[AArch64]SME2 single-multi and multi-multi INT/FP dot product instructions
authorCaroline Concatto <caroline.concatto@arm.com>
Sat, 22 Oct 2022 11:13:49 +0000 (12:13 +0100)
committerCaroline Concatto <caroline.concatto@arm.com>
Tue, 25 Oct 2022 17:28:11 +0000 (18:28 +0100)
This patch adds the assembly/disassembly for the following instruction:
INT:
  SDOT (2-way, multiple and single vector): Multi-vector signed integer dot-product by vector.
       (2-way, multiple vectors): Multi-vector signed integer dot-product.
  UDOT (2-way, multiple and single vector): Multi-vector unsigned integer dot-product by vector.
       (2-way, multiple vectors): Multi-vector unsigned integer dot-product.
  SUDOT (multiple and indexed vector): Multi-vector signed by unsigned integer dot-product by indexed elements.
        (multiple and single vector): Multi-vector signed by unsigned integer dot-product by vector.
  USDOT (multiple and single vector): Multi-vector unsigned by signed integer dot-product by vector.
        (multiple vectors): Multi-vector unsigned by signed integer dot-product.
FP:
  BFDOT(multiple and single vector): Multi-vector BFloat16 floating-point dot-product by vector.
        (multiple vectors): Multi-vector BFloat16 floating-point dot-product.

  FDOT (multiple and single vector): Multi-vector half-precision floating-point dot-product by vector.
       (multiple vectors): Multi-vector half-precision floating-point dot-product.
For set of 2 and 4 ZA registers

The reference can be found here:
        https://developer.arm.com/documentation/ddi0602/2022-09

Depends on:D135455

Differential Revision: https://reviews.llvm.org/D135683

llvm/lib/Target/AArch64/AArch64SMEInstrInfo.td
llvm/lib/Target/AArch64/SMEInstrFormats.td
llvm/test/MC/AArch64/SME2/bfdot-diagnostics.s
llvm/test/MC/AArch64/SME2/bfdot.s
llvm/test/MC/AArch64/SME2/fdot.s
llvm/test/MC/AArch64/SME2/sdot.s
llvm/test/MC/AArch64/SME2/sudot.s
llvm/test/MC/AArch64/SME2/udot.s
llvm/test/MC/AArch64/SME2/usdot.s

index 9bbcbb4..b092f25 100644 (file)
@@ -235,30 +235,30 @@ def : Pat<(i64 (AArch64ObscureCopy (i64 GPR64:$idx))),
 // SME2 Instructions
 //===----------------------------------------------------------------------===//
 let Predicates = [HasSME2] in {
-defm ADD_VG2_M2ZZ_S  : sme2_mla_add_sub_array_vg2_single_S<"add", 0b10>;
-defm ADD_VG4_M4ZZ_S  : sme2_mla_add_sub_array_vg4_single_S<"add", 0b10>;
-defm ADD_VG2_M2Z2Z_S : sme2_mla_add_sub_array_vg2_multi_S<"add", 0b10>;
-defm ADD_VG4_M4Z4Z_S : sme2_mla_add_sub_array_vg4_multi_S<"add", 0b10>;
+defm ADD_VG2_M2ZZ_S  : sme2_dot_mla_add_sub_array_vg24_single<"add", 0b0011010, MatrixOp32, ZZ_s, ZPR4b32>;
+defm ADD_VG4_M4ZZ_S  : sme2_dot_mla_add_sub_array_vg24_single<"add", 0b0111010, MatrixOp32, ZZZZ_s, ZPR4b32>;
+defm ADD_VG2_M2Z2Z_S : sme2_dot_mla_add_sub_array_vg2_multi<"add", 0b011010, MatrixOp32, ZZ_s_mul_r>;
+defm ADD_VG4_M4Z4Z_S : sme2_dot_mla_add_sub_array_vg4_multi<"add", 0b011010, MatrixOp32, ZZZZ_s_mul_r>;
 
 defm ADD_VG2_2ZZ  : sme2_int_sve_destructive_vector_vg2_single<"add", 0b0110000>;
 defm ADD_VG4_4ZZ  : sme2_int_sve_destructive_vector_vg4_single<"add", 0b0110000>;
 
-defm SUB_VG2_M2ZZ_S  : sme2_mla_add_sub_array_vg2_single_S<"sub", 0b11>;
-defm SUB_VG4_M4ZZ_S  : sme2_mla_add_sub_array_vg4_single_S<"sub", 0b11>;
-defm SUB_VG2_M2Z2Z_S : sme2_mla_add_sub_array_vg2_multi_S<"sub", 0b11>;
-defm SUB_VG4_M4Z4Z_S : sme2_mla_add_sub_array_vg4_multi_S<"sub", 0b11>;
+defm SUB_VG2_M2ZZ_S  : sme2_dot_mla_add_sub_array_vg24_single<"sub", 0b0011011, MatrixOp32, ZZ_s, ZPR4b32>;
+defm SUB_VG4_M4ZZ_S  : sme2_dot_mla_add_sub_array_vg24_single<"sub", 0b0111011, MatrixOp32, ZZZZ_s, ZPR4b32>;
+defm SUB_VG2_M2Z2Z_S : sme2_dot_mla_add_sub_array_vg2_multi<"sub", 0b011011, MatrixOp32, ZZ_s_mul_r>;
+defm SUB_VG4_M4Z4Z_S : sme2_dot_mla_add_sub_array_vg4_multi<"sub", 0b011011, MatrixOp32, ZZZZ_s_mul_r>;
 
-defm FMLA_VG2_M2ZZ_S  : sme2_mla_add_sub_array_vg2_single_S<"fmla", 0b00>;
-defm FMLA_VG4_M4ZZ_S  : sme2_mla_add_sub_array_vg4_single_S<"fmla", 0b00>;
-defm FMLA_VG2_M2Z2Z_S : sme2_mla_add_sub_array_vg2_multi_S<"fmla", 0b00>;
-defm FMLA_VG4_M4Z4Z_S : sme2_mla_add_sub_array_vg4_multi_S<"fmla", 0b00>;
+defm FMLA_VG2_M2ZZ_S  : sme2_dot_mla_add_sub_array_vg24_single<"fmla", 0b0011000, MatrixOp32, ZZ_s, ZPR4b32>;
+defm FMLA_VG4_M4ZZ_S  : sme2_dot_mla_add_sub_array_vg24_single<"fmla", 0b0111000, MatrixOp32, ZZZZ_s, ZPR4b32>;
+defm FMLA_VG2_M2Z2Z_S : sme2_dot_mla_add_sub_array_vg2_multi<"fmla", 0b011000, MatrixOp32, ZZ_s_mul_r>;
+defm FMLA_VG4_M4Z4Z_S : sme2_dot_mla_add_sub_array_vg4_multi<"fmla", 0b011000, MatrixOp32, ZZZZ_s_mul_r>;
 defm FMLA_VG2_M2ZZI_S : sme2_multi_vec_array_vg2_index_32b<"fmla", 0b0000, ZZ_s_mul_r, ZPR4b32>;
 defm FMLA_VG4_M4ZZI_S : sme2_multi_vec_array_vg4_index_32b<"fmla", 0b0000, ZZZZ_s_mul_r, ZPR4b32>;
 
-defm FMLS_VG2_M2ZZ_S  : sme2_mla_add_sub_array_vg2_single_S<"fmls", 0b01>;
-defm FMLS_VG4_M4ZZ_S  : sme2_mla_add_sub_array_vg4_single_S<"fmls", 0b01>;
-defm FMLS_VG2_M2Z2Z_S : sme2_mla_add_sub_array_vg2_multi_S<"fmls", 0b01>;
-defm FMLS_VG4_M4Z4Z_S : sme2_mla_add_sub_array_vg4_multi_S<"fmls", 0b01>;
+defm FMLS_VG2_M2ZZ_S  : sme2_dot_mla_add_sub_array_vg24_single<"fmls", 0b0011001, MatrixOp32, ZZ_s, ZPR4b32>;
+defm FMLS_VG4_M4ZZ_S  : sme2_dot_mla_add_sub_array_vg24_single<"fmls", 0b0111001, MatrixOp32, ZZZZ_s, ZPR4b32>;
+defm FMLS_VG2_M2Z2Z_S : sme2_dot_mla_add_sub_array_vg2_multi<"fmls", 0b011001, MatrixOp32, ZZ_s_mul_r>;
+defm FMLS_VG4_M4Z4Z_S : sme2_dot_mla_add_sub_array_vg4_multi<"fmls", 0b011001, MatrixOp32, ZZZZ_s_mul_r>;
 defm FMLS_VG2_M2ZZI_S : sme2_multi_vec_array_vg2_index_32b<"fmls", 0b0010, ZZ_s_mul_r, ZPR4b32>;
 defm FMLS_VG4_M4ZZI_S : sme2_multi_vec_array_vg4_index_32b<"fmls", 0b0010, ZZZZ_s_mul_r, ZPR4b32>;
 
@@ -436,9 +436,17 @@ defm UCLAMP_VG4_4Z4Z : sme2_int_clamp_vector_vg4_multi<"uclamp", 0b1>;
 
 defm FDOT_VG2_M2ZZI_HtoS : sme2_multi_vec_array_vg2_index_32b<"fdot", 0b1001, ZZ_h_mul_r, ZPR4b16>;
 defm FDOT_VG4_M4ZZI_HtoS : sme2_multi_vec_array_vg4_index_32b<"fdot", 0b1001, ZZZZ_h_mul_r, ZPR4b16>;
+defm FDOT_VG2_M2ZZ_HtoS  : sme2_dot_mla_add_sub_array_vg24_single<"fdot", 0b0010000, MatrixOp32, ZZ_h, ZPR4b16>;
+defm FDOT_VG4_M4ZZ_HtoS  : sme2_dot_mla_add_sub_array_vg24_single<"fdot", 0b0110000, MatrixOp32, ZZZZ_h, ZPR4b16>;
+defm FDOT_VG2_M2Z2Z_HtoS : sme2_dot_mla_add_sub_array_vg2_multi<"fdot",  0b010000, MatrixOp32, ZZ_h_mul_r>;
+defm FDOT_VG4_M4Z4Z_HtoS : sme2_dot_mla_add_sub_array_vg4_multi<"fdot",  0b010000, MatrixOp32, ZZZZ_h_mul_r>;
 
 defm BFDOT_VG2_M2ZZI_HtoS : sme2_multi_vec_array_vg2_index_32b<"bfdot", 0b1011, ZZ_h_mul_r, ZPR4b16>;
 defm BFDOT_VG4_M4ZZI_HtoS : sme2_multi_vec_array_vg4_index_32b<"bfdot", 0b1011, ZZZZ_h_mul_r, ZPR4b16>;
+defm BFDOT_VG2_M2ZZ_HtoS  : sme2_dot_mla_add_sub_array_vg24_single<"bfdot", 0b0010010, MatrixOp32, ZZ_h, ZPR4b16>;
+defm BFDOT_VG4_M4ZZ_HtoS  : sme2_dot_mla_add_sub_array_vg24_single<"bfdot", 0b0110010, MatrixOp32, ZZZZ_h, ZPR4b16>;
+defm BFDOT_VG2_M2Z2Z_HtoS : sme2_dot_mla_add_sub_array_vg2_multi<"bfdot",  0b010010, MatrixOp32, ZZ_h_mul_r>;
+defm BFDOT_VG4_M4Z4Z_HtoS : sme2_dot_mla_add_sub_array_vg4_multi<"bfdot",  0b010010, MatrixOp32, ZZZZ_h_mul_r>;
 
 defm BFVDOT_VG2_M2ZZI_HtoS : sme2_multi_vec_array_vg2_index_32b<"bfvdot", 0b0011, ZZ_h_mul_r, ZPR4b16>;
 
@@ -448,9 +456,15 @@ defm SDOT_VG2_M2ZZI_HToS : sme2_multi_vec_array_vg2_index_32b<"sdot", 0b1000, ZZ
 defm SDOT_VG2_M2ZZI_BToS : sme2_multi_vec_array_vg2_index_32b<"sdot", 0b1100, ZZ_b_mul_r, ZPR4b8>;
 defm SDOT_VG4_M4ZZI_HToS : sme2_multi_vec_array_vg4_index_32b<"sdot", 0b1000, ZZZZ_h_mul_r, ZPR4b16>;
 defm SDOT_VG4_M4ZZI_BToS : sme2_multi_vec_array_vg4_index_32b<"sdot", 0b1100, ZZZZ_b_mul_r, ZPR4b8>;
+defm SDOT_VG2_M2ZZ_HtoS  : sme2_dot_mla_add_sub_array_vg24_single<"sdot", 0b1010101, MatrixOp32, ZZ_h, ZPR4b16>;
+defm SDOT_VG4_M4ZZ_HtoS  : sme2_dot_mla_add_sub_array_vg24_single<"sdot", 0b1110101, MatrixOp32, ZZZZ_h, ZPR4b16>;
+defm SDOT_VG2_M2Z2Z_HtoS : sme2_dot_mla_add_sub_array_vg2_multi<"sdot", 0b110101, MatrixOp32, ZZ_h_mul_r>;
+defm SDOT_VG4_M4Z4Z_HtoS : sme2_dot_mla_add_sub_array_vg4_multi<"sdot", 0b110101, MatrixOp32, ZZZZ_h_mul_r>;
 
 defm SUDOT_VG2_M2ZZI_BToS : sme2_multi_vec_array_vg2_index_32b<"sudot", 0b1111, ZZ_b_mul_r, ZPR4b8>;
 defm SUDOT_VG4_M4ZZI_BToS : sme2_multi_vec_array_vg4_index_32b<"sudot", 0b1111, ZZZZ_b_mul_r, ZPR4b8>;
+defm SUDOT_VG2_M2ZZ_BToS  : sme2_dot_mla_add_sub_array_vg24_single<"sudot", 0b0010111, MatrixOp32, ZZ_b, ZPR4b8>;
+defm SUDOT_VG4_M4ZZ_BToS  : sme2_dot_mla_add_sub_array_vg24_single<"sudot", 0b0110111, MatrixOp32, ZZZZ_b, ZPR4b8>;
 
 defm SVDOT_VG2_M2ZZI_HtoS : sme2_multi_vec_array_vg2_index_32b<"svdot", 0b0100, ZZ_h_mul_r, ZPR4b16>;
 defm SVDOT_VG4_M4ZZI_BtoS : sme2_multi_vec_array_vg4_index_32b<"svdot", 0b0100, ZZZZ_b_mul_r, ZPR4b8>;
@@ -461,28 +475,35 @@ defm UDOT_VG2_M2ZZI_HToS : sme2_multi_vec_array_vg2_index_32b<"udot", 0b1010, ZZ
 defm UDOT_VG2_M2ZZI_BToS : sme2_multi_vec_array_vg2_index_32b<"udot", 0b1110, ZZ_b_mul_r, ZPR4b8>;
 defm UDOT_VG4_M4ZZI_BtoS : sme2_multi_vec_array_vg4_index_32b<"udot", 0b1110, ZZZZ_b_mul_r, ZPR4b8>;
 defm UDOT_VG4_M4ZZI_HToS : sme2_multi_vec_array_vg4_index_32b<"udot", 0b1010, ZZZZ_h_mul_r, ZPR4b16>;
-
-defm USDOT_VG2_M2ZZI_BToS : sme2_multi_vec_array_vg2_index_32b<"usdot", 0b1101, ZZ_b_mul_r, ZPR4b8>;
-defm USDOT_VG4_M4ZZI_BToS : sme2_multi_vec_array_vg4_index_32b<"usdot", 0b1101, ZZZZ_b_mul_r, ZPR4b8>;
+defm UDOT_VG2_M2ZZ_HtoS  : sme2_dot_mla_add_sub_array_vg24_single<"udot", 0b1010111, MatrixOp32, ZZ_h, ZPR4b16>;
+defm UDOT_VG4_M4ZZ_HtoS  : sme2_dot_mla_add_sub_array_vg24_single<"udot", 0b1110111, MatrixOp32, ZZZZ_h, ZPR4b16>;
+defm UDOT_VG2_M2Z2Z_HtoS : sme2_dot_mla_add_sub_array_vg2_multi<"udot", 0b110111, MatrixOp32, ZZ_h_mul_r>;
+defm UDOT_VG4_M4Z4Z_HtoS : sme2_dot_mla_add_sub_array_vg4_multi<"udot", 0b110111, MatrixOp32, ZZZZ_h_mul_r>;
+
+defm USDOT_VG2_M2ZZI_BToS: sme2_multi_vec_array_vg2_index_32b<"usdot", 0b1101, ZZ_b_mul_r, ZPR4b8>;
+defm USDOT_VG4_M4ZZI_BToS: sme2_multi_vec_array_vg4_index_32b<"usdot", 0b1101, ZZZZ_b_mul_r, ZPR4b8>;
+defm USDOT_VG2_M2ZZ_BToS : sme2_dot_mla_add_sub_array_vg24_single<"usdot",  0b0010101, MatrixOp32, ZZ_b, ZPR4b8>;
+defm USDOT_VG4_M4ZZ_BToS  : sme2_dot_mla_add_sub_array_vg24_single<"usdot", 0b0110101, MatrixOp32, ZZZZ_b, ZPR4b8>;
+defm USDOT_VG2_M2Z2Z_BToS : sme2_dot_mla_add_sub_array_vg2_multi<"usdot", 0b010101, MatrixOp32, ZZ_b_mul_r>;
+defm USDOT_VG4_M4Z4Z_BToS : sme2_dot_mla_add_sub_array_vg4_multi<"usdot", 0b010101, MatrixOp32, ZZZZ_b_mul_r>;
 
 defm USVDOT_VG4_M4ZZI_BToS : sme2_multi_vec_array_vg4_index_32b<"usvdot", 0b0101, ZZZZ_b_mul_r, ZPR4b8>;
 
 defm UVDOT_VG2_M2ZZI_HtoS : sme2_multi_vec_array_vg2_index_32b<"uvdot", 0b0110, ZZ_h_mul_r, ZPR4b16>;
 defm UVDOT_VG4_M4ZZI_BtoS : sme2_multi_vec_array_vg4_index_32b<"uvdot", 0b0110, ZZZZ_b_mul_r, ZPR4b8>;
-
 }
 
 
 let Predicates = [HasSME2, HasSMEI16I64] in {
-defm ADD_VG2_M2ZZ_D  : sme2_mla_add_sub_array_vg2_single_D<"add", 0b10>;
-defm ADD_VG4_M4ZZ_D  : sme2_mla_add_sub_array_vg4_single_D<"add", 0b10>;
-defm ADD_VG2_M2Z2Z_D : sme2_mla_add_sub_array_vg2_multi_D<"add", 0b10>;
-defm ADD_VG4_M4Z4Z_D : sme2_mla_add_sub_array_vg4_multi_D<"add", 0b10>;
+defm ADD_VG2_M2ZZ_D  : sme2_dot_mla_add_sub_array_vg24_single<"add", 0b1011010, MatrixOp64, ZZ_d, ZPR4b64>;
+defm ADD_VG4_M4ZZ_D  : sme2_dot_mla_add_sub_array_vg24_single<"add", 0b1111010, MatrixOp64, ZZZZ_d, ZPR4b64>;
+defm ADD_VG2_M2Z2Z_D : sme2_dot_mla_add_sub_array_vg2_multi<"add", 0b111010, MatrixOp64, ZZ_d_mul_r>;
+defm ADD_VG4_M4Z4Z_D : sme2_dot_mla_add_sub_array_vg4_multi<"add", 0b111010, MatrixOp64, ZZZZ_d_mul_r>;
 
-defm SUB_VG2_M2ZZ_D  : sme2_mla_add_sub_array_vg2_single_D<"sub", 0b11>;
-defm SUB_VG4_M4ZZ_D  : sme2_mla_add_sub_array_vg4_single_D<"sub", 0b11>;
-defm SUB_VG2_M2Z2Z_D : sme2_mla_add_sub_array_vg2_multi_D<"sub", 0b11>;
-defm SUB_VG4_M4Z4Z_D : sme2_mla_add_sub_array_vg4_multi_D<"sub", 0b11>;
+defm SUB_VG2_M2ZZ_D  : sme2_dot_mla_add_sub_array_vg24_single<"sub", 0b1011011, MatrixOp64, ZZ_d, ZPR4b64>;
+defm SUB_VG4_M4ZZ_D  : sme2_dot_mla_add_sub_array_vg24_single<"sub", 0b1111011, MatrixOp64, ZZZZ_d, ZPR4b64>;
+defm SUB_VG2_M2Z2Z_D : sme2_dot_mla_add_sub_array_vg2_multi<"sub", 0b111011, MatrixOp64, ZZ_d_mul_r>;
+defm SUB_VG4_M4Z4Z_D : sme2_dot_mla_add_sub_array_vg4_multi<"sub", 0b111011, MatrixOp64, ZZZZ_d_mul_r>;
 
 defm ADDA_VG2_M2Z2Z_D : sme2_multivec_accum_add_sub_vg2_D<"add", 0b10>;
 defm ADDA_VG4_M4Z4Z_D : sme2_multivec_accum_add_sub_vg4_D<"add", 0b10>;
@@ -502,19 +523,19 @@ defm UVDOT_VG4_M4ZZI_HtoD : sme2_multi_vec_array_vg4_index_64b<"uvdot", 0b111, Z
 }
 
 let Predicates = [HasSME2, HasSMEF64F64] in {
-defm FMLA_VG2_M2ZZ_D  : sme2_mla_add_sub_array_vg2_single_D<"fmla", 0b00>;
-defm FMLA_VG4_M4ZZ_D  : sme2_mla_add_sub_array_vg4_single_D<"fmla", 0b00>;
-defm FMLA_VG2_M2Z2Z_D : sme2_mla_add_sub_array_vg2_multi_D<"fmla", 0b00>;
-defm FMLA_VG4_M4Z4Z_D : sme2_mla_add_sub_array_vg4_multi_D<"fmla", 0b00>;
 defm FMLA_VG2_M2ZZI_D : sme2_multi_vec_array_vg2_index_64b<"fmla", 0b00, ZZ_d_mul_r, ZPR4b64>;
 defm FMLA_VG4_M4ZZI_D : sme2_multi_vec_array_vg4_index_64b<"fmla", 0b000, ZZZZ_d_mul_r, ZPR4b64>;
+defm FMLA_VG2_M2ZZ_D  : sme2_dot_mla_add_sub_array_vg24_single<"fmla", 0b1011000, MatrixOp64, ZZ_d, ZPR4b64>;
+defm FMLA_VG4_M4ZZ_D  : sme2_dot_mla_add_sub_array_vg24_single<"fmla", 0b1111000, MatrixOp64, ZZZZ_d, ZPR4b64>;
+defm FMLA_VG2_M2Z2Z_D : sme2_dot_mla_add_sub_array_vg2_multi<"fmla", 0b111000, MatrixOp64, ZZ_d_mul_r>;
+defm FMLA_VG4_M4Z4Z_D : sme2_dot_mla_add_sub_array_vg4_multi<"fmla", 0b111000, MatrixOp64, ZZZZ_d_mul_r>;
 
-defm FMLS_VG2_M2ZZ_D  : sme2_mla_add_sub_array_vg2_single_D<"fmls", 0b01>;
-defm FMLS_VG4_M4ZZ_D  : sme2_mla_add_sub_array_vg4_single_D<"fmls", 0b01>;
-defm FMLS_VG2_M2Z2Z_D : sme2_mla_add_sub_array_vg2_multi_D<"fmls", 0b01>;
-defm FMLS_VG4_M4Z4Z_D : sme2_mla_add_sub_array_vg4_multi_D<"fmls", 0b01>;
 defm FMLS_VG2_M2ZZI_D : sme2_multi_vec_array_vg2_index_64b<"fmls", 0b10, ZZ_d_mul_r, ZPR4b64>;
 defm FMLS_VG4_M4ZZI_D : sme2_multi_vec_array_vg4_index_64b<"fmls", 0b010, ZZZZ_d_mul_r, ZPR4b64>;
+defm FMLS_VG2_M2ZZ_D  : sme2_dot_mla_add_sub_array_vg24_single<"fmls", 0b1011001, MatrixOp64, ZZ_d, ZPR4b64>;
+defm FMLS_VG4_M4ZZ_D  : sme2_dot_mla_add_sub_array_vg24_single<"fmls", 0b1111001, MatrixOp64, ZZZZ_d, ZPR4b64>;
+defm FMLS_VG2_M2Z2Z_D : sme2_dot_mla_add_sub_array_vg2_multi<"fmls", 0b111001, MatrixOp64, ZZ_d_mul_r>;
+defm FMLS_VG4_M4Z4Z_D : sme2_dot_mla_add_sub_array_vg4_multi<"fmls", 0b111001, MatrixOp64, ZZZZ_d_mul_r>;
 
 defm FADD_VG2_M2Z2Z_D : sme2_multivec_accum_add_sub_vg2_D<"fadd", 0b00>;
 defm FADD_VG4_M4Z4Z_D : sme2_multivec_accum_add_sub_vg4_D<"fadd", 0b00>;
index 4ef8168..ed99d18 100644 (file)
@@ -1194,7 +1194,7 @@ multiclass sve2_int_perm_sel_p<string asm, SDPatternOperator op> {
 //===----------------------------------------------------------------------===//
 // SME2 single-multi ternary int/fp, two/four registers
 
-class sme2_mla_add_sub_array_vg24_single<bit vg4, bit sz, bits<2> op,
+class sme2_dot_mla_add_sub_array_vg24_single<bits<7> op,
                                          MatrixOperand matrix_ty,
                                          RegisterOperand multi_vector_ty,
                                          ZPRRegOp zpr_ty,
@@ -1202,61 +1202,39 @@ class sme2_mla_add_sub_array_vg24_single<bit vg4, bit sz, bits<2> op,
    : I<(outs matrix_ty:$ZAd),
        (ins  matrix_ty:$_ZAd, MatrixIndexGPR32Op8_11:$Rv,
        sme_elm_idx0_7:$imm3, multi_vector_ty:$Zn, zpr_ty:$Zm),
-       mnemonic,"\t$ZAd[$Rv, $imm3, " # !if(vg4, "vgx4", "vgx2") # "], $Zn, $Zm",
+       mnemonic,"\t$ZAd[$Rv, $imm3, " # !if(op{5}, "vgx4", "vgx2") # "], $Zn, $Zm",
        "", []> , Sched<[]> {
   bits<4> Zm;
   bits<5> Zn;
   bits<2> Rv;
   bits<3> imm3;
   let Inst{31-23} = 0b110000010;
-  let Inst{22}    = sz;
+  let Inst{22}    = op{6}; //sz
   let Inst{21}    = 0b1;
-  let Inst{20}    = vg4;
+  let Inst{20}    = op{5}; //vgx4
   let Inst{19-16} = Zm;
   let Inst{15}    = 0b0;
   let Inst{14-13} = Rv;
-  let Inst{12-10} = 0b110;
+  let Inst{12-10} = op{4-2};
   let Inst{9-5}   = Zn;
-  let Inst{4-3}   = op;
+  let Inst{4-3}   = op{1-0};
   let Inst{2-0}   = imm3;
   let Constraints = "$ZAd = $_ZAd";
 }
 
-multiclass sme2_mla_add_sub_array_vg2_single_S<string mnemonic, bits<2> op>{
-  def NAME : sme2_mla_add_sub_array_vg24_single<0b0, 0b0, op, MatrixOp32, ZZ_s,
-                                                ZPR4b32, mnemonic>;
-
-  def : InstAlias<mnemonic # "\t$ZAd[$Rv, $imm3], $Zn, $Zm",
-                 (!cast<Instruction>(NAME) MatrixOp32:$ZAd,  MatrixIndexGPR32Op8_11:$Rv, sme_elm_idx0_7:$imm3, ZZ_s:$Zn, ZPR4b32:$Zm), 0>;
-}
-
-multiclass sme2_mla_add_sub_array_vg2_single_D<string mnemonic, bits<2> op>{
-  def NAME : sme2_mla_add_sub_array_vg24_single<0b0, 0b1, op, MatrixOp64,
-                                                ZZ_d, ZPR4b64, mnemonic>;
-
-  def : InstAlias<mnemonic # "\t$ZAd[$Rv, $imm3], $Zn, $Zm",
-                 (!cast<Instruction>(NAME) MatrixOp64:$ZAd,  MatrixIndexGPR32Op8_11:$Rv, sme_elm_idx0_7:$imm3, ZZ_d:$Zn, ZPR4b64:$Zm), 0>;
-}
-
-multiclass sme2_mla_add_sub_array_vg4_single_S<string mnemonic, bits<2> op>{
-  def NAME : sme2_mla_add_sub_array_vg24_single<0b1, 0b0, op, MatrixOp32, ZZZZ_s,
-                                                ZPR4b32, mnemonic>;
-
-  def : InstAlias<mnemonic # "\t$ZAd[$Rv, $imm3], $Zn, $Zm",
-                 (!cast<Instruction>(NAME) MatrixOp32:$ZAd,  MatrixIndexGPR32Op8_11:$Rv, sme_elm_idx0_7:$imm3, ZZZZ_s:$Zn, ZPR4b32:$Zm), 0>;
-}
-
-multiclass sme2_mla_add_sub_array_vg4_single_D<string mnemonic, bits<2> op>{
-  def NAME : sme2_mla_add_sub_array_vg24_single<0b1, 0b1, op, MatrixOp64, ZZZZ_d,
-                                                ZPR4b64, mnemonic>;
+multiclass sme2_dot_mla_add_sub_array_vg24_single<string mnemonic, bits<7> op,
+                                              MatrixOperand matrix_ty,
+                                              RegisterOperand multi_vector_ty,
+                                              ZPRRegOp zpr_ty>{
+  def NAME: sme2_dot_mla_add_sub_array_vg24_single<op, matrix_ty, multi_vector_ty, zpr_ty, mnemonic>;
 
   def : InstAlias<mnemonic # "\t$ZAd[$Rv, $imm3], $Zn, $Zm",
-                 (!cast<Instruction>(NAME) MatrixOp64:$ZAd,  MatrixIndexGPR32Op8_11:$Rv, sme_elm_idx0_7:$imm3, ZZZZ_d:$Zn, ZPR4b64:$Zm), 0>;
+                 (!cast<Instruction>(NAME) matrix_ty:$ZAd,  MatrixIndexGPR32Op8_11:$Rv, sme_elm_idx0_7:$imm3, multi_vector_ty:$Zn, zpr_ty:$Zm), 0>;
 }
 
 //===----------------------------------------------------------------------===//
 // SME2 multiple vectors ternary INT/FP  two and four registers
-class sme2_mla_add_sub_array_vg2_multi<bit sz, bits<2> op,
+class sme2_dot_mla_add_sub_array_vg2_multi<bits<6> op,
                                        MatrixOperand matrix_ty,
                                        RegisterOperand multi_vector_ty,
                                        string mnemonic>
@@ -1270,34 +1248,30 @@ class sme2_mla_add_sub_array_vg2_multi<bit sz, bits<2> op,
   bits<2> Rv;
   bits<3> imm3;
   let Inst{31-23} = 0b110000011;
-  let Inst{22}    = sz;
+  let Inst{22}    = op{5}; //sz
   let Inst{21}    = 0b1;
   let Inst{20-17} = Zm;
   let Inst{16-15} = 0b00;
   let Inst{14-13} = Rv;
-  let Inst{12-10} = 0b110;
+  let Inst{12-10} = op{4-2};
   let Inst{9-6}   = Zn;
   let Inst{5}     = 0b0;
-  let Inst{4-3}   = op;
+  let Inst{4-3}   = op{1-0};
   let Inst{2-0}   = imm3;
   let Constraints = "$ZAd = $_ZAd";
 }
 
-multiclass sme2_mla_add_sub_array_vg2_multi_S<string mnemonic, bits<2> op>{
-  def NAME : sme2_mla_add_sub_array_vg2_multi<0b0, op, MatrixOp32, ZZ_s_mul_r, mnemonic>;
-
-  def : InstAlias<mnemonic # "\t$ZAd[$Rv, $imm3], $Zn, $Zm",
-                 (!cast<Instruction>(NAME) MatrixOp32:$ZAd,  MatrixIndexGPR32Op8_11:$Rv, sme_elm_idx0_7:$imm3, ZZ_s_mul_r:$Zn, ZZ_s_mul_r:$Zm), 0>;
-}
+multiclass sme2_dot_mla_add_sub_array_vg2_multi<string mnemonic, bits<6> op,
+                                            MatrixOperand  matrix_ty,
+                                            RegisterOperand multi_vector_ty>{
+ def NAME : sme2_dot_mla_add_sub_array_vg2_multi<op, matrix_ty, multi_vector_ty, mnemonic>;
 
-multiclass sme2_mla_add_sub_array_vg2_multi_D<string mnemonic, bits<2> op>{
-  def NAME : sme2_mla_add_sub_array_vg2_multi<0b1, op, MatrixOp64, ZZ_d_mul_r, mnemonic>;
+   def : InstAlias<mnemonic # "\t$ZAd[$Rv, $imm3], $Zn, $Zm",
+                 (!cast<Instruction>(NAME) matrix_ty:$ZAd,  MatrixIndexGPR32Op8_11:$Rv, sme_elm_idx0_7:$imm3, multi_vector_ty:$Zn, multi_vector_ty:$Zm), 0>;
 
-  def : InstAlias<mnemonic # "\t$ZAd[$Rv, $imm3], $Zn, $Zm",
-                 (!cast<Instruction>(NAME) MatrixOp64:$ZAd,  MatrixIndexGPR32Op8_11:$Rv, sme_elm_idx0_7:$imm3, ZZ_d_mul_r:$Zn, ZZ_d_mul_r:$Zm), 0>;
 }
 
-class sme2_mla_add_sub_array_vg4_multi_base<bit sz, bits<2> op,
+class sme2_dot_mla_add_sub_array_vg4_multi<bits<6> op,
                                             MatrixOperand matrix_ty,
                                             RegisterOperand multi_vector_ty,
                                             string mnemonic>
@@ -1311,31 +1285,28 @@ class sme2_mla_add_sub_array_vg4_multi_base<bit sz, bits<2> op,
   bits<2> Rv;
   bits<3> imm3;
   let Inst{31-23} = 0b110000011;
-  let Inst{22}    = sz;
+  let Inst{22}    = op{5}; //sz
   let Inst{21}    = 0b1;
   let Inst{20-18} = Zm;
   let Inst{17-15} = 0b010;
   let Inst{14-13} = Rv;
-  let Inst{12-10} = 0b110;
+  let Inst{12-10} = op{4-2};
   let Inst{9-7}   = Zn;
   let Inst{6-5}   = 0b00;
-  let Inst{4-3}   = op;
+  let Inst{4-3}   = op{1-0};
   let Inst{2-0}   = imm3;
   let Constraints = "$ZAd = $_ZAd";
 }
 
-multiclass sme2_mla_add_sub_array_vg4_multi_S<string mnemonic, bits<2> op>{
-  def NAME : sme2_mla_add_sub_array_vg4_multi_base<0b0, op, MatrixOp32, ZZZZ_s_mul_r, mnemonic>;
 
-  def : InstAlias<mnemonic # "\t$ZAd[$Rv, $imm3], $Zn, $Zm",
-                 (!cast<Instruction>(NAME) MatrixOp32:$ZAd,  MatrixIndexGPR32Op8_11:$Rv, sme_elm_idx0_7:$imm3, ZZZZ_s_mul_r:$Zn, ZZZZ_s_mul_r:$Zm), 0>;
-}
+multiclass sme2_dot_mla_add_sub_array_vg4_multi<string mnemonic, bits<6> op,
+                                            MatrixOperand  matrix_ty,
+                                            RegisterOperand multi_vector_ty>{
+ def NAME : sme2_dot_mla_add_sub_array_vg4_multi<op, matrix_ty, multi_vector_ty, mnemonic>;
 
-multiclass sme2_mla_add_sub_array_vg4_multi_D<string mnemonic, bits<2> op>{
-  def NAME : sme2_mla_add_sub_array_vg4_multi_base<0b1, op, MatrixOp64, ZZZZ_d_mul_r, mnemonic>;
+   def : InstAlias<mnemonic # "\t$ZAd[$Rv, $imm3], $Zn, $Zm",
+                 (!cast<Instruction>(NAME) matrix_ty:$ZAd,  MatrixIndexGPR32Op8_11:$Rv, sme_elm_idx0_7:$imm3, multi_vector_ty:$Zn, multi_vector_ty:$Zm), 0>;
 
-  def : InstAlias<mnemonic # "\t$ZAd[$Rv, $imm3], $Zn, $Zm",
-                 (!cast<Instruction>(NAME) MatrixOp64:$ZAd,  MatrixIndexGPR32Op8_11:$Rv, sme_elm_idx0_7:$imm3, ZZZZ_d_mul_r:$Zn, ZZZZ_d_mul_r:$Zm), 0>;
 }
 
 //===----------------------------------------------------------------------===//
index df1a049..3b75ae3 100644 (file)
@@ -8,9 +8,32 @@ bfdot za.s[w8, 0, vgx2], {z0.h-z2.h}, z0.h[0]
 // CHECK-NEXT: bfdot za.s[w8, 0, vgx2], {z0.h-z2.h}, z0.h[0]
 // CHECK-NOT: [[@LINE-1]]:{{[0-9]+}}:
 
+bfdot za.s[w8, 0, vgx4], {z1.h-z5.h}, z0.h
+// CHECK: [[@LINE-1]]:{{[0-9]+}}: error: invalid number of vectors
+// CHECK-NEXT: bfdot za.s[w8, 0, vgx4], {z1.h-z5.h}, z0.h
+// CHECK-NOT: [[@LINE-1]]:{{[0-9]+}}:
+
+bfdot za.s[w8, 0, vgx2], {z0.h-z1.h}, {z3.h-z4.h}
+// CHECK: [[@LINE-1]]:{{[0-9]+}}: error: Invalid vector list, expected list with 2 consecutive SVE vectors, where the first vector is a multiple of 2 and with matching element types
+// CHECK-NEXT: bfdot za.s[w8, 0, vgx2], {z0.h-z1.h}, {z3.h-z4.h}
+// CHECK-NOT: [[@LINE-1]]:{{[0-9]+}}:
+
+// --------------------------------------------------------------------------//
+// Invalid single vector register
+
+bfdot za.s[w8, 0, vgx4], {z0.h-z3.h}, z16.h
+// CHECK: [[@LINE-1]]:{{[0-9]+}}: error: Invalid restricted vector register, expected z0.h..z15.h
+// CHECK-NEXT: bfdot za.s[w8, 0, vgx4], {z0.h-z3.h}, z16.h
+// CHECK-NOT: [[@LINE-1]]:{{[0-9]+}}:
+
 // --------------------------------------------------------------------------//
 // Invalid vector select register
 
+bfdot za.s[w7, 0, vgx2], {z0.h-z1.h}, {z3.h-z4.h}
+// CHECK: [[@LINE-1]]:{{[0-9]+}}: error: operand must be a register in range [w8, w11]
+// CHECK-NEXT: bfdot za.s[w7, 0, vgx2], {z0.h-z1.h}, {z3.h-z4.h}
+// CHECK-NOT: [[@LINE-1]]:{{[0-9]+}}:
+
 bfdot za.s[w12, 0, vgx4], {z0.h-z3.h}, z0.h[0]
 // CHECK: [[@LINE-1]]:{{[0-9]+}}: error: operand must be a register in range [w8, w11]
 // CHECK-NEXT: bfdot za.s[w12, 0, vgx4], {z0.h-z3.h}, z0.h[0]
@@ -29,6 +52,10 @@ bfdot za.s[w8, 8, vgx4], {z0.h-z3.h}, z0.h[0]
 // CHECK-NEXT: bfdot za.s[w8, 8, vgx4], {z0.h-z3.h}, z0.h[0]
 // CHECK-NOT: [[@LINE-1]]:{{[0-9]+}}:
 
+bfdot za.s[w8, -1, vgx2], {z0.h-z1.h}, {z3.h-z4.h}
+// CHECK: [[@LINE-1]]:{{[0-9]+}}: error: immediate must be an integer in range [0, 7].
+// CHECK-NEXT: bfdot za.s[w8, -1, vgx2], {z0.h-z1.h}, {z3.h-z4.h}
+// CHECK-NOT: [[@LINE-1]]:{{[0-9]+}}:
 
 // --------------------------------------------------------------------------//
 // Invalid Register Suffix
index 711c3f3..25d10bc 100644 (file)
 // RUN:        | FileCheck %s --check-prefixes=CHECK-ENCODING,CHECK-INST
 
 
+bfdot   za.s[w8, 0, vgx2], {z0.h, z1.h}, z0.h  // 11000001-00100000-00010000-00010000
+// CHECK-INST: bfdot   za.s[w8, 0, vgx2], { z0.h, z1.h }, z0.h
+// CHECK-ENCODING: [0x10,0x10,0x20,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1201010 <unknown>
+
+bfdot   za.s[w8, 0], {z0.h, z1.h}, z0.h  // 11000001-00100000-00010000-00010000
+// CHECK-INST: bfdot   za.s[w8, 0, vgx2], { z0.h, z1.h }, z0.h
+// CHECK-ENCODING: [0x10,0x10,0x20,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1201010 <unknown>
+
+bfdot   za.s[w10, 5, vgx2], {z10.h, z11.h}, z5.h  // 11000001-00100101-01010001-01010101
+// CHECK-INST: bfdot   za.s[w10, 5, vgx2], { z10.h, z11.h }, z5.h
+// CHECK-ENCODING: [0x55,0x51,0x25,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1255155 <unknown>
+
+bfdot   za.s[w10, 5], {z10.h, z11.h}, z5.h  // 11000001-00100101-01010001-01010101
+// CHECK-INST: bfdot   za.s[w10, 5, vgx2], { z10.h, z11.h }, z5.h
+// CHECK-ENCODING: [0x55,0x51,0x25,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1255155 <unknown>
+
+bfdot   za.s[w11, 7, vgx2], {z13.h, z14.h}, z8.h  // 11000001-00101000-01110001-10110111
+// CHECK-INST: bfdot   za.s[w11, 7, vgx2], { z13.h, z14.h }, z8.h
+// CHECK-ENCODING: [0xb7,0x71,0x28,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c12871b7 <unknown>
+
+bfdot   za.s[w11, 7], {z13.h, z14.h}, z8.h  // 11000001-00101000-01110001-10110111
+// CHECK-INST: bfdot   za.s[w11, 7, vgx2], { z13.h, z14.h }, z8.h
+// CHECK-ENCODING: [0xb7,0x71,0x28,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c12871b7 <unknown>
+
+bfdot   za.s[w11, 7, vgx2], {z31.h, z0.h}, z15.h  // 11000001-00101111-01110011-11110111
+// CHECK-INST: bfdot   za.s[w11, 7, vgx2], { z31.h, z0.h }, z15.h
+// CHECK-ENCODING: [0xf7,0x73,0x2f,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c12f73f7 <unknown>
+
+bfdot   za.s[w11, 7], {z31.h, z0.h}, z15.h  // 11000001-00101111-01110011-11110111
+// CHECK-INST: bfdot   za.s[w11, 7, vgx2], { z31.h, z0.h }, z15.h
+// CHECK-ENCODING: [0xf7,0x73,0x2f,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c12f73f7 <unknown>
+
+bfdot   za.s[w8, 5, vgx2], {z17.h, z18.h}, z0.h  // 11000001-00100000-00010010-00110101
+// CHECK-INST: bfdot   za.s[w8, 5, vgx2], { z17.h, z18.h }, z0.h
+// CHECK-ENCODING: [0x35,0x12,0x20,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1201235 <unknown>
+
+bfdot   za.s[w8, 5], {z17.h, z18.h}, z0.h  // 11000001-00100000-00010010-00110101
+// CHECK-INST: bfdot   za.s[w8, 5, vgx2], { z17.h, z18.h }, z0.h
+// CHECK-ENCODING: [0x35,0x12,0x20,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1201235 <unknown>
+
+bfdot   za.s[w8, 1, vgx2], {z1.h, z2.h}, z14.h  // 11000001-00101110-00010000-00110001
+// CHECK-INST: bfdot   za.s[w8, 1, vgx2], { z1.h, z2.h }, z14.h
+// CHECK-ENCODING: [0x31,0x10,0x2e,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c12e1031 <unknown>
+
+bfdot   za.s[w8, 1], {z1.h, z2.h}, z14.h  // 11000001-00101110-00010000-00110001
+// CHECK-INST: bfdot   za.s[w8, 1, vgx2], { z1.h, z2.h }, z14.h
+// CHECK-ENCODING: [0x31,0x10,0x2e,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c12e1031 <unknown>
+
+bfdot   za.s[w10, 0, vgx2], {z19.h, z20.h}, z4.h  // 11000001-00100100-01010010-01110000
+// CHECK-INST: bfdot   za.s[w10, 0, vgx2], { z19.h, z20.h }, z4.h
+// CHECK-ENCODING: [0x70,0x52,0x24,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1245270 <unknown>
+
+bfdot   za.s[w10, 0], {z19.h, z20.h}, z4.h  // 11000001-00100100-01010010-01110000
+// CHECK-INST: bfdot   za.s[w10, 0, vgx2], { z19.h, z20.h }, z4.h
+// CHECK-ENCODING: [0x70,0x52,0x24,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1245270 <unknown>
+
+bfdot   za.s[w8, 0, vgx2], {z12.h, z13.h}, z2.h  // 11000001-00100010-00010001-10010000
+// CHECK-INST: bfdot   za.s[w8, 0, vgx2], { z12.h, z13.h }, z2.h
+// CHECK-ENCODING: [0x90,0x11,0x22,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1221190 <unknown>
+
+bfdot   za.s[w8, 0], {z12.h, z13.h}, z2.h  // 11000001-00100010-00010001-10010000
+// CHECK-INST: bfdot   za.s[w8, 0, vgx2], { z12.h, z13.h }, z2.h
+// CHECK-ENCODING: [0x90,0x11,0x22,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1221190 <unknown>
+
+bfdot   za.s[w10, 1, vgx2], {z1.h, z2.h}, z10.h  // 11000001-00101010-01010000-00110001
+// CHECK-INST: bfdot   za.s[w10, 1, vgx2], { z1.h, z2.h }, z10.h
+// CHECK-ENCODING: [0x31,0x50,0x2a,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c12a5031 <unknown>
+
+bfdot   za.s[w10, 1], {z1.h, z2.h}, z10.h  // 11000001-00101010-01010000-00110001
+// CHECK-INST: bfdot   za.s[w10, 1, vgx2], { z1.h, z2.h }, z10.h
+// CHECK-ENCODING: [0x31,0x50,0x2a,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c12a5031 <unknown>
+
+bfdot   za.s[w8, 5, vgx2], {z22.h, z23.h}, z14.h  // 11000001-00101110-00010010-11010101
+// CHECK-INST: bfdot   za.s[w8, 5, vgx2], { z22.h, z23.h }, z14.h
+// CHECK-ENCODING: [0xd5,0x12,0x2e,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c12e12d5 <unknown>
+
+bfdot   za.s[w8, 5], {z22.h, z23.h}, z14.h  // 11000001-00101110-00010010-11010101
+// CHECK-INST: bfdot   za.s[w8, 5, vgx2], { z22.h, z23.h }, z14.h
+// CHECK-ENCODING: [0xd5,0x12,0x2e,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c12e12d5 <unknown>
+
+bfdot   za.s[w11, 2, vgx2], {z9.h, z10.h}, z1.h  // 11000001-00100001-01110001-00110010
+// CHECK-INST: bfdot   za.s[w11, 2, vgx2], { z9.h, z10.h }, z1.h
+// CHECK-ENCODING: [0x32,0x71,0x21,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1217132 <unknown>
+
+bfdot   za.s[w11, 2], {z9.h, z10.h}, z1.h  // 11000001-00100001-01110001-00110010
+// CHECK-INST: bfdot   za.s[w11, 2, vgx2], { z9.h, z10.h }, z1.h
+// CHECK-ENCODING: [0x32,0x71,0x21,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1217132 <unknown>
+
+bfdot   za.s[w9, 7, vgx2], {z12.h, z13.h}, z11.h  // 11000001-00101011-00110001-10010111
+// CHECK-INST: bfdot   za.s[w9, 7, vgx2], { z12.h, z13.h }, z11.h
+// CHECK-ENCODING: [0x97,0x31,0x2b,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c12b3197 <unknown>
+
+bfdot   za.s[w9, 7], {z12.h, z13.h}, z11.h  // 11000001-00101011-00110001-10010111
+// CHECK-INST: bfdot   za.s[w9, 7, vgx2], { z12.h, z13.h }, z11.h
+// CHECK-ENCODING: [0x97,0x31,0x2b,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c12b3197 <unknown>
+
+
 bfdot   za.s[w8, 0, vgx2], {z0.h, z1.h}, z0.h[0]  // 11000001-01010000-00010000-00011000
 // CHECK-INST: bfdot   za.s[w8, 0, vgx2], { z0.h, z1.h }, z0.h[0]
 // CHECK-ENCODING: [0x18,0x10,0x50,0xc1]
@@ -157,6 +302,296 @@ bfdot   za.s[w9, 7], {z12.h, z13.h}, z11.h[2]  // 11000001-01011011-00111001-100
 // CHECK-UNKNOWN: c15b399f <unknown>
 
 
+bfdot   za.s[w8, 0, vgx2], {z0.h, z1.h}, {z0.h, z1.h}  // 11000001-10100000-00010000-00010000
+// CHECK-INST: bfdot   za.s[w8, 0, vgx2], { z0.h, z1.h }, { z0.h, z1.h }
+// CHECK-ENCODING: [0x10,0x10,0xa0,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1a01010 <unknown>
+
+bfdot   za.s[w8, 0], {z0.h, z1.h}, {z0.h, z1.h}  // 11000001-10100000-00010000-00010000
+// CHECK-INST: bfdot   za.s[w8, 0, vgx2], { z0.h, z1.h }, { z0.h, z1.h }
+// CHECK-ENCODING: [0x10,0x10,0xa0,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1a01010 <unknown>
+
+bfdot   za.s[w10, 5, vgx2], {z10.h, z11.h}, {z20.h, z21.h}  // 11000001-10110100-01010001-01010101
+// CHECK-INST: bfdot   za.s[w10, 5, vgx2], { z10.h, z11.h }, { z20.h, z21.h }
+// CHECK-ENCODING: [0x55,0x51,0xb4,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1b45155 <unknown>
+
+bfdot   za.s[w10, 5], {z10.h, z11.h}, {z20.h, z21.h}  // 11000001-10110100-01010001-01010101
+// CHECK-INST: bfdot   za.s[w10, 5, vgx2], { z10.h, z11.h }, { z20.h, z21.h }
+// CHECK-ENCODING: [0x55,0x51,0xb4,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1b45155 <unknown>
+
+bfdot   za.s[w11, 7, vgx2], {z12.h, z13.h}, {z8.h, z9.h}  // 11000001-10101000-01110001-10010111
+// CHECK-INST: bfdot   za.s[w11, 7, vgx2], { z12.h, z13.h }, { z8.h, z9.h }
+// CHECK-ENCODING: [0x97,0x71,0xa8,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1a87197 <unknown>
+
+bfdot   za.s[w11, 7], {z12.h, z13.h}, {z8.h, z9.h}  // 11000001-10101000-01110001-10010111
+// CHECK-INST: bfdot   za.s[w11, 7, vgx2], { z12.h, z13.h }, { z8.h, z9.h }
+// CHECK-ENCODING: [0x97,0x71,0xa8,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1a87197 <unknown>
+
+bfdot   za.s[w11, 7, vgx2], {z30.h, z31.h}, {z30.h, z31.h}  // 11000001-10111110-01110011-11010111
+// CHECK-INST: bfdot   za.s[w11, 7, vgx2], { z30.h, z31.h }, { z30.h, z31.h }
+// CHECK-ENCODING: [0xd7,0x73,0xbe,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1be73d7 <unknown>
+
+bfdot   za.s[w11, 7], {z30.h, z31.h}, {z30.h, z31.h}  // 11000001-10111110-01110011-11010111
+// CHECK-INST: bfdot   za.s[w11, 7, vgx2], { z30.h, z31.h }, { z30.h, z31.h }
+// CHECK-ENCODING: [0xd7,0x73,0xbe,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1be73d7 <unknown>
+
+bfdot   za.s[w8, 5, vgx2], {z16.h, z17.h}, {z16.h, z17.h}  // 11000001-10110000-00010010-00010101
+// CHECK-INST: bfdot   za.s[w8, 5, vgx2], { z16.h, z17.h }, { z16.h, z17.h }
+// CHECK-ENCODING: [0x15,0x12,0xb0,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1b01215 <unknown>
+
+bfdot   za.s[w8, 5], {z16.h, z17.h}, {z16.h, z17.h}  // 11000001-10110000-00010010-00010101
+// CHECK-INST: bfdot   za.s[w8, 5, vgx2], { z16.h, z17.h }, { z16.h, z17.h }
+// CHECK-ENCODING: [0x15,0x12,0xb0,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1b01215 <unknown>
+
+bfdot   za.s[w8, 1, vgx2], {z0.h, z1.h}, {z30.h, z31.h}  // 11000001-10111110-00010000-00010001
+// CHECK-INST: bfdot   za.s[w8, 1, vgx2], { z0.h, z1.h }, { z30.h, z31.h }
+// CHECK-ENCODING: [0x11,0x10,0xbe,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1be1011 <unknown>
+
+bfdot   za.s[w8, 1], {z0.h, z1.h}, {z30.h, z31.h}  // 11000001-10111110-00010000-00010001
+// CHECK-INST: bfdot   za.s[w8, 1, vgx2], { z0.h, z1.h }, { z30.h, z31.h }
+// CHECK-ENCODING: [0x11,0x10,0xbe,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1be1011 <unknown>
+
+bfdot   za.s[w10, 0, vgx2], {z18.h, z19.h}, {z20.h, z21.h}  // 11000001-10110100-01010010-01010000
+// CHECK-INST: bfdot   za.s[w10, 0, vgx2], { z18.h, z19.h }, { z20.h, z21.h }
+// CHECK-ENCODING: [0x50,0x52,0xb4,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1b45250 <unknown>
+
+bfdot   za.s[w10, 0], {z18.h, z19.h}, {z20.h, z21.h}  // 11000001-10110100-01010010-01010000
+// CHECK-INST: bfdot   za.s[w10, 0, vgx2], { z18.h, z19.h }, { z20.h, z21.h }
+// CHECK-ENCODING: [0x50,0x52,0xb4,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1b45250 <unknown>
+
+bfdot   za.s[w8, 0, vgx2], {z12.h, z13.h}, {z2.h, z3.h}  // 11000001-10100010-00010001-10010000
+// CHECK-INST: bfdot   za.s[w8, 0, vgx2], { z12.h, z13.h }, { z2.h, z3.h }
+// CHECK-ENCODING: [0x90,0x11,0xa2,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1a21190 <unknown>
+
+bfdot   za.s[w8, 0], {z12.h, z13.h}, {z2.h, z3.h}  // 11000001-10100010-00010001-10010000
+// CHECK-INST: bfdot   za.s[w8, 0, vgx2], { z12.h, z13.h }, { z2.h, z3.h }
+// CHECK-ENCODING: [0x90,0x11,0xa2,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1a21190 <unknown>
+
+bfdot   za.s[w10, 1, vgx2], {z0.h, z1.h}, {z26.h, z27.h}  // 11000001-10111010-01010000-00010001
+// CHECK-INST: bfdot   za.s[w10, 1, vgx2], { z0.h, z1.h }, { z26.h, z27.h }
+// CHECK-ENCODING: [0x11,0x50,0xba,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1ba5011 <unknown>
+
+bfdot   za.s[w10, 1], {z0.h, z1.h}, {z26.h, z27.h}  // 11000001-10111010-01010000-00010001
+// CHECK-INST: bfdot   za.s[w10, 1, vgx2], { z0.h, z1.h }, { z26.h, z27.h }
+// CHECK-ENCODING: [0x11,0x50,0xba,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1ba5011 <unknown>
+
+bfdot   za.s[w8, 5, vgx2], {z22.h, z23.h}, {z30.h, z31.h}  // 11000001-10111110-00010010-11010101
+// CHECK-INST: bfdot   za.s[w8, 5, vgx2], { z22.h, z23.h }, { z30.h, z31.h }
+// CHECK-ENCODING: [0xd5,0x12,0xbe,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1be12d5 <unknown>
+
+bfdot   za.s[w8, 5], {z22.h, z23.h}, {z30.h, z31.h}  // 11000001-10111110-00010010-11010101
+// CHECK-INST: bfdot   za.s[w8, 5, vgx2], { z22.h, z23.h }, { z30.h, z31.h }
+// CHECK-ENCODING: [0xd5,0x12,0xbe,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1be12d5 <unknown>
+
+bfdot   za.s[w11, 2, vgx2], {z8.h, z9.h}, {z0.h, z1.h}  // 11000001-10100000-01110001-00010010
+// CHECK-INST: bfdot   za.s[w11, 2, vgx2], { z8.h, z9.h }, { z0.h, z1.h }
+// CHECK-ENCODING: [0x12,0x71,0xa0,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1a07112 <unknown>
+
+bfdot   za.s[w11, 2], {z8.h, z9.h}, {z0.h, z1.h}  // 11000001-10100000-01110001-00010010
+// CHECK-INST: bfdot   za.s[w11, 2, vgx2], { z8.h, z9.h }, { z0.h, z1.h }
+// CHECK-ENCODING: [0x12,0x71,0xa0,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1a07112 <unknown>
+
+bfdot   za.s[w9, 7, vgx2], {z12.h, z13.h}, {z10.h, z11.h}  // 11000001-10101010-00110001-10010111
+// CHECK-INST: bfdot   za.s[w9, 7, vgx2], { z12.h, z13.h }, { z10.h, z11.h }
+// CHECK-ENCODING: [0x97,0x31,0xaa,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1aa3197 <unknown>
+
+bfdot   za.s[w9, 7], {z12.h, z13.h}, {z10.h, z11.h}  // 11000001-10101010-00110001-10010111
+// CHECK-INST: bfdot   za.s[w9, 7, vgx2], { z12.h, z13.h }, { z10.h, z11.h }
+// CHECK-ENCODING: [0x97,0x31,0xaa,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1aa3197 <unknown>
+
+
+bfdot   za.s[w8, 0, vgx4], {z0.h - z3.h}, z0.h  // 11000001-00110000-00010000-00010000
+// CHECK-INST: bfdot   za.s[w8, 0, vgx4], { z0.h - z3.h }, z0.h
+// CHECK-ENCODING: [0x10,0x10,0x30,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1301010 <unknown>
+
+bfdot   za.s[w8, 0], {z0.h - z3.h}, z0.h  // 11000001-00110000-00010000-00010000
+// CHECK-INST: bfdot   za.s[w8, 0, vgx4], { z0.h - z3.h }, z0.h
+// CHECK-ENCODING: [0x10,0x10,0x30,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1301010 <unknown>
+
+bfdot   za.s[w10, 5, vgx4], {z10.h - z13.h}, z5.h  // 11000001-00110101-01010001-01010101
+// CHECK-INST: bfdot   za.s[w10, 5, vgx4], { z10.h - z13.h }, z5.h
+// CHECK-ENCODING: [0x55,0x51,0x35,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1355155 <unknown>
+
+bfdot   za.s[w10, 5], {z10.h - z13.h}, z5.h  // 11000001-00110101-01010001-01010101
+// CHECK-INST: bfdot   za.s[w10, 5, vgx4], { z10.h - z13.h }, z5.h
+// CHECK-ENCODING: [0x55,0x51,0x35,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1355155 <unknown>
+
+bfdot   za.s[w11, 7, vgx4], {z13.h - z16.h}, z8.h  // 11000001-00111000-01110001-10110111
+// CHECK-INST: bfdot   za.s[w11, 7, vgx4], { z13.h - z16.h }, z8.h
+// CHECK-ENCODING: [0xb7,0x71,0x38,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c13871b7 <unknown>
+
+bfdot   za.s[w11, 7], {z13.h - z16.h}, z8.h  // 11000001-00111000-01110001-10110111
+// CHECK-INST: bfdot   za.s[w11, 7, vgx4], { z13.h - z16.h }, z8.h
+// CHECK-ENCODING: [0xb7,0x71,0x38,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c13871b7 <unknown>
+
+bfdot   za.s[w11, 7, vgx4], {z31.h - z2.h}, z15.h  // 11000001-00111111-01110011-11110111
+// CHECK-INST: bfdot   za.s[w11, 7, vgx4], {  z31.h, z0.h, z1.h, z2.h }, z15.h
+// CHECK-ENCODING: [0xf7,0x73,0x3f,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c13f73f7 <unknown>
+
+bfdot   za.s[w11, 7], {z31.h - z2.h}, z15.h  // 11000001-00111111-01110011-11110111
+// CHECK-INST: bfdot   za.s[w11, 7, vgx4], {  z31.h, z0.h, z1.h, z2.h }, z15.h
+// CHECK-ENCODING: [0xf7,0x73,0x3f,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c13f73f7 <unknown>
+
+bfdot   za.s[w8, 5, vgx4], {z17.h - z20.h}, z0.h  // 11000001-00110000-00010010-00110101
+// CHECK-INST: bfdot   za.s[w8, 5, vgx4], { z17.h - z20.h }, z0.h
+// CHECK-ENCODING: [0x35,0x12,0x30,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1301235 <unknown>
+
+bfdot   za.s[w8, 5], {z17.h - z20.h}, z0.h  // 11000001-00110000-00010010-00110101
+// CHECK-INST: bfdot   za.s[w8, 5, vgx4], { z17.h - z20.h }, z0.h
+// CHECK-ENCODING: [0x35,0x12,0x30,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1301235 <unknown>
+
+bfdot   za.s[w8, 1, vgx4], {z1.h - z4.h}, z14.h  // 11000001-00111110-00010000-00110001
+// CHECK-INST: bfdot   za.s[w8, 1, vgx4], { z1.h - z4.h }, z14.h
+// CHECK-ENCODING: [0x31,0x10,0x3e,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c13e1031 <unknown>
+
+bfdot   za.s[w8, 1], {z1.h - z4.h}, z14.h  // 11000001-00111110-00010000-00110001
+// CHECK-INST: bfdot   za.s[w8, 1, vgx4], { z1.h - z4.h }, z14.h
+// CHECK-ENCODING: [0x31,0x10,0x3e,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c13e1031 <unknown>
+
+bfdot   za.s[w10, 0, vgx4], {z19.h - z22.h}, z4.h  // 11000001-00110100-01010010-01110000
+// CHECK-INST: bfdot   za.s[w10, 0, vgx4], { z19.h - z22.h }, z4.h
+// CHECK-ENCODING: [0x70,0x52,0x34,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1345270 <unknown>
+
+bfdot   za.s[w10, 0], {z19.h - z22.h}, z4.h  // 11000001-00110100-01010010-01110000
+// CHECK-INST: bfdot   za.s[w10, 0, vgx4], { z19.h - z22.h }, z4.h
+// CHECK-ENCODING: [0x70,0x52,0x34,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1345270 <unknown>
+
+bfdot   za.s[w8, 0, vgx4], {z12.h - z15.h}, z2.h  // 11000001-00110010-00010001-10010000
+// CHECK-INST: bfdot   za.s[w8, 0, vgx4], { z12.h - z15.h }, z2.h
+// CHECK-ENCODING: [0x90,0x11,0x32,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1321190 <unknown>
+
+bfdot   za.s[w8, 0], {z12.h - z15.h}, z2.h  // 11000001-00110010-00010001-10010000
+// CHECK-INST: bfdot   za.s[w8, 0, vgx4], { z12.h - z15.h }, z2.h
+// CHECK-ENCODING: [0x90,0x11,0x32,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1321190 <unknown>
+
+bfdot   za.s[w10, 1, vgx4], {z1.h - z4.h}, z10.h  // 11000001-00111010-01010000-00110001
+// CHECK-INST: bfdot   za.s[w10, 1, vgx4], { z1.h - z4.h }, z10.h
+// CHECK-ENCODING: [0x31,0x50,0x3a,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c13a5031 <unknown>
+
+bfdot   za.s[w10, 1], {z1.h - z4.h}, z10.h  // 11000001-00111010-01010000-00110001
+// CHECK-INST: bfdot   za.s[w10, 1, vgx4], { z1.h - z4.h }, z10.h
+// CHECK-ENCODING: [0x31,0x50,0x3a,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c13a5031 <unknown>
+
+bfdot   za.s[w8, 5, vgx4], {z22.h - z25.h}, z14.h  // 11000001-00111110-00010010-11010101
+// CHECK-INST: bfdot   za.s[w8, 5, vgx4], { z22.h - z25.h }, z14.h
+// CHECK-ENCODING: [0xd5,0x12,0x3e,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c13e12d5 <unknown>
+
+bfdot   za.s[w8, 5], {z22.h - z25.h}, z14.h  // 11000001-00111110-00010010-11010101
+// CHECK-INST: bfdot   za.s[w8, 5, vgx4], { z22.h - z25.h }, z14.h
+// CHECK-ENCODING: [0xd5,0x12,0x3e,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c13e12d5 <unknown>
+
+bfdot   za.s[w11, 2, vgx4], {z9.h - z12.h}, z1.h  // 11000001-00110001-01110001-00110010
+// CHECK-INST: bfdot   za.s[w11, 2, vgx4], { z9.h - z12.h }, z1.h
+// CHECK-ENCODING: [0x32,0x71,0x31,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1317132 <unknown>
+
+bfdot   za.s[w11, 2], {z9.h - z12.h}, z1.h  // 11000001-00110001-01110001-00110010
+// CHECK-INST: bfdot   za.s[w11, 2, vgx4], { z9.h - z12.h }, z1.h
+// CHECK-ENCODING: [0x32,0x71,0x31,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1317132 <unknown>
+
+bfdot   za.s[w9, 7, vgx4], {z12.h - z15.h}, z11.h  // 11000001-00111011-00110001-10010111
+// CHECK-INST: bfdot   za.s[w9, 7, vgx4], { z12.h - z15.h }, z11.h
+// CHECK-ENCODING: [0x97,0x31,0x3b,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c13b3197 <unknown>
+
+bfdot   za.s[w9, 7], {z12.h - z15.h}, z11.h  // 11000001-00111011-00110001-10010111
+// CHECK-INST: bfdot   za.s[w9, 7, vgx4], { z12.h - z15.h }, z11.h
+// CHECK-ENCODING: [0x97,0x31,0x3b,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c13b3197 <unknown>
+
+
 bfdot   za.s[w8, 0, vgx4], {z0.h - z3.h}, z0.h[0]  // 11000001-01010000-10010000-00011000
 // CHECK-INST: bfdot   za.s[w8, 0, vgx4], { z0.h - z3.h }, z0.h[0]
 // CHECK-ENCODING: [0x18,0x90,0x50,0xc1]
@@ -301,3 +736,148 @@ bfdot   za.s[w9, 7], {z12.h - z15.h}, z11.h[2]  // 11000001-01011011-10111001-10
 // CHECK-ERROR: instruction requires: sme2
 // CHECK-UNKNOWN: c15bb99f <unknown>
 
+
+bfdot   za.s[w8, 0, vgx4], {z0.h - z3.h}, {z0.h - z3.h}  // 11000001-10100001-00010000-00010000
+// CHECK-INST: bfdot   za.s[w8, 0, vgx4], { z0.h - z3.h }, { z0.h - z3.h }
+// CHECK-ENCODING: [0x10,0x10,0xa1,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1a11010 <unknown>
+
+bfdot   za.s[w8, 0], {z0.h - z3.h}, {z0.h - z3.h}  // 11000001-10100001-00010000-00010000
+// CHECK-INST: bfdot   za.s[w8, 0, vgx4], { z0.h - z3.h }, { z0.h - z3.h }
+// CHECK-ENCODING: [0x10,0x10,0xa1,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1a11010 <unknown>
+
+bfdot   za.s[w10, 5, vgx4], {z8.h - z11.h}, {z20.h - z23.h}  // 11000001-10110101-01010001-00010101
+// CHECK-INST: bfdot   za.s[w10, 5, vgx4], { z8.h - z11.h }, { z20.h - z23.h }
+// CHECK-ENCODING: [0x15,0x51,0xb5,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1b55115 <unknown>
+
+bfdot   za.s[w10, 5], {z8.h - z11.h}, {z20.h - z23.h}  // 11000001-10110101-01010001-00010101
+// CHECK-INST: bfdot   za.s[w10, 5, vgx4], { z8.h - z11.h }, { z20.h - z23.h }
+// CHECK-ENCODING: [0x15,0x51,0xb5,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1b55115 <unknown>
+
+bfdot   za.s[w11, 7, vgx4], {z12.h - z15.h}, {z8.h - z11.h}  // 11000001-10101001-01110001-10010111
+// CHECK-INST: bfdot   za.s[w11, 7, vgx4], { z12.h - z15.h }, { z8.h - z11.h }
+// CHECK-ENCODING: [0x97,0x71,0xa9,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1a97197 <unknown>
+
+bfdot   za.s[w11, 7], {z12.h - z15.h}, {z8.h - z11.h}  // 11000001-10101001-01110001-10010111
+// CHECK-INST: bfdot   za.s[w11, 7, vgx4], { z12.h - z15.h }, { z8.h - z11.h }
+// CHECK-ENCODING: [0x97,0x71,0xa9,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1a97197 <unknown>
+
+bfdot   za.s[w11, 7, vgx4], {z28.h - z31.h}, {z28.h - z31.h}  // 11000001-10111101-01110011-10010111
+// CHECK-INST: bfdot   za.s[w11, 7, vgx4], { z28.h - z31.h }, { z28.h - z31.h }
+// CHECK-ENCODING: [0x97,0x73,0xbd,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1bd7397 <unknown>
+
+bfdot   za.s[w11, 7], {z28.h - z31.h}, {z28.h - z31.h}  // 11000001-10111101-01110011-10010111
+// CHECK-INST: bfdot   za.s[w11, 7, vgx4], { z28.h - z31.h }, { z28.h - z31.h }
+// CHECK-ENCODING: [0x97,0x73,0xbd,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1bd7397 <unknown>
+
+bfdot   za.s[w8, 5, vgx4], {z16.h - z19.h}, {z16.h - z19.h}  // 11000001-10110001-00010010-00010101
+// CHECK-INST: bfdot   za.s[w8, 5, vgx4], { z16.h - z19.h }, { z16.h - z19.h }
+// CHECK-ENCODING: [0x15,0x12,0xb1,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1b11215 <unknown>
+
+bfdot   za.s[w8, 5], {z16.h - z19.h}, {z16.h - z19.h}  // 11000001-10110001-00010010-00010101
+// CHECK-INST: bfdot   za.s[w8, 5, vgx4], { z16.h - z19.h }, { z16.h - z19.h }
+// CHECK-ENCODING: [0x15,0x12,0xb1,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1b11215 <unknown>
+
+bfdot   za.s[w8, 1, vgx4], {z0.h - z3.h}, {z28.h - z31.h}  // 11000001-10111101-00010000-00010001
+// CHECK-INST: bfdot   za.s[w8, 1, vgx4], { z0.h - z3.h }, { z28.h - z31.h }
+// CHECK-ENCODING: [0x11,0x10,0xbd,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1bd1011 <unknown>
+
+bfdot   za.s[w8, 1], {z0.h - z3.h}, {z28.h - z31.h}  // 11000001-10111101-00010000-00010001
+// CHECK-INST: bfdot   za.s[w8, 1, vgx4], { z0.h - z3.h }, { z28.h - z31.h }
+// CHECK-ENCODING: [0x11,0x10,0xbd,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1bd1011 <unknown>
+
+bfdot   za.s[w10, 0, vgx4], {z16.h - z19.h}, {z20.h - z23.h}  // 11000001-10110101-01010010-00010000
+// CHECK-INST: bfdot   za.s[w10, 0, vgx4], { z16.h - z19.h }, { z20.h - z23.h }
+// CHECK-ENCODING: [0x10,0x52,0xb5,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1b55210 <unknown>
+
+bfdot   za.s[w10, 0], {z16.h - z19.h}, {z20.h - z23.h}  // 11000001-10110101-01010010-00010000
+// CHECK-INST: bfdot   za.s[w10, 0, vgx4], { z16.h - z19.h }, { z20.h - z23.h }
+// CHECK-ENCODING: [0x10,0x52,0xb5,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1b55210 <unknown>
+
+bfdot   za.s[w8, 0, vgx4], {z12.h - z15.h}, {z0.h - z3.h}  // 11000001-10100001-00010001-10010000
+// CHECK-INST: bfdot   za.s[w8, 0, vgx4], { z12.h - z15.h }, { z0.h - z3.h }
+// CHECK-ENCODING: [0x90,0x11,0xa1,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1a11190 <unknown>
+
+bfdot   za.s[w8, 0], {z12.h - z15.h}, {z0.h - z3.h}  // 11000001-10100001-00010001-10010000
+// CHECK-INST: bfdot   za.s[w8, 0, vgx4], { z12.h - z15.h }, { z0.h - z3.h }
+// CHECK-ENCODING: [0x90,0x11,0xa1,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1a11190 <unknown>
+
+bfdot   za.s[w10, 1, vgx4], {z0.h - z3.h}, {z24.h - z27.h}  // 11000001-10111001-01010000-00010001
+// CHECK-INST: bfdot   za.s[w10, 1, vgx4], { z0.h - z3.h }, { z24.h - z27.h }
+// CHECK-ENCODING: [0x11,0x50,0xb9,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1b95011 <unknown>
+
+bfdot   za.s[w10, 1], {z0.h - z3.h}, {z24.h - z27.h}  // 11000001-10111001-01010000-00010001
+// CHECK-INST: bfdot   za.s[w10, 1, vgx4], { z0.h - z3.h }, { z24.h - z27.h }
+// CHECK-ENCODING: [0x11,0x50,0xb9,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1b95011 <unknown>
+
+bfdot   za.s[w8, 5, vgx4], {z20.h - z23.h}, {z28.h - z31.h}  // 11000001-10111101-00010010-10010101
+// CHECK-INST: bfdot   za.s[w8, 5, vgx4], { z20.h - z23.h }, { z28.h - z31.h }
+// CHECK-ENCODING: [0x95,0x12,0xbd,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1bd1295 <unknown>
+
+bfdot   za.s[w8, 5], {z20.h - z23.h}, {z28.h - z31.h}  // 11000001-10111101-00010010-10010101
+// CHECK-INST: bfdot   za.s[w8, 5, vgx4], { z20.h - z23.h }, { z28.h - z31.h }
+// CHECK-ENCODING: [0x95,0x12,0xbd,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1bd1295 <unknown>
+
+bfdot   za.s[w11, 2, vgx4], {z8.h - z11.h}, {z0.h - z3.h}  // 11000001-10100001-01110001-00010010
+// CHECK-INST: bfdot   za.s[w11, 2, vgx4], { z8.h - z11.h }, { z0.h - z3.h }
+// CHECK-ENCODING: [0x12,0x71,0xa1,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1a17112 <unknown>
+
+bfdot   za.s[w11, 2], {z8.h - z11.h}, {z0.h - z3.h}  // 11000001-10100001-01110001-00010010
+// CHECK-INST: bfdot   za.s[w11, 2, vgx4], { z8.h - z11.h }, { z0.h - z3.h }
+// CHECK-ENCODING: [0x12,0x71,0xa1,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1a17112 <unknown>
+
+bfdot   za.s[w9, 7, vgx4], {z12.h - z15.h}, {z8.h - z11.h}  // 11000001-10101001-00110001-10010111
+// CHECK-INST: bfdot   za.s[w9, 7, vgx4], { z12.h - z15.h }, { z8.h - z11.h }
+// CHECK-ENCODING: [0x97,0x31,0xa9,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1a93197 <unknown>
+
+bfdot   za.s[w9, 7], {z12.h - z15.h}, {z8.h - z11.h}  // 11000001-10101001-00110001-10010111
+// CHECK-INST: bfdot   za.s[w9, 7, vgx4], { z12.h - z15.h }, { z8.h - z11.h }
+// CHECK-ENCODING: [0x97,0x31,0xa9,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1a93197 <unknown>
+
index 115962c..268a1b6 100644 (file)
 // RUN:        | FileCheck %s --check-prefixes=CHECK-ENCODING,CHECK-INST
 
 
+fdot    za.s[w8, 0, vgx2], {z0.h, z1.h}, z0.h  // 11000001-00100000-00010000-00000000
+// CHECK-INST: fdot    za.s[w8, 0, vgx2], { z0.h, z1.h }, z0.h
+// CHECK-ENCODING: [0x00,0x10,0x20,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1201000 <unknown>
+
+fdot    za.s[w8, 0], {z0.h, z1.h}, z0.h  // 11000001-00100000-00010000-00000000
+// CHECK-INST: fdot    za.s[w8, 0, vgx2], { z0.h, z1.h }, z0.h
+// CHECK-ENCODING: [0x00,0x10,0x20,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1201000 <unknown>
+
+fdot    za.s[w10, 5, vgx2], {z10.h, z11.h}, z5.h  // 11000001-00100101-01010001-01000101
+// CHECK-INST: fdot    za.s[w10, 5, vgx2], { z10.h, z11.h }, z5.h
+// CHECK-ENCODING: [0x45,0x51,0x25,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1255145 <unknown>
+
+fdot    za.s[w10, 5], {z10.h, z11.h}, z5.h  // 11000001-00100101-01010001-01000101
+// CHECK-INST: fdot    za.s[w10, 5, vgx2], { z10.h, z11.h }, z5.h
+// CHECK-ENCODING: [0x45,0x51,0x25,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1255145 <unknown>
+
+fdot    za.s[w11, 7, vgx2], {z13.h, z14.h}, z8.h  // 11000001-00101000-01110001-10100111
+// CHECK-INST: fdot    za.s[w11, 7, vgx2], { z13.h, z14.h }, z8.h
+// CHECK-ENCODING: [0xa7,0x71,0x28,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c12871a7 <unknown>
+
+fdot    za.s[w11, 7], {z13.h, z14.h}, z8.h  // 11000001-00101000-01110001-10100111
+// CHECK-INST: fdot    za.s[w11, 7, vgx2], { z13.h, z14.h }, z8.h
+// CHECK-ENCODING: [0xa7,0x71,0x28,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c12871a7 <unknown>
+
+fdot    za.s[w11, 7, vgx2], {z31.h, z0.h}, z15.h  // 11000001-00101111-01110011-11100111
+// CHECK-INST: fdot    za.s[w11, 7, vgx2], { z31.h, z0.h }, z15.h
+// CHECK-ENCODING: [0xe7,0x73,0x2f,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c12f73e7 <unknown>
+
+fdot    za.s[w11, 7], {z31.h, z0.h}, z15.h  // 11000001-00101111-01110011-11100111
+// CHECK-INST: fdot    za.s[w11, 7, vgx2], { z31.h, z0.h }, z15.h
+// CHECK-ENCODING: [0xe7,0x73,0x2f,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c12f73e7 <unknown>
+
+fdot    za.s[w8, 5, vgx2], {z17.h, z18.h}, z0.h  // 11000001-00100000-00010010-00100101
+// CHECK-INST: fdot    za.s[w8, 5, vgx2], { z17.h, z18.h }, z0.h
+// CHECK-ENCODING: [0x25,0x12,0x20,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1201225 <unknown>
+
+fdot    za.s[w8, 5], {z17.h, z18.h}, z0.h  // 11000001-00100000-00010010-00100101
+// CHECK-INST: fdot    za.s[w8, 5, vgx2], { z17.h, z18.h }, z0.h
+// CHECK-ENCODING: [0x25,0x12,0x20,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1201225 <unknown>
+
+fdot    za.s[w8, 1, vgx2], {z1.h, z2.h}, z14.h  // 11000001-00101110-00010000-00100001
+// CHECK-INST: fdot    za.s[w8, 1, vgx2], { z1.h, z2.h }, z14.h
+// CHECK-ENCODING: [0x21,0x10,0x2e,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c12e1021 <unknown>
+
+fdot    za.s[w8, 1], {z1.h, z2.h}, z14.h  // 11000001-00101110-00010000-00100001
+// CHECK-INST: fdot    za.s[w8, 1, vgx2], { z1.h, z2.h }, z14.h
+// CHECK-ENCODING: [0x21,0x10,0x2e,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c12e1021 <unknown>
+
+fdot    za.s[w10, 0, vgx2], {z19.h, z20.h}, z4.h  // 11000001-00100100-01010010-01100000
+// CHECK-INST: fdot    za.s[w10, 0, vgx2], { z19.h, z20.h }, z4.h
+// CHECK-ENCODING: [0x60,0x52,0x24,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1245260 <unknown>
+
+fdot    za.s[w10, 0], {z19.h, z20.h}, z4.h  // 11000001-00100100-01010010-01100000
+// CHECK-INST: fdot    za.s[w10, 0, vgx2], { z19.h, z20.h }, z4.h
+// CHECK-ENCODING: [0x60,0x52,0x24,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1245260 <unknown>
+
+fdot    za.s[w8, 0, vgx2], {z12.h, z13.h}, z2.h  // 11000001-00100010-00010001-10000000
+// CHECK-INST: fdot    za.s[w8, 0, vgx2], { z12.h, z13.h }, z2.h
+// CHECK-ENCODING: [0x80,0x11,0x22,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1221180 <unknown>
+
+fdot    za.s[w8, 0], {z12.h, z13.h}, z2.h  // 11000001-00100010-00010001-10000000
+// CHECK-INST: fdot    za.s[w8, 0, vgx2], { z12.h, z13.h }, z2.h
+// CHECK-ENCODING: [0x80,0x11,0x22,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1221180 <unknown>
+
+fdot    za.s[w10, 1, vgx2], {z1.h, z2.h}, z10.h  // 11000001-00101010-01010000-00100001
+// CHECK-INST: fdot    za.s[w10, 1, vgx2], { z1.h, z2.h }, z10.h
+// CHECK-ENCODING: [0x21,0x50,0x2a,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c12a5021 <unknown>
+
+fdot    za.s[w10, 1], {z1.h, z2.h}, z10.h  // 11000001-00101010-01010000-00100001
+// CHECK-INST: fdot    za.s[w10, 1, vgx2], { z1.h, z2.h }, z10.h
+// CHECK-ENCODING: [0x21,0x50,0x2a,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c12a5021 <unknown>
+
+fdot    za.s[w8, 5, vgx2], {z22.h, z23.h}, z14.h  // 11000001-00101110-00010010-11000101
+// CHECK-INST: fdot    za.s[w8, 5, vgx2], { z22.h, z23.h }, z14.h
+// CHECK-ENCODING: [0xc5,0x12,0x2e,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c12e12c5 <unknown>
+
+fdot    za.s[w8, 5], {z22.h, z23.h}, z14.h  // 11000001-00101110-00010010-11000101
+// CHECK-INST: fdot    za.s[w8, 5, vgx2], { z22.h, z23.h }, z14.h
+// CHECK-ENCODING: [0xc5,0x12,0x2e,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c12e12c5 <unknown>
+
+fdot    za.s[w11, 2, vgx2], {z9.h, z10.h}, z1.h  // 11000001-00100001-01110001-00100010
+// CHECK-INST: fdot    za.s[w11, 2, vgx2], { z9.h, z10.h }, z1.h
+// CHECK-ENCODING: [0x22,0x71,0x21,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1217122 <unknown>
+
+fdot    za.s[w11, 2], {z9.h, z10.h}, z1.h  // 11000001-00100001-01110001-00100010
+// CHECK-INST: fdot    za.s[w11, 2, vgx2], { z9.h, z10.h }, z1.h
+// CHECK-ENCODING: [0x22,0x71,0x21,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1217122 <unknown>
+
+fdot    za.s[w9, 7, vgx2], {z12.h, z13.h}, z11.h  // 11000001-00101011-00110001-10000111
+// CHECK-INST: fdot    za.s[w9, 7, vgx2], { z12.h, z13.h }, z11.h
+// CHECK-ENCODING: [0x87,0x31,0x2b,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c12b3187 <unknown>
+
+fdot    za.s[w9, 7], {z12.h, z13.h}, z11.h  // 11000001-00101011-00110001-10000111
+// CHECK-INST: fdot    za.s[w9, 7, vgx2], { z12.h, z13.h }, z11.h
+// CHECK-ENCODING: [0x87,0x31,0x2b,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c12b3187 <unknown>
+
+
 fdot    za.s[w8, 0, vgx2], {z0.h, z1.h}, z0.h[0]  // 11000001-01010000-00010000-00001000
 // CHECK-INST: fdot    za.s[w8, 0, vgx2], { z0.h, z1.h }, z0.h[0]
 // CHECK-ENCODING: [0x08,0x10,0x50,0xc1]
@@ -157,6 +302,296 @@ fdot    za.s[w9, 7], {z12.h, z13.h}, z11.h[2]  // 11000001-01011011-00111001-100
 // CHECK-UNKNOWN: c15b398f <unknown>
 
 
+fdot    za.s[w8, 0, vgx2], {z0.h, z1.h}, {z0.h, z1.h}  // 11000001-10100000-00010000-00000000
+// CHECK-INST: fdot    za.s[w8, 0, vgx2], { z0.h, z1.h }, { z0.h, z1.h }
+// CHECK-ENCODING: [0x00,0x10,0xa0,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1a01000 <unknown>
+
+fdot    za.s[w8, 0], {z0.h, z1.h}, {z0.h, z1.h}  // 11000001-10100000-00010000-00000000
+// CHECK-INST: fdot    za.s[w8, 0, vgx2], { z0.h, z1.h }, { z0.h, z1.h }
+// CHECK-ENCODING: [0x00,0x10,0xa0,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1a01000 <unknown>
+
+fdot    za.s[w10, 5, vgx2], {z10.h, z11.h}, {z20.h, z21.h}  // 11000001-10110100-01010001-01000101
+// CHECK-INST: fdot    za.s[w10, 5, vgx2], { z10.h, z11.h }, { z20.h, z21.h }
+// CHECK-ENCODING: [0x45,0x51,0xb4,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1b45145 <unknown>
+
+fdot    za.s[w10, 5], {z10.h, z11.h}, {z20.h, z21.h}  // 11000001-10110100-01010001-01000101
+// CHECK-INST: fdot    za.s[w10, 5, vgx2], { z10.h, z11.h }, { z20.h, z21.h }
+// CHECK-ENCODING: [0x45,0x51,0xb4,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1b45145 <unknown>
+
+fdot    za.s[w11, 7, vgx2], {z12.h, z13.h}, {z8.h, z9.h}  // 11000001-10101000-01110001-10000111
+// CHECK-INST: fdot    za.s[w11, 7, vgx2], { z12.h, z13.h }, { z8.h, z9.h }
+// CHECK-ENCODING: [0x87,0x71,0xa8,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1a87187 <unknown>
+
+fdot    za.s[w11, 7], {z12.h, z13.h}, {z8.h, z9.h}  // 11000001-10101000-01110001-10000111
+// CHECK-INST: fdot    za.s[w11, 7, vgx2], { z12.h, z13.h }, { z8.h, z9.h }
+// CHECK-ENCODING: [0x87,0x71,0xa8,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1a87187 <unknown>
+
+fdot    za.s[w11, 7, vgx2], {z30.h, z31.h}, {z30.h, z31.h}  // 11000001-10111110-01110011-11000111
+// CHECK-INST: fdot    za.s[w11, 7, vgx2], { z30.h, z31.h }, { z30.h, z31.h }
+// CHECK-ENCODING: [0xc7,0x73,0xbe,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1be73c7 <unknown>
+
+fdot    za.s[w11, 7], {z30.h, z31.h}, {z30.h, z31.h}  // 11000001-10111110-01110011-11000111
+// CHECK-INST: fdot    za.s[w11, 7, vgx2], { z30.h, z31.h }, { z30.h, z31.h }
+// CHECK-ENCODING: [0xc7,0x73,0xbe,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1be73c7 <unknown>
+
+fdot    za.s[w8, 5, vgx2], {z16.h, z17.h}, {z16.h, z17.h}  // 11000001-10110000-00010010-00000101
+// CHECK-INST: fdot    za.s[w8, 5, vgx2], { z16.h, z17.h }, { z16.h, z17.h }
+// CHECK-ENCODING: [0x05,0x12,0xb0,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1b01205 <unknown>
+
+fdot    za.s[w8, 5], {z16.h, z17.h}, {z16.h, z17.h}  // 11000001-10110000-00010010-00000101
+// CHECK-INST: fdot    za.s[w8, 5, vgx2], { z16.h, z17.h }, { z16.h, z17.h }
+// CHECK-ENCODING: [0x05,0x12,0xb0,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1b01205 <unknown>
+
+fdot    za.s[w8, 1, vgx2], {z0.h, z1.h}, {z30.h, z31.h}  // 11000001-10111110-00010000-00000001
+// CHECK-INST: fdot    za.s[w8, 1, vgx2], { z0.h, z1.h }, { z30.h, z31.h }
+// CHECK-ENCODING: [0x01,0x10,0xbe,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1be1001 <unknown>
+
+fdot    za.s[w8, 1], {z0.h, z1.h}, {z30.h, z31.h}  // 11000001-10111110-00010000-00000001
+// CHECK-INST: fdot    za.s[w8, 1, vgx2], { z0.h, z1.h }, { z30.h, z31.h }
+// CHECK-ENCODING: [0x01,0x10,0xbe,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1be1001 <unknown>
+
+fdot    za.s[w10, 0, vgx2], {z18.h, z19.h}, {z20.h, z21.h}  // 11000001-10110100-01010010-01000000
+// CHECK-INST: fdot    za.s[w10, 0, vgx2], { z18.h, z19.h }, { z20.h, z21.h }
+// CHECK-ENCODING: [0x40,0x52,0xb4,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1b45240 <unknown>
+
+fdot    za.s[w10, 0], {z18.h, z19.h}, {z20.h, z21.h}  // 11000001-10110100-01010010-01000000
+// CHECK-INST: fdot    za.s[w10, 0, vgx2], { z18.h, z19.h }, { z20.h, z21.h }
+// CHECK-ENCODING: [0x40,0x52,0xb4,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1b45240 <unknown>
+
+fdot    za.s[w8, 0, vgx2], {z12.h, z13.h}, {z2.h, z3.h}  // 11000001-10100010-00010001-10000000
+// CHECK-INST: fdot    za.s[w8, 0, vgx2], { z12.h, z13.h }, { z2.h, z3.h }
+// CHECK-ENCODING: [0x80,0x11,0xa2,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1a21180 <unknown>
+
+fdot    za.s[w8, 0], {z12.h, z13.h}, {z2.h, z3.h}  // 11000001-10100010-00010001-10000000
+// CHECK-INST: fdot    za.s[w8, 0, vgx2], { z12.h, z13.h }, { z2.h, z3.h }
+// CHECK-ENCODING: [0x80,0x11,0xa2,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1a21180 <unknown>
+
+fdot    za.s[w10, 1, vgx2], {z0.h, z1.h}, {z26.h, z27.h}  // 11000001-10111010-01010000-00000001
+// CHECK-INST: fdot    za.s[w10, 1, vgx2], { z0.h, z1.h }, { z26.h, z27.h }
+// CHECK-ENCODING: [0x01,0x50,0xba,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1ba5001 <unknown>
+
+fdot    za.s[w10, 1], {z0.h, z1.h}, {z26.h, z27.h}  // 11000001-10111010-01010000-00000001
+// CHECK-INST: fdot    za.s[w10, 1, vgx2], { z0.h, z1.h }, { z26.h, z27.h }
+// CHECK-ENCODING: [0x01,0x50,0xba,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1ba5001 <unknown>
+
+fdot    za.s[w8, 5, vgx2], {z22.h, z23.h}, {z30.h, z31.h}  // 11000001-10111110-00010010-11000101
+// CHECK-INST: fdot    za.s[w8, 5, vgx2], { z22.h, z23.h }, { z30.h, z31.h }
+// CHECK-ENCODING: [0xc5,0x12,0xbe,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1be12c5 <unknown>
+
+fdot    za.s[w8, 5], {z22.h, z23.h}, {z30.h, z31.h}  // 11000001-10111110-00010010-11000101
+// CHECK-INST: fdot    za.s[w8, 5, vgx2], { z22.h, z23.h }, { z30.h, z31.h }
+// CHECK-ENCODING: [0xc5,0x12,0xbe,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1be12c5 <unknown>
+
+fdot    za.s[w11, 2, vgx2], {z8.h, z9.h}, {z0.h, z1.h}  // 11000001-10100000-01110001-00000010
+// CHECK-INST: fdot    za.s[w11, 2, vgx2], { z8.h, z9.h }, { z0.h, z1.h }
+// CHECK-ENCODING: [0x02,0x71,0xa0,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1a07102 <unknown>
+
+fdot    za.s[w11, 2], {z8.h, z9.h}, {z0.h, z1.h}  // 11000001-10100000-01110001-00000010
+// CHECK-INST: fdot    za.s[w11, 2, vgx2], { z8.h, z9.h }, { z0.h, z1.h }
+// CHECK-ENCODING: [0x02,0x71,0xa0,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1a07102 <unknown>
+
+fdot    za.s[w9, 7, vgx2], {z12.h, z13.h}, {z10.h, z11.h}  // 11000001-10101010-00110001-10000111
+// CHECK-INST: fdot    za.s[w9, 7, vgx2], { z12.h, z13.h }, { z10.h, z11.h }
+// CHECK-ENCODING: [0x87,0x31,0xaa,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1aa3187 <unknown>
+
+fdot    za.s[w9, 7], {z12.h, z13.h}, {z10.h, z11.h}  // 11000001-10101010-00110001-10000111
+// CHECK-INST: fdot    za.s[w9, 7, vgx2], { z12.h, z13.h }, { z10.h, z11.h }
+// CHECK-ENCODING: [0x87,0x31,0xaa,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1aa3187 <unknown>
+
+
+fdot    za.s[w8, 0, vgx4], {z0.h - z3.h}, z0.h  // 11000001-00110000-00010000-00000000
+// CHECK-INST: fdot    za.s[w8, 0, vgx4], { z0.h - z3.h }, z0.h
+// CHECK-ENCODING: [0x00,0x10,0x30,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1301000 <unknown>
+
+fdot    za.s[w8, 0], {z0.h - z3.h}, z0.h  // 11000001-00110000-00010000-00000000
+// CHECK-INST: fdot    za.s[w8, 0, vgx4], { z0.h - z3.h }, z0.h
+// CHECK-ENCODING: [0x00,0x10,0x30,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1301000 <unknown>
+
+fdot    za.s[w10, 5, vgx4], {z10.h - z13.h}, z5.h  // 11000001-00110101-01010001-01000101
+// CHECK-INST: fdot    za.s[w10, 5, vgx4], { z10.h - z13.h }, z5.h
+// CHECK-ENCODING: [0x45,0x51,0x35,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1355145 <unknown>
+
+fdot    za.s[w10, 5], {z10.h - z13.h}, z5.h  // 11000001-00110101-01010001-01000101
+// CHECK-INST: fdot    za.s[w10, 5, vgx4], { z10.h - z13.h }, z5.h
+// CHECK-ENCODING: [0x45,0x51,0x35,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1355145 <unknown>
+
+fdot    za.s[w11, 7, vgx4], {z13.h - z16.h}, z8.h  // 11000001-00111000-01110001-10100111
+// CHECK-INST: fdot    za.s[w11, 7, vgx4], { z13.h - z16.h }, z8.h
+// CHECK-ENCODING: [0xa7,0x71,0x38,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c13871a7 <unknown>
+
+fdot    za.s[w11, 7], {z13.h - z16.h}, z8.h  // 11000001-00111000-01110001-10100111
+// CHECK-INST: fdot    za.s[w11, 7, vgx4], { z13.h - z16.h }, z8.h
+// CHECK-ENCODING: [0xa7,0x71,0x38,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c13871a7 <unknown>
+
+fdot    za.s[w11, 7, vgx4], {z31.h - z2.h}, z15.h  // 11000001-00111111-01110011-11100111
+// CHECK-INST: fdot    za.s[w11, 7, vgx4], { z31.h, z0.h, z1.h, z2.h }, z15.h
+// CHECK-ENCODING: [0xe7,0x73,0x3f,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c13f73e7 <unknown>
+
+fdot    za.s[w11, 7], {z31.h - z2.h}, z15.h  // 11000001-00111111-01110011-11100111
+// CHECK-INST: fdot    za.s[w11, 7, vgx4], { z31.h, z0.h, z1.h, z2.h }, z15.h
+// CHECK-ENCODING: [0xe7,0x73,0x3f,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c13f73e7 <unknown>
+
+fdot    za.s[w8, 5, vgx4], {z17.h - z20.h}, z0.h  // 11000001-00110000-00010010-00100101
+// CHECK-INST: fdot    za.s[w8, 5, vgx4], { z17.h - z20.h }, z0.h
+// CHECK-ENCODING: [0x25,0x12,0x30,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1301225 <unknown>
+
+fdot    za.s[w8, 5], {z17.h - z20.h}, z0.h  // 11000001-00110000-00010010-00100101
+// CHECK-INST: fdot    za.s[w8, 5, vgx4], { z17.h - z20.h }, z0.h
+// CHECK-ENCODING: [0x25,0x12,0x30,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1301225 <unknown>
+
+fdot    za.s[w8, 1, vgx4], {z1.h - z4.h}, z14.h  // 11000001-00111110-00010000-00100001
+// CHECK-INST: fdot    za.s[w8, 1, vgx4], { z1.h - z4.h }, z14.h
+// CHECK-ENCODING: [0x21,0x10,0x3e,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c13e1021 <unknown>
+
+fdot    za.s[w8, 1], {z1.h - z4.h}, z14.h  // 11000001-00111110-00010000-00100001
+// CHECK-INST: fdot    za.s[w8, 1, vgx4], { z1.h - z4.h }, z14.h
+// CHECK-ENCODING: [0x21,0x10,0x3e,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c13e1021 <unknown>
+
+fdot    za.s[w10, 0, vgx4], {z19.h - z22.h}, z4.h  // 11000001-00110100-01010010-01100000
+// CHECK-INST: fdot    za.s[w10, 0, vgx4], { z19.h - z22.h }, z4.h
+// CHECK-ENCODING: [0x60,0x52,0x34,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1345260 <unknown>
+
+fdot    za.s[w10, 0], {z19.h - z22.h}, z4.h  // 11000001-00110100-01010010-01100000
+// CHECK-INST: fdot    za.s[w10, 0, vgx4], { z19.h - z22.h }, z4.h
+// CHECK-ENCODING: [0x60,0x52,0x34,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1345260 <unknown>
+
+fdot    za.s[w8, 0, vgx4], {z12.h - z15.h}, z2.h  // 11000001-00110010-00010001-10000000
+// CHECK-INST: fdot    za.s[w8, 0, vgx4], { z12.h - z15.h }, z2.h
+// CHECK-ENCODING: [0x80,0x11,0x32,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1321180 <unknown>
+
+fdot    za.s[w8, 0], {z12.h - z15.h}, z2.h  // 11000001-00110010-00010001-10000000
+// CHECK-INST: fdot    za.s[w8, 0, vgx4], { z12.h - z15.h }, z2.h
+// CHECK-ENCODING: [0x80,0x11,0x32,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1321180 <unknown>
+
+fdot    za.s[w10, 1, vgx4], {z1.h - z4.h}, z10.h  // 11000001-00111010-01010000-00100001
+// CHECK-INST: fdot    za.s[w10, 1, vgx4], { z1.h - z4.h }, z10.h
+// CHECK-ENCODING: [0x21,0x50,0x3a,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c13a5021 <unknown>
+
+fdot    za.s[w10, 1], {z1.h - z4.h}, z10.h  // 11000001-00111010-01010000-00100001
+// CHECK-INST: fdot    za.s[w10, 1, vgx4], { z1.h - z4.h }, z10.h
+// CHECK-ENCODING: [0x21,0x50,0x3a,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c13a5021 <unknown>
+
+fdot    za.s[w8, 5, vgx4], {z22.h - z25.h}, z14.h  // 11000001-00111110-00010010-11000101
+// CHECK-INST: fdot    za.s[w8, 5, vgx4], { z22.h - z25.h }, z14.h
+// CHECK-ENCODING: [0xc5,0x12,0x3e,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c13e12c5 <unknown>
+
+fdot    za.s[w8, 5], {z22.h - z25.h}, z14.h  // 11000001-00111110-00010010-11000101
+// CHECK-INST: fdot    za.s[w8, 5, vgx4], { z22.h - z25.h }, z14.h
+// CHECK-ENCODING: [0xc5,0x12,0x3e,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c13e12c5 <unknown>
+
+fdot    za.s[w11, 2, vgx4], {z9.h - z12.h}, z1.h  // 11000001-00110001-01110001-00100010
+// CHECK-INST: fdot    za.s[w11, 2, vgx4], { z9.h - z12.h }, z1.h
+// CHECK-ENCODING: [0x22,0x71,0x31,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1317122 <unknown>
+
+fdot    za.s[w11, 2], {z9.h - z12.h}, z1.h  // 11000001-00110001-01110001-00100010
+// CHECK-INST: fdot    za.s[w11, 2, vgx4], { z9.h - z12.h }, z1.h
+// CHECK-ENCODING: [0x22,0x71,0x31,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1317122 <unknown>
+
+fdot    za.s[w9, 7, vgx4], {z12.h - z15.h}, z11.h  // 11000001-00111011-00110001-10000111
+// CHECK-INST: fdot    za.s[w9, 7, vgx4], { z12.h - z15.h }, z11.h
+// CHECK-ENCODING: [0x87,0x31,0x3b,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c13b3187 <unknown>
+
+fdot    za.s[w9, 7], {z12.h - z15.h}, z11.h  // 11000001-00111011-00110001-10000111
+// CHECK-INST: fdot    za.s[w9, 7, vgx4], { z12.h - z15.h }, z11.h
+// CHECK-ENCODING: [0x87,0x31,0x3b,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c13b3187 <unknown>
+
+
 fdot    za.s[w8, 0, vgx4], {z0.h - z3.h}, z0.h[0]  // 11000001-01010000-10010000-00001000
 // CHECK-INST: fdot    za.s[w8, 0, vgx4], { z0.h - z3.h }, z0.h[0]
 // CHECK-ENCODING: [0x08,0x90,0x50,0xc1]
@@ -301,3 +736,148 @@ fdot    za.s[w9, 7], {z12.h - z15.h}, z11.h[2]  // 11000001-01011011-10111001-10
 // CHECK-ERROR: instruction requires: sme2
 // CHECK-UNKNOWN: c15bb98f <unknown>
 
+
+fdot    za.s[w8, 0, vgx4], {z0.h - z3.h}, {z0.h - z3.h}  // 11000001-10100001-00010000-00000000
+// CHECK-INST: fdot    za.s[w8, 0, vgx4], { z0.h - z3.h }, { z0.h - z3.h }
+// CHECK-ENCODING: [0x00,0x10,0xa1,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1a11000 <unknown>
+
+fdot    za.s[w8, 0], {z0.h - z3.h}, {z0.h - z3.h}  // 11000001-10100001-00010000-00000000
+// CHECK-INST: fdot    za.s[w8, 0, vgx4], { z0.h - z3.h }, { z0.h - z3.h }
+// CHECK-ENCODING: [0x00,0x10,0xa1,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1a11000 <unknown>
+
+fdot    za.s[w10, 5, vgx4], {z8.h - z11.h}, {z20.h - z23.h}  // 11000001-10110101-01010001-00000101
+// CHECK-INST: fdot    za.s[w10, 5, vgx4], { z8.h - z11.h }, { z20.h - z23.h }
+// CHECK-ENCODING: [0x05,0x51,0xb5,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1b55105 <unknown>
+
+fdot    za.s[w10, 5], {z8.h - z11.h}, {z20.h - z23.h}  // 11000001-10110101-01010001-00000101
+// CHECK-INST: fdot    za.s[w10, 5, vgx4], { z8.h - z11.h }, { z20.h - z23.h }
+// CHECK-ENCODING: [0x05,0x51,0xb5,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1b55105 <unknown>
+
+fdot    za.s[w11, 7, vgx4], {z12.h - z15.h}, {z8.h - z11.h}  // 11000001-10101001-01110001-10000111
+// CHECK-INST: fdot    za.s[w11, 7, vgx4], { z12.h - z15.h }, { z8.h - z11.h }
+// CHECK-ENCODING: [0x87,0x71,0xa9,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1a97187 <unknown>
+
+fdot    za.s[w11, 7], {z12.h - z15.h}, {z8.h - z11.h}  // 11000001-10101001-01110001-10000111
+// CHECK-INST: fdot    za.s[w11, 7, vgx4], { z12.h - z15.h }, { z8.h - z11.h }
+// CHECK-ENCODING: [0x87,0x71,0xa9,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1a97187 <unknown>
+
+fdot    za.s[w11, 7, vgx4], {z28.h - z31.h}, {z28.h - z31.h}  // 11000001-10111101-01110011-10000111
+// CHECK-INST: fdot    za.s[w11, 7, vgx4], { z28.h - z31.h }, { z28.h - z31.h }
+// CHECK-ENCODING: [0x87,0x73,0xbd,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1bd7387 <unknown>
+
+fdot    za.s[w11, 7], {z28.h - z31.h}, {z28.h - z31.h}  // 11000001-10111101-01110011-10000111
+// CHECK-INST: fdot    za.s[w11, 7, vgx4], { z28.h - z31.h }, { z28.h - z31.h }
+// CHECK-ENCODING: [0x87,0x73,0xbd,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1bd7387 <unknown>
+
+fdot    za.s[w8, 5, vgx4], {z16.h - z19.h}, {z16.h - z19.h}  // 11000001-10110001-00010010-00000101
+// CHECK-INST: fdot    za.s[w8, 5, vgx4], { z16.h - z19.h }, { z16.h - z19.h }
+// CHECK-ENCODING: [0x05,0x12,0xb1,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1b11205 <unknown>
+
+fdot    za.s[w8, 5], {z16.h - z19.h}, {z16.h - z19.h}  // 11000001-10110001-00010010-00000101
+// CHECK-INST: fdot    za.s[w8, 5, vgx4], { z16.h - z19.h }, { z16.h - z19.h }
+// CHECK-ENCODING: [0x05,0x12,0xb1,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1b11205 <unknown>
+
+fdot    za.s[w8, 1, vgx4], {z0.h - z3.h}, {z28.h - z31.h}  // 11000001-10111101-00010000-00000001
+// CHECK-INST: fdot    za.s[w8, 1, vgx4], { z0.h - z3.h }, { z28.h - z31.h }
+// CHECK-ENCODING: [0x01,0x10,0xbd,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1bd1001 <unknown>
+
+fdot    za.s[w8, 1], {z0.h - z3.h}, {z28.h - z31.h}  // 11000001-10111101-00010000-00000001
+// CHECK-INST: fdot    za.s[w8, 1, vgx4], { z0.h - z3.h }, { z28.h - z31.h }
+// CHECK-ENCODING: [0x01,0x10,0xbd,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1bd1001 <unknown>
+
+fdot    za.s[w10, 0, vgx4], {z16.h - z19.h}, {z20.h - z23.h}  // 11000001-10110101-01010010-00000000
+// CHECK-INST: fdot    za.s[w10, 0, vgx4], { z16.h - z19.h }, { z20.h - z23.h }
+// CHECK-ENCODING: [0x00,0x52,0xb5,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1b55200 <unknown>
+
+fdot    za.s[w10, 0], {z16.h - z19.h}, {z20.h - z23.h}  // 11000001-10110101-01010010-00000000
+// CHECK-INST: fdot    za.s[w10, 0, vgx4], { z16.h - z19.h }, { z20.h - z23.h }
+// CHECK-ENCODING: [0x00,0x52,0xb5,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1b55200 <unknown>
+
+fdot    za.s[w8, 0, vgx4], {z12.h - z15.h}, {z0.h - z3.h}  // 11000001-10100001-00010001-10000000
+// CHECK-INST: fdot    za.s[w8, 0, vgx4], { z12.h - z15.h }, { z0.h - z3.h }
+// CHECK-ENCODING: [0x80,0x11,0xa1,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1a11180 <unknown>
+
+fdot    za.s[w8, 0], {z12.h - z15.h}, {z0.h - z3.h}  // 11000001-10100001-00010001-10000000
+// CHECK-INST: fdot    za.s[w8, 0, vgx4], { z12.h - z15.h }, { z0.h - z3.h }
+// CHECK-ENCODING: [0x80,0x11,0xa1,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1a11180 <unknown>
+
+fdot    za.s[w10, 1, vgx4], {z0.h - z3.h}, {z24.h - z27.h}  // 11000001-10111001-01010000-00000001
+// CHECK-INST: fdot    za.s[w10, 1, vgx4], { z0.h - z3.h }, { z24.h - z27.h }
+// CHECK-ENCODING: [0x01,0x50,0xb9,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1b95001 <unknown>
+
+fdot    za.s[w10, 1], {z0.h - z3.h}, {z24.h - z27.h}  // 11000001-10111001-01010000-00000001
+// CHECK-INST: fdot    za.s[w10, 1, vgx4], { z0.h - z3.h }, { z24.h - z27.h }
+// CHECK-ENCODING: [0x01,0x50,0xb9,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1b95001 <unknown>
+
+fdot    za.s[w8, 5, vgx4], {z20.h - z23.h}, {z28.h - z31.h}  // 11000001-10111101-00010010-10000101
+// CHECK-INST: fdot    za.s[w8, 5, vgx4], { z20.h - z23.h }, { z28.h - z31.h }
+// CHECK-ENCODING: [0x85,0x12,0xbd,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1bd1285 <unknown>
+
+fdot    za.s[w8, 5], {z20.h - z23.h}, {z28.h - z31.h}  // 11000001-10111101-00010010-10000101
+// CHECK-INST: fdot    za.s[w8, 5, vgx4], { z20.h - z23.h }, { z28.h - z31.h }
+// CHECK-ENCODING: [0x85,0x12,0xbd,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1bd1285 <unknown>
+
+fdot    za.s[w11, 2, vgx4], {z8.h - z11.h}, {z0.h - z3.h}  // 11000001-10100001-01110001-00000010
+// CHECK-INST: fdot    za.s[w11, 2, vgx4], { z8.h - z11.h }, { z0.h - z3.h }
+// CHECK-ENCODING: [0x02,0x71,0xa1,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1a17102 <unknown>
+
+fdot    za.s[w11, 2], {z8.h - z11.h}, {z0.h - z3.h}  // 11000001-10100001-01110001-00000010
+// CHECK-INST: fdot    za.s[w11, 2, vgx4], { z8.h - z11.h }, { z0.h - z3.h }
+// CHECK-ENCODING: [0x02,0x71,0xa1,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1a17102 <unknown>
+
+fdot    za.s[w9, 7, vgx4], {z12.h - z15.h}, {z8.h - z11.h}  // 11000001-10101001-00110001-10000111
+// CHECK-INST: fdot    za.s[w9, 7, vgx4], { z12.h - z15.h }, { z8.h - z11.h }
+// CHECK-ENCODING: [0x87,0x31,0xa9,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1a93187 <unknown>
+
+fdot    za.s[w9, 7], {z12.h - z15.h}, {z8.h - z11.h}  // 11000001-10101001-00110001-10000111
+// CHECK-INST: fdot    za.s[w9, 7, vgx4], { z12.h - z15.h }, { z8.h - z11.h }
+// CHECK-ENCODING: [0x87,0x31,0xa9,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1a93187 <unknown>
+
index 5e621f6..742a97a 100644 (file)
 // RUN:        | FileCheck %s --check-prefixes=CHECK-ENCODING,CHECK-INST
 
 
+sdot    za.s[w8, 0, vgx2], {z0.h, z1.h}, z0.h  // 11000001-01100000-00010100-00001000
+// CHECK-INST: sdot    za.s[w8, 0, vgx2], { z0.h, z1.h }, z0.h
+// CHECK-ENCODING: [0x08,0x14,0x60,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1601408 <unknown>
+
+sdot    za.s[w8, 0], {z0.h, z1.h}, z0.h  // 11000001-01100000-00010100-00001000
+// CHECK-INST: sdot    za.s[w8, 0, vgx2], { z0.h, z1.h }, z0.h
+// CHECK-ENCODING: [0x08,0x14,0x60,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1601408 <unknown>
+
+sdot    za.s[w10, 5, vgx2], {z10.h, z11.h}, z5.h  // 11000001-01100101-01010101-01001101
+// CHECK-INST: sdot    za.s[w10, 5, vgx2], { z10.h, z11.h }, z5.h
+// CHECK-ENCODING: [0x4d,0x55,0x65,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c165554d <unknown>
+
+sdot    za.s[w10, 5], {z10.h, z11.h}, z5.h  // 11000001-01100101-01010101-01001101
+// CHECK-INST: sdot    za.s[w10, 5, vgx2], { z10.h, z11.h }, z5.h
+// CHECK-ENCODING: [0x4d,0x55,0x65,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c165554d <unknown>
+
+sdot    za.s[w11, 7, vgx2], {z13.h, z14.h}, z8.h  // 11000001-01101000-01110101-10101111
+// CHECK-INST: sdot    za.s[w11, 7, vgx2], { z13.h, z14.h }, z8.h
+// CHECK-ENCODING: [0xaf,0x75,0x68,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c16875af <unknown>
+
+sdot    za.s[w11, 7], {z13.h, z14.h}, z8.h  // 11000001-01101000-01110101-10101111
+// CHECK-INST: sdot    za.s[w11, 7, vgx2], { z13.h, z14.h }, z8.h
+// CHECK-ENCODING: [0xaf,0x75,0x68,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c16875af <unknown>
+
+sdot    za.s[w11, 7, vgx2], {z31.h, z0.h}, z15.h  // 11000001-01101111-01110111-11101111
+// CHECK-INST: sdot    za.s[w11, 7, vgx2], { z31.h, z0.h }, z15.h
+// CHECK-ENCODING: [0xef,0x77,0x6f,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c16f77ef <unknown>
+
+sdot    za.s[w11, 7], {z31.h, z0.h}, z15.h  // 11000001-01101111-01110111-11101111
+// CHECK-INST: sdot    za.s[w11, 7, vgx2], { z31.h, z0.h }, z15.h
+// CHECK-ENCODING: [0xef,0x77,0x6f,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c16f77ef <unknown>
+
+sdot    za.s[w8, 5, vgx2], {z17.h, z18.h}, z0.h  // 11000001-01100000-00010110-00101101
+// CHECK-INST: sdot    za.s[w8, 5, vgx2], { z17.h, z18.h }, z0.h
+// CHECK-ENCODING: [0x2d,0x16,0x60,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c160162d <unknown>
+
+sdot    za.s[w8, 5], {z17.h, z18.h}, z0.h  // 11000001-01100000-00010110-00101101
+// CHECK-INST: sdot    za.s[w8, 5, vgx2], { z17.h, z18.h }, z0.h
+// CHECK-ENCODING: [0x2d,0x16,0x60,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c160162d <unknown>
+
+sdot    za.s[w8, 1, vgx2], {z1.h, z2.h}, z14.h  // 11000001-01101110-00010100-00101001
+// CHECK-INST: sdot    za.s[w8, 1, vgx2], { z1.h, z2.h }, z14.h
+// CHECK-ENCODING: [0x29,0x14,0x6e,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c16e1429 <unknown>
+
+sdot    za.s[w8, 1], {z1.h, z2.h}, z14.h  // 11000001-01101110-00010100-00101001
+// CHECK-INST: sdot    za.s[w8, 1, vgx2], { z1.h, z2.h }, z14.h
+// CHECK-ENCODING: [0x29,0x14,0x6e,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c16e1429 <unknown>
+
+sdot    za.s[w10, 0, vgx2], {z19.h, z20.h}, z4.h  // 11000001-01100100-01010110-01101000
+// CHECK-INST: sdot    za.s[w10, 0, vgx2], { z19.h, z20.h }, z4.h
+// CHECK-ENCODING: [0x68,0x56,0x64,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1645668 <unknown>
+
+sdot    za.s[w10, 0], {z19.h, z20.h}, z4.h  // 11000001-01100100-01010110-01101000
+// CHECK-INST: sdot    za.s[w10, 0, vgx2], { z19.h, z20.h }, z4.h
+// CHECK-ENCODING: [0x68,0x56,0x64,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1645668 <unknown>
+
+sdot    za.s[w8, 0, vgx2], {z12.h, z13.h}, z2.h  // 11000001-01100010-00010101-10001000
+// CHECK-INST: sdot    za.s[w8, 0, vgx2], { z12.h, z13.h }, z2.h
+// CHECK-ENCODING: [0x88,0x15,0x62,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1621588 <unknown>
+
+sdot    za.s[w8, 0], {z12.h, z13.h}, z2.h  // 11000001-01100010-00010101-10001000
+// CHECK-INST: sdot    za.s[w8, 0, vgx2], { z12.h, z13.h }, z2.h
+// CHECK-ENCODING: [0x88,0x15,0x62,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1621588 <unknown>
+
+sdot    za.s[w10, 1, vgx2], {z1.h, z2.h}, z10.h  // 11000001-01101010-01010100-00101001
+// CHECK-INST: sdot    za.s[w10, 1, vgx2], { z1.h, z2.h }, z10.h
+// CHECK-ENCODING: [0x29,0x54,0x6a,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c16a5429 <unknown>
+
+sdot    za.s[w10, 1], {z1.h, z2.h}, z10.h  // 11000001-01101010-01010100-00101001
+// CHECK-INST: sdot    za.s[w10, 1, vgx2], { z1.h, z2.h }, z10.h
+// CHECK-ENCODING: [0x29,0x54,0x6a,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c16a5429 <unknown>
+
+sdot    za.s[w8, 5, vgx2], {z22.h, z23.h}, z14.h  // 11000001-01101110-00010110-11001101
+// CHECK-INST: sdot    za.s[w8, 5, vgx2], { z22.h, z23.h }, z14.h
+// CHECK-ENCODING: [0xcd,0x16,0x6e,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c16e16cd <unknown>
+
+sdot    za.s[w8, 5], {z22.h, z23.h}, z14.h  // 11000001-01101110-00010110-11001101
+// CHECK-INST: sdot    za.s[w8, 5, vgx2], { z22.h, z23.h }, z14.h
+// CHECK-ENCODING: [0xcd,0x16,0x6e,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c16e16cd <unknown>
+
+sdot    za.s[w11, 2, vgx2], {z9.h, z10.h}, z1.h  // 11000001-01100001-01110101-00101010
+// CHECK-INST: sdot    za.s[w11, 2, vgx2], { z9.h, z10.h }, z1.h
+// CHECK-ENCODING: [0x2a,0x75,0x61,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c161752a <unknown>
+
+sdot    za.s[w11, 2], {z9.h, z10.h}, z1.h  // 11000001-01100001-01110101-00101010
+// CHECK-INST: sdot    za.s[w11, 2, vgx2], { z9.h, z10.h }, z1.h
+// CHECK-ENCODING: [0x2a,0x75,0x61,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c161752a <unknown>
+
+sdot    za.s[w9, 7, vgx2], {z12.h, z13.h}, z11.h  // 11000001-01101011-00110101-10001111
+// CHECK-INST: sdot    za.s[w9, 7, vgx2], { z12.h, z13.h }, z11.h
+// CHECK-ENCODING: [0x8f,0x35,0x6b,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c16b358f <unknown>
+
+sdot    za.s[w9, 7], {z12.h, z13.h}, z11.h  // 11000001-01101011-00110101-10001111
+// CHECK-INST: sdot    za.s[w9, 7, vgx2], { z12.h, z13.h }, z11.h
+// CHECK-ENCODING: [0x8f,0x35,0x6b,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c16b358f <unknown>
+
+
 sdot    za.s[w8, 0, vgx2], {z0.h, z1.h}, z0.h[0]  // 11000001-01010000-00010000-00000000
 // CHECK-INST: sdot    za.s[w8, 0, vgx2], { z0.h, z1.h }, z0.h[0]
 // CHECK-ENCODING: [0x00,0x10,0x50,0xc1]
@@ -157,6 +302,151 @@ sdot    za.s[w9, 7], {z12.h, z13.h}, z11.h[2]  // 11000001-01011011-00111001-100
 // CHECK-UNKNOWN: c15b3987 <unknown>
 
 
+sdot    za.s[w8, 0, vgx2], {z0.h, z1.h}, {z0.h, z1.h}  // 11000001-11100000-00010100-00001000
+// CHECK-INST: sdot    za.s[w8, 0, vgx2], { z0.h, z1.h }, { z0.h, z1.h }
+// CHECK-ENCODING: [0x08,0x14,0xe0,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1e01408 <unknown>
+
+sdot    za.s[w8, 0], {z0.h, z1.h}, {z0.h, z1.h}  // 11000001-11100000-00010100-00001000
+// CHECK-INST: sdot    za.s[w8, 0, vgx2], { z0.h, z1.h }, { z0.h, z1.h }
+// CHECK-ENCODING: [0x08,0x14,0xe0,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1e01408 <unknown>
+
+sdot    za.s[w10, 5, vgx2], {z10.h, z11.h}, {z20.h, z21.h}  // 11000001-11110100-01010101-01001101
+// CHECK-INST: sdot    za.s[w10, 5, vgx2], { z10.h, z11.h }, { z20.h, z21.h }
+// CHECK-ENCODING: [0x4d,0x55,0xf4,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1f4554d <unknown>
+
+sdot    za.s[w10, 5], {z10.h, z11.h}, {z20.h, z21.h}  // 11000001-11110100-01010101-01001101
+// CHECK-INST: sdot    za.s[w10, 5, vgx2], { z10.h, z11.h }, { z20.h, z21.h }
+// CHECK-ENCODING: [0x4d,0x55,0xf4,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1f4554d <unknown>
+
+sdot    za.s[w11, 7, vgx2], {z12.h, z13.h}, {z8.h, z9.h}  // 11000001-11101000-01110101-10001111
+// CHECK-INST: sdot    za.s[w11, 7, vgx2], { z12.h, z13.h }, { z8.h, z9.h }
+// CHECK-ENCODING: [0x8f,0x75,0xe8,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1e8758f <unknown>
+
+sdot    za.s[w11, 7], {z12.h, z13.h}, {z8.h, z9.h}  // 11000001-11101000-01110101-10001111
+// CHECK-INST: sdot    za.s[w11, 7, vgx2], { z12.h, z13.h }, { z8.h, z9.h }
+// CHECK-ENCODING: [0x8f,0x75,0xe8,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1e8758f <unknown>
+
+sdot    za.s[w11, 7, vgx2], {z30.h, z31.h}, {z30.h, z31.h}  // 11000001-11111110-01110111-11001111
+// CHECK-INST: sdot    za.s[w11, 7, vgx2], { z30.h, z31.h }, { z30.h, z31.h }
+// CHECK-ENCODING: [0xcf,0x77,0xfe,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1fe77cf <unknown>
+
+sdot    za.s[w11, 7], {z30.h, z31.h}, {z30.h, z31.h}  // 11000001-11111110-01110111-11001111
+// CHECK-INST: sdot    za.s[w11, 7, vgx2], { z30.h, z31.h }, { z30.h, z31.h }
+// CHECK-ENCODING: [0xcf,0x77,0xfe,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1fe77cf <unknown>
+
+sdot    za.s[w8, 5, vgx2], {z16.h, z17.h}, {z16.h, z17.h}  // 11000001-11110000-00010110-00001101
+// CHECK-INST: sdot    za.s[w8, 5, vgx2], { z16.h, z17.h }, { z16.h, z17.h }
+// CHECK-ENCODING: [0x0d,0x16,0xf0,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1f0160d <unknown>
+
+sdot    za.s[w8, 5], {z16.h, z17.h}, {z16.h, z17.h}  // 11000001-11110000-00010110-00001101
+// CHECK-INST: sdot    za.s[w8, 5, vgx2], { z16.h, z17.h }, { z16.h, z17.h }
+// CHECK-ENCODING: [0x0d,0x16,0xf0,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1f0160d <unknown>
+
+sdot    za.s[w8, 1, vgx2], {z0.h, z1.h}, {z30.h, z31.h}  // 11000001-11111110-00010100-00001001
+// CHECK-INST: sdot    za.s[w8, 1, vgx2], { z0.h, z1.h }, { z30.h, z31.h }
+// CHECK-ENCODING: [0x09,0x14,0xfe,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1fe1409 <unknown>
+
+sdot    za.s[w8, 1], {z0.h, z1.h}, {z30.h, z31.h}  // 11000001-11111110-00010100-00001001
+// CHECK-INST: sdot    za.s[w8, 1, vgx2], { z0.h, z1.h }, { z30.h, z31.h }
+// CHECK-ENCODING: [0x09,0x14,0xfe,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1fe1409 <unknown>
+
+sdot    za.s[w10, 0, vgx2], {z18.h, z19.h}, {z20.h, z21.h}  // 11000001-11110100-01010110-01001000
+// CHECK-INST: sdot    za.s[w10, 0, vgx2], { z18.h, z19.h }, { z20.h, z21.h }
+// CHECK-ENCODING: [0x48,0x56,0xf4,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1f45648 <unknown>
+
+sdot    za.s[w10, 0], {z18.h, z19.h}, {z20.h, z21.h}  // 11000001-11110100-01010110-01001000
+// CHECK-INST: sdot    za.s[w10, 0, vgx2], { z18.h, z19.h }, { z20.h, z21.h }
+// CHECK-ENCODING: [0x48,0x56,0xf4,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1f45648 <unknown>
+
+sdot    za.s[w8, 0, vgx2], {z12.h, z13.h}, {z2.h, z3.h}  // 11000001-11100010-00010101-10001000
+// CHECK-INST: sdot    za.s[w8, 0, vgx2], { z12.h, z13.h }, { z2.h, z3.h }
+// CHECK-ENCODING: [0x88,0x15,0xe2,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1e21588 <unknown>
+
+sdot    za.s[w8, 0], {z12.h, z13.h}, {z2.h, z3.h}  // 11000001-11100010-00010101-10001000
+// CHECK-INST: sdot    za.s[w8, 0, vgx2], { z12.h, z13.h }, { z2.h, z3.h }
+// CHECK-ENCODING: [0x88,0x15,0xe2,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1e21588 <unknown>
+
+sdot    za.s[w10, 1, vgx2], {z0.h, z1.h}, {z26.h, z27.h}  // 11000001-11111010-01010100-00001001
+// CHECK-INST: sdot    za.s[w10, 1, vgx2], { z0.h, z1.h }, { z26.h, z27.h }
+// CHECK-ENCODING: [0x09,0x54,0xfa,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1fa5409 <unknown>
+
+sdot    za.s[w10, 1], {z0.h, z1.h}, {z26.h, z27.h}  // 11000001-11111010-01010100-00001001
+// CHECK-INST: sdot    za.s[w10, 1, vgx2], { z0.h, z1.h }, { z26.h, z27.h }
+// CHECK-ENCODING: [0x09,0x54,0xfa,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1fa5409 <unknown>
+
+sdot    za.s[w8, 5, vgx2], {z22.h, z23.h}, {z30.h, z31.h}  // 11000001-11111110-00010110-11001101
+// CHECK-INST: sdot    za.s[w8, 5, vgx2], { z22.h, z23.h }, { z30.h, z31.h }
+// CHECK-ENCODING: [0xcd,0x16,0xfe,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1fe16cd <unknown>
+
+sdot    za.s[w8, 5], {z22.h, z23.h}, {z30.h, z31.h}  // 11000001-11111110-00010110-11001101
+// CHECK-INST: sdot    za.s[w8, 5, vgx2], { z22.h, z23.h }, { z30.h, z31.h }
+// CHECK-ENCODING: [0xcd,0x16,0xfe,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1fe16cd <unknown>
+
+sdot    za.s[w11, 2, vgx2], {z8.h, z9.h}, {z0.h, z1.h}  // 11000001-11100000-01110101-00001010
+// CHECK-INST: sdot    za.s[w11, 2, vgx2], { z8.h, z9.h }, { z0.h, z1.h }
+// CHECK-ENCODING: [0x0a,0x75,0xe0,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1e0750a <unknown>
+
+sdot    za.s[w11, 2], {z8.h, z9.h}, {z0.h, z1.h}  // 11000001-11100000-01110101-00001010
+// CHECK-INST: sdot    za.s[w11, 2, vgx2], { z8.h, z9.h }, { z0.h, z1.h }
+// CHECK-ENCODING: [0x0a,0x75,0xe0,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1e0750a <unknown>
+
+sdot    za.s[w9, 7, vgx2], {z12.h, z13.h}, {z10.h, z11.h}  // 11000001-11101010-00110101-10001111
+// CHECK-INST: sdot    za.s[w9, 7, vgx2], { z12.h, z13.h }, { z10.h, z11.h }
+// CHECK-ENCODING: [0x8f,0x35,0xea,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1ea358f <unknown>
+
+sdot    za.s[w9, 7], {z12.h, z13.h}, {z10.h, z11.h}  // 11000001-11101010-00110101-10001111
+// CHECK-INST: sdot    za.s[w9, 7, vgx2], { z12.h, z13.h }, { z10.h, z11.h }
+// CHECK-ENCODING: [0x8f,0x35,0xea,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1ea358f <unknown>
+
+
 sdot    za.s[w8, 0, vgx2], {z0.b, z1.b}, z0.b[0]  // 11000001-01010000-00010000-00100000
 // CHECK-INST: sdot    za.s[w8, 0, vgx2], { z0.b, z1.b }, z0.b[0]
 // CHECK-ENCODING: [0x20,0x10,0x50,0xc1]
@@ -447,6 +737,151 @@ sdot    za.d[w9, 7], {z12.h, z13.h}, z11.h[0]  // 11000001-11011011-00100001-100
 // CHECK-UNKNOWN: c1db218f <unknown>
 
 
+sdot    za.s[w8, 0, vgx4], {z0.h - z3.h}, z0.h  // 11000001-01110000-00010100-00001000
+// CHECK-INST: sdot    za.s[w8, 0, vgx4], { z0.h - z3.h }, z0.h
+// CHECK-ENCODING: [0x08,0x14,0x70,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1701408 <unknown>
+
+sdot    za.s[w8, 0], {z0.h - z3.h}, z0.h  // 11000001-01110000-00010100-00001000
+// CHECK-INST: sdot    za.s[w8, 0, vgx4], { z0.h - z3.h }, z0.h
+// CHECK-ENCODING: [0x08,0x14,0x70,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1701408 <unknown>
+
+sdot    za.s[w10, 5, vgx4], {z10.h - z13.h}, z5.h  // 11000001-01110101-01010101-01001101
+// CHECK-INST: sdot    za.s[w10, 5, vgx4], { z10.h - z13.h }, z5.h
+// CHECK-ENCODING: [0x4d,0x55,0x75,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c175554d <unknown>
+
+sdot    za.s[w10, 5], {z10.h - z13.h}, z5.h  // 11000001-01110101-01010101-01001101
+// CHECK-INST: sdot    za.s[w10, 5, vgx4], { z10.h - z13.h }, z5.h
+// CHECK-ENCODING: [0x4d,0x55,0x75,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c175554d <unknown>
+
+sdot    za.s[w11, 7, vgx4], {z13.h - z16.h}, z8.h  // 11000001-01111000-01110101-10101111
+// CHECK-INST: sdot    za.s[w11, 7, vgx4], { z13.h - z16.h }, z8.h
+// CHECK-ENCODING: [0xaf,0x75,0x78,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c17875af <unknown>
+
+sdot    za.s[w11, 7], {z13.h - z16.h}, z8.h  // 11000001-01111000-01110101-10101111
+// CHECK-INST: sdot    za.s[w11, 7, vgx4], { z13.h - z16.h }, z8.h
+// CHECK-ENCODING: [0xaf,0x75,0x78,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c17875af <unknown>
+
+sdot    za.s[w11, 7, vgx4], {z31.h - z2.h}, z15.h  // 11000001-01111111-01110111-11101111
+// CHECK-INST: sdot    za.s[w11, 7, vgx4], { z31.h, z0.h, z1.h, z2.h }, z15.h
+// CHECK-ENCODING: [0xef,0x77,0x7f,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c17f77ef <unknown>
+
+sdot    za.s[w11, 7], {z31.h - z2.h}, z15.h  // 11000001-01111111-01110111-11101111
+// CHECK-INST: sdot    za.s[w11, 7, vgx4], { z31.h, z0.h, z1.h, z2.h }, z15.h
+// CHECK-ENCODING: [0xef,0x77,0x7f,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c17f77ef <unknown>
+
+sdot    za.s[w8, 5, vgx4], {z17.h - z20.h}, z0.h  // 11000001-01110000-00010110-00101101
+// CHECK-INST: sdot    za.s[w8, 5, vgx4], { z17.h - z20.h }, z0.h
+// CHECK-ENCODING: [0x2d,0x16,0x70,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c170162d <unknown>
+
+sdot    za.s[w8, 5], {z17.h - z20.h}, z0.h  // 11000001-01110000-00010110-00101101
+// CHECK-INST: sdot    za.s[w8, 5, vgx4], { z17.h - z20.h }, z0.h
+// CHECK-ENCODING: [0x2d,0x16,0x70,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c170162d <unknown>
+
+sdot    za.s[w8, 1, vgx4], {z1.h - z4.h}, z14.h  // 11000001-01111110-00010100-00101001
+// CHECK-INST: sdot    za.s[w8, 1, vgx4], { z1.h - z4.h }, z14.h
+// CHECK-ENCODING: [0x29,0x14,0x7e,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c17e1429 <unknown>
+
+sdot    za.s[w8, 1], {z1.h - z4.h}, z14.h  // 11000001-01111110-00010100-00101001
+// CHECK-INST: sdot    za.s[w8, 1, vgx4], { z1.h - z4.h }, z14.h
+// CHECK-ENCODING: [0x29,0x14,0x7e,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c17e1429 <unknown>
+
+sdot    za.s[w10, 0, vgx4], {z19.h - z22.h}, z4.h  // 11000001-01110100-01010110-01101000
+// CHECK-INST: sdot    za.s[w10, 0, vgx4], { z19.h - z22.h }, z4.h
+// CHECK-ENCODING: [0x68,0x56,0x74,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1745668 <unknown>
+
+sdot    za.s[w10, 0], {z19.h - z22.h}, z4.h  // 11000001-01110100-01010110-01101000
+// CHECK-INST: sdot    za.s[w10, 0, vgx4], { z19.h - z22.h }, z4.h
+// CHECK-ENCODING: [0x68,0x56,0x74,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1745668 <unknown>
+
+sdot    za.s[w8, 0, vgx4], {z12.h - z15.h}, z2.h  // 11000001-01110010-00010101-10001000
+// CHECK-INST: sdot    za.s[w8, 0, vgx4], { z12.h - z15.h }, z2.h
+// CHECK-ENCODING: [0x88,0x15,0x72,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1721588 <unknown>
+
+sdot    za.s[w8, 0], {z12.h - z15.h}, z2.h  // 11000001-01110010-00010101-10001000
+// CHECK-INST: sdot    za.s[w8, 0, vgx4], { z12.h - z15.h }, z2.h
+// CHECK-ENCODING: [0x88,0x15,0x72,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1721588 <unknown>
+
+sdot    za.s[w10, 1, vgx4], {z1.h - z4.h}, z10.h  // 11000001-01111010-01010100-00101001
+// CHECK-INST: sdot    za.s[w10, 1, vgx4], { z1.h - z4.h }, z10.h
+// CHECK-ENCODING: [0x29,0x54,0x7a,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c17a5429 <unknown>
+
+sdot    za.s[w10, 1], {z1.h - z4.h}, z10.h  // 11000001-01111010-01010100-00101001
+// CHECK-INST: sdot    za.s[w10, 1, vgx4], { z1.h - z4.h }, z10.h
+// CHECK-ENCODING: [0x29,0x54,0x7a,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c17a5429 <unknown>
+
+sdot    za.s[w8, 5, vgx4], {z22.h - z25.h}, z14.h  // 11000001-01111110-00010110-11001101
+// CHECK-INST: sdot    za.s[w8, 5, vgx4], { z22.h - z25.h }, z14.h
+// CHECK-ENCODING: [0xcd,0x16,0x7e,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c17e16cd <unknown>
+
+sdot    za.s[w8, 5], {z22.h - z25.h}, z14.h  // 11000001-01111110-00010110-11001101
+// CHECK-INST: sdot    za.s[w8, 5, vgx4], { z22.h - z25.h }, z14.h
+// CHECK-ENCODING: [0xcd,0x16,0x7e,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c17e16cd <unknown>
+
+sdot    za.s[w11, 2, vgx4], {z9.h - z12.h}, z1.h  // 11000001-01110001-01110101-00101010
+// CHECK-INST: sdot    za.s[w11, 2, vgx4], { z9.h - z12.h }, z1.h
+// CHECK-ENCODING: [0x2a,0x75,0x71,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c171752a <unknown>
+
+sdot    za.s[w11, 2], {z9.h - z12.h}, z1.h  // 11000001-01110001-01110101-00101010
+// CHECK-INST: sdot    za.s[w11, 2, vgx4], { z9.h - z12.h }, z1.h
+// CHECK-ENCODING: [0x2a,0x75,0x71,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c171752a <unknown>
+
+sdot    za.s[w9, 7, vgx4], {z12.h - z15.h}, z11.h  // 11000001-01111011-00110101-10001111
+// CHECK-INST: sdot    za.s[w9, 7, vgx4], { z12.h - z15.h }, z11.h
+// CHECK-ENCODING: [0x8f,0x35,0x7b,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c17b358f <unknown>
+
+sdot    za.s[w9, 7], {z12.h - z15.h}, z11.h  // 11000001-01111011-00110101-10001111
+// CHECK-INST: sdot    za.s[w9, 7, vgx4], { z12.h - z15.h }, z11.h
+// CHECK-ENCODING: [0x8f,0x35,0x7b,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c17b358f <unknown>
+
+
 sdot    za.s[w8, 0, vgx4], {z0.h - z3.h}, z0.h[0]  // 11000001-01010000-10010000-00000000
 // CHECK-INST: sdot    za.s[w8, 0, vgx4], { z0.h - z3.h }, z0.h[0]
 // CHECK-ENCODING: [0x00,0x90,0x50,0xc1]
@@ -592,6 +1027,152 @@ sdot    za.s[w9, 7], {z12.h - z15.h}, z11.h[2]  // 11000001-01011011-10111001-10
 // CHECK-UNKNOWN: c15bb987 <unknown>
 
 
+sdot    za.s[w8, 0, vgx4], {z0.h - z3.h}, {z0.h - z3.h}  // 11000001-11100001-00010100-00001000
+// CHECK-INST: sdot    za.s[w8, 0, vgx4], { z0.h - z3.h }, { z0.h - z3.h }
+// CHECK-ENCODING: [0x08,0x14,0xe1,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1e11408 <unknown>
+
+sdot    za.s[w8, 0], {z0.h - z3.h}, {z0.h - z3.h}  // 11000001-11100001-00010100-00001000
+// CHECK-INST: sdot    za.s[w8, 0, vgx4], { z0.h - z3.h }, { z0.h - z3.h }
+// CHECK-ENCODING: [0x08,0x14,0xe1,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1e11408 <unknown>
+
+sdot    za.s[w10, 5, vgx4], {z8.h - z11.h}, {z20.h - z23.h}  // 11000001-11110101-01010101-00001101
+// CHECK-INST: sdot    za.s[w10, 5, vgx4], { z8.h - z11.h }, { z20.h - z23.h }
+// CHECK-ENCODING: [0x0d,0x55,0xf5,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1f5550d <unknown>
+
+sdot    za.s[w10, 5], {z8.h - z11.h}, {z20.h - z23.h}  // 11000001-11110101-01010101-00001101
+// CHECK-INST: sdot    za.s[w10, 5, vgx4], { z8.h - z11.h }, { z20.h - z23.h }
+// CHECK-ENCODING: [0x0d,0x55,0xf5,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1f5550d <unknown>
+
+sdot    za.s[w11, 7, vgx4], {z12.h - z15.h}, {z8.h - z11.h}  // 11000001-11101001-01110101-10001111
+// CHECK-INST: sdot    za.s[w11, 7, vgx4], { z12.h - z15.h }, { z8.h - z11.h }
+// CHECK-ENCODING: [0x8f,0x75,0xe9,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1e9758f <unknown>
+
+sdot    za.s[w11, 7], {z12.h - z15.h}, {z8.h - z11.h}  // 11000001-11101001-01110101-10001111
+// CHECK-INST: sdot    za.s[w11, 7, vgx4], { z12.h - z15.h }, { z8.h - z11.h }
+// CHECK-ENCODING: [0x8f,0x75,0xe9,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1e9758f <unknown>
+
+sdot    za.s[w11, 7, vgx4], {z28.h - z31.h}, {z28.h - z31.h}  // 11000001-11111101-01110111-10001111
+// CHECK-INST: sdot    za.s[w11, 7, vgx4], { z28.h - z31.h }, { z28.h - z31.h }
+// CHECK-ENCODING: [0x8f,0x77,0xfd,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1fd778f <unknown>
+
+sdot    za.s[w11, 7], {z28.h - z31.h}, {z28.h - z31.h}  // 11000001-11111101-01110111-10001111
+// CHECK-INST: sdot    za.s[w11, 7, vgx4], { z28.h - z31.h }, { z28.h - z31.h }
+// CHECK-ENCODING: [0x8f,0x77,0xfd,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1fd778f <unknown>
+
+sdot    za.s[w8, 5, vgx4], {z16.h - z19.h}, {z16.h - z19.h}  // 11000001-11110001-00010110-00001101
+// CHECK-INST: sdot    za.s[w8, 5, vgx4], { z16.h - z19.h }, { z16.h - z19.h }
+// CHECK-ENCODING: [0x0d,0x16,0xf1,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1f1160d <unknown>
+
+sdot    za.s[w8, 5], {z16.h - z19.h}, {z16.h - z19.h}  // 11000001-11110001-00010110-00001101
+// CHECK-INST: sdot    za.s[w8, 5, vgx4], { z16.h - z19.h }, { z16.h - z19.h }
+// CHECK-ENCODING: [0x0d,0x16,0xf1,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1f1160d <unknown>
+
+sdot    za.s[w8, 1, vgx4], {z0.h - z3.h}, {z28.h - z31.h}  // 11000001-11111101-00010100-00001001
+// CHECK-INST: sdot    za.s[w8, 1, vgx4], { z0.h - z3.h }, { z28.h - z31.h }
+// CHECK-ENCODING: [0x09,0x14,0xfd,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1fd1409 <unknown>
+
+sdot    za.s[w8, 1], {z0.h - z3.h}, {z28.h - z31.h}  // 11000001-11111101-00010100-00001001
+// CHECK-INST: sdot    za.s[w8, 1, vgx4], { z0.h - z3.h }, { z28.h - z31.h }
+// CHECK-ENCODING: [0x09,0x14,0xfd,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1fd1409 <unknown>
+
+
+sdot    za.s[w10, 0, vgx4], {z16.h - z19.h}, {z20.h - z23.h}  // 11000001-11110101-01010110-00001000
+// CHECK-INST: sdot    za.s[w10, 0, vgx4], { z16.h - z19.h }, { z20.h - z23.h }
+// CHECK-ENCODING: [0x08,0x56,0xf5,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1f55608 <unknown>
+
+sdot    za.s[w10, 0], {z16.h - z19.h}, {z20.h - z23.h}  // 11000001-11110101-01010110-00001000
+// CHECK-INST: sdot    za.s[w10, 0, vgx4], { z16.h - z19.h }, { z20.h - z23.h }
+// CHECK-ENCODING: [0x08,0x56,0xf5,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1f55608 <unknown>
+
+sdot    za.s[w8, 0, vgx4], {z12.h - z15.h}, {z0.h - z3.h}  // 11000001-11100001-00010101-10001000
+// CHECK-INST: sdot    za.s[w8, 0, vgx4], { z12.h - z15.h }, { z0.h - z3.h }
+// CHECK-ENCODING: [0x88,0x15,0xe1,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1e11588 <unknown>
+
+sdot    za.s[w8, 0], {z12.h - z15.h}, {z0.h - z3.h}  // 11000001-11100001-00010101-10001000
+// CHECK-INST: sdot    za.s[w8, 0, vgx4], { z12.h - z15.h }, { z0.h - z3.h }
+// CHECK-ENCODING: [0x88,0x15,0xe1,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1e11588 <unknown>
+
+sdot    za.s[w10, 1, vgx4], {z0.h - z3.h}, {z24.h - z27.h}  // 11000001-11111001-01010100-00001001
+// CHECK-INST: sdot    za.s[w10, 1, vgx4], { z0.h - z3.h }, { z24.h - z27.h }
+// CHECK-ENCODING: [0x09,0x54,0xf9,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1f95409 <unknown>
+
+sdot    za.s[w10, 1], {z0.h - z3.h}, {z24.h - z27.h}  // 11000001-11111001-01010100-00001001
+// CHECK-INST: sdot    za.s[w10, 1, vgx4], { z0.h - z3.h }, { z24.h - z27.h }
+// CHECK-ENCODING: [0x09,0x54,0xf9,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1f95409 <unknown>
+
+sdot    za.s[w8, 5, vgx4], {z20.h - z23.h}, {z28.h - z31.h}  // 11000001-11111101-00010110-10001101
+// CHECK-INST: sdot    za.s[w8, 5, vgx4], { z20.h - z23.h }, { z28.h - z31.h }
+// CHECK-ENCODING: [0x8d,0x16,0xfd,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1fd168d <unknown>
+
+sdot    za.s[w8, 5], {z20.h - z23.h}, {z28.h - z31.h}  // 11000001-11111101-00010110-10001101
+// CHECK-INST: sdot    za.s[w8, 5, vgx4], { z20.h - z23.h }, { z28.h - z31.h }
+// CHECK-ENCODING: [0x8d,0x16,0xfd,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1fd168d <unknown>
+
+sdot    za.s[w11, 2, vgx4], {z8.h - z11.h}, {z0.h - z3.h}  // 11000001-11100001-01110101-00001010
+// CHECK-INST: sdot    za.s[w11, 2, vgx4], { z8.h - z11.h }, { z0.h - z3.h }
+// CHECK-ENCODING: [0x0a,0x75,0xe1,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1e1750a <unknown>
+
+sdot    za.s[w11, 2], {z8.h - z11.h}, {z0.h - z3.h}  // 11000001-11100001-01110101-00001010
+// CHECK-INST: sdot    za.s[w11, 2, vgx4], { z8.h - z11.h }, { z0.h - z3.h }
+// CHECK-ENCODING: [0x0a,0x75,0xe1,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1e1750a <unknown>
+
+sdot    za.s[w9, 7, vgx4], {z12.h - z15.h}, {z8.h - z11.h}  // 11000001-11101001-00110101-10001111
+// CHECK-INST: sdot    za.s[w9, 7, vgx4], { z12.h - z15.h }, { z8.h - z11.h }
+// CHECK-ENCODING: [0x8f,0x35,0xe9,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1e9358f <unknown>
+
+sdot    za.s[w9, 7], {z12.h - z15.h}, {z8.h - z11.h}  // 11000001-11101001-00110101-10001111
+// CHECK-INST: sdot    za.s[w9, 7, vgx4], { z12.h - z15.h }, { z8.h - z11.h }
+// CHECK-ENCODING: [0x8f,0x35,0xe9,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1e9358f <unknown>
+
+
 sdot    za.s[w8, 0, vgx4], {z0.b - z3.b}, z0.b[0]  // 11000001-01010000-10010000-00100000
 // CHECK-INST: sdot    za.s[w8, 0, vgx4], { z0.b - z3.b }, z0.b[0]
 // CHECK-ENCODING: [0x20,0x90,0x50,0xc1]
index 6aa16ee..cee00ce 100644 (file)
 // RUN:        | FileCheck %s --check-prefixes=CHECK-ENCODING,CHECK-INST
 
 
+sudot   za.s[w8, 0, vgx2], {z0.b, z1.b}, z0.b  // 11000001-00100000-00010100-00011000
+// CHECK-INST: sudot   za.s[w8, 0, vgx2], { z0.b, z1.b }, z0.b
+// CHECK-ENCODING: [0x18,0x14,0x20,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1201418 <unknown>
+
+sudot   za.s[w8, 0], {z0.b, z1.b}, z0.b  // 11000001-00100000-00010100-00011000
+// CHECK-INST: sudot   za.s[w8, 0, vgx2], { z0.b, z1.b }, z0.b
+// CHECK-ENCODING: [0x18,0x14,0x20,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1201418 <unknown>
+
+sudot   za.s[w10, 5, vgx2], {z10.b, z11.b}, z5.b  // 11000001-00100101-01010101-01011101
+// CHECK-INST: sudot   za.s[w10, 5, vgx2], { z10.b, z11.b }, z5.b
+// CHECK-ENCODING: [0x5d,0x55,0x25,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c125555d <unknown>
+
+sudot   za.s[w10, 5], {z10.b, z11.b}, z5.b  // 11000001-00100101-01010101-01011101
+// CHECK-INST: sudot   za.s[w10, 5, vgx2], { z10.b, z11.b }, z5.b
+// CHECK-ENCODING: [0x5d,0x55,0x25,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c125555d <unknown>
+
+sudot   za.s[w11, 7, vgx2], {z13.b, z14.b}, z8.b  // 11000001-00101000-01110101-10111111
+// CHECK-INST: sudot   za.s[w11, 7, vgx2], { z13.b, z14.b }, z8.b
+// CHECK-ENCODING: [0xbf,0x75,0x28,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c12875bf <unknown>
+
+sudot   za.s[w11, 7], {z13.b, z14.b}, z8.b  // 11000001-00101000-01110101-10111111
+// CHECK-INST: sudot   za.s[w11, 7, vgx2], { z13.b, z14.b }, z8.b
+// CHECK-ENCODING: [0xbf,0x75,0x28,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c12875bf <unknown>
+
+sudot   za.s[w11, 7, vgx2], {z31.b, z0.b}, z15.b  // 11000001-00101111-01110111-11111111
+// CHECK-INST: sudot   za.s[w11, 7, vgx2], { z31.b, z0.b }, z15.b
+// CHECK-ENCODING: [0xff,0x77,0x2f,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c12f77ff <unknown>
+
+sudot   za.s[w11, 7], {z31.b, z0.b}, z15.b  // 11000001-00101111-01110111-11111111
+// CHECK-INST: sudot   za.s[w11, 7, vgx2], { z31.b, z0.b }, z15.b
+// CHECK-ENCODING: [0xff,0x77,0x2f,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c12f77ff <unknown>
+
+sudot   za.s[w8, 5, vgx2], {z17.b, z18.b}, z0.b  // 11000001-00100000-00010110-00111101
+// CHECK-INST: sudot   za.s[w8, 5, vgx2], { z17.b, z18.b }, z0.b
+// CHECK-ENCODING: [0x3d,0x16,0x20,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c120163d <unknown>
+
+sudot   za.s[w8, 5], {z17.b, z18.b}, z0.b  // 11000001-00100000-00010110-00111101
+// CHECK-INST: sudot   za.s[w8, 5, vgx2], { z17.b, z18.b }, z0.b
+// CHECK-ENCODING: [0x3d,0x16,0x20,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c120163d <unknown>
+
+sudot   za.s[w8, 1, vgx2], {z1.b, z2.b}, z14.b  // 11000001-00101110-00010100-00111001
+// CHECK-INST: sudot   za.s[w8, 1, vgx2], { z1.b, z2.b }, z14.b
+// CHECK-ENCODING: [0x39,0x14,0x2e,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c12e1439 <unknown>
+
+sudot   za.s[w8, 1], {z1.b, z2.b}, z14.b  // 11000001-00101110-00010100-00111001
+// CHECK-INST: sudot   za.s[w8, 1, vgx2], { z1.b, z2.b }, z14.b
+// CHECK-ENCODING: [0x39,0x14,0x2e,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c12e1439 <unknown>
+
+
+sudot   za.s[w10, 0, vgx2], {z19.b, z20.b}, z4.b  // 11000001-00100100-01010110-01111000
+// CHECK-INST: sudot   za.s[w10, 0, vgx2], { z19.b, z20.b }, z4.b
+// CHECK-ENCODING: [0x78,0x56,0x24,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1245678 <unknown>
+
+sudot   za.s[w10, 0], {z19.b, z20.b}, z4.b  // 11000001-00100100-01010110-01111000
+// CHECK-INST: sudot   za.s[w10, 0, vgx2], { z19.b, z20.b }, z4.b
+// CHECK-ENCODING: [0x78,0x56,0x24,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1245678 <unknown>
+
+sudot   za.s[w8, 0, vgx2], {z12.b, z13.b}, z2.b  // 11000001-00100010-00010101-10011000
+// CHECK-INST: sudot   za.s[w8, 0, vgx2], { z12.b, z13.b }, z2.b
+// CHECK-ENCODING: [0x98,0x15,0x22,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1221598 <unknown>
+
+sudot   za.s[w8, 0], {z12.b, z13.b}, z2.b  // 11000001-00100010-00010101-10011000
+// CHECK-INST: sudot   za.s[w8, 0, vgx2], { z12.b, z13.b }, z2.b
+// CHECK-ENCODING: [0x98,0x15,0x22,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1221598 <unknown>
+
+sudot   za.s[w10, 1, vgx2], {z1.b, z2.b}, z10.b  // 11000001-00101010-01010100-00111001
+// CHECK-INST: sudot   za.s[w10, 1, vgx2], { z1.b, z2.b }, z10.b
+// CHECK-ENCODING: [0x39,0x54,0x2a,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c12a5439 <unknown>
+
+sudot   za.s[w10, 1], {z1.b, z2.b}, z10.b  // 11000001-00101010-01010100-00111001
+// CHECK-INST: sudot   za.s[w10, 1, vgx2], { z1.b, z2.b }, z10.b
+// CHECK-ENCODING: [0x39,0x54,0x2a,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c12a5439 <unknown>
+
+sudot   za.s[w8, 5, vgx2], {z22.b, z23.b}, z14.b  // 11000001-00101110-00010110-11011101
+// CHECK-INST: sudot   za.s[w8, 5, vgx2], { z22.b, z23.b }, z14.b
+// CHECK-ENCODING: [0xdd,0x16,0x2e,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c12e16dd <unknown>
+
+sudot   za.s[w8, 5], {z22.b, z23.b}, z14.b  // 11000001-00101110-00010110-11011101
+// CHECK-INST: sudot   za.s[w8, 5, vgx2], { z22.b, z23.b }, z14.b
+// CHECK-ENCODING: [0xdd,0x16,0x2e,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c12e16dd <unknown>
+
+sudot   za.s[w11, 2, vgx2], {z9.b, z10.b}, z1.b  // 11000001-00100001-01110101-00111010
+// CHECK-INST: sudot   za.s[w11, 2, vgx2], { z9.b, z10.b }, z1.b
+// CHECK-ENCODING: [0x3a,0x75,0x21,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c121753a <unknown>
+
+sudot   za.s[w11, 2], {z9.b, z10.b}, z1.b  // 11000001-00100001-01110101-00111010
+// CHECK-INST: sudot   za.s[w11, 2, vgx2], { z9.b, z10.b }, z1.b
+// CHECK-ENCODING: [0x3a,0x75,0x21,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c121753a <unknown>
+
+sudot   za.s[w9, 7, vgx2], {z12.b, z13.b}, z11.b  // 11000001-00101011-00110101-10011111
+// CHECK-INST: sudot   za.s[w9, 7, vgx2], { z12.b, z13.b }, z11.b
+// CHECK-ENCODING: [0x9f,0x35,0x2b,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c12b359f <unknown>
+
+sudot   za.s[w9, 7], {z12.b, z13.b}, z11.b  // 11000001-00101011-00110101-10011111
+// CHECK-INST: sudot   za.s[w9, 7, vgx2], { z12.b, z13.b }, z11.b
+// CHECK-ENCODING: [0x9f,0x35,0x2b,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c12b359f <unknown>
+
+
 sudot   za.s[w8, 0, vgx2], {z0.b, z1.b}, z0.b[0]  // 11000001-01010000-00010000-00111000
 // CHECK-INST: sudot   za.s[w8, 0, vgx2], { z0.b, z1.b }, z0.b[0]
 // CHECK-ENCODING: [0x38,0x10,0x50,0xc1]
@@ -156,6 +302,150 @@ sudot   za.s[w9, 7], {z12.b, z13.b}, z11.b[2]  // 11000001-01011011-00111001-101
 // CHECK-ERROR: instruction requires: sme2
 // CHECK-UNKNOWN: c15b39bf <unknown>
 
+sudot   za.s[w8, 0, vgx4], {z0.b - z3.b}, z0.b  // 11000001-00110000-00010100-00011000
+// CHECK-INST: sudot   za.s[w8, 0, vgx4], { z0.b - z3.b }, z0.b
+// CHECK-ENCODING: [0x18,0x14,0x30,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1301418 <unknown>
+
+sudot   za.s[w8, 0], {z0.b - z3.b}, z0.b  // 11000001-00110000-00010100-00011000
+// CHECK-INST: sudot   za.s[w8, 0, vgx4], { z0.b - z3.b }, z0.b
+// CHECK-ENCODING: [0x18,0x14,0x30,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1301418 <unknown>
+
+sudot   za.s[w10, 5, vgx4], {z10.b - z13.b}, z5.b  // 11000001-00110101-01010101-01011101
+// CHECK-INST: sudot   za.s[w10, 5, vgx4], { z10.b - z13.b }, z5.b
+// CHECK-ENCODING: [0x5d,0x55,0x35,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c135555d <unknown>
+
+sudot   za.s[w10, 5], {z10.b - z13.b}, z5.b  // 11000001-00110101-01010101-01011101
+// CHECK-INST: sudot   za.s[w10, 5, vgx4], { z10.b - z13.b }, z5.b
+// CHECK-ENCODING: [0x5d,0x55,0x35,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c135555d <unknown>
+
+sudot   za.s[w11, 7, vgx4], {z13.b - z16.b}, z8.b  // 11000001-00111000-01110101-10111111
+// CHECK-INST: sudot   za.s[w11, 7, vgx4], { z13.b - z16.b }, z8.b
+// CHECK-ENCODING: [0xbf,0x75,0x38,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c13875bf <unknown>
+
+sudot   za.s[w11, 7], {z13.b - z16.b}, z8.b  // 11000001-00111000-01110101-10111111
+// CHECK-INST: sudot   za.s[w11, 7, vgx4], { z13.b - z16.b }, z8.b
+// CHECK-ENCODING: [0xbf,0x75,0x38,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c13875bf <unknown>
+
+sudot   za.s[w11, 7, vgx4], {z31.b - z2.b}, z15.b  // 11000001-00111111-01110111-11111111
+// CHECK-INST: sudot   za.s[w11, 7, vgx4], { z31.b, z0.b, z1.b, z2.b }, z15.b
+// CHECK-ENCODING: [0xff,0x77,0x3f,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c13f77ff <unknown>
+
+sudot   za.s[w11, 7], {z31.b - z2.b}, z15.b  // 11000001-00111111-01110111-11111111
+// CHECK-INST: sudot   za.s[w11, 7, vgx4], { z31.b, z0.b, z1.b, z2.b }, z15.b
+// CHECK-ENCODING: [0xff,0x77,0x3f,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c13f77ff <unknown>
+
+sudot   za.s[w8, 5, vgx4], {z17.b - z20.b}, z0.b  // 11000001-00110000-00010110-00111101
+// CHECK-INST: sudot   za.s[w8, 5, vgx4], { z17.b - z20.b }, z0.b
+// CHECK-ENCODING: [0x3d,0x16,0x30,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c130163d <unknown>
+
+sudot   za.s[w8, 5], {z17.b - z20.b}, z0.b  // 11000001-00110000-00010110-00111101
+// CHECK-INST: sudot   za.s[w8, 5, vgx4], { z17.b - z20.b }, z0.b
+// CHECK-ENCODING: [0x3d,0x16,0x30,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c130163d <unknown>
+
+sudot   za.s[w8, 1, vgx4], {z1.b - z4.b}, z14.b  // 11000001-00111110-00010100-00111001
+// CHECK-INST: sudot   za.s[w8, 1, vgx4], { z1.b - z4.b }, z14.b
+// CHECK-ENCODING: [0x39,0x14,0x3e,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c13e1439 <unknown>
+
+sudot   za.s[w8, 1], {z1.b - z4.b}, z14.b  // 11000001-00111110-00010100-00111001
+// CHECK-INST: sudot   za.s[w8, 1, vgx4], { z1.b - z4.b }, z14.b
+// CHECK-ENCODING: [0x39,0x14,0x3e,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c13e1439 <unknown>
+
+sudot   za.s[w10, 0, vgx4], {z19.b - z22.b}, z4.b  // 11000001-00110100-01010110-01111000
+// CHECK-INST: sudot   za.s[w10, 0, vgx4], { z19.b - z22.b }, z4.b
+// CHECK-ENCODING: [0x78,0x56,0x34,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1345678 <unknown>
+
+sudot   za.s[w10, 0], {z19.b - z22.b}, z4.b  // 11000001-00110100-01010110-01111000
+// CHECK-INST: sudot   za.s[w10, 0, vgx4], { z19.b - z22.b }, z4.b
+// CHECK-ENCODING: [0x78,0x56,0x34,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1345678 <unknown>
+
+sudot   za.s[w8, 0, vgx4], {z12.b - z15.b}, z2.b  // 11000001-00110010-00010101-10011000
+// CHECK-INST: sudot   za.s[w8, 0, vgx4], { z12.b - z15.b }, z2.b
+// CHECK-ENCODING: [0x98,0x15,0x32,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1321598 <unknown>
+
+sudot   za.s[w8, 0], {z12.b - z15.b}, z2.b  // 11000001-00110010-00010101-10011000
+// CHECK-INST: sudot   za.s[w8, 0, vgx4], { z12.b - z15.b }, z2.b
+// CHECK-ENCODING: [0x98,0x15,0x32,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1321598 <unknown>
+
+sudot   za.s[w10, 1, vgx4], {z1.b - z4.b}, z10.b  // 11000001-00111010-01010100-00111001
+// CHECK-INST: sudot   za.s[w10, 1, vgx4], { z1.b - z4.b }, z10.b
+// CHECK-ENCODING: [0x39,0x54,0x3a,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c13a5439 <unknown>
+
+sudot   za.s[w10, 1], {z1.b - z4.b}, z10.b  // 11000001-00111010-01010100-00111001
+// CHECK-INST: sudot   za.s[w10, 1, vgx4], { z1.b - z4.b }, z10.b
+// CHECK-ENCODING: [0x39,0x54,0x3a,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c13a5439 <unknown>
+
+sudot   za.s[w8, 5, vgx4], {z22.b - z25.b}, z14.b  // 11000001-00111110-00010110-11011101
+// CHECK-INST: sudot   za.s[w8, 5, vgx4], { z22.b - z25.b }, z14.b
+// CHECK-ENCODING: [0xdd,0x16,0x3e,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c13e16dd <unknown>
+
+sudot   za.s[w8, 5], {z22.b - z25.b}, z14.b  // 11000001-00111110-00010110-11011101
+// CHECK-INST: sudot   za.s[w8, 5, vgx4], { z22.b - z25.b }, z14.b
+// CHECK-ENCODING: [0xdd,0x16,0x3e,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c13e16dd <unknown>
+
+sudot   za.s[w11, 2, vgx4], {z9.b - z12.b}, z1.b  // 11000001-00110001-01110101-00111010
+// CHECK-INST: sudot   za.s[w11, 2, vgx4], { z9.b - z12.b }, z1.b
+// CHECK-ENCODING: [0x3a,0x75,0x31,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c131753a <unknown>
+
+sudot   za.s[w11, 2], {z9.b - z12.b}, z1.b  // 11000001-00110001-01110101-00111010
+// CHECK-INST: sudot   za.s[w11, 2, vgx4], { z9.b - z12.b }, z1.b
+// CHECK-ENCODING: [0x3a,0x75,0x31,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c131753a <unknown>
+
+sudot   za.s[w9, 7, vgx4], {z12.b - z15.b}, z11.b  // 11000001-00111011-00110101-10011111
+// CHECK-INST: sudot   za.s[w9, 7, vgx4], { z12.b - z15.b }, z11.b
+// CHECK-ENCODING: [0x9f,0x35,0x3b,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c13b359f <unknown>
+
+sudot   za.s[w9, 7], {z12.b - z15.b}, z11.b  // 11000001-00111011-00110101-10011111
+// CHECK-INST: sudot   za.s[w9, 7, vgx4], { z12.b - z15.b }, z11.b
+// CHECK-ENCODING: [0x9f,0x35,0x3b,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c13b359f <unknown>
+
 
 sudot   za.s[w8, 0, vgx4], {z0.b - z3.b}, z0.b[0]  // 11000001-01010000-10010000-00111000
 // CHECK-INST: sudot   za.s[w8, 0, vgx4], { z0.b - z3.b }, z0.b[0]
index 3132a4c..6246b52 100644 (file)
 // RUN:        | FileCheck %s --check-prefixes=CHECK-ENCODING,CHECK-INST
 
 
+udot    za.s[w8, 0, vgx2], {z0.h, z1.h}, z0.h  // 11000001-01100000-00010100-00011000
+// CHECK-INST: udot    za.s[w8, 0, vgx2], { z0.h, z1.h }, z0.h
+// CHECK-ENCODING: [0x18,0x14,0x60,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1601418 <unknown>
+
+udot    za.s[w8, 0], {z0.h, z1.h}, z0.h  // 11000001-01100000-00010100-00011000
+// CHECK-INST: udot    za.s[w8, 0, vgx2], { z0.h, z1.h }, z0.h
+// CHECK-ENCODING: [0x18,0x14,0x60,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1601418 <unknown>
+
+udot    za.s[w10, 5, vgx2], {z10.h, z11.h}, z5.h  // 11000001-01100101-01010101-01011101
+// CHECK-INST: udot    za.s[w10, 5, vgx2], { z10.h, z11.h }, z5.h
+// CHECK-ENCODING: [0x5d,0x55,0x65,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c165555d <unknown>
+
+udot    za.s[w10, 5], {z10.h, z11.h}, z5.h  // 11000001-01100101-01010101-01011101
+// CHECK-INST: udot    za.s[w10, 5, vgx2], { z10.h, z11.h }, z5.h
+// CHECK-ENCODING: [0x5d,0x55,0x65,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c165555d <unknown>
+
+udot    za.s[w11, 7, vgx2], {z13.h, z14.h}, z8.h  // 11000001-01101000-01110101-10111111
+// CHECK-INST: udot    za.s[w11, 7, vgx2], { z13.h, z14.h }, z8.h
+// CHECK-ENCODING: [0xbf,0x75,0x68,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c16875bf <unknown>
+
+udot    za.s[w11, 7], {z13.h, z14.h}, z8.h  // 11000001-01101000-01110101-10111111
+// CHECK-INST: udot    za.s[w11, 7, vgx2], { z13.h, z14.h }, z8.h
+// CHECK-ENCODING: [0xbf,0x75,0x68,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c16875bf <unknown>
+
+udot    za.s[w11, 7, vgx2], {z31.h, z0.h}, z15.h  // 11000001-01101111-01110111-11111111
+// CHECK-INST: udot    za.s[w11, 7, vgx2], { z31.h, z0.h }, z15.h
+// CHECK-ENCODING: [0xff,0x77,0x6f,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c16f77ff <unknown>
+
+udot    za.s[w11, 7], {z31.h, z0.h}, z15.h  // 11000001-01101111-01110111-11111111
+// CHECK-INST: udot    za.s[w11, 7, vgx2], { z31.h, z0.h }, z15.h
+// CHECK-ENCODING: [0xff,0x77,0x6f,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c16f77ff <unknown>
+
+udot    za.s[w8, 5, vgx2], {z17.h, z18.h}, z0.h  // 11000001-01100000-00010110-00111101
+// CHECK-INST: udot    za.s[w8, 5, vgx2], { z17.h, z18.h }, z0.h
+// CHECK-ENCODING: [0x3d,0x16,0x60,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c160163d <unknown>
+
+udot    za.s[w8, 5], {z17.h, z18.h}, z0.h  // 11000001-01100000-00010110-00111101
+// CHECK-INST: udot    za.s[w8, 5, vgx2], { z17.h, z18.h }, z0.h
+// CHECK-ENCODING: [0x3d,0x16,0x60,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c160163d <unknown>
+
+udot    za.s[w8, 1, vgx2], {z1.h, z2.h}, z14.h  // 11000001-01101110-00010100-00111001
+// CHECK-INST: udot    za.s[w8, 1, vgx2], { z1.h, z2.h }, z14.h
+// CHECK-ENCODING: [0x39,0x14,0x6e,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c16e1439 <unknown>
+
+udot    za.s[w8, 1], {z1.h, z2.h}, z14.h  // 11000001-01101110-00010100-00111001
+// CHECK-INST: udot    za.s[w8, 1, vgx2], { z1.h, z2.h }, z14.h
+// CHECK-ENCODING: [0x39,0x14,0x6e,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c16e1439 <unknown>
+
+udot    za.s[w10, 0, vgx2], {z19.h, z20.h}, z4.h  // 11000001-01100100-01010110-01111000
+// CHECK-INST: udot    za.s[w10, 0, vgx2], { z19.h, z20.h }, z4.h
+// CHECK-ENCODING: [0x78,0x56,0x64,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1645678 <unknown>
+
+udot    za.s[w10, 0], {z19.h, z20.h}, z4.h  // 11000001-01100100-01010110-01111000
+// CHECK-INST: udot    za.s[w10, 0, vgx2], { z19.h, z20.h }, z4.h
+// CHECK-ENCODING: [0x78,0x56,0x64,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1645678 <unknown>
+
+udot    za.s[w8, 0, vgx2], {z12.h, z13.h}, z2.h  // 11000001-01100010-00010101-10011000
+// CHECK-INST: udot    za.s[w8, 0, vgx2], { z12.h, z13.h }, z2.h
+// CHECK-ENCODING: [0x98,0x15,0x62,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1621598 <unknown>
+
+udot    za.s[w8, 0], {z12.h, z13.h}, z2.h  // 11000001-01100010-00010101-10011000
+// CHECK-INST: udot    za.s[w8, 0, vgx2], { z12.h, z13.h }, z2.h
+// CHECK-ENCODING: [0x98,0x15,0x62,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1621598 <unknown>
+
+udot    za.s[w10, 1, vgx2], {z1.h, z2.h}, z10.h  // 11000001-01101010-01010100-00111001
+// CHECK-INST: udot    za.s[w10, 1, vgx2], { z1.h, z2.h }, z10.h
+// CHECK-ENCODING: [0x39,0x54,0x6a,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c16a5439 <unknown>
+
+udot    za.s[w10, 1], {z1.h, z2.h}, z10.h  // 11000001-01101010-01010100-00111001
+// CHECK-INST: udot    za.s[w10, 1, vgx2], { z1.h, z2.h }, z10.h
+// CHECK-ENCODING: [0x39,0x54,0x6a,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c16a5439 <unknown>
+
+udot    za.s[w8, 5, vgx2], {z22.h, z23.h}, z14.h  // 11000001-01101110-00010110-11011101
+// CHECK-INST: udot    za.s[w8, 5, vgx2], { z22.h, z23.h }, z14.h
+// CHECK-ENCODING: [0xdd,0x16,0x6e,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c16e16dd <unknown>
+
+udot    za.s[w8, 5], {z22.h, z23.h}, z14.h  // 11000001-01101110-00010110-11011101
+// CHECK-INST: udot    za.s[w8, 5, vgx2], { z22.h, z23.h }, z14.h
+// CHECK-ENCODING: [0xdd,0x16,0x6e,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c16e16dd <unknown>
+
+udot    za.s[w11, 2, vgx2], {z9.h, z10.h}, z1.h  // 11000001-01100001-01110101-00111010
+// CHECK-INST: udot    za.s[w11, 2, vgx2], { z9.h, z10.h }, z1.h
+// CHECK-ENCODING: [0x3a,0x75,0x61,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c161753a <unknown>
+
+udot    za.s[w11, 2], {z9.h, z10.h}, z1.h  // 11000001-01100001-01110101-00111010
+// CHECK-INST: udot    za.s[w11, 2, vgx2], { z9.h, z10.h }, z1.h
+// CHECK-ENCODING: [0x3a,0x75,0x61,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c161753a <unknown>
+
+udot    za.s[w9, 7, vgx2], {z12.h, z13.h}, z11.h  // 11000001-01101011-00110101-10011111
+// CHECK-INST: udot    za.s[w9, 7, vgx2], { z12.h, z13.h }, z11.h
+// CHECK-ENCODING: [0x9f,0x35,0x6b,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c16b359f <unknown>
+
+udot    za.s[w9, 7], {z12.h, z13.h}, z11.h  // 11000001-01101011-00110101-10011111
+// CHECK-INST: udot    za.s[w9, 7, vgx2], { z12.h, z13.h }, z11.h
+// CHECK-ENCODING: [0x9f,0x35,0x6b,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c16b359f <unknown>
+
+
 udot    za.s[w8, 0, vgx2], {z0.h, z1.h}, z0.h[0]  // 11000001-01010000-00010000-00010000
 // CHECK-INST: udot    za.s[w8, 0, vgx2], { z0.h, z1.h }, z0.h[0]
 // CHECK-ENCODING: [0x10,0x10,0x50,0xc1]
@@ -157,6 +302,151 @@ udot    za.s[w9, 7], {z12.h, z13.h}, z11.h[2]  // 11000001-01011011-00111001-100
 // CHECK-UNKNOWN: c15b3997 <unknown>
 
 
+udot    za.s[w8, 0, vgx2], {z0.h, z1.h}, {z0.h, z1.h}  // 11000001-11100000-00010100-00011000
+// CHECK-INST: udot    za.s[w8, 0, vgx2], { z0.h, z1.h }, { z0.h, z1.h }
+// CHECK-ENCODING: [0x18,0x14,0xe0,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1e01418 <unknown>
+
+udot    za.s[w8, 0], {z0.h, z1.h}, {z0.h, z1.h}  // 11000001-11100000-00010100-00011000
+// CHECK-INST: udot    za.s[w8, 0, vgx2], { z0.h, z1.h }, { z0.h, z1.h }
+// CHECK-ENCODING: [0x18,0x14,0xe0,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1e01418 <unknown>
+
+udot    za.s[w10, 5, vgx2], {z10.h, z11.h}, {z20.h, z21.h}  // 11000001-11110100-01010101-01011101
+// CHECK-INST: udot    za.s[w10, 5, vgx2], { z10.h, z11.h }, { z20.h, z21.h }
+// CHECK-ENCODING: [0x5d,0x55,0xf4,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1f4555d <unknown>
+
+udot    za.s[w10, 5], {z10.h, z11.h}, {z20.h, z21.h}  // 11000001-11110100-01010101-01011101
+// CHECK-INST: udot    za.s[w10, 5, vgx2], { z10.h, z11.h }, { z20.h, z21.h }
+// CHECK-ENCODING: [0x5d,0x55,0xf4,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1f4555d <unknown>
+
+udot    za.s[w11, 7, vgx2], {z12.h, z13.h}, {z8.h, z9.h}  // 11000001-11101000-01110101-10011111
+// CHECK-INST: udot    za.s[w11, 7, vgx2], { z12.h, z13.h }, { z8.h, z9.h }
+// CHECK-ENCODING: [0x9f,0x75,0xe8,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1e8759f <unknown>
+
+udot    za.s[w11, 7], {z12.h, z13.h}, {z8.h, z9.h}  // 11000001-11101000-01110101-10011111
+// CHECK-INST: udot    za.s[w11, 7, vgx2], { z12.h, z13.h }, { z8.h, z9.h }
+// CHECK-ENCODING: [0x9f,0x75,0xe8,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1e8759f <unknown>
+
+udot    za.s[w11, 7, vgx2], {z30.h, z31.h}, {z30.h, z31.h}  // 11000001-11111110-01110111-11011111
+// CHECK-INST: udot    za.s[w11, 7, vgx2], { z30.h, z31.h }, { z30.h, z31.h }
+// CHECK-ENCODING: [0xdf,0x77,0xfe,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1fe77df <unknown>
+
+udot    za.s[w11, 7], {z30.h, z31.h}, {z30.h, z31.h}  // 11000001-11111110-01110111-11011111
+// CHECK-INST: udot    za.s[w11, 7, vgx2], { z30.h, z31.h }, { z30.h, z31.h }
+// CHECK-ENCODING: [0xdf,0x77,0xfe,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1fe77df <unknown>
+
+udot    za.s[w8, 5, vgx2], {z16.h, z17.h}, {z16.h, z17.h}  // 11000001-11110000-00010110-00011101
+// CHECK-INST: udot    za.s[w8, 5, vgx2], { z16.h, z17.h }, { z16.h, z17.h }
+// CHECK-ENCODING: [0x1d,0x16,0xf0,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1f0161d <unknown>
+
+udot    za.s[w8, 5], {z16.h, z17.h}, {z16.h, z17.h}  // 11000001-11110000-00010110-00011101
+// CHECK-INST: udot    za.s[w8, 5, vgx2], { z16.h, z17.h }, { z16.h, z17.h }
+// CHECK-ENCODING: [0x1d,0x16,0xf0,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1f0161d <unknown>
+
+udot    za.s[w8, 1, vgx2], {z0.h, z1.h}, {z30.h, z31.h}  // 11000001-11111110-00010100-00011001
+// CHECK-INST: udot    za.s[w8, 1, vgx2], { z0.h, z1.h }, { z30.h, z31.h }
+// CHECK-ENCODING: [0x19,0x14,0xfe,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1fe1419 <unknown>
+
+udot    za.s[w8, 1], {z0.h, z1.h}, {z30.h, z31.h}  // 11000001-11111110-00010100-00011001
+// CHECK-INST: udot    za.s[w8, 1, vgx2], { z0.h, z1.h }, { z30.h, z31.h }
+// CHECK-ENCODING: [0x19,0x14,0xfe,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1fe1419 <unknown>
+
+udot    za.s[w10, 0, vgx2], {z18.h, z19.h}, {z20.h, z21.h}  // 11000001-11110100-01010110-01011000
+// CHECK-INST: udot    za.s[w10, 0, vgx2], { z18.h, z19.h }, { z20.h, z21.h }
+// CHECK-ENCODING: [0x58,0x56,0xf4,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1f45658 <unknown>
+
+udot    za.s[w10, 0], {z18.h, z19.h}, {z20.h, z21.h}  // 11000001-11110100-01010110-01011000
+// CHECK-INST: udot    za.s[w10, 0, vgx2], { z18.h, z19.h }, { z20.h, z21.h }
+// CHECK-ENCODING: [0x58,0x56,0xf4,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1f45658 <unknown>
+
+udot    za.s[w8, 0, vgx2], {z12.h, z13.h}, {z2.h, z3.h}  // 11000001-11100010-00010101-10011000
+// CHECK-INST: udot    za.s[w8, 0, vgx2], { z12.h, z13.h }, { z2.h, z3.h }
+// CHECK-ENCODING: [0x98,0x15,0xe2,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1e21598 <unknown>
+
+udot    za.s[w8, 0], {z12.h, z13.h}, {z2.h, z3.h}  // 11000001-11100010-00010101-10011000
+// CHECK-INST: udot    za.s[w8, 0, vgx2], { z12.h, z13.h }, { z2.h, z3.h }
+// CHECK-ENCODING: [0x98,0x15,0xe2,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1e21598 <unknown>
+
+udot    za.s[w10, 1, vgx2], {z0.h, z1.h}, {z26.h, z27.h}  // 11000001-11111010-01010100-00011001
+// CHECK-INST: udot    za.s[w10, 1, vgx2], { z0.h, z1.h }, { z26.h, z27.h }
+// CHECK-ENCODING: [0x19,0x54,0xfa,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1fa5419 <unknown>
+
+udot    za.s[w10, 1], {z0.h, z1.h}, {z26.h, z27.h}  // 11000001-11111010-01010100-00011001
+// CHECK-INST: udot    za.s[w10, 1, vgx2], { z0.h, z1.h }, { z26.h, z27.h }
+// CHECK-ENCODING: [0x19,0x54,0xfa,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1fa5419 <unknown>
+
+udot    za.s[w8, 5, vgx2], {z22.h, z23.h}, {z30.h, z31.h}  // 11000001-11111110-00010110-11011101
+// CHECK-INST: udot    za.s[w8, 5, vgx2], { z22.h, z23.h }, { z30.h, z31.h }
+// CHECK-ENCODING: [0xdd,0x16,0xfe,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1fe16dd <unknown>
+
+udot    za.s[w8, 5], {z22.h, z23.h}, {z30.h, z31.h}  // 11000001-11111110-00010110-11011101
+// CHECK-INST: udot    za.s[w8, 5, vgx2], { z22.h, z23.h }, { z30.h, z31.h }
+// CHECK-ENCODING: [0xdd,0x16,0xfe,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1fe16dd <unknown>
+
+udot    za.s[w11, 2, vgx2], {z8.h, z9.h}, {z0.h, z1.h}  // 11000001-11100000-01110101-00011010
+// CHECK-INST: udot    za.s[w11, 2, vgx2], { z8.h, z9.h }, { z0.h, z1.h }
+// CHECK-ENCODING: [0x1a,0x75,0xe0,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1e0751a <unknown>
+
+udot    za.s[w11, 2], {z8.h, z9.h}, {z0.h, z1.h}  // 11000001-11100000-01110101-00011010
+// CHECK-INST: udot    za.s[w11, 2, vgx2], { z8.h, z9.h }, { z0.h, z1.h }
+// CHECK-ENCODING: [0x1a,0x75,0xe0,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1e0751a <unknown>
+
+udot    za.s[w9, 7, vgx2], {z12.h, z13.h}, {z10.h, z11.h}  // 11000001-11101010-00110101-10011111
+// CHECK-INST: udot    za.s[w9, 7, vgx2], { z12.h, z13.h }, { z10.h, z11.h }
+// CHECK-ENCODING: [0x9f,0x35,0xea,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1ea359f <unknown>
+
+udot    za.s[w9, 7], {z12.h, z13.h}, {z10.h, z11.h}  // 11000001-11101010-00110101-10011111
+// CHECK-INST: udot    za.s[w9, 7, vgx2], { z12.h, z13.h }, { z10.h, z11.h }
+// CHECK-ENCODING: [0x9f,0x35,0xea,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1ea359f <unknown>
+
+
 udot    za.s[w8, 0, vgx2], {z0.b, z1.b}, z0.b[0]  // 11000001-01010000-00010000-00110000
 // CHECK-INST: udot    za.s[w8, 0, vgx2], { z0.b, z1.b }, z0.b[0]
 // CHECK-ENCODING: [0x30,0x10,0x50,0xc1]
@@ -447,6 +737,151 @@ udot    za.d[w9, 7], {z12.h, z13.h}, z11.h[0]  // 11000001-11011011-00100001-100
 // CHECK-UNKNOWN: c1db219f <unknown>
 
 
+udot    za.s[w8, 0, vgx4], {z0.h - z3.h}, z0.h  // 11000001-01110000-00010100-00011000
+// CHECK-INST: udot    za.s[w8, 0, vgx4], { z0.h - z3.h }, z0.h
+// CHECK-ENCODING: [0x18,0x14,0x70,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1701418 <unknown>
+
+udot    za.s[w8, 0], {z0.h - z3.h}, z0.h  // 11000001-01110000-00010100-00011000
+// CHECK-INST: udot    za.s[w8, 0, vgx4], { z0.h - z3.h }, z0.h
+// CHECK-ENCODING: [0x18,0x14,0x70,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1701418 <unknown>
+
+udot    za.s[w10, 5, vgx4], {z10.h - z13.h}, z5.h  // 11000001-01110101-01010101-01011101
+// CHECK-INST: udot    za.s[w10, 5, vgx4], { z10.h - z13.h }, z5.h
+// CHECK-ENCODING: [0x5d,0x55,0x75,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c175555d <unknown>
+
+udot    za.s[w10, 5], {z10.h - z13.h}, z5.h  // 11000001-01110101-01010101-01011101
+// CHECK-INST: udot    za.s[w10, 5, vgx4], { z10.h - z13.h }, z5.h
+// CHECK-ENCODING: [0x5d,0x55,0x75,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c175555d <unknown>
+
+udot    za.s[w11, 7, vgx4], {z13.h - z16.h}, z8.h  // 11000001-01111000-01110101-10111111
+// CHECK-INST: udot    za.s[w11, 7, vgx4], { z13.h - z16.h }, z8.h
+// CHECK-ENCODING: [0xbf,0x75,0x78,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c17875bf <unknown>
+
+udot    za.s[w11, 7], {z13.h - z16.h}, z8.h  // 11000001-01111000-01110101-10111111
+// CHECK-INST: udot    za.s[w11, 7, vgx4], { z13.h - z16.h }, z8.h
+// CHECK-ENCODING: [0xbf,0x75,0x78,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c17875bf <unknown>
+
+udot    za.s[w11, 7, vgx4], {z31.h - z2.h}, z15.h  // 11000001-01111111-01110111-11111111
+// CHECK-INST: udot    za.s[w11, 7, vgx4], { z31.h, z0.h, z1.h, z2.h }, z15.h
+// CHECK-ENCODING: [0xff,0x77,0x7f,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c17f77ff <unknown>
+
+udot    za.s[w11, 7], {z31.h - z2.h}, z15.h  // 11000001-01111111-01110111-11111111
+// CHECK-INST: udot    za.s[w11, 7, vgx4], { z31.h, z0.h, z1.h, z2.h }, z15.h
+// CHECK-ENCODING: [0xff,0x77,0x7f,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c17f77ff <unknown>
+
+udot    za.s[w8, 5, vgx4], {z17.h - z20.h}, z0.h  // 11000001-01110000-00010110-00111101
+// CHECK-INST: udot    za.s[w8, 5, vgx4], { z17.h - z20.h }, z0.h
+// CHECK-ENCODING: [0x3d,0x16,0x70,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c170163d <unknown>
+
+udot    za.s[w8, 5], {z17.h - z20.h}, z0.h  // 11000001-01110000-00010110-00111101
+// CHECK-INST: udot    za.s[w8, 5, vgx4], { z17.h - z20.h }, z0.h
+// CHECK-ENCODING: [0x3d,0x16,0x70,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c170163d <unknown>
+
+udot    za.s[w8, 1, vgx4], {z1.h - z4.h}, z14.h  // 11000001-01111110-00010100-00111001
+// CHECK-INST: udot    za.s[w8, 1, vgx4], { z1.h - z4.h }, z14.h
+// CHECK-ENCODING: [0x39,0x14,0x7e,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c17e1439 <unknown>
+
+udot    za.s[w8, 1], {z1.h - z4.h}, z14.h  // 11000001-01111110-00010100-00111001
+// CHECK-INST: udot    za.s[w8, 1, vgx4], { z1.h - z4.h }, z14.h
+// CHECK-ENCODING: [0x39,0x14,0x7e,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c17e1439 <unknown>
+
+udot    za.s[w10, 0, vgx4], {z19.h - z22.h}, z4.h  // 11000001-01110100-01010110-01111000
+// CHECK-INST: udot    za.s[w10, 0, vgx4], { z19.h - z22.h }, z4.h
+// CHECK-ENCODING: [0x78,0x56,0x74,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1745678 <unknown>
+
+udot    za.s[w10, 0], {z19.h - z22.h}, z4.h  // 11000001-01110100-01010110-01111000
+// CHECK-INST: udot    za.s[w10, 0, vgx4], { z19.h - z22.h }, z4.h
+// CHECK-ENCODING: [0x78,0x56,0x74,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1745678 <unknown>
+
+udot    za.s[w8, 0, vgx4], {z12.h - z15.h}, z2.h  // 11000001-01110010-00010101-10011000
+// CHECK-INST: udot    za.s[w8, 0, vgx4], { z12.h - z15.h }, z2.h
+// CHECK-ENCODING: [0x98,0x15,0x72,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1721598 <unknown>
+
+udot    za.s[w8, 0], {z12.h - z15.h}, z2.h  // 11000001-01110010-00010101-10011000
+// CHECK-INST: udot    za.s[w8, 0, vgx4], { z12.h - z15.h }, z2.h
+// CHECK-ENCODING: [0x98,0x15,0x72,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1721598 <unknown>
+
+udot    za.s[w10, 1, vgx4], {z1.h - z4.h}, z10.h  // 11000001-01111010-01010100-00111001
+// CHECK-INST: udot    za.s[w10, 1, vgx4], { z1.h - z4.h }, z10.h
+// CHECK-ENCODING: [0x39,0x54,0x7a,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c17a5439 <unknown>
+
+udot    za.s[w10, 1], {z1.h - z4.h}, z10.h  // 11000001-01111010-01010100-00111001
+// CHECK-INST: udot    za.s[w10, 1, vgx4], { z1.h - z4.h }, z10.h
+// CHECK-ENCODING: [0x39,0x54,0x7a,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c17a5439 <unknown>
+
+udot    za.s[w8, 5, vgx4], {z22.h - z25.h}, z14.h  // 11000001-01111110-00010110-11011101
+// CHECK-INST: udot    za.s[w8, 5, vgx4], { z22.h - z25.h }, z14.h
+// CHECK-ENCODING: [0xdd,0x16,0x7e,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c17e16dd <unknown>
+
+udot    za.s[w8, 5], {z22.h - z25.h}, z14.h  // 11000001-01111110-00010110-11011101
+// CHECK-INST: udot    za.s[w8, 5, vgx4], { z22.h - z25.h }, z14.h
+// CHECK-ENCODING: [0xdd,0x16,0x7e,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c17e16dd <unknown>
+
+udot    za.s[w11, 2, vgx4], {z9.h - z12.h}, z1.h  // 11000001-01110001-01110101-00111010
+// CHECK-INST: udot    za.s[w11, 2, vgx4], { z9.h - z12.h }, z1.h
+// CHECK-ENCODING: [0x3a,0x75,0x71,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c171753a <unknown>
+
+udot    za.s[w11, 2], {z9.h - z12.h}, z1.h  // 11000001-01110001-01110101-00111010
+// CHECK-INST: udot    za.s[w11, 2, vgx4], { z9.h - z12.h }, z1.h
+// CHECK-ENCODING: [0x3a,0x75,0x71,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c171753a <unknown>
+
+udot    za.s[w9, 7, vgx4], {z12.h - z15.h}, z11.h  // 11000001-01111011-00110101-10011111
+// CHECK-INST: udot    za.s[w9, 7, vgx4], { z12.h - z15.h }, z11.h
+// CHECK-ENCODING: [0x9f,0x35,0x7b,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c17b359f <unknown>
+
+udot    za.s[w9, 7], {z12.h - z15.h}, z11.h  // 11000001-01111011-00110101-10011111
+// CHECK-INST: udot    za.s[w9, 7, vgx4], { z12.h - z15.h }, z11.h
+// CHECK-ENCODING: [0x9f,0x35,0x7b,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c17b359f <unknown>
+
+
 udot    za.s[w8, 0, vgx4], {z0.h - z3.h}, z0.h[0]  // 11000001-01010000-10010000-00010000
 // CHECK-INST: udot    za.s[w8, 0, vgx4], { z0.h - z3.h }, z0.h[0]
 // CHECK-ENCODING: [0x10,0x90,0x50,0xc1]
@@ -592,6 +1027,151 @@ udot    za.s[w9, 7], {z12.h - z15.h}, z11.h[2]  // 11000001-01011011-10111001-10
 // CHECK-UNKNOWN: c15bb997 <unknown>
 
 
+udot    za.s[w8, 0, vgx4], {z0.h - z3.h}, {z0.h - z3.h}  // 11000001-11100001-00010100-00011000
+// CHECK-INST: udot    za.s[w8, 0, vgx4], { z0.h - z3.h }, { z0.h - z3.h }
+// CHECK-ENCODING: [0x18,0x14,0xe1,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1e11418 <unknown>
+
+udot    za.s[w8, 0], {z0.h - z3.h}, {z0.h - z3.h}  // 11000001-11100001-00010100-00011000
+// CHECK-INST: udot    za.s[w8, 0, vgx4], { z0.h - z3.h }, { z0.h - z3.h }
+// CHECK-ENCODING: [0x18,0x14,0xe1,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1e11418 <unknown>
+
+udot    za.s[w10, 5, vgx4], {z8.h - z11.h}, {z20.h - z23.h}  // 11000001-11110101-01010101-00011101
+// CHECK-INST: udot    za.s[w10, 5, vgx4], { z8.h - z11.h }, { z20.h - z23.h }
+// CHECK-ENCODING: [0x1d,0x55,0xf5,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1f5551d <unknown>
+
+udot    za.s[w10, 5], {z8.h - z11.h}, {z20.h - z23.h}  // 11000001-11110101-01010101-00011101
+// CHECK-INST: udot    za.s[w10, 5, vgx4], { z8.h - z11.h }, { z20.h - z23.h }
+// CHECK-ENCODING: [0x1d,0x55,0xf5,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1f5551d <unknown>
+
+udot    za.s[w11, 7, vgx4], {z12.h - z15.h}, {z8.h - z11.h}  // 11000001-11101001-01110101-10011111
+// CHECK-INST: udot    za.s[w11, 7, vgx4], { z12.h - z15.h }, { z8.h - z11.h }
+// CHECK-ENCODING: [0x9f,0x75,0xe9,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1e9759f <unknown>
+
+udot    za.s[w11, 7], {z12.h - z15.h}, {z8.h - z11.h}  // 11000001-11101001-01110101-10011111
+// CHECK-INST: udot    za.s[w11, 7, vgx4], { z12.h - z15.h }, { z8.h - z11.h }
+// CHECK-ENCODING: [0x9f,0x75,0xe9,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1e9759f <unknown>
+
+udot    za.s[w11, 7, vgx4], {z28.h - z31.h}, {z28.h - z31.h}  // 11000001-11111101-01110111-10011111
+// CHECK-INST: udot    za.s[w11, 7, vgx4], { z28.h - z31.h }, { z28.h - z31.h }
+// CHECK-ENCODING: [0x9f,0x77,0xfd,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1fd779f <unknown>
+
+udot    za.s[w11, 7], {z28.h - z31.h}, {z28.h - z31.h}  // 11000001-11111101-01110111-10011111
+// CHECK-INST: udot    za.s[w11, 7, vgx4], { z28.h - z31.h }, { z28.h - z31.h }
+// CHECK-ENCODING: [0x9f,0x77,0xfd,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1fd779f <unknown>
+
+udot    za.s[w8, 5, vgx4], {z16.h - z19.h}, {z16.h - z19.h}  // 11000001-11110001-00010110-00011101
+// CHECK-INST: udot    za.s[w8, 5, vgx4], { z16.h - z19.h }, { z16.h - z19.h }
+// CHECK-ENCODING: [0x1d,0x16,0xf1,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1f1161d <unknown>
+
+udot    za.s[w8, 5], {z16.h - z19.h}, {z16.h - z19.h}  // 11000001-11110001-00010110-00011101
+// CHECK-INST: udot    za.s[w8, 5, vgx4], { z16.h - z19.h }, { z16.h - z19.h }
+// CHECK-ENCODING: [0x1d,0x16,0xf1,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1f1161d <unknown>
+
+udot    za.s[w8, 1, vgx4], {z0.h - z3.h}, {z28.h - z31.h}  // 11000001-11111101-00010100-00011001
+// CHECK-INST: udot    za.s[w8, 1, vgx4], { z0.h - z3.h }, { z28.h - z31.h }
+// CHECK-ENCODING: [0x19,0x14,0xfd,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1fd1419 <unknown>
+
+udot    za.s[w8, 1], {z0.h - z3.h}, {z28.h - z31.h}  // 11000001-11111101-00010100-00011001
+// CHECK-INST: udot    za.s[w8, 1, vgx4], { z0.h - z3.h }, { z28.h - z31.h }
+// CHECK-ENCODING: [0x19,0x14,0xfd,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1fd1419 <unknown>
+
+udot    za.s[w10, 0, vgx4], {z16.h - z19.h}, {z20.h - z23.h}  // 11000001-11110101-01010110-00011000
+// CHECK-INST: udot    za.s[w10, 0, vgx4], { z16.h - z19.h }, { z20.h - z23.h }
+// CHECK-ENCODING: [0x18,0x56,0xf5,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1f55618 <unknown>
+
+udot    za.s[w10, 0], {z16.h - z19.h}, {z20.h - z23.h}  // 11000001-11110101-01010110-00011000
+// CHECK-INST: udot    za.s[w10, 0, vgx4], { z16.h - z19.h }, { z20.h - z23.h }
+// CHECK-ENCODING: [0x18,0x56,0xf5,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1f55618 <unknown>
+
+udot    za.s[w8, 0, vgx4], {z12.h - z15.h}, {z0.h - z3.h}  // 11000001-11100001-00010101-10011000
+// CHECK-INST: udot    za.s[w8, 0, vgx4], { z12.h - z15.h }, { z0.h - z3.h }
+// CHECK-ENCODING: [0x98,0x15,0xe1,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1e11598 <unknown>
+
+udot    za.s[w8, 0], {z12.h - z15.h}, {z0.h - z3.h}  // 11000001-11100001-00010101-10011000
+// CHECK-INST: udot    za.s[w8, 0, vgx4], { z12.h - z15.h }, { z0.h - z3.h }
+// CHECK-ENCODING: [0x98,0x15,0xe1,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1e11598 <unknown>
+
+udot    za.s[w10, 1, vgx4], {z0.h - z3.h}, {z24.h - z27.h}  // 11000001-11111001-01010100-00011001
+// CHECK-INST: udot    za.s[w10, 1, vgx4], { z0.h - z3.h }, { z24.h - z27.h }
+// CHECK-ENCODING: [0x19,0x54,0xf9,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1f95419 <unknown>
+
+udot    za.s[w10, 1], {z0.h - z3.h}, {z24.h - z27.h}  // 11000001-11111001-01010100-00011001
+// CHECK-INST: udot    za.s[w10, 1, vgx4], { z0.h - z3.h }, { z24.h - z27.h }
+// CHECK-ENCODING: [0x19,0x54,0xf9,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1f95419 <unknown>
+
+udot    za.s[w8, 5, vgx4], {z20.h - z23.h}, {z28.h - z31.h}  // 11000001-11111101-00010110-10011101
+// CHECK-INST: udot    za.s[w8, 5, vgx4], { z20.h - z23.h }, { z28.h - z31.h }
+// CHECK-ENCODING: [0x9d,0x16,0xfd,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1fd169d <unknown>
+
+udot    za.s[w8, 5], {z20.h - z23.h}, {z28.h - z31.h}  // 11000001-11111101-00010110-10011101
+// CHECK-INST: udot    za.s[w8, 5, vgx4], { z20.h - z23.h }, { z28.h - z31.h }
+// CHECK-ENCODING: [0x9d,0x16,0xfd,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1fd169d <unknown>
+
+udot    za.s[w11, 2, vgx4], {z8.h - z11.h}, {z0.h - z3.h}  // 11000001-11100001-01110101-00011010
+// CHECK-INST: udot    za.s[w11, 2, vgx4], { z8.h - z11.h }, { z0.h - z3.h }
+// CHECK-ENCODING: [0x1a,0x75,0xe1,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1e1751a <unknown>
+
+udot    za.s[w11, 2], {z8.h - z11.h}, {z0.h - z3.h}  // 11000001-11100001-01110101-00011010
+// CHECK-INST: udot    za.s[w11, 2, vgx4], { z8.h - z11.h }, { z0.h - z3.h }
+// CHECK-ENCODING: [0x1a,0x75,0xe1,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1e1751a <unknown>
+
+udot    za.s[w9, 7, vgx4], {z12.h - z15.h}, {z8.h - z11.h}  // 11000001-11101001-00110101-10011111
+// CHECK-INST: udot    za.s[w9, 7, vgx4], { z12.h - z15.h }, { z8.h - z11.h }
+// CHECK-ENCODING: [0x9f,0x35,0xe9,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1e9359f <unknown>
+
+udot    za.s[w9, 7], {z12.h - z15.h}, {z8.h - z11.h}  // 11000001-11101001-00110101-10011111
+// CHECK-INST: udot    za.s[w9, 7, vgx4], { z12.h - z15.h }, { z8.h - z11.h }
+// CHECK-ENCODING: [0x9f,0x35,0xe9,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1e9359f <unknown>
+
+
 udot    za.s[w8, 0, vgx4], {z0.b - z3.b}, z0.b[0]  // 11000001-01010000-10010000-00110000
 // CHECK-INST: udot    za.s[w8, 0, vgx4], { z0.b - z3.b }, z0.b[0]
 // CHECK-ENCODING: [0x30,0x90,0x50,0xc1]
index 2b05a70..9158b9d 100644 (file)
 // RUN:        | FileCheck %s --check-prefixes=CHECK-ENCODING,CHECK-INST
 
 
+usdot   za.s[w8, 0, vgx2], {z0.b, z1.b}, z0.b  // 11000001-00100000-00010100-00001000
+// CHECK-INST: usdot   za.s[w8, 0, vgx2], { z0.b, z1.b }, z0.b
+// CHECK-ENCODING: [0x08,0x14,0x20,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1201408 <unknown>
+
+usdot   za.s[w8, 0], {z0.b, z1.b}, z0.b  // 11000001-00100000-00010100-00001000
+// CHECK-INST: usdot   za.s[w8, 0, vgx2], { z0.b, z1.b }, z0.b
+// CHECK-ENCODING: [0x08,0x14,0x20,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1201408 <unknown>
+
+usdot   za.s[w10, 5, vgx2], {z10.b, z11.b}, z5.b  // 11000001-00100101-01010101-01001101
+// CHECK-INST: usdot   za.s[w10, 5, vgx2], { z10.b, z11.b }, z5.b
+// CHECK-ENCODING: [0x4d,0x55,0x25,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c125554d <unknown>
+
+usdot   za.s[w10, 5], {z10.b, z11.b}, z5.b  // 11000001-00100101-01010101-01001101
+// CHECK-INST: usdot   za.s[w10, 5, vgx2], { z10.b, z11.b }, z5.b
+// CHECK-ENCODING: [0x4d,0x55,0x25,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c125554d <unknown>
+
+usdot   za.s[w11, 7, vgx2], {z13.b, z14.b}, z8.b  // 11000001-00101000-01110101-10101111
+// CHECK-INST: usdot   za.s[w11, 7, vgx2], { z13.b, z14.b }, z8.b
+// CHECK-ENCODING: [0xaf,0x75,0x28,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c12875af <unknown>
+
+usdot   za.s[w11, 7], {z13.b, z14.b}, z8.b  // 11000001-00101000-01110101-10101111
+// CHECK-INST: usdot   za.s[w11, 7, vgx2], { z13.b, z14.b }, z8.b
+// CHECK-ENCODING: [0xaf,0x75,0x28,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c12875af <unknown>
+
+usdot   za.s[w11, 7, vgx2], {z31.b, z0.b}, z15.b  // 11000001-00101111-01110111-11101111
+// CHECK-INST: usdot   za.s[w11, 7, vgx2], { z31.b, z0.b }, z15.b
+// CHECK-ENCODING: [0xef,0x77,0x2f,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c12f77ef <unknown>
+
+usdot   za.s[w11, 7], {z31.b, z0.b}, z15.b  // 11000001-00101111-01110111-11101111
+// CHECK-INST: usdot   za.s[w11, 7, vgx2], { z31.b, z0.b }, z15.b
+// CHECK-ENCODING: [0xef,0x77,0x2f,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c12f77ef <unknown>
+
+usdot   za.s[w8, 5, vgx2], {z17.b, z18.b}, z0.b  // 11000001-00100000-00010110-00101101
+// CHECK-INST: usdot   za.s[w8, 5, vgx2], { z17.b, z18.b }, z0.b
+// CHECK-ENCODING: [0x2d,0x16,0x20,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c120162d <unknown>
+
+usdot   za.s[w8, 5], {z17.b, z18.b}, z0.b  // 11000001-00100000-00010110-00101101
+// CHECK-INST: usdot   za.s[w8, 5, vgx2], { z17.b, z18.b }, z0.b
+// CHECK-ENCODING: [0x2d,0x16,0x20,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c120162d <unknown>
+
+usdot   za.s[w8, 1, vgx2], {z1.b, z2.b}, z14.b  // 11000001-00101110-00010100-00101001
+// CHECK-INST: usdot   za.s[w8, 1, vgx2], { z1.b, z2.b }, z14.b
+// CHECK-ENCODING: [0x29,0x14,0x2e,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c12e1429 <unknown>
+
+usdot   za.s[w8, 1], {z1.b, z2.b}, z14.b  // 11000001-00101110-00010100-00101001
+// CHECK-INST: usdot   za.s[w8, 1, vgx2], { z1.b, z2.b }, z14.b
+// CHECK-ENCODING: [0x29,0x14,0x2e,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c12e1429 <unknown>
+
+usdot   za.s[w10, 0, vgx2], {z19.b, z20.b}, z4.b  // 11000001-00100100-01010110-01101000
+// CHECK-INST: usdot   za.s[w10, 0, vgx2], { z19.b, z20.b }, z4.b
+// CHECK-ENCODING: [0x68,0x56,0x24,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1245668 <unknown>
+
+usdot   za.s[w10, 0], {z19.b, z20.b}, z4.b  // 11000001-00100100-01010110-01101000
+// CHECK-INST: usdot   za.s[w10, 0, vgx2], { z19.b, z20.b }, z4.b
+// CHECK-ENCODING: [0x68,0x56,0x24,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1245668 <unknown>
+
+usdot   za.s[w8, 0, vgx2], {z12.b, z13.b}, z2.b  // 11000001-00100010-00010101-10001000
+// CHECK-INST: usdot   za.s[w8, 0, vgx2], { z12.b, z13.b }, z2.b
+// CHECK-ENCODING: [0x88,0x15,0x22,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1221588 <unknown>
+
+usdot   za.s[w8, 0], {z12.b, z13.b}, z2.b  // 11000001-00100010-00010101-10001000
+// CHECK-INST: usdot   za.s[w8, 0, vgx2], { z12.b, z13.b }, z2.b
+// CHECK-ENCODING: [0x88,0x15,0x22,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1221588 <unknown>
+
+usdot   za.s[w10, 1, vgx2], {z1.b, z2.b}, z10.b  // 11000001-00101010-01010100-00101001
+// CHECK-INST: usdot   za.s[w10, 1, vgx2], { z1.b, z2.b }, z10.b
+// CHECK-ENCODING: [0x29,0x54,0x2a,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c12a5429 <unknown>
+
+usdot   za.s[w10, 1], {z1.b, z2.b}, z10.b  // 11000001-00101010-01010100-00101001
+// CHECK-INST: usdot   za.s[w10, 1, vgx2], { z1.b, z2.b }, z10.b
+// CHECK-ENCODING: [0x29,0x54,0x2a,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c12a5429 <unknown>
+
+usdot   za.s[w8, 5, vgx2], {z22.b, z23.b}, z14.b  // 11000001-00101110-00010110-11001101
+// CHECK-INST: usdot   za.s[w8, 5, vgx2], { z22.b, z23.b }, z14.b
+// CHECK-ENCODING: [0xcd,0x16,0x2e,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c12e16cd <unknown>
+
+usdot   za.s[w8, 5], {z22.b, z23.b}, z14.b  // 11000001-00101110-00010110-11001101
+// CHECK-INST: usdot   za.s[w8, 5, vgx2], { z22.b, z23.b }, z14.b
+// CHECK-ENCODING: [0xcd,0x16,0x2e,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c12e16cd <unknown>
+
+usdot   za.s[w11, 2, vgx2], {z9.b, z10.b}, z1.b  // 11000001-00100001-01110101-00101010
+// CHECK-INST: usdot   za.s[w11, 2, vgx2], { z9.b, z10.b }, z1.b
+// CHECK-ENCODING: [0x2a,0x75,0x21,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c121752a <unknown>
+
+usdot   za.s[w11, 2], {z9.b, z10.b}, z1.b  // 11000001-00100001-01110101-00101010
+// CHECK-INST: usdot   za.s[w11, 2, vgx2], { z9.b, z10.b }, z1.b
+// CHECK-ENCODING: [0x2a,0x75,0x21,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c121752a <unknown>
+
+usdot   za.s[w9, 7, vgx2], {z12.b, z13.b}, z11.b  // 11000001-00101011-00110101-10001111
+// CHECK-INST: usdot   za.s[w9, 7, vgx2], { z12.b, z13.b }, z11.b
+// CHECK-ENCODING: [0x8f,0x35,0x2b,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c12b358f <unknown>
+
+usdot   za.s[w9, 7], {z12.b, z13.b}, z11.b  // 11000001-00101011-00110101-10001111
+// CHECK-INST: usdot   za.s[w9, 7, vgx2], { z12.b, z13.b }, z11.b
+// CHECK-ENCODING: [0x8f,0x35,0x2b,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c12b358f <unknown>
+
+
 usdot   za.s[w8, 0, vgx2], {z0.b, z1.b}, z0.b[0]  // 11000001-01010000-00010000-00101000
 // CHECK-INST: usdot   za.s[w8, 0, vgx2], { z0.b, z1.b }, z0.b[0]
 // CHECK-ENCODING: [0x28,0x10,0x50,0xc1]
@@ -157,6 +302,295 @@ usdot   za.s[w9, 7], {z12.b, z13.b}, z11.b[2]  // 11000001-01011011-00111001-101
 // CHECK-UNKNOWN: c15b39af <unknown>
 
 
+usdot   za.s[w8, 0, vgx2], {z0.b, z1.b}, {z0.b, z1.b}  // 11000001-10100000-00010100-00001000
+// CHECK-INST: usdot   za.s[w8, 0, vgx2], { z0.b, z1.b }, { z0.b, z1.b }
+// CHECK-ENCODING: [0x08,0x14,0xa0,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1a01408 <unknown>
+
+usdot   za.s[w8, 0], {z0.b, z1.b}, {z0.b, z1.b}  // 11000001-10100000-00010100-00001000
+// CHECK-INST: usdot   za.s[w8, 0, vgx2], { z0.b, z1.b }, { z0.b, z1.b }
+// CHECK-ENCODING: [0x08,0x14,0xa0,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1a01408 <unknown>
+
+usdot   za.s[w10, 5, vgx2], {z10.b, z11.b}, {z20.b, z21.b}  // 11000001-10110100-01010101-01001101
+// CHECK-INST: usdot   za.s[w10, 5, vgx2], { z10.b, z11.b }, { z20.b, z21.b }
+// CHECK-ENCODING: [0x4d,0x55,0xb4,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1b4554d <unknown>
+
+usdot   za.s[w10, 5], {z10.b, z11.b}, {z20.b, z21.b}  // 11000001-10110100-01010101-01001101
+// CHECK-INST: usdot   za.s[w10, 5, vgx2], { z10.b, z11.b }, { z20.b, z21.b }
+// CHECK-ENCODING: [0x4d,0x55,0xb4,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1b4554d <unknown>
+
+usdot   za.s[w11, 7, vgx2], {z12.b, z13.b}, {z8.b, z9.b}  // 11000001-10101000-01110101-10001111
+// CHECK-INST: usdot   za.s[w11, 7, vgx2], { z12.b, z13.b }, { z8.b, z9.b }
+// CHECK-ENCODING: [0x8f,0x75,0xa8,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1a8758f <unknown>
+
+usdot   za.s[w11, 7], {z12.b, z13.b}, {z8.b, z9.b}  // 11000001-10101000-01110101-10001111
+// CHECK-INST: usdot   za.s[w11, 7, vgx2], { z12.b, z13.b }, { z8.b, z9.b }
+// CHECK-ENCODING: [0x8f,0x75,0xa8,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1a8758f <unknown>
+
+usdot   za.s[w11, 7, vgx2], {z30.b, z31.b}, {z30.b, z31.b}  // 11000001-10111110-01110111-11001111
+// CHECK-INST: usdot   za.s[w11, 7, vgx2], { z30.b, z31.b }, { z30.b, z31.b }
+// CHECK-ENCODING: [0xcf,0x77,0xbe,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1be77cf <unknown>
+
+usdot   za.s[w11, 7], {z30.b, z31.b}, {z30.b, z31.b}  // 11000001-10111110-01110111-11001111
+// CHECK-INST: usdot   za.s[w11, 7, vgx2], { z30.b, z31.b }, { z30.b, z31.b }
+// CHECK-ENCODING: [0xcf,0x77,0xbe,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1be77cf <unknown>
+
+usdot   za.s[w8, 5, vgx2], {z16.b, z17.b}, {z16.b, z17.b}  // 11000001-10110000-00010110-00001101
+// CHECK-INST: usdot   za.s[w8, 5, vgx2], { z16.b, z17.b }, { z16.b, z17.b }
+// CHECK-ENCODING: [0x0d,0x16,0xb0,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1b0160d <unknown>
+
+usdot   za.s[w8, 5], {z16.b, z17.b}, {z16.b, z17.b}  // 11000001-10110000-00010110-00001101
+// CHECK-INST: usdot   za.s[w8, 5, vgx2], { z16.b, z17.b }, { z16.b, z17.b }
+// CHECK-ENCODING: [0x0d,0x16,0xb0,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1b0160d <unknown>
+
+usdot   za.s[w8, 1, vgx2], {z0.b, z1.b}, {z30.b, z31.b}  // 11000001-10111110-00010100-00001001
+// CHECK-INST: usdot   za.s[w8, 1, vgx2], { z0.b, z1.b }, { z30.b, z31.b }
+// CHECK-ENCODING: [0x09,0x14,0xbe,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1be1409 <unknown>
+
+usdot   za.s[w8, 1], {z0.b, z1.b}, {z30.b, z31.b}  // 11000001-10111110-00010100-00001001
+// CHECK-INST: usdot   za.s[w8, 1, vgx2], { z0.b, z1.b }, { z30.b, z31.b }
+// CHECK-ENCODING: [0x09,0x14,0xbe,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1be1409 <unknown>
+
+usdot   za.s[w10, 0, vgx2], {z18.b, z19.b}, {z20.b, z21.b}  // 11000001-10110100-01010110-01001000
+// CHECK-INST: usdot   za.s[w10, 0, vgx2], { z18.b, z19.b }, { z20.b, z21.b }
+// CHECK-ENCODING: [0x48,0x56,0xb4,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1b45648 <unknown>
+
+usdot   za.s[w10, 0], {z18.b, z19.b}, {z20.b, z21.b}  // 11000001-10110100-01010110-01001000
+// CHECK-INST: usdot   za.s[w10, 0, vgx2], { z18.b, z19.b }, { z20.b, z21.b }
+// CHECK-ENCODING: [0x48,0x56,0xb4,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1b45648 <unknown>
+
+usdot   za.s[w8, 0, vgx2], {z12.b, z13.b}, {z2.b, z3.b}  // 11000001-10100010-00010101-10001000
+// CHECK-INST: usdot   za.s[w8, 0, vgx2], { z12.b, z13.b }, { z2.b, z3.b }
+// CHECK-ENCODING: [0x88,0x15,0xa2,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1a21588 <unknown>
+
+usdot   za.s[w8, 0], {z12.b, z13.b}, {z2.b, z3.b}  // 11000001-10100010-00010101-10001000
+// CHECK-INST: usdot   za.s[w8, 0, vgx2], { z12.b, z13.b }, { z2.b, z3.b }
+// CHECK-ENCODING: [0x88,0x15,0xa2,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1a21588 <unknown>
+
+usdot   za.s[w10, 1, vgx2], {z0.b, z1.b}, {z26.b, z27.b}  // 11000001-10111010-01010100-00001001
+// CHECK-INST: usdot   za.s[w10, 1, vgx2], { z0.b, z1.b }, { z26.b, z27.b }
+// CHECK-ENCODING: [0x09,0x54,0xba,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1ba5409 <unknown>
+
+usdot   za.s[w10, 1], {z0.b, z1.b}, {z26.b, z27.b}  // 11000001-10111010-01010100-00001001
+// CHECK-INST: usdot   za.s[w10, 1, vgx2], { z0.b, z1.b }, { z26.b, z27.b }
+// CHECK-ENCODING: [0x09,0x54,0xba,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1ba5409 <unknown>
+
+usdot   za.s[w8, 5, vgx2], {z22.b, z23.b}, {z30.b, z31.b}  // 11000001-10111110-00010110-11001101
+// CHECK-INST: usdot   za.s[w8, 5, vgx2], { z22.b, z23.b }, { z30.b, z31.b }
+// CHECK-ENCODING: [0xcd,0x16,0xbe,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1be16cd <unknown>
+
+usdot   za.s[w8, 5], {z22.b, z23.b}, {z30.b, z31.b}  // 11000001-10111110-00010110-11001101
+// CHECK-INST: usdot   za.s[w8, 5, vgx2], { z22.b, z23.b }, { z30.b, z31.b }
+// CHECK-ENCODING: [0xcd,0x16,0xbe,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1be16cd <unknown>
+
+usdot   za.s[w11, 2, vgx2], {z8.b, z9.b}, {z0.b, z1.b}  // 11000001-10100000-01110101-00001010
+// CHECK-INST: usdot   za.s[w11, 2, vgx2], { z8.b, z9.b }, { z0.b, z1.b }
+// CHECK-ENCODING: [0x0a,0x75,0xa0,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1a0750a <unknown>
+
+usdot   za.s[w11, 2], {z8.b, z9.b}, {z0.b, z1.b}  // 11000001-10100000-01110101-00001010
+// CHECK-INST: usdot   za.s[w11, 2, vgx2], { z8.b, z9.b }, { z0.b, z1.b }
+// CHECK-ENCODING: [0x0a,0x75,0xa0,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1a0750a <unknown>
+
+usdot   za.s[w9, 7, vgx2], {z12.b, z13.b}, {z10.b, z11.b}  // 11000001-10101010-00110101-10001111
+// CHECK-INST: usdot   za.s[w9, 7, vgx2], { z12.b, z13.b }, { z10.b, z11.b }
+// CHECK-ENCODING: [0x8f,0x35,0xaa,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1aa358f <unknown>
+
+usdot   za.s[w9, 7], {z12.b, z13.b}, {z10.b, z11.b}  // 11000001-10101010-00110101-10001111
+// CHECK-INST: usdot   za.s[w9, 7, vgx2], { z12.b, z13.b }, { z10.b, z11.b }
+// CHECK-ENCODING: [0x8f,0x35,0xaa,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1aa358f <unknown>
+
+usdot   za.s[w8, 0, vgx4], {z0.b - z3.b}, z0.b  // 11000001-00110000-00010100-00001000
+// CHECK-INST: usdot   za.s[w8, 0, vgx4], { z0.b - z3.b }, z0.b
+// CHECK-ENCODING: [0x08,0x14,0x30,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1301408 <unknown>
+
+usdot   za.s[w8, 0], {z0.b - z3.b}, z0.b  // 11000001-00110000-00010100-00001000
+// CHECK-INST: usdot   za.s[w8, 0, vgx4], { z0.b - z3.b }, z0.b
+// CHECK-ENCODING: [0x08,0x14,0x30,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1301408 <unknown>
+
+usdot   za.s[w10, 5, vgx4], {z10.b - z13.b}, z5.b  // 11000001-00110101-01010101-01001101
+// CHECK-INST: usdot   za.s[w10, 5, vgx4], { z10.b - z13.b }, z5.b
+// CHECK-ENCODING: [0x4d,0x55,0x35,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c135554d <unknown>
+
+usdot   za.s[w10, 5], {z10.b - z13.b}, z5.b  // 11000001-00110101-01010101-01001101
+// CHECK-INST: usdot   za.s[w10, 5, vgx4], { z10.b - z13.b }, z5.b
+// CHECK-ENCODING: [0x4d,0x55,0x35,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c135554d <unknown>
+
+usdot   za.s[w11, 7, vgx4], {z13.b - z16.b}, z8.b  // 11000001-00111000-01110101-10101111
+// CHECK-INST: usdot   za.s[w11, 7, vgx4], { z13.b - z16.b }, z8.b
+// CHECK-ENCODING: [0xaf,0x75,0x38,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c13875af <unknown>
+
+usdot   za.s[w11, 7], {z13.b - z16.b}, z8.b  // 11000001-00111000-01110101-10101111
+// CHECK-INST: usdot   za.s[w11, 7, vgx4], { z13.b - z16.b }, z8.b
+// CHECK-ENCODING: [0xaf,0x75,0x38,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c13875af <unknown>
+
+usdot   za.s[w11, 7, vgx4], {z31.b - z2.b}, z15.b  // 11000001-00111111-01110111-11101111
+// CHECK-INST: usdot   za.s[w11, 7, vgx4], { z31.b, z0.b, z1.b, z2.b }, z15.b
+// CHECK-ENCODING: [0xef,0x77,0x3f,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c13f77ef <unknown>
+
+usdot   za.s[w11, 7], {z31.b - z2.b}, z15.b  // 11000001-00111111-01110111-11101111
+// CHECK-INST: usdot   za.s[w11, 7, vgx4], { z31.b, z0.b, z1.b, z2.b }, z15.b
+// CHECK-ENCODING: [0xef,0x77,0x3f,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c13f77ef <unknown>
+
+usdot   za.s[w8, 5, vgx4], {z17.b - z20.b}, z0.b  // 11000001-00110000-00010110-00101101
+// CHECK-INST: usdot   za.s[w8, 5, vgx4], { z17.b - z20.b }, z0.b
+// CHECK-ENCODING: [0x2d,0x16,0x30,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c130162d <unknown>
+
+usdot   za.s[w8, 5], {z17.b - z20.b}, z0.b  // 11000001-00110000-00010110-00101101
+// CHECK-INST: usdot   za.s[w8, 5, vgx4], { z17.b - z20.b }, z0.b
+// CHECK-ENCODING: [0x2d,0x16,0x30,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c130162d <unknown>
+
+usdot   za.s[w8, 1, vgx4], {z1.b - z4.b}, z14.b  // 11000001-00111110-00010100-00101001
+// CHECK-INST: usdot   za.s[w8, 1, vgx4], { z1.b - z4.b }, z14.b
+// CHECK-ENCODING: [0x29,0x14,0x3e,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c13e1429 <unknown>
+
+usdot   za.s[w8, 1], {z1.b - z4.b}, z14.b  // 11000001-00111110-00010100-00101001
+// CHECK-INST: usdot   za.s[w8, 1, vgx4], { z1.b - z4.b }, z14.b
+// CHECK-ENCODING: [0x29,0x14,0x3e,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c13e1429 <unknown>
+
+usdot   za.s[w10, 0, vgx4], {z19.b - z22.b}, z4.b  // 11000001-00110100-01010110-01101000
+// CHECK-INST: usdot   za.s[w10, 0, vgx4], { z19.b - z22.b }, z4.b
+// CHECK-ENCODING: [0x68,0x56,0x34,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1345668 <unknown>
+
+usdot   za.s[w10, 0], {z19.b - z22.b}, z4.b  // 11000001-00110100-01010110-01101000
+// CHECK-INST: usdot   za.s[w10, 0, vgx4], { z19.b - z22.b }, z4.b
+// CHECK-ENCODING: [0x68,0x56,0x34,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1345668 <unknown>
+
+usdot   za.s[w8, 0, vgx4], {z12.b - z15.b}, z2.b  // 11000001-00110010-00010101-10001000
+// CHECK-INST: usdot   za.s[w8, 0, vgx4], { z12.b - z15.b }, z2.b
+// CHECK-ENCODING: [0x88,0x15,0x32,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1321588 <unknown>
+
+usdot   za.s[w8, 0], {z12.b - z15.b}, z2.b  // 11000001-00110010-00010101-10001000
+// CHECK-INST: usdot   za.s[w8, 0, vgx4], { z12.b - z15.b }, z2.b
+// CHECK-ENCODING: [0x88,0x15,0x32,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1321588 <unknown>
+
+usdot   za.s[w10, 1, vgx4], {z1.b - z4.b}, z10.b  // 11000001-00111010-01010100-00101001
+// CHECK-INST: usdot   za.s[w10, 1, vgx4], { z1.b - z4.b }, z10.b
+// CHECK-ENCODING: [0x29,0x54,0x3a,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c13a5429 <unknown>
+
+usdot   za.s[w10, 1], {z1.b - z4.b}, z10.b  // 11000001-00111010-01010100-00101001
+// CHECK-INST: usdot   za.s[w10, 1, vgx4], { z1.b - z4.b }, z10.b
+// CHECK-ENCODING: [0x29,0x54,0x3a,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c13a5429 <unknown>
+
+usdot   za.s[w8, 5, vgx4], {z22.b - z25.b}, z14.b  // 11000001-00111110-00010110-11001101
+// CHECK-INST: usdot   za.s[w8, 5, vgx4], { z22.b - z25.b }, z14.b
+// CHECK-ENCODING: [0xcd,0x16,0x3e,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c13e16cd <unknown>
+
+usdot   za.s[w8, 5], {z22.b - z25.b}, z14.b  // 11000001-00111110-00010110-11001101
+// CHECK-INST: usdot   za.s[w8, 5, vgx4], { z22.b - z25.b }, z14.b
+// CHECK-ENCODING: [0xcd,0x16,0x3e,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c13e16cd <unknown>
+
+usdot   za.s[w11, 2, vgx4], {z9.b - z12.b}, z1.b  // 11000001-00110001-01110101-00101010
+// CHECK-INST: usdot   za.s[w11, 2, vgx4], { z9.b - z12.b }, z1.b
+// CHECK-ENCODING: [0x2a,0x75,0x31,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c131752a <unknown>
+
+usdot   za.s[w11, 2], {z9.b - z12.b}, z1.b  // 11000001-00110001-01110101-00101010
+// CHECK-INST: usdot   za.s[w11, 2, vgx4], { z9.b - z12.b }, z1.b
+// CHECK-ENCODING: [0x2a,0x75,0x31,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c131752a <unknown>
+
+usdot   za.s[w9, 7, vgx4], {z12.b - z15.b}, z11.b  // 11000001-00111011-00110101-10001111
+// CHECK-INST: usdot   za.s[w9, 7, vgx4], { z12.b - z15.b }, z11.b
+// CHECK-ENCODING: [0x8f,0x35,0x3b,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c13b358f <unknown>
+
+usdot   za.s[w9, 7], {z12.b - z15.b}, z11.b  // 11000001-00111011-00110101-10001111
+// CHECK-INST: usdot   za.s[w9, 7, vgx4], { z12.b - z15.b }, z11.b
+// CHECK-ENCODING: [0x8f,0x35,0x3b,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c13b358f <unknown>
+
+
 usdot   za.s[w8, 0, vgx4], {z0.b - z3.b}, z0.b[0]  // 11000001-01010000-10010000-00101000
 // CHECK-INST: usdot   za.s[w8, 0, vgx4], { z0.b - z3.b }, z0.b[0]
 // CHECK-ENCODING: [0x28,0x90,0x50,0xc1]
@@ -301,3 +735,148 @@ usdot   za.s[w9, 7], {z12.b - z15.b}, z11.b[2]  // 11000001-01011011-10111001-10
 // CHECK-ERROR: instruction requires: sme2
 // CHECK-UNKNOWN: c15bb9af <unknown>
 
+
+usdot   za.s[w8, 0, vgx4], {z0.b - z3.b}, {z0.b - z3.b}  // 11000001-10100001-00010100-00001000
+// CHECK-INST: usdot   za.s[w8, 0, vgx4], { z0.b - z3.b }, { z0.b - z3.b }
+// CHECK-ENCODING: [0x08,0x14,0xa1,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1a11408 <unknown>
+
+usdot   za.s[w8, 0], {z0.b - z3.b}, {z0.b - z3.b}  // 11000001-10100001-00010100-00001000
+// CHECK-INST: usdot   za.s[w8, 0, vgx4], { z0.b - z3.b }, { z0.b - z3.b }
+// CHECK-ENCODING: [0x08,0x14,0xa1,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1a11408 <unknown>
+
+usdot   za.s[w10, 5, vgx4], {z8.b - z11.b}, {z20.b - z23.b}  // 11000001-10110101-01010101-00001101
+// CHECK-INST: usdot   za.s[w10, 5, vgx4], { z8.b - z11.b }, { z20.b - z23.b }
+// CHECK-ENCODING: [0x0d,0x55,0xb5,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1b5550d <unknown>
+
+usdot   za.s[w10, 5], {z8.b - z11.b}, {z20.b - z23.b}  // 11000001-10110101-01010101-00001101
+// CHECK-INST: usdot   za.s[w10, 5, vgx4], { z8.b - z11.b }, { z20.b - z23.b }
+// CHECK-ENCODING: [0x0d,0x55,0xb5,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1b5550d <unknown>
+
+usdot   za.s[w11, 7, vgx4], {z12.b - z15.b}, {z8.b - z11.b}  // 11000001-10101001-01110101-10001111
+// CHECK-INST: usdot   za.s[w11, 7, vgx4], { z12.b - z15.b }, { z8.b - z11.b }
+// CHECK-ENCODING: [0x8f,0x75,0xa9,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1a9758f <unknown>
+
+usdot   za.s[w11, 7], {z12.b - z15.b}, {z8.b - z11.b}  // 11000001-10101001-01110101-10001111
+// CHECK-INST: usdot   za.s[w11, 7, vgx4], { z12.b - z15.b }, { z8.b - z11.b }
+// CHECK-ENCODING: [0x8f,0x75,0xa9,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1a9758f <unknown>
+
+usdot   za.s[w11, 7, vgx4], {z28.b - z31.b}, {z28.b - z31.b}  // 11000001-10111101-01110111-10001111
+// CHECK-INST: usdot   za.s[w11, 7, vgx4], { z28.b - z31.b }, { z28.b - z31.b }
+// CHECK-ENCODING: [0x8f,0x77,0xbd,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1bd778f <unknown>
+
+usdot   za.s[w11, 7], {z28.b - z31.b}, {z28.b - z31.b}  // 11000001-10111101-01110111-10001111
+// CHECK-INST: usdot   za.s[w11, 7, vgx4], { z28.b - z31.b }, { z28.b - z31.b }
+// CHECK-ENCODING: [0x8f,0x77,0xbd,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1bd778f <unknown>
+
+usdot   za.s[w8, 5, vgx4], {z16.b - z19.b}, {z16.b - z19.b}  // 11000001-10110001-00010110-00001101
+// CHECK-INST: usdot   za.s[w8, 5, vgx4], { z16.b - z19.b }, { z16.b - z19.b }
+// CHECK-ENCODING: [0x0d,0x16,0xb1,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1b1160d <unknown>
+
+usdot   za.s[w8, 5], {z16.b - z19.b}, {z16.b - z19.b}  // 11000001-10110001-00010110-00001101
+// CHECK-INST: usdot   za.s[w8, 5, vgx4], { z16.b - z19.b }, { z16.b - z19.b }
+// CHECK-ENCODING: [0x0d,0x16,0xb1,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1b1160d <unknown>
+
+usdot   za.s[w8, 1, vgx4], {z0.b - z3.b}, {z28.b - z31.b}  // 11000001-10111101-00010100-00001001
+// CHECK-INST: usdot   za.s[w8, 1, vgx4], { z0.b - z3.b }, { z28.b - z31.b }
+// CHECK-ENCODING: [0x09,0x14,0xbd,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1bd1409 <unknown>
+
+usdot   za.s[w8, 1], {z0.b - z3.b}, {z28.b - z31.b}  // 11000001-10111101-00010100-00001001
+// CHECK-INST: usdot   za.s[w8, 1, vgx4], { z0.b - z3.b }, { z28.b - z31.b }
+// CHECK-ENCODING: [0x09,0x14,0xbd,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1bd1409 <unknown>
+
+usdot   za.s[w10, 0, vgx4], {z16.b - z19.b}, {z20.b - z23.b}  // 11000001-10110101-01010110-00001000
+// CHECK-INST: usdot   za.s[w10, 0, vgx4], { z16.b - z19.b }, { z20.b - z23.b }
+// CHECK-ENCODING: [0x08,0x56,0xb5,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1b55608 <unknown>
+
+usdot   za.s[w10, 0], {z16.b - z19.b}, {z20.b - z23.b}  // 11000001-10110101-01010110-00001000
+// CHECK-INST: usdot   za.s[w10, 0, vgx4], { z16.b - z19.b }, { z20.b - z23.b }
+// CHECK-ENCODING: [0x08,0x56,0xb5,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1b55608 <unknown>
+
+usdot   za.s[w8, 0, vgx4], {z12.b - z15.b}, {z0.b - z3.b}  // 11000001-10100001-00010101-10001000
+// CHECK-INST: usdot   za.s[w8, 0, vgx4], { z12.b - z15.b }, { z0.b - z3.b }
+// CHECK-ENCODING: [0x88,0x15,0xa1,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1a11588 <unknown>
+
+usdot   za.s[w8, 0], {z12.b - z15.b}, {z0.b - z3.b}  // 11000001-10100001-00010101-10001000
+// CHECK-INST: usdot   za.s[w8, 0, vgx4], { z12.b - z15.b }, { z0.b - z3.b }
+// CHECK-ENCODING: [0x88,0x15,0xa1,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1a11588 <unknown>
+
+usdot   za.s[w10, 1, vgx4], {z0.b - z3.b}, {z24.b - z27.b}  // 11000001-10111001-01010100-00001001
+// CHECK-INST: usdot   za.s[w10, 1, vgx4], { z0.b - z3.b }, { z24.b - z27.b }
+// CHECK-ENCODING: [0x09,0x54,0xb9,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1b95409 <unknown>
+
+usdot   za.s[w10, 1], {z0.b - z3.b}, {z24.b - z27.b}  // 11000001-10111001-01010100-00001001
+// CHECK-INST: usdot   za.s[w10, 1, vgx4], { z0.b - z3.b }, { z24.b - z27.b }
+// CHECK-ENCODING: [0x09,0x54,0xb9,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1b95409 <unknown>
+
+usdot   za.s[w8, 5, vgx4], {z20.b - z23.b}, {z28.b - z31.b}  // 11000001-10111101-00010110-10001101
+// CHECK-INST: usdot   za.s[w8, 5, vgx4], { z20.b - z23.b }, { z28.b - z31.b }
+// CHECK-ENCODING: [0x8d,0x16,0xbd,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1bd168d <unknown>
+
+usdot   za.s[w8, 5], {z20.b - z23.b}, {z28.b - z31.b}  // 11000001-10111101-00010110-10001101
+// CHECK-INST: usdot   za.s[w8, 5, vgx4], { z20.b - z23.b }, { z28.b - z31.b }
+// CHECK-ENCODING: [0x8d,0x16,0xbd,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1bd168d <unknown>
+
+usdot   za.s[w11, 2, vgx4], {z8.b - z11.b}, {z0.b - z3.b}  // 11000001-10100001-01110101-00001010
+// CHECK-INST: usdot   za.s[w11, 2, vgx4], { z8.b - z11.b }, { z0.b - z3.b }
+// CHECK-ENCODING: [0x0a,0x75,0xa1,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1a1750a <unknown>
+
+usdot   za.s[w11, 2], {z8.b - z11.b}, {z0.b - z3.b}  // 11000001-10100001-01110101-00001010
+// CHECK-INST: usdot   za.s[w11, 2, vgx4], { z8.b - z11.b }, { z0.b - z3.b }
+// CHECK-ENCODING: [0x0a,0x75,0xa1,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1a1750a <unknown>
+
+usdot   za.s[w9, 7, vgx4], {z12.b - z15.b}, {z8.b - z11.b}  // 11000001-10101001-00110101-10001111
+// CHECK-INST: usdot   za.s[w9, 7, vgx4], { z12.b - z15.b }, { z8.b - z11.b }
+// CHECK-ENCODING: [0x8f,0x35,0xa9,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1a9358f <unknown>
+
+usdot   za.s[w9, 7], {z12.b - z15.b}, {z8.b - z11.b}  // 11000001-10101001-00110101-10001111
+// CHECK-INST: usdot   za.s[w9, 7, vgx4], { z12.b - z15.b }, { z8.b - z11.b }
+// CHECK-ENCODING: [0x8f,0x35,0xa9,0xc1]
+// CHECK-ERROR: instruction requires: sme2
+// CHECK-UNKNOWN: c1a9358f <unknown>
+