xref: /rk3399_ARM-atf/docs/porting-guide.rst (revision 1e8b5354574ac389bb3d29fdfcb9631cc8108ccb)
1Porting Guide
2=============
3
4Introduction
5------------
6
7Porting Trusted Firmware-A (TF-A) to a new platform involves making some
8mandatory and optional modifications for both the cold and warm boot paths.
9Modifications consist of:
10
11-  Implementing a platform-specific function or variable,
12-  Setting up the execution context in a certain way, or
13-  Defining certain constants (for example #defines).
14
15The platform-specific functions and variables are declared in
16``include/plat/common/platform.h``. The firmware provides a default
17implementation of variables and functions to fulfill the optional requirements
18in order to ease the porting effort. Each platform port can use them as is or
19provide their own implementation if the default implementation is inadequate.
20
21   .. note::
22
23      TF-A historically provided default implementations of platform interfaces
24      as *weak* functions. This practice is now discouraged and new platform
25      interfaces as they get introduced in the code base should be *strongly*
26      defined. We intend to convert existing weak functions over time. Until
27      then, you will find references to *weak* functions in this document.
28
29Please review the :ref:`Threat Model` documents as part of the porting
30effort. Some platform interfaces play a key role in mitigating against some of
31the threats. Failing to fulfill these expectations could undermine the security
32guarantees offered by TF-A. These platform responsibilities are highlighted in
33the threat assessment section, under the "`Mitigations implemented?`" box for
34each threat.
35
36Some modifications are common to all Boot Loader (BL) stages. Section 2
37discusses these in detail. The subsequent sections discuss the remaining
38modifications for each BL stage in detail.
39
40Please refer to the :ref:`Platform Ports Policy` for the policy regarding
41compatibility and deprecation of these porting interfaces.
42
43Only Arm development platforms (such as FVP and Juno) may use the
44functions/definitions in ``include/plat/arm/common/`` and the corresponding
45source files in ``plat/arm/common/``. This is done so that there are no
46dependencies between platforms maintained by different people/companies. If you
47want to use any of the functionality present in ``plat/arm`` files, please
48propose a patch that moves the code to ``plat/common`` so that it can be
49discussed.
50
51Common modifications
52--------------------
53
54This section covers the modifications that should be made by the platform for
55each BL stage to correctly port the firmware stack. They are categorized as
56either mandatory or optional.
57
58Common mandatory modifications
59------------------------------
60
61A platform port must enable the Memory Management Unit (MMU) as well as the
62instruction and data caches for each BL stage. Setting up the translation
63tables is the responsibility of the platform port because memory maps differ
64across platforms. A memory translation library (see ``lib/xlat_tables_v2/``) is
65provided to help in this setup.
66
67Note that although this library supports non-identity mappings, this is intended
68only for re-mapping peripheral physical addresses and allows platforms with high
69I/O addresses to reduce their virtual address space. All other addresses
70corresponding to code and data must currently use an identity mapping.
71
72Also, the only translation granule size supported in TF-A is 4KB, as various
73parts of the code assume that is the case. It is not possible to switch to
7416 KB or 64 KB granule sizes at the moment.
75
76In Arm standard platforms, each BL stage configures the MMU in the
77platform-specific architecture setup function, ``blX_plat_arch_setup()``, and uses
78an identity mapping for all addresses.
79
80If the build option ``USE_COHERENT_MEM`` is enabled, each platform can allocate a
81block of identity mapped secure memory with Device-nGnRE attributes aligned to
82page boundary (4K) for each BL stage. All sections which allocate coherent
83memory are grouped under ``.coherent_ram``. For ex: Bakery locks are placed in a
84section identified by name ``.bakery_lock`` inside ``.coherent_ram`` so that its
85possible for the firmware to place variables in it using the following C code
86directive:
87
88::
89
90    __section(".bakery_lock")
91
92Or alternatively the following assembler code directive:
93
94::
95
96    .section .bakery_lock
97
98The ``.coherent_ram`` section is a sum of all sections like ``.bakery_lock`` which are
99used to allocate any data structures that are accessed both when a CPU is
100executing with its MMU and caches enabled, and when it's running with its MMU
101and caches disabled. Examples are given below.
102
103The following variables, functions and constants must be defined by the platform
104for the firmware to work correctly.
105
106.. _platform_def_mandatory:
107
108File : platform_def.h [mandatory]
109~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
110
111Each platform must ensure that a header file of this name is in the system
112include path with the following constants defined. This will require updating
113the list of ``PLAT_INCLUDES`` in the ``platform.mk`` file.
114
115Platform ports may optionally use the file ``include/plat/common/common_def.h``,
116which provides typical values for some of the constants below. These values are
117likely to be suitable for all platform ports.
118
119-  **#define : PLATFORM_LINKER_FORMAT**
120
121   Defines the linker format used by the platform, for example
122   ``elf64-littleaarch64``.
123
124-  **#define : PLATFORM_LINKER_ARCH**
125
126   Defines the processor architecture for the linker by the platform, for
127   example ``aarch64``.
128
129-  **#define : PLATFORM_STACK_SIZE**
130
131   Defines the normal stack memory available to each CPU. This constant is used
132   by ``plat/common/aarch64/platform_mp_stack.S`` and
133   ``plat/common/aarch64/platform_up_stack.S``.
134
135-  **#define : CACHE_WRITEBACK_GRANULE**
136
137   Defines the size in bytes of the largest cache line across all the cache
138   levels in the platform.
139
140-  **#define : FIRMWARE_WELCOME_STR**
141
142   Defines the character string printed by BL1 upon entry into the ``bl1_main()``
143   function.
144
145-  **#define : PLATFORM_CORE_COUNT**
146
147   Defines the total number of CPUs implemented by the platform across all
148   clusters in the system.
149
150-  **#define : PLAT_NUM_PWR_DOMAINS**
151
152   Defines the total number of nodes in the power domain topology
153   tree at all the power domain levels used by the platform.
154   This macro is used by the PSCI implementation to allocate
155   data structures to represent power domain topology.
156
157-  **#define : PLAT_MAX_PWR_LVL**
158
159   Defines the maximum power domain level that the power management operations
160   should apply to. More often, but not always, the power domain level
161   corresponds to affinity level. This macro allows the PSCI implementation
162   to know the highest power domain level that it should consider for power
163   management operations in the system that the platform implements. For
164   example, the Base AEM FVP implements two clusters with a configurable
165   number of CPUs and it reports the maximum power domain level as 1.
166
167-  **#define : PLAT_MAX_OFF_STATE**
168
169   Defines the local power state corresponding to the deepest power down
170   possible at every power domain level in the platform. The local power
171   states for each level may be sparsely allocated between 0 and this value
172   with 0 being reserved for the RUN state. The PSCI implementation uses this
173   value to initialize the local power states of the power domain nodes and
174   to specify the requested power state for a PSCI_CPU_OFF call.
175
176-  **#define : PLAT_MAX_RET_STATE**
177
178   Defines the local power state corresponding to the deepest retention state
179   possible at every power domain level in the platform. This macro should be
180   a value less than PLAT_MAX_OFF_STATE and greater than 0. It is used by the
181   PSCI implementation to distinguish between retention and power down local
182   power states within PSCI_CPU_SUSPEND call.
183
184-  **#define : PLAT_MAX_PWR_LVL_STATES**
185
186   Defines the maximum number of local power states per power domain level
187   that the platform supports. The default value of this macro is 2 since
188   most platforms just support a maximum of two local power states at each
189   power domain level (power-down and retention). If the platform needs to
190   account for more local power states, then it must redefine this macro.
191
192   Currently, this macro is used by the Generic PSCI implementation to size
193   the array used for PSCI_STAT_COUNT/RESIDENCY accounting.
194
195-  **#define : BL1_RO_BASE**
196
197   Defines the base address in secure ROM where BL1 originally lives. Must be
198   aligned on a page-size boundary.
199
200-  **#define : BL1_RO_LIMIT**
201
202   Defines the maximum address in secure ROM that BL1's actual content (i.e.
203   excluding any data section allocated at runtime) can occupy.
204
205-  **#define : BL1_RW_BASE**
206
207   Defines the base address in secure RAM where BL1's read-write data will live
208   at runtime. Must be aligned on a page-size boundary.
209
210-  **#define : BL1_RW_LIMIT**
211
212   Defines the maximum address in secure RAM that BL1's read-write data can
213   occupy at runtime.
214
215-  **#define : BL2_BASE**
216
217   Defines the base address in secure RAM where BL1 loads the BL2 binary image.
218   Must be aligned on a page-size boundary. This constant is not applicable
219   when BL2_IN_XIP_MEM is set to '1'.
220
221-  **#define : BL2_LIMIT**
222
223   Defines the maximum address in secure RAM that the BL2 image can occupy.
224   This constant is not applicable when BL2_IN_XIP_MEM is set to '1'.
225
226-  **#define : BL2_RO_BASE**
227
228   Defines the base address in secure XIP memory where BL2 RO section originally
229   lives. Must be aligned on a page-size boundary. This constant is only needed
230   when BL2_IN_XIP_MEM is set to '1'.
231
232-  **#define : BL2_RO_LIMIT**
233
234   Defines the maximum address in secure XIP memory that BL2's actual content
235   (i.e. excluding any data section allocated at runtime) can occupy. This
236   constant is only needed when BL2_IN_XIP_MEM is set to '1'.
237
238-  **#define : BL2_RW_BASE**
239
240   Defines the base address in secure RAM where BL2's read-write data will live
241   at runtime. Must be aligned on a page-size boundary. This constant is only
242   needed when BL2_IN_XIP_MEM is set to '1'.
243
244-  **#define : BL2_RW_LIMIT**
245
246   Defines the maximum address in secure RAM that BL2's read-write data can
247   occupy at runtime. This constant is only needed when BL2_IN_XIP_MEM is set
248   to '1'.
249
250-  **#define : BL31_BASE**
251
252   Defines the base address in secure RAM where BL2 loads the BL31 binary
253   image. Must be aligned on a page-size boundary.
254
255-  **#define : BL31_LIMIT**
256
257   Defines the maximum address in secure RAM that the BL31 image can occupy.
258
259-  **#define : PLAT_RSE_COMMS_PAYLOAD_MAX_SIZE**
260
261   Defines the maximum message size between AP and RSE. Need to define if
262   platform supports RSE.
263
264For every image, the platform must define individual identifiers that will be
265used by BL1 or BL2 to load the corresponding image into memory from non-volatile
266storage. For the sake of performance, integer numbers will be used as
267identifiers. The platform will use those identifiers to return the relevant
268information about the image to be loaded (file handler, load address,
269authentication information, etc.). The following image identifiers are
270mandatory:
271
272-  **#define : BL2_IMAGE_ID**
273
274   BL2 image identifier, used by BL1 to load BL2.
275
276-  **#define : BL31_IMAGE_ID**
277
278   BL31 image identifier, used by BL2 to load BL31.
279
280-  **#define : BL33_IMAGE_ID**
281
282   BL33 image identifier, used by BL2 to load BL33.
283
284If Trusted Board Boot is enabled, the following certificate identifiers must
285also be defined:
286
287-  **#define : TRUSTED_BOOT_FW_CERT_ID**
288
289   BL2 content certificate identifier, used by BL1 to load the BL2 content
290   certificate.
291
292-  **#define : TRUSTED_KEY_CERT_ID**
293
294   Trusted key certificate identifier, used by BL2 to load the trusted key
295   certificate.
296
297-  **#define : SOC_FW_KEY_CERT_ID**
298
299   BL31 key certificate identifier, used by BL2 to load the BL31 key
300   certificate.
301
302-  **#define : SOC_FW_CONTENT_CERT_ID**
303
304   BL31 content certificate identifier, used by BL2 to load the BL31 content
305   certificate.
306
307-  **#define : NON_TRUSTED_FW_KEY_CERT_ID**
308
309   BL33 key certificate identifier, used by BL2 to load the BL33 key
310   certificate.
311
312-  **#define : NON_TRUSTED_FW_CONTENT_CERT_ID**
313
314   BL33 content certificate identifier, used by BL2 to load the BL33 content
315   certificate.
316
317-  **#define : FWU_CERT_ID**
318
319   Firmware Update (FWU) certificate identifier, used by NS_BL1U to load the
320   FWU content certificate.
321
322If the AP Firmware Updater Configuration image, BL2U is used, the following
323must also be defined:
324
325-  **#define : BL2U_BASE**
326
327   Defines the base address in secure memory where BL1 copies the BL2U binary
328   image. Must be aligned on a page-size boundary.
329
330-  **#define : BL2U_LIMIT**
331
332   Defines the maximum address in secure memory that the BL2U image can occupy.
333
334-  **#define : BL2U_IMAGE_ID**
335
336   BL2U image identifier, used by BL1 to fetch an image descriptor
337   corresponding to BL2U.
338
339If the SCP Firmware Update Configuration Image, SCP_BL2U is used, the following
340must also be defined:
341
342-  **#define : SCP_BL2U_IMAGE_ID**
343
344   SCP_BL2U image identifier, used by BL1 to fetch an image descriptor
345   corresponding to SCP_BL2U.
346
347   .. note::
348      TF-A does not provide source code for this image.
349
350If the Non-Secure Firmware Updater ROM, NS_BL1U is used, the following must
351also be defined:
352
353-  **#define : NS_BL1U_BASE**
354
355   Defines the base address in non-secure ROM where NS_BL1U executes.
356   Must be aligned on a page-size boundary.
357
358   .. note::
359      TF-A does not provide source code for this image.
360
361-  **#define : NS_BL1U_IMAGE_ID**
362
363   NS_BL1U image identifier, used by BL1 to fetch an image descriptor
364   corresponding to NS_BL1U.
365
366If the Non-Secure Firmware Updater, NS_BL2U is used, the following must also
367be defined:
368
369-  **#define : NS_BL2U_BASE**
370
371   Defines the base address in non-secure memory where NS_BL2U executes.
372   Must be aligned on a page-size boundary.
373
374   .. note::
375      TF-A does not provide source code for this image.
376
377-  **#define : NS_BL2U_IMAGE_ID**
378
379   NS_BL2U image identifier, used by BL1 to fetch an image descriptor
380   corresponding to NS_BL2U.
381
382For the the Firmware update capability of TRUSTED BOARD BOOT, the following
383macros may also be defined:
384
385-  **#define : PLAT_FWU_MAX_SIMULTANEOUS_IMAGES**
386
387   Total number of images that can be loaded simultaneously. If the platform
388   doesn't specify any value, it defaults to 10.
389
390If a SCP_BL2 image is supported by the platform, the following constants must
391also be defined:
392
393-  **#define : SCP_BL2_IMAGE_ID**
394
395   SCP_BL2 image identifier, used by BL2 to load SCP_BL2 into secure memory
396   from platform storage before being transferred to the SCP.
397
398-  **#define : SCP_FW_KEY_CERT_ID**
399
400   SCP_BL2 key certificate identifier, used by BL2 to load the SCP_BL2 key
401   certificate (mandatory when Trusted Board Boot is enabled).
402
403-  **#define : SCP_FW_CONTENT_CERT_ID**
404
405   SCP_BL2 content certificate identifier, used by BL2 to load the SCP_BL2
406   content certificate (mandatory when Trusted Board Boot is enabled).
407
408If a BL32 image is supported by the platform, the following constants must
409also be defined:
410
411-  **#define : BL32_IMAGE_ID**
412
413   BL32 image identifier, used by BL2 to load BL32.
414
415-  **#define : TRUSTED_OS_FW_KEY_CERT_ID**
416
417   BL32 key certificate identifier, used by BL2 to load the BL32 key
418   certificate (mandatory when Trusted Board Boot is enabled).
419
420-  **#define : TRUSTED_OS_FW_CONTENT_CERT_ID**
421
422   BL32 content certificate identifier, used by BL2 to load the BL32 content
423   certificate (mandatory when Trusted Board Boot is enabled).
424
425-  **#define : BL32_BASE**
426
427   Defines the base address in secure memory where BL2 loads the BL32 binary
428   image. Must be aligned on a page-size boundary.
429
430-  **#define : BL32_LIMIT**
431
432   Defines the maximum address that the BL32 image can occupy.
433
434If the Test Secure-EL1 Payload (TSP) instantiation of BL32 is supported by the
435platform, the following constants must also be defined:
436
437-  **#define : TSP_SEC_MEM_BASE**
438
439   Defines the base address of the secure memory used by the TSP image on the
440   platform. This must be at the same address or below ``BL32_BASE``.
441
442-  **#define : TSP_SEC_MEM_SIZE**
443
444   Defines the size of the secure memory used by the BL32 image on the
445   platform. ``TSP_SEC_MEM_BASE`` and ``TSP_SEC_MEM_SIZE`` must fully
446   accommodate the memory required by the BL32 image, defined by ``BL32_BASE``
447   and ``BL32_LIMIT``.
448
449-  **#define : TSP_IRQ_SEC_PHY_TIMER**
450
451   Defines the ID of the secure physical generic timer interrupt used by the
452   TSP's interrupt handling code.
453
454If the platform port uses the translation table library code, the following
455constants must also be defined:
456
457-  **#define : PLAT_XLAT_TABLES_DYNAMIC**
458
459   Optional flag that can be set per-image to enable the dynamic allocation of
460   regions even when the MMU is enabled. If not defined, only static
461   functionality will be available, if defined and set to 1 it will also
462   include the dynamic functionality.
463
464-  **#define : MAX_XLAT_TABLES**
465
466   Defines the maximum number of translation tables that are allocated by the
467   translation table library code. To minimize the amount of runtime memory
468   used, choose the smallest value needed to map the required virtual addresses
469   for each BL stage. If ``PLAT_XLAT_TABLES_DYNAMIC`` flag is enabled for a BL
470   image, ``MAX_XLAT_TABLES`` must be defined to accommodate the dynamic regions
471   as well.
472
473-  **#define : MAX_MMAP_REGIONS**
474
475   Defines the maximum number of regions that are allocated by the translation
476   table library code. A region consists of physical base address, virtual base
477   address, size and attributes (Device/Memory, RO/RW, Secure/Non-Secure), as
478   defined in the ``mmap_region_t`` structure. The platform defines the regions
479   that should be mapped. Then, the translation table library will create the
480   corresponding tables and descriptors at runtime. To minimize the amount of
481   runtime memory used, choose the smallest value needed to register the
482   required regions for each BL stage. If ``PLAT_XLAT_TABLES_DYNAMIC`` flag is
483   enabled for a BL image, ``MAX_MMAP_REGIONS`` must be defined to accommodate
484   the dynamic regions as well.
485
486-  **#define : PLAT_VIRT_ADDR_SPACE_SIZE**
487
488   Defines the total size of the virtual address space in bytes. For example,
489   for a 32 bit virtual address space, this value should be ``(1ULL << 32)``.
490
491-  **#define : PLAT_PHY_ADDR_SPACE_SIZE**
492
493   Defines the total size of the physical address space in bytes. For example,
494   for a 32 bit physical address space, this value should be ``(1ULL << 32)``.
495
496If the platform port uses the IO storage framework, the following constants
497must also be defined:
498
499-  **#define : MAX_IO_DEVICES**
500
501   Defines the maximum number of registered IO devices. Attempting to register
502   more devices than this value using ``io_register_device()`` will fail with
503   -ENOMEM.
504
505-  **#define : MAX_IO_HANDLES**
506
507   Defines the maximum number of open IO handles. Attempting to open more IO
508   entities than this value using ``io_open()`` will fail with -ENOMEM.
509
510-  **#define : MAX_IO_BLOCK_DEVICES**
511
512   Defines the maximum number of registered IO block devices. Attempting to
513   register more devices this value using ``io_dev_open()`` will fail
514   with -ENOMEM. MAX_IO_BLOCK_DEVICES should be less than MAX_IO_DEVICES.
515   With this macro, multiple block devices could be supported at the same
516   time.
517
518If the platform needs to allocate data within the per-cpu data framework in
519BL31, it should define the following macro. Currently this is only required if
520the platform decides not to use the coherent memory section by undefining the
521``USE_COHERENT_MEM`` build flag. In this case, the framework allocates the
522required memory within the the per-cpu data to minimize wastage.
523
524-  **#define : PLAT_PCPU_DATA_SIZE**
525
526   Defines the memory (in bytes) to be reserved within the per-cpu data
527   structure for use by the platform layer.
528
529The following constants are optional. They should be defined when the platform
530memory layout implies some image overlaying like in Arm standard platforms.
531
532-  **#define : BL31_PROGBITS_LIMIT**
533
534   Defines the maximum address in secure RAM that the BL31's progbits sections
535   can occupy.
536
537-  **#define : TSP_PROGBITS_LIMIT**
538
539   Defines the maximum address that the TSP's progbits sections can occupy.
540
541If the platform supports OS-initiated mode, i.e. the build option
542``PSCI_OS_INIT_MODE`` is enabled, and if the platform's maximum power domain
543level for PSCI_CPU_SUSPEND differs from ``PLAT_MAX_PWR_LVL``, the following
544constant must be defined.
545
546-  **#define : PLAT_MAX_CPU_SUSPEND_PWR_LVL**
547
548   Defines the maximum power domain level that PSCI_CPU_SUSPEND should apply to.
549
550If the platform port uses the PL061 GPIO driver, the following constant may
551optionally be defined:
552
553-  **PLAT_PL061_MAX_GPIOS**
554   Maximum number of GPIOs required by the platform. This allows control how
555   much memory is allocated for PL061 GPIO controllers. The default value is
556
557   #. $(eval $(call add_define,PLAT_PL061_MAX_GPIOS))
558
559If the platform port uses the partition driver, the following constant may
560optionally be defined:
561
562-  **PLAT_PARTITION_MAX_ENTRIES**
563   Maximum number of partition entries required by the platform. This allows
564   control how much memory is allocated for partition entries. The default
565   value is 128.
566   For example, define the build flag in ``platform.mk``:
567   PLAT_PARTITION_MAX_ENTRIES := 12
568   $(eval $(call add_define,PLAT_PARTITION_MAX_ENTRIES))
569
570-  **PLAT_PARTITION_BLOCK_SIZE**
571   The size of partition block. It could be either 512 bytes or 4096 bytes.
572   The default value is 512.
573   For example, define the build flag in ``platform.mk``:
574   PLAT_PARTITION_BLOCK_SIZE := 4096
575   $(eval $(call add_define,PLAT_PARTITION_BLOCK_SIZE))
576
577If the platform port uses the Arm® Ethos™-N NPU driver, the following
578configuration must be performed:
579
580- The NPU SiP service handler must be hooked up. This consists of both the
581  initial setup (``ethosn_smc_setup``) and the handler itself
582  (``ethosn_smc_handler``)
583
584If the platform port uses the Arm® Ethos™-N NPU driver with TZMP1 support
585enabled, the following constants and configuration must also be defined:
586
587- **ETHOSN_NPU_PROT_FW_NSAID**
588
589  Defines the Non-secure Access IDentity (NSAID) that the NPU shall use to
590  access the protected memory that contains the NPU's firmware.
591
592- **ETHOSN_NPU_PROT_DATA_RW_NSAID**
593
594  Defines the Non-secure Access IDentity (NSAID) that the NPU shall use for
595  read/write access to the protected memory that contains inference data.
596
597- **ETHOSN_NPU_PROT_DATA_RO_NSAID**
598
599  Defines the Non-secure Access IDentity (NSAID) that the NPU shall use for
600  read-only access to the protected memory that contains inference data.
601
602- **ETHOSN_NPU_NS_RW_DATA_NSAID**
603
604  Defines the Non-secure Access IDentity (NSAID) that the NPU shall use for
605  read/write access to the non-protected memory.
606
607- **ETHOSN_NPU_NS_RO_DATA_NSAID**
608
609  Defines the Non-secure Access IDentity (NSAID) that the NPU shall use for
610  read-only access to the non-protected memory.
611
612- **ETHOSN_NPU_FW_IMAGE_BASE** and **ETHOSN_NPU_FW_IMAGE_LIMIT**
613
614  Defines the physical address range that the NPU's firmware will be loaded
615  into and executed from.
616
617- Configure the platforms TrustZone Controller (TZC) with appropriate regions
618  of protected memory. At minimum this must include a region for the NPU's
619  firmware code and a region for protected inference data, and these must be
620  accessible using the NSAIDs defined above.
621
622- Include the NPU firmware and certificates in the FIP.
623
624- Provide FCONF entries to configure the image source for the NPU firmware
625  and certificates.
626
627- Add MMU mappings such that:
628
629 - BL2 can write the NPU firmware into the region defined by
630   ``ETHOSN_NPU_FW_IMAGE_BASE`` and ``ETHOSN_NPU_FW_IMAGE_LIMIT``
631 - BL31 (SiP service) can read the NPU firmware from the same region
632
633- Add the firmware image ID ``ETHOSN_NPU_FW_IMAGE_ID`` to the list of images
634  loaded by BL2.
635
636Please see the reference implementation code for the Juno platform as an example.
637
638
639The following constant is optional. It should be defined to override the default
640behaviour of the ``assert()`` function (for example, to save memory).
641
642-  **PLAT_LOG_LEVEL_ASSERT**
643   If ``PLAT_LOG_LEVEL_ASSERT`` is higher or equal than ``LOG_LEVEL_VERBOSE``,
644   ``assert()`` prints the name of the file, the line number and the asserted
645   expression. Else if it is higher than ``LOG_LEVEL_INFO``, it prints the file
646   name and the line number. Else if it is lower than ``LOG_LEVEL_INFO``, it
647   doesn't print anything to the console. If ``PLAT_LOG_LEVEL_ASSERT`` isn't
648   defined, it defaults to ``LOG_LEVEL``.
649
650If the platform port uses the DRTM feature, the following constants must be
651defined:
652
653-  **#define : PLAT_DRTM_EVENT_LOG_MAX_SIZE**
654
655   Maximum Event Log size used by the platform. Platform can decide the maximum
656   size of the Event Log buffer, depending upon the highest hash algorithm
657   chosen and the number of components selected to measure during the DRTM
658   execution flow.
659
660-  **#define : PLAT_DRTM_MMAP_ENTRIES**
661
662   Number of the MMAP entries used by the DRTM implementation to calculate the
663   size of address map region of the platform.
664
665File : plat_macros.S [mandatory]
666~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
667
668Each platform must ensure a file of this name is in the system include path with
669the following macro defined. In the Arm development platforms, this file is
670found in ``plat/arm/board/<plat_name>/include/plat_macros.S``.
671
672-  **Macro : plat_crash_print_regs**
673
674   This macro allows the crash reporting routine to print relevant platform
675   registers in case of an unhandled exception in BL31. This aids in debugging
676   and this macro can be defined to be empty in case register reporting is not
677   desired.
678
679   For instance, GIC or interconnect registers may be helpful for
680   troubleshooting.
681
682Handling Reset
683--------------
684
685BL1 by default implements the reset vector where execution starts from a cold
686or warm boot. BL31 can be optionally set as a reset vector using the
687``RESET_TO_BL31`` make variable.
688
689For each CPU, the reset vector code is responsible for the following tasks:
690
691#. Distinguishing between a cold boot and a warm boot.
692
693#. In the case of a cold boot and the CPU being a secondary CPU, ensuring that
694   the CPU is placed in a platform-specific state until the primary CPU
695   performs the necessary steps to remove it from this state.
696
697#. In the case of a warm boot, ensuring that the CPU jumps to a platform-
698   specific address in the BL31 image in the same processor mode as it was
699   when released from reset.
700
701The following functions need to be implemented by the platform port to enable
702reset vector code to perform the above tasks.
703
704Function : plat_get_my_entrypoint() [mandatory when PROGRAMMABLE_RESET_ADDRESS == 0]
705~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
706
707::
708
709    Argument : void
710    Return   : uintptr_t
711
712This function is called with the MMU and caches disabled
713(``SCTLR_EL3.M`` = 0 and ``SCTLR_EL3.C`` = 0). The function is responsible for
714distinguishing between a warm and cold reset for the current CPU using
715platform-specific means. If it's a warm reset, then it returns the warm
716reset entrypoint point provided to ``plat_setup_psci_ops()`` during
717BL31 initialization. If it's a cold reset then this function must return zero.
718
719This function does not follow the Procedure Call Standard used by the
720Application Binary Interface for the Arm 64-bit architecture. The caller should
721not assume that callee saved registers are preserved across a call to this
722function.
723
724This function fulfills requirement 1 and 3 listed above.
725
726Note that for platforms that support programming the reset address, it is
727expected that a CPU will start executing code directly at the right address,
728both on a cold and warm reset. In this case, there is no need to identify the
729type of reset nor to query the warm reset entrypoint. Therefore, implementing
730this function is not required on such platforms.
731
732Function : plat_secondary_cold_boot_setup() [mandatory when COLD_BOOT_SINGLE_CPU == 0]
733~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
734
735::
736
737    Argument : void
738
739This function is called with the MMU and data caches disabled. It is responsible
740for placing the executing secondary CPU in a platform-specific state until the
741primary CPU performs the necessary actions to bring it out of that state and
742allow entry into the OS. This function must not return.
743
744In the Arm FVP port, when using the normal boot flow, each secondary CPU powers
745itself off. The primary CPU is responsible for powering up the secondary CPUs
746when normal world software requires them. When booting an EL3 payload instead,
747they stay powered on and are put in a holding pen until their mailbox gets
748populated.
749
750This function fulfills requirement 2 above.
751
752Note that for platforms that can't release secondary CPUs out of reset, only the
753primary CPU will execute the cold boot code. Therefore, implementing this
754function is not required on such platforms.
755
756Function : plat_is_my_cpu_primary() [mandatory when COLD_BOOT_SINGLE_CPU == 0]
757~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
758
759::
760
761    Argument : void
762    Return   : unsigned int
763
764This function identifies whether the current CPU is the primary CPU or a
765secondary CPU. A return value of zero indicates that the CPU is not the
766primary CPU, while a non-zero return value indicates that the CPU is the
767primary CPU.
768
769Note that for platforms that can't release secondary CPUs out of reset, only the
770primary CPU will execute the cold boot code. Therefore, there is no need to
771distinguish between primary and secondary CPUs and implementing this function is
772not required.
773
774Function : platform_mem_init() [mandatory]
775~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
776
777::
778
779    Argument : void
780    Return   : void
781
782This function is called before any access to data is made by the firmware, in
783order to carry out any essential memory initialization.
784
785Function: plat_get_rotpk_info()
786~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
787
788::
789
790    Argument : void *, void **, unsigned int *, unsigned int *
791    Return   : int
792
793This function is mandatory when Trusted Board Boot is enabled. It returns a
794pointer to the ROTPK stored in the platform (or a hash of it) and its length.
795The ROTPK must be encoded in DER format according to the following ASN.1
796structure:
797
798::
799
800    AlgorithmIdentifier  ::=  SEQUENCE  {
801        algorithm         OBJECT IDENTIFIER,
802        parameters        ANY DEFINED BY algorithm OPTIONAL
803    }
804
805    SubjectPublicKeyInfo  ::=  SEQUENCE  {
806        algorithm         AlgorithmIdentifier,
807        subjectPublicKey  BIT STRING
808    }
809
810In case the function returns a hash of the key:
811
812::
813
814    DigestInfo ::= SEQUENCE {
815        digestAlgorithm   AlgorithmIdentifier,
816        digest            OCTET STRING
817    }
818
819The function returns 0 on success. Any other value is treated as error by the
820Trusted Board Boot. The function also reports extra information related
821to the ROTPK in the flags parameter:
822
823::
824
825    ROTPK_IS_HASH      : Indicates that the ROTPK returned by the platform is a
826                         hash.
827    ROTPK_NOT_DEPLOYED : This allows the platform to skip certificate ROTPK
828                         verification while the platform ROTPK is not deployed.
829                         When this flag is set, the function does not need to
830                         return a platform ROTPK, and the authentication
831                         framework uses the ROTPK in the certificate without
832                         verifying it against the platform value. This flag
833                         must not be used in a deployed production environment.
834
835Function: plat_get_nv_ctr()
836~~~~~~~~~~~~~~~~~~~~~~~~~~~
837
838::
839
840    Argument : void *, unsigned int *
841    Return   : int
842
843This function is mandatory when Trusted Board Boot is enabled. It returns the
844non-volatile counter value stored in the platform in the second argument. The
845cookie in the first argument may be used to select the counter in case the
846platform provides more than one (for example, on platforms that use the default
847TBBR CoT, the cookie will correspond to the OID values defined in
848TRUSTED_FW_NVCOUNTER_OID or NON_TRUSTED_FW_NVCOUNTER_OID).
849
850The function returns 0 on success. Any other value means the counter value could
851not be retrieved from the platform.
852
853Function: plat_set_nv_ctr()
854~~~~~~~~~~~~~~~~~~~~~~~~~~~
855
856::
857
858    Argument : void *, unsigned int
859    Return   : int
860
861This function is mandatory when Trusted Board Boot is enabled. It sets a new
862counter value in the platform. The cookie in the first argument may be used to
863select the counter (as explained in plat_get_nv_ctr()). The second argument is
864the updated counter value to be written to the NV counter.
865
866The function returns 0 on success. Any other value means the counter value could
867not be updated.
868
869Function: plat_set_nv_ctr2()
870~~~~~~~~~~~~~~~~~~~~~~~~~~~~
871
872::
873
874    Argument : void *, const auth_img_desc_t *, unsigned int
875    Return   : int
876
877This function is optional when Trusted Board Boot is enabled. If this
878interface is defined, then ``plat_set_nv_ctr()`` need not be defined. The
879first argument passed is a cookie and is typically used to
880differentiate between a Non Trusted NV Counter and a Trusted NV
881Counter. The second argument is a pointer to an authentication image
882descriptor and may be used to decide if the counter is allowed to be
883updated or not. The third argument is the updated counter value to
884be written to the NV counter.
885
886The function returns 0 on success. Any other value means the counter value
887either could not be updated or the authentication image descriptor indicates
888that it is not allowed to be updated.
889
890Dynamic Root of Trust for Measurement support (in BL31)
891-------------------------------------------------------
892
893The functions mentioned in this section are mandatory, when platform enables
894DRTM_SUPPORT build flag.
895
896Function : plat_get_addr_mmap()
897~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
898
899::
900
901    Argument : void
902    Return   : const mmap_region_t *
903
904This function is used to return the address of the platform *address-map* table,
905which describes the regions of normal memory, memory mapped I/O
906and non-volatile memory.
907
908Function : plat_has_non_host_platforms()
909~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
910
911::
912
913    Argument : void
914    Return   : bool
915
916This function returns *true* if the platform has any trusted devices capable of
917DMA, otherwise returns *false*.
918
919Function : plat_has_unmanaged_dma_peripherals()
920~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
921
922::
923
924    Argument : void
925    Return   : bool
926
927This function returns *true* if platform uses peripherals whose DMA is not
928managed by an SMMU, otherwise returns *false*.
929
930Note -
931If the platform has peripherals that are not managed by the SMMU, then the
932platform should investigate such peripherals to determine whether they can
933be trusted, and such peripherals should be moved under "Non-host platforms"
934if they can be trusted.
935
936Function : plat_get_total_num_smmus()
937~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
938
939::
940
941    Argument : void
942    Return   : unsigned int
943
944This function returns the total number of SMMUs in the platform.
945
946Function : plat_enumerate_smmus()
947~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
948::
949
950
951    Argument : void
952    Return   : const uintptr_t *, size_t
953
954This function returns an array of SMMU addresses and the actual number of SMMUs
955reported by the platform.
956
957Function : plat_drtm_get_dma_prot_features()
958~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
959
960::
961
962    Argument : void
963    Return   : const plat_drtm_dma_prot_features_t*
964
965This function returns the address of plat_drtm_dma_prot_features_t structure
966containing the maximum number of protected regions and bitmap with the types
967of DMA protection supported by the platform.
968For more details see section 3.3 Table 6 of `DRTM`_ specification.
969
970Function : plat_drtm_dma_prot_get_max_table_bytes()
971~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
972
973::
974
975    Argument : void
976    Return   : uint64_t
977
978This function returns the maximum size of DMA protected regions table in
979bytes.
980
981Function : plat_drtm_get_tpm_features()
982~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
983
984::
985
986    Argument : void
987    Return   : const plat_drtm_tpm_features_t*
988
989This function returns the address of *plat_drtm_tpm_features_t* structure
990containing PCR usage schema, TPM-based hash, and firmware hash algorithm
991supported by the platform.
992
993Function : plat_drtm_get_min_size_normal_world_dce()
994~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
995
996::
997
998    Argument : void
999    Return   : uint64_t
1000
1001This function returns the size normal-world DCE of the platform.
1002
1003Function : plat_drtm_get_imp_def_dlme_region_size()
1004~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1005
1006::
1007
1008    Argument : void
1009    Return   : uint64_t
1010
1011This function returns the size of implementation defined DLME region
1012of the platform.
1013
1014Function : plat_drtm_get_tcb_hash_table_size()
1015~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1016
1017::
1018
1019    Argument : void
1020    Return   : uint64_t
1021
1022This function returns the size of TCB hash table of the platform.
1023
1024Function : plat_drtm_get_acpi_tables_region_size()
1025~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1026
1027::
1028
1029    Argument : void
1030    Return   : uint64_t
1031
1032This function returns the size of ACPI tables region of the platform.
1033
1034Function : plat_drtm_get_tcb_hash_features()
1035~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1036
1037::
1038
1039    Argument : void
1040    Return   : uint64_t
1041
1042This function returns the Maximum number of TCB hashes recorded by the
1043platform.
1044For more details see section 3.3 Table 6 of `DRTM`_ specification.
1045
1046Function : plat_drtm_get_dlme_img_auth_features()
1047~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1048
1049::
1050
1051    Argument : void
1052    Return   : uint64_t
1053
1054This function returns the DLME image authentication features.
1055For more details see section 3.3 Table 6 of `DRTM`_ specification.
1056
1057Function : plat_drtm_validate_ns_region()
1058~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1059
1060::
1061
1062    Argument : uintptr_t, uintptr_t
1063    Return   : int
1064
1065This function validates that given region is within the Non-Secure region
1066of DRAM. This function takes a region start address and size an input
1067arguments, and returns 0 on success and -1 on failure.
1068
1069Function : plat_set_drtm_error()
1070~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1071
1072::
1073
1074    Argument : uint64_t
1075    Return   : int
1076
1077This function writes a 64 bit error code received as input into
1078non-volatile storage and returns 0 on success and -1 on failure.
1079
1080Function : plat_get_drtm_error()
1081~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1082
1083::
1084
1085    Argument : uint64_t*
1086    Return   : int
1087
1088This function reads a 64 bit error code from the non-volatile storage
1089into the received address, and returns 0 on success and -1 on failure.
1090
1091Common mandatory function modifications
1092---------------------------------------
1093
1094The following functions are mandatory functions which need to be implemented
1095by the platform port.
1096
1097Function : plat_my_core_pos()
1098~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1099
1100::
1101
1102    Argument : void
1103    Return   : unsigned int
1104
1105This function returns the index of the calling CPU which is used as a
1106CPU-specific linear index into blocks of memory (for example while allocating
1107per-CPU stacks). This function will be invoked very early in the
1108initialization sequence which mandates that this function should be
1109implemented in assembly and should not rely on the availability of a C
1110runtime environment. This function can clobber x0 - x8 and must preserve
1111x9 - x29.
1112
1113This function plays a crucial role in the power domain topology framework in
1114PSCI and details of this can be found in
1115:ref:`PSCI Power Domain Tree Structure`.
1116
1117Function : plat_core_pos_by_mpidr()
1118~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1119
1120::
1121
1122    Argument : u_register_t
1123    Return   : int
1124
1125This function validates the ``MPIDR`` of a CPU and converts it to an index,
1126which can be used as a CPU-specific linear index into blocks of memory. In
1127case the ``MPIDR`` is invalid, this function returns -1. This function will only
1128be invoked by BL31 after the power domain topology is initialized and can
1129utilize the C runtime environment. For further details about how TF-A
1130represents the power domain topology and how this relates to the linear CPU
1131index, please refer :ref:`PSCI Power Domain Tree Structure`.
1132
1133Function : plat_get_mbedtls_heap() [when TRUSTED_BOARD_BOOT == 1]
1134~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1135
1136::
1137
1138    Arguments : void **heap_addr, size_t *heap_size
1139    Return    : int
1140
1141This function is invoked during Mbed TLS library initialisation to get a heap,
1142by means of a starting address and a size. This heap will then be used
1143internally by the Mbed TLS library. Hence, each BL stage that utilises Mbed TLS
1144must be able to provide a heap to it.
1145
1146A helper function can be found in `drivers/auth/mbedtls/mbedtls_common.c` in
1147which a heap is statically reserved during compile time inside every image
1148(i.e. every BL stage) that utilises Mbed TLS. In this default implementation,
1149the function simply returns the address and size of this "pre-allocated" heap.
1150For a platform to use this default implementation, only a call to the helper
1151from inside plat_get_mbedtls_heap() body is enough and nothing else is needed.
1152
1153However, by writting their own implementation, platforms have the potential to
1154optimise memory usage. For example, on some Arm platforms, the Mbed TLS heap is
1155shared between BL1 and BL2 stages and, thus, the necessary space is not reserved
1156twice.
1157
1158On success the function should return 0 and a negative error code otherwise.
1159
1160Function : plat_get_enc_key_info() [when FW_ENC_STATUS == 0 or 1]
1161~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1162
1163::
1164
1165    Arguments : enum fw_enc_status_t fw_enc_status, uint8_t *key,
1166                size_t *key_len, unsigned int *flags, const uint8_t *img_id,
1167                size_t img_id_len
1168    Return    : int
1169
1170This function provides a symmetric key (either SSK or BSSK depending on
1171fw_enc_status) which is invoked during runtime decryption of encrypted
1172firmware images. `plat/common/plat_bl_common.c` provides a dummy weak
1173implementation for testing purposes which must be overridden by the platform
1174trying to implement a real world firmware encryption use-case.
1175
1176It also allows the platform to pass symmetric key identifier rather than
1177actual symmetric key which is useful in cases where the crypto backend provides
1178secure storage for the symmetric key. So in this case ``ENC_KEY_IS_IDENTIFIER``
1179flag must be set in ``flags``.
1180
1181In addition to above a platform may also choose to provide an image specific
1182symmetric key/identifier using img_id.
1183
1184On success the function should return 0 and a negative error code otherwise.
1185
1186Note that this API depends on ``DECRYPTION_SUPPORT`` build flag.
1187
1188Function : plat_fwu_set_images_source() [when PSA_FWU_SUPPORT == 1]
1189~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1190
1191::
1192
1193    Argument : const struct fwu_metadata *metadata
1194    Return   : void
1195
1196This function is mandatory when PSA_FWU_SUPPORT is enabled.
1197It provides a means to retrieve image specification (offset in
1198non-volatile storage and length) of active/updated images using the passed
1199FWU metadata, and update I/O policies of active/updated images using retrieved
1200image specification information.
1201Further I/O layer operations such as I/O open, I/O read, etc. on these
1202images rely on this function call.
1203
1204In Arm platforms, this function is used to set an I/O policy of the FIP image,
1205container of all active/updated secure and non-secure images.
1206
1207Function : plat_fwu_set_metadata_image_source() [when PSA_FWU_SUPPORT == 1]
1208~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1209
1210::
1211
1212    Argument : unsigned int image_id, uintptr_t *dev_handle,
1213               uintptr_t *image_spec
1214    Return   : int
1215
1216This function is mandatory when PSA_FWU_SUPPORT is enabled. It is
1217responsible for setting up the platform I/O policy of the requested metadata
1218image (either FWU_METADATA_IMAGE_ID or BKUP_FWU_METADATA_IMAGE_ID) that will
1219be used to load this image from the platform's non-volatile storage.
1220
1221FWU metadata can not be always stored as a raw image in non-volatile storage
1222to define its image specification (offset in non-volatile storage and length)
1223statically in I/O policy.
1224For example, the FWU metadata image is stored as a partition inside the GUID
1225partition table image. Its specification is defined in the partition table
1226that needs to be parsed dynamically.
1227This function provides a means to retrieve such dynamic information to set
1228the I/O policy of the FWU metadata image.
1229Further I/O layer operations such as I/O open, I/O read, etc. on FWU metadata
1230image relies on this function call.
1231
1232It returns '0' on success, otherwise a negative error value on error.
1233Alongside, returns device handle and image specification from the I/O policy
1234of the requested FWU metadata image.
1235
1236Function : plat_fwu_get_boot_idx() [when PSA_FWU_SUPPORT == 1]
1237~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1238
1239::
1240
1241    Argument : void
1242    Return   : uint32_t
1243
1244This function is mandatory when PSA_FWU_SUPPORT is enabled. It provides the
1245means to retrieve the boot index value from the platform. The boot index is the
1246bank from which the platform has booted the firmware images.
1247
1248By default, the platform will read the metadata structure and try to boot from
1249the active bank. If the platform fails to boot from the active bank due to
1250reasons like an Authentication failure, or on crossing a set number of watchdog
1251resets while booting from the active bank, the platform can then switch to boot
1252from a different bank. This function then returns the bank that the platform
1253should boot its images from.
1254
1255Common optional modifications
1256-----------------------------
1257
1258The following are helper functions implemented by the firmware that perform
1259common platform-specific tasks. A platform may choose to override these
1260definitions.
1261
1262Function : plat_set_my_stack()
1263~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1264
1265::
1266
1267    Argument : void
1268    Return   : void
1269
1270This function sets the current stack pointer to the normal memory stack that
1271has been allocated for the current CPU. For BL images that only require a
1272stack for the primary CPU, the UP version of the function is used. The size
1273of the stack allocated to each CPU is specified by the platform defined
1274constant ``PLATFORM_STACK_SIZE``.
1275
1276Common implementations of this function for the UP and MP BL images are
1277provided in ``plat/common/aarch64/platform_up_stack.S`` and
1278``plat/common/aarch64/platform_mp_stack.S``
1279
1280Function : plat_get_my_stack()
1281~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1282
1283::
1284
1285    Argument : void
1286    Return   : uintptr_t
1287
1288This function returns the base address of the normal memory stack that
1289has been allocated for the current CPU. For BL images that only require a
1290stack for the primary CPU, the UP version of the function is used. The size
1291of the stack allocated to each CPU is specified by the platform defined
1292constant ``PLATFORM_STACK_SIZE``.
1293
1294Common implementations of this function for the UP and MP BL images are
1295provided in ``plat/common/aarch64/platform_up_stack.S`` and
1296``plat/common/aarch64/platform_mp_stack.S``
1297
1298Function : plat_report_exception()
1299~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1300
1301::
1302
1303    Argument : unsigned int
1304    Return   : void
1305
1306A platform may need to report various information about its status when an
1307exception is taken, for example the current exception level, the CPU security
1308state (secure/non-secure), the exception type, and so on. This function is
1309called in the following circumstances:
1310
1311-  In BL1, whenever an exception is taken.
1312-  In BL2, whenever an exception is taken.
1313
1314The default implementation doesn't do anything, to avoid making assumptions
1315about the way the platform displays its status information.
1316
1317For AArch64, this function receives the exception type as its argument.
1318Possible values for exceptions types are listed in the
1319``include/common/bl_common.h`` header file. Note that these constants are not
1320related to any architectural exception code; they are just a TF-A convention.
1321
1322For AArch32, this function receives the exception mode as its argument.
1323Possible values for exception modes are listed in the
1324``include/lib/aarch32/arch.h`` header file.
1325
1326Function : plat_reset_handler()
1327~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1328
1329::
1330
1331    Argument : void
1332    Return   : void
1333
1334A platform may need to do additional initialization after reset. This function
1335allows the platform to do the platform specific initializations. Platform
1336specific errata workarounds could also be implemented here. The API should
1337preserve the values of callee saved registers x19 to x29.
1338
1339The default implementation doesn't do anything. If a platform needs to override
1340the default implementation, refer to the :ref:`Firmware Design` for general
1341guidelines.
1342
1343Function : plat_disable_acp()
1344~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1345
1346::
1347
1348    Argument : void
1349    Return   : void
1350
1351This API allows a platform to disable the Accelerator Coherency Port (if
1352present) during a cluster power down sequence. The default weak implementation
1353doesn't do anything. Since this API is called during the power down sequence,
1354it has restrictions for stack usage and it can use the registers x0 - x17 as
1355scratch registers. It should preserve the value in x18 register as it is used
1356by the caller to store the return address.
1357
1358Function : plat_error_handler()
1359~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1360
1361::
1362
1363    Argument : int
1364    Return   : void
1365
1366This API is called when the generic code encounters an error situation from
1367which it cannot continue. It allows the platform to perform error reporting or
1368recovery actions (for example, reset the system). This function must not return.
1369
1370The parameter indicates the type of error using standard codes from ``errno.h``.
1371Possible errors reported by the generic code are:
1372
1373-  ``-EAUTH``: a certificate or image could not be authenticated (when Trusted
1374   Board Boot is enabled)
1375-  ``-ENOENT``: the requested image or certificate could not be found or an IO
1376   error was detected
1377-  ``-ENOMEM``: resources exhausted. TF-A does not use dynamic memory, so this
1378   error is usually an indication of an incorrect array size
1379
1380The default implementation simply spins.
1381
1382Function : plat_panic_handler()
1383~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1384
1385::
1386
1387    Argument : void
1388    Return   : void
1389
1390This API is called when the generic code encounters an unexpected error
1391situation from which it cannot recover. This function must not return,
1392and must be implemented in assembly because it may be called before the C
1393environment is initialized.
1394
1395.. note::
1396   The address from where it was called is stored in x30 (Link Register).
1397   The default implementation simply spins.
1398
1399Function : plat_system_reset()
1400~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1401
1402::
1403
1404    Argument : void
1405    Return   : void
1406
1407This function is used by the platform to resets the system. It can be used
1408in any specific use-case where system needs to be resetted. For example,
1409in case of DRTM implementation this function reset the system after
1410writing the DRTM error code in the non-volatile storage. This function
1411never returns. Failure in reset results in panic.
1412
1413Function : plat_get_bl_image_load_info()
1414~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1415
1416::
1417
1418    Argument : void
1419    Return   : bl_load_info_t *
1420
1421This function returns pointer to the list of images that the platform has
1422populated to load. This function is invoked in BL2 to load the
1423BL3xx images.
1424
1425Function : plat_get_next_bl_params()
1426~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1427
1428::
1429
1430    Argument : void
1431    Return   : bl_params_t *
1432
1433This function returns a pointer to the shared memory that the platform has
1434kept aside to pass TF-A related information that next BL image needs. This
1435function is invoked in BL2 to pass this information to the next BL
1436image.
1437
1438Function : plat_get_stack_protector_canary()
1439~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1440
1441::
1442
1443    Argument : void
1444    Return   : u_register_t
1445
1446This function returns a random value that is used to initialize the canary used
1447when the stack protector is enabled with ENABLE_STACK_PROTECTOR. A predictable
1448value will weaken the protection as the attacker could easily write the right
1449value as part of the attack most of the time. Therefore, it should return a
1450true random number.
1451
1452.. warning::
1453   For the protection to be effective, the global data need to be placed at
1454   a lower address than the stack bases. Failure to do so would allow an
1455   attacker to overwrite the canary as part of the stack buffer overflow attack.
1456
1457Function : plat_flush_next_bl_params()
1458~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1459
1460::
1461
1462    Argument : void
1463    Return   : void
1464
1465This function flushes to main memory all the image params that are passed to
1466next image. This function is invoked in BL2 to flush this information
1467to the next BL image.
1468
1469Function : plat_log_get_prefix()
1470~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1471
1472::
1473
1474    Argument : unsigned int
1475    Return   : const char *
1476
1477This function defines the prefix string corresponding to the `log_level` to be
1478prepended to all the log output from TF-A. The `log_level` (argument) will
1479correspond to one of the standard log levels defined in debug.h. The platform
1480can override the common implementation to define a different prefix string for
1481the log output. The implementation should be robust to future changes that
1482increase the number of log levels.
1483
1484Function : plat_get_soc_version()
1485~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1486
1487::
1488
1489    Argument : void
1490    Return   : int32_t
1491
1492This function returns soc version which mainly consist of below fields
1493
1494::
1495
1496    soc_version[30:24] = JEP-106 continuation code for the SiP
1497    soc_version[23:16] = JEP-106 identification code with parity bit for the SiP
1498    soc_version[15:0]  = Implementation defined SoC ID
1499
1500Function : plat_get_soc_revision()
1501~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1502
1503::
1504
1505    Argument : void
1506    Return   : int32_t
1507
1508This function returns soc revision in below format
1509
1510::
1511
1512    soc_revision[0:30] = SOC revision of specific SOC
1513
1514Function : plat_is_smccc_feature_available()
1515~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1516
1517::
1518
1519    Argument : u_register_t
1520    Return   : int32_t
1521
1522This function returns SMC_ARCH_CALL_SUCCESS if the platform supports
1523the SMCCC function specified in the argument; otherwise returns
1524SMC_ARCH_CALL_NOT_SUPPORTED.
1525
1526Function : plat_can_cmo()
1527~~~~~~~~~~~~~~~~~~~~~~~~~
1528
1529::
1530
1531    Argument : void
1532    Return   : uint64_t
1533
1534When CONDITIONAL_CMO flag is enabled:
1535
1536- This function indicates whether cache management operations should be
1537  performed. It returns 0 if CMOs should be skipped and non-zero
1538  otherwise.
1539- The function must not clobber x1, x2 and x3. It's also not safe to rely on
1540  stack. Otherwise obey AAPCS.
1541
1542Struct: plat_try_images_ops [optional]
1543~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1544
1545This optional structure holds platform hooks for alternative images load.
1546It has to be defined in platform code and registered by calling
1547plat_setup_try_img_ops() function, passing it the address of the
1548plat_try_images_ops struct.
1549
1550Function : plat_setup_try_img_ops [optional]
1551............................................
1552
1553::
1554
1555    Argument : const struct plat_try_images_ops *
1556    Return   : void
1557
1558This optional function is called to register platform try images ops, given
1559as argument.
1560
1561Function : plat_try_images_ops.next_instance [optional]
1562.......................................................
1563
1564::
1565
1566    Argument : unsigned int image_id
1567    Return   : int
1568
1569This optional function tries to load images from alternative places.
1570In case PSA FWU is not used, it can be any instance or media. If PSA FWU is
1571used, it is mandatory that the backup image is on the same media.
1572This is required for MTD devices like NAND.
1573The argument is the ID of the image for which we are looking for an alternative
1574place. It returns 0 in case of success and a negative errno value otherwise.
1575
1576Modifications specific to a Boot Loader stage
1577---------------------------------------------
1578
1579Boot Loader Stage 1 (BL1)
1580-------------------------
1581
1582BL1 implements the reset vector where execution starts from after a cold or
1583warm boot. For each CPU, BL1 is responsible for the following tasks:
1584
1585#. Handling the reset as described in section 2.2
1586
1587#. In the case of a cold boot and the CPU being the primary CPU, ensuring that
1588   only this CPU executes the remaining BL1 code, including loading and passing
1589   control to the BL2 stage.
1590
1591#. Identifying and starting the Firmware Update process (if required).
1592
1593#. Loading the BL2 image from non-volatile storage into secure memory at the
1594   address specified by the platform defined constant ``BL2_BASE``.
1595
1596#. Populating a ``meminfo`` structure with the following information in memory,
1597   accessible by BL2 immediately upon entry.
1598
1599   ::
1600
1601       meminfo.total_base = Base address of secure RAM visible to BL2
1602       meminfo.total_size = Size of secure RAM visible to BL2
1603
1604   By default, BL1 places this ``meminfo`` structure at the end of secure
1605   memory visible to BL2.
1606
1607   It is possible for the platform to decide where it wants to place the
1608   ``meminfo`` structure for BL2 or restrict the amount of memory visible to
1609   BL2 by overriding the weak default implementation of
1610   ``bl1_plat_handle_post_image_load`` API.
1611
1612The following functions need to be implemented by the platform port to enable
1613BL1 to perform the above tasks.
1614
1615Function : bl1_early_platform_setup() [mandatory]
1616~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1617
1618::
1619
1620    Argument : void
1621    Return   : void
1622
1623This function executes with the MMU and data caches disabled. It is only called
1624by the primary CPU.
1625
1626On Arm standard platforms, this function:
1627
1628-  Enables a secure instance of SP805 to act as the Trusted Watchdog.
1629
1630-  Initializes a UART (PL011 console), which enables access to the ``printf``
1631   family of functions in BL1.
1632
1633-  Enables issuing of snoop and DVM (Distributed Virtual Memory) requests to
1634   the CCI slave interface corresponding to the cluster that includes the
1635   primary CPU.
1636
1637Function : bl1_plat_arch_setup() [mandatory]
1638~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1639
1640::
1641
1642    Argument : void
1643    Return   : void
1644
1645This function performs any platform-specific and architectural setup that the
1646platform requires. Platform-specific setup might include configuration of
1647memory controllers and the interconnect.
1648
1649In Arm standard platforms, this function enables the MMU.
1650
1651This function helps fulfill requirement 2 above.
1652
1653Function : bl1_platform_setup() [mandatory]
1654~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1655
1656::
1657
1658    Argument : void
1659    Return   : void
1660
1661This function executes with the MMU and data caches enabled. It is responsible
1662for performing any remaining platform-specific setup that can occur after the
1663MMU and data cache have been enabled.
1664
1665In Arm standard platforms, this function initializes the storage abstraction
1666layer used to load the next bootloader image.
1667
1668This function helps fulfill requirement 4 above.
1669
1670Function : bl1_plat_sec_mem_layout() [mandatory]
1671~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1672
1673::
1674
1675    Argument : void
1676    Return   : meminfo *
1677
1678This function should only be called on the cold boot path. It executes with the
1679MMU and data caches enabled. The pointer returned by this function must point to
1680a ``meminfo`` structure containing the extents and availability of secure RAM for
1681the BL1 stage.
1682
1683::
1684
1685    meminfo.total_base = Base address of secure RAM visible to BL1
1686    meminfo.total_size = Size of secure RAM visible to BL1
1687
1688This information is used by BL1 to load the BL2 image in secure RAM. BL1 also
1689populates a similar structure to tell BL2 the extents of memory available for
1690its own use.
1691
1692This function helps fulfill requirements 4 and 5 above.
1693
1694Function : bl1_plat_prepare_exit() [optional]
1695~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1696
1697::
1698
1699    Argument : entry_point_info_t *
1700    Return   : void
1701
1702This function is called prior to exiting BL1 in response to the
1703``BL1_SMC_RUN_IMAGE`` SMC request raised by BL2. It should be used to perform
1704platform specific clean up or bookkeeping operations before transferring
1705control to the next image. It receives the address of the ``entry_point_info_t``
1706structure passed from BL2. This function runs with MMU disabled.
1707
1708Function : bl1_plat_set_ep_info() [optional]
1709~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1710
1711::
1712
1713    Argument : unsigned int image_id, entry_point_info_t *ep_info
1714    Return   : void
1715
1716This function allows platforms to override ``ep_info`` for the given ``image_id``.
1717
1718The default implementation just returns.
1719
1720Function : bl1_plat_get_next_image_id() [optional]
1721~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1722
1723::
1724
1725    Argument : void
1726    Return   : unsigned int
1727
1728This and the following function must be overridden to enable the FWU feature.
1729
1730BL1 calls this function after platform setup to identify the next image to be
1731loaded and executed. If the platform returns ``BL2_IMAGE_ID`` then BL1 proceeds
1732with the normal boot sequence, which loads and executes BL2. If the platform
1733returns a different image id, BL1 assumes that Firmware Update is required.
1734
1735The default implementation always returns ``BL2_IMAGE_ID``. The Arm development
1736platforms override this function to detect if firmware update is required, and
1737if so, return the first image in the firmware update process.
1738
1739Function : bl1_plat_get_image_desc() [optional]
1740~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1741
1742::
1743
1744    Argument : unsigned int image_id
1745    Return   : image_desc_t *
1746
1747BL1 calls this function to get the image descriptor information ``image_desc_t``
1748for the provided ``image_id`` from the platform.
1749
1750The default implementation always returns a common BL2 image descriptor. Arm
1751standard platforms return an image descriptor corresponding to BL2 or one of
1752the firmware update images defined in the Trusted Board Boot Requirements
1753specification.
1754
1755Function : bl1_plat_handle_pre_image_load() [optional]
1756~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1757
1758::
1759
1760    Argument : unsigned int image_id
1761    Return   : int
1762
1763This function can be used by the platforms to update/use image information
1764corresponding to ``image_id``. This function is invoked in BL1, both in cold
1765boot and FWU code path, before loading the image.
1766
1767Function : bl1_plat_calc_bl2_layout() [optional]
1768~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1769
1770::
1771
1772    Argument : const meminfo_t *bl1_mem_layout, meminfo_t *bl2_mem_layout
1773    Return   : void
1774
1775This utility function calculates the memory layout of BL2, representing it in a
1776`meminfo_t` structure. The default implementation derives this layout from the
1777positioning of BL1’s RW data at the top of the memory layout.
1778
1779Function : bl1_plat_handle_post_image_load() [optional]
1780~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1781
1782::
1783
1784    Argument : unsigned int image_id
1785    Return   : int
1786
1787This function can be used by the platforms to update/use image information
1788corresponding to ``image_id``. This function is invoked in BL1, both in cold
1789boot and FWU code path, after loading and authenticating the image.
1790
1791The default weak implementation of this function calculates the amount of
1792Trusted SRAM that can be used by BL2 and allocates a ``meminfo_t``
1793structure at the beginning of this free memory and populates it. The address
1794of ``meminfo_t`` structure is updated in ``arg1`` of the entrypoint
1795information to BL2.
1796
1797Function : bl1_plat_fwu_done() [optional]
1798~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1799
1800::
1801
1802    Argument : unsigned int image_id, uintptr_t image_src,
1803               unsigned int image_size
1804    Return   : void
1805
1806BL1 calls this function when the FWU process is complete. It must not return.
1807The platform may override this function to take platform specific action, for
1808example to initiate the normal boot flow.
1809
1810The default implementation spins forever.
1811
1812Function : bl1_plat_mem_check() [mandatory]
1813~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1814
1815::
1816
1817    Argument : uintptr_t mem_base, unsigned int mem_size,
1818               unsigned int flags
1819    Return   : int
1820
1821BL1 calls this function while handling FWU related SMCs, more specifically when
1822copying or authenticating an image. Its responsibility is to ensure that the
1823region of memory identified by ``mem_base`` and ``mem_size`` is mapped in BL1, and
1824that this memory corresponds to either a secure or non-secure memory region as
1825indicated by the security state of the ``flags`` argument.
1826
1827This function can safely assume that the value resulting from the addition of
1828``mem_base`` and ``mem_size`` fits into a ``uintptr_t`` type variable and does not
1829overflow.
1830
1831This function must return 0 on success, a non-null error code otherwise.
1832
1833The default implementation of this function asserts therefore platforms must
1834override it when using the FWU feature.
1835
1836Boot Loader Stage 2 (BL2)
1837-------------------------
1838
1839The BL2 stage is executed only by the primary CPU, which is determined in BL1
1840using the ``platform_is_primary_cpu()`` function. BL1 passed control to BL2 at
1841``BL2_BASE``. BL2 executes in Secure EL1 and and invokes
1842``plat_get_bl_image_load_info()`` to retrieve the list of images to load from
1843non-volatile storage to secure/non-secure RAM. After all the images are loaded
1844then BL2 invokes ``plat_get_next_bl_params()`` to get the list of executable
1845images to be passed to the next BL image.
1846
1847The following functions must be implemented by the platform port to enable BL2
1848to perform the above tasks.
1849
1850Function : bl2_early_platform_setup2() [mandatory]
1851~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1852
1853::
1854
1855    Argument : u_register_t, u_register_t, u_register_t, u_register_t
1856    Return   : void
1857
1858This function executes with the MMU and data caches disabled. It is only called
1859by the primary CPU. The 4 arguments are passed by BL1 to BL2 and these arguments
1860are platform specific.
1861
1862On Arm standard platforms, the arguments received are :
1863
1864    arg0 - Points to load address of FW_CONFIG
1865
1866    arg1 - ``meminfo`` structure populated by BL1. The platform copies
1867    the contents of ``meminfo`` as it may be subsequently overwritten by BL2.
1868
1869On Arm standard platforms, this function also:
1870
1871-  Initializes a UART (PL011 console), which enables access to the ``printf``
1872   family of functions in BL2.
1873
1874-  Initializes the storage abstraction layer used to load further bootloader
1875   images. It is necessary to do this early on platforms with a SCP_BL2 image,
1876   since the later ``bl2_platform_setup`` must be done after SCP_BL2 is loaded.
1877
1878Function : bl2_plat_arch_setup() [mandatory]
1879~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1880
1881::
1882
1883    Argument : void
1884    Return   : void
1885
1886This function executes with the MMU and data caches disabled. It is only called
1887by the primary CPU.
1888
1889The purpose of this function is to perform any architectural initialization
1890that varies across platforms.
1891
1892On Arm standard platforms, this function enables the MMU.
1893
1894Function : bl2_platform_setup() [mandatory]
1895~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1896
1897::
1898
1899    Argument : void
1900    Return   : void
1901
1902This function may execute with the MMU and data caches enabled if the platform
1903port does the necessary initialization in ``bl2_plat_arch_setup()``. It is only
1904called by the primary CPU.
1905
1906The purpose of this function is to perform any platform initialization
1907specific to BL2.
1908
1909In Arm standard platforms, this function performs security setup, including
1910configuration of the TrustZone controller to allow non-secure masters access
1911to most of DRAM. Part of DRAM is reserved for secure world use.
1912
1913Function : bl2_plat_handle_pre_image_load() [optional]
1914~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1915
1916::
1917
1918    Argument : unsigned int
1919    Return   : int
1920
1921This function can be used by the platforms to update/use image information
1922for given ``image_id``. This function is currently invoked in BL2 before
1923loading each image.
1924
1925Function : bl2_plat_handle_post_image_load() [optional]
1926~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1927
1928::
1929
1930    Argument : unsigned int
1931    Return   : int
1932
1933This function can be used by the platforms to update/use image information
1934for given ``image_id``. This function is currently invoked in BL2 after
1935loading each image.
1936
1937Function : bl2_plat_preload_setup [optional]
1938~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1939
1940::
1941
1942    Argument : void
1943    Return   : void
1944
1945This optional function performs any BL2 platform initialization
1946required before image loading, that is not done later in
1947bl2_platform_setup().
1948
1949Boot Loader Stage 2 (BL2) at EL3
1950--------------------------------
1951
1952When the platform has a non-TF-A Boot ROM it is desirable to jump
1953directly to BL2 instead of TF-A BL1. In this case BL2 is expected to
1954execute at EL3 instead of executing at EL1. Refer to the :ref:`Firmware Design`
1955document for more information.
1956
1957All mandatory functions of BL2 must be implemented, except the functions
1958bl2_early_platform_setup and bl2_el3_plat_arch_setup, because
1959their work is done now by bl2_el3_early_platform_setup and
1960bl2_el3_plat_arch_setup. These functions should generally implement
1961the bl1_plat_xxx() and bl2_plat_xxx() functionality combined.
1962
1963
1964Function : bl2_el3_early_platform_setup() [mandatory]
1965~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1966
1967::
1968
1969	Argument : u_register_t, u_register_t, u_register_t, u_register_t
1970	Return   : void
1971
1972This function executes with the MMU and data caches disabled. It is only called
1973by the primary CPU. This function receives four parameters which can be used
1974by the platform to pass any needed information from the Boot ROM to BL2.
1975
1976On Arm standard platforms, this function does the following:
1977
1978-  Initializes a UART (PL011 console), which enables access to the ``printf``
1979   family of functions in BL2.
1980
1981-  Initializes the storage abstraction layer used to load further bootloader
1982   images. It is necessary to do this early on platforms with a SCP_BL2 image,
1983   since the later ``bl2_platform_setup`` must be done after SCP_BL2 is loaded.
1984
1985- Initializes the private variables that define the memory layout used.
1986
1987Function : bl2_el3_plat_arch_setup() [mandatory]
1988~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1989
1990::
1991
1992	Argument : void
1993	Return   : void
1994
1995This function executes with the MMU and data caches disabled. It is only called
1996by the primary CPU.
1997
1998The purpose of this function is to perform any architectural initialization
1999that varies across platforms.
2000
2001On Arm standard platforms, this function enables the MMU.
2002
2003Function : bl2_el3_plat_prepare_exit() [optional]
2004~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2005
2006::
2007
2008	Argument : void
2009	Return   : void
2010
2011This function is called prior to exiting BL2 and run the next image.
2012It should be used to perform platform specific clean up or bookkeeping
2013operations before transferring control to the next image. This function
2014runs with MMU disabled.
2015
2016FWU Boot Loader Stage 2 (BL2U)
2017------------------------------
2018
2019The AP Firmware Updater Configuration, BL2U, is an optional part of the FWU
2020process and is executed only by the primary CPU. BL1 passes control to BL2U at
2021``BL2U_BASE``. BL2U executes in Secure-EL1 and is responsible for:
2022
2023#. (Optional) Transferring the optional SCP_BL2U binary image from AP secure
2024   memory to SCP RAM. BL2U uses the SCP_BL2U ``image_info`` passed by BL1.
2025   ``SCP_BL2U_BASE`` defines the address in AP secure memory where SCP_BL2U
2026   should be copied from. Subsequent handling of the SCP_BL2U image is
2027   implemented by the platform specific ``bl2u_plat_handle_scp_bl2u()`` function.
2028   If ``SCP_BL2U_BASE`` is not defined then this step is not performed.
2029
2030#. Any platform specific setup required to perform the FWU process. For
2031   example, Arm standard platforms initialize the TZC controller so that the
2032   normal world can access DDR memory.
2033
2034The following functions must be implemented by the platform port to enable
2035BL2U to perform the tasks mentioned above.
2036
2037Function : bl2u_early_platform_setup() [mandatory]
2038~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2039
2040::
2041
2042    Argument : meminfo *mem_info, void *plat_info
2043    Return   : void
2044
2045This function executes with the MMU and data caches disabled. It is only
2046called by the primary CPU. The arguments to this function is the address
2047of the ``meminfo`` structure and platform specific info provided by BL1.
2048
2049The platform may copy the contents of the ``mem_info`` and ``plat_info`` into
2050private storage as the original memory may be subsequently overwritten by BL2U.
2051
2052On Arm CSS platforms ``plat_info`` is interpreted as an ``image_info_t`` structure,
2053to extract SCP_BL2U image information, which is then copied into a private
2054variable.
2055
2056Function : bl2u_plat_arch_setup() [mandatory]
2057~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2058
2059::
2060
2061    Argument : void
2062    Return   : void
2063
2064This function executes with the MMU and data caches disabled. It is only
2065called by the primary CPU.
2066
2067The purpose of this function is to perform any architectural initialization
2068that varies across platforms, for example enabling the MMU (since the memory
2069map differs across platforms).
2070
2071Function : bl2u_platform_setup() [mandatory]
2072~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2073
2074::
2075
2076    Argument : void
2077    Return   : void
2078
2079This function may execute with the MMU and data caches enabled if the platform
2080port does the necessary initialization in ``bl2u_plat_arch_setup()``. It is only
2081called by the primary CPU.
2082
2083The purpose of this function is to perform any platform initialization
2084specific to BL2U.
2085
2086In Arm standard platforms, this function performs security setup, including
2087configuration of the TrustZone controller to allow non-secure masters access
2088to most of DRAM. Part of DRAM is reserved for secure world use.
2089
2090Function : bl2u_plat_handle_scp_bl2u() [optional]
2091~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2092
2093::
2094
2095    Argument : void
2096    Return   : int
2097
2098This function is used to perform any platform-specific actions required to
2099handle the SCP firmware. Typically it transfers the image into SCP memory using
2100a platform-specific protocol and waits until SCP executes it and signals to the
2101Application Processor (AP) for BL2U execution to continue.
2102
2103This function returns 0 on success, a negative error code otherwise.
2104This function is included if SCP_BL2U_BASE is defined.
2105
2106Boot Loader Stage 3-1 (BL31)
2107----------------------------
2108
2109During cold boot, the BL31 stage is executed only by the primary CPU. This is
2110determined in BL1 using the ``platform_is_primary_cpu()`` function. BL1 passes
2111control to BL31 at ``BL31_BASE``. During warm boot, BL31 is executed by all
2112CPUs. BL31 executes at EL3 and is responsible for:
2113
2114#. Re-initializing all architectural and platform state. Although BL1 performs
2115   some of this initialization, BL31 remains resident in EL3 and must ensure
2116   that EL3 architectural and platform state is completely initialized. It
2117   should make no assumptions about the system state when it receives control.
2118
2119#. Passing control to a normal world BL image, pre-loaded at a platform-
2120   specific address by BL2. On ARM platforms, BL31 uses the ``bl_params`` list
2121   populated by BL2 in memory to do this.
2122
2123#. Providing runtime firmware services. Currently, BL31 only implements a
2124   subset of the Power State Coordination Interface (PSCI) API as a runtime
2125   service. See :ref:`psci_in_bl31` below for details of porting the PSCI
2126   implementation.
2127
2128#. Optionally passing control to the BL32 image, pre-loaded at a platform-
2129   specific address by BL2. BL31 exports a set of APIs that allow runtime
2130   services to specify the security state in which the next image should be
2131   executed and run the corresponding image. On ARM platforms, BL31 uses the
2132   ``bl_params`` list populated by BL2 in memory to do this.
2133
2134If BL31 is a reset vector, It also needs to handle the reset as specified in
2135section 2.2 before the tasks described above.
2136
2137The following functions must be implemented by the platform port to enable BL31
2138to perform the above tasks.
2139
2140Function : bl31_early_platform_setup2() [mandatory]
2141~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2142
2143::
2144
2145    Argument : u_register_t, u_register_t, u_register_t, u_register_t
2146    Return   : void
2147
2148This function executes with the MMU and data caches disabled. It is only called
2149by the primary CPU. BL2 can pass 4 arguments to BL31 and these arguments are
2150platform specific.
2151
2152In Arm standard platforms, the arguments received are :
2153
2154    arg0 - The pointer to the head of `bl_params_t` list
2155    which is list of executable images following BL31,
2156
2157    arg1 - Points to load address of SOC_FW_CONFIG if present
2158           except in case of Arm FVP and Juno platform.
2159
2160           In case of Arm FVP and Juno platform, points to load address
2161           of FW_CONFIG.
2162
2163    arg2 - Points to load address of HW_CONFIG if present
2164
2165    arg3 - A special value to verify platform parameters from BL2 to BL31. Not
2166    used in release builds.
2167
2168The function runs through the `bl_param_t` list and extracts the entry point
2169information for BL32 and BL33. It also performs the following:
2170
2171-  Initialize a UART (PL011 console), which enables access to the ``printf``
2172   family of functions in BL31.
2173
2174-  Enable issuing of snoop and DVM (Distributed Virtual Memory) requests to the
2175   CCI slave interface corresponding to the cluster that includes the primary
2176   CPU.
2177
2178Function : bl31_plat_arch_setup() [mandatory]
2179~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2180
2181::
2182
2183    Argument : void
2184    Return   : void
2185
2186This function executes with the MMU and data caches disabled. It is only called
2187by the primary CPU.
2188
2189The purpose of this function is to perform any architectural initialization
2190that varies across platforms.
2191
2192On Arm standard platforms, this function enables the MMU.
2193
2194Function : bl31_platform_setup() [mandatory]
2195~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2196
2197::
2198
2199    Argument : void
2200    Return   : void
2201
2202This function may execute with the MMU and data caches enabled if the platform
2203port does the necessary initialization in ``bl31_plat_arch_setup()``. It is only
2204called by the primary CPU.
2205
2206The purpose of this function is to complete platform initialization so that both
2207BL31 runtime services and normal world software can function correctly.
2208
2209On Arm standard platforms, this function does the following:
2210
2211-  Initialize the generic interrupt controller.
2212
2213   Depending on the GIC driver selected by the platform, the appropriate GICv2
2214   or GICv3 initialization will be done, which mainly consists of:
2215
2216   -  Enable secure interrupts in the GIC CPU interface.
2217   -  Disable the legacy interrupt bypass mechanism.
2218   -  Configure the priority mask register to allow interrupts of all priorities
2219      to be signaled to the CPU interface.
2220   -  Mark SGIs 8-15 and the other secure interrupts on the platform as secure.
2221   -  Target all secure SPIs to CPU0.
2222   -  Enable these secure interrupts in the GIC distributor.
2223   -  Configure all other interrupts as non-secure.
2224   -  Enable signaling of secure interrupts in the GIC distributor.
2225
2226-  Enable system-level implementation of the generic timer counter through the
2227   memory mapped interface.
2228
2229-  Grant access to the system counter timer module
2230
2231-  Initialize the power controller device.
2232
2233   In particular, initialise the locks that prevent concurrent accesses to the
2234   power controller device.
2235
2236Function : bl31_plat_runtime_setup() [optional]
2237~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2238
2239::
2240
2241    Argument : void
2242    Return   : void
2243
2244The purpose of this function is to allow the platform to perform any BL31 runtime
2245setup just prior to BL31 exit during cold boot. The default weak implementation
2246of this function is empty. Any platform that needs to perform additional runtime
2247setup, before BL31 exits, will need to override this function.
2248
2249Function : bl31_plat_get_next_image_ep_info() [mandatory]
2250~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2251
2252::
2253
2254    Argument : uint32_t
2255    Return   : entry_point_info *
2256
2257This function may execute with the MMU and data caches enabled if the platform
2258port does the necessary initializations in ``bl31_plat_arch_setup()``.
2259
2260This function is called by ``bl31_main()`` to retrieve information provided by
2261BL2 for the next image in the security state specified by the argument. BL31
2262uses this information to pass control to that image in the specified security
2263state. This function must return a pointer to the ``entry_point_info`` structure
2264(that was copied during ``bl31_early_platform_setup()``) if the image exists. It
2265should return NULL otherwise.
2266
2267Function : plat_rmmd_get_cca_attest_token() [mandatory when ENABLE_RME == 1]
2268~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2269
2270::
2271
2272    Argument : uintptr_t, size_t *, uintptr_t, size_t, size_t *
2273    Return   : int
2274
2275This function returns the Platform attestation token. If the full token does
2276not fit in the buffer, the function will return a hunk of the token and
2277indicate how many bytes were copied and how many are pending. Multiple calls
2278to this function may be needed to retrieve the entire token.
2279
2280The parameters of the function are:
2281
2282    arg0 - A pointer to the buffer where the Platform token should be copied by
2283           this function. If the platform token does not completely fit in the
2284           buffer, the function may return a piece of the token only.
2285
2286    arg1 - Contains the size (in bytes) of the buffer passed in arg0. In
2287           addition, this parameter is used by the function to return the size
2288           of the platform token length hunk copied to the buffer.
2289
2290    arg2 - A pointer to the buffer where the challenge object is stored.
2291
2292    arg3 - The length of the challenge object in bytes. Possible values are 32,
2293           48 and 64. This argument must be zero for subsequent calls to
2294           retrieve the remaining hunks of the token.
2295
2296    arg4 - Returns the remaining length of the token (in bytes) that is yet to
2297           be returned in further calls.
2298
2299The function returns 0 on success, -EINVAL on failure and -EAGAIN if the
2300resource associated with the platform token retrieval is busy.
2301
2302Function : plat_rmmd_get_cca_realm_attest_key() [mandatory when ENABLE_RME == 1]
2303~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2304
2305::
2306
2307    Argument : uintptr_t, size_t *, unsigned int
2308    Return   : int
2309
2310This function returns the delegated realm attestation key which will be used to
2311sign Realm attestation token. The API currently only supports P-384 ECC curve
2312key.
2313
2314The parameters of the function are:
2315
2316    arg0 - A pointer to the buffer where the attestation key should be copied
2317           by this function. The buffer must be big enough to hold the
2318           attestation key.
2319
2320    arg1 - Contains the size (in bytes) of the buffer passed in arg0. The
2321           function returns the attestation key length in this parameter.
2322
2323    arg2 - The type of the elliptic curve to which the requested attestation key
2324           belongs.
2325
2326The function returns 0 on success, -EINVAL on failure.
2327
2328Function : plat_rmmd_get_el3_rmm_shared_mem() [when ENABLE_RME == 1]
2329~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2330
2331::
2332
2333   Argument : uintptr_t *
2334   Return   : size_t
2335
2336This function returns the size of the shared area between EL3 and RMM (or 0 on
2337failure). A pointer to the shared area (or a NULL pointer on failure) is stored
2338in the pointer passed as argument.
2339
2340Function : plat_rmmd_load_manifest() [when ENABLE_RME == 1]
2341~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2342
2343::
2344
2345    Arguments : rmm_manifest_t *manifest
2346    Return    : int
2347
2348When ENABLE_RME is enabled, this function populates a boot manifest for the
2349RMM image and stores it in the area specified by manifest.
2350
2351When ENABLE_RME is disabled, this function is not used.
2352
2353Function : plat_rmm_mecid_key_update() [when ENABLE_RME == 1]
2354~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2355
2356::
2357
2358    Argument : uint16_t
2359    Return   : int
2360
2361This function is invoked by BL31's RMMD when there is a request from the RMM
2362monitor to update the tweak for the encryption key associated to a MECID.
2363
2364The first parameter (``uint16_t mecid``) contains the MECID for which the
2365encryption key is to be updated.
2366
2367Return value is 0 upon success and -EFAULT otherwise.
2368
2369This function needs to be implemented by a platform if it enables RME.
2370
2371Function : plat_rmmd_el3_token_sign_push_req() [mandatory when RMMD_ENABLE_EL3_TOKEN_SIGN == 1]
2372~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2373
2374::
2375
2376    Arguments : const struct el3_token_sign_request *req
2377    Return    : int
2378
2379Queue realm attestation token signing request from the RMM in EL3. The interface between
2380the RMM and EL3 is modeled as a queue but the underlying implementation may be different,
2381so long as the semantics of queuing and the error codes are used as defined below.
2382
2383See :ref:`el3_token_sign_request_struct` for definition of the request structure.
2384
2385Optional interface from the RMM-EL3 interface v0.4 onwards.
2386
2387The parameters of the functions are:
2388      arg0: Pointer to the token sign request to be pushed to EL3.
2389      The structure must be located in the RMM-EL3 shared
2390      memory buffer and must be locked before use.
2391
2392Return codes:
2393        - E_RMM_OK	On Success.
2394        - E_RMM_INVAL   If the arguments are invalid.
2395        - E_RMM_AGAIN   Indicates that the request was not queued since the
2396	  queue in EL3 is full. This may also be returned for any reason
2397	  or situation in the system, that prevents accepting the request
2398	  from the RMM.
2399        - E_RMM_UNK     If the SMC is not implemented or if interface
2400	  version is < 0.4.
2401
2402Function : plat_rmmd_el3_token_sign_pull_resp() [mandatory when RMMD_ENABLE_EL3_TOKEN_SIGN == 1]
2403~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2404
2405::
2406
2407    Arguments : struct el3_token_sign_response *resp
2408    Return    : int
2409
2410Populate the attestation signing response in the ``resp`` parameter. The interface between
2411the RMM and EL3 is modeled as a queue for responses but the underlying implementation may
2412be different, so long as the semantics of queuing and the error codes are used as defined
2413below.
2414
2415See :ref:`el3_token_sign_response_struct` for definition of the response structure.
2416
2417Optional interface from the RMM-EL3 interface v0.4 onwards.
2418
2419The parameters of the functions are:
2420          resp: Pointer to the token sign response to get from EL3.
2421	  The structure must be located in the RMM-EL3 shared
2422	  memory buffer and must be locked before use.
2423
2424Return:
2425        - E_RMM_OK      On Success.
2426        - E_RMM_INVAL   If the arguments are invalid.
2427        - E_RMM_AGAIN   Indicates that a response is not ready yet.
2428        - E_RMM_UNK     If the SMC is not implemented or if interface
2429	  version is < 0.4.
2430
2431Function : plat_rmmd_el3_token_sign_get_rak_pub() [mandatory when RMMD_ENABLE_EL3_TOKEN_SIGN == 1]
2432~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2433
2434::
2435
2436    Argument : uintptr_t, size_t *, unsigned int
2437    Return   : int
2438
2439This function returns the public portion of the realm attestation key which will be used to
2440sign Realm attestation token. Typically, with delegated attestation, the private key is
2441returned, however, there may be platforms where the private key bits are better protected
2442in a platform specific manner such that the private key is not exposed. In such cases,
2443the RMM will only cache the public key and forward any requests such as signing, that
2444uses the private key to EL3. The API currently only supports P-384 ECC curve key.
2445
2446This is an optional interface from the RMM-EL3 interface v0.4 onwards.
2447
2448The parameters of the function are:
2449
2450    arg0 - A pointer to the buffer where the public key should be copied
2451    by this function. The buffer must be big enough to hold the
2452    attestation key.
2453
2454    arg1 - Contains the size (in bytes) of the buffer passed in arg0. The
2455    function returns the attestation key length in this parameter.
2456
2457    arg2 - The type of the elliptic curve to which the requested attestation key
2458    belongs.
2459
2460The function returns E_RMM_OK on success, RMM_E_INVAL if arguments are invalid and
2461E_RMM_UNK if the SMC is not implemented or if interface version is < 0.4.
2462
2463Function : plat_rmmd_el3_ide_key_program() [mandatory when RMMD_ENABLE_IDE_KEY_PROG == 1]
2464~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2465
2466::
2467
2468    Argument : uint64_t, uint64_t, uint64_t, struct rp_ide_key_info_t *, uint64_t, uint64_t
2469    Return   : int
2470
2471This function sets the key/IV info for an IDE stream at the Root port. The key is 256 bits
2472and IV is 96 bits. The caller calls this SMC to program this key to the Rx and Tx ports
2473and for each substream corresponding to a single keyset. The platform should validate
2474the arguments `Ecam address` and `Rootport ID` before acting on it. The arguments `request ID`
2475and `cookie` are to be ignored for blocking mode and are pass-through to the response for
2476non-blocking mode.
2477
2478The platform needs to ensure proper exclusives are in place when accessed from multiple CPUs.
2479Depending on the expected latency for IDE-KM interface, the platform should choose blocking
2480or non-blocking semantics. More details about IDE Setup flow can be found
2481in this `RFC <https://github.com/TF-RMM/tf-rmm/wiki/RFC:-EL3-RMM-IDE-KM-Interface>`_.
2482
2483The parameters of the function are:
2484
2485    arg0 - The ecam address, to access and configure PCI devices in a system.
2486
2487    arg1 - The rootport ID used to identify the PCIe rootport of a connected device.
2488
2489    arg2 - The IDE stream info associated with a physical device, this parameter packs the
2490    the keyset, direction, substream and stream ID info.
2491
2492    arg3 - Structure with key and IV info.
2493
2494    arg4 - The request ID, is used in non-blocking mode only and can be ignored in blocking mode.
2495
2496    arg5 - The cookie variable, is used in non-blocking mode only and can be ignored in blocking
2497    mode.
2498
2499The function returns E_RMM_OK on success, E_RMM_INVAL if arguments are invalid, E_RMM_FAULT
2500if the key programming is unsuccesful, E_RMM_UNK for an unknown error, E_RMM_AGAIN returned
2501only for non-blocking mode if the IDE-KM interface is busy or the request queue is full.
2502E_RMM_INPROGRESS returned if the request is queued successfully and used only in non-blocking
2503mode.
2504
2505Function : plat_rmmd_el3_ide_key_set_go() [mandatory when RMMD_ENABLE_IDE_KEY_PROG == 1]
2506~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2507
2508::
2509
2510    Argument : uint64_t, uint64_t, uint64_t, uint64_t, uint64_t
2511    Return   : int
2512
2513This function activates the IDE stream at the Root Port once all the keys have been
2514programmed. The platform should validate the arguments `Ecam address` and `Rootport ID`
2515before acting on it. The arguments `request ID` and `cookie` are to be ignored for blocking
2516mode and are pass-through to the response for non-blocking mode.
2517
2518The platform needs to ensure proper exclusives are in place when accessed from multiple CPUs.
2519Depending on the expected latency for IDE-KM interface, the platform should choose blocking
2520or non-blocking semantics. More details about IDE Setup flow can be found
2521in this `RFC <https://github.com/TF-RMM/tf-rmm/wiki/RFC:-EL3-RMM-IDE-KM-Interface>`_.
2522
2523The parameters of the function are:
2524
2525    arg0 - The ecam address, to access and configure PCI devices in a system.
2526
2527    arg1 - The rootport ID used to identify the PCIe rootport of a connected device.
2528
2529    arg2 - The IDE stream info associated with a physical device, this parameter packs the
2530    the keyset, direction, substream and stream ID info.
2531
2532    arg3 - The request ID, is used in non-blocking mode only and can be ignored in blocking mode.
2533
2534    arg4 - The cookie variable, is used in non-blocking mode only and can be ignored in blocking
2535    mode.
2536
2537The function returns E_RMM_OK on success, E_RMM_INVAL if arguments are invalid, E_RMM_FAULT
2538if the key programming is unsuccesful, E_RMM_UNK for an unknown error, E_RMM_AGAIN returned
2539only for non-blocking mode if the IDE-KM interface is busy or the request queue is full.
2540E_RMM_INPROGRESS returned if the request is queued successfully and used only in non-blocking
2541mode.
2542
2543Function : plat_rmmd_el3_ide_key_set_stop() [mandatory when RMMD_ENABLE_IDE_KEY_PROG == 1]
2544~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2545
2546::
2547
2548    Argument : uint64_t, uint64_t, uint64_t, uint64_t, uint64_t
2549    Return   : int
2550
2551This function stops the IDE stream and is used to tear down the IDE stream at Root Port.
2552The platform should validate the arguments `Ecam address` and `Rootport ID` before acting
2553on it. The arguments `request ID` and `cookie` are to be ignored for blocking
2554mode and are pass-through to the response for non-blocking mode.
2555
2556The platform needs to ensure proper exclusives are in place when accessed from multiple CPUs.
2557Depending on the expected latency for IDE-KM interface, the platform should choose blocking
2558or non-blocking semantics. More details about IDE Setup flow can be found
2559in this `RFC <https://github.com/TF-RMM/tf-rmm/wiki/RFC:-EL3-RMM-IDE-KM-Interface>`_.
2560
2561The parameters of the function are:
2562
2563    arg0 - The ecam address, to access and configure PCI devices in a system.
2564
2565    arg1 - The rootport ID used to identify the PCIe rootport of a connected device.
2566
2567    arg2 - The IDE stream info associated with a physical device, this parameter packs the
2568    the keyset, direction, substream and stream ID info.
2569
2570    arg3 - The request ID, is used in non-blocking mode only and can be ignored in blocking mode.
2571
2572    arg4 - The cookie variable, is used in non-blocking mode only and can be ignored in blocking
2573    mode.
2574
2575The function returns E_RMM_OK on success, E_RMM_INVAL if arguments are invalid, E_RMM_FAULT
2576if the key programming is unsuccesful, E_RMM_UNK for an unknown error, E_RMM_AGAIN returned
2577only for non-blocking mode if the IDE-KM interface is busy or the request queue is full.
2578E_RMM_INPROGRESS returned if the request is queued successfully and used only in non-blocking
2579mode.
2580
2581Function : plat_rmmd_el3_ide_km_pull_response() [mandatory when RMMD_ENABLE_IDE_KEY_PROG == 1]
2582~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2583
2584::
2585
2586    Argument : uint64_t, uint64_t, uint64_t *, uint64_t *, uint64_t *
2587    Return   : int
2588
2589This function retrieves a reponse for any of the prior non-blocking IDE-KM requests. The
2590caller has to identify the request and populate the accurate response. For blocking calls,
2591this function always returns E_RMM_UNK.
2592
2593The platform needs to ensure proper exclusives are in place when accessed from multiple CPUs.
2594Depending on the expected latency for IDE-KM interface, the platform should choose blocking
2595or non-blocking semantics. More details about IDE Setup flow can be found
2596in this `RFC <https://github.com/TF-RMM/tf-rmm/wiki/RFC:-EL3-RMM-IDE-KM-Interface>`_.
2597
2598The parameters of the function are:
2599
2600    arg0 - The ecam address, to access and configure PCI devices in a system.
2601
2602    arg1 - The rootport ID used to identify the PCIe rootport of a connected device.
2603
2604    arg2 - Retrieved response corresponding to the previous IDE_KM request.
2605
2606    arg3 - returns the passthrough request ID of the retrieved response.
2607
2608    arg4 - returns the passthrough cookie of the retrieved response.
2609
2610The function returns E_RMM_OK if response is retrieved successfully, E_RMM_INVAL if arguments
2611to this function are invalid, E_RMM_UNK if response retrieval failed for an unknown error or
2612IDE-KM interface is having blocking semantics, E_RMM_AGAIN if the response queue is empty.
2613
2614The `arg2` return parameter can return the following values:
2615E_RMM_OK - The previous request was successful.
2616E_RMM_FAULT - The previous request was not successful.
2617E_RMM_INVAL - Arguments to previous request were incorrect.
2618E_RMM_UNK - Previous request returned Unknown error.
2619
2620Function : bl31_plat_enable_mmu [optional]
2621~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2622
2623::
2624
2625    Argument : uint32_t
2626    Return   : void
2627
2628This function enables the MMU. The boot code calls this function with MMU and
2629caches disabled. This function should program necessary registers to enable
2630translation, and upon return, the MMU on the calling PE must be enabled.
2631
2632The function must honor flags passed in the first argument. These flags are
2633defined by the translation library, and can be found in the file
2634``include/lib/xlat_tables/xlat_mmu_helpers.h``.
2635
2636On DynamIQ systems, this function must not use stack while enabling MMU, which
2637is how the function in xlat table library version 2 is implemented.
2638
2639Function : plat_init_apkey [optional]
2640~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2641
2642::
2643
2644    Argument : void
2645    Return   : uint128_t
2646
2647This function returns the 128-bit value which can be used to program ARMv8.3
2648pointer authentication keys.
2649
2650The value should be obtained from a reliable source of randomness.
2651
2652This function is only needed if ARMv8.3 pointer authentication is used in the
2653Trusted Firmware by building with ``BRANCH_PROTECTION`` option set to 1, 2 or 3.
2654
2655Function : plat_get_syscnt_freq2() [mandatory]
2656~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2657
2658::
2659
2660    Argument : void
2661    Return   : unsigned int
2662
2663This function is used by the architecture setup code to retrieve the counter
2664frequency for the CPU's generic timer. This value will be programmed into the
2665``CNTFRQ_EL0`` register. In Arm standard platforms, it returns the base frequency
2666of the system counter, which is retrieved from the first entry in the frequency
2667modes table.
2668
2669#define : PLAT_PERCPU_BAKERY_LOCK_SIZE [optional]
2670~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2671
2672When ``USE_COHERENT_MEM = 0``, this constant defines the total memory (in
2673bytes) aligned to the cache line boundary that should be allocated per-cpu to
2674accommodate all the bakery locks.
2675
2676If this constant is not defined when ``USE_COHERENT_MEM = 0``, the linker
2677calculates the size of the ``.bakery_lock`` input section, aligns it to the
2678nearest ``CACHE_WRITEBACK_GRANULE``, multiplies it with ``PLATFORM_CORE_COUNT``
2679and stores the result in a linker symbol. This constant prevents a platform
2680from relying on the linker and provide a more efficient mechanism for
2681accessing per-cpu bakery lock information.
2682
2683If this constant is defined and its value is not equal to the value
2684calculated by the linker then a link time assertion is raised. A compile time
2685assertion is raised if the value of the constant is not aligned to the cache
2686line boundary.
2687
2688.. _porting_guide_sdei_requirements:
2689
2690SDEI porting requirements
2691~~~~~~~~~~~~~~~~~~~~~~~~~
2692
2693The |SDEI| dispatcher requires the platform to provide the following macros
2694and functions, of which some are optional, and some others mandatory.
2695
2696Macros
2697......
2698
2699Macro: PLAT_SDEI_NORMAL_PRI [mandatory]
2700^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
2701
2702This macro must be defined to the EL3 exception priority level associated with
2703Normal |SDEI| events on the platform. This must have a higher value
2704(therefore of lower priority) than ``PLAT_SDEI_CRITICAL_PRI``.
2705
2706Macro: PLAT_SDEI_CRITICAL_PRI [mandatory]
2707^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
2708
2709This macro must be defined to the EL3 exception priority level associated with
2710Critical |SDEI| events on the platform. This must have a lower value
2711(therefore of higher priority) than ``PLAT_SDEI_NORMAL_PRI``.
2712
2713**Note**: |SDEI| exception priorities must be the lowest among Secure
2714priorities. Among the |SDEI| exceptions, Critical |SDEI| priority must
2715be higher than Normal |SDEI| priority.
2716
2717Functions
2718.........
2719
2720Function: int plat_sdei_validate_entry_point() [optional]
2721^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
2722
2723::
2724
2725  Argument: uintptr_t ep, unsigned int client_mode
2726  Return: int
2727
2728This function validates the entry point address of the event handler provided by
2729the client for both event registration and *Complete and Resume* |SDEI| calls.
2730The function ensures that the address is valid in the client translation regime.
2731
2732The second argument is the exception level that the client is executing in. It
2733can be Non-Secure EL1 or Non-Secure EL2.
2734
2735The function must return ``0`` for successful validation, or ``-1`` upon failure.
2736
2737The default implementation always returns ``0``. On Arm platforms, this function
2738translates the entry point address within the client translation regime and
2739further ensures that the resulting physical address is located in Non-secure
2740DRAM.
2741
2742Function: void plat_sdei_handle_masked_trigger(uint64_t mpidr, unsigned int intr) [optional]
2743^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
2744
2745::
2746
2747  Argument: uint64_t
2748  Argument: unsigned int
2749  Return: void
2750
2751|SDEI| specification requires that a PE comes out of reset with the events
2752masked. The client therefore is expected to call ``PE_UNMASK`` to unmask
2753|SDEI| events on the PE. No |SDEI| events can be dispatched until such
2754time.
2755
2756Should a PE receive an interrupt that was bound to an |SDEI| event while the
2757events are masked on the PE, the dispatcher implementation invokes the function
2758``plat_sdei_handle_masked_trigger``. The MPIDR of the PE that received the
2759interrupt and the interrupt ID are passed as parameters.
2760
2761The default implementation only prints out a warning message.
2762
2763.. _porting_guide_trng_requirements:
2764
2765TRNG porting requirements
2766~~~~~~~~~~~~~~~~~~~~~~~~~
2767
2768The |TRNG| backend requires the platform to provide the following values
2769and mandatory functions.
2770
2771Values
2772......
2773
2774value: uuid_t plat_trng_uuid [mandatory]
2775^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
2776
2777This value must be defined to the UUID of the TRNG backend that is specific to
2778the hardware after ``plat_entropy_setup`` function is called. This value must
2779conform to the SMCCC calling convention; The most significant 32 bits of the
2780UUID must not equal ``0xffffffff`` or the signed integer ``-1`` as this value in
2781w0 indicates failure to get a TRNG source.
2782
2783Functions
2784.........
2785
2786Function: void plat_entropy_setup(void) [mandatory]
2787^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
2788
2789::
2790
2791  Argument: none
2792  Return: none
2793
2794This function is expected to do platform-specific initialization of any TRNG
2795hardware. This may include generating a UUID from a hardware-specific seed.
2796
2797Function: bool plat_get_entropy(uint64_t \*out) [mandatory]
2798^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
2799
2800::
2801
2802  Argument: uint64_t *
2803  Return: bool
2804  Out : when the return value is true, the entropy has been written into the
2805  storage pointed to
2806
2807This function writes entropy into storage provided by the caller. If no entropy
2808is available, it must return false and the storage must not be written.
2809
2810.. _psci_in_bl31:
2811
2812Power State Coordination Interface (in BL31)
2813--------------------------------------------
2814
2815The TF-A implementation of the PSCI API is based around the concept of a
2816*power domain*. A *power domain* is a CPU or a logical group of CPUs which
2817share some state on which power management operations can be performed as
2818specified by `PSCI`_. Each CPU in the system is assigned a cpu index which is
2819a unique number between ``0`` and ``PLATFORM_CORE_COUNT - 1``. The
2820*power domains* are arranged in a hierarchical tree structure and each
2821*power domain* can be identified in a system by the cpu index of any CPU that
2822is part of that domain and a *power domain level*. A processing element (for
2823example, a CPU) is at level 0. If the *power domain* node above a CPU is a
2824logical grouping of CPUs that share some state, then level 1 is that group of
2825CPUs (for example, a cluster), and level 2 is a group of clusters (for
2826example, the system). More details on the power domain topology and its
2827organization can be found in :ref:`PSCI Power Domain Tree Structure`.
2828
2829BL31's platform initialization code exports a pointer to the platform-specific
2830power management operations required for the PSCI implementation to function
2831correctly. This information is populated in the ``plat_psci_ops`` structure. The
2832PSCI implementation calls members of the ``plat_psci_ops`` structure for performing
2833power management operations on the power domains. For example, the target
2834CPU is specified by its ``MPIDR`` in a PSCI ``CPU_ON`` call. The ``pwr_domain_on()``
2835handler (if present) is called for the CPU power domain.
2836
2837The ``power-state`` parameter of a PSCI ``CPU_SUSPEND`` call can be used to
2838describe composite power states specific to a platform. The PSCI implementation
2839defines a generic representation of the power-state parameter, which is an
2840array of local power states where each index corresponds to a power domain
2841level. Each entry contains the local power state the power domain at that power
2842level could enter. It depends on the ``validate_power_state()`` handler to
2843convert the power-state parameter (possibly encoding a composite power state)
2844passed in a PSCI ``CPU_SUSPEND`` call to this representation.
2845
2846The following functions form part of platform port of PSCI functionality.
2847
2848Function : plat_psci_stat_accounting_start() [optional]
2849~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2850
2851::
2852
2853    Argument : const psci_power_state_t *
2854    Return   : void
2855
2856This is an optional hook that platforms can implement for residency statistics
2857accounting before entering a low power state. The ``pwr_domain_state`` field of
2858``state_info`` (first argument) can be inspected if stat accounting is done
2859differently at CPU level versus higher levels. As an example, if the element at
2860index 0 (CPU power level) in the ``pwr_domain_state`` array indicates a power down
2861state, special hardware logic may be programmed in order to keep track of the
2862residency statistics. For higher levels (array indices > 0), the residency
2863statistics could be tracked in software using PMF. If ``ENABLE_PMF`` is set, the
2864default implementation will use PMF to capture timestamps.
2865
2866Function : plat_psci_stat_accounting_stop() [optional]
2867~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2868
2869::
2870
2871    Argument : const psci_power_state_t *
2872    Return   : void
2873
2874This is an optional hook that platforms can implement for residency statistics
2875accounting after exiting from a low power state. The ``pwr_domain_state`` field
2876of ``state_info`` (first argument) can be inspected if stat accounting is done
2877differently at CPU level versus higher levels. As an example, if the element at
2878index 0 (CPU power level) in the ``pwr_domain_state`` array indicates a power down
2879state, special hardware logic may be programmed in order to keep track of the
2880residency statistics. For higher levels (array indices > 0), the residency
2881statistics could be tracked in software using PMF. If ``ENABLE_PMF`` is set, the
2882default implementation will use PMF to capture timestamps.
2883
2884Function : plat_psci_stat_get_residency() [optional]
2885~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2886
2887::
2888
2889    Argument : unsigned int, const psci_power_state_t *, unsigned int
2890    Return   : u_register_t
2891
2892This is an optional interface that is is invoked after resuming from a low power
2893state and provides the time spent resident in that low power state by the power
2894domain at a particular power domain level. When a CPU wakes up from suspend,
2895all its parent power domain levels are also woken up. The generic PSCI code
2896invokes this function for each parent power domain that is resumed and it
2897identified by the ``lvl`` (first argument) parameter. The ``state_info`` (second
2898argument) describes the low power state that the power domain has resumed from.
2899The current CPU is the first CPU in the power domain to resume from the low
2900power state and the ``last_cpu_idx`` (third parameter) is the index of the last
2901CPU in the power domain to suspend and may be needed to calculate the residency
2902for that power domain.
2903
2904Function : plat_get_target_pwr_state() [optional]
2905~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2906
2907::
2908
2909    Argument : unsigned int, const plat_local_state_t *, unsigned int
2910    Return   : plat_local_state_t
2911
2912The PSCI generic code uses this function to let the platform participate in
2913state coordination during a power management operation. The function is passed
2914a pointer to an array of platform specific local power state ``states`` (second
2915argument) which contains the requested power state for each CPU at a particular
2916power domain level ``lvl`` (first argument) within the power domain. The function
2917is expected to traverse this array of upto ``ncpus`` (third argument) and return
2918a coordinated target power state by the comparing all the requested power
2919states. The target power state should not be deeper than any of the requested
2920power states.
2921
2922A weak definition of this API is provided by default wherein it assumes
2923that the platform assigns a local state value in order of increasing depth
2924of the power state i.e. for two power states X & Y, if X < Y
2925then X represents a shallower power state than Y. As a result, the
2926coordinated target local power state for a power domain will be the minimum
2927of the requested local power state values.
2928
2929Function : plat_get_power_domain_tree_desc() [mandatory]
2930~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2931
2932::
2933
2934    Argument : void
2935    Return   : const unsigned char *
2936
2937This function returns a pointer to the byte array containing the power domain
2938topology tree description. The format and method to construct this array are
2939described in :ref:`PSCI Power Domain Tree Structure`. The BL31 PSCI
2940initialization code requires this array to be described by the platform, either
2941statically or dynamically, to initialize the power domain topology tree. In case
2942the array is populated dynamically, then plat_core_pos_by_mpidr() and
2943plat_my_core_pos() should also be implemented suitably so that the topology tree
2944description matches the CPU indices returned by these APIs. These APIs together
2945form the platform interface for the PSCI topology framework.
2946
2947Function : plat_setup_psci_ops() [mandatory]
2948~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2949
2950::
2951
2952    Argument : uintptr_t, const plat_psci_ops **
2953    Return   : int
2954
2955This function may execute with the MMU and data caches enabled if the platform
2956port does the necessary initializations in ``bl31_plat_arch_setup()``. It is only
2957called by the primary CPU.
2958
2959This function is called by PSCI initialization code. Its purpose is to let
2960the platform layer know about the warm boot entrypoint through the
2961``sec_entrypoint`` (first argument) and to export handler routines for
2962platform-specific psci power management actions by populating the passed
2963pointer with a pointer to BL31's private ``plat_psci_ops`` structure.
2964
2965A description of each member of this structure is given below. Please refer to
2966the Arm FVP specific implementation of these handlers in
2967``plat/arm/board/fvp/fvp_pm.c`` as an example. For each PSCI function that the
2968platform wants to support, the associated operation or operations in this
2969structure must be provided and implemented (Refer section 4 of
2970:ref:`Firmware Design` for the PSCI API supported in TF-A). To disable a PSCI
2971function in a platform port, the operation should be removed from this
2972structure instead of providing an empty implementation.
2973
2974plat_psci_ops.cpu_standby()
2975...........................
2976
2977Perform the platform-specific actions to enter the standby state for a cpu
2978indicated by the passed argument. This provides a fast path for CPU standby
2979wherein overheads of PSCI state management and lock acquisition is avoided.
2980For this handler to be invoked by the PSCI ``CPU_SUSPEND`` API implementation,
2981the suspend state type specified in the ``power-state`` parameter should be
2982STANDBY and the target power domain level specified should be the CPU. The
2983handler should put the CPU into a low power retention state (usually by
2984issuing a wfi instruction) and ensure that it can be woken up from that
2985state by a normal interrupt. The generic code expects the handler to succeed.
2986
2987plat_psci_ops.pwr_domain_on()
2988.............................
2989
2990Perform the platform specific actions to power on a CPU, specified
2991by the ``MPIDR`` (first argument). The generic code expects the platform to
2992return PSCI_E_SUCCESS on success or PSCI_E_INTERN_FAIL for any failure.
2993
2994plat_psci_ops.pwr_domain_off_early() [optional]
2995...............................................
2996
2997This optional function performs the platform specific actions to check if
2998powering off the calling CPU and its higher parent power domain levels as
2999indicated by the ``target_state`` (first argument) is possible or allowed.
3000
3001The ``target_state`` encodes the platform coordinated target local power states
3002for the CPU power domain and its parent power domain levels.
3003
3004For this handler, the local power state for the CPU power domain will be a
3005power down state where as it could be either power down, retention or run state
3006for the higher power domain levels depending on the result of state
3007coordination. The generic code expects PSCI_E_DENIED return code if the
3008platform thinks that CPU_OFF should not proceed on the calling CPU.
3009
3010plat_psci_ops.pwr_domain_off()
3011..............................
3012
3013Perform the platform specific actions to prepare to power off the calling CPU
3014and its higher parent power domain levels as indicated by the ``target_state``
3015(first argument). It is called by the PSCI ``CPU_OFF`` API implementation.
3016
3017The ``target_state`` encodes the platform coordinated target local power states
3018for the CPU power domain and its parent power domain levels. The handler
3019needs to perform power management operation corresponding to the local state
3020at each power level.
3021
3022For this handler, the local power state for the CPU power domain will be a
3023power down state where as it could be either power down, retention or run state
3024for the higher power domain levels depending on the result of state
3025coordination. The generic code expects the handler to succeed.
3026
3027plat_psci_ops.pwr_domain_validate_suspend() [optional]
3028......................................................
3029
3030This is an optional function that is only compiled into the build if the build
3031option ``PSCI_OS_INIT_MODE`` is enabled.
3032
3033If implemented, this function allows the platform to perform platform specific
3034validations based on hardware states. The generic code expects this function to
3035return PSCI_E_SUCCESS on success, or either PSCI_E_DENIED or
3036PSCI_E_INVALID_PARAMS as appropriate for any invalid requests.
3037
3038plat_psci_ops.pwr_domain_suspend_pwrdown_early() [optional]
3039...........................................................
3040
3041This optional function may be used as a performance optimization to replace
3042or complement pwr_domain_suspend() on some platforms. Its calling semantics
3043are identical to pwr_domain_suspend(), except the PSCI implementation only
3044calls this function when suspending to a power down state, and it guarantees
3045that data caches are enabled.
3046
3047When HW_ASSISTED_COHERENCY = 0, the PSCI implementation disables data caches
3048before calling pwr_domain_suspend(). If the target_state corresponds to a
3049power down state and it is safe to perform some or all of the platform
3050specific actions in that function with data caches enabled, it may be more
3051efficient to move those actions to this function. When HW_ASSISTED_COHERENCY
3052= 1, data caches remain enabled throughout, and so there is no advantage to
3053moving platform specific actions to this function.
3054
3055plat_psci_ops.pwr_domain_suspend()
3056..................................
3057
3058Perform the platform specific actions to prepare to suspend the calling
3059CPU and its higher parent power domain levels as indicated by the
3060``target_state`` (first argument). It is called by the PSCI ``CPU_SUSPEND``
3061API implementation.
3062
3063The ``target_state`` has a similar meaning as described in
3064the ``pwr_domain_off()`` operation. It encodes the platform coordinated
3065target local power states for the CPU power domain and its parent
3066power domain levels. The handler needs to perform power management operation
3067corresponding to the local state at each power level. The generic code
3068expects the handler to succeed.
3069
3070The difference between turning a power domain off versus suspending it is that
3071in the former case, the power domain is expected to re-initialize its state
3072when it is next powered on (see ``pwr_domain_on_finish()``). In the latter
3073case, the power domain is expected to save enough state so that it can resume
3074execution by restoring this state when its powered on (see
3075``pwr_domain_suspend_finish()``).
3076
3077When suspending a core, the platform can also choose to power off the GICv3
3078Redistributor and ITS through an implementation-defined sequence. To achieve
3079this safely, the ITS context must be saved first. The architectural part is
3080implemented by the ``gicv3_its_save_disable()`` helper, but most of the needed
3081sequence is implementation defined and it is therefore the responsibility of
3082the platform code to implement the necessary sequence. Then the GIC
3083Redistributor context can be saved using the ``gicv3_rdistif_save()`` helper.
3084Powering off the Redistributor requires the implementation to support it and it
3085is the responsibility of the platform code to execute the right implementation
3086defined sequence.
3087
3088When a system suspend is requested, the platform can also make use of the
3089``gicv3_distif_save()`` helper to save the context of the GIC Distributor after
3090it has saved the context of the Redistributors and ITS of all the cores in the
3091system. The context of the Distributor can be large and may require it to be
3092allocated in a special area if it cannot fit in the platform's global static
3093data, for example in DRAM. The Distributor can then be powered down using an
3094implementation-defined sequence.
3095
3096plat_psci_ops.pwr_domain_pwr_down()
3097.......................................
3098
3099This is an optional function and, if implemented, is expected to perform
3100platform specific actions before the CPU is powered down. Since this function is
3101invoked outside the PSCI locks, the actions performed in this hook must be local
3102to the CPU or the platform must ensure that races between multiple CPUs cannot
3103occur.
3104
3105The ``target_state`` has a similar meaning as described in the ``pwr_domain_off()``
3106operation and it encodes the platform coordinated target local power states for
3107the CPU power domain and its parent power domain levels.
3108
3109It is preferred that this function returns. The caller will invoke
3110``psci_power_down_wfi()`` to powerdown the CPU, mitigate any powerdown errata,
3111and handle any wakeups that may arise. Previously, this function did not return
3112and instead called ``wfi`` (in an infinite loop) directly. This is still
3113possible on platforms where this is guaranteed to be terminal, however, it is
3114strongly discouraged going forward.
3115
3116Previously this function was called ``pwr_domain_pwr_down_wfi()``.
3117
3118plat_psci_ops.pwr_domain_on_finish()
3119....................................
3120
3121This function is called by the PSCI implementation after the calling CPU is
3122powered on and released from reset in response to an earlier PSCI ``CPU_ON`` call.
3123It performs the platform-specific setup required to initialize enough state for
3124this CPU to enter the normal world and also provide secure runtime firmware
3125services.
3126
3127The ``target_state`` (first argument) is the prior state of the power domains
3128immediately before the CPU was turned on. It indicates which power domains
3129above the CPU might require initialization due to having previously been in
3130low power states. The generic code expects the handler to succeed.
3131
3132plat_psci_ops.pwr_domain_on_finish_late() [optional]
3133...........................................................
3134
3135This optional function is called by the PSCI implementation after the calling
3136CPU is fully powered on with respective data caches enabled. The calling CPU and
3137the associated cluster are guaranteed to be participating in coherency. This
3138function gives the flexibility to perform any platform-specific actions safely,
3139such as initialization or modification of shared data structures, without the
3140overhead of explicit cache maintainace operations.
3141
3142The ``target_state`` has a similar meaning as described in the ``pwr_domain_on_finish()``
3143operation. The generic code expects the handler to succeed.
3144
3145plat_psci_ops.pwr_domain_suspend_finish()
3146.........................................
3147
3148This function is called by the PSCI implementation after the calling CPU is
3149powered on and released from reset in response to an asynchronous wakeup
3150event, for example a timer interrupt that was programmed by the CPU during the
3151``CPU_SUSPEND`` call or ``SYSTEM_SUSPEND`` call. It performs the platform-specific
3152setup required to restore the saved state for this CPU to resume execution
3153in the normal world and also provide secure runtime firmware services.
3154
3155The ``target_state`` (first argument) has a similar meaning as described in
3156the ``pwr_domain_on_finish()`` operation. The generic code expects the platform
3157to succeed.
3158
3159If the Distributor, Redistributors or ITS have been powered off as part of a
3160suspend, their context must be restored in this function in the reverse order
3161to how they were saved during suspend sequence.
3162
3163plat_psci_ops.system_off()
3164..........................
3165
3166This function is called by PSCI implementation in response to a ``SYSTEM_OFF``
3167call. It performs the platform-specific system poweroff sequence after
3168notifying the Secure Payload Dispatcher. The caller will call ``wfi`` if this
3169function returns, similar to `plat_psci_ops.pwr_domain_pwr_down()`_.
3170
3171plat_psci_ops.system_reset()
3172............................
3173
3174This function is called by PSCI implementation in response to a ``SYSTEM_RESET``
3175call. It performs the platform-specific system reset sequence after
3176notifying the Secure Payload Dispatcher. The caller will call ``wfi`` if this
3177function returns, similar to `plat_psci_ops.pwr_domain_pwr_down()`_.
3178
3179plat_psci_ops.validate_power_state()
3180....................................
3181
3182This function is called by the PSCI implementation during the ``CPU_SUSPEND``
3183call to validate the ``power_state`` parameter of the PSCI API and if valid,
3184populate it in ``req_state`` (second argument) array as power domain level
3185specific local states. If the ``power_state`` is invalid, the platform must
3186return PSCI_E_INVALID_PARAMS as error, which is propagated back to the
3187normal world PSCI client.
3188
3189plat_psci_ops.validate_ns_entrypoint()
3190......................................
3191
3192This function is called by the PSCI implementation during the ``CPU_SUSPEND``,
3193``SYSTEM_SUSPEND`` and ``CPU_ON`` calls to validate the non-secure ``entry_point``
3194parameter passed by the normal world. If the ``entry_point`` is invalid,
3195the platform must return PSCI_E_INVALID_ADDRESS as error, which is
3196propagated back to the normal world PSCI client.
3197
3198plat_psci_ops.get_sys_suspend_power_state()
3199...........................................
3200
3201This function is called by the PSCI implementation during the ``SYSTEM_SUSPEND``
3202call to get the ``req_state`` parameter from platform which encodes the power
3203domain level specific local states to suspend to system affinity level. The
3204``req_state`` will be utilized to do the PSCI state coordination and
3205``pwr_domain_suspend()`` will be invoked with the coordinated target state to
3206enter system suspend.
3207
3208plat_psci_ops.get_pwr_lvl_state_idx()
3209.....................................
3210
3211This is an optional function and, if implemented, is invoked by the PSCI
3212implementation to convert the ``local_state`` (first argument) at a specified
3213``pwr_lvl`` (second argument) to an index between 0 and
3214``PLAT_MAX_PWR_LVL_STATES`` - 1. This function is only needed if the platform
3215supports more than two local power states at each power domain level, that is
3216``PLAT_MAX_PWR_LVL_STATES`` is greater than 2, and needs to account for these
3217local power states.
3218
3219plat_psci_ops.translate_power_state_by_mpidr()
3220..............................................
3221
3222This is an optional function and, if implemented, verifies the ``power_state``
3223(second argument) parameter of the PSCI API corresponding to a target power
3224domain. The target power domain is identified by using both ``MPIDR`` (first
3225argument) and the power domain level encoded in ``power_state``. The power domain
3226level specific local states are to be extracted from ``power_state`` and be
3227populated in the ``output_state`` (third argument) array. The functionality
3228is similar to the ``validate_power_state`` function described above and is
3229envisaged to be used in case the validity of ``power_state`` depend on the
3230targeted power domain. If the ``power_state`` is invalid for the targeted power
3231domain, the platform must return PSCI_E_INVALID_PARAMS as error. If this
3232function is not implemented, then the generic implementation relies on
3233``validate_power_state`` function to translate the ``power_state``.
3234
3235This function can also be used in case the platform wants to support local
3236power state encoding for ``power_state`` parameter of PSCI_STAT_COUNT/RESIDENCY
3237APIs as described in Section 5.18 of `PSCI`_.
3238
3239plat_psci_ops.get_node_hw_state()
3240.................................
3241
3242This is an optional function. If implemented this function is intended to return
3243the power state of a node (identified by the first parameter, the ``MPIDR``) in
3244the power domain topology (identified by the second parameter, ``power_level``),
3245as retrieved from a power controller or equivalent component on the platform.
3246Upon successful completion, the implementation must map and return the final
3247status among ``HW_ON``, ``HW_OFF`` or ``HW_STANDBY``. Upon encountering failures, it
3248must return either ``PSCI_E_INVALID_PARAMS`` or ``PSCI_E_NOT_SUPPORTED`` as
3249appropriate.
3250
3251Implementations are not expected to handle ``power_levels`` greater than
3252``PLAT_MAX_PWR_LVL``.
3253
3254plat_psci_ops.system_reset2()
3255.............................
3256
3257This is an optional function. If implemented this function is
3258called during the ``SYSTEM_RESET2`` call to perform a reset
3259based on the first parameter ``reset_type`` as specified in
3260`PSCI`_. The parameter ``cookie`` can be used to pass additional
3261reset information. If the ``reset_type`` is not supported, the
3262function must return ``PSCI_E_NOT_SUPPORTED``. For architectural
3263resets, all failures must return ``PSCI_E_INVALID_PARAMETERS``
3264and vendor reset can return other PSCI error codes as defined
3265in `PSCI`_. If this function returns success, the caller will call
3266``wfi`` similar to `plat_psci_ops.pwr_domain_pwr_down()`_.
3267
3268plat_psci_ops.write_mem_protect()
3269.................................
3270
3271This is an optional function. If implemented it enables or disables the
3272``MEM_PROTECT`` functionality based on the value of ``val``.
3273A non-zero value enables ``MEM_PROTECT`` and a value of zero
3274disables it. Upon encountering failures it must return a negative value
3275and on success it must return 0.
3276
3277plat_psci_ops.read_mem_protect()
3278................................
3279
3280This is an optional function. If implemented it returns the current
3281state of ``MEM_PROTECT`` via the ``val`` parameter.  Upon encountering
3282failures it must return a negative value and on success it must
3283return 0.
3284
3285plat_psci_ops.mem_protect_chk()
3286...............................
3287
3288This is an optional function. If implemented it checks if a memory
3289region defined by a base address ``base`` and with a size of ``length``
3290bytes is protected by ``MEM_PROTECT``.  If the region is protected
3291then it must return 0, otherwise it must return a negative number.
3292
3293.. _porting_guide_imf_in_bl31:
3294
3295Interrupt Management framework (in BL31)
3296----------------------------------------
3297
3298BL31 implements an Interrupt Management Framework (IMF) to manage interrupts
3299generated in either security state and targeted to EL1 or EL2 in the non-secure
3300state or EL3/S-EL1 in the secure state. The design of this framework is
3301described in the :ref:`Interrupt Management Framework`
3302
3303A platform should export the following APIs to support the IMF. The following
3304text briefly describes each API and its implementation in Arm standard
3305platforms. The API implementation depends upon the type of interrupt controller
3306present in the platform. Arm standard platform layer supports both
3307`Arm Generic Interrupt Controller version 2.0 (GICv2)`_
3308and `3.0 (GICv3)`_. Juno builds the Arm platform layer to use GICv2 and the
3309FVP can be configured to use either GICv2 or GICv3 depending on the build flag
3310``FVP_USE_GIC_DRIVER`` (See :ref:`build_options_arm_fvp_platform` for more
3311details).
3312
3313See also: :ref:`Interrupt Controller Abstraction APIs<Platform Interrupt Controller API>`.
3314
3315Function : plat_interrupt_type_to_line() [mandatory]
3316~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3317
3318::
3319
3320    Argument : uint32_t, uint32_t
3321    Return   : uint32_t
3322
3323The Arm processor signals an interrupt exception either through the IRQ or FIQ
3324interrupt line. The specific line that is signaled depends on how the interrupt
3325controller (IC) reports different interrupt types from an execution context in
3326either security state. The IMF uses this API to determine which interrupt line
3327the platform IC uses to signal each type of interrupt supported by the framework
3328from a given security state. This API must be invoked at EL3.
3329
3330The first parameter will be one of the ``INTR_TYPE_*`` values (see
3331:ref:`Interrupt Management Framework`) indicating the target type of the
3332interrupt, the second parameter is the security state of the originating
3333execution context. The return result is the bit position in the ``SCR_EL3``
3334register of the respective interrupt trap: IRQ=1, FIQ=2.
3335
3336In the case of Arm standard platforms using GICv2, S-EL1 interrupts are
3337configured as FIQs and Non-secure interrupts as IRQs from either security
3338state.
3339
3340In the case of Arm standard platforms using GICv3, the interrupt line to be
3341configured depends on the security state of the execution context when the
3342interrupt is signalled and are as follows:
3343
3344-  The S-EL1 interrupts are signaled as IRQ in S-EL0/1 context and as FIQ in
3345   NS-EL0/1/2 context.
3346-  The Non secure interrupts are signaled as FIQ in S-EL0/1 context and as IRQ
3347   in the NS-EL0/1/2 context.
3348-  The EL3 interrupts are signaled as FIQ in both S-EL0/1 and NS-EL0/1/2
3349   context.
3350
3351Function : plat_ic_get_pending_interrupt_type() [mandatory]
3352~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3353
3354::
3355
3356    Argument : void
3357    Return   : uint32_t
3358
3359This API returns the type of the highest priority pending interrupt at the
3360platform IC. The IMF uses the interrupt type to retrieve the corresponding
3361handler function. ``INTR_TYPE_INVAL`` is returned when there is no interrupt
3362pending. The valid interrupt types that can be returned are ``INTR_TYPE_EL3``,
3363``INTR_TYPE_S_EL1`` and ``INTR_TYPE_NS``. This API must be invoked at EL3.
3364
3365In the case of Arm standard platforms using GICv2, the *Highest Priority
3366Pending Interrupt Register* (``GICC_HPPIR``) is read to determine the id of
3367the pending interrupt. The type of interrupt depends upon the id value as
3368follows.
3369
3370#. id < 1022 is reported as a S-EL1 interrupt
3371#. id = 1022 is reported as a Non-secure interrupt.
3372#. id = 1023 is reported as an invalid interrupt type.
3373
3374In the case of Arm standard platforms using GICv3, the system register
3375``ICC_HPPIR0_EL1``, *Highest Priority Pending group 0 Interrupt Register*,
3376is read to determine the id of the pending interrupt. The type of interrupt
3377depends upon the id value as follows.
3378
3379#. id = ``PENDING_G1S_INTID`` (1020) is reported as a S-EL1 interrupt
3380#. id = ``PENDING_G1NS_INTID`` (1021) is reported as a Non-secure interrupt.
3381#. id = ``GIC_SPURIOUS_INTERRUPT`` (1023) is reported as an invalid interrupt type.
3382#. All other interrupt id's are reported as EL3 interrupt.
3383
3384Function : plat_ic_get_pending_interrupt_id() [mandatory]
3385~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3386
3387::
3388
3389    Argument : void
3390    Return   : uint32_t
3391
3392This API returns the id of the highest priority pending interrupt at the
3393platform IC. ``INTR_ID_UNAVAILABLE`` is returned when there is no interrupt
3394pending.
3395
3396In the case of Arm standard platforms using GICv2, the *Highest Priority
3397Pending Interrupt Register* (``GICC_HPPIR``) is read to determine the id of the
3398pending interrupt. The id that is returned by API depends upon the value of
3399the id read from the interrupt controller as follows.
3400
3401#. id < 1022. id is returned as is.
3402#. id = 1022. The *Aliased Highest Priority Pending Interrupt Register*
3403   (``GICC_AHPPIR``) is read to determine the id of the non-secure interrupt.
3404   This id is returned by the API.
3405#. id = 1023. ``INTR_ID_UNAVAILABLE`` is returned.
3406
3407In the case of Arm standard platforms using GICv3, if the API is invoked from
3408EL3, the system register ``ICC_HPPIR0_EL1``, *Highest Priority Pending Interrupt
3409group 0 Register*, is read to determine the id of the pending interrupt. The id
3410that is returned by API depends upon the value of the id read from the
3411interrupt controller as follows.
3412
3413#. id < ``PENDING_G1S_INTID`` (1020). id is returned as is.
3414#. id = ``PENDING_G1S_INTID`` (1020) or ``PENDING_G1NS_INTID`` (1021). The system
3415   register ``ICC_HPPIR1_EL1``, *Highest Priority Pending Interrupt group 1
3416   Register* is read to determine the id of the group 1 interrupt. This id
3417   is returned by the API as long as it is a valid interrupt id
3418#. If the id is any of the special interrupt identifiers,
3419   ``INTR_ID_UNAVAILABLE`` is returned.
3420
3421When the API invoked from S-EL1 for GICv3 systems, the id read from system
3422register ``ICC_HPPIR1_EL1``, *Highest Priority Pending group 1 Interrupt
3423Register*, is returned if is not equal to GIC_SPURIOUS_INTERRUPT (1023) else
3424``INTR_ID_UNAVAILABLE`` is returned.
3425
3426Function : plat_ic_acknowledge_interrupt() [mandatory]
3427~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3428
3429::
3430
3431    Argument : void
3432    Return   : uint32_t
3433
3434This API is used by the CPU to indicate to the platform IC that processing of
3435the highest pending interrupt has begun. It should return the raw, unmodified
3436value obtained from the interrupt controller when acknowledging an interrupt.
3437The actual interrupt number shall be extracted from this raw value using the API
3438`plat_ic_get_interrupt_id()<plat_ic_get_interrupt_id>`.
3439
3440This function in Arm standard platforms using GICv2, reads the *Interrupt
3441Acknowledge Register* (``GICC_IAR``). This changes the state of the highest
3442priority pending interrupt from pending to active in the interrupt controller.
3443It returns the value read from the ``GICC_IAR``, unmodified.
3444
3445In the case of Arm standard platforms using GICv3, if the API is invoked
3446from EL3, the function reads the system register ``ICC_IAR0_EL1``, *Interrupt
3447Acknowledge Register group 0*. If the API is invoked from S-EL1, the function
3448reads the system register ``ICC_IAR1_EL1``, *Interrupt Acknowledge Register
3449group 1*. The read changes the state of the highest pending interrupt from
3450pending to active in the interrupt controller. The value read is returned
3451unmodified.
3452
3453The TSP uses this API to start processing of the secure physical timer
3454interrupt.
3455
3456Function : plat_ic_end_of_interrupt() [mandatory]
3457~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3458
3459::
3460
3461    Argument : uint32_t
3462    Return   : void
3463
3464This API is used by the CPU to indicate to the platform IC that processing of
3465the interrupt corresponding to the id (passed as the parameter) has
3466finished. The id should be the same as the id returned by the
3467``plat_ic_acknowledge_interrupt()`` API.
3468
3469Arm standard platforms write the id to the *End of Interrupt Register*
3470(``GICC_EOIR``) in case of GICv2, and to ``ICC_EOIR0_EL1`` or ``ICC_EOIR1_EL1``
3471system register in case of GICv3 depending on where the API is invoked from,
3472EL3 or S-EL1. This deactivates the corresponding interrupt in the interrupt
3473controller.
3474
3475The TSP uses this API to finish processing of the secure physical timer
3476interrupt.
3477
3478Function : plat_ic_get_interrupt_type() [mandatory]
3479~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3480
3481::
3482
3483    Argument : uint32_t
3484    Return   : uint32_t
3485
3486This API returns the type of the interrupt id passed as the parameter.
3487``INTR_TYPE_INVAL`` is returned if the id is invalid. If the id is valid, a valid
3488interrupt type (one of ``INTR_TYPE_EL3``, ``INTR_TYPE_S_EL1`` and ``INTR_TYPE_NS``) is
3489returned depending upon how the interrupt has been configured by the platform
3490IC. This API must be invoked at EL3.
3491
3492Arm standard platforms using GICv2 configures S-EL1 interrupts as Group0 interrupts
3493and Non-secure interrupts as Group1 interrupts. It reads the group value
3494corresponding to the interrupt id from the relevant *Interrupt Group Register*
3495(``GICD_IGROUPRn``). It uses the group value to determine the type of interrupt.
3496
3497In the case of Arm standard platforms using GICv3, both the *Interrupt Group
3498Register* (``GICD_IGROUPRn``) and *Interrupt Group Modifier Register*
3499(``GICD_IGRPMODRn``) is read to figure out whether the interrupt is configured
3500as Group 0 secure interrupt, Group 1 secure interrupt or Group 1 NS interrupt.
3501
3502Registering a console
3503---------------------
3504
3505Platforms will need to implement the TF-A console framework to register and use
3506a console for visual data output in TF-A. These can be used for data output during
3507the different stages of the firmware boot process and also for debugging purposes.
3508
3509The console framework can be used to output data on to a console using a number of
3510TF-A supported UARTs. Multiple consoles can be registered at the same time with
3511different output scopes (BOOT, RUNTIME, CRASH) so that data can be displayed on
3512their respective consoles without unnecessary cluttering of a single console.
3513
3514Information for registering a console can be found in the :ref:`Console Framework` section
3515of the :ref:`System Design` documentation.
3516
3517Common helper functions
3518-----------------------
3519Function : elx_panic()
3520~~~~~~~~~~~~~~~~~~~~~~
3521
3522::
3523
3524    Argument : void
3525    Return   : void
3526
3527This API is called from assembly files when reporting a critical failure
3528that has occured in lower EL and is been trapped in EL3. This call
3529**must not** return.
3530
3531Function : el3_panic()
3532~~~~~~~~~~~~~~~~~~~~~~
3533
3534::
3535
3536    Argument : void
3537    Return   : void
3538
3539This API is called from assembly files when encountering a critical failure that
3540cannot be recovered from. This function assumes that it is invoked from a C
3541runtime environment i.e. valid stack exists. This call **must not** return.
3542
3543Function : panic()
3544~~~~~~~~~~~~~~~~~~
3545
3546::
3547
3548    Argument : void
3549    Return   : void
3550
3551This API called from C files when encountering a critical failure that cannot
3552be recovered from. This function in turn prints backtrace (if enabled) and calls
3553el3_panic(). This call **must not** return.
3554
3555Crash Reporting mechanism (in BL31)
3556-----------------------------------
3557
3558BL31 implements a crash reporting mechanism which prints the various registers
3559of the CPU to enable quick crash analysis and debugging. This mechanism relies
3560on the platform implementing ``plat_crash_console_init``,
3561``plat_crash_console_putc`` and ``plat_crash_console_flush``.
3562
3563The file ``plat/common/aarch64/crash_console_helpers.S`` contains sample
3564implementation of all of them. Platforms may include this file to their
3565makefiles in order to benefit from them. By default, they will cause the crash
3566output to be routed over the normal console infrastructure and get printed on
3567consoles configured to output in crash state. ``console_set_scope()`` can be
3568used to control whether a console is used for crash output.
3569
3570.. note::
3571   Platforms are responsible for making sure that they only mark consoles for
3572   use in the crash scope that are able to support this, i.e. that are written
3573   in assembly and conform with the register clobber rules for putc()
3574   (x0-x2, x16-x17) and flush() (x0-x3, x16-x17) crash callbacks.
3575
3576In some cases (such as debugging very early crashes that happen before the
3577normal boot console can be set up), platforms may want to control crash output
3578more explicitly. These platforms may instead provide custom implementations for
3579these. They are executed outside of a C environment and without a stack. Many
3580console drivers provide functions named ``console_xxx_core_init/putc/flush``
3581that are designed to be used by these functions. See Arm platforms (like juno)
3582for an example of this.
3583
3584Function : plat_crash_console_init [mandatory]
3585~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3586
3587::
3588
3589    Argument : void
3590    Return   : int
3591
3592This API is used by the crash reporting mechanism to initialize the crash
3593console. It must only use the general purpose registers x0 through x7 to do the
3594initialization and returns 1 on success.
3595
3596Function : plat_crash_console_putc [mandatory]
3597~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3598
3599::
3600
3601    Argument : int
3602    Return   : int
3603
3604This API is used by the crash reporting mechanism to print a character on the
3605designated crash console. It must only use general purpose registers x1 and
3606x2 to do its work. The parameter and the return value are in general purpose
3607register x0.
3608
3609Function : plat_crash_console_flush [mandatory]
3610~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3611
3612::
3613
3614    Argument : void
3615    Return   : void
3616
3617This API is used by the crash reporting mechanism to force write of all buffered
3618data on the designated crash console. It should only use general purpose
3619registers x0 through x5 to do its work.
3620
3621Function : plat_setup_early_console [optional]
3622~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3623
3624::
3625
3626    Argument : void
3627    Return   : void
3628
3629This API is used to setup the early console, it is required only if the flag
3630``EARLY_CONSOLE`` is enabled.
3631
3632.. _External Abort handling and RAS Support:
3633
3634External Abort handling and RAS Support
3635---------------------------------------
3636
3637If any cores on the platform support powerdown abandon (i.e. ``FEAT_PABANDON``
3638is set, check the "Core powerup and powerdown sequence" in their TRMs), then
3639these functions should be able to handle being called with power domains off and
3640after the powerdown ``wfi``. In other words it may run after a call to
3641``pwr_domain_suspend()`` and before a call to ``pwr_domain_suspend_finish()``
3642(and their power off counterparts).
3643
3644Should this not be desirable, or if there is no powerdown abandon support, then
3645RAS errors should be masked by writing any relevant error records in any
3646powerdown hooks to prevent deadlocks due to a RAS error after the point of no
3647return. See the core's TRM for further information.
3648
3649Function : plat_ea_handler
3650~~~~~~~~~~~~~~~~~~~~~~~~~~
3651
3652::
3653
3654    Argument : int
3655    Argument : uint64_t
3656    Argument : void *
3657    Argument : void *
3658    Argument : uint64_t
3659    Return   : void
3660
3661This function is invoked by the runtime exception handling framework for the
3662platform to handle an External Abort received at EL3. The intention of the
3663function is to attempt to resolve the cause of External Abort and return;
3664if that's not possible then an orderly shutdown of the system is initiated.
3665
3666The first parameter (``int ea_reason``) indicates the reason for External Abort.
3667Its value is one of ``ERROR_EA_*`` constants defined in ``ea_handle.h``.
3668
3669The second parameter (``uint64_t syndrome``) is the respective syndrome
3670presented to EL3 after having received the External Abort. Depending on the
3671nature of the abort (as can be inferred from the ``ea_reason`` parameter), this
3672can be the content of either ``ESR_EL3`` or ``DISR_EL1``.
3673
3674The third parameter (``void *cookie``) is unused for now. The fourth parameter
3675(``void *handle``) is a pointer to the preempted context. The fifth parameter
3676(``uint64_t flags``) indicates the preempted security state. These parameters
3677are received from the top-level exception handler.
3678
3679This function must be implemented if a platform expects Firmware First handling
3680of External Aborts.
3681
3682Function : plat_handle_uncontainable_ea
3683~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3684
3685::
3686
3687    Argument : int
3688    Argument : uint64_t
3689    Return   : void
3690
3691This function is invoked by the RAS framework when an External Abort of
3692Uncontainable type is received at EL3. Due to the critical nature of
3693Uncontainable errors, the intention of this function is to initiate orderly
3694shutdown of the system, and is not expected to return.
3695
3696This function must be implemented in assembly.
3697
3698The first and second parameters are the same as that of ``plat_ea_handler``.
3699
3700The default implementation of this function calls
3701``report_unhandled_exception``.
3702
3703Function : plat_handle_double_fault
3704~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3705
3706::
3707
3708    Argument : int
3709    Argument : uint64_t
3710    Return   : void
3711
3712This function is invoked by the RAS framework when another External Abort is
3713received at EL3 while one is already being handled. I.e., a call to
3714``plat_ea_handler`` is outstanding. Due to its critical nature, the intention of
3715this function is to initiate orderly shutdown of the system, and is not expected
3716recover or return.
3717
3718This function must be implemented in assembly.
3719
3720The first and second parameters are the same as that of ``plat_ea_handler``.
3721
3722The default implementation of this function calls
3723``report_unhandled_exception``.
3724
3725Function : plat_handle_el3_ea
3726~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3727
3728::
3729
3730    Return   : void
3731
3732This function is invoked when an External Abort is received while executing in
3733EL3. Due to its critical nature, the intention of this function is to initiate
3734orderly shutdown of the system, and is not expected recover or return.
3735
3736This function must be implemented in assembly.
3737
3738The default implementation of this function calls
3739``report_unhandled_exception``.
3740
3741Function : plat_handle_rng_trap
3742~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3743
3744::
3745
3746    Argument : uint64_t
3747    Argument : cpu_context_t *
3748    Return   : int
3749
3750This function is invoked by BL31's exception handler when there is a synchronous
3751system register trap caused by access to the RNDR or RNDRRS registers. It allows
3752platforms implementing ``FEAT_RNG_TRAP`` and enabling ``ENABLE_FEAT_RNG_TRAP`` to
3753emulate those system registers by returing back some entropy to the lower EL.
3754
3755The first parameter (``uint64_t esr_el3``) contains the content of the ESR_EL3
3756syndrome register, which encodes the instruction that was trapped. The interesting
3757information in there is the target register (``get_sysreg_iss_rt()``).
3758
3759The second parameter (``cpu_context_t *ctx``) represents the CPU state in the
3760lower exception level, at the time when the execution of the ``mrs`` instruction
3761was trapped. Its content can be changed, to put the entropy into the target
3762register.
3763
3764The return value indicates how to proceed:
3765
3766-  When returning ``TRAP_RET_UNHANDLED`` (-1), the machine will panic.
3767-  When returning ``TRAP_RET_REPEAT`` (0), the exception handler will return
3768   to the same instruction, so its execution will be repeated.
3769-  When returning ``TRAP_RET_CONTINUE`` (1), the exception handler will return
3770   to the next instruction.
3771
3772This function needs to be implemented by a platform if it enables FEAT_RNG_TRAP.
3773
3774Function : plat_handle_impdef_trap
3775~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3776
3777::
3778
3779    Argument : uint64_t
3780    Argument : cpu_context_t *
3781    Return   : int
3782
3783This function is invoked by BL31's exception handler when there is a synchronous
3784system register trap caused by access to the implementation defined registers.
3785It allows platforms enabling ``IMPDEF_SYSREG_TRAP`` to emulate those system
3786registers choosing to program bits of their choice. If using in combination with
3787``ARCH_FEATURE_AVAILABILITY``, the macros
3788{SCR,MDCR,CPTR}_PLAT_{BITS,IGNORED,FLIPPED} should be defined to report correct
3789results.
3790
3791The first parameter (``uint64_t esr_el3``) contains the content of the ESR_EL3
3792syndrome register, which encodes the instruction that was trapped.
3793
3794The second parameter (``cpu_context_t *ctx``) represents the CPU state in the
3795lower exception level, at the time when the execution of the ``mrs`` instruction
3796was trapped.
3797
3798The return value indicates how to proceed:
3799
3800-  When returning ``TRAP_RET_UNHANDLED`` (-1), the machine will panic.
3801-  When returning ``TRAP_RET_REPEAT`` (0), the exception handler will return
3802   to the same instruction, so its execution will be repeated.
3803-  When returning ``TRAP_RET_CONTINUE`` (1), the exception handler will return
3804   to the next instruction.
3805
3806This function needs to be implemented by a platform if it enables
3807IMPDEF_SYSREG_TRAP.
3808
3809Build flags
3810-----------
3811
3812There are some build flags which can be defined by the platform to control
3813inclusion or exclusion of certain BL stages from the FIP image. These flags
3814need to be defined in the platform makefile which will get included by the
3815build system.
3816
3817-  **NEED_BL33**
3818   By default, this flag is defined ``yes`` by the build system and ``BL33``
3819   build option should be supplied as a build option. The platform has the
3820   option of excluding the BL33 image in the ``fip`` image by defining this flag
3821   to ``no``. If any of the options ``EL3_PAYLOAD_BASE`` or ``PRELOADED_BL33_BASE``
3822   are used, this flag will be set to ``no`` automatically.
3823
3824-  **ARM_ARCH_MAJOR and ARM_ARCH_MINOR**
3825   By default, ARM_ARCH_MAJOR.ARM_ARCH_MINOR is set to 8.0 in ``defaults.mk``,
3826   if the platform makefile/build defines or uses the correct ARM_ARCH_MAJOR and
3827   ARM_ARCH_MINOR then mandatory Architectural features available for that Arch
3828   version will be enabled by default and any optional Arch feature supported by
3829   the Architecture and available in TF-A can be enabled from platform specific
3830   makefile. Look up to ``arch_features.mk`` for details pertaining to mandatory
3831   and optional Arch specific features.
3832
3833Platform include paths
3834----------------------
3835
3836Platforms are allowed to add more include paths to be passed to the compiler.
3837The ``PLAT_INCLUDES`` variable is used for this purpose. This is needed in
3838particular for the file ``platform_def.h``.
3839
3840Example:
3841
3842.. code:: c
3843
3844  PLAT_INCLUDES  += -Iinclude/plat/myplat/include
3845
3846C Library
3847---------
3848
3849To avoid subtle toolchain behavioral dependencies, the header files provided
3850by the compiler are not used. The software is built with the ``-nostdinc`` flag
3851to ensure no headers are included from the toolchain inadvertently. Instead the
3852required headers are included in the TF-A source tree. The library only
3853contains those C library definitions required by the local implementation. If
3854more functionality is required, the needed library functions will need to be
3855added to the local implementation.
3856
3857Some C headers have been obtained from `FreeBSD`_ and `SCC`_, while others have
3858been written specifically for TF-A. Some implementation files have been obtained
3859from `FreeBSD`_, others have been written specifically for TF-A as well. The
3860files can be found in ``include/lib/libc`` and ``lib/libc``.
3861
3862SCC can be found in http://www.simple-cc.org/. A copy of the `FreeBSD`_ sources
3863can be obtained from http://github.com/freebsd/freebsd.
3864
3865Storage abstraction layer
3866-------------------------
3867
3868In order to improve platform independence and portability a storage abstraction
3869layer is used to load data from non-volatile platform storage. Currently
3870storage access is only required by BL1 and BL2 phases and performed inside the
3871``load_image()`` function in ``bl_common.c``.
3872
3873.. uml:: resources/diagrams/plantuml/io_framework_usage_overview.puml
3874
3875It is mandatory to implement at least one storage driver. For the Arm
3876development platforms the Firmware Image Package (FIP) driver is provided as
3877the default means to load data from storage (see :ref:`firmware_design_fip`).
3878The storage layer is described in the header file
3879``include/drivers/io/io_storage.h``. The implementation of the common library is
3880in ``drivers/io/io_storage.c`` and the driver files are located in
3881``drivers/io/``.
3882
3883.. uml:: resources/diagrams/plantuml/io_arm_class_diagram.puml
3884
3885Each IO driver must provide ``io_dev_*`` structures, as described in
3886``drivers/io/io_driver.h``. These are returned via a mandatory registration
3887function that is called on platform initialization. The semi-hosting driver
3888implementation in ``io_semihosting.c`` can be used as an example.
3889
3890Each platform should register devices and their drivers via the storage
3891abstraction layer. These drivers then need to be initialized by bootloader
3892phases as required in their respective ``blx_platform_setup()`` functions.
3893
3894.. uml:: resources/diagrams/plantuml/io_dev_registration.puml
3895
3896The storage abstraction layer provides mechanisms (``io_dev_init()``) to
3897initialize storage devices before IO operations are called.
3898
3899.. uml:: resources/diagrams/plantuml/io_dev_init_and_check.puml
3900
3901The basic operations supported by the layer
3902include ``open()``, ``close()``, ``read()``, ``write()``, ``size()`` and ``seek()``.
3903Drivers do not have to implement all operations, but each platform must
3904provide at least one driver for a device capable of supporting generic
3905operations such as loading a bootloader image.
3906
3907The current implementation only allows for known images to be loaded by the
3908firmware. These images are specified by using their identifiers, as defined in
3909``include/plat/common/common_def.h`` (or a separate header file included from
3910there). The platform layer (``plat_get_image_source()``) then returns a reference
3911to a device and a driver-specific ``spec`` which will be understood by the driver
3912to allow access to the image data.
3913
3914The layer is designed in such a way that is it possible to chain drivers with
3915other drivers. For example, file-system drivers may be implemented on top of
3916physical block devices, both represented by IO devices with corresponding
3917drivers. In such a case, the file-system "binding" with the block device may
3918be deferred until the file-system device is initialised.
3919
3920The abstraction currently depends on structures being statically allocated
3921by the drivers and callers, as the system does not yet provide a means of
3922dynamically allocating memory. This may also have the affect of limiting the
3923amount of open resources per driver.
3924
3925Measured Boot Platform Interface
3926--------------------------------
3927
3928Enabling the MEASURED_BOOT flag adds extra platform requirements. Please refer
3929to :ref:`Measured Boot Design` for more details.
3930
3931Live Firmware Activation Interface
3932----------------------------------
3933
3934Function : plat_lfa_get_components()
3935~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3936
3937::
3938
3939    Argument : plat_lfa_component_info_t **
3940    Return   : int
3941
3942This platform API provides the list of LFA components available for activation.
3943It populates a pointer to an array of ``plat_lfa_component_info_t`` structures,
3944which contain information about each component (like UUID, ID, etc.). It returns
39450 on success, or a standard error code on failure.
3946
3947Function : is_plat_lfa_activation_pending()
3948~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3949
3950::
3951
3952    Argument : uint32_t
3953    Return   : bool
3954
3955This platform API checks if the specified LFA component, identified
3956by its ``lfa_component_id``, is available for activation. It returns
3957true if available, otherwise false.
3958
3959Function : plat_lfa_cancel()
3960~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3961
3962::
3963
3964    Argument : uint32_t
3965    Return   : int
3966
3967This platform API allows the platform to cancel an ongoing update or activation
3968process for the specified ``lfa_component_id``. It returns 0 on success or
3969a standard error code on failure.
3970
3971Function : plat_lfa_load_auth_image()
3972~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3973
3974::
3975
3976    Argument : uint32_t
3977    Return   : int
3978
3979The platform uses this API to load, authenticate and measure the component
3980specified by ``lfa_component_id``. It should return 0 on success or appropriate
3981error codes for load/authentication failures.
3982
3983--------------
3984
3985*Copyright (c) 2013-2025, Arm Limited and Contributors. All rights reserved.*
3986
3987.. _PSCI: https://developer.arm.com/documentation/den0022/latest/
3988.. _Arm Generic Interrupt Controller version 2.0 (GICv2): http://infocenter.arm.com/help/topic/com.arm.doc.ihi0048b/index.html
3989.. _3.0 (GICv3): http://infocenter.arm.com/help/topic/com.arm.doc.ihi0069b/index.html
3990.. _FreeBSD: https://www.freebsd.org
3991.. _SCC: http://www.simple-cc.org/
3992.. _DRTM: https://developer.arm.com/documentation/den0113/a
3993