project/bcm63xx/atf.git
6 years agorockchip/rk3399: Fix memory corruptions or illegal memory access
Caesar Wang [Tue, 30 Jan 2018 01:11:24 +0000 (09:11 +0800)]
rockchip/rk3399: Fix memory corruptions or illegal memory access

Coverity scan done for the coreboot project found the issue:
Coverity (*** CID 1385418: Memory - illegal accesses (OVERRUN))
Coverity (*** CID 1385419: Memory - corruptions  (OVERRUN))

Fix the Converity error issue with store_cru[] loop needs to be one
element bigger.

Fixes: ARM-software/tf-issues#544
Change-Id: I420f0a660b24baaa5fc5e78fca242cf750c9bbc7
Signed-off-by: Caesar Wang <wxt@rock-chips.com>
6 years agoMerge pull request #1237 from sandrine-bailleux-arm/sb/spm-timer
davidcunado-arm [Mon, 29 Jan 2018 23:16:27 +0000 (23:16 +0000)]
Merge pull request #1237 from sandrine-bailleux-arm/sb/spm-timer

SPM: Map devices in the 1st GB

6 years agoMerge pull request #1246 from sandrine-bailleux-arm/topics/sb/fix-cnp-doc
davidcunado-arm [Mon, 29 Jan 2018 22:46:27 +0000 (22:46 +0000)]
Merge pull request #1246 from sandrine-bailleux-arm/topics/sb/fix-cnp-doc

Fix documentation for CnP bit

6 years agoMerge pull request #1243 from afaerber/hikey-docs
davidcunado-arm [Mon, 29 Jan 2018 22:45:35 +0000 (22:45 +0000)]
Merge pull request #1243 from afaerber/hikey-docs

docs: hikey: Fix typo

6 years agoFix documentation for CnP bit
Sandrine Bailleux [Mon, 29 Jan 2018 13:48:15 +0000 (14:48 +0100)]
Fix documentation for CnP bit

The CnP bit documentation in the Firmware Design Guide incorrectly
used the term "Page Entries" instead of "Processing Elements".
Fix that.

Change-Id: Ie44ee99c281b7b1a9ad90fba2c7d109f12425507
Signed-off-by: Sandrine Bailleux <sandrine.bailleux@arm.com>
6 years agoMerge pull request #1241 from vchong/fixmemprot
davidcunado-arm [Mon, 29 Jan 2018 09:54:52 +0000 (09:54 +0000)]
Merge pull request #1241 from vchong/fixmemprot

hikey: fix assert in sec_protect()

6 years agoSPM: Map devices in the 1st GB
Sandrine Bailleux [Fri, 12 Jan 2018 14:50:12 +0000 (15:50 +0100)]
SPM: Map devices in the 1st GB

This patch maps the devices in the first GB of the system address map
on the FVP into the S-EL1&0 translation regime when SPM support is
enabled. This grants the Secure Partition access to the devices in
this region, for example the memory-mapped Generic Timer device.

Change-Id: I3aeea65f859ecbe83efde2acee20c55500c451bc
Signed-off-by: Sandrine Bailleux <sandrine.bailleux@arm.com>
6 years agodocs: hikey: Fix typo
Andreas Färber [Mon, 29 Jan 2018 01:47:10 +0000 (02:47 +0100)]
docs: hikey: Fix typo

The correct name of the manufacturer is LeMaker.

Signed-off-by: Andreas Färber <afaerber@suse.de>
6 years agohikey: fix assert in sec_protect()
Victor Chong [Sat, 27 Jan 2018 12:36:12 +0000 (21:36 +0900)]
hikey: fix assert in sec_protect()

`assert(e)` was used in place of `if (e) ERROR()` when sec_protect()
was ported from hikey fork so the logic should have been reversed.

Fixes: 3d5d9f5a ("hikey: configure the top 16MB of DRAM as secure")
Fixes: 52988b38 ("hikey: configure 4 MB of secure DRAM for OP-TEE
Secure Data Path")
Signed-off-by: Victor Chong <victor.chong@linaro.org>
Tested-by: Victor Chong <victor.chong@linaro.org>
6 years agoMerge pull request #1205 from petegriffin/hikey-protect-memory
davidcunado-arm [Fri, 26 Jan 2018 17:29:36 +0000 (17:29 +0000)]
Merge pull request #1205 from petegriffin/hikey-protect-memory

Hikey protect optee / sdp memory

6 years agoMerge pull request #1232 from masahir0y/uniphier
davidcunado-arm [Thu, 25 Jan 2018 16:36:43 +0000 (16:36 +0000)]
Merge pull request #1232 from masahir0y/uniphier

uniphier: migrate to BL2-AT-EL3

6 years agoMerge pull request #1234 from SNG-ARM/master
davidcunado-arm [Thu, 25 Jan 2018 00:52:01 +0000 (00:52 +0000)]
Merge pull request #1234 from SNG-ARM/master

SPM: Declare explicit width based types in secure_partition_boot_info…

6 years agoMerge pull request #1231 from Leo-Yan/hikey960_enable_fiq_handling
davidcunado-arm [Thu, 25 Jan 2018 00:07:06 +0000 (00:07 +0000)]
Merge pull request #1231 from Leo-Yan/hikey960_enable_fiq_handling

Hikey960: Enable invalid FIQ handling

6 years agoMerge pull request #1228 from dp-arm/dp/cve_2017_5715
davidcunado-arm [Thu, 25 Jan 2018 00:06:50 +0000 (00:06 +0000)]
Merge pull request #1228 from dp-arm/dp/cve_2017_5715

Workarounds for CVE-2017-5715 on A9/A15 and A17 + serial console reporting

6 years agoMerge pull request #1223 from vchong/poplar_bl1loadsfip
davidcunado-arm [Wed, 24 Jan 2018 23:00:01 +0000 (23:00 +0000)]
Merge pull request #1223 from vchong/poplar_bl1loadsfip

poplar: Enable emmc and recovery build support

6 years agoMerge pull request #1193 from jwerner-chromium/JW_coreboot
davidcunado-arm [Wed, 24 Jan 2018 14:31:53 +0000 (14:31 +0000)]
Merge pull request #1193 from jwerner-chromium/JW_coreboot

New console API and coreboot support [v4]

6 years agouniphier: switch to BL2-AT-EL3 and remove BL1 support
Masahiro Yamada [Tue, 19 Dec 2017 02:56:05 +0000 (11:56 +0900)]
uniphier: switch to BL2-AT-EL3 and remove BL1 support

UniPhier platform implements non-TF boot ROM.  Prior to the BL2-AT-EL3
support, BL1 (worked as a pseudo ROM) was needed just for ensuring BL2
is entered at EL1-S.  Now, this platform is able to avoid this waste.

Enable the BL2_AT_EL3 option, and remove BL1.

Signed-off-by: Masahiro Yamada <yamada.masahiro@socionext.com>
6 years agouniphier: set PROGRAMMABLE_RESET_ADDRESS to disable warm boot mailbox
Masahiro Yamada [Mon, 22 Jan 2018 09:35:16 +0000 (18:35 +0900)]
uniphier: set PROGRAMMABLE_RESET_ADDRESS to disable warm boot mailbox

The warm boot mailbox code is compiled if PROGRAMMABLE_RESET_ADDRESS
is disabled.

The warm boot mailbox is useless for UniPhier SoC family because BL1
is not the first image.  The UniPhier platform implements non-TF ROM,
then BL1 works as a pseudo ROM, so it is never executed in the warm
boot.

The reset vector address is not actually programmable for UniPhier
platform, but it should not hurt to enable PROGRAMMABLE_RESET_ADDRESS
to disable the mailbox and remove pointless plat_get_my_entrypoint.

Signed-off-by: Masahiro Yamada <yamada.masahiro@socionext.com>
6 years agoMerge pull request #1233 from soby-mathew/sm/rm_uniphier_override
davidcunado-arm [Wed, 24 Jan 2018 12:31:06 +0000 (12:31 +0000)]
Merge pull request #1233 from soby-mathew/sm/rm_uniphier_override

Allow API deprecation for uniphier platform

6 years agohikey: configure 4 MB of secure DRAM for OP-TEE Secure Data Path
Peter Griffin [Thu, 21 Dec 2017 18:03:46 +0000 (18:03 +0000)]
hikey: configure 4 MB of secure DRAM for OP-TEE Secure Data Path

Update the memory firewall configuration to reserve 4 MB of secure RAM
for use by the kernel and OP-TEE as the Secure Data Path pool.
Note that this address range (0x3E800000 - 0x3EC00000) falls in the
range already set aside by UEFI (which reserves the upper 32 MB of the
1GB DRAM for OP-TEE [1]) and was previously unused.

[1] https://github.com/96boards-hikey/edk2/blob/hikey/HisiPkg/HiKeyPkg/Library/HiKeyLib/HiKeyMem.c#L44
Signed-off-by: Jerome Forissier <jerome.forissier@linaro.org>
Signed-off-by: Peter Griffin <peter.griffin@linaro.org>
Acked-by: Victor Chong <victor.chong@linaro.org>
6 years agohikey: configure the top 16MB of DRAM as secure
Jerome Forissier [Mon, 4 May 2015 07:40:03 +0000 (09:40 +0200)]
hikey: configure the top 16MB of DRAM as secure

DRAM region 0x3f000000 - 0x3fffffff is reserved for OP-TEE and should
therefore be accessible only from secure world.

Signed-off-by: Jerome Forissier <jerome.forissier@linaro.org>
Signed-off-by: Peter Griffin <peter.griffin@linaro.org>
Acked-by: Victor Chong <victor.chong@linaro.org>
6 years agoSPM: Declare explicit width based types in secure_partition_boot_info_t structure
Sughosh Ganu [Mon, 11 Dec 2017 13:33:19 +0000 (19:03 +0530)]
SPM: Declare explicit width based types in secure_partition_boot_info_t structure

The secure_partition_boot_info_t structure is used to communicate boot
parameters with the StandaloneMM code executing at S-EL0 through a
shared buffer. Certain data types used for members of this structure
are opaque with their size depending on the toolchain being used.

Declare the members of the structure with explicit width based data
types, which would maintain compatibility across toolchains.

Signed-off-by: Sughosh Ganu <sughosh.ganu@arm.com>
6 years agoAllow API deprecation for uniphier platform
Soby Mathew [Fri, 10 Nov 2017 12:08:46 +0000 (12:08 +0000)]
Allow API deprecation for uniphier platform

The `override ERROR_DEPRECATION = 1` setting in uniphier platform
makes deprecation of API difficult. Hence removing the same. This
flag should be specified on the command line if needed.

Change-Id: I8c82d8d13944e450a8cd636de3326137c04d7560
Signed-off-by: Soby Mathew <soby.mathew@arm.com>
6 years agopoplar: Add support for recovery build
Victor Chong [Mon, 15 Jan 2018 15:29:47 +0000 (00:29 +0900)]
poplar: Add support for recovery build

Signed-off-by: Victor Chong <victor.chong@linaro.org>
Acked-by: Shawn Guo <shawn.guo@linaro.org>
Tested-by: Shawn Guo <shawn.guo@linaro.org>
6 years agopoplar: Enable emmc support
Victor Chong [Tue, 2 Jan 2018 16:53:08 +0000 (01:53 +0900)]
poplar: Enable emmc support

Let bl1 and bl2 have the ability to load images from emmc instead of
dram (mmap).

Signed-off-by: Victor Chong <victor.chong@linaro.org>
Acked-by: Shawn Guo <shawn.guo@linaro.org>
Tested-by: Shawn Guo <shawn.guo@linaro.org>
6 years agoHikey960: Enable invalid FIQ handling
Leo Yan [Mon, 22 Jan 2018 04:40:25 +0000 (12:40 +0800)]
Hikey960: Enable invalid FIQ handling

When some interrupts are configured as group 0 in GICv2, these
interrupts trigger FIQ signal; this results in the Linux kernel panic
by reporting log: "Bad mode in FIQ handler detected on CPU0, code
0x00000000 -- Unknown/Uncategorized".  Unfortunately from kernel side it
has no permission to read the GIC register for group 0 interrupts so we
have no chance to get to know which interrupt is configured as secure
interrupt and cause the kernel panic.

For upper reason, this commit enables FIQ exception handling for
SPD_none case.  If the system has not enabled SPD the FIQ interrupt is
trapped into EL3 and the FIQ handler can report the interrupt number so
we can easily narrow down which FIQ introduce unexpected interrupt.
After enable SPD we can rely on SPD interrupt route model to handle FIQ.

Signed-off-by: Leo Yan <leo.yan@linaro.org>
6 years agoMerge pull request #1229 from manojkumar-arm/manojkumar-arm/ca72-aarch32-reset-fix
davidcunado-arm [Sat, 20 Jan 2018 17:04:49 +0000 (17:04 +0000)]
Merge pull request #1229 from manojkumar-arm/manojkumar-arm/ca72-aarch32-reset-fix

lib/cpus: fix branching in reset function for cortex-a72 AARCH32 mode

6 years agoMerge pull request #1227 from geesun/qx/emmc_macros
davidcunado-arm [Fri, 19 Jan 2018 23:56:58 +0000 (23:56 +0000)]
Merge pull request #1227 from geesun/qx/emmc_macros

emmc: add macros CMD21, BUS_WIDTH_DDR_4 and BUS_WIDTH_DDR_8

6 years agocoreboot: Add support for CBMEM console
Julius Werner [Tue, 13 Jun 2017 22:53:45 +0000 (15:53 -0700)]
coreboot: Add support for CBMEM console

coreboot supports an in-memory console to store firmware logs even when
no serial console is available. It is widely supported by
coreboot-compatible bootloaders (including SeaBIOS and GRUB) and can be
read by the Linux kernel.

This patch allows BL31 to add its own log messages to this console. The
driver will be registered automatically if coreboot support is compiled
in and detects the presence of a console buffer in the coreboot tables.

Change-Id: I31254dfa0c2fdeb7454634134b5707b4b4154907
Signed-off-by: Julius Werner <jwerner@chromium.org>
6 years agorockchip: Move to MULTI_CONSOLE_API
Julius Werner [Wed, 2 Aug 2017 23:31:04 +0000 (16:31 -0700)]
rockchip: Move to MULTI_CONSOLE_API

This patch changes all Rockchip platforms to use the new
MULTI_CONSOLE_API. The platform-specific plat_crash_console
implementations are removed so that the platform can use the ones from
the common platform code instead.

Also change the registers used in plat_crash_print_regs. The existing
use of x16 and x17 has always been illegal, since those registers are
reserved for use by the linker as a temporary scratch registers in
intra-procedure-call veneers and can never be expected to maintain their
values across a function call.

Change-Id: I8249424150be8d5543ed4af93b56756795a5288f
Signed-off-by: Julius Werner <jwerner@chromium.org>
6 years agorockchip: Use coreboot-supplied serial console on coreboot systems
Julius Werner [Fri, 9 Jun 2017 22:22:44 +0000 (15:22 -0700)]
rockchip: Use coreboot-supplied serial console on coreboot systems

This patch changes all Rockchip platforms to initialize the serial
console with information supplied by coreboot rather than hardcoded
base address and divisor values if BL31 is run on top of coreboot.
Moving the BL2-to-BL31 parameter parsing as early as possible to ensure
that the console is available for all following code.

Also update the Rockchip platform to use MULTI_CONSOLE_API.

Change-Id: I670d350fa2f8b8133539f91ac14977ab47db60d9
Signed-off-by: Julius Werner <jwerner@chromium.org>
6 years agoAdd platform-independent coreboot support library
Julius Werner [Fri, 9 Jun 2017 22:17:15 +0000 (15:17 -0700)]
Add platform-independent coreboot support library

This patch adds the foundation for a platform-independent coreboot
support library that can be shared by all platforms that boot BL31 from
coreboot (acting as BL2). It adds code to parse the "coreboot table", a
data structure that coreboot uses to communicate different kinds of
information to later-stage firmware and certain OS drivers.

As a first small use case for this information, allow platforms to
access the serial console configuration used by coreboot, removing the
need to hardcode base address and divisors and allowing Trusted Firmware
to benefit from coreboot's user configuration (e.g. which UART to pick
and which baud rate to use).

Change-Id: I2bfb39cd2609ce6640b844ab68df6c9ae3f28e9e
Signed-off-by: Julius Werner <jwerner@chromium.org>
6 years agodrivers: cadence: cdns: Update CDNS driver to support MULTI_CONSOLE_API
Julius Werner [Tue, 19 Sep 2017 00:01:06 +0000 (17:01 -0700)]
drivers: cadence: cdns: Update CDNS driver to support MULTI_CONSOLE_API

This patch updates the Cadence CDNS console driver to support the new
console API. The driver will continue to support the old API as well by
checking the MULTI_CONSOLE_API compile-time flag.

Change-Id: I2ef8fb0d6ab72696997db1e0243a533499569d6b
Signed-off-by: Julius Werner <jwerner@chromium.org>
6 years agodrivers: arm: pl011: Update PL011 driver to support MULTI_CONSOLE_API
Julius Werner [Mon, 18 Sep 2017 23:59:43 +0000 (16:59 -0700)]
drivers: arm: pl011: Update PL011 driver to support MULTI_CONSOLE_API

This patch updates the ARM PL011 console driver to support the new
console API. The driver will continue to support the old API as well by
checking the MULTI_CONSOLE_API compile-time flag.

Change-Id: Ic34e4158addbb0c5fae500c9cff899c05a4f4206
Signed-off-by: Julius Werner <jwerner@chromium.org>
6 years agodrivers: ti: uart: Update 16550 UART driver to support MULTI_CONSOLE_API
Julius Werner [Mon, 18 Sep 2017 23:57:51 +0000 (16:57 -0700)]
drivers: ti: uart: Update 16550 UART driver to support MULTI_CONSOLE_API

This patch updates the TI 16550 console driver to support the new
console API. The driver will continue to support the old API as well by
checking the MULTI_CONSOLE_API compile-time flag.

Change-Id: I60a44b7ba3c35c74561824c04b8dbe3e3039324c
Signed-off-by: Julius Werner <jwerner@chromium.org>
6 years agoAdd default crash console code to hook up to new console API
Julius Werner [Mon, 18 Sep 2017 23:49:48 +0000 (16:49 -0700)]
Add default crash console code to hook up to new console API

This patch expands the weak stubs for the plat_crash_console_xxx
functions in common platform code to use the new console API for crash
output. This should make crash console output "just work" for most cases
without the need for the platform to explicitly set up a crash console.
For cases where the normal console framework doesn't work (e.g. very
early crashes, before the platform can register any consoles), platforms
are still able to override the functions just like before.

This feature requires the MULTI_CONSOLE_API compile-time flag to work.
For builds which don't have it set, this patch has no practical effect.

Change-Id: I80dd161cb43f9db59a0bad2dae33c6560cfac584
Signed-off-by: Julius Werner <jwerner@chromium.org>
6 years agoMerge pull request #1200 from robertovargas-arm/bl2-el3
davidcunado-arm [Fri, 19 Jan 2018 13:40:12 +0000 (13:40 +0000)]
Merge pull request #1200 from robertovargas-arm/bl2-el3

Add BL2_AT_EL3 build option

6 years agolib/cpus: fix branching in reset function for cortex-a72 AARCH32 mode
Manoj Kumar [Fri, 19 Jan 2018 12:21:31 +0000 (17:51 +0530)]
lib/cpus: fix branching in reset function for cortex-a72 AARCH32 mode

In AARCH32 mode, cortex_a72_reset_func branches to address in lr
register instead of r5 register. This leads to linux boot failure
of Cortex-A72 cores in AARCH32 mode on Juno-R2 board.

This patch fixes the branching of cortex_a72_reset_func to r5
register as in cortex_a57_reset_func implementation.

Signed-off-by: Manoj Kumar <manoj.kumar3@arm.com>
6 years agobl2-el3: Don't compile BL1 when BL2_AT_EL3 is defined in FVP
Roberto Vargas [Tue, 16 Jan 2018 10:35:23 +0000 (10:35 +0000)]
bl2-el3: Don't compile BL1 when BL2_AT_EL3 is defined in FVP

This patch modifies the makefiles to avoid the definition
of BL1_SOURCES and BL2_SOURCES in the tbbr makefiles, and
it lets to the platform makefiles to define them if they
actually need these images. In the case of BL2_AT_EL3
BL1 will not be needed usually because the Boot ROM will
jump directly to BL2.

Change-Id: Ib6845a260633a22a646088629bcd7387fe35dcf9
Signed-off-by: Roberto Vargas <roberto.vargas@arm.com>
6 years agobl2-el3: Don't include BL2 in fip for BL2 at EL3
Roberto Vargas [Tue, 2 Jan 2018 11:23:41 +0000 (11:23 +0000)]
bl2-el3: Don't include BL2 in fip for BL2 at EL3

It is better to not include BL2 in FIP when using `BL2 at EL3` as
platforms using this config would not have the capability to parse the
FIP format in Boot ROM and BL2 needs to be loaded independently. This
patch does the required changes for the same.

Change-Id: Iad285c247b3440e2d827fef97c3dd81f5c09cabc
Signed-off-by: Roberto Vargas <roberto.vargas@arm.com>
6 years agobl2-el3: Add documentation for BL2 at EL3
Roberto Vargas [Mon, 20 Nov 2017 13:36:10 +0000 (13:36 +0000)]
bl2-el3: Add documentation for BL2 at EL3

Update firmware-design.rst, porting-guide.rst and user-guide.rst
with the information about BL2 at EL3. Firmware-design.rst is
also update to explain how to test this feauture with FVP.

Change-Id: I86d64bc64594e13eb041cea9cefa3f7f3fa745bd
Signed-off-by: Roberto Vargas <roberto.vargas@arm.com>
6 years agoWorkaround for CVE-2017-5715 for Cortex A9, A15 and A17
Dimitris Papastamos [Wed, 3 Jan 2018 10:48:59 +0000 (10:48 +0000)]
Workaround for CVE-2017-5715 for Cortex A9, A15 and A17

A per-cpu vbar is installed that implements the workaround by
invalidating the branch target buffer (BTB) directly in the case of A9
and A17 and indirectly by invalidating the icache in the case of A15.

For Cortex A57 and A72 there is currently no workaround implemented
when EL3 is in AArch32 mode so report it as missing.

For other vulnerable CPUs (e.g. Cortex A73 and Cortex A75), there are
no changes since there is currently no upstream AArch32 EL3 support
for these CPUs.

Change-Id: Ib42c6ef0b3c9ff2878a9e53839de497ff736258f
Signed-off-by: Dimitris Papastamos <dimitris.papastamos@arm.com>
6 years agosp_min: Implement workaround for CVE-2017-5715
Dimitris Papastamos [Tue, 2 Jan 2018 11:37:02 +0000 (11:37 +0000)]
sp_min: Implement workaround for CVE-2017-5715

This patch introduces two workarounds for ARMv7 systems.  The
workarounds need to be applied prior to any `branch` instruction in
secure world.  This is achieved using a custom vector table where each
entry is an `add sp, sp, #1` instruction.

On entry to monitor mode, once the sequence of `ADD` instructions is
executed, the branch target buffer (BTB) is invalidated.  The bottom
bits of `SP` are then used to decode the exception entry type.

A side effect of this change is that the exception vectors are
installed before the CPU specific reset function.  This is now
consistent with how it is done on AArch64.

Note, on AArch32 systems, the exception vectors are typically tightly
integrated with the secure payload (e.g. the Trusted OS).  This
workaround will need porting to each secure payload that requires it.

The patch to modify the AArch32 per-cpu vbar to the corresponding
workaround vector table according to the CPU type will be done in a
later patch.

Change-Id: I5786872497d359e496ebe0757e8017fa98f753fa
Signed-off-by: Dimitris Papastamos <dimitris.papastamos@arm.com>
6 years agoPrint erratum application report for CVE-2017-5715
Dimitris Papastamos [Tue, 16 Jan 2018 10:32:47 +0000 (10:32 +0000)]
Print erratum application report for CVE-2017-5715

Even though the workaround for CVE-2017-5715 is not a CPU erratum, the
code is piggybacking on the errata framework to print whether the
workaround was applied, missing or not needed.

Change-Id: I821197a4b8560c73fd894cd7cd9ecf9503c72fa3
Signed-off-by: Dimitris Papastamos <dimitris.papastamos@arm.com>
6 years agoChange the default errata format string
Dimitris Papastamos [Tue, 16 Jan 2018 10:42:20 +0000 (10:42 +0000)]
Change the default errata format string

As we are using the errata framework to handle workarounds in a more
general sense, change the default string to reflect that.

Change-Id: I2e266af2392c9d95e18fe4e965f9a1d46fd0e95e
Signed-off-by: Dimitris Papastamos <dimitris.papastamos@arm.com>
6 years agobl2-el3: Mark all the assembly functions in bl2 at el3
Roberto Vargas [Fri, 17 Nov 2017 10:51:54 +0000 (10:51 +0000)]
bl2-el3: Mark all the assembly functions in bl2 at el3

When BL2_AT_EL3 option is enabled some platforms are going to
need a resident part in BL2 because the boot rom may jump to it
after a reset. This patch introduces __TEXT_RESIDENT_START__ and
__TEXT_RESIDENT_END__ linker symbols that mark the resident region.

Change-Id: Ib20c1b8ee257831bcc0ca7d3df98d0cb617a04f8
Signed-off-by: Roberto Vargas <roberto.vargas@arm.com>
6 years agoMark functions defined in assembly files
Roberto Vargas [Thu, 2 Nov 2017 16:36:51 +0000 (16:36 +0000)]
Mark functions defined in assembly files

This patch change the name of the section containing the functions
defined in assembly files from text.* to text.asm.*. This change
makes possible to select in the linker script the functions
defined in those files.

Change-Id: If35e44ef1b43ffd951dfac5e052db75d7198e2e0
Signed-off-by: Roberto Vargas <roberto.vargas@arm.com>
6 years agobl2-el3: Add BL2 at EL3 support in FVP
Roberto Vargas [Fri, 17 Nov 2017 13:22:18 +0000 (13:22 +0000)]
bl2-el3: Add BL2 at EL3 support in FVP

This patch add supports for the new API added for BL2 at EL3 for
FVP. We don't have a non-TF Boot ROM for FVP, but this option can be
tested setting specific parameters in the model.

The bl2 image is loaded directly in memory instead of being loaded
by a non-TF Boot ROM and the reset address is changed:

--data cluster0.cpu0=bl2.bin@0x4001000
-C cluster0.cpu0.RVBAR=0x4001000

These parameters mean that in the cold boot path the processor will
jump to BL2 again. For this reason, BL2 is loaded in dram in this
case, to avoid other images reclaiming BL2 memory.

Change-Id: Ieb2ff8535a9e67ccebcd8c2212cad366e7776422
Signed-off-by: Roberto Vargas <roberto.vargas@arm.com>
6 years agobl2-el3: Add BL2_EL3 image
Roberto Vargas [Mon, 30 Oct 2017 14:43:43 +0000 (14:43 +0000)]
bl2-el3: Add BL2_EL3 image

This patch enables BL2 to execute at the highest exception level
without any dependancy on TF BL1. This enables platforms which already
have a non-TF Boot ROM to directly load and execute BL2 and subsequent BL
stages without need for BL1.  This is not currently possible because
BL2 executes at S-EL1 and cannot jump straight to EL3.

Change-Id: Ief1efca4598560b1b8c8e61fbe26d1f44e929d69
Signed-off-by: Roberto Vargas <roberto.vargas@arm.com>
6 years agoMerge pull request #1226 from masahir0y/utils
davidcunado-arm [Wed, 17 Jan 2018 20:22:57 +0000 (20:22 +0000)]
Merge pull request #1226 from masahir0y/utils

utils: rename struct mem_region_t to struct mem_region

6 years agoMerge pull request #1211 from Leo-Yan/remove_ca73_cpu_nap_state
davidcunado-arm [Wed, 17 Jan 2018 17:34:29 +0000 (17:34 +0000)]
Merge pull request #1211 from Leo-Yan/remove_ca73_cpu_nap_state

Hikey960: Remove ca73 cpu nap state

6 years agoemmc: add macros CMD21, BUS_WIDTH_DDR_4 and BUS_WIDTH_DDR_8
Qixiang Xu [Wed, 17 Jan 2018 05:31:21 +0000 (13:31 +0800)]
emmc: add macros CMD21, BUS_WIDTH_DDR_4 and BUS_WIDTH_DDR_8

Add some macros according to JEDEC Standard Embedded Multi-Media
Card (eMMC) Electrical Standard (5.1)": Table 145 - Bus Mode
Selection.

Change-Id: Iaa45e0582653ef4290efd60d039f0bdc420eeb47
Signed-off-by: Qixiang Xu <qixiang.xu@arm.com>
6 years agoutils: rename struct mem_region_t to struct mem_region
Masahiro Yamada [Tue, 16 Jan 2018 14:15:38 +0000 (23:15 +0900)]
utils: rename struct mem_region_t to struct mem_region

typedef mem_region_t mem_region_t;

... seems to work because they belong to different name-spaces,
but humans are confused even if compilers are not.

Signed-off-by: Masahiro Yamada <yamada.masahiro@socionext.com>
6 years agoHikey960: Change CPU standby state for WFI
Leo Yan [Thu, 4 Jan 2018 01:32:50 +0000 (09:32 +0800)]
Hikey960: Change CPU standby state for WFI

At early time, the CPU CA73 retention state has been supported on
Hikey960.  Later we found the system has the hang issue and for
resolving this issue Hisilicon released new MCU firmware, but
unfortunately the new MCU firmware has side effect and results in the
CA73 CPU cannot really enter retention state and roll back to WFI state.

After discussion we cannot see the possibility to enable CA73 retention
state anymore on Hikey960, based on this conclusion we should remove
this state supporting from ARM-TF and roll back to WFI state only.  We
will commit one patch to remove CA73 CPU retention state in kernel DT
binding as well.

Cc: Daniel Lezcano <daniel.lezcano@linaro.org>
Cc: Haojian Zhuang <haojian.zhuang@linaro.org>
Cc: Kevin Wang <jean.wangtao@linaro.org>
Cc: Vincent Guittot <vincent.guittot@linaro.org>
Signed-off-by: Leo Yan <leo.yan@linaro.org>
6 years agoRevert "Hikey960: Change to use recommended power state id format"
Leo Yan [Wed, 3 Jan 2018 06:52:19 +0000 (14:52 +0800)]
Revert "Hikey960: Change to use recommended power state id format"

This reverts commit fdae60b6ba27c216fd86d13b7432a1ff4f57dd84.

The commit fdae60b6ba27c216fd86d13b7432a1ff4f57dd84 changed the
parameter encoding for the hikey960.  However that implies a DT change
in the kernel side.  After submitting the DT change for upstreaming,
the backward compatibility issue and the interface change raise some
concerns from the Linux community about the issues related to kernel <->
ATF alignment.  There is no way to detect a mis-alignment of those
without a deep knowledge of the ATF and the kernel.  Furthermore, the
failing calls to PSCI in the idle path (because of bad parameters), will
lead to busy looping, implying: thermal issues and extra energy
consumption.

In regard of the Linux community concerns, the potential issues when the
ATF and the kernel are not aligned, it is preferable to revert the
commit.

Cc: Vincent Guittot <vincent.guittot@linaro.org>
Cc: Haojian Zhuang <haojian.zhuang@linaro.org>
Cc: Kevin Wang <jean.wangtao@linaro.org>
Co-authored-by: Daniel Lezcano <daniel.lezcano@linaro.org>
Signed-off-by: Leo Yan <leo.yan@linaro.org>
6 years agoMerge pull request #1221 from hzhuang1/fix_HLE
davidcunado-arm [Tue, 16 Jan 2018 10:44:30 +0000 (10:44 +0000)]
Merge pull request #1221 from hzhuang1/fix_HLE

emmc/dw_mmc: fix the assert on HLE bit

6 years agoMerge pull request #1219 from antonio-nino-diaz-arm/an/mm-version
davidcunado-arm [Tue, 16 Jan 2018 01:10:26 +0000 (01:10 +0000)]
Merge pull request #1219 from antonio-nino-diaz-arm/an/mm-version

SPM: Fix version header definitions

6 years agoMerge pull request #1218 from antonio-nino-diaz-arm/an/xlat-fix
davidcunado-arm [Tue, 16 Jan 2018 01:10:13 +0000 (01:10 +0000)]
Merge pull request #1218 from antonio-nino-diaz-arm/an/xlat-fix

xlat v2: Correctly unmap regions on map error

6 years agoMerge pull request #1216 from hzhuang1/fix_psci
davidcunado-arm [Tue, 16 Jan 2018 01:09:57 +0000 (01:09 +0000)]
Merge pull request #1216 from hzhuang1/fix_psci

Hikey960: Fix hikey960 pcie mount fail

6 years agoMerge pull request #1215 from vwadekar/tlkd-ns-dram-ranges
davidcunado-arm [Tue, 16 Jan 2018 01:09:37 +0000 (01:09 +0000)]
Merge pull request #1215 from vwadekar/tlkd-ns-dram-ranges

spd: tlkd: support for "NS memory ranges" function ID

6 years agoMerge pull request #1217 from robertovargas-arm/doc-plat_try_next_boot_source
davidcunado-arm [Mon, 15 Jan 2018 23:58:52 +0000 (23:58 +0000)]
Merge pull request #1217 from robertovargas-arm/doc-plat_try_next_boot_source

Add documentation about plat_try_next_boot_source to bl1_platform_setup

6 years agoMerge pull request #1213 from masahir0y/uniphier
davidcunado-arm [Mon, 15 Jan 2018 23:58:20 +0000 (23:58 +0000)]
Merge pull request #1213 from masahir0y/uniphier

uniphier: clean-up platform makefile

6 years agoMerge pull request #1225 from dp-arm/dp/amu-remove-warn
davidcunado-arm [Mon, 15 Jan 2018 16:37:39 +0000 (16:37 +0000)]
Merge pull request #1225 from dp-arm/dp/amu-remove-warn

AMU: Remove unnecessary WARN()

6 years agoAMU: Remove unnecessary WARN()
Dimitris Papastamos [Mon, 15 Jan 2018 14:52:57 +0000 (14:52 +0000)]
AMU: Remove unnecessary WARN()

If AMU is not supported by the hardware but it is enabled in Trusted
Firmware, the console will be spammed with warnings every time a CPU
is brought up with a CPU ON call.

Remove the warning message as this is more in line with how other
extensions like SPE and SVE are handled.

Change-Id: Iba6d367e4d1375ab554d23d2eaceab3ae1362c5a
Signed-off-by: Dimitris Papastamos <dimitris.papastamos@arm.com>
6 years agoMerge pull request #1197 from dp-arm/dp/amu
davidcunado-arm [Fri, 12 Jan 2018 09:02:24 +0000 (09:02 +0000)]
Merge pull request #1197 from dp-arm/dp/amu

AMUv1 support

6 years agoMerge pull request #1214 from dp-arm/dp/cve_2017_5715
davidcunado-arm [Thu, 11 Jan 2018 23:39:30 +0000 (23:39 +0000)]
Merge pull request #1214 from dp-arm/dp/cve_2017_5715

Workarounds for CVE-2017-5715 on Cortex A57/A72/A73 and A75

6 years agoMerge pull request #1222 from davidcunado-arm/dp/bl31_mem
davidcunado-arm [Thu, 11 Jan 2018 18:33:43 +0000 (18:33 +0000)]
Merge pull request #1222 from davidcunado-arm/dp/bl31_mem

Increase BL31 memory space by 2 pages

6 years agoIncrease BL31 memory space by 2 pages
Dimitris Papastamos [Thu, 11 Jan 2018 15:32:32 +0000 (15:32 +0000)]
Increase BL31 memory space by 2 pages

On some build configurations BL31 is running out of space.  Now that
TSP is moved to secure dram, we have a bit of additional space to use
in BL31.

Change-Id: Ib89fcd8bae99c85c9c5e5d9228bb42fb7048dcb6
Signed-off-by: Dimitris Papastamos <dimitris.papastamos@arm.com>
Signed-off-by: David Cunado <david.cunado@arm.com>
6 years agoAdd hooks to save/restore AMU context for Cortex A75
Dimitris Papastamos [Mon, 11 Dec 2017 11:45:35 +0000 (11:45 +0000)]
Add hooks to save/restore AMU context for Cortex A75

Change-Id: I504d3f65ca5829bc1f4ebadb764931f8379ee81f
Signed-off-by: Dimitris Papastamos <dimitris.papastamos@arm.com>
6 years agoAMU: Add hooks to save/restore AMU context
Dimitris Papastamos [Tue, 28 Nov 2017 13:47:06 +0000 (13:47 +0000)]
AMU: Add hooks to save/restore AMU context

On some systems, the AMU counters might reset to 0 when a CPU
powerdown happens.  This behaviour conflicts with the intended
use-case of AMU as lower ELs are only expected to see non-decreasing
counter values.

Change-Id: If25519965d4e6e47e09225d0e732947986cbb5ec
Signed-off-by: Dimitris Papastamos <dimitris.papastamos@arm.com>
6 years agoAMU: Add configuration helpers for aarch64
Dimitris Papastamos [Mon, 13 Nov 2017 09:49:45 +0000 (09:49 +0000)]
AMU: Add configuration helpers for aarch64

Add some AMU helper functions to allow configuring, reading and
writing of the Group 0 and Group 1 counters.  Documentation for these
helpers will come in a separate patch.

Change-Id: I656e070d2dae830c22414f694aa655341d4e2c40
Signed-off-by: Dimitris Papastamos <dimitris.papastamos@arm.com>
6 years agoAMU: Add plat interface to select which group 1 counters to enable
Dimitris Papastamos [Wed, 13 Dec 2017 10:54:37 +0000 (10:54 +0000)]
AMU: Add plat interface to select which group 1 counters to enable

A new platform macro `PLAT_AMU_GROUP1_COUNTERS_MASK` controls which
group 1 counters should be enabled. The maximum number of group 1
counters supported by AMUv1 is 16 so the mask can be at most 0xffff.
If the platform does not define this mask, no group 1 counters are
enabled.

A related platform macro `PLAT_AMU_GROUP1_NR_COUNTERS` is used by
generic code to allocate an array to save and restore the counters on
CPU suspend.

Change-Id: I6d135badf4846292de931a43bb563077f42bb47b
Signed-off-by: Dimitris Papastamos <dimitris.papastamos@arm.com>
6 years agoAdd PubSub events for CPU powerdown/powerup
Dimitris Papastamos [Tue, 28 Nov 2017 15:16:00 +0000 (15:16 +0000)]
Add PubSub events for CPU powerdown/powerup

The suspend hook is published at the start of a CPU powerdown
operation.  The resume hook is published at the end of a CPU powerup
operation.

Change-Id: I50c05e2dde0d33834095ac41b4fcea4c161bb434
Signed-off-by: Dimitris Papastamos <dimitris.papastamos@arm.com>
6 years agoUse PFR0 to identify need for mitigation of CVE-2017-5915
Dimitris Papastamos [Tue, 2 Jan 2018 15:53:01 +0000 (15:53 +0000)]
Use PFR0 to identify need for mitigation of CVE-2017-5915

If the CSV2 field reads as 1 then branch targets trained in one
context cannot affect speculative execution in a different context.
In that case skip the workaround on Cortex A75.

Change-Id: I4d5504cba516a67311fb5f0657b08f72909cbd38
Signed-off-by: Dimitris Papastamos <dimitris.papastamos@arm.com>
6 years agoWorkaround for CVE-2017-5715 on Cortex A73 and A75
Dimitris Papastamos [Mon, 18 Dec 2017 13:46:21 +0000 (13:46 +0000)]
Workaround for CVE-2017-5715 on Cortex A73 and A75

Invalidate the Branch Target Buffer (BTB) on entry to EL3 by
temporarily dropping into AArch32 Secure-EL1 and executing the
`BPIALL` instruction.

This is achieved by using 3 vector tables.  There is the runtime
vector table which is used to handle exceptions and 2 additional
tables which are required to implement this workaround.  The
additional tables are `vbar0` and `vbar1`.

The sequence of events for handling a single exception is
as follows:

1) Install vector table `vbar0` which saves the CPU context on entry
   to EL3 and sets up the Secure-EL1 context to execute in AArch32 mode
   with the MMU disabled and I$ enabled.  This is the default vector table.

2) Before doing an ERET into Secure-EL1, switch vbar to point to
   another vector table `vbar1`.  This is required to restore EL3 state
   when returning from the workaround, before proceeding with normal EL3
   exception handling.

3) While in Secure-EL1, the `BPIALL` instruction is executed and an
   SMC call back to EL3 is performed.

4) On entry to EL3 from Secure-EL1, the saved context from step 1) is
   restored.  The vbar is switched to point to `vbar0` in preparation to
   handle further exceptions.  Finally a branch to the runtime vector
   table entry is taken to complete the handling of the original
   exception.

This workaround is enabled by default on the affected CPUs.

NOTE
====

There are 4 different stubs in Secure-EL1.  Each stub corresponds to
an exception type such as Sync/IRQ/FIQ/SError.  Each stub will move a
different value in `R0` before doing an SMC call back into EL3.
Without this piece of information it would not be possible to know
what the original exception type was as we cannot use `ESR_EL3` to
distinguish between IRQs and FIQs.

Change-Id: I90b32d14a3735290b48685d43c70c99daaa4b434
Signed-off-by: Dimitris Papastamos <dimitris.papastamos@arm.com>
6 years agoWorkaround for CVE-2017-5715 on Cortex A57 and A72
Dimitris Papastamos [Thu, 30 Nov 2017 14:53:53 +0000 (14:53 +0000)]
Workaround for CVE-2017-5715 on Cortex A57 and A72

Invalidate the Branch Target Buffer (BTB) on entry to EL3 by disabling
and enabling the MMU.  To achieve this without performing any branch
instruction, a per-cpu vbar is installed which executes the workaround
and then branches off to the corresponding vector entry in the main
vector table.  A side effect of this change is that the main vbar is
configured before any reset handling.  This is to allow the per-cpu
reset function to override the vbar setting.

This workaround is enabled by default on the affected CPUs.

Change-Id: I97788d38463a5840a410e3cea85ed297a1678265
Signed-off-by: Dimitris Papastamos <dimitris.papastamos@arm.com>
6 years agoemmc/dw_mmc: fix the assert on HLE bit
Haojian Zhuang [Thu, 11 Jan 2018 05:49:56 +0000 (13:49 +0800)]
emmc/dw_mmc: fix the assert on HLE bit

When check HLE bit in interrupt register, it should check whether
HLE bit is set, not clear.

Signed-off-by: Haojian Zhuang <haojian.zhuang@linaro.org>
6 years agoxlat v2: Correctly unmap regions on map error
Antonio Nino Diaz [Fri, 5 Jan 2018 11:30:36 +0000 (11:30 +0000)]
xlat v2: Correctly unmap regions on map error

`mm_cursor` doesn't have the needed data because the `memmove()` that
is called right before it overwrites that information. In order to get
the information of the region that was being mapped, `mm` has to be used
instead (like it is done to fill the fields of `unmap_mm`).

If the incorrect information is read, this check isn't reliable and
`xlat_tables_unmap_region` may be requested to unmap memory that isn't
mapped at all, triggering assertions.

Change-Id: I602d4ac83095d4e5dac9deb34aa5d00d00e6c289
Signed-off-by: Antonio Nino Diaz <antonio.ninodiaz@arm.com>
6 years agoSPM: Fix version header definitions
Antonio Nino Diaz [Mon, 8 Jan 2018 17:33:34 +0000 (17:33 +0000)]
SPM: Fix version header definitions

Rename SP_VERSION macros to MM_VERSION, which is the name used in the MM
specification [1]. Also, a few more helper macros have been added.

MM-specific definitions have been moved to their own header file.

[1] http://infocenter.arm.com/help/topic/com.arm.doc.den0060a/DEN0060A_ARM_MM_Interface_Specification.pdf

Change-Id: Ia10e48c7e81a7a1f5eeca29a5270cae740a4a88a
Signed-off-by: Antonio Nino Diaz <antonio.ninodiaz@arm.com>
6 years agoMerge pull request #1176 from wjliang/zynqmp-ipi-mb-svc
davidcunado-arm [Wed, 10 Jan 2018 14:57:22 +0000 (14:57 +0000)]
Merge pull request #1176 from wjliang/zynqmp-ipi-mb-svc

plat: xilinx: Add ZynqMP IPI mailbox service [v4]

6 years agoAdd documentation about plat_try_next_boot_source to bl1_platform_setup
Roberto Vargas [Tue, 12 Dec 2017 10:39:44 +0000 (10:39 +0000)]
Add documentation about plat_try_next_boot_source to bl1_platform_setup

If boot redundancy is required in BL1 then the initialization
of the boot sequence must be done in bl1_platform_setup. In BL2,
we had to add a new function, bl2_preload_setup, because
bl2_platform_setup is called after the images are loaded, making it
invalid for the boot sequence initialization.

Change-Id: I5c177ff142608ed38b4192288b06614343b2b83b
Signed-off-by: Roberto Vargas <roberto.vargas@arm.com>
6 years agoMerge pull request #1208 from masahir0y/build
davidcunado-arm [Wed, 10 Jan 2018 01:10:51 +0000 (01:10 +0000)]
Merge pull request #1208 from masahir0y/build

Build: trivial fixes

6 years agoMerge pull request #1207 from hzhuang1/isp_clk
davidcunado-arm [Wed, 10 Jan 2018 00:21:25 +0000 (00:21 +0000)]
Merge pull request #1207 from hzhuang1/isp_clk

hikey960: set isp clks as unsecure mode

6 years agoHikey960: Fix hikey960 pcie mount fail
Kaihua Zhong [Tue, 9 Jan 2018 08:51:38 +0000 (16:51 +0800)]
Hikey960: Fix hikey960 pcie mount fail

Set IOC_AO_IOMG_033 function from GPIO213 to PCIE_CLKREQ_N

bit[0-2]:  000: GPIO_213;
           001: PCIE_CLKREQ_N;
           010: GPIO_018_SH;
           100: GPIO_014_SE;
           110: FAC_TEST24;
           111: FAC_TEST24;
bit[3-31]: reserved

Signed-off-by: Guangtao Zhang <zhangguangtao@hisilicon.com>
Tested-by: Yao Chen <chenyao11@huawei.com>
Acked-by: Haojian Zhuang <haojian.zhuang@linaro.org>
6 years agoMerge pull request #1167 from Leo-Yan/hikey-fix-alignment
davidcunado-arm [Tue, 9 Jan 2018 08:58:41 +0000 (08:58 +0000)]
Merge pull request #1167 from Leo-Yan/hikey-fix-alignment

Set alignment size to 512B for Hikey/Hikey960

6 years agozynqmp: pm_service: use zynqmp_ipi APIs
Wendy Liang [Wed, 4 Oct 2017 06:21:11 +0000 (23:21 -0700)]
zynqmp: pm_service: use zynqmp_ipi APIs

Use zynqmp_ipi APIs to access IPI registers in pm_service.
As the zynqmp_ipi APIs doesn't cover IPI buffers, the pm_ipi
in pm_service will still directly access the IPI buffers.

Signed-off-by: Wendy Liang <jliang@xilinx.com>
6 years agoAdd Xilinx ZynqMP IPI mailbox service
Wendy Liang [Wed, 6 Sep 2017 16:39:55 +0000 (09:39 -0700)]
Add Xilinx ZynqMP IPI mailbox service

Add IPI mailbox service to manage Xilinx ZynqMP IPI(Inter Processors
Interrupt) access.

Signed-off-by: Wendy Liang <jliang@xilinx.com>
6 years agoIntroduce ZynqMP IPI implementation
Wendy Liang [Wed, 13 Sep 2017 18:02:42 +0000 (11:02 -0700)]
Introduce ZynqMP IPI implementation

Previously, ZynqMP IPI in ATF is only for ZynqMP PM,
This patch is to have a ZynqMP IPI implementation to handle
both ZynqMP PM IPI requirement and IPI mailbox service requirement
which will be introduced next.

We control IPI agents registers access but not IPI buffers access in
this implementation. Each IPI mailbox user will directly access the
IPI buffers.

Signed-off-by: Wendy Liang <jliang@xilinx.com>
6 years agospd: tlkd: support for "NS memory ranges" function ID
Varun Wadekar [Mon, 8 Jan 2018 19:35:40 +0000 (11:35 -0800)]
spd: tlkd: support for "NS memory ranges" function ID

This patch adds support to receive function ID with NS world's
memory ranges to provide the memory snapshot to TLK.

Signed-off-by: Varun Wadekar <vwadekar@nvidia.com>
6 years agoMerge pull request #1202 from antonio-nino-diaz-arm/an/spm-secondary-cores
davidcunado-arm [Mon, 8 Jan 2018 22:59:37 +0000 (22:59 +0000)]
Merge pull request #1202 from antonio-nino-diaz-arm/an/spm-secondary-cores

SPM: Allow secondary CPUs to use the Secure Partition

6 years agoSPM: Allow secondary CPUs to use the Secure Partition
Antonio Nino Diaz [Mon, 8 Jan 2018 09:59:33 +0000 (09:59 +0000)]
SPM: Allow secondary CPUs to use the Secure Partition

The Secure Partition should be able to be used from any CPU, not just
the lead one. This patch point the secure contexts of all secondary
CPUs to the same one used by the lead CPU for the Secure Partition. This
way, they can also use it.

In order to prevent more than one CPU from using the Secure Partition at
the same time, a lock has been added.

Change-Id: Ica76373127c3626498b06c558a4874ce72201ff7
Signed-off-by: Antonio Nino Diaz <antonio.ninodiaz@arm.com>
6 years agoSPM: Move initialization flag to context struct
Antonio Nino Diaz [Mon, 18 Dec 2017 10:51:58 +0000 (10:51 +0000)]
SPM: Move initialization flag to context struct

Whether a Secure Partition is being initialized or not is something
related to that specific partition, so it should be saved with the
rest of the information related to it.

Change-Id: Ie8a780f70df83fb03ef9c01ba37960208d9b5319
Signed-off-by: Antonio Nino Diaz <antonio.ninodiaz@arm.com>
6 years agouniphier: simplify GZIP compress rule
Masahiro Yamada [Thu, 4 Jan 2018 03:59:11 +0000 (12:59 +0900)]
uniphier: simplify GZIP compress rule

It is not necessary to read data from stdin.  The input file name
is ripped off by -n option, anyway.  I still use the redirect for
the output to specify the output file name.

Signed-off-by: Masahiro Yamada <yamada.masahiro@socionext.com>
6 years agoMerge pull request #1204 from davidcunado-arm/rv/fip_tool
davidcunado-arm [Wed, 3 Jan 2018 23:48:51 +0000 (23:48 +0000)]
Merge pull request #1204 from davidcunado-arm/rv/fip_tool

Add padding at the end of the last entry

6 years agoMerge pull request #1206 from davidcunado-arm/dc/update_userguide
davidcunado-arm [Wed, 3 Jan 2018 21:13:43 +0000 (21:13 +0000)]
Merge pull request #1206 from davidcunado-arm/dc/update_userguide

Update dependencies for ARM TF

6 years agodocs: Update the ToC end marker description in the document
Jett Zhou [Fri, 24 Nov 2017 08:03:58 +0000 (16:03 +0800)]
docs: Update the ToC end marker description in the document

Change-Id: I2e29a63f08aed3b8ea0bb10170a3d55b8d033e62
Signed-off-by: Jett Zhou <jett.zhou@arm.com>
Signed-off-by: David Cunado <david.cunado@arm.com>
6 years agoAdd padding at the end of the last entry
Roberto Vargas [Tue, 19 Dec 2017 11:56:57 +0000 (11:56 +0000)]
Add padding at the end of the last entry

This patch adds padding bytes at the end of the last image in the
fip to be able to transfer by DMA the last image.

Change-Id: I8c6f07dee389cb3d1dc919936d9d52841d7e5723
Signed-off-by: Roberto Vargas <roberto.vargas@arm.com>
Signed-off-by: Dimitris Papastamos <dimitris.papastamos@arm.com>
Signed-off-by: David Cunado <david.cunado@arm.com>
6 years agoMerge pull request #1212 from dp-arm/dp/tsp_dram
davidcunado-arm [Wed, 3 Jan 2018 11:20:56 +0000 (11:20 +0000)]
Merge pull request #1212 from dp-arm/dp/tsp_dram

Move TSP to TZC secured DRAM

6 years agoMove TSP to TZC secured DRAM
Dimitris Papastamos [Tue, 2 Jan 2018 10:25:50 +0000 (10:25 +0000)]
Move TSP to TZC secured DRAM

To allow BL31 to grow in SRAM, move TSP in TZC secured DRAM
by default.

Increase the BL31 max limit by one page.

Change-Id: Idd3479be02f0f9bafac2f275376d7db0c2015431
Signed-off-by: Dimitris Papastamos <dimitris.papastamos@arm.com>
6 years agoMerge pull request #1203 from masahir0y/uniphier
davidcunado-arm [Sun, 24 Dec 2017 19:52:17 +0000 (19:52 +0000)]
Merge pull request #1203 from masahir0y/uniphier

uniphier: a bundle of fixes