Diagnostics no longer repeat the generated table's ignorance of a
registered mnemonic: unknown-instruction and unencodable-instruction
against a statement the registry encodes are filtered from the server's
own presentation, driven by asm.LookupExtension directly. The filter is
a no-op once lint learns the registry, so the two compose unchanged.
Assisted-by: GLM 5.3 Flash
Completion merges the extension registry's mnemonics beside the toolchain
entries, with operand shapes measured against the layer's own encoder, and
hover documents a registered mnemonic from its metadata: the extension
notice, the forms, the features, the fixed encodings and the manual
references.
Assisted-by: GLM 5.3 Flash
A line-for-line byte comparison of the whole amd64enc.s corpus against
go tool asm surfaced divergences the pass-only accounting never showed:
PEXTRW's GPR form swapped its fields, PUSHW took an imm32 where the
toolchain bounds the immediate to 16 bits, the double shift wrote the
unmasked register number into the reg field, VCOMISS carried a 0x66
prefix, RORX dropped the destination's R bit, and the variable bit
shifts used the manual's per-width opcodes where the toolchain
consolidates each row on one opcode with the W bit. The VEX forms the
toolchain prefers for plain vector registers (the SSE2/SSSE3/SSE4.1
AVX twins, the compare-with-predicate family, VMOVUPS, VSHUFPS, the
variable shifts) now encode under VEX, with EVEX left to the ZMM,
opmask and index-16+ spellings, and the mnemonics whose rows never
offer the 2-byte prefix force it. Every line is pinned through the new
corpus parity test (793 lines); the whole corpus file now assembles to
the toolchain's bytes at every commented line (10022 of 10022).
Assisted-by: GLM 5.3 Flash
The SSE3 horizontal and add-subtract pairs, the SSSE3 sign and
horizontal integers, the masked moves in both directions, the
reciprocity and test pairs, the AVX imm8 tail (blends, dot products,
inserts, rounds, MPSADBW, the string compares), the four-operand
variable blends with their /is4 mask byte, the scalar three-operand
moves, the MXCSR accessors, the VPERMIL register controls and the
variable word shifts, plus the BMI2 count forms over memory. Every
encoding is pinned byte for byte against go tool asm through every
corpus line the toolchain's own amd64enc.s carries for the families
(852 lines); the /is4 byte carries the mask register number in its high
nibble, the layout the toolchain emits.
Assisted-by: GLM 5.3 Flash
The packed integer and float binaries, the imm8-controlled SSE4.1 forms,
the variable blends with their X0 mask, the high/low half moves, the
sign-mask extractions, the non-temporal stores, the MOVQ bank crossings
and their odd spellings, the MMX shifts and shuffle and the cache-line
mask stores, plus the scalar leaves LEAVE, INVPCID and the RTM controls.
Every encoding is pinned byte for byte against go tool asm through every
corpus line the toolchain's own amd64enc.s carries for the families
(1465 lines). Two corpus-wide gaps fell out of the comparison: the
64-bit MOV immediate uses the zero-extending form across the unsigned
32-bit span, and the MMX-to-GPR MOVQ puts the bank register in reg.
Assisted-by: GLM 5.3 Flash
XSAVE, XSAVEOPT, XSAVEC and XSAVES with their restore twins, plain and
64, each pinned byte for byte against go tool asm through every corpus
line the toolchain's own amd64enc.s carries for the family (24 lines).
The toolchain emits XSAVEOPT without the manual's 0x66 prefix; the bytes
are the oracle, so the family carries none.
Assisted-by: GLM 5.3 Flash
The no-operand flag and system controls, the sign-extension pair, the
string primitives, the multi-byte no-ops, the cache controls, MOVBE, the
compare-exchange doubles, the random source and FS/GS base pairs, the
descriptor-table accesses, the 0F 00/01 register controls and the
LAR/LSL selector reads and far-segment loads, each pinned byte for byte
against go tool asm through every corpus line the toolchain's own
amd64enc.s carries for the families (279 lines).
Assisted-by: GLM 5.3 Flash
The x87 stack controls, the D8/DC arithmetic pair, the conditional moves,
the register compares, FADDDP, the memory loads and the FXSAVE pair, each
pinned byte for byte against go tool asm through every corpus line the
toolchain's own amd64enc.s carries for the family (78 lines).
Assisted-by: GLM 5.3 Flash
petrbalvin