Loading...
Searching...
No Matches
bb::avm2::simulation::Sha256 Class Reference
#include <sha256.hpp>
Inheritance diagram for bb::avm2::simulation::Sha256:
Public Member Functions | |
| Sha256 (ExecutionIdGetterInterface &execution_id_manager, BitwiseInterface &bitwise, GreaterThanInterface >, EventEmitterInterface< Sha256CompressionEvent > &event_emitter) | |
| void | compression (MemoryInterface &memory, MemoryAddress state_addr, MemoryAddress input_addr, MemoryAddress output_addr) override |
| Execute the SHA-256 compression function: read state and input from memory, compress, and write output. | |
 Public Member Functions inherited from bb::avm2::simulation::Sha256Interface | |
| virtual | ~Sha256Interface ()=default |
Private Member Functions | |
| MemoryValue | shr (const MemoryValue &x, uint8_t shift) |
| Perform a 32-bit right shift on a MemoryValue. | |
| MemoryValue | ror (const MemoryValue &x, uint8_t shift) |
| Perform a 32-bit right rotation on a MemoryValue. | |
| MemoryValue | modulo_sum (std::span< const MemoryValue > values) |
| Sum a span of U32 MemoryValues and return the result modulo 2^32. | |
Private Attributes | |
| ExecutionIdGetterInterface & | execution_id_manager |
| BitwiseInterface & | bitwise |
| GreaterThanInterface & | gt |
| EventEmitterInterface< Sha256CompressionEvent > & | events |
Detailed Description
Definition at line 17 of file sha256.hpp.
Constructor & Destructor Documentation
◆ Sha256()
|
inline |
Definition at line 19 of file sha256.hpp.
Member Function Documentation
◆ compression()
|
overridevirtual |
Execute the SHA-256 compression function: read state and input from memory, compress, and write output.
Events are emitted in the following flavors:
- Normal execution: all fields populated (state, input of 16 elements, computed output).
- Error (address out of range, invalid state tag, or invalid input tag): state is partially populated (up to the point of failure), input contains elements read before the error, output is zeroed.
In all cases the event is emitted before re-throwing the exception.
- Parameters
-
memory The memory interface to read state/input from and write output to. state_addr The starting address of the 8-element hash state in memory. input_addr The starting address of the 16-element hash input in memory. output_addr The starting address where the 8-element output will be written.
- Exceptions
-
Sha256CompressionException If any of state/input/output address ranges exceed the maximum memory address (checked first). Sha256CompressionException If any of the 8 state values do not have tag U32 (checked after address validation). Sha256CompressionException If any of the 16 input values do not have tag U32 (checked after state tag validation, on first invalid element).
Implements bb::avm2::simulation::Sha256Interface.
Definition at line 135 of file sha256.cpp.
◆ modulo_sum()
|
private |
Sum a span of U32 MemoryValues and return the result modulo 2^32.
- Parameters
-
values A span of MemoryValue elements, each expected to hold a uint32_t.
- Returns
- The sum reduced modulo 2^32, wrapped in a MemoryValue.
- Precondition
- Every element in values must be tagged as U32. This is an internal helper; all callers (sha256_compress) guarantee the precondition by validating tags before invoking this function. The cast exception is therefore never thrown in practice.
- Note
- Asserts that both the low and high 32-bit halves of the 64-bit sum are in range.
Definition at line 93 of file sha256.cpp.
◆ ror()
|
private |
Perform a 32-bit right rotation on a MemoryValue.
- Parameters
-
x The value to rotate (must hold a uint32_t). shift The number of bits to rotate right (must be non-zero and < 32).
- Returns
- The rotated 32-bit result wrapped in a MemoryValue.
- Precondition
- x must be tagged as U32. This is an internal helper; all callers (sha256_compress) guarantee the precondition by validating tags before invoking this function. The cast exception is therefore never thrown in practice.
-
shift must satisfy 0 < shift < 32. A shift >= 32 causes undefined behavior per the C++ standard for 32-bit operands. A shift == 0 also causes undefined behavior because the reconstruction
lo << (32 - shift)becomes a left shift by 32. All callers use fixed SHA-256 rotation amounts (2, 6, 7, 11, 13, 17, 18, 19, 22, 25), so this precondition is always satisfied.
- Note
- Asserts that the lower bits extracted during decomposition are in range (lo < 2^shift).
Definition at line 43 of file sha256.cpp.
◆ shr()
|
private |
Perform a 32-bit right shift on a MemoryValue.
- Parameters
-
x The value to shift (must hold a uint32_t). shift The number of bits to shift right (only 3 and 10 are used in SHA-256).
- Returns
- The shifted 32-bit result wrapped in a MemoryValue.
- Precondition
- x must be tagged as U32. This is an internal helper; all callers (sha256_compress) guarantee the precondition by validating tags before invoking this function. The cast exception is therefore never thrown in practice.
- shift must satisfy shift < 32. A shift >= 32 would cause undefined behavior per the C++ standard for 32-bit operands. All callers use fixed SHA-256 shift amounts (3, 10), so this precondition is always satisfied.
- Note
- Asserts that the lower bits extracted during decomposition are in range (lo < 2^shift).
Definition at line 70 of file sha256.cpp.
Member Data Documentation
◆ bitwise
|
private |
Definition at line 41 of file sha256.hpp.
◆ events
|
private |
Definition at line 43 of file sha256.hpp.
◆ execution_id_manager
|
private |
Definition at line 40 of file sha256.hpp.
◆ gt
|
private |
Definition at line 42 of file sha256.hpp.
The documentation for this class was generated from the following files:
- src/barretenberg/vm2/simulation/gadgets/sha256.hpp
- src/barretenberg/vm2/simulation/gadgets/sha256.cpp
