import data from llvm 9

1 files changed, 678 insertions, 946 deletions

diff --git a/lib/std/target/x86.zig b/lib/std/target/x86.zig
index 50b332f5e1..9d3b574401 100644
--- a/lib/std/target/x86.zig
+++ b/lib/std/target/x86.zig
@@ -108,12 +108,12 @@ pub const Feature = enum {
     slow_unaligned_mem_32,
     soft_float,
     sse,
+    sse_unaligned_mem,
     sse2,
     sse3,
     sse4_1,
     sse4_2,
     sse4a,
-    sse_unaligned_mem,
     ssse3,
     tbm,
     vaes,
@@ -134,7 +134,6 @@ pub const all_features = blk: {
     const len = @typeInfo(Feature).Enum.fields.len;
     std.debug.assert(len <= @typeInfo(Cpu.Feature.Set).Int.bits);
     var result: [len]Cpu.Feature = undefined;
-
     result[@enumToInt(Feature.@"16bit_mode")] = .{
         .index = @enumToInt(Feature.@"16bit_mode"),
         .name = @tagName(Feature.@"16bit_mode"),
@@ -142,7 +141,6 @@ pub const all_features = blk: {
         .description = "16-bit mode (i8086)",
         .dependencies = 0,
     };
-
     result[@enumToInt(Feature.@"32bit_mode")] = .{
         .index = @enumToInt(Feature.@"32bit_mode"),
         .name = @tagName(Feature.@"32bit_mode"),
@@ -150,7 +148,6 @@ pub const all_features = blk: {
         .description = "32-bit mode (80386)",
         .dependencies = 0,
     };
-
     result[@enumToInt(Feature.@"3dnow")] = .{
         .index = @enumToInt(Feature.@"3dnow"),
         .name = @tagName(Feature.@"3dnow"),
@@ -160,17 +157,15 @@ pub const all_features = blk: {
             .mmx,
         }),
     };
-
     result[@enumToInt(Feature.@"3dnowa")] = .{
         .index = @enumToInt(Feature.@"3dnowa"),
         .name = @tagName(Feature.@"3dnowa"),
         .llvm_name = "3dnowa",
         .description = "Enable 3DNow! Athlon instructions",
         .dependencies = featureSet(&[_]Feature{
-            .mmx,
+            .@"3dnow",
         }),
     };
-
     result[@enumToInt(Feature.@"64bit")] = .{
         .index = @enumToInt(Feature.@"64bit"),
         .name = @tagName(Feature.@"64bit"),
@@ -178,7 +173,6 @@ pub const all_features = blk: {
         .description = "Support 64-bit instructions",
         .dependencies = 0,
     };
-
     result[@enumToInt(Feature.@"64bit_mode")] = .{
         .index = @enumToInt(Feature.@"64bit_mode"),
         .name = @tagName(Feature.@"64bit_mode"),
@@ -186,167 +180,233 @@ pub const all_features = blk: {
         .description = "64-bit mode (x86_64)",
         .dependencies = 0,
     };
-
     result[@enumToInt(Feature.adx)] = .{
         .index = @enumToInt(Feature.adx),
         .name = @tagName(Feature.adx),
         .llvm_name = "adx",
         .description = "Support ADX instructions",
-        .dependencies = featureSet(&[_]Feature{}),
+        .dependencies = 0,
     };
-
     result[@enumToInt(Feature.aes)] = .{
         .index = @enumToInt(Feature.aes),
         .name = @tagName(Feature.aes),
         .llvm_name = "aes",
         .description = "Enable AES instructions",
         .dependencies = featureSet(&[_]Feature{
-            .sse,
+            .sse2,
         }),
     };
-
     result[@enumToInt(Feature.avx)] = .{
         .index = @enumToInt(Feature.avx),
         .name = @tagName(Feature.avx),
         .llvm_name = "avx",
         .description = "Enable AVX instructions",
         .dependencies = featureSet(&[_]Feature{
-            .sse,
+            .sse4_2,
         }),
     };
-
     result[@enumToInt(Feature.avx2)] = .{
         .index = @enumToInt(Feature.avx2),
         .name = @tagName(Feature.avx2),
         .llvm_name = "avx2",
         .description = "Enable AVX2 instructions",
         .dependencies = featureSet(&[_]Feature{
-            .sse,
-        }),
-    };
-
-    result[@enumToInt(Feature.avx512f)] = .{
-        .index = @enumToInt(Feature.avx512f),
-        .name = @tagName(Feature.avx512f),
-        .llvm_name = "avx512f",
-        .description = "Enable AVX-512 instructions",
-        .dependencies = featureSet(&[_]Feature{
-            .sse,
+            .avx,
         }),
     };
-
     result[@enumToInt(Feature.avx512bf16)] = .{
         .index = @enumToInt(Feature.avx512bf16),
         .name = @tagName(Feature.avx512bf16),
         .llvm_name = "avx512bf16",
         .description = "Support bfloat16 floating point",
         .dependencies = featureSet(&[_]Feature{
-            .sse,
+            .avx512bw,
         }),
     };
-
     result[@enumToInt(Feature.avx512bitalg)] = .{
         .index = @enumToInt(Feature.avx512bitalg),
         .name = @tagName(Feature.avx512bitalg),
         .llvm_name = "avx512bitalg",
         .description = "Enable AVX-512 Bit Algorithms",
         .dependencies = featureSet(&[_]Feature{
-            .sse,
+            .avx512bw,
+        }),
+    };
+    result[@enumToInt(Feature.avx512bw)] = .{
+        .index = @enumToInt(Feature.avx512bw),
+        .name = @tagName(Feature.avx512bw),
+        .llvm_name = "avx512bw",
+        .description = "Enable AVX-512 Byte and Word Instructions",
+        .dependencies = featureSet(&[_]Feature{
+            .avx512f,
+        }),
+    };
+    result[@enumToInt(Feature.avx512cd)] = .{
+        .index = @enumToInt(Feature.avx512cd),
+        .name = @tagName(Feature.avx512cd),
+        .llvm_name = "avx512cd",
+        .description = "Enable AVX-512 Conflict Detection Instructions",
+        .dependencies = featureSet(&[_]Feature{
+            .avx512f,
+        }),
+    };
+    result[@enumToInt(Feature.avx512dq)] = .{
+        .index = @enumToInt(Feature.avx512dq),
+        .name = @tagName(Feature.avx512dq),
+        .llvm_name = "avx512dq",
+        .description = "Enable AVX-512 Doubleword and Quadword Instructions",
+        .dependencies = featureSet(&[_]Feature{
+            .avx512f,
+        }),
+    };
+    result[@enumToInt(Feature.avx512er)] = .{
+        .index = @enumToInt(Feature.avx512er),
+        .name = @tagName(Feature.avx512er),
+        .llvm_name = "avx512er",
+        .description = "Enable AVX-512 Exponential and Reciprocal Instructions",
+        .dependencies = featureSet(&[_]Feature{
+            .avx512f,
+        }),
+    };
+    result[@enumToInt(Feature.avx512f)] = .{
+        .index = @enumToInt(Feature.avx512f),
+        .name = @tagName(Feature.avx512f),
+        .llvm_name = "avx512f",
+        .description = "Enable AVX-512 instructions",
+        .dependencies = featureSet(&[_]Feature{
+            .avx2,
+            .f16c,
+            .fma,
+        }),
+    };
+    result[@enumToInt(Feature.avx512ifma)] = .{
+        .index = @enumToInt(Feature.avx512ifma),
+        .name = @tagName(Feature.avx512ifma),
+        .llvm_name = "avx512ifma",
+        .description = "Enable AVX-512 Integer Fused Multiple-Add",
+        .dependencies = featureSet(&[_]Feature{
+            .avx512f,
+        }),
+    };
+    result[@enumToInt(Feature.avx512pf)] = .{
+        .index = @enumToInt(Feature.avx512pf),
+        .name = @tagName(Feature.avx512pf),
+        .llvm_name = "avx512pf",
+        .description = "Enable AVX-512 PreFetch Instructions",
+        .dependencies = featureSet(&[_]Feature{
+            .avx512f,
+        }),
+    };
+    result[@enumToInt(Feature.avx512vbmi)] = .{
+        .index = @enumToInt(Feature.avx512vbmi),
+        .name = @tagName(Feature.avx512vbmi),
+        .llvm_name = "avx512vbmi",
+        .description = "Enable AVX-512 Vector Byte Manipulation Instructions",
+        .dependencies = featureSet(&[_]Feature{
+            .avx512bw,
+        }),
+    };
+    result[@enumToInt(Feature.avx512vbmi2)] = .{
+        .index = @enumToInt(Feature.avx512vbmi2),
+        .name = @tagName(Feature.avx512vbmi2),
+        .llvm_name = "avx512vbmi2",
+        .description = "Enable AVX-512 further Vector Byte Manipulation Instructions",
+        .dependencies = featureSet(&[_]Feature{
+            .avx512bw,
+        }),
+    };
+    result[@enumToInt(Feature.avx512vl)] = .{
+        .index = @enumToInt(Feature.avx512vl),
+        .name = @tagName(Feature.avx512vl),
+        .llvm_name = "avx512vl",
+        .description = "Enable AVX-512 Vector Length eXtensions",
+        .dependencies = featureSet(&[_]Feature{
+            .avx512f,
+        }),
+    };
+    result[@enumToInt(Feature.avx512vnni)] = .{
+        .index = @enumToInt(Feature.avx512vnni),
+        .name = @tagName(Feature.avx512vnni),
+        .llvm_name = "avx512vnni",
+        .description = "Enable AVX-512 Vector Neural Network Instructions",
+        .dependencies = featureSet(&[_]Feature{
+            .avx512f,
+        }),
+    };
+    result[@enumToInt(Feature.avx512vp2intersect)] = .{
+        .index = @enumToInt(Feature.avx512vp2intersect),
+        .name = @tagName(Feature.avx512vp2intersect),
+        .llvm_name = "avx512vp2intersect",
+        .description = "Enable AVX-512 vp2intersect",
+        .dependencies = featureSet(&[_]Feature{
+            .avx512f,
+        }),
+    };
+    result[@enumToInt(Feature.avx512vpopcntdq)] = .{
+        .index = @enumToInt(Feature.avx512vpopcntdq),
+        .name = @tagName(Feature.avx512vpopcntdq),
+        .llvm_name = "avx512vpopcntdq",
+        .description = "Enable AVX-512 Population Count Instructions",
+        .dependencies = featureSet(&[_]Feature{
+            .avx512f,
         }),
     };
-
     result[@enumToInt(Feature.bmi)] = .{
         .index = @enumToInt(Feature.bmi),
         .name = @tagName(Feature.bmi),
         .llvm_name = "bmi",
         .description = "Support BMI instructions",
-        .dependencies = featureSet(&[_]Feature{}),
+        .dependencies = 0,
     };
-
     result[@enumToInt(Feature.bmi2)] = .{
         .index = @enumToInt(Feature.bmi2),
         .name = @tagName(Feature.bmi2),
         .llvm_name = "bmi2",
         .description = "Support BMI2 instructions",
-        .dependencies = featureSet(&[_]Feature{}),
-    };
-
-    result[@enumToInt(Feature.avx512bw)] = .{
-        .index = @enumToInt(Feature.avx512bw),
-        .name = @tagName(Feature.avx512bw),
-        .llvm_name = "avx512bw",
-        .description = "Enable AVX-512 Byte and Word Instructions",
-        .dependencies = featureSet(&[_]Feature{
-            .sse,
-        }),
+        .dependencies = 0,
     };
-
     result[@enumToInt(Feature.branchfusion)] = .{
         .index = @enumToInt(Feature.branchfusion),
         .name = @tagName(Feature.branchfusion),
         .llvm_name = "branchfusion",
         .description = "CMP/TEST can be fused with conditional branches",
-        .dependencies = featureSet(&[_]Feature{}),
-    };
-
-    result[@enumToInt(Feature.avx512cd)] = .{
-        .index = @enumToInt(Feature.avx512cd),
-        .name = @tagName(Feature.avx512cd),
-        .llvm_name = "avx512cd",
-        .description = "Enable AVX-512 Conflict Detection Instructions",
-        .dependencies = featureSet(&[_]Feature{
-            .sse,
-        }),
+        .dependencies = 0,
     };
-
     result[@enumToInt(Feature.cldemote)] = .{
         .index = @enumToInt(Feature.cldemote),
         .name = @tagName(Feature.cldemote),
         .llvm_name = "cldemote",
         .description = "Enable Cache Demote",
-        .dependencies = featureSet(&[_]Feature{}),
+        .dependencies = 0,
     };
-
     result[@enumToInt(Feature.clflushopt)] = .{
         .index = @enumToInt(Feature.clflushopt),
         .name = @tagName(Feature.clflushopt),
         .llvm_name = "clflushopt",
         .description = "Flush A Cache Line Optimized",
-        .dependencies = featureSet(&[_]Feature{}),
+        .dependencies = 0,
     };
-
     result[@enumToInt(Feature.clwb)] = .{
         .index = @enumToInt(Feature.clwb),
         .name = @tagName(Feature.clwb),
         .llvm_name = "clwb",
         .description = "Cache Line Write Back",
-        .dependencies = featureSet(&[_]Feature{}),
+        .dependencies = 0,
     };
-
     result[@enumToInt(Feature.clzero)] = .{
         .index = @enumToInt(Feature.clzero),
         .name = @tagName(Feature.clzero),
         .llvm_name = "clzero",
         .description = "Enable Cache Line Zero",
-        .dependencies = featureSet(&[_]Feature{}),
+        .dependencies = 0,
     };
-
     result[@enumToInt(Feature.cmov)] = .{
         .index = @enumToInt(Feature.cmov),
         .name = @tagName(Feature.cmov),
         .llvm_name = "cmov",
         .description = "Enable conditional move instructions",
-        .dependencies = featureSet(&[_]Feature{}),
-    };
-
-    result[@enumToInt(Feature.cx8)] = .{
-        .index = @enumToInt(Feature.cx8),
-        .name = @tagName(Feature.cx8),
-        .llvm_name = "cx8",
-        .description = "Support CMPXCHG8B instructions",
-        .dependencies = featureSet(&[_]Feature{}),
+        .dependencies = 0,
     };
-
     result[@enumToInt(Feature.cx16)] = .{
         .index = @enumToInt(Feature.cx16),
         .name = @tagName(Feature.cx16),
@@ -356,17 +416,13 @@ pub const all_features = blk: {
             .cx8,
         }),
     };
-
-    result[@enumToInt(Feature.avx512dq)] = .{
-        .index = @enumToInt(Feature.avx512dq),
-        .name = @tagName(Feature.avx512dq),
-        .llvm_name = "avx512dq",
-        .description = "Enable AVX-512 Doubleword and Quadword Instructions",
-        .dependencies = featureSet(&[_]Feature{
-            .sse,
-        }),
+    result[@enumToInt(Feature.cx8)] = .{
+        .index = @enumToInt(Feature.cx8),
+        .name = @tagName(Feature.cx8),
+        .llvm_name = "cx8",
+        .description = "Support CMPXCHG8B instructions",
+        .dependencies = 0,
     };
-
     result[@enumToInt(Feature.enqcmd)] = .{
         .index = @enumToInt(Feature.enqcmd),
         .name = @tagName(Feature.enqcmd),
@@ -374,17 +430,6 @@ pub const all_features = blk: {
         .description = "Has ENQCMD instructions",
         .dependencies = 0,
     };
-
-    result[@enumToInt(Feature.avx512er)] = .{
-        .index = @enumToInt(Feature.avx512er),
-        .name = @tagName(Feature.avx512er),
-        .llvm_name = "avx512er",
-        .description = "Enable AVX-512 Exponential and Reciprocal Instructions",
-        .dependencies = featureSet(&[_]Feature{
-            .sse,
-        }),
-    };
-
     result[@enumToInt(Feature.ermsb)] = .{
         .index = @enumToInt(Feature.ermsb),
         .name = @tagName(Feature.ermsb),
@@ -392,53 +437,29 @@ pub const all_features = blk: {
         .description = "REP MOVS/STOS are fast",
         .dependencies = 0,
     };
-
     result[@enumToInt(Feature.f16c)] = .{
         .index = @enumToInt(Feature.f16c),
         .name = @tagName(Feature.f16c),
         .llvm_name = "f16c",
         .description = "Support 16-bit floating point conversion instructions",
         .dependencies = featureSet(&[_]Feature{
-            .sse,
-        }),
-    };
-
-    result[@enumToInt(Feature.fma)] = .{
-        .index = @enumToInt(Feature.fma),
-        .name = @tagName(Feature.fma),
-        .llvm_name = "fma",
-        .description = "Enable three-operand fused multiple-add",
-        .dependencies = featureSet(&[_]Feature{
-            .sse,
-        }),
-    };
-
-    result[@enumToInt(Feature.fma4)] = .{
-        .index = @enumToInt(Feature.fma4),
-        .name = @tagName(Feature.fma4),
-        .llvm_name = "fma4",
-        .description = "Enable four-operand fused multiple-add",
-        .dependencies = featureSet(&[_]Feature{
-            .sse,
+            .avx,
         }),
     };
-
-    result[@enumToInt(Feature.fsgsbase)] = .{
-        .index = @enumToInt(Feature.fsgsbase),
-        .name = @tagName(Feature.fsgsbase),
-        .llvm_name = "fsgsbase",
-        .description = "Support FS/GS Base instructions",
+    result[@enumToInt(Feature.false_deps_lzcnt_tzcnt)] = .{
+        .index = @enumToInt(Feature.false_deps_lzcnt_tzcnt),
+        .name = @tagName(Feature.false_deps_lzcnt_tzcnt),
+        .llvm_name = "false-deps-lzcnt-tzcnt",
+        .description = "LZCNT/TZCNT have a false dependency on dest register",
         .dependencies = 0,
     };
-
-    result[@enumToInt(Feature.fxsr)] = .{
-        .index = @enumToInt(Feature.fxsr),
-        .name = @tagName(Feature.fxsr),
-        .llvm_name = "fxsr",
-        .description = "Support fxsave/fxrestore instructions",
+    result[@enumToInt(Feature.false_deps_popcnt)] = .{
+        .index = @enumToInt(Feature.false_deps_popcnt),
+        .name = @tagName(Feature.false_deps_popcnt),
+        .llvm_name = "false-deps-popcnt",
+        .description = "POPCNT has a false dependency on dest register",
         .dependencies = 0,
     };
-
     result[@enumToInt(Feature.fast_11bytenop)] = .{
         .index = @enumToInt(Feature.fast_11bytenop),
         .name = @tagName(Feature.fast_11bytenop),
@@ -446,7 +467,6 @@ pub const all_features = blk: {
         .description = "Target can quickly decode up to 11 byte NOPs",
         .dependencies = 0,
     };
-
     result[@enumToInt(Feature.fast_15bytenop)] = .{
         .index = @enumToInt(Feature.fast_15bytenop),
         .name = @tagName(Feature.fast_15bytenop),
@@ -454,7 +474,6 @@ pub const all_features = blk: {
         .description = "Target can quickly decode up to 15 byte NOPs",
         .dependencies = 0,
     };
-
     result[@enumToInt(Feature.fast_bextr)] = .{
         .index = @enumToInt(Feature.fast_bextr),
         .name = @tagName(Feature.fast_bextr),
@@ -462,17 +481,22 @@ pub const all_features = blk: {
         .description = "Indicates that the BEXTR instruction is implemented as a single uop with good throughput",
         .dependencies = 0,
     };
-
+    result[@enumToInt(Feature.fast_gather)] = .{
+        .index = @enumToInt(Feature.fast_gather),
+        .name = @tagName(Feature.fast_gather),
+        .llvm_name = "fast-gather",
+        .description = "Indicates if gather is reasonably fast",
+        .dependencies = 0,
+    };
     result[@enumToInt(Feature.fast_hops)] = .{
         .index = @enumToInt(Feature.fast_hops),
         .name = @tagName(Feature.fast_hops),
         .llvm_name = "fast-hops",
         .description = "Prefer horizontal vector math instructions (haddp, phsub, etc.) over normal vector instructions with shuffles",
         .dependencies = featureSet(&[_]Feature{
-            .sse,
+            .sse3,
         }),
     };
-
     result[@enumToInt(Feature.fast_lzcnt)] = .{
         .index = @enumToInt(Feature.fast_lzcnt),
         .name = @tagName(Feature.fast_lzcnt),
@@ -480,7 +504,6 @@ pub const all_features = blk: {
         .description = "LZCNT instructions are as fast as most simple integer ops",
         .dependencies = 0,
     };
-
     result[@enumToInt(Feature.fast_partial_ymm_or_zmm_write)] = .{
         .index = @enumToInt(Feature.fast_partial_ymm_or_zmm_write),
         .name = @tagName(Feature.fast_partial_ymm_or_zmm_write),
@@ -488,15 +511,6 @@ pub const all_features = blk: {
         .description = "Partial writes to YMM/ZMM registers are fast",
         .dependencies = 0,
     };
-
-    result[@enumToInt(Feature.fast_shld_rotate)] = .{
-        .index = @enumToInt(Feature.fast_shld_rotate),
-        .name = @tagName(Feature.fast_shld_rotate),
-        .llvm_name = "fast-shld-rotate",
-        .description = "SHLD can be used as a faster rotate",
-        .dependencies = 0,
-    };
-
     result[@enumToInt(Feature.fast_scalar_fsqrt)] = .{
         .index = @enumToInt(Feature.fast_scalar_fsqrt),
         .name = @tagName(Feature.fast_scalar_fsqrt),
@@ -504,7 +518,6 @@ pub const all_features = blk: {
         .description = "Scalar SQRT is fast (disable Newton-Raphson)",
         .dependencies = 0,
     };
-
     result[@enumToInt(Feature.fast_scalar_shift_masks)] = .{
         .index = @enumToInt(Feature.fast_scalar_shift_masks),
         .name = @tagName(Feature.fast_scalar_shift_masks),
@@ -512,7 +525,13 @@ pub const all_features = blk: {
         .description = "Prefer a left/right scalar logical shift pair over a shift+and pair",
         .dependencies = 0,
     };
-
+    result[@enumToInt(Feature.fast_shld_rotate)] = .{
+        .index = @enumToInt(Feature.fast_shld_rotate),
+        .name = @tagName(Feature.fast_shld_rotate),
+        .llvm_name = "fast-shld-rotate",
+        .description = "SHLD can be used as a faster rotate",
+        .dependencies = 0,
+    };
     result[@enumToInt(Feature.fast_variable_shuffle)] = .{
         .index = @enumToInt(Feature.fast_variable_shuffle),
         .name = @tagName(Feature.fast_variable_shuffle),
@@ -520,7 +539,6 @@ pub const all_features = blk: {
         .description = "Shuffles with variable masks are fast",
         .dependencies = 0,
     };
-
     result[@enumToInt(Feature.fast_vector_fsqrt)] = .{
         .index = @enumToInt(Feature.fast_vector_fsqrt),
         .name = @tagName(Feature.fast_vector_fsqrt),
@@ -528,7 +546,6 @@ pub const all_features = blk: {
         .description = "Vector SQRT is fast (disable Newton-Raphson)",
         .dependencies = 0,
     };
-
     result[@enumToInt(Feature.fast_vector_shift_masks)] = .{
         .index = @enumToInt(Feature.fast_vector_shift_masks),
         .name = @tagName(Feature.fast_vector_shift_masks),
@@ -536,35 +553,62 @@ pub const all_features = blk: {
         .description = "Prefer a left/right vector logical shift pair over a shift+and pair",
         .dependencies = 0,
     };
-
+    result[@enumToInt(Feature.fma)] = .{
+        .index = @enumToInt(Feature.fma),
+        .name = @tagName(Feature.fma),
+        .llvm_name = "fma",
+        .description = "Enable three-operand fused multiple-add",
+        .dependencies = featureSet(&[_]Feature{
+            .avx,
+        }),
+    };
+    result[@enumToInt(Feature.fma4)] = .{
+        .index = @enumToInt(Feature.fma4),
+        .name = @tagName(Feature.fma4),
+        .llvm_name = "fma4",
+        .description = "Enable four-operand fused multiple-add",
+        .dependencies = featureSet(&[_]Feature{
+            .avx,
+            .sse4a,
+        }),
+    };
+    result[@enumToInt(Feature.fsgsbase)] = .{
+        .index = @enumToInt(Feature.fsgsbase),
+        .name = @tagName(Feature.fsgsbase),
+        .llvm_name = "fsgsbase",
+        .description = "Support FS/GS Base instructions",
+        .dependencies = 0,
+    };
+    result[@enumToInt(Feature.fxsr)] = .{
+        .index = @enumToInt(Feature.fxsr),
+        .name = @tagName(Feature.fxsr),
+        .llvm_name = "fxsr",
+        .description = "Support fxsave/fxrestore instructions",
+        .dependencies = 0,
+    };
     result[@enumToInt(Feature.gfni)] = .{
         .index = @enumToInt(Feature.gfni),
         .name = @tagName(Feature.gfni),
         .llvm_name = "gfni",
         .description = "Enable Galois Field Arithmetic Instructions",
         .dependencies = featureSet(&[_]Feature{
-            .sse,
+            .sse2,
         }),
     };
-
-    result[@enumToInt(Feature.fast_gather)] = .{
-        .index = @enumToInt(Feature.fast_gather),
-        .name = @tagName(Feature.fast_gather),
-        .llvm_name = "fast-gather",
-        .description = "Indicates if gather is reasonably fast",
+    result[@enumToInt(Feature.idivl_to_divb)] = .{
+        .index = @enumToInt(Feature.idivl_to_divb),
+        .name = @tagName(Feature.idivl_to_divb),
+        .llvm_name = "idivl-to-divb",
+        .description = "Use 8-bit divide for positive values less than 256",
         .dependencies = 0,
     };
-
-    result[@enumToInt(Feature.avx512ifma)] = .{
-        .index = @enumToInt(Feature.avx512ifma),
-        .name = @tagName(Feature.avx512ifma),
-        .llvm_name = "avx512ifma",
-        .description = "Enable AVX-512 Integer Fused Multiple-Add",
-        .dependencies = featureSet(&[_]Feature{
-            .sse,
-        }),
+    result[@enumToInt(Feature.idivq_to_divl)] = .{
+        .index = @enumToInt(Feature.idivq_to_divl),
+        .name = @tagName(Feature.idivq_to_divl),
+        .llvm_name = "idivq-to-divl",
+        .description = "Use 32-bit divide for positive values less than 2^32",
+        .dependencies = 0,
     };
-
     result[@enumToInt(Feature.invpcid)] = .{
         .index = @enumToInt(Feature.invpcid),
         .name = @tagName(Feature.invpcid),
@@ -572,15 +616,6 @@ pub const all_features = blk: {
         .description = "Invalidate Process-Context Identifier",
         .dependencies = 0,
     };
-
-    result[@enumToInt(Feature.sahf)] = .{
-        .index = @enumToInt(Feature.sahf),
-        .name = @tagName(Feature.sahf),
-        .llvm_name = "sahf",
-        .description = "Support LAHF and SAHF instructions",
-        .dependencies = 0,
-    };
-
     result[@enumToInt(Feature.lea_sp)] = .{
         .index = @enumToInt(Feature.lea_sp),
         .name = @tagName(Feature.lea_sp),
@@ -588,7 +623,6 @@ pub const all_features = blk: {
         .description = "Use LEA for adjusting the stack pointer",
         .dependencies = 0,
     };
-
     result[@enumToInt(Feature.lea_uses_ag)] = .{
         .index = @enumToInt(Feature.lea_uses_ag),
         .name = @tagName(Feature.lea_uses_ag),
@@ -596,7 +630,6 @@ pub const all_features = blk: {
         .description = "LEA instruction needs inputs at AG stage",
         .dependencies = 0,
     };
-
     result[@enumToInt(Feature.lwp)] = .{
         .index = @enumToInt(Feature.lwp),
         .name = @tagName(Feature.lwp),
@@ -604,7 +637,6 @@ pub const all_features = blk: {
         .description = "Enable LWP instructions",
         .dependencies = 0,
     };
-
     result[@enumToInt(Feature.lzcnt)] = .{
         .index = @enumToInt(Feature.lzcnt),
         .name = @tagName(Feature.lzcnt),
@@ -612,15 +644,20 @@ pub const all_features = blk: {
         .description = "Support LZCNT instruction",
         .dependencies = 0,
     };
-
-    result[@enumToInt(Feature.false_deps_lzcnt_tzcnt)] = .{
-        .index = @enumToInt(Feature.false_deps_lzcnt_tzcnt),
-        .name = @tagName(Feature.false_deps_lzcnt_tzcnt),
-        .llvm_name = "false-deps-lzcnt-tzcnt",
-        .description = "LZCNT/TZCNT have a false dependency on dest register",
+    result[@enumToInt(Feature.macrofusion)] = .{
+        .index = @enumToInt(Feature.macrofusion),
+        .name = @tagName(Feature.macrofusion),
+        .llvm_name = "macrofusion",
+        .description = "Various instructions can be fused with conditional branches",
+        .dependencies = 0,
+    };
+    result[@enumToInt(Feature.merge_to_threeway_branch)] = .{
+        .index = @enumToInt(Feature.merge_to_threeway_branch),
+        .name = @tagName(Feature.merge_to_threeway_branch),
+        .llvm_name = "merge-to-threeway-branch",
+        .description = "Merge branches to a three-way conditional branch",
         .dependencies = 0,
     };
-
     result[@enumToInt(Feature.mmx)] = .{
         .index = @enumToInt(Feature.mmx),
         .name = @tagName(Feature.mmx),
@@ -628,7 +665,6 @@ pub const all_features = blk: {
         .description = "Enable MMX instructions",
         .dependencies = 0,
     };
-
     result[@enumToInt(Feature.movbe)] = .{
         .index = @enumToInt(Feature.movbe),
         .name = @tagName(Feature.movbe),
@@ -636,7 +672,6 @@ pub const all_features = blk: {
         .description = "Support MOVBE instruction",
         .dependencies = 0,
     };
-
     result[@enumToInt(Feature.movdir64b)] = .{
         .index = @enumToInt(Feature.movdir64b),
         .name = @tagName(Feature.movdir64b),
@@ -644,7 +679,6 @@ pub const all_features = blk: {
         .description = "Support movdir64b instruction",
         .dependencies = 0,
     };
-
     result[@enumToInt(Feature.movdiri)] = .{
         .index = @enumToInt(Feature.movdiri),
         .name = @tagName(Feature.movdiri),
@@ -652,7 +686,6 @@ pub const all_features = blk: {
         .description = "Support movdiri instruction",
         .dependencies = 0,
     };
-
     result[@enumToInt(Feature.mpx)] = .{
         .index = @enumToInt(Feature.mpx),
         .name = @tagName(Feature.mpx),
@@ -660,7 +693,6 @@ pub const all_features = blk: {
         .description = "Support MPX instructions",
         .dependencies = 0,
     };
-
     result[@enumToInt(Feature.mwaitx)] = .{
         .index = @enumToInt(Feature.mwaitx),
         .name = @tagName(Feature.mwaitx),
@@ -668,23 +700,6 @@ pub const all_features = blk: {
         .description = "Enable MONITORX/MWAITX timer functionality",
         .dependencies = 0,
     };
-
-    result[@enumToInt(Feature.macrofusion)] = .{
-        .index = @enumToInt(Feature.macrofusion),
-        .name = @tagName(Feature.macrofusion),
-        .llvm_name = "macrofusion",
-        .description = "Various instructions can be fused with conditional branches",
-        .dependencies = 0,
-    };
-
-    result[@enumToInt(Feature.merge_to_threeway_branch)] = .{
-        .index = @enumToInt(Feature.merge_to_threeway_branch),
-        .name = @tagName(Feature.merge_to_threeway_branch),
-        .llvm_name = "merge-to-threeway-branch",
-        .description = "Merge branches to a three-way conditional branch",
-        .dependencies = 0,
-    };
-
     result[@enumToInt(Feature.nopl)] = .{
         .index = @enumToInt(Feature.nopl),
         .name = @tagName(Feature.nopl),
@@ -692,17 +707,22 @@ pub const all_features = blk: {
         .description = "Enable NOPL instruction",
         .dependencies = 0,
     };
-
+    result[@enumToInt(Feature.pad_short_functions)] = .{
+        .index = @enumToInt(Feature.pad_short_functions),
+        .name = @tagName(Feature.pad_short_functions),
+        .llvm_name = "pad-short-functions",
+        .description = "Pad short functions",
+        .dependencies = 0,
+    };
     result[@enumToInt(Feature.pclmul)] = .{
         .index = @enumToInt(Feature.pclmul),
         .name = @tagName(Feature.pclmul),
         .llvm_name = "pclmul",
         .description = "Enable packed carry-less multiplication instructions",
         .dependencies = featureSet(&[_]Feature{
-            .sse,
+            .sse2,
         }),
     };
-
     result[@enumToInt(Feature.pconfig)] = .{
         .index = @enumToInt(Feature.pconfig),
         .name = @tagName(Feature.pconfig),
@@ -710,17 +730,6 @@ pub const all_features = blk: {
         .description = "platform configuration instruction",
         .dependencies = 0,
     };
-
-    result[@enumToInt(Feature.avx512pf)] = .{
-        .index = @enumToInt(Feature.avx512pf),
-        .name = @tagName(Feature.avx512pf),
-        .llvm_name = "avx512pf",
-        .description = "Enable AVX-512 PreFetch Instructions",
-        .dependencies = featureSet(&[_]Feature{
-            .sse,
-        }),
-    };
-
     result[@enumToInt(Feature.pku)] = .{
         .index = @enumToInt(Feature.pku),
         .name = @tagName(Feature.pku),
@@ -728,7 +737,6 @@ pub const all_features = blk: {
         .description = "Enable protection keys",
         .dependencies = 0,
     };
-
     result[@enumToInt(Feature.popcnt)] = .{
         .index = @enumToInt(Feature.popcnt),
         .name = @tagName(Feature.popcnt),
@@ -736,15 +744,13 @@ pub const all_features = blk: {
         .description = "Support POPCNT instruction",
         .dependencies = 0,
     };
-
-    result[@enumToInt(Feature.false_deps_popcnt)] = .{
-        .index = @enumToInt(Feature.false_deps_popcnt),
-        .name = @tagName(Feature.false_deps_popcnt),
-        .llvm_name = "false-deps-popcnt",
-        .description = "POPCNT has a false dependency on dest register",
+    result[@enumToInt(Feature.prefer_256_bit)] = .{
+        .index = @enumToInt(Feature.prefer_256_bit),
+        .name = @tagName(Feature.prefer_256_bit),
+        .llvm_name = "prefer-256-bit",
+        .description = "Prefer 256-bit AVX instructions",
         .dependencies = 0,
     };
-
     result[@enumToInt(Feature.prefetchwt1)] = .{
         .index = @enumToInt(Feature.prefetchwt1),
         .name = @tagName(Feature.prefetchwt1),
@@ -752,7 +758,6 @@ pub const all_features = blk: {
         .description = "Prefetch with Intent to Write and T1 Hint",
         .dependencies = 0,
     };
-
     result[@enumToInt(Feature.prfchw)] = .{
         .index = @enumToInt(Feature.prfchw),
         .name = @tagName(Feature.prfchw),
@@ -760,7 +765,6 @@ pub const all_features = blk: {
         .description = "Support PRFCHW instructions",
         .dependencies = 0,
     };
-
     result[@enumToInt(Feature.ptwrite)] = .{
         .index = @enumToInt(Feature.ptwrite),
         .name = @tagName(Feature.ptwrite),
@@ -768,23 +772,6 @@ pub const all_features = blk: {
         .description = "Support ptwrite instruction",
         .dependencies = 0,
     };
-
-    result[@enumToInt(Feature.pad_short_functions)] = .{
-        .index = @enumToInt(Feature.pad_short_functions),
-        .name = @tagName(Feature.pad_short_functions),
-        .llvm_name = "pad-short-functions",
-        .description = "Pad short functions",
-        .dependencies = 0,
-    };
-
-    result[@enumToInt(Feature.prefer_256_bit)] = .{
-        .index = @enumToInt(Feature.prefer_256_bit),
-        .name = @tagName(Feature.prefer_256_bit),
-        .llvm_name = "prefer-256-bit",
-        .description = "Prefer 256-bit AVX instructions",
-        .dependencies = 0,
-    };
-
     result[@enumToInt(Feature.rdpid)] = .{
         .index = @enumToInt(Feature.rdpid),
         .name = @tagName(Feature.rdpid),
@@ -792,7 +779,6 @@ pub const all_features = blk: {
         .description = "Support RDPID instructions",
         .dependencies = 0,
     };
-
     result[@enumToInt(Feature.rdrnd)] = .{
         .index = @enumToInt(Feature.rdrnd),
         .name = @tagName(Feature.rdrnd),
@@ -800,7 +786,6 @@ pub const all_features = blk: {
         .description = "Support RDRAND instruction",
         .dependencies = 0,
     };
-
     result[@enumToInt(Feature.rdseed)] = .{
         .index = @enumToInt(Feature.rdseed),
         .name = @tagName(Feature.rdseed),
@@ -808,26 +793,16 @@ pub const all_features = blk: {
         .description = "Support RDSEED instruction",
         .dependencies = 0,
     };
-
-    result[@enumToInt(Feature.rtm)] = .{
-        .index = @enumToInt(Feature.rtm),
-        .name = @tagName(Feature.rtm),
-        .llvm_name = "rtm",
-        .description = "Support RTM instructions",
-        .dependencies = 0,
-    };
-
     result[@enumToInt(Feature.retpoline)] = .{
         .index = @enumToInt(Feature.retpoline),
         .name = @tagName(Feature.retpoline),
         .llvm_name = "retpoline",
         .description = "Remove speculation of indirect branches from the generated code, either by avoiding them entirely or lowering them with a speculation blocking construct",
         .dependencies = featureSet(&[_]Feature{
-            .retpoline_indirect_calls,
             .retpoline_indirect_branches,
+            .retpoline_indirect_calls,
         }),
     };
-
     result[@enumToInt(Feature.retpoline_external_thunk)] = .{
         .index = @enumToInt(Feature.retpoline_external_thunk),
         .name = @tagName(Feature.retpoline_external_thunk),
@@ -837,7 +812,6 @@ pub const all_features = blk: {
             .retpoline_indirect_calls,
         }),
     };
-
     result[@enumToInt(Feature.retpoline_indirect_branches)] = .{
         .index = @enumToInt(Feature.retpoline_indirect_branches),
         .name = @tagName(Feature.retpoline_indirect_branches),
@@ -845,7 +819,6 @@ pub const all_features = blk: {
         .description = "Remove speculation of indirect branches from the generated code",
         .dependencies = 0,
     };
-
     result[@enumToInt(Feature.retpoline_indirect_calls)] = .{
         .index = @enumToInt(Feature.retpoline_indirect_calls),
         .name = @tagName(Feature.retpoline_indirect_calls),
@@ -853,7 +826,20 @@ pub const all_features = blk: {
         .description = "Remove speculation of indirect calls from the generated code",
         .dependencies = 0,
     };
-
+    result[@enumToInt(Feature.rtm)] = .{
+        .index = @enumToInt(Feature.rtm),
+        .name = @tagName(Feature.rtm),
+        .llvm_name = "rtm",
+        .description = "Support RTM instructions",
+        .dependencies = 0,
+    };
+    result[@enumToInt(Feature.sahf)] = .{
+        .index = @enumToInt(Feature.sahf),
+        .name = @tagName(Feature.sahf),
+        .llvm_name = "sahf",
+        .description = "Support LAHF and SAHF instructions",
+        .dependencies = 0,
+    };
     result[@enumToInt(Feature.sgx)] = .{
         .index = @enumToInt(Feature.sgx),
         .name = @tagName(Feature.sgx),
@@ -861,17 +847,15 @@ pub const all_features = blk: {
         .description = "Enable Software Guard Extensions",
         .dependencies = 0,
     };
-
     result[@enumToInt(Feature.sha)] = .{
         .index = @enumToInt(Feature.sha),
         .name = @tagName(Feature.sha),
         .llvm_name = "sha",
         .description = "Enable SHA instructions",
         .dependencies = featureSet(&[_]Feature{
-            .sse,
+            .sse2,
         }),
     };
-
     result[@enumToInt(Feature.shstk)] = .{
         .index = @enumToInt(Feature.shstk),
         .name = @tagName(Feature.shstk),
@@ -879,83 +863,6 @@ pub const all_features = blk: {
         .description = "Support CET Shadow-Stack instructions",
         .dependencies = 0,
     };
-
-    result[@enumToInt(Feature.sse)] = .{
-        .index = @enumToInt(Feature.sse),
-        .name = @tagName(Feature.sse),
-        .llvm_name = "sse",
-        .description = "Enable SSE instructions",
-        .dependencies = 0,
-    };
-
-    result[@enumToInt(Feature.sse2)] = .{
-        .index = @enumToInt(Feature.sse2),
-        .name = @tagName(Feature.sse2),
-        .llvm_name = "sse2",
-        .description = "Enable SSE2 instructions",
-        .dependencies = featureSet(&[_]Feature{
-            .sse,
-        }),
-    };
-
-    result[@enumToInt(Feature.sse3)] = .{
-        .index = @enumToInt(Feature.sse3),
-        .name = @tagName(Feature.sse3),
-        .llvm_name = "sse3",
-        .description = "Enable SSE3 instructions",
-        .dependencies = featureSet(&[_]Feature{
-            .sse,
-        }),
-    };
-
-    result[@enumToInt(Feature.sse4a)] = .{
-        .index = @enumToInt(Feature.sse4a),
-        .name = @tagName(Feature.sse4a),
-        .llvm_name = "sse4a",
-        .description = "Support SSE 4a instructions",
-        .dependencies = featureSet(&[_]Feature{
-            .sse,
-        }),
-    };
-
-    result[@enumToInt(Feature.sse4_1)] = .{
-        .index = @enumToInt(Feature.sse4_1),
-        .name = @tagName(Feature.sse4_1),
-        .llvm_name = "sse4.1",
-        .description = "Enable SSE 4.1 instructions",
-        .dependencies = featureSet(&[_]Feature{
-            .sse,
-        }),
-    };
-
-    result[@enumToInt(Feature.sse4_2)] = .{
-        .index = @enumToInt(Feature.sse4_2),
-        .name = @tagName(Feature.sse4_2),
-        .llvm_name = "sse4.2",
-        .description = "Enable SSE 4.2 instructions",
-        .dependencies = featureSet(&[_]Feature{
-            .sse,
-        }),
-    };
-
-    result[@enumToInt(Feature.sse_unaligned_mem)] = .{
-        .index = @enumToInt(Feature.sse_unaligned_mem),
-        .name = @tagName(Feature.sse_unaligned_mem),
-        .llvm_name = "sse-unaligned-mem",
-        .description = "Allow unaligned memory operands with SSE instructions",
-        .dependencies = 0,
-    };
-
-    result[@enumToInt(Feature.ssse3)] = .{
-        .index = @enumToInt(Feature.ssse3),
-        .name = @tagName(Feature.ssse3),
-        .llvm_name = "ssse3",
-        .description = "Enable SSSE3 instructions",
-        .dependencies = featureSet(&[_]Feature{
-            .sse,
-        }),
-    };
-
     result[@enumToInt(Feature.slow_3ops_lea)] = .{
         .index = @enumToInt(Feature.slow_3ops_lea),
         .name = @tagName(Feature.slow_3ops_lea),
@@ -963,23 +870,6 @@ pub const all_features = blk: {
         .description = "LEA instruction with 3 ops or certain registers is slow",
         .dependencies = 0,
     };
-
-    result[@enumToInt(Feature.idivl_to_divb)] = .{
-        .index = @enumToInt(Feature.idivl_to_divb),
-        .name = @tagName(Feature.idivl_to_divb),
-        .llvm_name = "idivl-to-divb",
-        .description = "Use 8-bit divide for positive values less than 256",
-        .dependencies = 0,
-    };
-
-    result[@enumToInt(Feature.idivq_to_divl)] = .{
-        .index = @enumToInt(Feature.idivq_to_divl),
-        .name = @tagName(Feature.idivq_to_divl),
-        .llvm_name = "idivq-to-divl",
-        .description = "Use 32-bit divide for positive values less than 2^32",
-        .dependencies = 0,
-    };
-
     result[@enumToInt(Feature.slow_incdec)] = .{
         .index = @enumToInt(Feature.slow_incdec),
         .name = @tagName(Feature.slow_incdec),
@@ -987,7 +877,6 @@ pub const all_features = blk: {
         .description = "INC and DEC instructions are slower than ADD and SUB",
         .dependencies = 0,
     };
-
     result[@enumToInt(Feature.slow_lea)] = .{
         .index = @enumToInt(Feature.slow_lea),
         .name = @tagName(Feature.slow_lea),
@@ -995,7 +884,6 @@ pub const all_features = blk: {
         .description = "LEA instruction with certain arguments is slow",
         .dependencies = 0,
     };
-
     result[@enumToInt(Feature.slow_pmaddwd)] = .{
         .index = @enumToInt(Feature.slow_pmaddwd),
         .name = @tagName(Feature.slow_pmaddwd),
@@ -1003,7 +891,6 @@ pub const all_features = blk: {
         .description = "PMADDWD is slower than PMULLD",
         .dependencies = 0,
     };
-
     result[@enumToInt(Feature.slow_pmulld)] = .{
         .index = @enumToInt(Feature.slow_pmulld),
         .name = @tagName(Feature.slow_pmulld),
@@ -1011,7 +898,6 @@ pub const all_features = blk: {
         .description = "PMULLD instruction is slow",
         .dependencies = 0,
     };
-
     result[@enumToInt(Feature.slow_shld)] = .{
         .index = @enumToInt(Feature.slow_shld),
         .name = @tagName(Feature.slow_shld),
@@ -1019,7 +905,6 @@ pub const all_features = blk: {
         .description = "SHLD instruction is slow",
         .dependencies = 0,
     };
-
     result[@enumToInt(Feature.slow_two_mem_ops)] = .{
         .index = @enumToInt(Feature.slow_two_mem_ops),
         .name = @tagName(Feature.slow_two_mem_ops),
@@ -1027,7 +912,6 @@ pub const all_features = blk: {
         .description = "Two memory operand instructions are slow",
         .dependencies = 0,
     };
-
     result[@enumToInt(Feature.slow_unaligned_mem_16)] = .{
         .index = @enumToInt(Feature.slow_unaligned_mem_16),
         .name = @tagName(Feature.slow_unaligned_mem_16),
@@ -1035,7 +919,6 @@ pub const all_features = blk: {
         .description = "Slow unaligned 16-byte memory access",
         .dependencies = 0,
     };
-
     result[@enumToInt(Feature.slow_unaligned_mem_32)] = .{
         .index = @enumToInt(Feature.slow_unaligned_mem_32),
         .name = @tagName(Feature.slow_unaligned_mem_32),
@@ -1043,7 +926,6 @@ pub const all_features = blk: {
         .description = "Slow unaligned 32-byte memory access",
         .dependencies = 0,
     };
-
     result[@enumToInt(Feature.soft_float)] = .{
         .index = @enumToInt(Feature.soft_float),
         .name = @tagName(Feature.soft_float),
@@ -1051,95 +933,101 @@ pub const all_features = blk: {
         .description = "Use software floating point features",
         .dependencies = 0,
     };
-
-    result[@enumToInt(Feature.tbm)] = .{
-        .index = @enumToInt(Feature.tbm),
-        .name = @tagName(Feature.tbm),
-        .llvm_name = "tbm",
-        .description = "Enable TBM instructions",
+    result[@enumToInt(Feature.sse)] = .{
+        .index = @enumToInt(Feature.sse),
+        .name = @tagName(Feature.sse),
+        .llvm_name = "sse",
+        .description = "Enable SSE instructions",
         .dependencies = 0,
     };
-
-    result[@enumToInt(Feature.vaes)] = .{
-        .index = @enumToInt(Feature.vaes),
-        .name = @tagName(Feature.vaes),
-        .llvm_name = "vaes",
-        .description = "Promote selected AES instructions to AVX512/AVX registers",
+    result[@enumToInt(Feature.sse_unaligned_mem)] = .{
+        .index = @enumToInt(Feature.sse_unaligned_mem),
+        .name = @tagName(Feature.sse_unaligned_mem),
+        .llvm_name = "sse-unaligned-mem",
+        .description = "Allow unaligned memory operands with SSE instructions",
+        .dependencies = 0,
+    };
+    result[@enumToInt(Feature.sse2)] = .{
+        .index = @enumToInt(Feature.sse2),
+        .name = @tagName(Feature.sse2),
+        .llvm_name = "sse2",
+        .description = "Enable SSE2 instructions",
         .dependencies = featureSet(&[_]Feature{
             .sse,
         }),
     };
-
-    result[@enumToInt(Feature.avx512vbmi)] = .{
-        .index = @enumToInt(Feature.avx512vbmi),
-        .name = @tagName(Feature.avx512vbmi),
-        .llvm_name = "avx512vbmi",
-        .description = "Enable AVX-512 Vector Byte Manipulation Instructions",
+    result[@enumToInt(Feature.sse3)] = .{
+        .index = @enumToInt(Feature.sse3),
+        .name = @tagName(Feature.sse3),
+        .llvm_name = "sse3",
+        .description = "Enable SSE3 instructions",
         .dependencies = featureSet(&[_]Feature{
-            .sse,
+            .sse2,
         }),
     };
-
-    result[@enumToInt(Feature.avx512vbmi2)] = .{
-        .index = @enumToInt(Feature.avx512vbmi2),
-        .name = @tagName(Feature.avx512vbmi2),
-        .llvm_name = "avx512vbmi2",
-        .description = "Enable AVX-512 further Vector Byte Manipulation Instructions",
+    result[@enumToInt(Feature.sse4_1)] = .{
+        .index = @enumToInt(Feature.sse4_1),
+        .name = @tagName(Feature.sse4_1),
+        .llvm_name = "sse4.1",
+        .description = "Enable SSE 4.1 instructions",
         .dependencies = featureSet(&[_]Feature{
-            .sse,
+            .ssse3,
         }),
     };
-
-    result[@enumToInt(Feature.avx512vl)] = .{
-        .index = @enumToInt(Feature.avx512vl),
-        .name = @tagName(Feature.avx512vl),
-        .llvm_name = "avx512vl",
-        .description = "Enable AVX-512 Vector Length eXtensions",
+    result[@enumToInt(Feature.sse4_2)] = .{
+        .index = @enumToInt(Feature.sse4_2),
+        .name = @tagName(Feature.sse4_2),
+        .llvm_name = "sse4.2",
+        .description = "Enable SSE 4.2 instructions",
         .dependencies = featureSet(&[_]Feature{
-            .sse,
+            .sse4_1,
         }),
     };
-
-    result[@enumToInt(Feature.avx512vnni)] = .{
-        .index = @enumToInt(Feature.avx512vnni),
-        .name = @tagName(Feature.avx512vnni),
-        .llvm_name = "avx512vnni",
-        .description = "Enable AVX-512 Vector Neural Network Instructions",
+    result[@enumToInt(Feature.sse4a)] = .{
+        .index = @enumToInt(Feature.sse4a),
+        .name = @tagName(Feature.sse4a),
+        .llvm_name = "sse4a",
+        .description = "Support SSE 4a instructions",
         .dependencies = featureSet(&[_]Feature{
-            .sse,
+            .sse3,
         }),
     };
-
-    result[@enumToInt(Feature.avx512vp2intersect)] = .{
-        .index = @enumToInt(Feature.avx512vp2intersect),
-        .name = @tagName(Feature.avx512vp2intersect),
-        .llvm_name = "avx512vp2intersect",
-        .description = "Enable AVX-512 vp2intersect",
+    result[@enumToInt(Feature.ssse3)] = .{
+        .index = @enumToInt(Feature.ssse3),
+        .name = @tagName(Feature.ssse3),
+        .llvm_name = "ssse3",
+        .description = "Enable SSSE3 instructions",
         .dependencies = featureSet(&[_]Feature{
-            .sse,
+            .sse3,
+        }),
+    };
+    result[@enumToInt(Feature.tbm)] = .{
+        .index = @enumToInt(Feature.tbm),
+        .name = @tagName(Feature.tbm),
+        .llvm_name = "tbm",
+        .description = "Enable TBM instructions",
+        .dependencies = 0,
+    };
+    result[@enumToInt(Feature.vaes)] = .{
+        .index = @enumToInt(Feature.vaes),
+        .name = @tagName(Feature.vaes),
+        .llvm_name = "vaes",
+        .description = "Promote selected AES instructions to AVX512/AVX registers",
+        .dependencies = featureSet(&[_]Feature{
+            .aes,
+            .avx,
         }),
     };
-
     result[@enumToInt(Feature.vpclmulqdq)] = .{
         .index = @enumToInt(Feature.vpclmulqdq),
         .name = @tagName(Feature.vpclmulqdq),
         .llvm_name = "vpclmulqdq",
         .description = "Enable vpclmulqdq instructions",
         .dependencies = featureSet(&[_]Feature{
-            .sse,
-        }),
-    };
-
-    result[@enumToInt(Feature.avx512vpopcntdq)] = .{
-        .index = @enumToInt(Feature.avx512vpopcntdq),
-        .name = @tagName(Feature.avx512vpopcntdq),
-        .llvm_name = "avx512vpopcntdq",
-        .description = "Enable AVX-512 Population Count Instructions",
-        .dependencies = featureSet(&[_]Feature{
-            .sse,
+            .avx,
+            .pclmul,
         }),
     };
-
     result[@enumToInt(Feature.waitpkg)] = .{
         .index = @enumToInt(Feature.waitpkg),
         .name = @tagName(Feature.waitpkg),
@@ -1147,7 +1035,6 @@ pub const all_features = blk: {
         .description = "Wait and pause enhancements",
         .dependencies = 0,
     };
-
     result[@enumToInt(Feature.wbnoinvd)] = .{
         .index = @enumToInt(Feature.wbnoinvd),
         .name = @tagName(Feature.wbnoinvd),
@@ -1155,7 +1042,6 @@ pub const all_features = blk: {
         .description = "Write Back No Invalidate",
         .dependencies = 0,
     };
-
     result[@enumToInt(Feature.x87)] = .{
         .index = @enumToInt(Feature.x87),
         .name = @tagName(Feature.x87),
@@ -1163,17 +1049,15 @@ pub const all_features = blk: {
         .description = "Enable X87 float instructions",
         .dependencies = 0,
     };
-
     result[@enumToInt(Feature.xop)] = .{
         .index = @enumToInt(Feature.xop),
         .name = @tagName(Feature.xop),
         .llvm_name = "xop",
         .description = "Enable XOP instructions",
         .dependencies = featureSet(&[_]Feature{
-            .sse,
+            .fma4,
         }),
     };
-
     result[@enumToInt(Feature.xsave)] = .{
         .index = @enumToInt(Feature.xsave),
         .name = @tagName(Feature.xsave),
@@ -1181,7 +1065,6 @@ pub const all_features = blk: {
         .description = "Support xsave instructions",
         .dependencies = 0,
     };
-
     result[@enumToInt(Feature.xsavec)] = .{
         .index = @enumToInt(Feature.xsavec),
         .name = @tagName(Feature.xsavec),
@@ -1189,7 +1072,6 @@ pub const all_features = blk: {
         .description = "Support xsavec instructions",
         .dependencies = 0,
     };
-
     result[@enumToInt(Feature.xsaveopt)] = .{
         .index = @enumToInt(Feature.xsaveopt),
         .name = @tagName(Feature.xsaveopt),
@@ -1197,7 +1079,6 @@ pub const all_features = blk: {
         .description = "Support xsaveopt instructions",
         .dependencies = 0,
     };
-
     result[@enumToInt(Feature.xsaves)] = .{
         .index = @enumToInt(Feature.xsaves),
         .name = @tagName(Feature.xsaves),
@@ -1205,7 +1086,6 @@ pub const all_features = blk: {
         .description = "Support xsaves instructions",
         .dependencies = 0,
     };
-
     break :blk result;
 };
 
@@ -1214,30 +1094,26 @@ pub const cpu = struct {
         .name = "amdfam10",
         .llvm_name = "amdfam10",
         .features = featureSet(&[_]Feature{
-            .mmx,
             .@"3dnowa",
             .@"64bit",
             .cmov,
-            .cx8,
             .cx16,
-            .fxsr,
+            .cx8,
             .fast_scalar_shift_masks,
-            .sahf,
+            .fxsr,
             .lzcnt,
             .nopl,
             .popcnt,
-            .sse,
-            .sse4a,
+            .sahf,
             .slow_shld,
+            .sse4a,
             .x87,
         }),
     };
-
     pub const athlon = Cpu{
         .name = "athlon",
         .llvm_name = "athlon",
         .features = featureSet(&[_]Feature{
-            .mmx,
             .@"3dnowa",
             .cmov,
             .cx8,
@@ -1247,66 +1123,57 @@ pub const cpu = struct {
             .x87,
         }),
     };
-
-    pub const athlon4 = Cpu{
+    pub const athlon_4 = Cpu{
         .name = "athlon_4",
         .llvm_name = "athlon-4",
         .features = featureSet(&[_]Feature{
-            .mmx,
             .@"3dnowa",
             .cmov,
             .cx8,
             .fxsr,
             .nopl,
-            .sse,
             .slow_shld,
             .slow_unaligned_mem_16,
+            .sse,
             .x87,
         }),
     };
-
     pub const athlon_fx = Cpu{
         .name = "athlon_fx",
         .llvm_name = "athlon-fx",
         .features = featureSet(&[_]Feature{
-            .mmx,
             .@"3dnowa",
             .@"64bit",
             .cmov,
             .cx8,
-            .fxsr,
             .fast_scalar_shift_masks,
+            .fxsr,
             .nopl,
-            .sse,
-            .sse2,
             .slow_shld,
             .slow_unaligned_mem_16,
+            .sse2,
             .x87,
         }),
     };
-
     pub const athlon_mp = Cpu{
         .name = "athlon_mp",
         .llvm_name = "athlon-mp",
         .features = featureSet(&[_]Feature{
-            .mmx,
             .@"3dnowa",
             .cmov,
             .cx8,
             .fxsr,
             .nopl,
-            .sse,
             .slow_shld,
             .slow_unaligned_mem_16,
+            .sse,
             .x87,
         }),
     };
-
     pub const athlon_tbird = Cpu{
         .name = "athlon_tbird",
         .llvm_name = "athlon-tbird",
         .features = featureSet(&[_]Feature{
-            .mmx,
             .@"3dnowa",
             .cmov,
             .cx8,
@@ -1316,129 +1183,113 @@ pub const cpu = struct {
             .x87,
         }),
     };
-
     pub const athlon_xp = Cpu{
         .name = "athlon_xp",
         .llvm_name = "athlon-xp",
         .features = featureSet(&[_]Feature{
-            .mmx,
             .@"3dnowa",
             .cmov,
             .cx8,
             .fxsr,
             .nopl,
-            .sse,
             .slow_shld,
             .slow_unaligned_mem_16,
+            .sse,
             .x87,
         }),
     };
-
     pub const athlon64 = Cpu{
         .name = "athlon64",
         .llvm_name = "athlon64",
         .features = featureSet(&[_]Feature{
-            .mmx,
             .@"3dnowa",
             .@"64bit",
             .cmov,
             .cx8,
-            .fxsr,
             .fast_scalar_shift_masks,
+            .fxsr,
             .nopl,
-            .sse,
-            .sse2,
             .slow_shld,
             .slow_unaligned_mem_16,
+            .sse2,
             .x87,
         }),
     };
-
     pub const athlon64_sse3 = Cpu{
         .name = "athlon64_sse3",
         .llvm_name = "athlon64-sse3",
         .features = featureSet(&[_]Feature{
-            .mmx,
             .@"3dnowa",
             .@"64bit",
             .cmov,
-            .cx8,
             .cx16,
-            .fxsr,
+            .cx8,
             .fast_scalar_shift_masks,
+            .fxsr,
             .nopl,
-            .sse,
-            .sse3,
             .slow_shld,
             .slow_unaligned_mem_16,
+            .sse3,
             .x87,
         }),
     };
-
     pub const atom = Cpu{
         .name = "atom",
         .llvm_name = "atom",
         .features = featureSet(&[_]Feature{
             .@"64bit",
             .cmov,
-            .cx8,
             .cx16,
+            .cx8,
             .fxsr,
-            .sahf,
+            .idivl_to_divb,
+            .idivq_to_divl,
             .lea_sp,
             .lea_uses_ag,
             .mmx,
             .movbe,
             .nopl,
             .pad_short_functions,
-            .sse,
-            .ssse3,
-            .idivl_to_divb,
-            .idivq_to_divl,
+            .sahf,
             .slow_two_mem_ops,
             .slow_unaligned_mem_16,
+            .ssse3,
             .x87,
         }),
     };
-
     pub const barcelona = Cpu{
         .name = "barcelona",
         .llvm_name = "barcelona",
         .features = featureSet(&[_]Feature{
-            .mmx,
             .@"3dnowa",
             .@"64bit",
             .cmov,
-            .cx8,
             .cx16,
-            .fxsr,
+            .cx8,
             .fast_scalar_shift_masks,
-            .sahf,
+            .fxsr,
             .lzcnt,
             .nopl,
             .popcnt,
-            .sse,
-            .sse4a,
+            .sahf,
             .slow_shld,
+            .sse4a,
             .x87,
         }),
     };
-
     pub const bdver1 = Cpu{
         .name = "bdver1",
         .llvm_name = "bdver1",
         .features = featureSet(&[_]Feature{
             .@"64bit",
-            .sse,
             .aes,
             .branchfusion,
             .cmov,
-            .cx8,
             .cx16,
-            .fxsr,
+            .cx8,
             .fast_11bytenop,
             .fast_scalar_shift_masks,
-            .sahf,
+            .fxsr,
             .lwp,
             .lzcnt,
             .mmx,
@@ -1446,32 +1297,30 @@ pub const cpu = struct {
             .pclmul,
             .popcnt,
             .prfchw,
+            .sahf,
             .slow_shld,
             .x87,
             .xop,
             .xsave,
         }),
     };
-
     pub const bdver2 = Cpu{
         .name = "bdver2",
         .llvm_name = "bdver2",
         .features = featureSet(&[_]Feature{
             .@"64bit",
-            .sse,
             .aes,
             .bmi,
             .branchfusion,
             .cmov,
-            .cx8,
             .cx16,
+            .cx8,
             .f16c,
-            .fma,
-            .fxsr,
             .fast_11bytenop,
             .fast_bextr,
             .fast_scalar_shift_masks,
-            .sahf,
+            .fma,
+            .fxsr,
             .lwp,
             .lzcnt,
             .mmx,
@@ -1479,6 +1328,7 @@ pub const cpu = struct {
             .pclmul,
             .popcnt,
             .prfchw,
+            .sahf,
             .slow_shld,
             .tbm,
             .x87,
@@ -1486,27 +1336,24 @@ pub const cpu = struct {
             .xsave,
         }),
     };
-
     pub const bdver3 = Cpu{
         .name = "bdver3",
         .llvm_name = "bdver3",
         .features = featureSet(&[_]Feature{
             .@"64bit",
-            .sse,
             .aes,
             .bmi,
             .branchfusion,
             .cmov,
-            .cx8,
             .cx16,
+            .cx8,
             .f16c,
-            .fma,
-            .fsgsbase,
-            .fxsr,
             .fast_11bytenop,
             .fast_bextr,
             .fast_scalar_shift_masks,
-            .sahf,
+            .fma,
+            .fsgsbase,
+            .fxsr,
             .lwp,
             .lzcnt,
             .mmx,
@@ -1514,6 +1361,7 @@ pub const cpu = struct {
             .pclmul,
             .popcnt,
             .prfchw,
+            .sahf,
             .slow_shld,
             .tbm,
             .x87,
@@ -1522,29 +1370,26 @@ pub const cpu = struct {
             .xsaveopt,
         }),
     };
-
     pub const bdver4 = Cpu{
         .name = "bdver4",
         .llvm_name = "bdver4",
         .features = featureSet(&[_]Feature{
             .@"64bit",
-            .sse,
             .aes,
             .avx2,
             .bmi,
             .bmi2,
             .branchfusion,
             .cmov,
-            .cx8,
             .cx16,
+            .cx8,
             .f16c,
-            .fma,
-            .fsgsbase,
-            .fxsr,
             .fast_11bytenop,
             .fast_bextr,
             .fast_scalar_shift_masks,
-            .sahf,
+            .fma,
+            .fsgsbase,
+            .fxsr,
             .lwp,
             .lzcnt,
             .mmx,
@@ -1553,6 +1398,7 @@ pub const cpu = struct {
             .pclmul,
             .popcnt,
             .prfchw,
+            .sahf,
             .slow_shld,
             .tbm,
             .x87,
@@ -1561,119 +1407,110 @@ pub const cpu = struct {
             .xsaveopt,
         }),
     };
-
     pub const bonnell = Cpu{
         .name = "bonnell",
         .llvm_name = "bonnell",
         .features = featureSet(&[_]Feature{
             .@"64bit",
             .cmov,
-            .cx8,
             .cx16,
+            .cx8,
             .fxsr,
-            .sahf,
+            .idivl_to_divb,
+            .idivq_to_divl,
             .lea_sp,
             .lea_uses_ag,
             .mmx,
             .movbe,
             .nopl,
             .pad_short_functions,
-            .sse,
-            .ssse3,
-            .idivl_to_divb,
-            .idivq_to_divl,
+            .sahf,
             .slow_two_mem_ops,
             .slow_unaligned_mem_16,
+            .ssse3,
             .x87,
         }),
     };
-
     pub const broadwell = Cpu{
         .name = "broadwell",
         .llvm_name = "broadwell",
         .features = featureSet(&[_]Feature{
             .@"64bit",
             .adx,
-            .sse,
             .avx,
             .avx2,
             .bmi,
             .bmi2,
             .cmov,
-            .cx8,
             .cx16,
+            .cx8,
             .ermsb,
             .f16c,
+            .false_deps_lzcnt_tzcnt,
+            .false_deps_popcnt,
+            .fast_scalar_fsqrt,
+            .fast_shld_rotate,
+            .fast_variable_shuffle,
             .fma,
             .fsgsbase,
             .fxsr,
-            .fast_shld_rotate,
-            .fast_scalar_fsqrt,
-            .fast_variable_shuffle,
+            .idivq_to_divl,
             .invpcid,
-            .sahf,
             .lzcnt,
-            .false_deps_lzcnt_tzcnt,
-            .mmx,
-            .movbe,
             .macrofusion,
             .merge_to_threeway_branch,
+            .mmx,
+            .movbe,
             .nopl,
             .pclmul,
             .popcnt,
-            .false_deps_popcnt,
             .prfchw,
             .rdrnd,
             .rdseed,
-            .sse4_2,
+            .sahf,
             .slow_3ops_lea,
-            .idivq_to_divl,
+            .sse4_2,
             .x87,
             .xsave,
             .xsaveopt,
         }),
     };
-
     pub const btver1 = Cpu{
         .name = "btver1",
         .llvm_name = "btver1",
         .features = featureSet(&[_]Feature{
             .@"64bit",
             .cmov,
-            .cx8,
             .cx16,
-            .fxsr,
+            .cx8,
             .fast_15bytenop,
             .fast_scalar_shift_masks,
             .fast_vector_shift_masks,
-            .sahf,
+            .fxsr,
             .lzcnt,
             .mmx,
             .nopl,
             .popcnt,
             .prfchw,
-            .sse,
+            .sahf,
+            .slow_shld,
             .sse4a,
             .ssse3,
-            .slow_shld,
             .x87,
         }),
     };
-
     pub const btver2 = Cpu{
         .name = "btver2",
         .llvm_name = "btver2",
         .features = featureSet(&[_]Feature{
             .@"64bit",
-            .sse,
             .aes,
             .avx,
             .bmi,
             .cmov,
-            .cx8,
             .cx16,
+            .cx8,
             .f16c,
-            .fxsr,
             .fast_15bytenop,
             .fast_bextr,
             .fast_hops,
@@ -1681,7 +1518,7 @@ pub const cpu = struct {
             .fast_partial_ymm_or_zmm_write,
             .fast_scalar_shift_masks,
             .fast_vector_shift_masks,
-            .sahf,
+            .fxsr,
             .lzcnt,
             .mmx,
             .movbe,
@@ -1689,27 +1526,25 @@ pub const cpu = struct {
             .pclmul,
             .popcnt,
             .prfchw,
+            .sahf,
+            .slow_shld,
             .sse4a,
             .ssse3,
-            .slow_shld,
             .x87,
             .xsave,
             .xsaveopt,
         }),
     };
-
     pub const c3 = Cpu{
         .name = "c3",
         .llvm_name = "c3",
         .features = featureSet(&[_]Feature{
-            .mmx,
             .@"3dnow",
             .slow_unaligned_mem_16,
             .x87,
         }),
     };
-
-    pub const c32 = Cpu{
+    pub const c3_2 = Cpu{
         .name = "c3_2",
         .llvm_name = "c3-2",
         .features = featureSet(&[_]Feature{
@@ -1717,51 +1552,51 @@ pub const cpu = struct {
             .cx8,
             .fxsr,
             .mmx,
-            .sse,
             .slow_unaligned_mem_16,
+            .sse,
             .x87,
         }),
     };
-
     pub const cannonlake = Cpu{
         .name = "cannonlake",
         .llvm_name = "cannonlake",
         .features = featureSet(&[_]Feature{
             .@"64bit",
             .adx,
-            .sse,
             .aes,
             .avx,
             .avx2,
+            .avx512bw,
+            .avx512cd,
+            .avx512dq,
             .avx512f,
+            .avx512ifma,
+            .avx512vbmi,
+            .avx512vl,
             .bmi,
             .bmi2,
-            .avx512bw,
-            .avx512cd,
             .clflushopt,
             .cmov,
-            .cx8,
             .cx16,
-            .avx512dq,
+            .cx8,
             .ermsb,
             .f16c,
-            .fma,
-            .fsgsbase,
-            .fxsr,
-            .fast_shld_rotate,
+            .fast_gather,
             .fast_scalar_fsqrt,
+            .fast_shld_rotate,
             .fast_variable_shuffle,
             .fast_vector_fsqrt,
-            .fast_gather,
-            .avx512ifma,
+            .fma,
+            .fsgsbase,
+            .fxsr,
+            .idivq_to_divl,
             .invpcid,
-            .sahf,
             .lzcnt,
+            .macrofusion,
+            .merge_to_threeway_branch,
             .mmx,
             .movbe,
             .mpx,
-            .macrofusion,
-            .merge_to_threeway_branch,
             .nopl,
             .pclmul,
             .pku,
@@ -1769,13 +1604,11 @@ pub const cpu = struct {
             .prfchw,
             .rdrnd,
             .rdseed,
+            .sahf,
             .sgx,
             .sha,
-            .sse4_2,
             .slow_3ops_lea,
-            .idivq_to_divl,
-            .avx512vbmi,
-            .avx512vl,
+            .sse4_2,
             .x87,
             .xsave,
             .xsavec,
@@ -1783,59 +1616,57 @@ pub const cpu = struct {
             .xsaves,
         }),
     };
-
     pub const cascadelake = Cpu{
         .name = "cascadelake",
         .llvm_name = "cascadelake",
         .features = featureSet(&[_]Feature{
             .@"64bit",
             .adx,
-            .sse,
             .aes,
             .avx,
             .avx2,
+            .avx512bw,
+            .avx512cd,
+            .avx512dq,
             .avx512f,
+            .avx512vl,
+            .avx512vnni,
             .bmi,
             .bmi2,
-            .avx512bw,
-            .avx512cd,
             .clflushopt,
             .clwb,
             .cmov,
-            .cx8,
             .cx16,
-            .avx512dq,
+            .cx8,
             .ermsb,
             .f16c,
-            .fma,
-            .fsgsbase,
-            .fxsr,
-            .fast_shld_rotate,
+            .false_deps_popcnt,
+            .fast_gather,
             .fast_scalar_fsqrt,
+            .fast_shld_rotate,
             .fast_variable_shuffle,
             .fast_vector_fsqrt,
-            .fast_gather,
+            .fma,
+            .fsgsbase,
+            .fxsr,
+            .idivq_to_divl,
             .invpcid,
-            .sahf,
             .lzcnt,
+            .macrofusion,
+            .merge_to_threeway_branch,
             .mmx,
             .movbe,
             .mpx,
-            .macrofusion,
-            .merge_to_threeway_branch,
             .nopl,
             .pclmul,
             .pku,
             .popcnt,
-            .false_deps_popcnt,
             .prfchw,
             .rdrnd,
             .rdseed,
-            .sse4_2,
+            .sahf,
             .slow_3ops_lea,
-            .idivq_to_divl,
-            .avx512vl,
-            .avx512vnni,
+            .sse4_2,
             .x87,
             .xsave,
             .xsavec,
@@ -1843,60 +1674,58 @@ pub const cpu = struct {
             .xsaves,
         }),
     };
-
     pub const cooperlake = Cpu{
         .name = "cooperlake",
         .llvm_name = "cooperlake",
         .features = featureSet(&[_]Feature{
             .@"64bit",
             .adx,
-            .sse,
             .aes,
             .avx,
             .avx2,
-            .avx512f,
             .avx512bf16,
-            .bmi,
-            .bmi2,
             .avx512bw,
             .avx512cd,
+            .avx512dq,
+            .avx512f,
+            .avx512vl,
+            .avx512vnni,
+            .bmi,
+            .bmi2,
             .clflushopt,
             .clwb,
             .cmov,
-            .cx8,
             .cx16,
-            .avx512dq,
+            .cx8,
             .ermsb,
             .f16c,
-            .fma,
-            .fsgsbase,
-            .fxsr,
-            .fast_shld_rotate,
+            .false_deps_popcnt,
+            .fast_gather,
             .fast_scalar_fsqrt,
+            .fast_shld_rotate,
             .fast_variable_shuffle,
             .fast_vector_fsqrt,
-            .fast_gather,
+            .fma,
+            .fsgsbase,
+            .fxsr,
+            .idivq_to_divl,
             .invpcid,
-            .sahf,
             .lzcnt,
+            .macrofusion,
+            .merge_to_threeway_branch,
             .mmx,
             .movbe,
             .mpx,
-            .macrofusion,
-            .merge_to_threeway_branch,
             .nopl,
             .pclmul,
             .pku,
             .popcnt,
-            .false_deps_popcnt,
             .prfchw,
             .rdrnd,
             .rdseed,
-            .sse4_2,
+            .sahf,
             .slow_3ops_lea,
-            .idivq_to_divl,
-            .avx512vl,
-            .avx512vnni,
+            .sse4_2,
             .x87,
             .xsave,
             .xsavec,
@@ -1904,155 +1733,144 @@ pub const cpu = struct {
             .xsaves,
         }),
     };
-
     pub const core_avx_i = Cpu{
         .name = "core_avx_i",
         .llvm_name = "core-avx-i",
         .features = featureSet(&[_]Feature{
             .@"64bit",
-            .sse,
             .avx,
             .cmov,
-            .cx8,
             .cx16,
+            .cx8,
             .f16c,
+            .false_deps_popcnt,
+            .fast_scalar_fsqrt,
+            .fast_shld_rotate,
             .fsgsbase,
             .fxsr,
-            .fast_shld_rotate,
-            .fast_scalar_fsqrt,
-            .sahf,
-            .mmx,
+            .idivq_to_divl,
             .macrofusion,
             .merge_to_threeway_branch,
+            .mmx,
             .nopl,
             .pclmul,
             .popcnt,
-            .false_deps_popcnt,
             .rdrnd,
-            .sse4_2,
+            .sahf,
             .slow_3ops_lea,
-            .idivq_to_divl,
             .slow_unaligned_mem_32,
+            .sse4_2,
             .x87,
             .xsave,
             .xsaveopt,
         }),
     };
-
     pub const core_avx2 = Cpu{
         .name = "core_avx2",
         .llvm_name = "core-avx2",
         .features = featureSet(&[_]Feature{
             .@"64bit",
-            .sse,
             .avx,
             .avx2,
             .bmi,
             .bmi2,
             .cmov,
-            .cx8,
             .cx16,
+            .cx8,
             .ermsb,
             .f16c,
+            .false_deps_lzcnt_tzcnt,
+            .false_deps_popcnt,
+            .fast_scalar_fsqrt,
+            .fast_shld_rotate,
+            .fast_variable_shuffle,
             .fma,
             .fsgsbase,
             .fxsr,
-            .fast_shld_rotate,
-            .fast_scalar_fsqrt,
-            .fast_variable_shuffle,
+            .idivq_to_divl,
             .invpcid,
-            .sahf,
             .lzcnt,
-            .false_deps_lzcnt_tzcnt,
-            .mmx,
-            .movbe,
             .macrofusion,
             .merge_to_threeway_branch,
+            .mmx,
+            .movbe,
             .nopl,
             .pclmul,
             .popcnt,
-            .false_deps_popcnt,
             .rdrnd,
-            .sse4_2,
+            .sahf,
             .slow_3ops_lea,
-            .idivq_to_divl,
+            .sse4_2,
             .x87,
             .xsave,
             .xsaveopt,
         }),
     };
-
     pub const core2 = Cpu{
         .name = "core2",
         .llvm_name = "core2",
         .features = featureSet(&[_]Feature{
             .@"64bit",
             .cmov,
-            .cx8,
             .cx16,
+            .cx8,
             .fxsr,
-            .sahf,
-            .mmx,
             .macrofusion,
+            .mmx,
             .nopl,
-            .sse,
-            .ssse3,
+            .sahf,
             .slow_unaligned_mem_16,
+            .ssse3,
             .x87,
         }),
     };
-
     pub const corei7 = Cpu{
         .name = "corei7",
         .llvm_name = "corei7",
         .features = featureSet(&[_]Feature{
             .@"64bit",
             .cmov,
-            .cx8,
             .cx16,
+            .cx8,
             .fxsr,
-            .sahf,
-            .mmx,
             .macrofusion,
+            .mmx,
             .nopl,
             .popcnt,
-            .sse,
+            .sahf,
             .sse4_2,
             .x87,
         }),
     };
-
     pub const corei7_avx = Cpu{
         .name = "corei7_avx",
         .llvm_name = "corei7-avx",
         .features = featureSet(&[_]Feature{
             .@"64bit",
-            .sse,
             .avx,
             .cmov,
-            .cx8,
             .cx16,
-            .fxsr,
-            .fast_shld_rotate,
+            .cx8,
+            .false_deps_popcnt,
             .fast_scalar_fsqrt,
-            .sahf,
-            .mmx,
+            .fast_shld_rotate,
+            .fxsr,
+            .idivq_to_divl,
             .macrofusion,
             .merge_to_threeway_branch,
+            .mmx,
             .nopl,
             .pclmul,
             .popcnt,
-            .false_deps_popcnt,
-            .sse4_2,
+            .sahf,
             .slow_3ops_lea,
-            .idivq_to_divl,
             .slow_unaligned_mem_32,
+            .sse4_2,
             .x87,
             .xsave,
             .xsaveopt,
         }),
     };
-
     pub const generic = Cpu{
         .name = "generic",
         .llvm_name = "generic",
@@ -2062,49 +1880,45 @@ pub const cpu = struct {
             .x87,
         }),
     };
-
     pub const geode = Cpu{
         .name = "geode",
         .llvm_name = "geode",
         .features = featureSet(&[_]Feature{
-            .mmx,
             .@"3dnowa",
             .cx8,
             .slow_unaligned_mem_16,
             .x87,
         }),
     };
-
     pub const goldmont = Cpu{
         .name = "goldmont",
         .llvm_name = "goldmont",
         .features = featureSet(&[_]Feature{
             .@"64bit",
-            .sse,
             .aes,
             .clflushopt,
             .cmov,
-            .cx8,
             .cx16,
+            .cx8,
+            .false_deps_popcnt,
             .fsgsbase,
             .fxsr,
-            .sahf,
             .mmx,
             .movbe,
             .mpx,
             .nopl,
             .pclmul,
             .popcnt,
-            .false_deps_popcnt,
             .prfchw,
             .rdrnd,
             .rdseed,
+            .sahf,
             .sha,
-            .sse4_2,
-            .ssse3,
             .slow_incdec,
             .slow_lea,
             .slow_two_mem_ops,
+            .sse4_2,
+            .ssse3,
             .x87,
             .xsave,
             .xsavec,
@@ -2112,21 +1926,18 @@ pub const cpu = struct {
             .xsaves,
         }),
     };
-
     pub const goldmont_plus = Cpu{
         .name = "goldmont_plus",
         .llvm_name = "goldmont-plus",
         .features = featureSet(&[_]Feature{
             .@"64bit",
-            .sse,
             .aes,
             .clflushopt,
             .cmov,
-            .cx8,
             .cx16,
+            .cx8,
             .fsgsbase,
             .fxsr,
-            .sahf,
             .mmx,
             .movbe,
             .mpx,
@@ -2138,13 +1949,14 @@ pub const cpu = struct {
             .rdpid,
             .rdrnd,
             .rdseed,
+            .sahf,
             .sgx,
             .sha,
-            .sse4_2,
-            .ssse3,
             .slow_incdec,
             .slow_lea,
             .slow_two_mem_ops,
+            .sse4_2,
+            .ssse3,
             .x87,
             .xsave,
             .xsavec,
@@ -2152,50 +1964,47 @@ pub const cpu = struct {
             .xsaves,
         }),
     };
-
     pub const haswell = Cpu{
         .name = "haswell",
         .llvm_name = "haswell",
         .features = featureSet(&[_]Feature{
             .@"64bit",
-            .sse,
             .avx,
             .avx2,
             .bmi,
             .bmi2,
             .cmov,
-            .cx8,
             .cx16,
+            .cx8,
             .ermsb,
             .f16c,
+            .false_deps_lzcnt_tzcnt,
+            .false_deps_popcnt,
+            .fast_scalar_fsqrt,
+            .fast_shld_rotate,
+            .fast_variable_shuffle,
             .fma,
             .fsgsbase,
             .fxsr,
-            .fast_shld_rotate,
-            .fast_scalar_fsqrt,
-            .fast_variable_shuffle,
+            .idivq_to_divl,
             .invpcid,
-            .sahf,
             .lzcnt,
-            .false_deps_lzcnt_tzcnt,
-            .mmx,
-            .movbe,
             .macrofusion,
             .merge_to_threeway_branch,
+            .mmx,
+            .movbe,
             .nopl,
             .pclmul,
             .popcnt,
-            .false_deps_popcnt,
             .rdrnd,
-            .sse4_2,
+            .sahf,
             .slow_3ops_lea,
-            .idivq_to_divl,
+            .sse4_2,
             .x87,
             .xsave,
             .xsaveopt,
         }),
     };
-
     pub const _i386 = Cpu{
         .name = "_i386",
         .llvm_name = "i386",
@@ -2204,7 +2013,6 @@ pub const cpu = struct {
             .x87,
         }),
     };
-
     pub const _i486 = Cpu{
         .name = "_i486",
         .llvm_name = "i486",
@@ -2213,7 +2021,6 @@ pub const cpu = struct {
             .x87,
         }),
     };
-
     pub const _i586 = Cpu{
         .name = "_i586",
         .llvm_name = "i586",
@@ -2223,7 +2030,6 @@ pub const cpu = struct {
             .x87,
         }),
     };
-
     pub const _i686 = Cpu{
         .name = "_i686",
         .llvm_name = "i686",
@@ -2234,49 +2040,52 @@ pub const cpu = struct {
             .x87,
         }),
     };
-
     pub const icelake_client = Cpu{
         .name = "icelake_client",
         .llvm_name = "icelake-client",
         .features = featureSet(&[_]Feature{
             .@"64bit",
             .adx,
-            .sse,
             .aes,
             .avx,
             .avx2,
-            .avx512f,
             .avx512bitalg,
-            .bmi,
-            .bmi2,
             .avx512bw,
             .avx512cd,
+            .avx512dq,
+            .avx512f,
+            .avx512ifma,
+            .avx512vbmi,
+            .avx512vbmi2,
+            .avx512vl,
+            .avx512vnni,
+            .avx512vpopcntdq,
+            .bmi,
+            .bmi2,
             .clflushopt,
             .clwb,
             .cmov,
-            .cx8,
             .cx16,
-            .avx512dq,
+            .cx8,
             .ermsb,
             .f16c,
-            .fma,
-            .fsgsbase,
-            .fxsr,
-            .fast_shld_rotate,
+            .fast_gather,
             .fast_scalar_fsqrt,
+            .fast_shld_rotate,
             .fast_variable_shuffle,
             .fast_vector_fsqrt,
+            .fma,
+            .fsgsbase,
+            .fxsr,
             .gfni,
-            .fast_gather,
-            .avx512ifma,
+            .idivq_to_divl,
             .invpcid,
-            .sahf,
             .lzcnt,
+            .macrofusion,
+            .merge_to_threeway_branch,
             .mmx,
             .movbe,
             .mpx,
-            .macrofusion,
-            .merge_to_threeway_branch,
             .nopl,
             .pclmul,
             .pku,
@@ -2285,18 +2094,13 @@ pub const cpu = struct {
             .rdpid,
             .rdrnd,
             .rdseed,
+            .sahf,
             .sgx,
             .sha,
-            .sse4_2,
             .slow_3ops_lea,
-            .idivq_to_divl,
+            .sse4_2,
             .vaes,
-            .avx512vbmi,
-            .avx512vbmi2,
-            .avx512vl,
-            .avx512vnni,
             .vpclmulqdq,
-            .avx512vpopcntdq,
             .x87,
             .xsave,
             .xsavec,
@@ -2304,49 +2108,52 @@ pub const cpu = struct {
             .xsaves,
         }),
     };
-
     pub const icelake_server = Cpu{
         .name = "icelake_server",
         .llvm_name = "icelake-server",
         .features = featureSet(&[_]Feature{
             .@"64bit",
             .adx,
-            .sse,
             .aes,
             .avx,
             .avx2,
-            .avx512f,
             .avx512bitalg,
-            .bmi,
-            .bmi2,
             .avx512bw,
             .avx512cd,
+            .avx512dq,
+            .avx512f,
+            .avx512ifma,
+            .avx512vbmi,
+            .avx512vbmi2,
+            .avx512vl,
+            .avx512vnni,
+            .avx512vpopcntdq,
+            .bmi,
+            .bmi2,
             .clflushopt,
             .clwb,
             .cmov,
-            .cx8,
             .cx16,
-            .avx512dq,
+            .cx8,
             .ermsb,
             .f16c,
-            .fma,
-            .fsgsbase,
-            .fxsr,
-            .fast_shld_rotate,
+            .fast_gather,
             .fast_scalar_fsqrt,
+            .fast_shld_rotate,
             .fast_variable_shuffle,
             .fast_vector_fsqrt,
+            .fma,
+            .fsgsbase,
+            .fxsr,
             .gfni,
-            .fast_gather,
-            .avx512ifma,
+            .idivq_to_divl,
             .invpcid,
-            .sahf,
             .lzcnt,
+            .macrofusion,
+            .merge_to_threeway_branch,
             .mmx,
             .movbe,
             .mpx,
-            .macrofusion,
-            .merge_to_threeway_branch,
             .nopl,
             .pclmul,
             .pconfig,
@@ -2356,18 +2163,13 @@ pub const cpu = struct {
             .rdpid,
             .rdrnd,
             .rdseed,
+            .sahf,
             .sgx,
             .sha,
-            .sse4_2,
             .slow_3ops_lea,
-            .idivq_to_divl,
+            .sse4_2,
             .vaes,
-            .avx512vbmi,
-            .avx512vbmi2,
-            .avx512vl,
-            .avx512vnni,
             .vpclmulqdq,
-            .avx512vpopcntdq,
             .wbnoinvd,
             .x87,
             .xsave,
@@ -2376,41 +2178,38 @@ pub const cpu = struct {
             .xsaves,
         }),
     };
-
     pub const ivybridge = Cpu{
         .name = "ivybridge",
         .llvm_name = "ivybridge",
         .features = featureSet(&[_]Feature{
             .@"64bit",
-            .sse,
             .avx,
             .cmov,
-            .cx8,
             .cx16,
+            .cx8,
             .f16c,
+            .false_deps_popcnt,
+            .fast_scalar_fsqrt,
+            .fast_shld_rotate,
             .fsgsbase,
             .fxsr,
-            .fast_shld_rotate,
-            .fast_scalar_fsqrt,
-            .sahf,
-            .mmx,
+            .idivq_to_divl,
             .macrofusion,
             .merge_to_threeway_branch,
+            .mmx,
             .nopl,
             .pclmul,
             .popcnt,
-            .false_deps_popcnt,
             .rdrnd,
-            .sse4_2,
+            .sahf,
             .slow_3ops_lea,
-            .idivq_to_divl,
             .slow_unaligned_mem_32,
+            .sse4_2,
             .x87,
             .xsave,
             .xsaveopt,
         }),
     };
-
     pub const k6 = Cpu{
         .name = "k6",
         .llvm_name = "k6",
@@ -2421,108 +2220,96 @@ pub const cpu = struct {
             .x87,
         }),
     };
-
-    pub const k62 = Cpu{
+    pub const k6_2 = Cpu{
         .name = "k6_2",
         .llvm_name = "k6-2",
         .features = featureSet(&[_]Feature{
-            .mmx,
             .@"3dnow",
             .cx8,
             .slow_unaligned_mem_16,
             .x87,
         }),
     };
-
-    pub const k63 = Cpu{
+    pub const k6_3 = Cpu{
         .name = "k6_3",
         .llvm_name = "k6-3",
         .features = featureSet(&[_]Feature{
-            .mmx,
             .@"3dnow",
             .cx8,
             .slow_unaligned_mem_16,
             .x87,
         }),
     };
-
     pub const k8 = Cpu{
         .name = "k8",
         .llvm_name = "k8",
         .features = featureSet(&[_]Feature{
-            .mmx,
             .@"3dnowa",
             .@"64bit",
             .cmov,
             .cx8,
-            .fxsr,
             .fast_scalar_shift_masks,
+            .fxsr,
             .nopl,
-            .sse,
-            .sse2,
             .slow_shld,
             .slow_unaligned_mem_16,
+            .sse2,
             .x87,
         }),
     };
-
     pub const k8_sse3 = Cpu{
         .name = "k8_sse3",
         .llvm_name = "k8-sse3",
         .features = featureSet(&[_]Feature{
-            .mmx,
             .@"3dnowa",
             .@"64bit",
             .cmov,
-            .cx8,
             .cx16,
-            .fxsr,
+            .cx8,
             .fast_scalar_shift_masks,
+            .fxsr,
             .nopl,
-            .sse,
-            .sse3,
             .slow_shld,
             .slow_unaligned_mem_16,
+            .sse3,
             .x87,
         }),
     };
-
     pub const knl = Cpu{
         .name = "knl",
         .llvm_name = "knl",
         .features = featureSet(&[_]Feature{
             .@"64bit",
             .adx,
-            .sse,
             .aes,
+            .avx512cd,
+            .avx512er,
             .avx512f,
+            .avx512pf,
             .bmi,
             .bmi2,
-            .avx512cd,
             .cmov,
-            .cx8,
             .cx16,
-            .avx512er,
+            .cx8,
             .f16c,
+            .fast_gather,
+            .fast_partial_ymm_or_zmm_write,
             .fma,
             .fsgsbase,
             .fxsr,
-            .fast_partial_ymm_or_zmm_write,
-            .fast_gather,
-            .sahf,
+            .idivq_to_divl,
             .lzcnt,
             .mmx,
             .movbe,
             .nopl,
             .pclmul,
-            .avx512pf,
             .popcnt,
             .prefetchwt1,
             .prfchw,
             .rdrnd,
             .rdseed,
+            .sahf,
             .slow_3ops_lea,
-            .idivq_to_divl,
             .slow_incdec,
             .slow_pmaddwd,
             .slow_two_mem_ops,
@@ -2531,158 +2318,142 @@ pub const cpu = struct {
             .xsaveopt,
         }),
     };
-
     pub const knm = Cpu{
         .name = "knm",
         .llvm_name = "knm",
         .features = featureSet(&[_]Feature{
             .@"64bit",
             .adx,
-            .sse,
             .aes,
+            .avx512cd,
+            .avx512er,
             .avx512f,
+            .avx512pf,
+            .avx512vpopcntdq,
             .bmi,
             .bmi2,
-            .avx512cd,
             .cmov,
-            .cx8,
             .cx16,
-            .avx512er,
+            .cx8,
             .f16c,
+            .fast_gather,
+            .fast_partial_ymm_or_zmm_write,
             .fma,
             .fsgsbase,
             .fxsr,
-            .fast_partial_ymm_or_zmm_write,
-            .fast_gather,
-            .sahf,
+            .idivq_to_divl,
             .lzcnt,
             .mmx,
             .movbe,
             .nopl,
             .pclmul,
-            .avx512pf,
             .popcnt,
             .prefetchwt1,
             .prfchw,
             .rdrnd,
             .rdseed,
+            .sahf,
             .slow_3ops_lea,
-            .idivq_to_divl,
             .slow_incdec,
             .slow_pmaddwd,
             .slow_two_mem_ops,
-            .avx512vpopcntdq,
             .x87,
             .xsave,
             .xsaveopt,
         }),
     };
-
     pub const lakemont = Cpu{
         .name = "lakemont",
         .llvm_name = "lakemont",
         .features = 0,
     };
-
     pub const nehalem = Cpu{
         .name = "nehalem",
         .llvm_name = "nehalem",
         .features = featureSet(&[_]Feature{
             .@"64bit",
             .cmov,
-            .cx8,
             .cx16,
+            .cx8,
             .fxsr,
-            .sahf,
-            .mmx,
             .macrofusion,
+            .mmx,
             .nopl,
             .popcnt,
-            .sse,
+            .sahf,
             .sse4_2,
             .x87,
         }),
     };
-
     pub const nocona = Cpu{
         .name = "nocona",
         .llvm_name = "nocona",
         .features = featureSet(&[_]Feature{
             .@"64bit",
             .cmov,
-            .cx8,
             .cx16,
+            .cx8,
             .fxsr,
             .mmx,
             .nopl,
-            .sse,
-            .sse3,
             .slow_unaligned_mem_16,
+            .sse3,
             .x87,
         }),
     };
-
     pub const opteron = Cpu{
         .name = "opteron",
         .llvm_name = "opteron",
         .features = featureSet(&[_]Feature{
-            .mmx,
             .@"3dnowa",
             .@"64bit",
             .cmov,
             .cx8,
-            .fxsr,
             .fast_scalar_shift_masks,
+            .fxsr,
             .nopl,
-            .sse,
-            .sse2,
             .slow_shld,
             .slow_unaligned_mem_16,
+            .sse2,
             .x87,
         }),
     };
-
     pub const opteron_sse3 = Cpu{
         .name = "opteron_sse3",
         .llvm_name = "opteron-sse3",
         .features = featureSet(&[_]Feature{
-            .mmx,
             .@"3dnowa",
             .@"64bit",
             .cmov,
-            .cx8,
             .cx16,
-            .fxsr,
+            .cx8,
             .fast_scalar_shift_masks,
+            .fxsr,
             .nopl,
-            .sse,
-            .sse3,
             .slow_shld,
             .slow_unaligned_mem_16,
+            .sse3,
             .x87,
         }),
     };
-
     pub const penryn = Cpu{
         .name = "penryn",
         .llvm_name = "penryn",
         .features = featureSet(&[_]Feature{
             .@"64bit",
             .cmov,
-            .cx8,
             .cx16,
+            .cx8,
             .fxsr,
-            .sahf,
-            .mmx,
             .macrofusion,
+            .mmx,
             .nopl,
-            .sse,
-            .sse4_1,
+            .sahf,
             .slow_unaligned_mem_16,
+            .sse4_1,
             .x87,
         }),
     };
-
     pub const pentium = Cpu{
         .name = "pentium",
         .llvm_name = "pentium",
@@ -2692,7 +2463,6 @@ pub const cpu = struct {
             .x87,
         }),
     };
-
     pub const pentium_m = Cpu{
         .name = "pentium_m",
         .llvm_name = "pentium-m",
@@ -2702,13 +2472,11 @@ pub const cpu = struct {
             .fxsr,
             .mmx,
             .nopl,
-            .sse,
-            .sse2,
             .slow_unaligned_mem_16,
+            .sse2,
             .x87,
         }),
     };
-
     pub const pentium_mmx = Cpu{
         .name = "pentium_mmx",
         .llvm_name = "pentium-mmx",
@@ -2719,7 +2487,6 @@ pub const cpu = struct {
             .x87,
         }),
     };
-
     pub const pentium2 = Cpu{
         .name = "pentium2",
         .llvm_name = "pentium2",
@@ -2733,7 +2500,6 @@ pub const cpu = struct {
             .x87,
         }),
     };
-
     pub const pentium3 = Cpu{
         .name = "pentium3",
         .llvm_name = "pentium3",
@@ -2743,12 +2509,11 @@ pub const cpu = struct {
             .fxsr,
             .mmx,
             .nopl,
-            .sse,
             .slow_unaligned_mem_16,
+            .sse,
             .x87,
         }),
     };
-
     pub const pentium3m = Cpu{
         .name = "pentium3m",
         .llvm_name = "pentium3m",
@@ -2758,12 +2523,11 @@ pub const cpu = struct {
             .fxsr,
             .mmx,
             .nopl,
-            .sse,
             .slow_unaligned_mem_16,
+            .sse,
             .x87,
         }),
     };
-
     pub const pentium4 = Cpu{
         .name = "pentium4",
         .llvm_name = "pentium4",
@@ -2773,13 +2537,11 @@ pub const cpu = struct {
             .fxsr,
             .mmx,
             .nopl,
-            .sse,
-            .sse2,
             .slow_unaligned_mem_16,
+            .sse2,
             .x87,
         }),
     };
-
     pub const pentium4m = Cpu{
         .name = "pentium4m",
         .llvm_name = "pentium4m",
@@ -2789,13 +2551,11 @@ pub const cpu = struct {
             .fxsr,
             .mmx,
             .nopl,
-            .sse,
-            .sse2,
             .slow_unaligned_mem_16,
+            .sse2,
             .x87,
         }),
     };
-
     pub const pentiumpro = Cpu{
         .name = "pentiumpro",
         .llvm_name = "pentiumpro",
@@ -2807,7 +2567,6 @@ pub const cpu = struct {
             .x87,
         }),
     };
-
     pub const prescott = Cpu{
         .name = "prescott",
         .llvm_name = "prescott",
@@ -2817,125 +2576,118 @@ pub const cpu = struct {
             .fxsr,
             .mmx,
             .nopl,
-            .sse,
-            .sse3,
             .slow_unaligned_mem_16,
+            .sse3,
             .x87,
         }),
     };
-
     pub const sandybridge = Cpu{
         .name = "sandybridge",
         .llvm_name = "sandybridge",
         .features = featureSet(&[_]Feature{
             .@"64bit",
-            .sse,
             .avx,
             .cmov,
-            .cx8,
             .cx16,
-            .fxsr,
-            .fast_shld_rotate,
+            .cx8,
+            .false_deps_popcnt,
             .fast_scalar_fsqrt,
-            .sahf,
-            .mmx,
+            .fast_shld_rotate,
+            .fxsr,
+            .idivq_to_divl,
             .macrofusion,
             .merge_to_threeway_branch,
+            .mmx,
             .nopl,
             .pclmul,
             .popcnt,
-            .false_deps_popcnt,
-            .sse4_2,
+            .sahf,
             .slow_3ops_lea,
-            .idivq_to_divl,
             .slow_unaligned_mem_32,
+            .sse4_2,
             .x87,
             .xsave,
             .xsaveopt,
         }),
     };
-
     pub const silvermont = Cpu{
         .name = "silvermont",
         .llvm_name = "silvermont",
         .features = featureSet(&[_]Feature{
             .@"64bit",
             .cmov,
-            .cx8,
             .cx16,
+            .cx8,
+            .false_deps_popcnt,
             .fxsr,
-            .sahf,
+            .idivq_to_divl,
             .mmx,
             .movbe,
             .nopl,
-            .sse,
             .pclmul,
             .popcnt,
-            .false_deps_popcnt,
             .prfchw,
             .rdrnd,
-            .sse4_2,
-            .ssse3,
-            .idivq_to_divl,
+            .sahf,
             .slow_incdec,
             .slow_lea,
             .slow_pmulld,
             .slow_two_mem_ops,
+            .sse4_2,
+            .ssse3,
             .x87,
         }),
     };
-
     pub const skx = Cpu{
         .name = "skx",
         .llvm_name = "skx",
         .features = featureSet(&[_]Feature{
             .@"64bit",
             .adx,
-            .sse,
             .aes,
             .avx,
             .avx2,
+            .avx512bw,
+            .avx512cd,
+            .avx512dq,
             .avx512f,
+            .avx512vl,
             .bmi,
             .bmi2,
-            .avx512bw,
-            .avx512cd,
             .clflushopt,
             .clwb,
             .cmov,
-            .cx8,
             .cx16,
-            .avx512dq,
+            .cx8,
             .ermsb,
             .f16c,
-            .fma,
-            .fsgsbase,
-            .fxsr,
-            .fast_shld_rotate,
+            .false_deps_popcnt,
+            .fast_gather,
             .fast_scalar_fsqrt,
+            .fast_shld_rotate,
             .fast_variable_shuffle,
             .fast_vector_fsqrt,
-            .fast_gather,
+            .fma,
+            .fsgsbase,
+            .fxsr,
+            .idivq_to_divl,
             .invpcid,
-            .sahf,
             .lzcnt,
+            .macrofusion,
+            .merge_to_threeway_branch,
             .mmx,
             .movbe,
             .mpx,
-            .macrofusion,
-            .merge_to_threeway_branch,
             .nopl,
             .pclmul,
             .pku,
             .popcnt,
-            .false_deps_popcnt,
             .prfchw,
             .rdrnd,
             .rdseed,
-            .sse4_2,
+            .sahf,
             .slow_3ops_lea,
-            .idivq_to_divl,
-            .avx512vl,
+            .sse4_2,
             .x87,
             .xsave,
             .xsavec,
@@ -2943,14 +2695,12 @@ pub const cpu = struct {
             .xsaves,
         }),
     };
-
     pub const skylake = Cpu{
         .name = "skylake",
         .llvm_name = "skylake",
         .features = featureSet(&[_]Feature{
             .@"64bit",
             .adx,
-            .sse,
             .aes,
             .avx,
             .avx2,
@@ -2958,37 +2708,37 @@ pub const cpu = struct {
             .bmi2,
             .clflushopt,
             .cmov,
-            .cx8,
             .cx16,
+            .cx8,
             .ermsb,
             .f16c,
-            .fma,
-            .fsgsbase,
-            .fxsr,
-            .fast_shld_rotate,
+            .false_deps_popcnt,
+            .fast_gather,
             .fast_scalar_fsqrt,
+            .fast_shld_rotate,
             .fast_variable_shuffle,
             .fast_vector_fsqrt,
-            .fast_gather,
+            .fma,
+            .fsgsbase,
+            .fxsr,
+            .idivq_to_divl,
             .invpcid,
-            .sahf,
             .lzcnt,
+            .macrofusion,
+            .merge_to_threeway_branch,
             .mmx,
             .movbe,
             .mpx,
-            .macrofusion,
-            .merge_to_threeway_branch,
             .nopl,
             .pclmul,
             .popcnt,
-            .false_deps_popcnt,
             .prfchw,
             .rdrnd,
             .rdseed,
+            .sahf,
             .sgx,
-            .sse4_2,
             .slow_3ops_lea,
-            .idivq_to_divl,
+            .sse4_2,
             .x87,
             .xsave,
             .xsavec,
@@ -2996,58 +2746,56 @@ pub const cpu = struct {
             .xsaves,
         }),
     };
-
     pub const skylake_avx512 = Cpu{
         .name = "skylake_avx512",
         .llvm_name = "skylake-avx512",
         .features = featureSet(&[_]Feature{
             .@"64bit",
             .adx,
-            .sse,
             .aes,
             .avx,
             .avx2,
+            .avx512bw,
+            .avx512cd,
+            .avx512dq,
             .avx512f,
+            .avx512vl,
             .bmi,
             .bmi2,
-            .avx512bw,
-            .avx512cd,
             .clflushopt,
             .clwb,
             .cmov,
-            .cx8,
             .cx16,
-            .avx512dq,
+            .cx8,
             .ermsb,
             .f16c,
-            .fma,
-            .fsgsbase,
-            .fxsr,
-            .fast_shld_rotate,
+            .false_deps_popcnt,
+            .fast_gather,
             .fast_scalar_fsqrt,
+            .fast_shld_rotate,
             .fast_variable_shuffle,
             .fast_vector_fsqrt,
-            .fast_gather,
+            .fma,
+            .fsgsbase,
+            .fxsr,
+            .idivq_to_divl,
             .invpcid,
-            .sahf,
             .lzcnt,
+            .macrofusion,
+            .merge_to_threeway_branch,
             .mmx,
             .movbe,
             .mpx,
-            .macrofusion,
-            .merge_to_threeway_branch,
             .nopl,
             .pclmul,
             .pku,
             .popcnt,
-            .false_deps_popcnt,
             .prfchw,
             .rdrnd,
             .rdseed,
-            .sse4_2,
+            .sahf,
             .slow_3ops_lea,
-            .idivq_to_divl,
-            .avx512vl,
+            .sse4_2,
             .x87,
             .xsave,
             .xsavec,
@@ -3055,53 +2803,48 @@ pub const cpu = struct {
             .xsaves,
         }),
     };
-
     pub const slm = Cpu{
         .name = "slm",
         .llvm_name = "slm",
         .features = featureSet(&[_]Feature{
             .@"64bit",
             .cmov,
-            .cx8,
             .cx16,
+            .cx8,
+            .false_deps_popcnt,
             .fxsr,
-            .sahf,
+            .idivq_to_divl,
             .mmx,
             .movbe,
             .nopl,
-            .sse,
             .pclmul,
             .popcnt,
-            .false_deps_popcnt,
             .prfchw,
             .rdrnd,
-            .sse4_2,
-            .ssse3,
-            .idivq_to_divl,
+            .sahf,
             .slow_incdec,
             .slow_lea,
             .slow_pmulld,
             .slow_two_mem_ops,
+            .sse4_2,
+            .ssse3,
             .x87,
         }),
     };
-
     pub const tremont = Cpu{
         .name = "tremont",
         .llvm_name = "tremont",
         .features = featureSet(&[_]Feature{
             .@"64bit",
-            .sse,
             .aes,
             .cldemote,
             .clflushopt,
             .cmov,
-            .cx8,
             .cx16,
+            .cx8,
             .fsgsbase,
             .fxsr,
             .gfni,
-            .sahf,
             .mmx,
             .movbe,
             .movdir64b,
@@ -3115,13 +2858,14 @@ pub const cpu = struct {
             .rdpid,
             .rdrnd,
             .rdseed,
+            .sahf,
             .sgx,
             .sha,
-            .sse4_2,
-            .ssse3,
             .slow_incdec,
             .slow_lea,
             .slow_two_mem_ops,
+            .sse4_2,
+            .ssse3,
             .waitpkg,
             .x87,
             .xsave,
@@ -3130,28 +2874,25 @@ pub const cpu = struct {
             .xsaves,
         }),
     };
-
     pub const westmere = Cpu{
         .name = "westmere",
         .llvm_name = "westmere",
         .features = featureSet(&[_]Feature{
             .@"64bit",
             .cmov,
-            .cx8,
             .cx16,
+            .cx8,
             .fxsr,
-            .sahf,
-            .mmx,
             .macrofusion,
+            .mmx,
             .nopl,
-            .sse,
             .pclmul,
             .popcnt,
+            .sahf,
             .sse4_2,
             .x87,
         }),
     };
-
     pub const winchip_c6 = Cpu{
         .name = "winchip_c6",
         .llvm_name = "winchip-c6",
@@ -3161,18 +2902,15 @@ pub const cpu = struct {
             .x87,
         }),
     };
-
     pub const winchip2 = Cpu{
         .name = "winchip2",
         .llvm_name = "winchip2",
         .features = featureSet(&[_]Feature{
-            .mmx,
             .@"3dnow",
             .slow_unaligned_mem_16,
             .x87,
         }),
     };
-
     pub const x86_64 = Cpu{
         .name = "x86_64",
         .llvm_name = "x86-64",
@@ -3181,17 +2919,15 @@ pub const cpu = struct {
             .cmov,
             .cx8,
             .fxsr,
-            .mmx,
             .macrofusion,
+            .mmx,
             .nopl,
-            .sse,
-            .sse2,
             .slow_3ops_lea,
             .slow_incdec,
+            .sse2,
             .x87,
         }),
     };
-
     pub const yonah = Cpu{
         .name = "yonah",
         .llvm_name = "yonah",
@@ -3201,20 +2937,17 @@ pub const cpu = struct {
             .fxsr,
             .mmx,
             .nopl,
-            .sse,
-            .sse3,
             .slow_unaligned_mem_16,
+            .sse3,
             .x87,
         }),
     };
-
     pub const znver1 = Cpu{
         .name = "znver1",
         .llvm_name = "znver1",
         .features = featureSet(&[_]Feature{
             .@"64bit",
             .adx,
-            .sse,
             .aes,
             .avx2,
             .bmi,
@@ -3223,17 +2956,15 @@ pub const cpu = struct {
             .clflushopt,
             .clzero,
             .cmov,
-            .cx8,
             .cx16,
             .f16c,
-            .fma,
-            .fsgsbase,
-            .fxsr,
             .fast_15bytenop,
             .fast_bextr,
             .fast_lzcnt,
             .fast_scalar_shift_masks,
-            .sahf,
+            .fma,
+            .fsgsbase,
+            .fxsr,
             .lzcnt,
             .mmx,
             .movbe,
@@ -3244,9 +2975,10 @@ pub const cpu = struct {
             .prfchw,
             .rdrnd,
             .rdseed,
+            .sahf,
             .sha,
-            .sse4a,
             .slow_shld,
+            .sse4a,
             .x87,
             .xsave,
             .xsavec,
@@ -3254,14 +2986,12 @@ pub const cpu = struct {
             .xsaves,
         }),
     };
-
     pub const znver2 = Cpu{
         .name = "znver2",
         .llvm_name = "znver2",
         .features = featureSet(&[_]Feature{
             .@"64bit",
             .adx,
-            .sse,
             .aes,
             .avx2,
             .bmi,
@@ -3271,17 +3001,15 @@ pub const cpu = struct {
             .clwb,
             .clzero,
             .cmov,
-            .cx8,
             .cx16,
             .f16c,
-            .fma,
-            .fsgsbase,
-            .fxsr,
             .fast_15bytenop,
             .fast_bextr,
             .fast_lzcnt,
             .fast_scalar_shift_masks,
-            .sahf,
+            .fma,
+            .fsgsbase,
+            .fxsr,
             .lzcnt,
             .mmx,
             .movbe,
@@ -3293,9 +3021,10 @@ pub const cpu = struct {
             .rdpid,
             .rdrnd,
             .rdseed,
+            .sahf,
             .sha,
-            .sse4a,
             .slow_shld,
+            .sse4a,
             .wbnoinvd,
             .x87,
             .xsave,
@@ -3306,10 +3035,13 @@ pub const cpu = struct {
     };
 };
 
+/// All x86 CPUs, sorted alphabetically by name.
+/// TODO: Replace this with usage of `std.meta.declList`. It does work, but stage1
+/// compiler has inefficient memory and CPU usage, affecting build times.
 pub const all_cpus = &[_]*const Cpu{
     &cpu.amdfam10,
     &cpu.athlon,
-    &cpu.athlon4,
+    &cpu.athlon_4,
     &cpu.athlon_fx,
     &cpu.athlon_mp,
     &cpu.athlon_tbird,
@@ -3327,7 +3059,7 @@ pub const all_cpus = &[_]*const Cpu{
     &cpu.btver1,
     &cpu.btver2,
     &cpu.c3,
-    &cpu.c32,
+    &cpu.c3_2,
     &cpu.cannonlake,
     &cpu.cascadelake,
     &cpu.cooperlake,
@@ -3349,8 +3081,8 @@ pub const all_cpus = &[_]*const Cpu{
     &cpu.icelake_server,
     &cpu.ivybridge,
     &cpu.k6,
-    &cpu.k62,
-    &cpu.k63,
+    &cpu.k6_2,
+    &cpu.k6_3,
     &cpu.k8,
     &cpu.k8_sse3,
     &cpu.knl,