xref: /rk3399_ARM-atf/docs/plat/arm/fvp/index.rst (revision 831b0e9824e6c7cb07308830c12977acb79156c7)
1Arm Fixed Virtual Platforms (FVP)
2=================================
3
4Fixed Virtual Platform (FVP) Support
5------------------------------------
6
7This section lists the supported Arm |FVP| platforms. Please refer to the FVP
8documentation for a detailed description of the model parameter options.
9
10The latest version of the AArch64 build of TF-A has been tested on the following
11Arm FVPs without shifted affinities, and that do not support threaded CPU cores
12(64-bit host machine only).
13
14.. note::
15   The FVP models used are Version 11.12 Build 38, unless otherwise stated.
16
17-  ``FVP_Base_AEMvA``
18-  ``FVP_Base_AEMv8A-AEMv8A``
19-  ``FVP_Base_AEMv8A-AEMv8A-AEMv8A-AEMv8A-CCN502``
20-  ``FVP_Base_RevC-2xAEMv8A``
21-  ``FVP_Base_Cortex-A32x4``
22-  ``FVP_Base_Cortex-A35x4``
23-  ``FVP_Base_Cortex-A53x4``
24-  ``FVP_Base_Cortex-A55x4+Cortex-A75x4``
25-  ``FVP_Base_Cortex-A55x4``
26-  ``FVP_Base_Cortex-A57x1-A53x1``
27-  ``FVP_Base_Cortex-A57x2-A53x4``
28-  ``FVP_Base_Cortex-A57x4-A53x4``
29-  ``FVP_Base_Cortex-A57x4``
30-  ``FVP_Base_Cortex-A65x4``
31-  ``FVP_Base_Cortex-A65AEx8``
32-  ``FVP_Base_Cortex-A72x4-A53x4``
33-  ``FVP_Base_Cortex-A72x4``
34-  ``FVP_Base_Cortex-A73x4-A53x4``
35-  ``FVP_Base_Cortex-A73x4``
36-  ``FVP_Base_Cortex-A75x4``
37-  ``FVP_Base_Cortex-A76x4``
38-  ``FVP_Base_Cortex-A76AEx4``
39-  ``FVP_Base_Cortex-A76AEx8``
40-  ``FVP_Base_Cortex-A77x4``
41-  ``FVP_Base_Cortex-A78x4``
42-  ``FVP_Base_Neoverse-E1x1``
43-  ``FVP_Base_Neoverse-E1x2``
44-  ``FVP_Base_Neoverse-E1x4``
45-  ``FVP_Base_Neoverse-N1x4``
46-  ``FVP_Base_Neoverse-V1x4``
47-  ``FVP_CSS_SGI-575``     (Version 11.10 build 36)
48-  ``FVP_CSS_SGM-775``
49-  ``FVP_RD_E1_edge``      (Version 11.9 build 41)
50-  ``FVP_RD_N1_edge``      (Version 11.10 build 36)
51-  ``FVP_RD_N1_edge_dual`` (Version 11.10 build 36)
52-  ``FVP_RD_Daniel``       (Version 11.10 build 36)
53-  ``FVP_TC0``             (Version 0.0 build 6114)
54-  ``Foundation_Platform``
55
56The latest version of the AArch32 build of TF-A has been tested on the
57following Arm FVPs without shifted affinities, and that do not support threaded
58CPU cores (64-bit host machine only).
59
60-  ``FVP_Base_AEMvA``
61-  ``FVP_Base_AEMv8A-AEMv8A``
62-  ``FVP_Base_Cortex-A32x4``
63
64.. note::
65   The ``FVP_Base_RevC-2xAEMv8A`` FVP only supports shifted affinities, which
66   is not compatible with legacy GIC configurations. Therefore this FVP does not
67   support these legacy GIC configurations.
68
69The *Foundation* and *Base* FVPs can be downloaded free of charge. See the `Arm
70FVP website`_. The Cortex-A models listed above are also available to download
71from `Arm's website`_.
72
73.. note::
74   The build numbers quoted above are those reported by launching the FVP
75   with the ``--version`` parameter.
76
77.. note::
78   Linaro provides a ramdisk image in prebuilt FVP configurations and full
79   file systems that can be downloaded separately. To run an FVP with a virtio
80   file system image an additional FVP configuration option
81   ``-C bp.virtioblockdevice.image_path="<path-to>/<file-system-image>`` can be
82   used.
83
84.. note::
85   The software will not work on Version 1.0 of the Foundation FVP.
86   The commands below would report an ``unhandled argument`` error in this case.
87
88.. note::
89   FVPs can be launched with ``--cadi-server`` option such that a
90   CADI-compliant debugger (for example, Arm DS-5) can connect to and control
91   its execution.
92
93.. warning::
94   Since FVP model Version 11.0 Build 11.0.34 and Version 8.5 Build 0.8.5202
95   the internal synchronisation timings changed compared to older versions of
96   the models. The models can be launched with ``-Q 100`` option if they are
97   required to match the run time characteristics of the older versions.
98
99All the above platforms have been tested with `Linaro Release 19.06`_.
100
101.. _build_options_arm_fvp_platform:
102
103Arm FVP Platform Specific Build Options
104---------------------------------------
105
106-  ``FVP_CLUSTER_COUNT`` : Configures the cluster count to be used to
107   build the topology tree within TF-A. By default TF-A is configured for dual
108   cluster topology and this option can be used to override the default value.
109
110-  ``FVP_INTERCONNECT_DRIVER``: Selects the interconnect driver to be built. The
111   default interconnect driver depends on the value of ``FVP_CLUSTER_COUNT`` as
112   explained in the options below:
113
114   -  ``FVP_CCI`` : The CCI driver is selected. This is the default
115      if 0 < ``FVP_CLUSTER_COUNT`` <= 2.
116   -  ``FVP_CCN`` : The CCN driver is selected. This is the default
117      if ``FVP_CLUSTER_COUNT`` > 2.
118
119-  ``FVP_MAX_CPUS_PER_CLUSTER``: Sets the maximum number of CPUs implemented in
120   a single cluster.  This option defaults to 4.
121
122-  ``FVP_MAX_PE_PER_CPU``: Sets the maximum number of PEs implemented on any CPU
123   in the system. This option defaults to 1. Note that the build option
124   ``ARM_PLAT_MT`` doesn't have any effect on FVP platforms.
125
126-  ``FVP_USE_GIC_DRIVER`` : Selects the GIC driver to be built. Options:
127
128   -  ``FVP_GICV2`` : The GICv2 only driver is selected
129   -  ``FVP_GICV3`` : The GICv3 only driver is selected (default option)
130
131-  ``FVP_HW_CONFIG_DTS`` : Specify the path to the DTS file to be compiled
132   to DTB and packaged in FIP as the HW_CONFIG. See :ref:`Firmware Design` for
133   details on HW_CONFIG. By default, this is initialized to a sensible DTS
134   file in ``fdts/`` folder depending on other build options. But some cases,
135   like shifted affinity format for MPIDR, cannot be detected at build time
136   and this option is needed to specify the appropriate DTS file.
137
138-  ``FVP_HW_CONFIG`` : Specify the path to the HW_CONFIG blob to be packaged in
139   FIP. See :ref:`Firmware Design` for details on HW_CONFIG. This option is
140   similar to the ``FVP_HW_CONFIG_DTS`` option, but it directly specifies the
141   HW_CONFIG blob instead of the DTS file. This option is useful to override
142   the default HW_CONFIG selected by the build system.
143
144Booting Firmware Update images
145------------------------------
146
147When Firmware Update (FWU) is enabled there are at least 2 new images
148that have to be loaded, the Non-Secure FWU ROM (NS-BL1U), and the
149FWU FIP.
150
151The additional fip images must be loaded with:
152
153::
154
155    --data cluster0.cpu0="<path_to>/ns_bl1u.bin"@0x0beb8000	[ns_bl1u_base_address]
156    --data cluster0.cpu0="<path_to>/fwu_fip.bin"@0x08400000	[ns_bl2u_base_address]
157
158The address ns_bl1u_base_address is the value of NS_BL1U_BASE.
159In the same way, the address ns_bl2u_base_address is the value of
160NS_BL2U_BASE.
161
162Booting an EL3 payload
163----------------------
164
165The EL3 payloads boot flow requires the CPU's mailbox to be cleared at reset for
166the secondary CPUs holding pen to work properly. Unfortunately, its reset value
167is undefined on the FVP platform and the FVP platform code doesn't clear it.
168Therefore, one must modify the way the model is normally invoked in order to
169clear the mailbox at start-up.
170
171One way to do that is to create an 8-byte file containing all zero bytes using
172the following command:
173
174.. code:: shell
175
176    dd if=/dev/zero of=mailbox.dat bs=1 count=8
177
178and pre-load it into the FVP memory at the mailbox address (i.e. ``0x04000000``)
179using the following model parameters:
180
181::
182
183    --data cluster0.cpu0=mailbox.dat@0x04000000   [Base FVPs]
184    --data=mailbox.dat@0x04000000                 [Foundation FVP]
185
186To provide the model with the EL3 payload image, the following methods may be
187used:
188
189#. If the EL3 payload is able to execute in place, it may be programmed into
190   flash memory. On Base Cortex and AEM FVPs, the following model parameter
191   loads it at the base address of the NOR FLASH1 (the NOR FLASH0 is already
192   used for the FIP):
193
194   ::
195
196       -C bp.flashloader1.fname="<path-to>/<el3-payload>"
197
198   On Foundation FVP, there is no flash loader component and the EL3 payload
199   may be programmed anywhere in flash using method 3 below.
200
201#. When using the ``SPIN_ON_BL1_EXIT=1`` loading method, the following DS-5
202   command may be used to load the EL3 payload ELF image over JTAG:
203
204   ::
205
206       load <path-to>/el3-payload.elf
207
208#. The EL3 payload may be pre-loaded in volatile memory using the following
209   model parameters:
210
211   ::
212
213       --data cluster0.cpu0="<path-to>/el3-payload>"@address   [Base FVPs]
214       --data="<path-to>/<el3-payload>"@address                [Foundation FVP]
215
216   The address provided to the FVP must match the ``EL3_PAYLOAD_BASE`` address
217   used when building TF-A.
218
219Booting a preloaded kernel image (Base FVP)
220-------------------------------------------
221
222The following example uses a simplified boot flow by directly jumping from the
223TF-A to the Linux kernel, which will use a ramdisk as filesystem. This can be
224useful if both the kernel and the device tree blob (DTB) are already present in
225memory (like in FVP).
226
227For example, if the kernel is loaded at ``0x80080000`` and the DTB is loaded at
228address ``0x82000000``, the firmware can be built like this:
229
230.. code:: shell
231
232    CROSS_COMPILE=aarch64-none-elf-  \
233    make PLAT=fvp DEBUG=1             \
234    RESET_TO_BL31=1                   \
235    ARM_LINUX_KERNEL_AS_BL33=1        \
236    PRELOADED_BL33_BASE=0x80080000    \
237    ARM_PRELOADED_DTB_BASE=0x82000000 \
238    all fip
239
240Now, it is needed to modify the DTB so that the kernel knows the address of the
241ramdisk. The following script generates a patched DTB from the provided one,
242assuming that the ramdisk is loaded at address ``0x84000000``. Note that this
243script assumes that the user is using a ramdisk image prepared for U-Boot, like
244the ones provided by Linaro. If using a ramdisk without this header,the ``0x40``
245offset in ``INITRD_START`` has to be removed.
246
247.. code:: bash
248
249    #!/bin/bash
250
251    # Path to the input DTB
252    KERNEL_DTB=<path-to>/<fdt>
253    # Path to the output DTB
254    PATCHED_KERNEL_DTB=<path-to>/<patched-fdt>
255    # Base address of the ramdisk
256    INITRD_BASE=0x84000000
257    # Path to the ramdisk
258    INITRD=<path-to>/<ramdisk.img>
259
260    # Skip uboot header (64 bytes)
261    INITRD_START=$(printf "0x%x" $((${INITRD_BASE} + 0x40)) )
262    INITRD_SIZE=$(stat -Lc %s ${INITRD})
263    INITRD_END=$(printf "0x%x" $((${INITRD_BASE} + ${INITRD_SIZE})) )
264
265    CHOSEN_NODE=$(echo                                        \
266    "/ {                                                      \
267            chosen {                                          \
268                    linux,initrd-start = <${INITRD_START}>;   \
269                    linux,initrd-end = <${INITRD_END}>;       \
270            };                                                \
271    };")
272
273    echo $(dtc -O dts -I dtb ${KERNEL_DTB}) ${CHOSEN_NODE} |  \
274            dtc -O dtb -o ${PATCHED_KERNEL_DTB} -
275
276And the FVP binary can be run with the following command:
277
278.. code:: shell
279
280    <path-to>/FVP_Base_AEMv8A-AEMv8A                            \
281    -C pctl.startup=0.0.0.0                                     \
282    -C bp.secure_memory=1                                       \
283    -C cluster0.NUM_CORES=4                                     \
284    -C cluster1.NUM_CORES=4                                     \
285    -C cache_state_modelled=1                                   \
286    -C cluster0.cpu0.RVBAR=0x04001000                           \
287    -C cluster0.cpu1.RVBAR=0x04001000                           \
288    -C cluster0.cpu2.RVBAR=0x04001000                           \
289    -C cluster0.cpu3.RVBAR=0x04001000                           \
290    -C cluster1.cpu0.RVBAR=0x04001000                           \
291    -C cluster1.cpu1.RVBAR=0x04001000                           \
292    -C cluster1.cpu2.RVBAR=0x04001000                           \
293    -C cluster1.cpu3.RVBAR=0x04001000                           \
294    --data cluster0.cpu0="<path-to>/bl31.bin"@0x04001000        \
295    --data cluster0.cpu0="<path-to>/<patched-fdt>"@0x82000000   \
296    --data cluster0.cpu0="<path-to>/<kernel-binary>"@0x80080000 \
297    --data cluster0.cpu0="<path-to>/<ramdisk.img>"@0x84000000
298
299Obtaining the Flattened Device Trees
300^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
301
302Depending on the FVP configuration and Linux configuration used, different
303FDT files are required. FDT source files for the Foundation and Base FVPs can
304be found in the TF-A source directory under ``fdts/``. The Foundation FVP has
305a subset of the Base FVP components. For example, the Foundation FVP lacks
306CLCD and MMC support, and has only one CPU cluster.
307
308.. note::
309   It is not recommended to use the FDTs built along the kernel because not
310   all FDTs are available from there.
311
312The dynamic configuration capability is enabled in the firmware for FVPs.
313This means that the firmware can authenticate and load the FDT if present in
314FIP. A default FDT is packaged into FIP during the build based on
315the build configuration. This can be overridden by using the ``FVP_HW_CONFIG``
316or ``FVP_HW_CONFIG_DTS`` build options (refer to
317:ref:`build_options_arm_fvp_platform` for details on the options).
318
319-  ``fvp-base-gicv2-psci.dts``
320
321   For use with models such as the Cortex-A57-A53 Base FVPs without shifted
322   affinities and with Base memory map configuration.
323
324-  ``fvp-base-gicv2-psci-aarch32.dts``
325
326   For use with models such as the Cortex-A32 Base FVPs without shifted
327   affinities and running Linux in AArch32 state with Base memory map
328   configuration.
329
330-  ``fvp-base-gicv3-psci.dts``
331
332   For use with models such as the Cortex-A57-A53 Base FVPs without shifted
333   affinities and with Base memory map configuration and Linux GICv3 support.
334
335-  ``fvp-base-gicv3-psci-1t.dts``
336
337   For use with models such as the AEMv8-RevC Base FVP with shifted affinities,
338   single threaded CPUs, Base memory map configuration and Linux GICv3 support.
339
340-  ``fvp-base-gicv3-psci-dynamiq.dts``
341
342   For use with models as the Cortex-A55-A75 Base FVPs with shifted affinities,
343   single cluster, single threaded CPUs, Base memory map configuration and Linux
344   GICv3 support.
345
346-  ``fvp-base-gicv3-psci-aarch32.dts``
347
348   For use with models such as the Cortex-A32 Base FVPs without shifted
349   affinities and running Linux in AArch32 state with Base memory map
350   configuration and Linux GICv3 support.
351
352-  ``fvp-foundation-gicv2-psci.dts``
353
354   For use with Foundation FVP with Base memory map configuration.
355
356-  ``fvp-foundation-gicv3-psci.dts``
357
358   (Default) For use with Foundation FVP with Base memory map configuration
359   and Linux GICv3 support.
360
361
362Running on the Foundation FVP with reset to BL1 entrypoint
363^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
364
365The following ``Foundation_Platform`` parameters should be used to boot Linux with
3664 CPUs using the AArch64 build of TF-A.
367
368.. code:: shell
369
370    <path-to>/Foundation_Platform                   \
371    --cores=4                                       \
372    --arm-v8.0                                      \
373    --secure-memory                                 \
374    --visualization                                 \
375    --gicv3                                         \
376    --data="<path-to>/<bl1-binary>"@0x0             \
377    --data="<path-to>/<FIP-binary>"@0x08000000      \
378    --data="<path-to>/<kernel-binary>"@0x80080000   \
379    --data="<path-to>/<ramdisk-binary>"@0x84000000
380
381Notes:
382
383-  BL1 is loaded at the start of the Trusted ROM.
384-  The Firmware Image Package is loaded at the start of NOR FLASH0.
385-  The firmware loads the FDT packaged in FIP to the DRAM. The FDT load address
386   is specified via the ``hw_config_addr`` property in `TB_FW_CONFIG for FVP`_.
387-  The default use-case for the Foundation FVP is to use the ``--gicv3`` option
388   and enable the GICv3 device in the model. Note that without this option,
389   the Foundation FVP defaults to legacy (Versatile Express) memory map which
390   is not supported by TF-A.
391-  In order for TF-A to run correctly on the Foundation FVP, the architecture
392   versions must match. The Foundation FVP defaults to the highest v8.x
393   version it supports but the default build for TF-A is for v8.0. To avoid
394   issues either start the Foundation FVP to use v8.0 architecture using the
395   ``--arm-v8.0`` option, or build TF-A with an appropriate value for
396   ``ARM_ARCH_MINOR``.
397
398Running on the AEMv8 Base FVP with reset to BL1 entrypoint
399^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
400
401The following ``FVP_Base_RevC-2xAEMv8A`` parameters should be used to boot Linux
402with 8 CPUs using the AArch64 build of TF-A.
403
404.. code:: shell
405
406    <path-to>/FVP_Base_RevC-2xAEMv8A                            \
407    -C pctl.startup=0.0.0.0                                     \
408    -C bp.secure_memory=1                                       \
409    -C bp.tzc_400.diagnostics=1                                 \
410    -C cluster0.NUM_CORES=4                                     \
411    -C cluster1.NUM_CORES=4                                     \
412    -C cache_state_modelled=1                                   \
413    -C bp.secureflashloader.fname="<path-to>/<bl1-binary>"      \
414    -C bp.flashloader0.fname="<path-to>/<FIP-binary>"           \
415    --data cluster0.cpu0="<path-to>/<kernel-binary>"@0x80080000 \
416    --data cluster0.cpu0="<path-to>/<ramdisk>"@0x84000000
417
418.. note::
419   The ``FVP_Base_RevC-2xAEMv8A`` has shifted affinities and requires
420   a specific DTS for all the CPUs to be loaded.
421
422Running on the AEMv8 Base FVP (AArch32) with reset to BL1 entrypoint
423^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
424
425The following ``FVP_Base_AEMv8A-AEMv8A`` parameters should be used to boot Linux
426with 8 CPUs using the AArch32 build of TF-A.
427
428.. code:: shell
429
430    <path-to>/FVP_Base_AEMv8A-AEMv8A                            \
431    -C pctl.startup=0.0.0.0                                     \
432    -C bp.secure_memory=1                                       \
433    -C bp.tzc_400.diagnostics=1                                 \
434    -C cluster0.NUM_CORES=4                                     \
435    -C cluster1.NUM_CORES=4                                     \
436    -C cache_state_modelled=1                                   \
437    -C cluster0.cpu0.CONFIG64=0                                 \
438    -C cluster0.cpu1.CONFIG64=0                                 \
439    -C cluster0.cpu2.CONFIG64=0                                 \
440    -C cluster0.cpu3.CONFIG64=0                                 \
441    -C cluster1.cpu0.CONFIG64=0                                 \
442    -C cluster1.cpu1.CONFIG64=0                                 \
443    -C cluster1.cpu2.CONFIG64=0                                 \
444    -C cluster1.cpu3.CONFIG64=0                                 \
445    -C bp.secureflashloader.fname="<path-to>/<bl1-binary>"      \
446    -C bp.flashloader0.fname="<path-to>/<FIP-binary>"           \
447    --data cluster0.cpu0="<path-to>/<kernel-binary>"@0x80080000 \
448    --data cluster0.cpu0="<path-to>/<ramdisk>"@0x84000000
449
450Running on the Cortex-A57-A53 Base FVP with reset to BL1 entrypoint
451^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
452
453The following ``FVP_Base_Cortex-A57x4-A53x4`` model parameters should be used to
454boot Linux with 8 CPUs using the AArch64 build of TF-A.
455
456.. code:: shell
457
458    <path-to>/FVP_Base_Cortex-A57x4-A53x4                       \
459    -C pctl.startup=0.0.0.0                                     \
460    -C bp.secure_memory=1                                       \
461    -C bp.tzc_400.diagnostics=1                                 \
462    -C cache_state_modelled=1                                   \
463    -C bp.secureflashloader.fname="<path-to>/<bl1-binary>"      \
464    -C bp.flashloader0.fname="<path-to>/<FIP-binary>"           \
465    --data cluster0.cpu0="<path-to>/<kernel-binary>"@0x80080000 \
466    --data cluster0.cpu0="<path-to>/<ramdisk>"@0x84000000
467
468Running on the Cortex-A32 Base FVP (AArch32) with reset to BL1 entrypoint
469^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
470
471The following ``FVP_Base_Cortex-A32x4`` model parameters should be used to
472boot Linux with 4 CPUs using the AArch32 build of TF-A.
473
474.. code:: shell
475
476    <path-to>/FVP_Base_Cortex-A32x4                             \
477    -C pctl.startup=0.0.0.0                                     \
478    -C bp.secure_memory=1                                       \
479    -C bp.tzc_400.diagnostics=1                                 \
480    -C cache_state_modelled=1                                   \
481    -C bp.secureflashloader.fname="<path-to>/<bl1-binary>"      \
482    -C bp.flashloader0.fname="<path-to>/<FIP-binary>"           \
483    --data cluster0.cpu0="<path-to>/<kernel-binary>"@0x80080000 \
484    --data cluster0.cpu0="<path-to>/<ramdisk>"@0x84000000
485
486
487Running on the AEMv8 Base FVP with reset to BL31 entrypoint
488^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
489
490The following ``FVP_Base_RevC-2xAEMv8A`` parameters should be used to boot Linux
491with 8 CPUs using the AArch64 build of TF-A.
492
493.. code:: shell
494
495    <path-to>/FVP_Base_RevC-2xAEMv8A                             \
496    -C pctl.startup=0.0.0.0                                      \
497    -C bp.secure_memory=1                                        \
498    -C bp.tzc_400.diagnostics=1                                  \
499    -C cluster0.NUM_CORES=4                                      \
500    -C cluster1.NUM_CORES=4                                      \
501    -C cache_state_modelled=1                                    \
502    -C cluster0.cpu0.RVBAR=0x04010000                            \
503    -C cluster0.cpu1.RVBAR=0x04010000                            \
504    -C cluster0.cpu2.RVBAR=0x04010000                            \
505    -C cluster0.cpu3.RVBAR=0x04010000                            \
506    -C cluster1.cpu0.RVBAR=0x04010000                            \
507    -C cluster1.cpu1.RVBAR=0x04010000                            \
508    -C cluster1.cpu2.RVBAR=0x04010000                            \
509    -C cluster1.cpu3.RVBAR=0x04010000                            \
510    --data cluster0.cpu0="<path-to>/<bl31-binary>"@0x04010000    \
511    --data cluster0.cpu0="<path-to>/<bl32-binary>"@0xff000000    \
512    --data cluster0.cpu0="<path-to>/<bl33-binary>"@0x88000000    \
513    --data cluster0.cpu0="<path-to>/<fdt>"@0x82000000            \
514    --data cluster0.cpu0="<path-to>/<kernel-binary>"@0x80080000  \
515    --data cluster0.cpu0="<path-to>/<ramdisk>"@0x84000000
516
517Notes:
518
519-  If Position Independent Executable (PIE) support is enabled for BL31
520   in this config, it can be loaded at any valid address for execution.
521
522-  Since a FIP is not loaded when using BL31 as reset entrypoint, the
523   ``--data="<path-to><bl31|bl32|bl33-binary>"@<base-address-of-binary>``
524   parameter is needed to load the individual bootloader images in memory.
525   BL32 image is only needed if BL31 has been built to expect a Secure-EL1
526   Payload. For the same reason, the FDT needs to be compiled from the DT source
527   and loaded via the ``--data cluster0.cpu0="<path-to>/<fdt>"@0x82000000``
528   parameter.
529
530-  The ``FVP_Base_RevC-2xAEMv8A`` has shifted affinities and requires a
531   specific DTS for all the CPUs to be loaded.
532
533-  The ``-C cluster<X>.cpu<Y>.RVBAR=@<base-address-of-bl31>`` parameter, where
534   X and Y are the cluster and CPU numbers respectively, is used to set the
535   reset vector for each core.
536
537-  Changing the default value of ``ARM_TSP_RAM_LOCATION`` will also require
538   changing the value of
539   ``--data="<path-to><bl32-binary>"@<base-address-of-bl32>`` to the new value of
540   ``BL32_BASE``.
541
542
543Running on the AEMv8 Base FVP (AArch32) with reset to SP_MIN entrypoint
544^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
545
546The following ``FVP_Base_AEMv8A-AEMv8A`` parameters should be used to boot Linux
547with 8 CPUs using the AArch32 build of TF-A.
548
549.. code:: shell
550
551    <path-to>/FVP_Base_AEMv8A-AEMv8A                             \
552    -C pctl.startup=0.0.0.0                                      \
553    -C bp.secure_memory=1                                        \
554    -C bp.tzc_400.diagnostics=1                                  \
555    -C cluster0.NUM_CORES=4                                      \
556    -C cluster1.NUM_CORES=4                                      \
557    -C cache_state_modelled=1                                    \
558    -C cluster0.cpu0.CONFIG64=0                                  \
559    -C cluster0.cpu1.CONFIG64=0                                  \
560    -C cluster0.cpu2.CONFIG64=0                                  \
561    -C cluster0.cpu3.CONFIG64=0                                  \
562    -C cluster1.cpu0.CONFIG64=0                                  \
563    -C cluster1.cpu1.CONFIG64=0                                  \
564    -C cluster1.cpu2.CONFIG64=0                                  \
565    -C cluster1.cpu3.CONFIG64=0                                  \
566    -C cluster0.cpu0.RVBAR=0x04002000                            \
567    -C cluster0.cpu1.RVBAR=0x04002000                            \
568    -C cluster0.cpu2.RVBAR=0x04002000                            \
569    -C cluster0.cpu3.RVBAR=0x04002000                            \
570    -C cluster1.cpu0.RVBAR=0x04002000                            \
571    -C cluster1.cpu1.RVBAR=0x04002000                            \
572    -C cluster1.cpu2.RVBAR=0x04002000                            \
573    -C cluster1.cpu3.RVBAR=0x04002000                            \
574    --data cluster0.cpu0="<path-to>/<bl32-binary>"@0x04002000    \
575    --data cluster0.cpu0="<path-to>/<bl33-binary>"@0x88000000    \
576    --data cluster0.cpu0="<path-to>/<fdt>"@0x82000000            \
577    --data cluster0.cpu0="<path-to>/<kernel-binary>"@0x80080000  \
578    --data cluster0.cpu0="<path-to>/<ramdisk>"@0x84000000
579
580.. note::
581   The load address of ``<bl32-binary>`` depends on the value ``BL32_BASE``.
582   It should match the address programmed into the RVBAR register as well.
583
584Running on the Cortex-A57-A53 Base FVP with reset to BL31 entrypoint
585^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
586
587The following ``FVP_Base_Cortex-A57x4-A53x4`` model parameters should be used to
588boot Linux with 8 CPUs using the AArch64 build of TF-A.
589
590.. code:: shell
591
592    <path-to>/FVP_Base_Cortex-A57x4-A53x4                        \
593    -C pctl.startup=0.0.0.0                                      \
594    -C bp.secure_memory=1                                        \
595    -C bp.tzc_400.diagnostics=1                                  \
596    -C cache_state_modelled=1                                    \
597    -C cluster0.cpu0.RVBARADDR=0x04010000                        \
598    -C cluster0.cpu1.RVBARADDR=0x04010000                        \
599    -C cluster0.cpu2.RVBARADDR=0x04010000                        \
600    -C cluster0.cpu3.RVBARADDR=0x04010000                        \
601    -C cluster1.cpu0.RVBARADDR=0x04010000                        \
602    -C cluster1.cpu1.RVBARADDR=0x04010000                        \
603    -C cluster1.cpu2.RVBARADDR=0x04010000                        \
604    -C cluster1.cpu3.RVBARADDR=0x04010000                        \
605    --data cluster0.cpu0="<path-to>/<bl31-binary>"@0x04010000    \
606    --data cluster0.cpu0="<path-to>/<bl32-binary>"@0xff000000    \
607    --data cluster0.cpu0="<path-to>/<bl33-binary>"@0x88000000    \
608    --data cluster0.cpu0="<path-to>/<fdt>"@0x82000000            \
609    --data cluster0.cpu0="<path-to>/<kernel-binary>"@0x80080000  \
610    --data cluster0.cpu0="<path-to>/<ramdisk>"@0x84000000
611
612Running on the Cortex-A32 Base FVP (AArch32) with reset to SP_MIN entrypoint
613^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
614
615The following ``FVP_Base_Cortex-A32x4`` model parameters should be used to
616boot Linux with 4 CPUs using the AArch32 build of TF-A.
617
618.. code:: shell
619
620    <path-to>/FVP_Base_Cortex-A32x4                             \
621    -C pctl.startup=0.0.0.0                                     \
622    -C bp.secure_memory=1                                       \
623    -C bp.tzc_400.diagnostics=1                                 \
624    -C cache_state_modelled=1                                   \
625    -C cluster0.cpu0.RVBARADDR=0x04002000                       \
626    -C cluster0.cpu1.RVBARADDR=0x04002000                       \
627    -C cluster0.cpu2.RVBARADDR=0x04002000                       \
628    -C cluster0.cpu3.RVBARADDR=0x04002000                       \
629    --data cluster0.cpu0="<path-to>/<bl32-binary>"@0x04002000   \
630    --data cluster0.cpu0="<path-to>/<bl33-binary>"@0x88000000   \
631    --data cluster0.cpu0="<path-to>/<fdt>"@0x82000000           \
632    --data cluster0.cpu0="<path-to>/<kernel-binary>"@0x80080000 \
633    --data cluster0.cpu0="<path-to>/<ramdisk>"@0x84000000
634
635--------------
636
637*Copyright (c) 2019-2020, Arm Limited. All rights reserved.*
638
639.. _TB_FW_CONFIG for FVP: https://git.trustedfirmware.org/TF-A/trusted-firmware-a.git/tree/plat/arm/board/fvp/fdts/fvp_tb_fw_config.dts
640.. _Arm's website: `FVP models`_
641.. _FVP models: https://developer.arm.com/products/system-design/fixed-virtual-platforms
642.. _Linaro Release 19.06: http://releases.linaro.org/members/arm/platforms/19.06
643.. _Arm FVP website: https://developer.arm.com/products/system-design/fixed-virtual-platforms
644