1 Booting the Linux/ppc kernel without Open Firmware
2 --------------------------------------------------
5 (c) 2005 Benjamin Herrenschmidt <benh at kernel.crashing.org>,
7 (c) 2005 Becky Bruce <becky.bruce at freescale.com>,
8 Freescale Semiconductor, FSL SOC and 32-bit additions
10 May 18, 2005: Rev 0.1 - Initial draft, no chapter III yet.
12 May 19, 2005: Rev 0.2 - Add chapter III and bits & pieces here or
13 clarifies the fact that a lot of things are
14 optional, the kernel only requires a very
15 small device tree, though it is encouraged
16 to provide an as complete one as possible.
18 May 24, 2005: Rev 0.3 - Precise that DT block has to be in RAM
20 - Define version 3 and new format version 16
21 for the DT block (version 16 needs kernel
22 patches, will be fwd separately).
23 String block now has a size, and full path
24 is replaced by unit name for more
26 linux,phandle is made optional, only nodes
27 that are referenced by other nodes need it.
28 "name" property is now automatically
29 deduced from the unit name
31 June 1, 2005: Rev 0.4 - Correct confusion between OF_DT_END and
32 OF_DT_END_NODE in structure definition.
33 - Change version 16 format to always align
34 property data to 4 bytes. Since tokens are
35 already aligned, that means no specific
36 required alignement between property size
37 and property data. The old style variable
38 alignment would make it impossible to do
39 "simple" insertion of properties using
40 memove (thanks Milton for
41 noticing). Updated kernel patch as well
42 - Correct a few more alignement constraints
43 - Add a chapter about the device-tree
44 compiler and the textural representation of
45 the tree that can be "compiled" by dtc.
47 November 21, 2005: Rev 0.5
48 - Additions/generalizations for 32-bit
49 - Changed to reflect the new arch/powerpc
55 - Add some definitions of interrupt tree (simple/complex)
56 - Add some definitions for pci host bridges
57 - Add some common address format examples
58 - Add definitions for standard properties and "compatible"
59 names for cells that are not already defined by the existing
61 - Compare FSL SOC use of PCI to standard and make sure no new
62 node definition required.
63 - Add more information about node definitions for SOC devices
64 that currently have no standard, like the FSL CPM.
70 During the recent development of the Linux/ppc64 kernel, and more
71 specifically, the addition of new platform types outside of the old
72 IBM pSeries/iSeries pair, it was decided to enforce some strict rules
73 regarding the kernel entry and bootloader <-> kernel interfaces, in
74 order to avoid the degeneration that had become the ppc32 kernel entry
75 point and the way a new platform should be added to the kernel. The
76 legacy iSeries platform breaks those rules as it predates this scheme,
77 but no new board support will be accepted in the main tree that
78 doesn't follows them properly. In addition, since the advent of the
79 arch/powerpc merged architecture for ppc32 and ppc64, new 32-bit
80 platforms and 32-bit platforms which move into arch/powerpc will be
81 required to use these rules as well.
83 The main requirement that will be defined in more detail below is
84 the presence of a device-tree whose format is defined after Open
85 Firmware specification. However, in order to make life easier
86 to embedded board vendors, the kernel doesn't require the device-tree
87 to represent every device in the system and only requires some nodes
88 and properties to be present. This will be described in detail in
89 section III, but, for example, the kernel does not require you to
90 create a node for every PCI device in the system. It is a requirement
91 to have a node for PCI host bridges in order to provide interrupt
92 routing informations and memory/IO ranges, among others. It is also
93 recommended to define nodes for on chip devices and other busses that
94 don't specifically fit in an existing OF specification. This creates a
95 great flexibility in the way the kernel can then probe those and match
96 drivers to device, without having to hard code all sorts of tables. It
97 also makes it more flexible for board vendors to do minor hardware
98 upgrades without significantly impacting the kernel code or cluttering
99 it with special cases.
102 1) Entry point for arch/powerpc
103 -------------------------------
105 There is one and one single entry point to the kernel, at the start
106 of the kernel image. That entry point supports two calling
109 a) Boot from Open Firmware. If your firmware is compatible
110 with Open Firmware (IEEE 1275) or provides an OF compatible
111 client interface API (support for "interpret" callback of
112 forth words isn't required), you can enter the kernel with:
114 r5 : OF callback pointer as defined by IEEE 1275
115 bindings to powerpc. Only the 32 bit client interface
116 is currently supported
118 r3, r4 : address & length of an initrd if any or 0
120 The MMU is either on or off; the kernel will run the
121 trampoline located in arch/powerpc/kernel/prom_init.c to
122 extract the device-tree and other information from open
123 firmware and build a flattened device-tree as described
124 in b). prom_init() will then re-enter the kernel using
125 the second method. This trampoline code runs in the
126 context of the firmware, which is supposed to handle all
127 exceptions during that time.
129 b) Direct entry with a flattened device-tree block. This entry
130 point is called by a) after the OF trampoline and can also be
131 called directly by a bootloader that does not support the Open
132 Firmware client interface. It is also used by "kexec" to
133 implement "hot" booting of a new kernel from a previous
134 running one. This method is what I will describe in more
135 details in this document, as method a) is simply standard Open
136 Firmware, and thus should be implemented according to the
137 various standard documents defining it and its binding to the
138 PowerPC platform. The entry point definition then becomes:
140 r3 : physical pointer to the device-tree block
141 (defined in chapter II) in RAM
143 r4 : physical pointer to the kernel itself. This is
144 used by the assembly code to properly disable the MMU
145 in case you are entering the kernel with MMU enabled
146 and a non-1:1 mapping.
148 r5 : NULL (as to differentiate with method a)
150 Note about SMP entry: Either your firmware puts your other
151 CPUs in some sleep loop or spin loop in ROM where you can get
152 them out via a soft reset or some other means, in which case
153 you don't need to care, or you'll have to enter the kernel
154 with all CPUs. The way to do that with method b) will be
155 described in a later revision of this document.
163 Board supports (platforms) are not exclusive config options. An
164 arbitrary set of board supports can be built in a single kernel
165 image. The kernel will "know" what set of functions to use for a
166 given platform based on the content of the device-tree. Thus, you
169 a) add your platform support as a _boolean_ option in
170 arch/powerpc/Kconfig, following the example of PPC_PSERIES,
171 PPC_PMAC and PPC_MAPLE. The later is probably a good
172 example of a board support to start from.
174 b) create your main platform file as
175 "arch/powerpc/platforms/myplatform/myboard_setup.c" and add it
176 to the Makefile under the condition of your CONFIG_
177 option. This file will define a structure of type "ppc_md"
178 containing the various callbacks that the generic code will
179 use to get to your platform specific code
181 c) Add a reference to your "ppc_md" structure in the
182 "machines" table in arch/powerpc/kernel/setup_64.c if you are
185 d) request and get assigned a platform number (see PLATFORM_*
186 constants in include/asm-powerpc/processor.h
188 32-bit embedded kernels:
190 Currently, board support is essentially an exclusive config option.
191 The kernel is configured for a single platform. Part of the reason
192 for this is to keep kernels on embedded systems small and efficient;
193 part of this is due to the fact the code is already that way. In the
194 future, a kernel may support multiple platforms, but only if the
195 platforms feature the same core architectire. A single kernel build
196 cannot support both configurations with Book E and configurations
197 with classic Powerpc architectures.
199 32-bit embedded platforms that are moved into arch/powerpc using a
200 flattened device tree should adopt the merged tree practice of
201 setting ppc_md up dynamically, even though the kernel is currently
202 built with support for only a single platform at a time. This allows
203 unification of the setup code, and will make it easier to go to a
204 multiple-platform-support model in the future.
206 NOTE: I believe the above will be true once Ben's done with the merge
207 of the boot sequences.... someone speak up if this is wrong!
209 To add a 32-bit embedded platform support, follow the instructions
210 for 64-bit platforms above, with the exception that the Kconfig
211 option should be set up such that the kernel builds exclusively for
212 the platform selected. The processor type for the platform should
213 enable another config option to select the specific board
216 NOTE: If ben doesn't merge the setup files, may need to change this to
220 I will describe later the boot process and various callbacks that
221 your platform should implement.
224 II - The DT block format
225 ========================
228 This chapter defines the actual format of the flattened device-tree
229 passed to the kernel. The actual content of it and kernel requirements
230 are described later. You can find example of code manipulating that
231 format in various places, including arch/powerpc/kernel/prom_init.c
232 which will generate a flattened device-tree from the Open Firmware
233 representation, or the fs2dt utility which is part of the kexec tools
234 which will generate one from a filesystem representation. It is
235 expected that a bootloader like uboot provides a bit more support,
236 that will be discussed later as well.
238 Note: The block has to be in main memory. It has to be accessible in
239 both real mode and virtual mode with no mapping other than main
240 memory. If you are writing a simple flash bootloader, it should copy
241 the block to RAM before passing it to the kernel.
247 The kernel is entered with r3 pointing to an area of memory that is
248 roughly described in include/asm-powerpc/prom.h by the structure
251 struct boot_param_header {
252 u32 magic; /* magic word OF_DT_HEADER */
253 u32 totalsize; /* total size of DT block */
254 u32 off_dt_struct; /* offset to structure */
255 u32 off_dt_strings; /* offset to strings */
256 u32 off_mem_rsvmap; /* offset to memory reserve map
258 u32 version; /* format version */
259 u32 last_comp_version; /* last compatible version */
261 /* version 2 fields below */
262 u32 boot_cpuid_phys; /* Which physical CPU id we're
264 /* version 3 fields below */
265 u32 size_dt_strings; /* size of the strings block */
268 Along with the constants:
270 /* Definitions used by the flattened device tree */
271 #define OF_DT_HEADER 0xd00dfeed /* 4: version,
273 #define OF_DT_BEGIN_NODE 0x1 /* Start node: full name
275 #define OF_DT_END_NODE 0x2 /* End node */
276 #define OF_DT_PROP 0x3 /* Property: name off,
278 #define OF_DT_END 0x9
280 All values in this header are in big endian format, the various
281 fields in this header are defined more precisely below. All
282 "offset" values are in bytes from the start of the header; that is
283 from the value of r3.
287 This is a magic value that "marks" the beginning of the
288 device-tree block header. It contains the value 0xd00dfeed and is
289 defined by the constant OF_DT_HEADER
293 This is the total size of the DT block including the header. The
294 "DT" block should enclose all data structures defined in this
295 chapter (who are pointed to by offsets in this header). That is,
296 the device-tree structure, strings, and the memory reserve map.
300 This is an offset from the beginning of the header to the start
301 of the "structure" part the device tree. (see 2) device tree)
305 This is an offset from the beginning of the header to the start
306 of the "strings" part of the device-tree
310 This is an offset from the beginning of the header to the start
311 of the reserved memory map. This map is a list of pairs of 64
312 bit integers. Each pair is a physical address and a size. The
314 list is terminated by an entry of size 0. This map provides the
315 kernel with a list of physical memory areas that are "reserved"
316 and thus not to be used for memory allocations, especially during
317 early initialization. The kernel needs to allocate memory during
318 boot for things like un-flattening the device-tree, allocating an
319 MMU hash table, etc... Those allocations must be done in such a
320 way to avoid overriding critical things like, on Open Firmware
321 capable machines, the RTAS instance, or on some pSeries, the TCE
322 tables used for the iommu. Typically, the reserve map should
323 contain _at least_ this DT block itself (header,total_size). If
324 you are passing an initrd to the kernel, you should reserve it as
325 well. You do not need to reserve the kernel image itself. The map
326 should be 64 bit aligned.
330 This is the version of this structure. Version 1 stops
331 here. Version 2 adds an additional field boot_cpuid_phys.
332 Version 3 adds the size of the strings block, allowing the kernel
333 to reallocate it easily at boot and free up the unused flattened
334 structure after expansion. Version 16 introduces a new more
335 "compact" format for the tree itself that is however not backward
336 compatible. You should always generate a structure of the highest
337 version defined at the time of your implementation. Currently
338 that is version 16, unless you explicitly aim at being backward
343 Last compatible version. This indicates down to what version of
344 the DT block you are backward compatible. For example, version 2
345 is backward compatible with version 1 (that is, a kernel build
346 for version 1 will be able to boot with a version 2 format). You
347 should put a 1 in this field if you generate a device tree of
348 version 1 to 3, or 0x10 if you generate a tree of version 0x10
349 using the new unit name format.
353 This field only exist on version 2 headers. It indicate which
354 physical CPU ID is calling the kernel entry point. This is used,
355 among others, by kexec. If you are on an SMP system, this value
356 should match the content of the "reg" property of the CPU node in
357 the device-tree corresponding to the CPU calling the kernel entry
358 point (see further chapters for more informations on the required
359 device-tree contents)
362 So the typical layout of a DT block (though the various parts don't
363 need to be in that order) looks like this (addresses go from top to
367 ------------------------------
368 r3 -> | struct boot_param_header |
369 ------------------------------
370 | (alignment gap) (*) |
371 ------------------------------
372 | memory reserve map |
373 ------------------------------
375 ------------------------------
377 | device-tree structure |
379 ------------------------------
381 ------------------------------
383 | device-tree strings |
385 -----> ------------------------------
390 (*) The alignment gaps are not necessarily present; their presence
391 and size are dependent on the various alignment requirements of
392 the individual data blocks.
395 2) Device tree generalities
396 ---------------------------
398 This device-tree itself is separated in two different blocks, a
399 structure block and a strings block. Both need to be aligned to a 4
402 First, let's quickly describe the device-tree concept before detailing
403 the storage format. This chapter does _not_ describe the detail of the
404 required types of nodes & properties for the kernel, this is done
405 later in chapter III.
407 The device-tree layout is strongly inherited from the definition of
408 the Open Firmware IEEE 1275 device-tree. It's basically a tree of
409 nodes, each node having two or more named properties. A property can
412 It is a tree, so each node has one and only one parent except for the
413 root node who has no parent.
415 A node has 2 names. The actual node name is generally contained in a
416 property of type "name" in the node property list whose value is a
417 zero terminated string and is mandatory for version 1 to 3 of the
418 format definition (as it is in Open Firmware). Version 0x10 makes it
419 optional as it can generate it from the unit name defined below.
421 There is also a "unit name" that is used to differentiate nodes with
422 the same name at the same level, it is usually made of the node
423 names, the "@" sign, and a "unit address", which definition is
424 specific to the bus type the node sits on.
426 The unit name doesn't exist as a property per-se but is included in
427 the device-tree structure. It is typically used to represent "path" in
428 the device-tree. More details about the actual format of these will be
431 The kernel powerpc generic code does not make any formal use of the
432 unit address (though some board support code may do) so the only real
433 requirement here for the unit address is to ensure uniqueness of
434 the node unit name at a given level of the tree. Nodes with no notion
435 of address and no possible sibling of the same name (like /memory or
436 /cpus) may omit the unit address in the context of this specification,
437 or use the "@0" default unit address. The unit name is used to define
438 a node "full path", which is the concatenation of all parent node
439 unit names separated with "/".
441 The root node doesn't have a defined name, and isn't required to have
442 a name property either if you are using version 3 or earlier of the
443 format. It also has no unit address (no @ symbol followed by a unit
444 address). The root node unit name is thus an empty string. The full
445 path to the root node is "/".
447 Every node which actually represents an actual device (that is, a node
448 which isn't only a virtual "container" for more nodes, like "/cpus"
449 is) is also required to have a "device_type" property indicating the
452 Finally, every node that can be referenced from a property in another
453 node is required to have a "linux,phandle" property. Real open
454 firmware implementations provide a unique "phandle" value for every
455 node that the "prom_init()" trampoline code turns into
456 "linux,phandle" properties. However, this is made optional if the
457 flattened device tree is used directly. An example of a node
458 referencing another node via "phandle" is when laying out the
459 interrupt tree which will be described in a further version of this
462 This "linux, phandle" property is a 32 bit value that uniquely
463 identifies a node. You are free to use whatever values or system of
464 values, internal pointers, or whatever to generate these, the only
465 requirement is that every node for which you provide that property has
466 a unique value for it.
468 Here is an example of a simple device-tree. In this example, an "o"
469 designates a node followed by the node unit name. Properties are
470 presented with their name followed by their content. "content"
471 represents an ASCII string (zero terminated) value, while <content>
472 represents a 32 bit hexadecimal value. The various nodes in this
473 example will be discussed in a later chapter. At this point, it is
474 only meant to give you a idea of what a device-tree looks like. I have
475 purposefully kept the "name" and "linux,phandle" properties which
476 aren't necessary in order to give you a better idea of what the tree
477 looks like in practice.
480 |- name = "device-tree"
481 |- model = "MyBoardName"
482 |- compatible = "MyBoardFamilyName"
483 |- #address-cells = <2>
485 |- linux,phandle = <0>
489 | | - linux,phandle = <1>
490 | | - #address-cells = <1>
491 | | - #size-cells = <0>
494 | |- name = "PowerPC,970"
495 | |- device_type = "cpu"
497 | |- clock-frequency = <5f5e1000>
499 | |- linux,phandle = <2>
503 | |- device_type = "memory"
504 | |- reg = <00000000 00000000 00000000 20000000>
505 | |- linux,phandle = <3>
509 |- bootargs = "root=/dev/sda2"
510 |- linux,platform = <00000600>
511 |- linux,phandle = <4>
513 This tree is almost a minimal tree. It pretty much contains the
514 minimal set of required nodes and properties to boot a linux kernel;
515 that is, some basic model informations at the root, the CPUs, and the
516 physical memory layout. It also includes misc information passed
517 through /chosen, like in this example, the platform type (mandatory)
518 and the kernel command line arguments (optional).
520 The /cpus/PowerPC,970@0/linux,boot-cpu property is an example of a
521 property without a value. All other properties have a value. The
522 significance of the #address-cells and #size-cells properties will be
523 explained in chapter IV which defines precisely the required nodes and
524 properties and their content.
527 3) Device tree "structure" block
529 The structure of the device tree is a linearized tree structure. The
530 "OF_DT_BEGIN_NODE" token starts a new node, and the "OF_DT_END_NODE"
531 ends that node definition. Child nodes are simply defined before
532 "OF_DT_END_NODE" (that is nodes within the node). A 'token' is a 32
533 bit value. The tree has to be "finished" with a OF_DT_END token
535 Here's the basic structure of a single node:
537 * token OF_DT_BEGIN_NODE (that is 0x00000001)
538 * for version 1 to 3, this is the node full path as a zero
539 terminated string, starting with "/". For version 16 and later,
540 this is the node unit name only (or an empty string for the
542 * [align gap to next 4 bytes boundary]
544 * token OF_DT_PROP (that is 0x00000003)
545 * 32 bit value of property value size in bytes (or 0 of no
547 * 32 bit value of offset in string block of property name
548 * property value data if any
549 * [align gap to next 4 bytes boundary]
550 * [child nodes if any]
551 * token OF_DT_END_NODE (that is 0x00000002)
553 So the node content can be summarised as a start token, a full path,
554 a list of properties, a list of child nodes, and an end token. Every
555 child node is a full node structure itself as defined above.
557 4) Device tree "strings" block
559 In order to save space, property names, which are generally redundant,
560 are stored separately in the "strings" block. This block is simply the
561 whole bunch of zero terminated strings for all property names
562 concatenated together. The device-tree property definitions in the
563 structure block will contain offset values from the beginning of the
567 III - Required content of the device tree
568 =========================================
570 WARNING: All "linux,*" properties defined in this document apply only
571 to a flattened device-tree. If your platform uses a real
572 implementation of Open Firmware or an implementation compatible with
573 the Open Firmware client interface, those properties will be created
574 by the trampoline code in the kernel's prom_init() file. For example,
575 that's where you'll have to add code to detect your board model and
576 set the platform number. However, when using the flattened device-tree
577 entry point, there is no prom_init() pass, and thus you have to
578 provide those properties yourself.
581 1) Note about cells and address representation
582 ----------------------------------------------
584 The general rule is documented in the various Open Firmware
585 documentations. If you chose to describe a bus with the device-tree
586 and there exist an OF bus binding, then you should follow the
587 specification. However, the kernel does not require every single
588 device or bus to be described by the device tree.
590 In general, the format of an address for a device is defined by the
591 parent bus type, based on the #address-cells and #size-cells
592 property. In the absence of such a property, the parent's parent
593 values are used, etc... The kernel requires the root node to have
594 those properties defining addresses format for devices directly mapped
595 on the processor bus.
597 Those 2 properties define 'cells' for representing an address and a
598 size. A "cell" is a 32 bit number. For example, if both contain 2
599 like the example tree given above, then an address and a size are both
600 composed of 2 cells, and each is a 64 bit number (cells are
601 concatenated and expected to be in big endian format). Another example
602 is the way Apple firmware defines them, with 2 cells for an address
603 and one cell for a size. Most 32-bit implementations should define
604 #address-cells and #size-cells to 1, which represents a 32-bit value.
605 Some 32-bit processors allow for physical addresses greater than 32
606 bits; these processors should define #address-cells as 2.
608 "reg" properties are always a tuple of the type "address size" where
609 the number of cells of address and size is specified by the bus
610 #address-cells and #size-cells. When a bus supports various address
611 spaces and other flags relative to a given address allocation (like
612 prefetchable, etc...) those flags are usually added to the top level
613 bits of the physical address. For example, a PCI physical address is
614 made of 3 cells, the bottom two containing the actual address itself
615 while the top cell contains address space indication, flags, and pci
616 bus & device numbers.
618 For busses that support dynamic allocation, it's the accepted practice
619 to then not provide the address in "reg" (keep it 0) though while
620 providing a flag indicating the address is dynamically allocated, and
621 then, to provide a separate "assigned-addresses" property that
622 contains the fully allocated addresses. See the PCI OF bindings for
625 In general, a simple bus with no address space bits and no dynamic
626 allocation is preferred if it reflects your hardware, as the existing
627 kernel address parsing functions will work out of the box. If you
628 define a bus type with a more complex address format, including things
629 like address space bits, you'll have to add a bus translator to the
630 prom_parse.c file of the recent kernels for your bus type.
632 The "reg" property only defines addresses and sizes (if #size-cells
633 is non-0) within a given bus. In order to translate addresses upward
634 (that is into parent bus addresses, and possibly into cpu physical
635 addresses), all busses must contain a "ranges" property. If the
636 "ranges" property is missing at a given level, it's assumed that
637 translation isn't possible. The format of the "ranges" property for a
640 bus address, parent bus address, size
642 "bus address" is in the format of the bus this bus node is defining,
643 that is, for a PCI bridge, it would be a PCI address. Thus, (bus
644 address, size) defines a range of addresses for child devices. "parent
645 bus address" is in the format of the parent bus of this bus. For
646 example, for a PCI host controller, that would be a CPU address. For a
647 PCI<->ISA bridge, that would be a PCI address. It defines the base
648 address in the parent bus where the beginning of that range is mapped.
650 For a new 64 bit powerpc board, I recommend either the 2/2 format or
651 Apple's 2/1 format which is slightly more compact since sizes usually
652 fit in a single 32 bit word. New 32 bit powerpc boards should use a
653 1/1 format, unless the processor supports physical addresses greater
654 than 32-bits, in which case a 2/1 format is recommended.
657 2) Note about "compatible" properties
658 -------------------------------------
660 These properties are optional, but recommended in devices and the root
661 node. The format of a "compatible" property is a list of concatenated
662 zero terminated strings. They allow a device to express its
663 compatibility with a family of similar devices, in some cases,
664 allowing a single driver to match against several devices regardless
665 of their actual names.
667 3) Note about "name" properties
668 -------------------------------
670 While earlier users of Open Firmware like OldWorld macintoshes tended
671 to use the actual device name for the "name" property, it's nowadays
672 considered a good practice to use a name that is closer to the device
673 class (often equal to device_type). For example, nowadays, ethernet
674 controllers are named "ethernet", an additional "model" property
675 defining precisely the chip type/model, and "compatible" property
676 defining the family in case a single driver can driver more than one
677 of these chips. However, the kernel doesn't generally put any
678 restriction on the "name" property; it is simply considered good
679 practice to follow the standard and its evolutions as closely as
682 Note also that the new format version 16 makes the "name" property
683 optional. If it's absent for a node, then the node's unit name is then
684 used to reconstruct the name. That is, the part of the unit name
685 before the "@" sign is used (or the entire unit name if no "@" sign
688 4) Note about node and property names and character set
689 -------------------------------------------------------
691 While open firmware provides more flexible usage of 8859-1, this
692 specification enforces more strict rules. Nodes and properties should
693 be comprised only of ASCII characters 'a' to 'z', '0' to
694 '9', ',', '.', '_', '+', '#', '?', and '-'. Node names additionally
695 allow uppercase characters 'A' to 'Z' (property names should be
696 lowercase. The fact that vendors like Apple don't respect this rule is
697 irrelevant here). Additionally, node and property names should always
698 begin with a character in the range 'a' to 'z' (or 'A' to 'Z' for node
701 The maximum number of characters for both nodes and property names
702 is 31. In the case of node names, this is only the leftmost part of
703 a unit name (the pure "name" property), it doesn't include the unit
704 address which can extend beyond that limit.
707 5) Required nodes and properties
708 --------------------------------
709 These are all that are currently required. However, it is strongly
710 recommended that you expose PCI host bridges as documented in the
711 PCI binding to open firmware, and your interrupt tree as documented
712 in OF interrupt tree specification.
716 The root node requires some properties to be present:
718 - model : this is your board name/model
719 - #address-cells : address representation for "root" devices
720 - #size-cells: the size representation for "root" devices
721 - device_type : This property shouldn't be necessary. However, if
722 you decide to create a device_type for your root node, make sure it
723 is _not_ "chrp" unless your platform is a pSeries or PAPR compliant
724 one for 64-bit, or a CHRP-type machine for 32-bit as this will
725 matched by the kernel this way.
727 Additionally, some recommended properties are:
729 - compatible : the board "family" generally finds its way here,
730 for example, if you have 2 board models with a similar layout,
731 that typically get driven by the same platform code in the
732 kernel, you would use a different "model" property but put a
733 value in "compatible". The kernel doesn't directly use that
734 value (see /chosen/linux,platform for how the kernel chooses a
735 platform type) but it is generally useful.
737 The root node is also generally where you add additional properties
738 specific to your board like the serial number if any, that sort of
739 thing. It is recommended that if you add any "custom" property whose
740 name may clash with standard defined ones, you prefix them with your
741 vendor name and a comma.
745 This node is the parent of all individual CPU nodes. It doesn't
746 have any specific requirements, though it's generally good practice
749 #address-cells = <00000001>
750 #size-cells = <00000000>
752 This defines that the "address" for a CPU is a single cell, and has
753 no meaningful size. This is not necessary but the kernel will assume
754 that format when reading the "reg" properties of a CPU node, see
759 So under /cpus, you are supposed to create a node for every CPU on
760 the machine. There is no specific restriction on the name of the
761 CPU, though It's common practice to call it PowerPC,<name>. For
762 example, Apple uses PowerPC,G5 while IBM uses PowerPC,970FX.
766 - device_type : has to be "cpu"
767 - reg : This is the physical cpu number, it's a single 32 bit cell
768 and is also used as-is as the unit number for constructing the
769 unit name in the full path. For example, with 2 CPUs, you would
771 /cpus/PowerPC,970FX@0
772 /cpus/PowerPC,970FX@1
773 (unit addresses do not require leading zeroes)
774 - d-cache-line-size : one cell, L1 data cache line size in bytes
775 - i-cache-line-size : one cell, L1 instruction cache line size in
777 - d-cache-size : one cell, size of L1 data cache in bytes
778 - i-cache-size : one cell, size of L1 instruction cache in bytes
779 - linux, boot-cpu : Should be defined if this cpu is the boot cpu.
781 Recommended properties:
783 - timebase-frequency : a cell indicating the frequency of the
784 timebase in Hz. This is not directly used by the generic code,
785 but you are welcome to copy/paste the pSeries code for setting
786 the kernel timebase/decrementer calibration based on this
788 - clock-frequency : a cell indicating the CPU core clock frequency
789 in Hz. A new property will be defined for 64 bit values, but if
790 your frequency is < 4Ghz, one cell is enough. Here as well as
791 for the above, the common code doesn't use that property, but
792 you are welcome to re-use the pSeries or Maple one. A future
793 kernel version might provide a common function for this.
795 You are welcome to add any property you find relevant to your board,
796 like some information about the mechanism used to soft-reset the
797 CPUs. For example, Apple puts the GPIO number for CPU soft reset
798 lines in there as a "soft-reset" property since they start secondary
799 CPUs by soft-resetting them.
802 d) the /memory node(s)
804 To define the physical memory layout of your board, you should
805 create one or more memory node(s). You can either create a single
806 node with all memory ranges in its reg property, or you can create
807 several nodes, as you wish. The unit address (@ part) used for the
808 full path is the address of the first range of memory defined by a
809 given node. If you use a single memory node, this will typically be
814 - device_type : has to be "memory"
815 - reg : This property contains all the physical memory ranges of
816 your board. It's a list of addresses/sizes concatenated
817 together, with the number of cells of each defined by the
818 #address-cells and #size-cells of the root node. For example,
819 with both of these properties being 2 like in the example given
820 earlier, a 970 based machine with 6Gb of RAM could typically
821 have a "reg" property here that looks like:
823 00000000 00000000 00000000 80000000
824 00000001 00000000 00000001 00000000
826 That is a range starting at 0 of 0x80000000 bytes and a range
827 starting at 0x100000000 and of 0x100000000 bytes. You can see
828 that there is no memory covering the IO hole between 2Gb and
829 4Gb. Some vendors prefer splitting those ranges into smaller
830 segments, but the kernel doesn't care.
834 This node is a bit "special". Normally, that's where open firmware
835 puts some variable environment information, like the arguments, or
836 phandle pointers to nodes like the main interrupt controller, or the
837 default input/output devices.
839 This specification makes a few of these mandatory, but also defines
840 some linux-specific properties that would be normally constructed by
841 the prom_init() trampoline when booting with an OF client interface,
842 but that you have to provide yourself when using the flattened format.
846 - linux,platform : This is your platform number as assigned by the
847 architecture maintainers
849 Recommended properties:
851 - bootargs : This zero-terminated string is passed as the kernel
853 - linux,stdout-path : This is the full path to your standard
854 console device if any. Typically, if you have serial devices on
855 your board, you may want to put the full path to the one set as
856 the default console in the firmware here, for the kernel to pick
857 it up as it's own default console. If you look at the funciton
858 set_preferred_console() in arch/ppc64/kernel/setup.c, you'll see
859 that the kernel tries to find out the default console and has
860 knowledge of various types like 8250 serial ports. You may want
861 to extend this function to add your own.
862 - interrupt-controller : This is one cell containing a phandle
863 value that matches the "linux,phandle" property of your main
864 interrupt controller node. May be used for interrupt routing.
867 Note that u-boot creates and fills in the chosen node for platforms
870 f) the /soc<SOCname> node
872 This node is used to represent a system-on-a-chip (SOC) and must be
873 present if the processor is a SOC. The top-level soc node contains
874 information that is global to all devices on the SOC. The node name
875 should contain a unit address for the SOC, which is the base address
876 of the memory-mapped register set for the SOC. The name of an soc
877 node should start with "soc", and the remainder of the name should
878 represent the part number for the soc. For example, the MPC8540's
879 soc node would be called "soc8540".
883 - device_type : Should be "soc"
884 - ranges : Should be defined as specified in 1) to describe the
885 translation of SOC addresses for memory mapped SOC registers.
886 - bus-frequency: Contains the bus frequency for the SOC node.
887 Typically, the value of this field is filled in by the boot
891 Recommended properties:
893 - reg : This property defines the address and size of the
894 memory-mapped registers that are used for the SOC node itself.
895 It does not include the child device registers - these will be
896 defined inside each child node. The address specified in the
897 "reg" property should match the unit address of the SOC node.
898 - #address-cells : Address representation for "soc" devices. The
899 format of this field may vary depending on whether or not the
900 device registers are memory mapped. For memory mapped
901 registers, this field represents the number of cells needed to
902 represent the address of the registers. For SOCs that do not
903 use MMIO, a special address format should be defined that
904 contains enough cells to represent the required information.
905 See 1) above for more details on defining #address-cells.
906 - #size-cells : Size representation for "soc" devices
907 - #interrupt-cells : Defines the width of cells used to represent
908 interrupts. Typically this value is <2>, which includes a
909 32-bit number that represents the interrupt number, and a
910 32-bit number that represents the interrupt sense and level.
911 This field is only needed if the SOC contains an interrupt
914 The SOC node may contain child nodes for each SOC device that the
915 platform uses. Nodes should not be created for devices which exist
916 on the SOC but are not used by a particular platform. See chapter VI
917 for more information on how to specify devices that are part of an
920 Example SOC node for the MPC8540:
923 #address-cells = <1>;
925 #interrupt-cells = <2>;
927 ranges = <00000000 e0000000 00100000>
928 reg = <e0000000 00003000>;
934 IV - "dtc", the device tree compiler
935 ====================================
938 dtc source code can be found at
939 <http://ozlabs.org/~dgibson/dtc/dtc.tar.gz>
941 WARNING: This version is still in early development stage; the
942 resulting device-tree "blobs" have not yet been validated with the
943 kernel. The current generated bloc lacks a useful reserve map (it will
944 be fixed to generate an empty one, it's up to the bootloader to fill
945 it up) among others. The error handling needs work, bugs are lurking,
948 dtc basically takes a device-tree in a given format and outputs a
949 device-tree in another format. The currently supported formats are:
954 - "dtb": "blob" format, that is a flattened device-tree block
956 header all in a binary blob.
957 - "dts": "source" format. This is a text file containing a
958 "source" for a device-tree. The format is defined later in this
960 - "fs" format. This is a representation equivalent to the
961 output of /proc/device-tree, that is nodes are directories and
967 - "dtb": "blob" format
968 - "dts": "source" format
969 - "asm": assembly language file. This is a file that can be
970 sourced by gas to generate a device-tree "blob". That file can
971 then simply be added to your Makefile. Additionally, the
972 assembly file exports some symbols that can be used.
975 The syntax of the dtc tool is
977 dtc [-I <input-format>] [-O <output-format>]
978 [-o output-filename] [-V output_version] input_filename
981 The "output_version" defines what versio of the "blob" format will be
982 generated. Supported versions are 1,2,3 and 16. The default is
983 currently version 3 but that may change in the future to version 16.
985 Additionally, dtc performs various sanity checks on the tree, like the
986 uniqueness of linux, phandle properties, validity of strings, etc...
988 The format of the .dts "source" file is "C" like, supports C and C++
994 The above is the "device-tree" definition. It's the only statement
995 supported currently at the toplevel.
998 property1 = "string_value"; /* define a property containing a 0
1002 property2 = <1234abcd>; /* define a property containing a
1003 * numerical 32 bits value (hexadecimal)
1006 property3 = <12345678 12345678 deadbeef>;
1007 /* define a property containing 3
1008 * numerical 32 bits values (cells) in
1011 property4 = [0a 0b 0c 0d de ea ad be ef];
1012 /* define a property whose content is
1013 * an arbitrary array of bytes
1016 childnode@addresss { /* define a child node named "childnode"
1017 * whose unit name is "childnode at
1021 childprop = "hello\n"; /* define a property "childprop" of
1022 * childnode (in this case, a string)
1027 Nodes can contain other nodes etc... thus defining the hierarchical
1028 structure of the tree.
1030 Strings support common escape sequences from C: "\n", "\t", "\r",
1031 "\(octal value)", "\x(hex value)".
1033 It is also suggested that you pipe your source file through cpp (gcc
1034 preprocessor) so you can use #include's, #define for constants, etc...
1036 Finally, various options are planned but not yet implemented, like
1037 automatic generation of phandles, labels (exported to the asm file so
1038 you can point to a property content and change it easily from whatever
1039 you link the device-tree with), label or path instead of numeric value
1040 in some cells to "point" to a node (replaced by a phandle at compile
1041 time), export of reserve map address to the asm file, ability to
1042 specify reserve map content at compile time, etc...
1044 We may provide a .h include file with common definitions of that
1045 proves useful for some properties (like building PCI properties or
1046 interrupt maps) though it may be better to add a notion of struct
1047 definitions to the compiler...
1050 V - Recommendations for a bootloader
1051 ====================================
1054 Here are some various ideas/recommendations that have been proposed
1055 while all this has been defined and implemented.
1057 - The bootloader may want to be able to use the device-tree itself
1058 and may want to manipulate it (to add/edit some properties,
1059 like physical memory size or kernel arguments). At this point, 2
1060 choices can be made. Either the bootloader works directly on the
1061 flattened format, or the bootloader has its own internal tree
1062 representation with pointers (similar to the kernel one) and
1063 re-flattens the tree when booting the kernel. The former is a bit
1064 more difficult to edit/modify, the later requires probably a bit
1065 more code to handle the tree structure. Note that the structure
1066 format has been designed so it's relatively easy to "insert"
1067 properties or nodes or delete them by just memmoving things
1068 around. It contains no internal offsets or pointers for this
1071 - An example of code for iterating nodes & retrieving properties
1072 directly from the flattened tree format can be found in the kernel
1073 file arch/ppc64/kernel/prom.c, look at scan_flat_dt() function,
1074 its usage in early_init_devtree(), and the corresponding various
1075 early_init_dt_scan_*() callbacks. That code can be re-used in a
1076 GPL bootloader, and as the author of that code, I would be happy
1077 to discuss possible free licencing to any vendor who wishes to
1078 integrate all or part of this code into a non-GPL bootloader.
1082 VI - System-on-a-chip devices and nodes
1083 =======================================
1085 Many companies are now starting to develop system-on-a-chip
1086 processors, where the processor core (cpu) and many peripheral devices
1087 exist on a single piece of silicon. For these SOCs, an SOC node
1088 should be used that defines child nodes for the devices that make
1089 up the SOC. While platforms are not required to use this model in
1090 order to boot the kernel, it is highly encouraged that all SOC
1091 implementations define as complete a flat-device-tree as possible to
1092 describe the devices on the SOC. This will allow for the
1093 genericization of much of the kernel code.
1096 1) Defining child nodes of an SOC
1097 ---------------------------------
1099 Each device that is part of an SOC may have its own node entry inside
1100 the SOC node. For each device that is included in the SOC, the unit
1101 address property represents the address offset for this device's
1102 memory-mapped registers in the parent's address space. The parent's
1103 address space is defined by the "ranges" property in the top-level soc
1104 node. The "reg" property for each node that exists directly under the
1105 SOC node should contain the address mapping from the child address space
1106 to the parent SOC address space and the size of the device's
1107 memory-mapped register file.
1109 For many devices that may exist inside an SOC, there are predefined
1110 specifications for the format of the device tree node. All SOC child
1111 nodes should follow these specifications, except where noted in this
1114 See appendix A for an example partial SOC node definition for the
1118 2) Specifying interrupt information for SOC devices
1119 ---------------------------------------------------
1121 Each device that is part of an SOC and which generates interrupts
1122 should have the following properties:
1124 - interrupt-parent : contains the phandle of the interrupt
1125 controller which handles interrupts for this device
1126 - interrupts : a list of tuples representing the interrupt
1127 number and the interrupt sense and level for each interupt
1130 This information is used by the kernel to build the interrupt table
1131 for the interrupt controllers in the system.
1133 Sense and level information should be encoded as follows:
1135 Devices connected to openPIC-compatible controllers should encode
1136 sense and polarity as follows:
1138 0 = low to high edge sensitive type enabled
1139 1 = active low level sensitive type enabled
1140 2 = active high level sensitive type enabled
1141 3 = high to low edge sensitive type enabled
1143 ISA PIC interrupt controllers should adhere to the ISA PIC
1144 encodings listed below:
1146 0 = active low level sensitive type enabled
1147 1 = active high level sensitive type enabled
1148 2 = high to low edge sensitive type enabled
1149 3 = low to high edge sensitive type enabled
1153 3) Representing devices without a current OF specification
1154 ----------------------------------------------------------
1156 Currently, there are many devices on SOCs that do not have a standard
1157 representation pre-defined as part of the open firmware
1158 specifications, mainly because the boards that contain these SOCs are
1159 not currently booted using open firmware. This section contains
1160 descriptions for the SOC devices for which new nodes have been
1161 defined; this list will expand as more and more SOC-containing
1162 platforms are moved over to use the flattened-device-tree model.
1166 The MDIO is a bus to which the PHY devices are connected. For each
1167 device that exists on this bus, a child node should be created. See
1168 the definition of the PHY node below for an example of how to define
1171 Required properties:
1172 - reg : Offset and length of the register set for the device
1173 - device_type : Should be "mdio"
1174 - compatible : Should define the compatible device type for the
1175 mdio. Currently, this is most likely to be "gianfar"
1181 device_type = "mdio";
1182 compatible = "gianfar";
1190 b) Gianfar-compatible ethernet nodes
1192 Required properties:
1194 - device_type : Should be "network"
1195 - model : Model of the device. Can be "TSEC", "eTSEC", or "FEC"
1196 - compatible : Should be "gianfar"
1197 - reg : Offset and length of the register set for the device
1198 - mac-address : List of bytes representing the ethernet address of
1200 - interrupts : <a b> where a is the interrupt number and b is a
1201 field that represents an encoding of the sense and level
1202 information for the interrupt. This should be encoded based on
1203 the information in section 2) depending on the type of interrupt
1204 controller you have.
1205 - interrupt-parent : the phandle for the interrupt controller that
1206 services interrupts for this device.
1207 - phy-handle : The phandle for the PHY connected to this ethernet
1214 device_type = "network";
1216 compatible = "gianfar";
1218 mac-address = [ 00 E0 0C 00 73 00 ];
1219 interrupts = <d 3 e 3 12 3>;
1220 interrupt-parent = <40000>;
1221 phy-handle = <2452000>
1228 Required properties:
1230 - device_type : Should be "ethernet-phy"
1231 - interrupts : <a b> where a is the interrupt number and b is a
1232 field that represents an encoding of the sense and level
1233 information for the interrupt. This should be encoded based on
1234 the information in section 2) depending on the type of interrupt
1235 controller you have.
1236 - interrupt-parent : the phandle for the interrupt controller that
1237 services interrupts for this device.
1238 - reg : The ID number for the phy, usually a small integer
1239 - linux,phandle : phandle for this node; likely referenced by an
1240 ethernet controller node.
1246 linux,phandle = <2452000>
1247 interrupt-parent = <40000>;
1248 interrupts = <35 1>;
1250 device_type = "ethernet-phy";
1254 d) Interrupt controllers
1256 Some SOC devices contain interrupt controllers that are different
1257 from the standard Open PIC specification. The SOC device nodes for
1258 these types of controllers should be specified just like a standard
1259 OpenPIC controller. Sense and level information should be encoded
1260 as specified in section 2) of this chapter for each device that
1261 specifies an interrupt.
1266 linux,phandle = <40000>;
1267 clock-frequency = <0>;
1268 interrupt-controller;
1269 #address-cells = <0>;
1270 reg = <40000 40000>;
1272 compatible = "chrp,open-pic";
1273 device_type = "open-pic";
1280 Required properties :
1282 - device_type : Should be "i2c"
1283 - reg : Offset and length of the register set for the device
1285 Recommended properties :
1287 - compatible : Should be "fsl-i2c" for parts compatible with
1288 Freescale I2C specifications.
1289 - interrupts : <a b> where a is the interrupt number and b is a
1290 field that represents an encoding of the sense and level
1291 information for the interrupt. This should be encoded based on
1292 the information in section 2) depending on the type of interrupt
1293 controller you have.
1294 - interrupt-parent : the phandle for the interrupt controller that
1295 services interrupts for this device.
1296 - dfsrr : boolean; if defined, indicates that this I2C device has
1297 a digital filter sampling rate register
1298 - fsl5200-clocking : boolean; if defined, indicated that this device
1299 uses the FSL 5200 clocking mechanism.
1304 interrupt-parent = <40000>;
1305 interrupts = <1b 3>;
1307 device_type = "i2c";
1308 compatible = "fsl-i2c";
1313 f) Freescale SOC USB controllers
1315 The device node for a USB controller that is part of a Freescale
1316 SOC is as described in the document "Open Firmware Recommended
1317 Practice : Universal Serial Bus" with the following modifications
1320 Required properties :
1321 - compatible : Should be "fsl-usb2-mph" for multi port host usb
1322 controllers, or "fsl-usb2-dr" for dual role usb controllers
1323 - phy_type : For multi port host usb controllers, should be one of
1324 "ulpi", or "serial". For dual role usb controllers, should be
1325 one of "ulpi", "utmi", "utmi_wide", or "serial".
1326 - reg : Offset and length of the register set for the device
1327 - port0 : boolean; if defined, indicates port0 is connected for
1328 fsl-usb2-mph compatible controllers. Either this property or
1329 "port1" (or both) must be defined for "fsl-usb2-mph" compatible
1331 - port1 : boolean; if defined, indicates port1 is connected for
1332 fsl-usb2-mph compatible controllers. Either this property or
1333 "port0" (or both) must be defined for "fsl-usb2-mph" compatible
1336 Recommended properties :
1337 - interrupts : <a b> where a is the interrupt number and b is a
1338 field that represents an encoding of the sense and level
1339 information for the interrupt. This should be encoded based on
1340 the information in section 2) depending on the type of interrupt
1341 controller you have.
1342 - interrupt-parent : the phandle for the interrupt controller that
1343 services interrupts for this device.
1345 Example multi port host usb controller device node :
1347 device_type = "usb";
1348 compatible = "fsl-usb2-mph";
1350 #address-cells = <1>;
1352 interrupt-parent = <700>;
1353 interrupts = <27 1>;
1359 Example dual role usb controller device node :
1361 device_type = "usb";
1362 compatible = "fsl-usb2-dr";
1364 #address-cells = <1>;
1366 interrupt-parent = <700>;
1367 interrupts = <26 1>;
1372 g) Freescale SOC SEC Security Engines
1374 Required properties:
1376 - device_type : Should be "crypto"
1377 - model : Model of the device. Should be "SEC1" or "SEC2"
1378 - compatible : Should be "talitos"
1379 - reg : Offset and length of the register set for the device
1380 - interrupts : <a b> where a is the interrupt number and b is a
1381 field that represents an encoding of the sense and level
1382 information for the interrupt. This should be encoded based on
1383 the information in section 2) depending on the type of interrupt
1384 controller you have.
1385 - interrupt-parent : the phandle for the interrupt controller that
1386 services interrupts for this device.
1387 - num-channels : An integer representing the number of channels
1389 - channel-fifo-len : An integer representing the number of
1390 descriptor pointers each channel fetch fifo can hold.
1391 - exec-units-mask : The bitmask representing what execution units
1392 (EUs) are available. It's a single 32 bit cell. EU information
1393 should be encoded following the SEC's Descriptor Header Dword
1394 EU_SEL0 field documentation, i.e. as follows:
1396 bit 0 = reserved - should be 0
1397 bit 1 = set if SEC has the ARC4 EU (AFEU)
1398 bit 2 = set if SEC has the DES/3DES EU (DEU)
1399 bit 3 = set if SEC has the message digest EU (MDEU)
1400 bit 4 = set if SEC has the random number generator EU (RNG)
1401 bit 5 = set if SEC has the public key EU (PKEU)
1402 bit 6 = set if SEC has the AES EU (AESU)
1403 bit 7 = set if SEC has the Kasumi EU (KEU)
1405 bits 8 through 31 are reserved for future SEC EUs.
1407 - descriptor-types-mask : The bitmask representing what descriptors
1408 are available. It's a single 32 bit cell. Descriptor type
1409 information should be encoded following the SEC's Descriptor
1410 Header Dword DESC_TYPE field documentation, i.e. as follows:
1412 bit 0 = set if SEC supports the aesu_ctr_nonsnoop desc. type
1413 bit 1 = set if SEC supports the ipsec_esp descriptor type
1414 bit 2 = set if SEC supports the common_nonsnoop desc. type
1415 bit 3 = set if SEC supports the 802.11i AES ccmp desc. type
1416 bit 4 = set if SEC supports the hmac_snoop_no_afeu desc. type
1417 bit 5 = set if SEC supports the srtp descriptor type
1418 bit 6 = set if SEC supports the non_hmac_snoop_no_afeu desc.type
1419 bit 7 = set if SEC supports the pkeu_assemble descriptor type
1420 bit 8 = set if SEC supports the aesu_key_expand_output desc.type
1421 bit 9 = set if SEC supports the pkeu_ptmul descriptor type
1422 bit 10 = set if SEC supports the common_nonsnoop_afeu desc. type
1423 bit 11 = set if SEC supports the pkeu_ptadd_dbl descriptor type
1425 ..and so on and so forth.
1431 device_type = "crypto";
1433 compatible = "talitos";
1434 reg = <30000 10000>;
1435 interrupts = <1d 3>;
1436 interrupt-parent = <40000>;
1438 channel-fifo-len = <18>;
1439 exec-units-mask = <000000fe>;
1440 descriptor-types-mask = <012b0ebf>;
1443 h) Board Control and Status (BCSR)
1445 Required properties:
1447 - device_type : Should be "board-control"
1448 - reg : Offset and length of the register set for the device
1453 device_type = "board-control";
1454 reg = <f8000000 8000>;
1457 i) Freescale QUICC Engine module (QE)
1458 This represents qe module that is installed on PowerQUICC II Pro.
1459 Hopefully it will merge backward compatibility with CPM/CPM2.
1460 Basically, it is a bus of devices, that could act more or less
1461 as a complete entity (UCC, USB etc ). All of them should be siblings on
1462 the "root" qe node, using the common properties from there.
1463 The description below applies to the the qe of MPC8360 and
1464 more nodes and properties would be extended in the future.
1468 Required properties:
1469 - device_type : should be "qe";
1470 - model : precise model of the QE, Can be "QE", "CPM", or "CPM2"
1471 - reg : offset and length of the device registers.
1472 - bus-frequency : the clock frequency for QUICC Engine.
1474 Recommended properties
1475 - brg-frequency : the internal clock source frequency for baud-rate
1480 #address-cells = <1>;
1482 #interrupt-cells = <2>;
1485 ranges = <0 e0100000 00100000>;
1486 reg = <e0100000 480>;
1487 brg-frequency = <0>;
1488 bus-frequency = <179A7B00>;
1492 ii) SPI (Serial Peripheral Interface)
1494 Required properties:
1495 - device_type : should be "spi".
1496 - compatible : should be "fsl_spi".
1497 - mode : the spi operation mode, it can be "cpu" or "qe".
1498 - reg : Offset and length of the register set for the device
1499 - interrupts : <a b> where a is the interrupt number and b is a
1500 field that represents an encoding of the sense and level
1501 information for the interrupt. This should be encoded based on
1502 the information in section 2) depending on the type of interrupt
1503 controller you have.
1504 - interrupt-parent : the phandle for the interrupt controller that
1505 services interrupts for this device.
1509 device_type = "spi";
1510 compatible = "fsl_spi";
1512 interrupts = <82 0>;
1513 interrupt-parent = <700>;
1518 iii) USB (Universal Serial Bus Controller)
1520 Required properties:
1521 - device_type : should be "usb".
1522 - compatible : could be "qe_udc" or "fhci-hcd".
1523 - mode : the could be "host" or "slave".
1524 - reg : Offset and length of the register set for the device
1525 - interrupts : <a b> where a is the interrupt number and b is a
1526 field that represents an encoding of the sense and level
1527 information for the interrupt. This should be encoded based on
1528 the information in section 2) depending on the type of interrupt
1529 controller you have.
1530 - interrupt-parent : the phandle for the interrupt controller that
1531 services interrupts for this device.
1535 device_type = "usb";
1536 compatible = "qe_udc";
1538 interrupts = <8b 0>;
1539 interrupt-parent = <700>;
1544 iv) UCC (Unified Communications Controllers)
1546 Required properties:
1547 - device_type : should be "network", "hldc", "uart", "transparent"
1549 - compatible : could be "ucc_geth" or "fsl_atm" and so on.
1550 - model : should be "UCC".
1551 - device-id : the ucc number(1-8), corresponding to UCCx in UM.
1552 - reg : Offset and length of the register set for the device
1553 - interrupts : <a b> where a is the interrupt number and b is a
1554 field that represents an encoding of the sense and level
1555 information for the interrupt. This should be encoded based on
1556 the information in section 2) depending on the type of interrupt
1557 controller you have.
1558 - interrupt-parent : the phandle for the interrupt controller that
1559 services interrupts for this device.
1560 - pio-handle : The phandle for the Parallel I/O port configuration.
1561 - rx-clock : represents the UCC receive clock source.
1562 0x00 : clock source is disabled;
1563 0x1~0x10 : clock source is BRG1~BRG16 respectively;
1564 0x11~0x28: clock source is QE_CLK1~QE_CLK24 respectively.
1565 - tx-clock: represents the UCC transmit clock source;
1566 0x00 : clock source is disabled;
1567 0x1~0x10 : clock source is BRG1~BRG16 respectively;
1568 0x11~0x28: clock source is QE_CLK1~QE_CLK24 respectively.
1570 Required properties for network device_type:
1571 - mac-address : list of bytes representing the ethernet address.
1572 - phy-handle : The phandle for the PHY connected to this controller.
1576 device_type = "network";
1577 compatible = "ucc_geth";
1581 interrupts = <a0 0>;
1582 interrupt-parent = <700>;
1583 mac-address = [ 00 04 9f 00 23 23 ];
1586 phy-handle = <212000>;
1587 pio-handle = <140001>;
1591 v) Parallel I/O Ports
1593 This node configures Parallel I/O ports for CPUs with QE support.
1594 The node should reside in the "soc" node of the tree. For each
1595 device that using parallel I/O ports, a child node should be created.
1596 See the definition of the Pin configuration nodes below for more
1599 Required properties:
1600 - device_type : should be "par_io".
1601 - reg : offset to the register set and its length.
1602 - num-ports : number of Parallel I/O ports
1607 #address-cells = <1>;
1609 device_type = "par_io";
1616 vi) Pin configuration nodes
1618 Required properties:
1619 - linux,phandle : phandle of this node; likely referenced by a QE
1621 - pio-map : array of pin configurations. Each pin is defined by 6
1622 integers. The six numbers are respectively: port, pin, dir,
1623 open_drain, assignment, has_irq.
1624 - port : port number of the pin; 0-6 represent port A-G in UM.
1625 - pin : pin number in the port.
1626 - dir : direction of the pin, should encode as follows:
1628 0 = The pin is disabled
1629 1 = The pin is an output
1630 2 = The pin is an input
1633 - open_drain : indicates the pin is normal or wired-OR:
1635 0 = The pin is actively driven as an output
1636 1 = The pin is an open-drain driver. As an output, the pin is
1637 driven active-low, otherwise it is three-stated.
1639 - assignment : function number of the pin according to the Pin Assignment
1640 tables in User Manual. Each pin can have up to 4 possible functions in
1641 QE and two options for CPM.
1642 - has_irq : indicates if the pin is used as source of exteral
1647 linux,phandle = <140001>;
1649 /* port pin dir open_drain assignment has_irq */
1650 0 3 1 0 1 0 /* TxD0 */
1651 0 4 1 0 1 0 /* TxD1 */
1652 0 5 1 0 1 0 /* TxD2 */
1653 0 6 1 0 1 0 /* TxD3 */
1654 1 6 1 0 3 0 /* TxD4 */
1655 1 7 1 0 1 0 /* TxD5 */
1656 1 9 1 0 2 0 /* TxD6 */
1657 1 a 1 0 2 0 /* TxD7 */
1658 0 9 2 0 1 0 /* RxD0 */
1659 0 a 2 0 1 0 /* RxD1 */
1660 0 b 2 0 1 0 /* RxD2 */
1661 0 c 2 0 1 0 /* RxD3 */
1662 0 d 2 0 1 0 /* RxD4 */
1663 1 1 2 0 2 0 /* RxD5 */
1664 1 0 2 0 2 0 /* RxD6 */
1665 1 4 2 0 2 0 /* RxD7 */
1666 0 7 1 0 1 0 /* TX_EN */
1667 0 8 1 0 1 0 /* TX_ER */
1668 0 f 2 0 1 0 /* RX_DV */
1669 0 10 2 0 1 0 /* RX_ER */
1670 0 0 2 0 1 0 /* RX_CLK */
1671 2 9 1 0 3 0 /* GTX_CLK - CLK10 */
1672 2 8 2 0 1 0>; /* GTX125 - CLK9 */
1675 vii) Multi-User RAM (MURAM)
1677 Required properties:
1678 - device_type : should be "muram".
1679 - mode : the could be "host" or "slave".
1680 - ranges : Should be defined as specified in 1) to describe the
1681 translation of MURAM addresses.
1682 - data-only : sub-node which defines the address area under MURAM
1683 bus that can be allocated as data/parameter
1688 device_type = "muram";
1689 ranges = <0 00010000 0000c000>;
1696 More devices will be defined as this spec matures.
1699 Appendix A - Sample SOC node for MPC8540
1700 ========================================
1702 Note that the #address-cells and #size-cells for the SoC node
1703 in this example have been explicitly listed; these are likely
1704 not necessary as they are usually the same as the root node.
1707 #address-cells = <1>;
1709 #interrupt-cells = <2>;
1710 device_type = "soc";
1711 ranges = <00000000 e0000000 00100000>
1712 reg = <e0000000 00003000>;
1713 bus-frequency = <0>;
1717 device_type = "mdio";
1718 compatible = "gianfar";
1721 linux,phandle = <2452000>
1722 interrupt-parent = <40000>;
1723 interrupts = <35 1>;
1725 device_type = "ethernet-phy";
1729 linux,phandle = <2452001>
1730 interrupt-parent = <40000>;
1731 interrupts = <35 1>;
1733 device_type = "ethernet-phy";
1737 linux,phandle = <2452002>
1738 interrupt-parent = <40000>;
1739 interrupts = <35 1>;
1741 device_type = "ethernet-phy";
1748 device_type = "network";
1750 compatible = "gianfar";
1752 mac-address = [ 00 E0 0C 00 73 00 ];
1753 interrupts = <d 3 e 3 12 3>;
1754 interrupt-parent = <40000>;
1755 phy-handle = <2452000>;
1759 #address-cells = <1>;
1761 device_type = "network";
1763 compatible = "gianfar";
1765 mac-address = [ 00 E0 0C 00 73 01 ];
1766 interrupts = <13 3 14 3 18 3>;
1767 interrupt-parent = <40000>;
1768 phy-handle = <2452001>;
1772 #address-cells = <1>;
1774 device_type = "network";
1776 compatible = "gianfar";
1778 mac-address = [ 00 E0 0C 00 73 02 ];
1779 interrupts = <19 3>;
1780 interrupt-parent = <40000>;
1781 phy-handle = <2452002>;
1785 device_type = "serial";
1786 compatible = "ns16550";
1788 clock-frequency = <0>;
1789 interrupts = <1a 3>;
1790 interrupt-parent = <40000>;
1794 linux,phandle = <40000>;
1795 clock-frequency = <0>;
1796 interrupt-controller;
1797 #address-cells = <0>;
1798 reg = <40000 40000>;
1800 compatible = "chrp,open-pic";
1801 device_type = "open-pic";
1806 interrupt-parent = <40000>;
1807 interrupts = <1b 3>;
1809 device_type = "i2c";
1810 compatible = "fsl-i2c";