The FEAT_MTPMU feature disable runs very early after reset. This means,
it needs to be written in assembly, since the C runtime has not been
initialised yet.
However, there is no need for it to be initialised so soon. The PMU
state is only relevant after TF-A has relinquished control. The code
to do this is also very verbose and difficult to read. Delaying the
initialisation allows for it to happen with the rest of the PMU. Align
with FEAT_STATE in the process.
BREAKING CHANGE: This patch explicitly breaks the EL2 entry path. It is
currently unsupported.
Signed-off-by: Boyan Karatotev <boyan.karatotev@arm.com>
Change-Id: I2aa659d026fbdb75152469f6d19812ece3488c6f
The enablement code for the PMU is scattered and difficult to track
down. Factor out the feature into its own lib/extensions folder and
consolidate the implementation. Treat it is as an architecturally
mandatory feature as it is currently.
Additionally, do some cleanup on AArch64. Setting overflow bits in
PMCR_EL0 is irrelevant for firmware so don't do it. Then delay the PMU
initialisation until the context management stage which simplifies the
early environment assembly. One side effect is that the PMU might count
before this happens so reset all counters to 0 to prevent any leakage.
Finally, add an enable to manage_extensions_realm() as realm world uses
the pmu. This introduces the HPMN fixup to realm world.
Signed-off-by: Boyan Karatotev <boyan.karatotev@arm.com>
Change-Id: Ie13a8625820ecc5fbfa467dc6ca18025bf6a9cd3
Make the default value for MTPME always be 1 to preserve the reset
behaviour on newer revisions and on older revisions where the bit is
RES0 it doesn't matter.
Before its introduction MDCR_EL3.MTPME was RES0. Upon its introduction
the field resets to 1, making the MTPMU architecturally "enabled". As
such, the logical action on TF-A's part is to "disable" it, which led to
the introduction of DISABLE_MTPMU.
This hinges on the assumption that MDCR_EL3.MTPME will always be 1
unless the above flag is set. Unfortunately this is not the case, as the
reset value is overwritten at reset with a macro that sets this bit to
0.
Signed-off-by: Boyan Karatotev <boyan.karatotev@arm.com>
Change-Id: Ie570774972f246b3aa41dc016ecbcc6fc2f581f6
manage_extensions_nonsecure() is problematic because it updates both
context and in-place registers (unlike its secure/realm counterparts).
The in-place register updates make it particularly tricky, as those
never change for the lifetime of TF-A. However, they are only set when
exiting to NS world. As such, all of TF-A's execution before that
operates under a different context. This is inconsistent and could cause
problems.
This patch Introduce a real manage_extensions_nonsecure() which only
operates on the context structure. It also introduces a
cm_manage_extensions_el3() which only operates on register in-place that
are not context switched. It is called in BL31's entrypoints so that all
of TF-A executes with the same environment once all features have been
converted.
Signed-off-by: Boyan Karatotev <boyan.karatotev@arm.com>
Change-Id: Ic579f86c41026d2054863ef44893e0ba4c591da9
* changes:
refactor(cpus): add Cortex-A72 errata information
refactor(cpus): convert Rainier to use errata framework
refactor(cpus): convert QEMU Max to use the errata framework
Read data from DeviceTree provided by QEMU, provide via SMC
to the next firmware level.
Signed-off-by: Marcin Juszkiewicz <marcin.juszkiewicz@linaro.org>
Change-Id: I95c5f00ab2cca3b5fda122dcc8d7704a7a82059b
Adding basic CPU library code to support the Hermes CPU.
Change-Id: I61946033fe5fafb56ceb2d14d4c796d85b30457e
Signed-off-by: Govindraj Raja <govindraj.raja@arm.com>
* adds add_erratum_etnry for all described erratas.
* replaces errata_report function with errata_report_shim to report
errata automatically
Change-Id: I7e3315d5cc77b77c328fff7f3988ec588b8f88b9
Signed-off-by: Maksims Svecovs <maksims.svecovs@arm.com>
This involves replacing:
* the reset_func with the standard cpu_reset_func_{start,end} to apply
errata automatically
* the <cpu>_errata_report with the errata_report_shim to report errata
automatically
as well as specifically related to single errata for this CPU:
* the prologue with the workaround_<type>_start to do the checks and
framework registration automatically
* the epilogue with the workaround_<type>_end
* the checker function with the check_erratum_<type> to make it more
descriptive
Change-Id: I31cacbbdd4caa12b32e2c65ec456b0ab6b1a9101
Signed-off-by: Maksims Svecovs <maksims.svecovs@arm.com>
This involves replacing:
* the <cpu>_errata_report with the errata_report_shim to report errata
automatically.
Change-Id: I78b65052dcfc1f29b7dec443bd0aaf67d0efb4eb
Signed-off-by: Maksims Svecovs <maksims.svecovs@arm.com>
As the unit testing project uses the host machine GCC version to
compile, it is marking non-casted references as errors. This patch adds
the proper casting, so it compiles correctly for both Arm platforms and
host machines for unit testing.
Change-Id: Iee96e9117301ba28b6f164aac2cd36dc0f8b6be8
Signed-off-by: Juan Pablo Conde <juanpablo.conde@arm.com>
* changes:
feat(board/rdn2): enable base element RAM RAS support on RD-N2 platform
feat(plat/arm): add memory map entry for CPER memory region
feat(plat/arm): firmware first error handling support for base RAMs
feat(plat/arm): update common platform RAS implementation
feat(plat/sgi): remove RAS setup call from common code
refactor(plat/sgi): deprecate DMC-620 RAS support
fix(plat/common): register PLAT_SP_PRI only if not already registered
fix(plat/sgi): update PLAT_SP_PRI macro definition
fix(plat/arm): add RAS_FFH_SUPPORT check for RAS EHF priority
TC0 is now a deprecated platform so remove it from the list
of supported FVPs as well as throwing an error if it is attempted
to be built.
Signed-off-by: Daniel Boulby <daniel.boulby@arm.com>
Change-Id: Id013fcecbe20700611463ef9eab8cb3ae09071cc
To enable firmware first support for base element RAMs on RD-N2 platform
this patch adds following support
- Includes SDEI header to enable SDEI feature on RD-N2 platform.
- Add TZC configuration for CPER memory region for RD-N2 platform
variants. This region is marked for non-secure access as OSPM and
firmware need to access this region.
- Defines all base element RAM errors for RD-N2 platform variants.
- Defines a platform RAS event map and respective RAS config data
structure.
Signed-off-by: Omkar Anand Kulkarni <omkar.kulkarni@arm.com>
Change-Id: Ideaed598f4924f3b9836d4d7e9ef76b9b7580b48
In firmware-first error handling approach the firmware consumes the
hardware fault interrupt, processes the error and notifies the fault to
OSPM. Firmware also shares the error information with the OSPM using a
standard format called Common Platform Error Record (CPER). The CPER is
placed in reserved memory that is shared between OSPM and the firmware.
On RD-N2 platform variants carve out a reserved memory space for the
CPER buffer. This patch enables CPER memory map region on RD-N2 platform
variants.
Signed-off-by: Omkar Anand Kulkarni <omkar.kulkarni@arm.com>
Change-Id: Ib2645c90d4dc975f57bb143795f61f74f4f81494
RD-N2 platform variants support base element RAM. The RAMs implement
ECC that detects ECC 1/2-bit errors and reports them via interrupts. The
error information is reported as part of error record frames defined for
base element RAMs.
This patch provides reference error handler implementation to handle
1/2-bit RAS errors that occur on base element RAM's. On error event the
error handler reads the error records information and forwards the event
to secure partition. Secure partition creates a CPER record from this
error information. Finally the handler notifies the OS about the RAS
error using the SDEI notification mechanism.
Signed-off-by: Omkar Anand Kulkarni <omkar.kulkarni@arm.com>
Change-Id: Ic209c714de6cd2d4c845198b03724940a2e1c240
Refactor the RAS implementation to be used as common platform RAS
implementation for all the platforms. As part of refactoring this patch
extends support to configure interrupt as PPI interrupt type in addition
to currently supported SPI interrupts.
This patch defines a RAS config data structure to be defined by each
platform. The RAS config data structure carries the event map and size
information. Each platform code during initialization phase must define
this RAS config and register it with common platform RAS module.
Signed-off-by: Omkar Anand Kulkarni <omkar.kulkarni@arm.com>
Change-Id: I4019b31386a7e9c197bcc83bdca47876ee854d0f
In preparation of refactoring the support for platform error handling,
remove the call to RAS platform setup call from SGI specific common
code. This function will be called from platform code after the
refactoring.
Signed-off-by: Omkar Anand Kulkarni <omkar.kulkarni@arm.com>
Change-Id: If4a87e0adf166b1c99bf5999f2f89efa6c7c6afc
Remove DMC-620 specific code from platform RAS implementation. DMC-620
RAS support is not supported on SGI and RD platforms. The rest of the
platform specific code maintained will be reused for supporting RAS
error handling on RD-N2 and later platforms.
Signed-off-by: Omkar Anand Kulkarni <omkar.kulkarni@arm.com>
Change-Id: Ic03ae0e3298628330c5f7c25bafb0131f7b9d5b6
Build fails when RAS and SPM are enabled together and when PLAT_SP_PRI
EHF priority is equal to PLAT_RAS_PRI EHF priority.
So add checks to register SPM priority with the EHF framework only when
the priority is different from RAS priority or when RAS is not enabled
on the platform.
Signed-off-by: Omkar Anand Kulkarni <omkar.kulkarni@arm.com>
Change-Id: Ie14f82d27c9835b24890cc4561a56821881cf0ec
PLAT_SP_PRI EHF priority is defined to be same as the PLAT_RAS_PRI EHF
priority. But PLAT_RAS_PRIORITY is defined only if RAS_FFH_SUPPORT is
enabled. This patch defines priority value for PLAT_SP_PRI if
RAS_FFH_SUPPORT is not enabled.
Signed-off-by: Omkar Anand Kulkarni <omkar.kulkarni@arm.com>
Change-Id: Ib3747317d2ecc088fbbf1f5f283726a330454c93
For TF-A, there is no format specified for functional documentation.
For AMD-Xilinx platforms, following kernel-doc format for the functional
documentation to make sure AMD-xilinx documentation is align with
actual code.
For example use kernel-doc from linux to call:
<linux>/scripts/kernel-doc -man -v 1 >/dev/null file...
Signed-off-by: Prasad Kummari <prasad.kummari@amd.com>
Change-Id: Idcc9def408b6c8da35b36f67ef82fc00890e998c
Both march32-directive and march64-directive eventually generate the
same march option that will passed to compiler.
Merge this two separate directives to a common one as march-directive.
Change-Id: I220d2b782eb3b54e13ffd5b6a581d0e6da68756a
Signed-off-by: Govindraj Raja <govindraj.raja@arm.com>
Previous implementation used common CSS interrupts, which do not match
the Morello platform interrupt map. Updated to configure Secure
interrupts according to the Morello TRM and InfraSYSDESIGN4.0
specification.
Signed-off-by: Werner Lewis <werner.lewis@arm.com>
Change-Id: I783a472d92601d86f1844f0d035dd0d036b2bfca
* changes:
docs(msm8916): document new build options
feat(msm8916): allow selecting which UART to use
feat(msm8916): add SP_MIN port for AArch32
refactor(msm8916): detect cold boot in plat_get_my_entrypoint
feat(msm8916): add Test Secure Payload (TSP) port
build(msm8916): place bl32 directly after bl31
refactor(msm8916): separate common platform setup code
* changes:
fix(tsp): fix destination ID in direct request
fix(el3-spm): fix LSP direct message response
fix(el3-spm): improve direct messaging validation
Moved the Juno-specific build option from the common build option
file to the Arm build option file.
Change-Id: I0f53203f0cfca4a3baadab2cee4339c9694cfe8b
Signed-off-by: Manish V Badarkhe <Manish.Badarkhe@arm.com>
Ensure the TSP ID is set as the source ID in a direct request.
Signed-off-by: Marc Bonnici <marc.bonnici@arm.com>
Change-Id: Ia082fe5a1da6f2994072ec70c6ba818212a52f20
Ensure that the example LSP correctly sets the
sender/receiver field in a direct response.
Signed-off-by: Marc Bonnici <marc.bonnici@arm.com>
Change-Id: I482c08d4657617adb00b0f3cf3c8ddc84f1bf7c8
Perform additional validation of the source and destination
IDs of direct messages.
Additionally track the sender of a direct request to allow
validating the target of the corresponding direct response.
Signed-off-by: Marc Bonnici <marc.bonnici@arm.com>
Change-Id: I8d39d53a02b8333246f1500c79ba04f149459c16
In some build configurations TF-A can exceed the existing 256KB SRAM,
triggering a build failure. More recent versions of the base FVP allow
you to configure a larger Trusted SRAM of 512KB.
This change introduces the `FVP_TRUSTED_SRAM_SIZE` build option, which
allows you to explicitly specify how much of the Trusted SRAM to
utilise, e.g.:
FVP_TRUSTED_SRAM_SIZE=384
This allows previously-failing configurations to build successfully by
utilising more than the originally-allocated 256KB of the Trusted SRAM
while maintaining compatibility with older configurations/models that
only require/have 256KB.
Change-Id: I8344d3718564cd2bd53f1e6860e2fe341ae240b0
Signed-off-by: Chris Kay <chris.kay@arm.com>
Update the MSM8916 platform documentation with the new build options
introduced in the previous changes:
- AArch32 (BL32/SP_MIN)
- UART selection
While at it, also document the build options that allow changing the
memory addresses (PRELOADED_BL33_BASE, BL31_BASE, BL32_BASE).
Change-Id: I2370c8264982317693f69fda0b03a255f12bafe2
Signed-off-by: Stephan Gerhold <stephan@gerhold.net>
At the moment the msm8916 platform port always uses UART number 2 for
debug output. In some situations it is necessary to change this, either
because only the other UART is exposed on the board or for runtime
debugging, to avoid conflicting with the normal world.
Make the UART to use configurable using QTI_UART_NUM on the make
command line and also add QTI_RUNTIME_UART as an option to keep using
the UART after early boot. The latter is disabled by default since it
requires reserving the UART and related clocks inside the normal world.
Change-Id: I14725f954bbcecebcf317e8601922a3d00f2ec28
Signed-off-by: Stephan Gerhold <stephan@gerhold.net>
Use the new shared msm8916 setup code to allow compiling the minimal
AArch32 Secure Payload (SP_MIN) as simple PSCI implementation.
AArch64 is preferred for the Cortex-A53 cores in MSM8916 but there are
some similar platforms with AArch32-only Cortex-A7 cores that can
benefit from this in future changes.
The AArch32 assembly implementation for msm8916_helpers.S and
uartdm_console.S is a direct port of the AArch64 implementation.
Only plat_get_my_entrypoint is slightly different because there is no
need to handle the "boot remapper" on cold boot for AArch32.
Change-Id: Idf160e86fb3e685fcedec3e051400e6273997b74
Signed-off-by: Stephan Gerhold <stephan@gerhold.net>
The msm8916 platform port needs to disable the TCM redirect to the L2
cache as early as possible during cold boot to avoid crashes. Right now
this is done in plat_reset_handler by checking if BL31 was started
through the "boot remapper", which redirects memory accesses around the
fixed CPU reset address (0x0) to the actual link address of BL31. On
AArch64 this is always the case during cold boot, since a CPU reset was
necessary to switch from AArch32 in the initial bootloader to AArch64.
On AArch32, SP_MIN starts running at the real link address immediately,
so the initial cold boot must be detected with a different approach.
To keep the AArch32 and AArch64 implementation of this functionality
consistent, move this functionality to plat_get_my_entrypoint, by
checking if the msm8916_entry_point is still zero or was already
updated for later warm boots by the PSCI code.
Also, avoid entering BL31 twice and instead add the BL31_BASE offset
to the return address in the link register. This allows preserving the
bootloader arguments in x0-x3 because they otherwise get lost.
Change-Id: I90286c6cacf23f44ed7930a3e7e33804ca63c391
Signed-off-by: Stephan Gerhold <stephan@gerhold.net>
