emmc: support CMD23 ...

Support CMD23. When CMD23 is used, CMD12 could be avoided.

Two scenarios:
1. CMD17 for single block, CMD18 + CMD12 for multiple blocks.
2. CMD23 + CMD18 for both single block and multiple blocks.

The emmc_init() should initialize whether CMD23 is supported
or not.

Signed-off-by: Haojian Zhuang <haojian.zhuang@linaro.org>

GICv3: Fix the GICD_IROUTER offset ...

This patch fixes the offset of GICD_IROUTER register defined in gicv3.h.
Although the GICv3 documention mentions that the offset for this register
is 0x6100-0x7FD8, the offset calculation for an interrupt id `n` is :

   0x6000 + 8n, where n >= 32

This requires the offset for GICD_IROUTER to be defined as 0x6000.

Fixes ARM-software/tf-issues#410

Change-Id: If9e91e30d946afe7f1f60fea4f065c7567093fa8

Rework type usage in Trusted Firmware ...

This patch reworks type usage in generic code, drivers and ARM platform files
to make it more portable. The major changes done with respect to
type usage are as listed below:

* Use uintptr_t for storing address instead of uint64_t or unsigned long.
* Review usage of unsigned long as it can no longer be assumed to be 64 bit.
* Use u_register_t for register values whose width varies depending on
  whether AArch64 or AArch32.
* Use generic C types where-ever possible.

In addition to the above changes, this patch also modifies format specifiers
in print invocations so that they are AArch64/AArch32 agnostic. Only files
related to upcoming feature development have been reworked.

Change-Id: I9f8c78347c5a52ba7027ff389791f1dad63ee5f8

Merge pull request #629 from ljerry/tf_issue_398 ...

Bring IO storage dummy driver

Merge pull request #632 from rockchip-linux/support-for-gpio-driver-v2 ...

rockchip/rk3399: Support the gpio driver and configure

gpio: support gpio set/get pull status ...

On some platform gpio can set/get pull status when input, add these
function so we can set/get gpio pull status when need it. And they are
optional function.

CCN: Add API to query the PART0 ID from CCN ...

This patch adds the API `ccn_get_part0_id` to query the PART0 ID from the
PERIPHERAL_ID 0 register in the CCN driver. This ID allows to distinguish
the variant of CCN present on the system and possibly enable dynamic
configuration of the IP based on the variant. Also added an assert in
`ccn_master_to_rn_id_map()` to ensure that the master map bitfield provided
by the platform is within the expected interface id.

Change-Id: I92d2db7bd93a9be8a7fbe72a522cbcba0aba2d0e

Implement generic delay timer ...

Add delay timer implementation based on the system generic counter.
This either uses the platform's implementation of
`plat_get_syscnt_freq()` or explicit clock multiplier/divider values
provided by the platform.

The current implementation of udelay has been modified to avoid
unnecessary calculations while waiting on the loop and to make it
easier to check for overflows.

Change-Id: I9062e1d506dc2f68367fd9289250b93444721732

Bring IO storage dummy driver ...

Allow to handle cases where some images are pre-loaded (by debugger for
instance) without introducing many switches in files calling load_* functions.

Fixes: arm-software/tf-issues#398
Signed-off-by: Gerald Lejeune <gerald.lejeune@st.com>

drivers: add emmc stack ...

In a lot of embedded platforms, eMMC device is the only one storage
device. So loading content from eMMC device is required in ATF.

Create the emmc stack that could co-work with IO block driver.
Support to read/write/erase eMMC blocks on both rpmb and normal
user area. Support to change the IO speed and bus width.

Signed-off-by: Haojian Zhuang <haojian.zhuang@linaro.org>

IO: support block device type ...

FIP is accessed as memory-mapped type. eMMC is block device type.
In order to support FIP based on eMMC, add the new io_block layer.

io_block always access eMMC device as block size. And it'll only
copy the required data into buffer in io_block driver. So preparing
an temporary buffer is required.

When use io_block device, MAX_IO_BLOCK_DEVICES should be declared
in platform_def.h. It's used to support multiple block devices.

Signed-off-by: Haojian Zhuang <haojian.zhuang@linaro.org>

Use unsigned long long instead of uintptr_t in TZC400/DMC500 drivers ...

Currently the `tzc400_configure_region` and `tzc_dmc500_configure_region`
functions uses uintptr_t as the data type for `region_top` and `region_base`
variables, which will be converted to 32/64 bits for AArch32/AArch64
respectively. But the expectation is to keep these addresses at least 64 bit.

This patch modifies the data types to make it at least 64 bit by using
unsigned long long instead of uintptr_t for the `region_top` and
`region_base` variables. It also modifies the associated macros
`_tzc##fn_name##_write_region_xxx` accordingly.

Change-Id: I4e3c6a8a39ad04205cf0f3bda336c3970b15a28b

Merge pull request #569 from Xilinx/zynqmp-v1 ...

Support for Xilinx Zynq UltraScale+ MPSoC

Merge pull request #575 from soby-mathew/sm/new_tzc_driver ...

Refactor the TZC driver and add DMC-500 driver

drivers: Add Cadence UART driver ...

Add a driver for the Cadence UART which is found in Xilinx Zynq SOCs.

Signed-off-by: Soren Brinkmann <soren.brinkmann@xilinx.com>

Add ARM CoreLink DMC-500 driver to program TrustZone protection ...

The ARM CoreLink DMC-500 Dynamic Memory Controller provides the
programmable address region control of a TrustZone Address Space
Controller. The access permissions can be defined for eight
separate address regions plus a background or default region.
This patch adds a DMC-500 driver to define address regions and
program their access permissions as per ARM 100131_0000_02_en
(r0p0) document.

Change-Id: I9d33120f9480d742bcf7937e4b876f9d40c727e6

Refactor the ARM CoreLink TZC-400 driver ...

TrustZone protection can be programmed by both memory and TrustZone
address space controllers like DMC-500 and TZC-400. These peripherals
share a similar programmer's view.

Furthermore, it is possible to have multiple instances of each type of
peripheral in a system resulting in multiple programmer's views.
For example, on the TZC-400 each of the 4 filter units can be enabled
or disabled for each region. There is a single set of registers to
program the region attributes. On the DMC-500, each filter unit has its
own programmer's view resulting in multiple sets of registers to program
the region attributes. The layout of the registers is almost the same
across all these variations.

Hence the existing driver in `tzc400\tzc400.c` is refactored into the
new driver in `tzc\tzc400.c`. The previous driver file is still maintained
for compatibility and it is now deprecated.

Change-Id: Ieabd0528e244582875bc7e65029a00517671216d

TBB: add non-volatile counter support ...

This patch adds support for non-volatile counter authentication to
the Authentication Module. This method consists of matching the
counter values provided in the certificates with the ones stored
in the platform. If the value from the certificate is lower than
the platform, the boot process is aborted. This mechanism protects
the system against rollback.

The TBBR CoT has been updated to include this method as part of the
authentication process. Two counters are used: one for the trusted
world images and another for the non trusted world images.

** NEW PLATFORM APIs (mandatory when TBB is enabled) **

int plat_get_nv_ctr(void *cookie, unsigned int *nv_ctr);

    This API returns the non-volatile counter value stored
    in the platform. The cookie in the first argument may be
    used to select the counter in case the platform provides
    more than one (i.e. TBSA compliant platforms must provide
    trusted and non-trusted counters). This cookie is specified
    in the CoT.

int plat_set_nv_ctr(void *cookie, unsigned int nv_ctr);

    This API sets a new counter value. The cookie may be
    used to select the counter to be updated.

An implementation of these new APIs for ARM platforms is also
provided. The values are obtained from the Trusted Non-Volatile
Counters peripheral. The cookie is used to pass the extension OID.
This OID may be interpreted by the platform to know which counter
must return. On Juno, The trusted and non-trusted counter values
have been tied to 31 and 223, respectively, and cannot be modified.

** IMPORTANT **

THIS PATCH BREAKS THE BUILD WHEN TRUSTED_BOARD_BOOT IS ENABLED. THE
NEW PLATFORM APIs INTRODUCED IN THIS PATCH MUST BE IMPLEMENTED IN
ORDER TO SUCCESSFULLY BUILD TF.

Change-Id: Ic943b76b25f2a37f490eaaab6d87b4a8b3cbc89a

Merge pull request #518 from hzhuang1/pl061_gpio_v5 ...

Pl061 gpio v5

arm: gpio: add pl061 driver ...

Add PL061 GPIO driver that is depend on gpio framework.

Signed-off-by: Haojian Zhuang <haojian.zhuang@linaro.org>

gpio: add gpio framework ...

Define the gpio ops in gpio driver.

Signed-off-by: Haojian Zhuang <haojian.zhuang@linaro.org>

Move private APIs in gic_common.h to a private header ...

This patch moves the private GIC common accessors from `gic_common.h` to
a new private header file `gic_common_private.h`. This patch also adds
additional comments to GIC register accessors to highlight the fact
that some of them access register values that correspond to multiple
interrupt IDs. The convention used is that the `set`, `get` and `clr`
accessors access and modify the values corresponding to a single interrupt
ID whereas the `read` and `write` GIC register accessors access the raw
GIC registers and it could correspond to multiple interrupt IDs depending
on the register accessed.

Change-Id: I2643ecb2533f01e3d3219fcedfb5f80c120622f9

Fix GIC_IPRIORITYR setting in new drivers ...

The code to set the interrupt priority for secure interrupts in the
new GICv2 and GICv3 drivers is incorrect. The setup code to configure
interrupt priorities of secure interrupts, one interrupt at a time, used
gicd_write_ipriorityr()/gicr_write_ipriority() function affecting
4 interrupts at a time. This bug did not manifest itself because all the
secure interrupts were configured to the highest secure priority(0) during
cold boot and the adjacent non secure interrupt priority would be configured
later by the normal world. This patch introduces new accessors,
gicd_set_ipriorityr() and gicr_set_ipriorityr(), for configuring priority
one interrupt at a time and fixes the the setup code to use the new
accessors.

Fixes ARM-software/tf-issues#344

Change-Id: I470fd74d2b7fce7058b55d83f604be05a27e1341

Merge pull request #501 from jcastillo-arm/jc/tf-issues/300 ...

Disable PL011 UART before configuring it

Disable PL011 UART before configuring it ...

The PL011 TRM (ARM DDI 0183G) specifies that the UART must be
disabled before any of the control registers are programmed. The
PL011 driver included in TF does not disable the UART, so the
initialization in BL2 and BL31 is violating this requirement
(and potentially in BL1 if the UART is enabled after reset).

This patch modifies the initialization function in the PL011
console driver to disable the UART before programming the
control registers.

Register clobber list and documentation updated.

Fixes ARM-software/tf-issues#300

Change-Id: I839b2d681d48b03f821ac53663a6a78e8b30a1a1

Remove direct usage of __attribute__((foo)) ...

Migrate all direct usage of __attribute__ to usage of their
corresponding macros from cdefs.h.
e.g.:
 - __attribute__((unused)) -> __unused

Signed-off-by: Soren Brinkmann <soren.brinkmann@xilinx.com>

Migrate __warn_deprecated -> __deprecated ...

Use the new __deprecated macro from the generic cdefs header and remove
the deprecated __warn_deprecated.

Signed-off-by: Soren Brinkmann <soren.brinkmann@xilinx.com>

Rearrange fields in TF data structures to reduce padding ...

This patch rearranges fields of the `image_desc_t` & `auth_img_desc_t`
data structures to reduce padding between the fields and thereby
save memory.

NOTE: Platform ports which anonymously initialize these structures
should be aware of the rearrangement and do the required
modification.

Change-Id: I428b5429632797b31d5bd306174599c07e24c060

Always build with '-pedantic' ...

By default ARM TF is built with the '-pedantic' compiler flag, which
helps detecting violations of the C standard. However, the mbed TLS
library and its associated authentication module in TF used to fail
building with this compiler flag. As a workaround, the mbed TLS
authentication module makefile used to set the 'DISABLE_PEDANTIC'
TF build flag.

The compiler errors flagged by '-pedantic' in the mbed TLS library
have been fixed between versions 1.3.9 and 2.2.0 and the library now
properly builds with this compiler flag.

This patch fixes the remaining compiler errors in the mbed TLS
authentication module in TF and unsets the 'DISABLE_PEDANTIC' TF
build flag. This means that TF is now always built with '-pedantic'.

In particular, this patch:

 * Removes the final semi-colon in REGISTER_COT() macro.

   This semi-colon was causing the following error message:

   drivers/auth/tbbr/tbbr_cot.c:544:23: error: ISO C does not allow
   extra ';' outside of a function [-Werror=pedantic]

   This has been fixed both in the mbed TLS authentication module
   as well as in the certificate generation tool. Note that the latter
   code didn't need fixing since it is not built with '-pedantic' but
   the change has been propagated for consistency.

   Also fixed the REGISTER_KEYS() and REGISTER_EXTENSIONS() macros,
   which were suffering from the same issue.

 * Fixes a pointer type.

   It was causing the following error message:

   drivers/auth/mbedtls/mbedtls_crypto.c: In function 'verify_hash':
   drivers/auth/mbedtls/mbedtls_crypto.c:177:42: error: pointer of
   type 'void *' used in arithmetic [-Werror=pointer-arith]

Change-Id: I7b7a04ef711efd65e17b5be26990d1a0d940257d

Merge pull request #465 from jcastillo-arm/jc/tbb_mbedtls_2_x ...

Move up to mbed TLS 2.x