Compare commits

...

2 Commits

Author SHA1 Message Date
07acc7ff1e Started to document instructions and updated encoding formats (colors + added conditions)
All checks were successful
Central(Architecture) Docs Build / Build Documentation (push) Successful in 13s
2024-10-20 23:09:13 +02:00
f3bf60db75 Fix typo and reformulate phrases 2024-10-20 18:23:00 +02:00
6 changed files with 187 additions and 30 deletions

View File

@ -4,7 +4,7 @@
[glossary]
[horizontal]
Processing Complex::
Composed of a _Processing Unit_, memory modules and subsystems that are not located inside of the _Processing Unit_ (e.g. Channel I/O Subsytem, Hardware Management Subsystem).
Composed of a _Processing Unit_, memory modules and subsystems that are not located inside of the _Processing Unit_ (e.g. Channel I/O Subsystem, Hardware Management Subsystem).
The name applies both to the hardware enclosure and the hardware inside said enclosure.
Processing Unit::

View File

@ -1,48 +1,60 @@
[wavedrom, ,svg]
....
{reg: [
{bits: 7, name: 'opcode'},
{bits: 5, name: 'rd'},
{bits: 5, name: 'rs1'},
{bits: 5, name: 'rs2'},
{bits: 10, name: 'imm'}
{bits: 7, name: 'opcode', type: 8},
{bits: 4, name: 'cond', type: 6},
{bits: 5, name: 'rd', type: 2},
{bits: 5, name: 'rs1', type: 4},
{bits: 5, name: 'rs2', type: 4},
{bits: 6, name: 'imm', type: 5}
], config: {label: {right: 'A-Type'}}}
....
[wavedrom, ,svg]
....
{reg: [
{bits: 7, name: 'opcode'},
{bits: 5, name: 'rd'},
{bits: 5, name: 'rs1'},
{bits: 15, name: 'imm'}
{bits: 7, name: 'opcode', type: 8},
{bits: 4, name: 'cond', type: 6},
{bits: 5, name: 'rd', type: 2},
{bits: 5, name: 'rs1', type: 4},
{bits: 11, name: 'imm', type: 5}
], config: {label: {right: 'B-Type'}}}
....
[wavedrom, ,svg]
....
{reg: [
{bits: 7, name: 'opcode'},
{bits: 5, name: 'rd'},
{bits: 20, name: 'imm'}
{bits: 7, name: 'opcode', type: 8},
{bits: 4, name: 'cond', type: 6},
{bits: 5, name: 'rd', type: 2},
{bits: 16, name: 'imm', type: 5}
], config: {label: {right: 'C-Type'}}}
....
[wavedrom, ,svg]
....
{reg: [
{bits: 7, name: 'opcode'},
{bits: 25, name: 'imm'}
], config: {label: {right: 'D-Type'}}}
{bits: 7, name: 'opcode', type: 8},
{bits: 4, name: 'cond', type: 6},
{bits: 21, name: 'imm', type: 5}
], config: {label: {right: 'D1-Type'}}}
....
[wavedrom, ,svg]
....
{reg: [
{bits: 7, name: 'opcode'},
{bits: 5, name: 'reg'},
{bits: 4, name: 'sid'},
{bits: 8, name: 'sre'},
{bits: 8, name: 'cmd'}
{bits: 7, name: 'opcode', type: 8},
{bits: 25, name: 'imm', type: 5}
], config: {label: {right: 'D2-Type'}}}
....
[wavedrom, ,svg]
....
{reg: [
{bits: 7, name: 'opcode', type: 8},
{bits: 5, name: 'reg', type: 2},
{bits: 4, name: 'sid', type: 7},
{bits: 8, name: 'sre', type: 2},
{bits: 8, name: 'cmd', type: 8}
], config: {label: {right: 'E-Type'}}}
....

View File

@ -7,17 +7,23 @@ For instruction encoding formats that contain an immediate value, not all immedi
include::images/instruction-formats.adoc[]
.Bit ranges legend
.Format fields
[horizontal]
opcode::
The operation to carry on.
cond::
Condition code. +
#TODO: Define conditions encoding.#
rd::
Destination register.
Destination register. +
#TODO: Define registers encoding.#
rs1::
Source register 1.
rs2::
Source register 2.
imm::
Immediate value.
Can be interpreted as signed or unsigned depending on the instruction.
reg::
Source/destination register on the _Execution Engine_ side.
sid::
@ -30,3 +36,135 @@ cmd::
=== Instruction list
#TODO: List instructions#
* NOP instruction
* memory load/store instructions
* register move instructions
* arithmetic instructions
* bitwise operations instructions (w/ bit shifts)
* comparison instructions
* jump instructions
* system mode instructions (svc, uret, sret, ...)
* subsystems instructions (ssr & ssw)
==== The NOP instruction
[wavedrom, ,svg]
....
{reg: [
{bits: 7, name: 0x0, type: 8, attr: '0x0'},
{bits: 25, name: 'unused'}
], config: {label: {right: 'NOP'}}}
....
Description::
Does nothing. Can be used to align a block of instructions.
Encoding:: D2-Type
Assembler syntax::
[source]
----
nop
----
Privileged instruction:: No.
Updates program state flags:: No.
Exceptions:: None.
==== Memory-related instructions
===== LDR: Load Register
[wavedrom, ,svg]
....
{reg: [
{bits: 7, name: 0x1, type: 8, attr: '0x1'},
{bits: 4, name: 'cond', type: 6},
{bits: 5, name: 'dst', type: 2},
{bits: 5, name: 'src', type: 4},
{bits: 11, name: 'off', type: 5}
], config: {label: {right: 'LDRIR'}}}
....
[wavedrom, ,svg]
....
{reg: [
{bits: 7, name: 0x2, type: 8, attr: '0x2'},
{bits: 4, name: 'cond', type: 6},
{bits: 5, name: 'dst', type: 2},
{bits: 5, name: 'src', type: 4},
{bits: 11, name: 'off', type: 5}
], config: {label: {right: 'LDRIRW'}}}
....
[wavedrom, ,svg]
....
{reg: [
{bits: 7, name: 0x3, type: 8, attr: '0x3'},
{bits: 4, name: 'cond', type: 6},
{bits: 5, name: 'dst', type: 2},
{bits: 5, name: 'src', type: 4},
{bits: 11, name: 'off', type: 5}
], config: {label: {right: 'LDRIOW'}}}
....
[wavedrom, ,svg]
....
{reg: [
{bits: 7, name: 0x4, type: 8, attr: '0x4'},
{bits: 4, name: 'cond', type: 6},
{bits: 5, name: 'dst', type: 2},
{bits: 5, name: 'src', type: 4},
{bits: 5, name: 'off', type: 4},
{bits: 6, name: 'unused'}
], config: {label: {right: 'LDRR'}}}
....
[wavedrom, ,svg]
....
{reg: [
{bits: 7, name: 0x5, type: 8, attr: '0x5'},
{bits: 4, name: 'cond', type: 6},
{bits: 5, name: 'dst', type: 2},
{bits: 5, name: 'src', type: 4},
{bits: 5, name: 'off', type: 4},
{bits: 6, name: 'unused'}
], config: {label: {right: 'LDRRW'}}}
....
[wavedrom, ,svg]
....
{reg: [
{bits: 7, name: 0x6, type: 8, attr: '0x6'},
{bits: 4, name: 'cond', type: 6},
{bits: 5, name: 'dst', type: 2},
{bits: 5, name: 'src', type: 4},
{bits: 5, name: 'off', type: 4},
{bits: 6, name: 'unused'}
], config: {label: {right: 'LDROW'}}}
....
====== LDRIR: Load Register Immediate Pre-indexed
Description::
Loads a word from memory into a register.
The immediate offset `off` is added to the address in the `src` register before reading memory.
Encoding:: A-Type
Assembler syntax::
+
[source]
----
ldr{cond} dst, [src, off]
----
Examples::
+
[source]
----
ldr r1, [r0] ; Reads a word from the memory address in r0 into r1.
ldr r3, [r2, 8] ; Reads a word from the memory address in r2, with a 8 bytes
; offset, into r3.
ldr.eq r5, [r4] ; If the last comparison resulted in an 'eq' condition status,
; then reads a word from the memory address in r4 into r5.
; Else, does nothing.
----
Privileged instruction:: No.
Updates program state flags:: No.
Exceptions::
[horizontal]
MemFault:::
If the memory address being accessed is invalid, non readable or not paged in.
The kernel may update the page table entries and re-execute the instruction without the user application being aware that it failed in the first place.

View File

@ -7,14 +7,14 @@ The {central-arch-name} uses byte-addressable memory.
While an _Execution engine_ handles data 32-bits wide, memory addresses are only 24-bits wide.
An _Execution Engine_ can thus address up to 16MB of main memory.
NOTE: We use the terms "`memory`" and "`main memory`" interchangably. Main memory refers to the RAM while we use the term "`secondary memory`" to refer to HDD or SSD storage.
NOTE: We use the terms "`memory`" and "`main memory`" interchangeably. Main memory refers to the RAM while we use the term "`secondary memory`" to refer to HDD or SSD storage.
At the hardware level, memory accesses are done on a memory-word boundary.
A memory-word is 32-bits wide and memory accesses should be done at a 32-bits alignment to avoid wasting cycles doing double the amount of memory operations.
NOTE: The term "`memory accesses`" encompasses both read and write operations.
Data is encoded in memory with the little endian scheme.
Data is encoded in memory in little endian.
For a given value, the least significant byte (LSB) is stored in the lowest address and the most significant byte (MSB) in the highest.
=== System Modes and Privilege Levels
@ -48,7 +48,7 @@ Supervisor-mode::
Software executing in this mode provides context switching, I/O, process management and inter-process communications.
Fault-mode::
Code executes in this mode when a double fault occurs, i.e. when an exception is generated in system-mode code.
Code executes in this mode when a double fault occurs, i.e. when an exception is generated in supervisor-mode code.
Code executing under this mode can be used to log/report double faults and then reset/halt the system.
Debug exceptions generated in supervisor-mode code are also handled in this mode, in which case control is passed back to supervisor-mode after handling.
@ -62,7 +62,7 @@ Exceptions thus always suspend user code for the duration of their handling.
Exceptions can be of two types: *synchronous* and *asynchronous*.
==== Synchronous Exceptions
Synchronous exceptions are generated from events originating from inside of the _Execution engine_.
Synchronous exceptions are generated from events originating from inside of the _Execution Engine_.
They are a conditional or unconditional response to the execution of an instruction.
SVC::
@ -84,11 +84,18 @@ Debug::
* Fetching an instruction located at an address matching a configured _Hardware Breakpoint_,
* Accessing a memory address matching a configured _Hardware Watchpoint_
SvcDebug::
This exception is the same as the *Debug* exception, except it is generated when debug events are encountered while in supervisor-mode.
UnknownInstr::
This exception is generated when an unknown instruction failed to be executed.
It is important not to ignore this exception as it can be the symptom of a memory corruption bug.
==== Asynchronous Exceptions
Asynchronous exceptions are generated from events originating from outside of the _Execution engine_.
These exceptions enable the system to react to its environment.
HardMgmt::
HdwrMgmt::
This exception is generated when the _Hardware Management Console_ communicates with the _Processing Unit_ and that the _Processing Unit_ relays the event to the _Execution Engine_.
Data can be passed alongside the exception and would be stored in main memory by the _Processing Unit_ communications controller.

View File

@ -1,7 +1,7 @@
=== Registers
==== General Purpose Registers
General purpose registers (GPRs) are used to perform calculations and store intermediate values.
There are 8 GPRs in an Execution Engine. These registers are named *_r0_* through *_r7_*.
There are 8 GPRs in an _Execution Engine_. These registers are named *_r0_* through *_r7_*.
==== Special Purpose Registers

View File

@ -3,6 +3,6 @@
:icons: font
:lang: en
:toc: left
:toclevels: 5
:toclevels: 4
:sectnums:
:sectnumlevels: 5