2015-10-27 |
Make: fix dependency files generation
...
Currently, if no make goal is specified in the command line, 'all'
is assumed by default, but the dependency files are not generated.
This might lead to a successful but inconsistent build. This patch
provides a fix to the problem.
Change-Id: I0148719e114dbdbe46f8a57c7d05da7cbc212c92
Juan Castillo
committed
on 27 Oct 2015
|
Rework Makefile
...
This patch is a complete rework of the main Makefile. Functionality
remains the same but the code has been reorganized in sections in
order to improve readability and facilitate adding future extensions.
A new file 'build_macros.mk' has been created and will contain common
definitions (variables, macros, etc) that may be used from the main
Makefile and other platform specific makefiles.
A new macro 'FIP_ADD_IMG' has been introduced and it will allow the
platform to specify binary images and the necessary checks for a
successful build. Platforms that require a BL30 image no longer need
to specify the NEED_BL30 option. The main Makefile is now completely
unaware of additional images not built as part of Trusted Firmware,
like BL30. It is the platform responsibility to specify images using
the macro 'FIP_ADD_IMG'. Juno uses this macro to include the BL30
image in the build.
BL33 image is specified in the main Makefile to preserve backward
compatibility with the NEED_BL33 option. Otherwise, platform ports
that rely on the definition of NEED_BL33 might break.
All Trusted Board Boot related definitions have been moved to a
separate file 'tbbr_tools.mk'. The main Makefile will include this
file unless the platform indicates otherwise by setting the variable
'INCLUDE_TBBR_MK := 0' in the corresponding platform.mk file. This
will keep backward compatibility but ideally each platform should
include the corresponding TBB .mk file in platform.mk.
Change-Id: I35e7bc9930d38132412e950e20aa2a01e2b26801
Juan Castillo
committed
on 27 Oct 2015
|
2015-10-21 |
Merge pull request #410 from soby-mathew/sm/psci_handler_reorg
...
Reorganise PSCI PM handler setup on ARM Standard platforms
danh-arm
committed
on 21 Oct 2015
|
Merge pull request #411 from jcastillo-arm/jc/plat_bl1_exit
...
Jc/plat bl1 exit
danh-arm
committed
on 21 Oct 2015
|
2015-10-20 |
Merge pull request #409 from sandrine-bailleux/sb/break-down-bl1-sync-exceptions
...
Break down BL1 AArch64 synchronous exception handler
danh-arm
committed
on 20 Oct 2015
|
Add optional bl1_plat_prepare_exit() API
...
This patch adds an optional API to the platform port:
void bl1_plat_prepare_exit(void);
This function is called prior to exiting BL1 in response to the
RUN_IMAGE_SMC request raised by BL2. It should be used to perform
platform specific clean up or bookkeeping operations before
transferring control to the next image.
A weak empty definition of this function has been provided to
preserve platform backwards compatibility.
Change-Id: Iec09697de5c449ae84601403795cdb6aca166ba1
Juan Castillo
authored
on 5 Oct 2015
Juan Castillo
committed
on 20 Oct 2015
|
Reorganise PSCI PM handler setup on ARM Standard platforms
...
This patch does the following reorganization to psci power management (PM)
handler setup for ARM standard platform ports :
1. The mailbox programming required during `plat_setup_psci_ops()` is identical
for all ARM platforms. Hence the implementation of this API is now moved
to the common `arm_pm.c` file. Each ARM platform now must define the
PLAT_ARM_TRUSTED_MAILBOX_BASE macro, which in current platforms is the same
as ARM_SHARED_RAM_BASE.
2. The PSCI PM handler callback structure, `plat_psci_ops`, must now be
exported via `plat_arm_psci_pm_ops`. This allows the common implementation
of `plat_setup_psci_ops()` to return a platform specific `plat_psci_ops`.
In the case of CSS platforms, a default weak implementation of the same is
provided in `css_pm.c` which can be overridden by each CSS platform.
3. For CSS platforms, the PSCI PM handlers defined in `css_pm.c` are now
made library functions and a new header file `css_pm.h` is added to export
these generic PM handlers. This allows the platform to reuse the
adequate CSS PM handlers and redefine others which need to be customized
when overriding the default `plat_arm_psci_pm_ops` in `css_pm.c`.
Change-Id: I277910f609e023ee5d5ff0129a80ecfce4356ede
Soby Mathew
committed
on 20 Oct 2015
|
2015-10-19 |
Break down BL1 AArch64 synchronous exception handler
...
The AArch64 synchronous exception vector code in BL1 is almost
reaching its architectural limit of 32 instructions. This means
there is very little space for this code to grow.
This patch reduces the size of the exception vector code by
moving most of its code in a function to which we branch from
SynchronousExceptionA64.
Change-Id: Ib35351767a685fb2c2398029d32e54026194f7ed
Sandrine Bailleux
committed
on 19 Oct 2015
|
Merge pull request #408 from sandrine-bailleux/sb/cassert
...
Make CASSERT() macro callable from anywhere
danh-arm
committed
on 19 Oct 2015
|
Merge pull request #407 from sandrine-bailleux/sb/fix-arm-bl1-setup-include
...
Fix #include path in ARM platform BL1 setup code
danh-arm
committed
on 19 Oct 2015
|
Make CASSERT() macro callable from anywhere
...
The CASSERT() macro introduces a typedef for the sole purpose of
triggering a compilation error if the condition to check is false.
This typedef is not used afterwards. As a consequence, when the
CASSERT() macro is called from withing a function block, the compiler
complains and outputs the following error message:
error: typedef 'msg' locally defined but not used [-Werror=unused-local-typedefs]
This patch adds the "unused" attribute for the aforementioned
typedef. This silences the compiler warning and thus makes the
CASSERT() macro callable from within function blocks as well.
Change-Id: Ie36b58fcddae01a21584c48bb6ef43ec85590479
Sandrine Bailleux
committed
on 19 Oct 2015
|
Fix #include path in ARM platform BL1 setup code
...
This patch fixes the relative path to the 'bl1_private.h' header file
included from 'arm_bl1_setup.c'. Note that, although the path was
incorrect, it wasn't causing a compilation error because the header
file still got included through an alternative include search path.
Change-Id: I28e4f3dbe50e3550ca6cad186502c88a9fb5e260
Sandrine Bailleux
committed
on 19 Oct 2015
|
2015-10-14 |
Merge pull request #405 from vwadekar/tlkd-resume-fid-v3
...
TLKD: pass results with TLK_RESUME_FID function ID
danh-arm
committed
on 14 Oct 2015
|
Merge pull request #406 from sandrine-bailleux/sb/cci-init-fix-assertion
...
Fix debug assertion in deprecated CCI-400 driver
danh-arm
committed
on 14 Oct 2015
|
2015-10-12 |
Fix debug assertion in deprecated CCI-400 driver
...
This patch fixes a copy and paste issue that resulted in the cluster
indexes not being checked as intended. Note that this fix applies to
the deprecated CCI-400 driver, not the unified one.
Change-Id: I497132a91c236690e5eaff908f2db5c8c65e85ab
Sandrine Bailleux
committed
on 12 Oct 2015
|
2015-10-09 |
TLKD: pass results with TLK_RESUME_FID function ID
...
TLK sends the "preempted" event to the NS world along with an
identifier for certain use cases. The NS world driver is then
expected to take appropriate action depending on the identifier
value. Upon completion, the NS world driver then sends the
results to TLK (via x1-x3) with the TLK_RESUME_FID function ID.
This patch uses the already present code to pass the results
from the NS world to TLK for the TLK_RESUME_FID function ID.
Signed-off-by: Varun Wadekar <vwadekar@nvidia.com>
Varun Wadekar
committed
on 9 Oct 2015
|
2015-10-07 |
Merge pull request #402 from soby-mathew/sm/psci_cpu_off
...
PSCI: Update state only if CPU_OFF is not denied by SPD
danh-arm
committed
on 7 Oct 2015
|
2015-10-06 |
PSCI: Update state only if CPU_OFF is not denied by SPD
...
This patch fixes an issue in the PSCI framework where the affinity info
state of a core was being set to OFF even when the SPD had denied the
CPU_OFF request. Now, the state remains set to ON instead.
Fixes ARM-software/tf-issues#323
Change-Id: Ia9042aa41fae574eaa07fd2ce3f50cf8cae1b6fc
Soby Mathew
committed
on 6 Oct 2015
|
2015-09-30 |
Merge pull request #401 from sandrine-bailleux/sb/fix-sp804-bug-v2
...
Bug fix in the SP804 dual timer driver
danh-arm
committed
on 30 Sep 2015
|
Merge pull request #400 from vwadekar/tlkd-pm-handlers-v5
...
Send power management events to the Trusted OS (TLK)
Achin Gupta
committed
on 30 Sep 2015
|
Merge pull request #393 from mtk09422/misc-updates
...
mt8173: Update SPM and fix watchdog setting
danh-arm
committed
on 30 Sep 2015
|
Send power management events to the Trusted OS (TLK)
...
This patch adds PM handlers to TLKD for the system suspend/resume and
system poweroff/reset cases. TLK expects all SMCs through a single
handler, which then fork out into multiple handlers depending on the
SMC. We tap into the same single entrypoint by restoring the S-EL1
context before passing the PM event via register 'x0'. On completion
of the PM event, TLK sends a completion SMC and TLKD then moves on
with the PM process.
Signed-off-by: Varun Wadekar <vwadekar@nvidia.com>
Varun Wadekar
committed
on 30 Sep 2015
|
2015-09-28 |
Bug fix in the SP804 dual timer driver
...
The generic delay timer driver expects a pointer to a timer_ops_t
structure containing the specific timer driver information. It
doesn't make a copy of the structure, instead it just keeps the
pointer. Therefore, this pointer must remain valid over time.
The SP804 driver doesn't satisfy this requirement. The
sp804_timer_init() macro creates a temporary instanciation of the
timer_ops_t structure on the fly and passes it to the generic
delay timer. When this temporary instanciation gets deallocated,
the generic delay timer is left with a pointer to invalid data.
This patch fixes this bug by statically allocating the SP804
timer_ops_t structure.
Change-Id: I8fbf75907583aef06701e3fd9fabe0b2c9bc95bf
Sandrine Bailleux
committed
on 28 Sep 2015
|
Merge pull request #398 from achingupta/vk/fix_bakery_lock_size
...
Fix relocation of __PERCPU_BAKERY_LOCK_SIZE__ in PR #390
Achin Gupta
committed
on 28 Sep 2015
|
2015-09-25 |
Fix relocation of __PERCPU_BAKERY_LOCK_SIZE__
...
When a platform port does not define PLAT_PERCPU_BAKERY_LOCK_SIZE, the total
memory that should be allocated per-cpu to accommodate all bakery locks is
calculated by the linker in bl31.ld.S. The linker stores this value in the
__PERCPU_BAKERY_LOCK_SIZE__ linker symbol. The runtime value of this symbol is
different from the link time value as the symbol is relocated into the current
section (.bss). This patch fixes this issue by marking the symbol as ABSOLUTE
which allows it to retain its correct value even at runtime.
The description of PLAT_PERCPU_BAKERY_LOCK_SIZE in the porting-guide.md has been
made clearer as well.
Change-Id: Ia0cfd42f51deaf739d792297e60cad5c6e6e610b
Vikram Kanigiri
authored
on 24 Sep 2015
Achin Gupta
committed
on 25 Sep 2015
|
2015-09-22 |
Merge pull request #394 from achingupta/ag/ccn_driver
...
Support for ARM CoreLink CCN interconnects
Achin Gupta
committed
on 22 Sep 2015
|
2015-09-14 |
Add a generic driver for ARM CCN IP
...
This patch adds a device driver which can be used to program the following
aspects of ARM CCN IP:
1. Specify the mapping between ACE/ACELite/ACELite+DVM/CHI master interfaces and
Request nodes.
2. Add and remove master interfaces from the snoop and dvm
domains.
3. Place the L3 cache in a given power state.
4. Configuring system adress map and enabling 3 SN striping mode of memory
controller operation.
Change-Id: I0f665c6a306938e5b66f6a92f8549b529aa8f325
Achin Gupta
committed
on 14 Sep 2015
|
Tegra: Perform cache maintenance on video carveout memory
...
Currently, the non-overlapping video memory carveout region is cleared after
disabling the MMU at EL3. If at any exception level the carveout region is being
marked as cacheable, this zeroing of memory will not have an affect on the
cached lines. Hence, we first invalidate the dirty lines and update the memory
and invalidate again so that both caches and memory is zeroed out.
Change-Id: If3b2d139ab7227f6799c0911d59e079849dc86aa
Vikram Kanigiri
authored
on 10 Sep 2015
Achin Gupta
committed
on 14 Sep 2015
|
Make generic code work in presence of system caches
...
On the ARMv8 architecture, cache maintenance operations by set/way on the last
level of integrated cache do not affect the system cache. This means that such a
flush or clean operation could result in the data being pushed out to the system
cache rather than main memory. Another CPU could access this data before it
enables its data cache or MMU. Such accesses could be serviced from the main
memory instead of the system cache. If the data in the sysem cache has not yet
been flushed or evicted to main memory then there could be a loss of
coherency. The only mechanism to guarantee that the main memory will be updated
is to use cache maintenance operations to the PoC by MVA(See section D3.4.11
(System level caches) of ARMv8-A Reference Manual (Issue A.g/ARM DDI0487A.G).
This patch removes the reliance of Trusted Firmware on the flush by set/way
operation to ensure visibility of data in the main memory. Cache maintenance
operations by MVA are now used instead. The following are the broad category of
changes:
1. The RW areas of BL2/BL31/BL32 are invalidated by MVA before the C runtime is
initialised. This ensures that any stale cache lines at any level of cache
are removed.
2. Updates to global data in runtime firmware (BL31) by the primary CPU are made
visible to secondary CPUs using a cache clean operation by MVA.
3. Cache maintenance by set/way operations are only used prior to power down.
NOTE: NON-UPSTREAM TRUSTED FIRMWARE CODE SHOULD MAKE EQUIVALENT CHANGES IN
ORDER TO FUNCTION CORRECTLY ON PLATFORMS WITH SUPPORT FOR SYSTEM CACHES.
Fixes ARM-software/tf-issues#205
Change-Id: I64f1b398de0432813a0e0881d70f8337681f6e9a
Achin Gupta
committed
on 14 Sep 2015
|
Merge pull request #390 from vikramkanigiri/at/unify_bakery_locks_v2
...
Re-design bakery lock allocation and algorithm
Achin Gupta
committed
on 14 Sep 2015
|