petrbalvin 71e8dd550d feat(asm): split wide arm64 load and store offsets into REGTMP ...

Offsets the single-instruction forms cannot carry lower the way the
toolchain lowers them: ADD or SUB moves the whole distance into REGTMP
within the ±4095 band, and the 24-bit band above it splits into an ADD of
the high half and an access of the low half, with the pair family taking
the two-ADD sequence.  The split band follows loadStoreClass per width,
byte accesses taking the full 24 bits and the Q width the widest, so an
offset the toolchain pools is never split instead.

Assisted-by: GLM 5.3 Flash

2026-10-07 02:36:24 +02:00

..

abi_amd64.s

style: purge em dashes from the produced text

2026-09-16 23:12:31 +02:00

abi_arm64.s

fix(riscv64): accept the g spelling of the goroutine register

2026-09-20 00:38:24 +02:00

abi_loong64.s

fix(riscv64): accept the g spelling of the goroutine register

2026-09-20 00:38:24 +02:00

abi_riscv64.s

fix(riscv64): accept the g spelling of the goroutine register

2026-09-20 00:38:24 +02:00

amdb_loong64.s

feat(loong64): encode the full LSX and LASX table

2026-09-20 19:14:43 +02:00

atomics_amd64.s

feat(amd64): encode the GOROOT instruction families

2026-09-20 06:44:51 +02:00

atomics_arm64.s

fix(asm): scale the FLDPQ and FSTPQ pair offsets by sixteen

2026-10-07 02:36:24 +02:00

atomics_loong64.s

feat(riscv64,loong64): encode AMO atomics, vector slices and bit ops

2026-09-20 06:44:51 +02:00

atomics_riscv64.s

feat(riscv64,loong64): encode AMO atomics, vector slices and bit ops

2026-09-20 06:44:51 +02:00

avx512_amd64.s

feat(amd64): encode the AVX-512 and BMI corpus families

2026-09-20 21:17:20 +02:00

avx_amd64.s

feat(amd64): encode the GOROOT instruction families

2026-09-20 06:44:51 +02:00

basic_amd64.s

style: put the repository assembly in gasm fmt canonical form

2026-09-17 20:33:18 +02:00

basic_arm64.s

style: put the repository assembly in gasm fmt canonical form

2026-09-17 20:33:18 +02:00

basic_loong64.s

feat(cmd): enable loong64 JIT execution, all trampolines qemu-validated

2026-09-20 00:57:02 +02:00

basic_riscv64.s

fix(asm): correct RISC-V frame layout and RVC encodings

2026-08-13 14:41:57 +02:00

bigframe2_arm64.s

fix(arm64): store-exclusive operand order and large-frame parity

2026-09-20 00:38:24 +02:00

bigframe_amd64.s

fix(asm): close the oracle parity gaps in frame addressing and calls

2026-09-14 23:25:14 +02:00

bigframe_arm64.s

fix(asm): close the oracle parity gaps in frame addressing and calls

2026-09-14 23:25:14 +02:00

bigframe_loong64.s

fix(asm): close the oracle parity gaps in frame addressing and calls

2026-09-14 23:25:14 +02:00

bigframe_riscv64.s

fix(asm): close the oracle parity gaps in frame addressing and calls

2026-09-14 23:25:14 +02:00

bitmanip_riscv64.s

feat(riscv64,loong64): encode AMO atomics, vector slices and bit ops

2026-09-20 06:44:51 +02:00

bookkeep_amd64.s

feat(amd64): floating-point immediates through a synthesised pool

2026-09-21 02:04:44 +02:00

bookkeep_arm64.s

feat(arm64): whole-vector moves, bookkeeping ops and truncating-move lowering

2026-09-20 21:17:20 +02:00

bookkeep_riscv64.s

feat(riscv64,loong64): operand tail, float DATA and honest port classification

2026-09-21 00:44:47 +02:00

branch2_loong64.s

feat(riscv64,loong64): operand tail, float DATA and honest port classification

2026-09-21 00:44:47 +02:00

branch_arm64.s

style: put the repository assembly in gasm fmt canonical form

2026-09-17 20:33:18 +02:00

branch_far_riscv64.s

feat(riscv64,loong64): PCALIGN, branch relaxation and operand shapes

2026-09-20 14:25:47 +02:00

branch_riscv64.s

style: put the repository assembly in gasm fmt canonical form

2026-09-17 20:33:18 +02:00

branchu_loong64.s

fix(loong64): FP immediates through R30 and unsigned branch forms

2026-09-19 23:49:13 +02:00

call_arm64.s

style: put the repository assembly in gasm fmt canonical form

2026-09-17 20:33:18 +02:00

call_riscv64.s

fix(asm): encode RISC-V CALL sym(SB) as JAL

2026-08-13 18:12:22 +02:00

carryshift_arm64.s

feat(arm64): wide immediates, SIMD compare and system operand forms

2026-09-20 14:25:47 +02:00

crypto_amd64.s

feat(amd64): encode the GOROOT instruction families

2026-09-20 06:44:51 +02:00

crypto_arm64.s

feat(arm64): encode pairs, atomics, crypto, system and NEON slices

2026-09-20 06:44:51 +02:00

datainit_loong64.s

feat(riscv64,loong64): operand tail, float DATA and honest port classification

2026-09-21 00:44:47 +02:00

datarel_amd64.s

feat(asm): symbol-valued DATA, division slash in symbols and plain semicolons

2026-09-20 19:15:05 +02:00

datarel_arm64.s

feat(asm): symbol-valued DATA, division slash in symbols and plain semicolons

2026-09-20 19:15:05 +02:00

divslash_amd64.s

feat(asm): symbol-valued DATA, division slash in symbols and plain semicolons

2026-09-20 19:15:05 +02:00

divslash_arm64.s

feat(asm): symbol-valued DATA, division slash in symbols and plain semicolons

2026-09-20 19:15:05 +02:00

doubleshift_amd64.s

feat(amd64): assemble the double-shift and static-SB operand shapes

2026-09-20 11:40:39 +02:00

exclusive_arm64.s

fix(arm64): store-exclusive operand order and large-frame parity

2026-09-20 00:38:24 +02:00

floatimm_amd64.s

feat(amd64): floating-point immediates through a synthesised pool

2026-09-21 02:04:44 +02:00

forms_amd64.s

style(testdata): canonicalise forms_amd64.s

2026-10-02 00:40:20 +02:00

fp_arm64.s

style: put the repository assembly in gasm fmt canonical form

2026-09-17 20:33:18 +02:00

fp_loong64.s

style: put the repository assembly in gasm fmt canonical form

2026-09-17 20:33:18 +02:00

guard_amd64.s

fix(asm): close the oracle parity gaps in frame addressing and calls

2026-09-14 23:25:14 +02:00

guard_arm64.s

fix(asm): close the oracle parity gaps in frame addressing and calls

2026-09-14 23:25:14 +02:00

guard_loong64.s

fix(asm): close the oracle parity gaps in frame addressing and calls

2026-09-14 23:25:14 +02:00

guard_riscv64.s

fix(asm): close the oracle parity gaps in frame addressing and calls

2026-09-14 23:25:14 +02:00

indirect_amd64.s

style(testdata): canonical gasm formatting for the verify kernels

2026-09-19 23:49:27 +02:00

indirect_arm64.s

style(testdata): canonical gasm formatting for the verify kernels

2026-09-19 23:49:27 +02:00

indirect_loong64.s

feat(asm): encode indirect JMP and CALL on all four architectures

2026-09-19 19:17:07 +02:00

indirect_riscv64.s

style(testdata): canonical gasm formatting for the verify kernels

2026-09-19 23:49:27 +02:00

integer_arm64.s

feat(arm64): encode pairs, atomics, crypto, system and NEON slices

2026-09-20 06:44:51 +02:00

l64forms_loong64.s

feat(riscv64,loong64): PCALIGN, branch relaxation and operand shapes

2026-09-20 14:25:47 +02:00

largeimm_riscv64.s

style: put the repository assembly in gasm fmt canonical form

2026-09-17 20:33:18 +02:00

loadstore_riscv64.s

style: put the repository assembly in gasm fmt canonical form

2026-09-17 20:33:18 +02:00

misc_riscv64.s

style(testdata): canonical gasm formatting for the verify kernels

2026-09-19 23:49:27 +02:00

movimm2_arm64.s

fix(arm64): encode shifts, divides and multiplies and align sizes with emission

2026-09-19 23:49:07 +02:00

movimm_arm64.s

style: put the repository assembly in gasm fmt canonical form

2026-09-17 20:33:18 +02:00

movimm_riscv64.s

fix(asm): materialise RISC-V MOV immediates

2026-08-13 17:41:16 +02:00

movwfp_loong64.s

fix(loong64): FP immediates through R30 and unsigned branch forms

2026-09-19 23:49:13 +02:00

pcalign_loong64.s

feat(riscv64,loong64): PCALIGN, branch relaxation and operand shapes

2026-09-20 14:25:47 +02:00

pcalign_riscv64.s

feat(riscv64,loong64): PCALIGN, branch relaxation and operand shapes

2026-09-20 14:25:47 +02:00

pfx_amd64.s

feat(amd64): LOCK and REP prefixes, literal data pseudo-ops and ADJSP

2026-09-20 14:25:47 +02:00

pseudos_loong64.s

feat(loong64): encode the full LSX and LASX table

2026-09-20 19:14:43 +02:00

qmov_arm64.s

feat(asm): encode the arm64 Q-width FP load and store

2026-10-07 02:36:24 +02:00

quadreg_amd64.s

feat(amd64): emit the quad-register EVEX families

2026-09-21 02:02:19 +02:00

rawdata_amd64.s

feat(amd64): LOCK and REP prefixes, literal data pseudo-ops and ADJSP

2026-09-20 14:25:47 +02:00

rvc_riscv64.s

style: put the repository assembly in gasm fmt canonical form

2026-09-17 20:33:18 +02:00

rvcstore_riscv64.s

fix(riscv64): compressed store offsets, FENCE and branch range checks

2026-09-19 23:49:13 +02:00

scalar_amd64.s

feat(amd64): encode the GOROOT instruction families

2026-09-20 06:44:51 +02:00

semicolons_amd64.s

feat(asm): symbol-valued DATA, division slash in symbols and plain semicolons

2026-09-20 19:15:05 +02:00

shifts_arm64.s

fix(arm64): encode shifts, divides and multiplies and align sizes with emission

2026-09-19 23:49:07 +02:00

simd_arm64.s

fix(format): preserve square brackets in SIMD operands

2026-09-20 09:58:25 +02:00

simdmove_arm64.s

feat(arm64): whole-vector moves, bookkeeping ops and truncating-move lowering

2026-09-20 21:17:20 +02:00

splits_arm64.s

feat(asm): split wide arm64 load and store offsets into REGTMP

2026-10-07 02:36:24 +02:00

sse_amd64.s

feat(amd64): encode the GOROOT instruction families

2026-09-20 06:44:51 +02:00

ssestatic_amd64.s

feat(amd64): assemble the double-shift and static-SB operand shapes

2026-09-20 11:40:39 +02:00

system_amd64.s

feat(amd64): encode the GOROOT instruction families

2026-09-20 06:44:51 +02:00

system_arm64.s

feat(arm64): encode pairs, atomics, crypto, system and NEON slices

2026-09-20 06:44:51 +02:00

vector_arith_add_loong64.s

feat(loong64): encode the full LSX and LASX table

2026-09-20 19:14:43 +02:00

vector_arith_sat_loong64.s

feat(loong64): encode the full LSX and LASX table

2026-09-20 19:14:43 +02:00

vector_arith_sub_loong64.s

feat(loong64): encode the full LSX and LASX table

2026-09-20 19:14:43 +02:00

vector_bitops_loong64.s

feat(loong64): encode the full LSX and LASX table

2026-09-20 19:14:43 +02:00

vector_bounds_loong64.s

feat(loong64): encode the full LSX and LASX table

2026-09-20 19:14:43 +02:00

vector_divmod_loong64.s

feat(loong64): encode the full LSX and LASX table

2026-09-20 19:14:43 +02:00

vector_fp_loong64.s

feat(loong64): encode the full LSX and LASX table

2026-09-20 19:14:43 +02:00

vector_interleave_loong64.s

feat(loong64): encode the full LSX and LASX table

2026-09-20 19:14:43 +02:00

vector_logic_loong64.s

feat(loong64): encode the full LSX and LASX table

2026-09-20 19:14:43 +02:00

vector_loong64.s

feat(riscv64,loong64): encode AMO atomics, vector slices and bit ops

2026-09-20 06:44:51 +02:00

vector_madd_loong64.s

feat(loong64): encode the full LSX and LASX table

2026-09-20 19:14:43 +02:00

vector_maddwiden_loong64.s

feat(loong64): encode the full LSX and LASX table

2026-09-20 19:14:43 +02:00

vector_msub_loong64.s

feat(loong64): encode the full LSX and LASX table

2026-09-20 19:14:43 +02:00

vector_mul_loong64.s

feat(loong64): encode the full LSX and LASX table

2026-09-20 19:14:43 +02:00

vector_mulwiden_loong64.s

feat(loong64): encode the full LSX and LASX table

2026-09-20 19:14:43 +02:00

vector_riscv64.s

feat(riscv64,loong64): encode AMO atomics, vector slices and bit ops

2026-09-20 06:44:51 +02:00

vector_shift_loong64.s

feat(loong64): encode the full LSX and LASX table

2026-09-20 19:14:43 +02:00

vector_shuffle_loong64.s

feat(loong64): encode the full LSX and LASX table

2026-09-20 19:14:43 +02:00

vector_unary_loong64.s

feat(loong64): encode the full LSX and LASX table

2026-09-20 19:14:43 +02:00

vsib_amd64.s

feat(parser): bracket register ranges, index-only VSIB and bare trailing immediates

2026-09-20 22:02:19 +02:00

wideimm_riscv64.s

feat(riscv64,loong64): operand tail, float DATA and honest port classification

2026-09-21 00:44:47 +02:00

widen_amd64.s

feat(amd64): encode the mixed-width extend family and PMOVMSKB

2026-09-20 00:38:24 +02:00

widenimm_arm64.s

feat(arm64): wide immediates, SIMD compare and system operand forms

2026-09-20 14:25:47 +02:00