s_inst_prefetch GPU Native ISA AMD Scalar
Branch & Control
Change instruction prefetch mode. This controls how many cachelines ahead of the current PC the shader attempts to prefetch.
Also written as
s_set_inst_prefetch_distance.
AMD's machine-readable ISA specification lists this name for the same instruction.
Encoding
The ENC_SOPP layout from AMD's machine-readable ISA specification (AMD RDNA 2). Opcode 32 in OP rebuilds 0xBFA00000, an identifier AMD lists for this encoding. The same in AMD RDNA 1.
Operands
-
SIMM16
Read. 16-bit integer constant. Data format: N/A (OPR_SIMM16, FMT_NUM_B16)
In AMD's order, as its machine-readable ISA specification lists them for the ENC_SOPP encoding (AMD RDNA 2).
GFX Target Compatibility
Per-target GFX compatibility has not yet been verified for this instruction.
Related
More in Branch & Control
Reference
Example
s_inst_prefetch 0x0A real instruction accepted by the LLVM assembler, taken verbatim from LLVM's own AMDGPU MC test suite (gfx10_asm_sop.s). Not from an AMD document, and not authored here.
Sources
- AMD Machine-Readable GPU ISA Specification ↗ - Advanced Micro Devices, Inc.
- LLVM MC assembler tests for AMDGPU ↗ - LLVM Project