Original-Reviewed-by: Julius Werner <jwerner@chromium.org>
(cherry picked from commit bdbe9ead46fa883618a4acedd1feaf676e2eb29b)
Signed-off-by: Marc Jones <marc.jones@se-eng.com>
Change-Id: Ic853e737fc106527eb3bb15c25bf801a36bbff57
Reviewed-on: http://review.coreboot.org/7412
Tested-by: build bot (Jenkins)
Reviewed-by: David Hendricks <dhendrix@chromium.org>
Fix the PLLU parameters to match the recommended values from the TRM,
and the values used by the kernel and LP0 blob. This includes adding
support for setting an LFCON value. It appears that changing the PLLU
parameters across suspend/resume causes XHCI stability issues after
resume.
BUG=chrome-os-partner:26326
TEST=XHCI works after LP0 suspend/resume on Nyan.
BRANCH=none
Original-Change-Id: Ia4af12fefeebe607803e7f2f03ee4802367b82c3
Original-Signed-off-by: Andrew Bresticker <abrestic@chromium.org>
Original-Reviewed-on: https://chromium-review.googlesource.com/188752
Original-Reviewed-by: Julius Werner <jwerner@chromium.org>
Original-Reviewed-by: Tom Warren <twarren@nvidia.com>
(cherry picked from commit bbc8d92eb462e165c2378bcb3055a3a74b47a19b)
Signed-off-by: Marc Jones <marc.jones@se-eng.com>
Change-Id: I687d1709befc2f5dec094ee423f2ff824412996e
Reviewed-on: http://review.coreboot.org/7411
Tested-by: build bot (Jenkins)
Reviewed-by: David Hendricks <dhendrix@chromium.org>
This indirectly selects an appropriate PLLX frequency so the main CPUs run as
fast as they can but not faster.
BUG=chrome-os-partner:25467
TEST=Booted on nyan rev1.
BRANCH=None
Original-Change-Id: Ibe61f5e35246b272771debf4fdf90c79b21eb5d0
Original-Signed-off-by: Gabe Black <gabeblack@google.com>
Original-Reviewed-on: https://chromium-review.googlesource.com/188603
Original-Commit-Queue: Gabe Black <gabeblack@chromium.org>
Original-Tested-by: Gabe Black <gabeblack@chromium.org>
Original-Reviewed-by: Gabe Black <gabeblack@chromium.org>
(cherry picked from commit 947ecbce3cb6e4d7ab07d3ffd5b4694ca6270cde)
Signed-off-by: Marc Jones <marc.jones@se-eng.com>
Change-Id: I9163ddea7f246ae7207a8a715ebae2c9627a7e37
Reviewed-on: http://review.coreboot.org/7410
Tested-by: build bot (Jenkins)
Reviewed-by: David Hendricks <dhendrix@chromium.org>
The PLLX provides the clock for the main cores which can run at different max
frequencies depending on the specific model of Tegra124. This change makes it
possible to select a model which will, in turn, select a frequency for PLLX.
The default is 2GHz which is the lowest maximum frequency.
BUG=chrome-os-partner:25467
TEST=Booted on nyan rev1. Verified that the selected PLLX frequency was 2GHz.
With a change that selects the right model for nyan, verified that the
corresponding frequency was selected.
BRANCH=None
Original-Change-Id: Iee3a615083dee97ad659ff41cbf867af2a0c325d
Original-Signed-off-by: Gabe Black <gabeblack@google.com>
Original-Reviewed-on: https://chromium-review.googlesource.com/188602
Original-Reviewed-by: Gabe Black <gabeblack@chromium.org>
Original-Commit-Queue: Gabe Black <gabeblack@chromium.org>
Original-Tested-by: Gabe Black <gabeblack@chromium.org>
(cherry picked from commit 1282015048420a518e6c6959ce982be70378211a)
Signed-off-by: Marc Jones <marc.jones@se-eng.com>
Change-Id: I448a830f3184ad1afeadbd1c2974c7a27b03a923
Reviewed-on: http://review.coreboot.org/7409
Tested-by: build bot (Jenkins)
Reviewed-by: David Hendricks <dhendrix@chromium.org>
BUG=none
BRANCH=nyan
TEST=built and booted coreboot on my Nyan-rev1, browsed, ran Youtube vids,
WebGL experiments, etc. Everything seemed OK.
Original-Change-Id: I877680c9329ed96a0b602f0690acaa12079786d7
Original-Signed-off-by: Tom Warren <twarren@nvidia.com>
Original-Reviewed-on: https://chromium-review.googlesource.com/188550
Original-Reviewed-by: Julius Werner <jwerner@chromium.org>
(cherry picked from commit b6ca59e9db26f7422fa43ade889c921257a36851)
Signed-off-by: Marc Jones <marc.jones@se-eng.com>
Change-Id: If166938f241e2a4a8670bfce2df6591b4b71ff67
Reviewed-on: http://review.coreboot.org/7408
Tested-by: build bot (Jenkins)
Reviewed-by: David Hendricks <dhendrix@chromium.org>
The SRAM is very likely faster than going all the way out to DRAM for data,
but I don't think it's part of the cores themselves and won't be as fast as
the L1 caches. Enabling caching for this region reduces the time it takes to
get to the payload by about 75% when serial output is disabled and the main
part of display init is commented out.
BUG=chrome-os-partner:25467
TEST=Built and booted on nyan.
BRANCH=None
Original-Change-Id: I7ff26dea9d50e7d9a76e598e5654488481286b35
Original-Signed-off-by: Gabe Black <gabeblack@google.com>
Original-Reviewed-on: https://chromium-review.googlesource.com/188459
Original-Reviewed-by: Tom Warren <twarren@nvidia.com>
Original-Reviewed-by: Hung-Te Lin <hungte@chromium.org>
Original-Reviewed-by: Julius Werner <jwerner@chromium.org>
Original-Tested-by: Gabe Black <gabeblack@chromium.org>
Original-Commit-Queue: Gabe Black <gabeblack@chromium.org>
(cherry picked from commit ac8b9b30490d511ca1b207af6845d50e08ac130f)
Signed-off-by: Marc Jones <marc.jones@se-eng.com>
Change-Id: If79dcd1b116f30b778788ba4fd45d362ff5d8e6e
Reviewed-on: http://review.coreboot.org/7407
Tested-by: build bot (Jenkins)
Reviewed-by: David Hendricks <dhendrix@chromium.org>
When setting up caching on nyan and big, we would set the region after DRAM to
the end of the address space as uncachable. DRAM may actually extend beyond
the end of the address space, so that may result in address aliasing or other
problems. This change adds a check to make sure there's actually space there.
BUG=None
TEST=Built for big.
BRANCH=None
Original-Change-Id: Ic0a98550222f9dfc0aeafd67a2dd1c0c8f4ece44
Original-Signed-off-by: Gabe Black <gabeblack@google.com>
Original-Reviewed-on: https://chromium-review.googlesource.com/186769
Original-Reviewed-by: David Hendricks <dhendrix@chromium.org>
Original-Commit-Queue: Gabe Black <gabeblack@chromium.org>
Original-Tested-by: Gabe Black <gabeblack@chromium.org>
(cherry picked from commit 1866a4d2a001beb97779b611b8b69c63175048f4)
Signed-off-by: Marc Jones <marc.jones@se-eng.com>
Change-Id: If1ca8b5bd4efab8962e03c0d9eaa70c0327ea6b5
Reviewed-on: http://review.coreboot.org/7405
Tested-by: build bot (Jenkins)
Reviewed-by: David Hendricks <dhendrix@chromium.org>
This creates a new PL011 config variable which avoids the
infinite busy wait on serial_putchar() because the register
mapping is not compatible with current implementation.
BUG=None
BRANCH=none
TEST=printf() works on the PL011 based ARMv8 foundation model
Original-Change-Id: I9feda35a50a3488fc504d1561444161e0889deda
Original-Signed-off-by: Marcelo Povoa <marcelogp@chromium.org>
Original-Reviewed-on: https://chromium-review.googlesource.com/187020
Original-Reviewed-by: David Hendricks <dhendrix@chromium.org>
(cherry picked from commit 85779a34a161c324cc8af995ada4393137275f20)
Signed-off-by: Marc Jones <marc.jones@se-eng.com>
Conflicts:
payloads/libpayload/Config.in
payloads/libpayload/drivers/serial.c
Change-Id: I23c8b3728cd7d2d7692b3e86a679e061e88f7bb5
Reviewed-on: http://review.coreboot.org/7422
Tested-by: build bot (Jenkins)
Reviewed-by: David Hendricks <dhendrix@chromium.org>
These drivers are needed right away and never really fit into depthcharge's
driver model anyway.
CQ-DEPEND=CL:194064
BUG=None
TEST=Built and booted nyan, link, and peach_pit and verified that timer values
in cbmem were reasonable. Built for nyan_big, nyan_blaze and daisy.
BRANCH=None
Original-Change-Id: Ia7953cfece57524262a6c7d6537082af7a00f4d6
Original-Signed-off-by: Gabe Black <gabeblack@google.com>
Original-Reviewed-on: https://chromium-review.googlesource.com/194058
Original-Reviewed-by: Gabe Black <gabeblack@chromium.org>
Original-Commit-Queue: Gabe Black <gabeblack@chromium.org>
Original-Tested-by: Gabe Black <gabeblack@chromium.org>
(cherry picked from commit f30a410f0a248c93bc34f5868af1596bf8ce3cdd)
Signed-off-by: Marc Jones <marc.jones@se-eng.com>
Change-Id: I782d20f3cd63210a87c712643c7a53753f5ef301
Reviewed-on: http://review.coreboot.org/7225
Reviewed-by: Patrick Georgi <pgeorgi@google.com>
Tested-by: build bot (Jenkins)
We've recently fixed a problem where an external hard drive would choke
due to one too many CLEAR_FEATURE(HALT) commands in the XHCI stack with
"libpayload: usb: xhci: Fix STALL endpoint handling". Clearing stall
conditions from within the transfer function is wrong in general... this
is really something that is host controller agnostic and should be left
to the higher-level driver to decide. The mass storage driver (the only
one that should really encounter stalls right now) already contains the
proper amount of clear_stall() calls... any more than that is redundant
and as we found out potentially dangerous.
This patch removes automatic clear stalls from UHCI and OHCI drivers as
well to make things consistent between host controllers.
BUG=chromium:192866
TEST=None. I could borrow the original hard drive from Shawn and compile
a Snow to only use the OHCI driver to reproduce/verify this, but alas, I
am lazy (and it's really not that important).
Original-Change-Id: Ie1e4d4d2d70fa4abf8b4dabd33b10d6d4012048a
Original-Signed-off-by: Julius Werner <jwerner@chromium.org>
Original-Reviewed-on: https://chromium-review.googlesource.com/193732
Original-Reviewed-by: Stefan Reinauer <reinauer@chromium.org>
(cherry picked from commit d46e183f3e7e0b0130becdefa6fd3ef8097df54b)
Signed-off-by: Marc Jones <marc.jones@se-eng.com>
Change-Id: Ie8f4ab3db8ec0d9a2d1e91c62967833e59c46700
Reviewed-on: http://review.coreboot.org/7223
Tested-by: build bot (Jenkins)
Reviewed-by: David Hendricks <dhendrix@chromium.org>
This patch combines a few minor fixes and refactoring to the various
host controller and root hub drivers to ensure they all do the right
thing on a call to usb_exit(). It puts a usb_detach_device(0) call
into detach_controller() so that the HCD doesn't need to remember to
tear down the root hub itself, and makes sure all root hubs properly
detach the subtree of devices connected to their ports first (as
generic_hub and by extension XHCI had already been doing).
It also fixes up some missing free() calls and replaces most 'ptr =
malloc(); if (!ptr) fatal()' idioms with the new x(z)alloc().
BUG=chromium:343415
TEST=Tested EHCI on Big and OHCI, EHCI, and XHCI on Snow. Could not test
UHCI (unless anyone volunteers to port coreboot to a ZGB? ;) ), but the
changes are really tame.
Original-Change-Id: I6eca51ff2685d0946fe4267ad7d3ec48ad7fc510
Original-Signed-off-by: Julius Werner <jwerner@chromium.org>
Original-Reviewed-on: https://chromium-review.googlesource.com/193731
Original-Reviewed-by: Stefan Reinauer <reinauer@chromium.org>
(cherry picked from commit 5791b546e5a21a360d0c65888a5b92d5f48f8178)
Signed-off-by: Marc Jones <marc.jones@se-eng.com>
Change-Id: I00138f0aeceb12ed721f7368c7788c9b6bee227d
Reviewed-on: http://review.coreboot.org/7222
Tested-by: build bot (Jenkins)
Reviewed-by: David Hendricks <dhendrix@chromium.org>
config.panther was added in a chromium upstream patch. We don't
want mainboard specific configs in libpayload, so remove it.
Change-Id: Ibfb894a0262911c13e88bc161749b78e2b5c5185
Signed-off-by: Marc Jones <marc.jones@se-eng.com>
Reviewed-on: http://review.coreboot.org/7450
Tested-by: build bot (Jenkins)
Reviewed-by: David Hendricks <dhendrix@chromium.org>
Update x86 and ARM defconfig. Adds default N to specific timer
and serial drivers.
Change-Id: Ida6b953565dc6053729c2a72c6342d86596c599b
Signed-off-by: Marc Jones <marc.jones@se-eng.com>
Reviewed-on: http://review.coreboot.org/7449
Tested-by: build bot (Jenkins)
Reviewed-by: David Hendricks <dhendrix@chromium.org>
In such setup there is no resource 5. find_resource die()s if no resource is
present. Use probe_resource instead.
Change-Id: I6eb4a9d8712295c58281ee69ab129276d784ca2e
Signed-off-by: Vladimir Serbinenko <phcoder@gmail.com>
Reviewed-on: http://review.coreboot.org/7438
Reviewed-by: Edward O'Callaghan <eocallaghan@alterapraxis.com>
Tested-by: build bot (Jenkins)
This patch brings in ipq806x source files from the vendor's u-boot
tree as it was published in the 'cs_banana' release.
The following files are being copied:
arch/arm/cpu/armv7/ipq/clock.c => src/soc/qualcomm/ipq806x/clock.c
arch/arm/cpu/armv7/ipq/gpio.c => src/soc/qualcomm/ipq806x/gpio.c
arch/arm/cpu/armv7/ipq/timer.c => src/soc/qualcomm/ipq806x/timer.c
arch/arm/include/asm/arch-ipq806x/clock.h => src/soc/qualcomm/ipq806x/clock.h
arch/arm/include/asm/arch-ipq806x/gpio.h => src/soc/qualcomm/ipq806x/gpio.h
arch/arm/include/asm/arch-ipq806x/gsbi.h => src/soc/qualcomm/ipq806x/gsbi.h
arch/arm/include/asm/arch-ipq806x/iomap.h => src/soc/qualcomm/ipq806x/iomap.h
arch/arm/include/asm/arch-ipq806x/timer.h src/soc/qualcomm/ipq806x/timer.h
arch/arm/include/asm/arch-ipq806x/uart.h => src/soc/qualcomm/ipq806x/uart.h
board/qcom/ipq806x_cdp/ipq806x_cdp.c => src/mainboard/google/storm/cdp.c
board/qcom/ipq806x_cdp/ipq806x_cdp.h => src/soc/qualcomm/ipq8064/cdp.h
drivers/serial/ipq806x_uart.c => src/console/ipq806x_console.c
Note that local timer.c gets overwritten with the original version. To
prevent a build breakage some shortly to be reverted modifications had
to be made to src/soc/qualcomm/ipq806x/Makefile.inc and
src/soc/qualcomm/ipq806x/cbfs.c.
BRANCH=none
BUG=chrome-os-partner:27784
TEST='emerge-storm coreboot' still succeeds
Original-Change-Id: I3f50bfbec2e18a3b5d2c640cff353a26f88c98c1
Original-Signed-off-by: Vadim Bendebury <vbendeb@chromium.org>
Original-Reviewed-on: https://chromium-review.googlesource.com/193722
Original-Reviewed-by: David Hendricks <dhendrix@chromium.org>
(cherry picked from commit 3c9c2ede7e97e330cad2c2f3e557cc9bcdaecdcc)
Signed-off-by: Marc Jones <marc.jones@se-eng.com>
Change-Id: Ia7bc66cecfc16f1dd4a9f3cb9840cbe91878adf4
Reviewed-on: http://review.coreboot.org/7263
Tested-by: build bot (Jenkins)
Reviewed-by: Stefan Reinauer <stefan.reinauer@coreboot.org>
We want the coreboot build produce an image which can be run on the
target, even if the remaining parts of the bootprom (recovery path,
read-write stages, gbb, etc.) are not available yet.
This is achieved by including the Qualcomm SBLs blob in the bootblock.
CQ-DEPEND=CL:193518
BRANCH=None
BUG=chrome-os-partner:27784
TEST=manual
. run the following commands inside chroot to confirm expected image
layout (no actual code is executed on the target yet):
$ emerge-storm coreboot
$ \od -Ax -t x1 -v /build/storm/firmware/coreboot.rom 2>/dev/null | head -1
000000 d1 dc 4b 84 34 10 d7 73 15 00 00 00 ff ff ff ff
$ \od -Ax -t x1 -v /build/storm/firmware/coreboot.rom | grep 220000
220000 05 00 00 00 03 00 00 00 00 00 00 00 00 00 01 2a
Original-Change-Id: I10e8b81c7bd90e4550a027573ad3a26c38c3808a
Original-Signed-off-by: Vadim Bendebury <vbendeb@chromium.org>
Original-Reviewed-on: https://chromium-review.googlesource.com/193540
(cherry picked from commit 64e193974ee448f78e0a5775a440094901590afb)
Signed-off-by: Marc Jones <marc.jones@se-eng.com>
Change-Id: Idbdbeb9d229eff94a7a94af5dc4844a295458200
Reviewed-on: http://review.coreboot.org/7262
Tested-by: build bot (Jenkins)
Reviewed-by: Stefan Reinauer <stefan.reinauer@coreboot.org>
Once SECURITY_MODE fuse is burned, JTAG is disabled by default.
To reenable JTAG, besides chip unique id and SecureJtagControl need
to be built into BCT, Jtag enable flag is also needed to be set.
BUG=None
TEST=Burn SECURITY_MODE fuse, build chip specific BCT, coreboot
comes up and jtag hooks up fine.
Original-Change-Id: Ic6b61be2c09b15541400f9766d486a4fcef192a8
Original-Signed-off-by: Jimmy Zhang <jimmzhang@nvidia.com>
Original-Reviewed-on: https://chromium-review.googlesource.com/186031
Original-Reviewed-by: Julius Werner <jwerner@chromium.org>
(cherry picked from commit ff962b81f424c840ef171d4287a65ab79b018a28)
Signed-off-by: Marc Jones <marc.jones@se-eng.com>
Change-Id: I14b496932dbc0ed184a2212a5b33d740e1f34a4e
Reviewed-on: http://review.coreboot.org/7403
Tested-by: build bot (Jenkins)
Reviewed-by: Edward O'Callaghan <eocallaghan@alterapraxis.com>
Repurpose config->pwm to mean the particular PWM device (we use PWM1 on
nyan), and add code to program the PWM device.
BUG=none
TEST=emerge-nyan chromeos-coreboot-nyan, regenerate bootimage, and boot.
See that the backlight comes up in the bootloader, and brightness can be
adjusted via pwm_bl driver in the kernel.
Original-Change-Id: I2db047e5ef23c0e8fb66dd05ad6339d60918d493
Original-Signed-off-by: Andrew Chew <achew@nvidia.com>
Original-Reviewed-on: https://chromium-review.googlesource.com/185772
Original-Reviewed-by: Andrew Bresticker <abrestic@chromium.org>
(cherry picked from commit 0dee98dd0c8510ecd630b5c6cb9ea49724dc8b55)
Signed-off-by: Marc Jones <marc.jones@se-eng.com>
Change-Id: Ie53610f3afa30b2d8f484685fb0e8c0b12cd8241
Reviewed-on: http://review.coreboot.org/7402
Tested-by: build bot (Jenkins)
Reviewed-by: Edward O'Callaghan <eocallaghan@alterapraxis.com>
The generic tegra124 code will use one of the PWMs to drive the backlight of
the display, but the PWM clock was enabled only for nyan. This change enables
it for big as well.
BUG=none
TEST=Built for Big
BRANCH=None
Original-Change-Id: I5171da7c41f4b4db931563ada3e8e4ebf74ec3d9
Original-Signed-off-by: Gabe Black <gabeblack@google.com>
Original-Reviewed-on: https://chromium-review.googlesource.com/186767
Original-Reviewed-by: Julius Werner <jwerner@chromium.org>
Original-Commit-Queue: Gabe Black <gabeblack@chromium.org>
Original-Tested-by: Gabe Black <gabeblack@chromium.org>
(cherry picked from commit 687f3771fb3e6b340a818fa7594b3ac0630fdeaf)
Signed-off-by: Marc Jones <marc.jones@se-eng.com>
Change-Id: Ifd14a22a98e7fe273ec28c460b928b8a83c84b66
Reviewed-on: http://review.coreboot.org/7404
Tested-by: build bot (Jenkins)
Reviewed-by: Edward O'Callaghan <eocallaghan@alterapraxis.com>
Add some defines and structs that describe what the PWM registers look like.
BUG=none
TEST=emerge-nyan chromeos-coreboot-nyan
Original-Change-Id: Ie10589e4cbf5292e543d205ac8a1c6b09a0f76d0
Original-Signed-off-by: Andrew Chew <achew@nvidia.com>
Original-Reviewed-on: https://chromium-review.googlesource.com/185771
Original-Reviewed-by: Andrew Bresticker <abrestic@chromium.org>
(cherry picked from commit fbbd2a5e148c1142aee100dbcde17c865b06b2bd)
Signed-off-by: Marc Jones <marc.jones@se-eng.com>
Change-Id: If4dc40c1dcdf1723e05923e2fea42ccc47766699
Reviewed-on: http://review.coreboot.org/7401
Tested-by: build bot (Jenkins)
Reviewed-by: Edward O'Callaghan <eocallaghan@alterapraxis.com>
It seems that someone just stuck the PM3 function for all of the potential
PWM pins. Fix this to be more specific to the particular PWM (of which
there are four).
BUG=none
TEST=emerge-nyan chromeos-coreboot-nyan
Original-Change-Id: Ic61a7321fbe28953b22007a1d0b522c3ca8714ad
Original-Signed-off-by: Andrew Chew <achew@nvidia.com>
Original-Reviewed-on: https://chromium-review.googlesource.com/185739
Original-Reviewed-by: Andrew Bresticker <abrestic@chromium.org>
(cherry picked from commit f19f897fe11a582cc240d98de88c5e2d4dc4e364)
Signed-off-by: Marc Jones <marc.jones@se-eng.com>
Change-Id: Ie10173413a5f00e06f5b1803fd93d6cb322cee3d
Reviewed-on: http://review.coreboot.org/7399
Tested-by: build bot (Jenkins)
Reviewed-by: Edward O'Callaghan <eocallaghan@alterapraxis.com>
The Tegra PWM base address was missing, so add it.
BUG=none
TEST=emerge-nyan chromeos-coreboot-nyan
Original-Change-Id: Iebf687c6644290e05ee72794cde697658ab6d7cb
Original-Signed-off-by: Andrew Chew <achew@nvidia.com>
Original-Reviewed-on: https://chromium-review.googlesource.com/185738
Original-Reviewed-by: Andrew Bresticker <abrestic@chromium.org>
(cherry picked from commit b62843f6cfbf870451f658e6df1a3b48256fa4e1)
Signed-off-by: Marc Jones <marc.jones@se-eng.com>
Change-Id: Ibb8578a130d5995345592caa610c57c1d7f28573
Reviewed-on: http://review.coreboot.org/7398
Tested-by: build bot (Jenkins)
Reviewed-by: Edward O'Callaghan <eocallaghan@alterapraxis.com>
rmodules were getting linked with libgcc and not libcompiler-rt.
Unfortunately this is pretty ugly however we do this else where
in the build system so its consistently ugly. The build system
will later need a unification pass between compilers once we are
tree stable on Clang.
Change-Id: I380f7386de2c5adfa9036311323ad9f703b6e712
Signed-off-by: Edward O'Callaghan <eocallaghan@alterapraxis.com>
Reviewed-on: http://review.coreboot.org/7440
Tested-by: build bot (Jenkins)
Reviewed-by: Patrick Georgi <pgeorgi@google.com>
We have no need for these sections winding up in the build
leading to possible overlaps, such as in the case of Clang
builds. Discard sections from inclusion into the resulting
binary.
Change-Id: Ie807e5809594dcc6e94660a64e359e3b2ca1a0f6
Signed-off-by: Edward O'Callaghan <eocallaghan@alterapraxis.com>
Reviewed-on: http://review.coreboot.org/7439
Tested-by: build bot (Jenkins)
Reviewed-by: Patrick Georgi <pgeorgi@google.com>
We'd been putting some data structures like the framebuffer and the cbmem at
the end of memory, but that may not actually be addressable as identity mapped
memory. This change clamps the addresses those structures are placed at so
they stay below 4GB.
BUG=None
TEST=Booted on nyan. Went into recovery mode and verified that there was a
recovery screen. Forced memory size to be 4GB and verified that the recovery
screen still shows up.
BRANCH=None
Original-Change-Id: I9e6b28212c113107d4f480b3dd846dd2349b3a91
Original-Signed-off-by: Gabe Black <gabeblack@google.com>
Original-Reviewed-on: https://chromium-review.googlesource.com/185571
Original-Reviewed-by: David Hendricks <dhendrix@chromium.org>
Original-Commit-Queue: Gabe Black <gabeblack@chromium.org>
Original-Tested-by: Gabe Black <gabeblack@chromium.org>
(cherry picked from commit 63ea1274a838dc739d302d7551f1db42034c5bd0)
Signed-off-by: Marc Jones <marc.jones@se-eng.com>
Change-Id: I970c1285270cb648bc67fa114d44c0841eab1615
Reviewed-on: http://review.coreboot.org/7397
Reviewed-by: Edward O'Callaghan <eocallaghan@alterapraxis.com>
Tested-by: build bot (Jenkins)
If an asm blob isn't marked as volatile, gcc is free to throw it out if it
doesn't think it produces any values that are actually used. To prevent that
from happening, add volatile to some asm blobs in the nyan romstage code.
BUG=None
TEST=Booted on nyan rev1.
BRANCH=None
Original-Change-Id: I819e068e738e94ea749fcb72bba2eee080e1dfb1
Original-Signed-off-by: Gabe Black <gabeblack@google.com>
Original-Reviewed-on: https://chromium-review.googlesource.com/185610
Original-Reviewed-by: Julius Werner <jwerner@chromium.org>
Original-Commit-Queue: Gabe Black <gabeblack@chromium.org>
Original-Tested-by: Gabe Black <gabeblack@chromium.org>
(cherry picked from commit 76c09581d6ca4dc6c2f9048f599822939f439d11)
Signed-off-by: Marc Jones <marc.jones@se-eng.com>
Change-Id: I0b32197abf0ddc5f454f9c2415a65d98c60ca48b
Reviewed-on: http://review.coreboot.org/7396
Reviewed-by: Edward O'Callaghan <eocallaghan@alterapraxis.com>
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
Tested-by: build bot (Jenkins)
Update the 3rdparty repo to the IPQ binary commit
This got updated in error by commit:39bbc8cb97e2de2423cc31bee014ef56884d9f3c
Original-Change-Id: I50fd7254eaf97ac44fb046e39ff1a81d2baad16f
Original-Signed-off-by: Marc Jones <marc.jones@se-eng.com>
Original-Reviewed-on: http://review.coreboot.org/7354
Original-Reviewed-by: Stefan Reinauer <stefan.reinauer@coreboot.org>
(cherry picked from commit cfa06c7460)
Signed-off-by: Marc Jones <marc.jones@se-eng.com>
Change-Id: Ibfa243d057f9a2d27e9e02e3e8d4fc6e1da61df0
Reviewed-on: http://review.coreboot.org/7437
Reviewed-by: Patrick Georgi <pgeorgi@google.com>
Tested-by: Patrick Georgi <pgeorgi@google.com>
Tested-by: build bot (Jenkins)
Version 0.9 contains a fix for a security issue. A more
detailed changelog is not available.
Change-Id: I1a66c9da900f89ba9b4c13f3457582278d3793e2
Signed-off-by: Zheng Bao <zheng.bao@amd.com>
Signed-off-by: Zheng Bao <fishbaozi@gmail.com>
Reviewed-on: http://review.coreboot.org/7293
Reviewed-by: Bruce Griffith <Bruce.Griffith@se-eng.com>
Reviewed-by: Rudolf Marek <r.marek@assembler.cz>
Tested-by: build bot (Jenkins)
The hudson handling alluded to in the original comment was implemented in
commit ea90963666, use it to disable the
SD controller so it doesn't show up in lspci.
Change-Id: Ib2ba79a11af06c6765dcad4070232a8a7c6d2751
Signed-off-by: Tobias Diedrich <ranma+coreboot@tdiedrich.de>
Reviewed-on: http://review.coreboot.org/7383
Tested-by: build bot (Jenkins)
Reviewed-by: Rudolf Marek <r.marek@assembler.cz>
The "Bus 0, Dev 21 PCIE Bridge" entry doesn't match the DSDT from my
BIOS. It looks like this entry was erroneously copied from the entry
for "Bus 0, Dev 20" without rotating the IRQ numbers.
The other entries match my ASUS BIOS and the usual rotation pattern.
Change-Id: I7401c3daaf0da78ba631791947e5a6bb045fc075
Signed-off-by: Tobias Diedrich <ranma+coreboot@tdiedrich.de>
Reviewed-on: http://review.coreboot.org/7384
Tested-by: build bot (Jenkins)
Reviewed-by: Rudolf Marek <r.marek@assembler.cz>
'%02hx' is unsigned short, where as the argument is typed as
uint8_t and so '%02hhx' is actually correct here.
Found-by: Clang
Change-Id: I40c48dcecf12845f4708e511236184908e90fb56
Signed-off-by: Edward O'Callaghan <eocallaghan@alterapraxis.com>
Reviewed-on: http://review.coreboot.org/7428
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
Reviewed-by: Marc Jones <marc.jones@se-eng.com>
Tested-by: build bot (Jenkins)
Broken linkage rule for Clang builds on one side of a
branch. Hence refactor out common rules from branch.
Change-Id: I00e5a2f5f9af1b7882a453caebb378ef74d2d51e
Signed-off-by: Edward O'Callaghan <eocallaghan@alterapraxis.com>
Reviewed-on: http://review.coreboot.org/7425
Reviewed-by: Patrick Georgi <pgeorgi@google.com>
Tested-by: build bot (Jenkins)
Guards around #includes only hide deeper issues.
Change-Id: I0a356360eb3919910a980966213a2c53e99e77eb
Signed-off-by: Edward O'Callaghan <eocallaghan@alterapraxis.com>
Reviewed-on: http://review.coreboot.org/7424
Reviewed-by: Marc Jones <marc.jones@se-eng.com>
Tested-by: build bot (Jenkins)
This patch changes several cache-related pieces to be cleaner, faster or
more correct. The largest point is removing the old
arm_invalidate_caches() function and surrounding bootblock code to
initialize SCTLR and replace it with an all-assembly function that takes
care of cache and SCTLR initialization to bring the system to a known
state. It runs without stack and before coreboot makes any write
accesses to be as compatible as possible with whatever state the system
was left in by preceeding code. This also finally fixes the dreaded
icache bug that wasted hundreds of milliseconds during boot.
Old-Change-Id: I7bb4995af8184f6383f8e3b1b870b0662bde8bd4
Signed-off-by: Julius Werner <jwerner@chromium.org>
Reviewed-on: https://chromium-review.googlesource.com/183890
(cherry picked from commit 07a35925dc957919bf88dfc90515971a36e81b97)
nyan_big: apply cache-related changes from nyan
This applies the same changes from 07a3592 that were applied to nyan.
Old-Change-Id: Idcbe85436d7a2f65fcd751954012eb5f4bec0b6c
Reviewed-on: https://chromium-review.googlesource.com/184551
Commit-Queue: David Hendricks <dhendrix@chromium.org>
Tested-by: David Hendricks <dhendrix@chromium.org>
Reviewed-by: David Hendricks <dhendrix@chromium.org>
(cherry picked from commit 4af27f02614da41c611aee2c6d175b1b948428ea)
Squashed the followup patch for nyan_big into the original patch.
Change-Id: Id14aef7846355ea2da496e55da227b635aca409e
Signed-off-by: Isaac Christensen <isaac.christensen@se-eng.com>
(cherry picked from commit 4cbf25f8eca3a12bbfec5b015953c0fc2b69c877)
Signed-off-by: Marc Jones <marc.jones@se-eng.com>
Reviewed-on: http://review.coreboot.org/6993
Tested-by: build bot (Jenkins)
Reviewed-by: Stefan Reinauer <stefan.reinauer@coreboot.org>
Apply commit 283ba78415 to f14 (literally, plus one adaptation).
Change-Id: Ieea47470e5852ec8a46596ce23a2d18444618624
Signed-off-by: Edward O'Callaghan <eocallaghan@alterapraxis.com>
Reviewed-on: http://review.coreboot.org/7361
Tested-by: build bot (Jenkins)
Reviewed-by: Patrick Georgi <pgeorgi@google.com>
This is required to run abuild parallely with clang
without the canonical coreboot toolchain installed.
Change-Id: Iea56d3f552d50ab6e762afa134091b0d8e38792c
Signed-off-by: Patrick Georgi <patrick@georgi-clan.de>
Reviewed-on: http://review.coreboot.org/7369
Reviewed-by: Edward O'Callaghan <eocallaghan@alterapraxis.com>
Tested-by: build bot (Jenkins)
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
Add IDs of some SNB and Haswell chips; use more descriptive names.
Add PCIEXBAR and PXPEPBAR read support for SNB/IVB/Haswell.
Change-Id: I16753bf90061fc2065b813b1c2169e7b7bcc89e8
Signed-off-by: Felix Held <felix-coreboot@felixheld.de>
Reviewed-on: http://review.coreboot.org/7360
Reviewed-by: Patrick Georgi <pgeorgi@google.com>
Tested-by: build bot (Jenkins)
Reviewed-by: Mathias Krause <minipli@googlemail.com>
These drivers need to be ready right away and never really fit into the
depthcharge driver model anyway.
CQ-DEPEND=CL:194063
BUG=None
TEST=Built and booted on nyan and peach_pit. Built for nyan_big, nyan_blaze,
and daisy.
BRANCH=None
Original-Change-Id: I9570dee53c57d42ef4cd956f66a878ce39a2dc20
Original-Signed-off-by: Gabe Black <gabeblack@google.com>
Original-Reviewed-on: https://chromium-review.googlesource.com/194057
Original-Reviewed-by: Gabe Black <gabeblack@chromium.org>
Original-Commit-Queue: Gabe Black <gabeblack@chromium.org>
Original-Tested-by: Gabe Black <gabeblack@chromium.org>
(cherry picked from commit 26e18f680c93fc990a3d1057c164f19859634a9f)
Signed-off-by: Marc Jones <marc.jones@se-eng.com>
Change-Id: Ia2233e2bd821d8de8d2d57a9423aeb74be7efd93
Reviewed-on: http://review.coreboot.org/7224
Reviewed-by: Patrick Georgi <pgeorgi@google.com>
Tested-by: build bot (Jenkins)
Otherwise clang feels free to optimize away that variable
(somewhat) and revive it in a different form inside .bss.
They probably have the language lawyery excuse for why
that's perfectly legal, so let's play it safe.
(relevant URL, sorry ron: http://llvm.org/bugs/show_bug.cgi?id=9520)
Change-Id: I603312ceea7207088dd29453cc8fb8f48c31af21
Signed-off-by: Patrick Georgi <patrick@georgi-clan.de>
Reviewed-on: http://review.coreboot.org/7357
Reviewed-by: Edward O'Callaghan <eocallaghan@alterapraxis.com>
Tested-by: build bot (Jenkins)
Daisuke Nojiri