microsoft · Mark Hildebrand (hildebrandmw) · Aug 4, 2026 · Aug 4, 2026 · Aug 4, 2026 · Aug 4, 2026
diff --git a/diskann-wide/src/arch/aarch64/double.rs b/diskann-wide/src/arch/aarch64/double.rs
@@ -243,6 +243,7 @@ mod tests {
 
         // Bit ops
         test_utils::ops::test_bitops!(u8x32, 0xd62d8de09f82ed4e, test_neon());
+        test_utils::ops::test_popcount!(u8x32, 0xf24be4e5bc1d95b6, test_neon());
         test_utils::ops::test_splitjoin!(u8x32 => u8x16, 0x2e301b7e12090d5c, test_neon());
         test_utils::ops::test_zipunzip!(u8x32 => u8x16, 0xa1b2c3d4e5f67890, test_neon());
     }
@@ -253,6 +254,7 @@ mod tests {
 
         // Bit ops
         test_utils::ops::test_bitops!(u8x64, 0xd62d8de09f82ed4e, test_neon());
+        test_utils::ops::test_popcount!(u8x64, 0x7207f32ecac3fbf3, test_neon());
         test_utils::ops::test_splitjoin!(u8x64 => u8x32, 0x2e301b7e12090d5c, test_neon());
     }
 
@@ -300,6 +302,7 @@ mod tests {
 
         // Bit ops
         test_utils::ops::test_bitops!(i8x32, 0xd62d8de09f82ed4e, test_neon());
+        test_utils::ops::test_popcount!(i8x32, 0x45f60286da125e84, test_neon());
         test_utils::ops::test_abs!(i8x32, 0xd62d8de09f82ed4e, test_neon());
         test_utils::ops::test_splitjoin!(i8x32 => i8x16, 0x2e301b7e12090d5c, test_neon());
         test_utils::ops::test_zipunzip!(i8x32 => i8x16, 0xc7e3a92f1d8b5604, test_neon());
@@ -311,6 +314,7 @@ mod tests {
 
         // Bit ops
         test_utils::ops::test_bitops!(i8x64, 0xd62d8de09f82ed4e, test_neon());
+        test_utils::ops::test_popcount!(i8x64, 0x73e6c7da94a679ac, test_neon());
         test_utils::ops::test_abs!(i8x64, 0xd62d8de09f82ed4e, test_neon());
         test_utils::ops::test_splitjoin!(i8x64 => i8x32, 0x2e301b7e12090d5c, test_neon());
     }

diff --git a/diskann-wide/src/arch/aarch64/i8x16_.rs b/diskann-wide/src/arch/aarch64/i8x16_.rs
@@ -4,8 +4,8 @@
  */
 
 use crate::{
-    Emulated, SIMDAbs, SIMDMask, SIMDMulAdd, SIMDPartialEq, SIMDPartialOrd, SIMDVector,
-    constant::Const, helpers,
+    Emulated, SIMDAbs, SIMDMask, SIMDMulAdd, SIMDPartialEq, SIMDPartialOrd, SIMDPopcount,
+    SIMDVector, constant::Const, helpers,
 };
 
 // AArch64 masks
@@ -31,6 +31,7 @@ helpers::unsafe_map_binary_op!(i8x16, std::ops::Add, add, vaddq_s8, "neon");
 helpers::unsafe_map_binary_op!(i8x16, std::ops::Sub, sub, vsubq_s8, "neon");
 helpers::unsafe_map_binary_op!(i8x16, std::ops::Mul, mul, vmulq_s8, "neon");
 helpers::unsafe_map_unary_op!(i8x16, SIMDAbs, abs_simd, vabsq_s8, "neon");
+helpers::unsafe_map_unary_op!(i8x16, SIMDPopcount, popcount_simd, vcntq_s8, "neon");
 macros::aarch64_define_fma!(i8x16, vmlaq_s8);
 
 macros::aarch64_define_cmp!(
@@ -102,4 +103,5 @@ mod tests {
 
     // Bit ops
     test_utils::ops::test_bitops!(i8x16, 0xd62d8de09f82ed4e, test_neon());
+    test_utils::ops::test_popcount!(i8x16, 0xddff81feaf77563f, test_neon());
 }
diff --git a/diskann-wide/src/arch/aarch64/i8x8_.rs b/diskann-wide/src/arch/aarch64/i8x8_.rs
@@ -4,8 +4,8 @@
  */
 
 use crate::{
-    Emulated, SIMDAbs, SIMDMask, SIMDMulAdd, SIMDPartialEq, SIMDPartialOrd, SIMDVector,
-    constant::Const, helpers,
+    Emulated, SIMDAbs, SIMDMask, SIMDMulAdd, SIMDPartialEq, SIMDPartialOrd, SIMDPopcount,
+    SIMDVector, constant::Const, helpers,
 };
 
 // AArch64 masks
@@ -30,6 +30,7 @@ helpers::unsafe_map_binary_op!(i8x8, std::ops::Add, add, vadd_s8, "neon");
 helpers::unsafe_map_binary_op!(i8x8, std::ops::Sub, sub, vsub_s8, "neon");
 helpers::unsafe_map_binary_op!(i8x8, std::ops::Mul, mul, vmul_s8, "neon");
 helpers::unsafe_map_unary_op!(i8x8, SIMDAbs, abs_simd, vabs_s8, "neon");
+helpers::unsafe_map_unary_op!(i8x8, SIMDPopcount, popcount_simd, vcnt_s8, "neon");
 macros::aarch64_define_fma!(i8x8, vmla_s8);
 
 macros::aarch64_define_cmp!(i8x8, vceq_s8, (vmvn_u8), vclt_s8, vcle_s8, vcgt_s8, vcge_s8);
@@ -92,4 +93,5 @@ mod tests {
 
     // Bit ops
     test_utils::ops::test_bitops!(i8x8, 0xd62d8de09f82ed4e, test_neon());
+    test_utils::ops::test_popcount!(i8x8, 0x46c84a6be23bc633, test_neon());
 }
diff --git a/diskann-wide/src/arch/aarch64/u8x16_.rs b/diskann-wide/src/arch/aarch64/u8x16_.rs
@@ -7,7 +7,7 @@ use crate::{
     Emulated,
     constant::Const,
     helpers,
-    traits::{SIMDMask, SIMDMulAdd, SIMDPartialEq, SIMDPartialOrd, SIMDVector},
+    traits::{SIMDMask, SIMDMulAdd, SIMDPartialEq, SIMDPartialOrd, SIMDPopcount, SIMDVector},
 };
 
 // AArch64 masks
@@ -33,6 +33,7 @@ macros::aarch64_splitjoin!(u8x16, u8x8, vget_low_u8, vget_high_u8, vcombine_u8);
 helpers::unsafe_map_binary_op!(u8x16, std::ops::Add, add, vaddq_u8, "neon");
 helpers::unsafe_map_binary_op!(u8x16, std::ops::Sub, sub, vsubq_u8, "neon");
 helpers::unsafe_map_binary_op!(u8x16, std::ops::Mul, mul, vmulq_u8, "neon");
+helpers::unsafe_map_unary_op!(u8x16, SIMDPopcount, popcount_simd, vcntq_u8, "neon");
 macros::aarch64_define_fma!(u8x16, vmlaq_u8);
 
 macros::aarch64_define_cmp!(
@@ -103,4 +104,5 @@ mod tests {
 
     // Bit ops
     test_utils::ops::test_bitops!(u8x16, 0xd62d8de09f82ed4e, test_neon());
+    test_utils::ops::test_popcount!(u8x16, 0xb801a142f098b25d, test_neon());
 }
diff --git a/diskann-wide/src/arch/aarch64/u8x8_.rs b/diskann-wide/src/arch/aarch64/u8x8_.rs
@@ -7,7 +7,7 @@ use crate::{
     Emulated,
     constant::Const,
     helpers,
-    traits::{SIMDMask, SIMDMulAdd, SIMDPartialEq, SIMDPartialOrd, SIMDVector},
+    traits::{SIMDMask, SIMDMulAdd, SIMDPartialEq, SIMDPartialOrd, SIMDPopcount, SIMDVector},
 };
 
 // AArch64 masks
@@ -31,6 +31,7 @@ macros::aarch64_define_loadstore!(u8x8, vld1_u8, internal::load_first::u8x8, vst
 helpers::unsafe_map_binary_op!(u8x8, std::ops::Add, add, vadd_u8, "neon");
 helpers::unsafe_map_binary_op!(u8x8, std::ops::Sub, sub, vsub_u8, "neon");
 helpers::unsafe_map_binary_op!(u8x8, std::ops::Mul, mul, vmul_u8, "neon");
+helpers::unsafe_map_unary_op!(u8x8, SIMDPopcount, popcount_simd, vcnt_u8, "neon");
 macros::aarch64_define_fma!(u8x8, vmla_u8);
 
 macros::aarch64_define_cmp!(u8x8, vceq_u8, (vmvn_u8), vclt_u8, vcle_u8, vcgt_u8, vcge_u8);
@@ -92,4 +93,5 @@ mod tests {
 
     // Bit ops
     test_utils::ops::test_bitops!(u8x8, 0xd62d8de09f82ed4e, test_neon());
+    test_utils::ops::test_popcount!(u8x8, 0xac222b463efd782c, test_neon());
 }
diff --git a/diskann-wide/src/arch/x86_64/v4/i16x16_.rs b/diskann-wide/src/arch/x86_64/v4/i16x16_.rs
@@ -18,7 +18,7 @@ use crate::{
     bitmask::BitMask,
     constant::Const,
     helpers,
-    traits::{SIMDAbs, SIMDMulAdd, SIMDVector},
+    traits::{SIMDAbs, SIMDMulAdd, SIMDPopcount, SIMDVector},
 };
 
 ///////////////////
@@ -47,6 +47,13 @@ helpers::unsafe_map_binary_op!(i16x16, std::ops::Add, add, _mm256_add_epi16, "av
 helpers::unsafe_map_binary_op!(i16x16, std::ops::Sub, sub, _mm256_sub_epi16, "avx2");
 helpers::unsafe_map_binary_op!(i16x16, std::ops::Mul, mul, _mm256_mullo_epi16, "avx2");
 helpers::unsafe_map_unary_op!(i16x16, SIMDAbs, abs_simd, _mm256_abs_epi16, "avx2");
+helpers::unsafe_map_unary_op!(
+    i16x16,
+    SIMDPopcount,
+    popcount_simd,
+    _mm256_popcnt_epi16,
+    "avx512bitalg,avx512vl"
+);
 
 helpers::unsafe_map_binary_op!(i16x16, std::ops::BitAnd, bitand, _mm256_and_si256, "avx2");
 helpers::unsafe_map_binary_op!(i16x16, std::ops::BitOr, bitor, _mm256_or_si256, "avx2");
@@ -142,4 +149,5 @@ mod test_x86_i16 {
 
     // Bit ops
     test_utils::ops::test_bitops!(i16x16, 0xba0be356b04d6427, V4::new_checked_uncached());
+    test_utils::ops::test_popcount!(i16x16, 0x781e87035cbf3527, V4::new_checked_uncached());
 }
diff --git a/diskann-wide/src/arch/x86_64/v4/i16x32_.rs b/diskann-wide/src/arch/x86_64/v4/i16x32_.rs
@@ -17,7 +17,7 @@ use crate::{
     bitmask::BitMask,
     constant::Const,
     helpers,
-    traits::{SIMDAbs, SIMDMulAdd, SIMDVector},
+    traits::{SIMDAbs, SIMDMulAdd, SIMDPopcount, SIMDVector},
 };
 
 ///////////////////
@@ -33,6 +33,13 @@ helpers::unsafe_map_binary_op!(i16x32, std::ops::Add, add, _mm512_add_epi16, "av
 helpers::unsafe_map_binary_op!(i16x32, std::ops::Sub, sub, _mm512_sub_epi16, "avx512bw");
 helpers::unsafe_map_binary_op!(i16x32, std::ops::Mul, mul, _mm512_mullo_epi16, "avx512bw");
 helpers::unsafe_map_unary_op!(i16x32, SIMDAbs, abs_simd, _mm512_abs_epi16, "avx512bw");
+helpers::unsafe_map_unary_op!(
+    i16x32,
+    SIMDPopcount,
+    popcount_simd,
+    _mm512_popcnt_epi16,
+    "avx512bitalg"
+);
 
 helpers::unsafe_map_binary_op!(
     i16x32,
@@ -126,4 +133,5 @@ mod test_x86_i16 {
 
     // Bit ops
     test_utils::ops::test_bitops!(i16x32, 0xba0be356b04d6427, V4::new_checked_uncached());
+    test_utils::ops::test_popcount!(i16x32, 0xa787d20c8f94b58b, V4::new_checked_uncached());
 }
diff --git a/diskann-wide/src/arch/x86_64/v4/i16x8_.rs b/diskann-wide/src/arch/x86_64/v4/i16x8_.rs
@@ -17,7 +17,7 @@ use crate::{
     bitmask::BitMask,
     constant::Const,
     helpers,
-    traits::{SIMDAbs, SIMDMulAdd, SIMDVector},
+    traits::{SIMDAbs, SIMDMulAdd, SIMDPopcount, SIMDVector},
 };
 
 ///////////////////
@@ -33,6 +33,13 @@ helpers::unsafe_map_binary_op!(i16x8, std::ops::Add, add, _mm_add_epi16, "sse2")
 helpers::unsafe_map_binary_op!(i16x8, std::ops::Sub, sub, _mm_sub_epi16, "sse2");
 helpers::unsafe_map_binary_op!(i16x8, std::ops::Mul, mul, _mm_mullo_epi16, "sse2");
 helpers::unsafe_map_unary_op!(i16x8, SIMDAbs, abs_simd, _mm_abs_epi16, "ssse3");
+helpers::unsafe_map_unary_op!(
+    i16x8,
+    SIMDPopcount,
+    popcount_simd,
+    _mm_popcnt_epi16,
+    "avx512bitalg,avx512vl"
+);
 
 helpers::unsafe_map_binary_op!(i16x8, std::ops::BitAnd, bitand, _mm_and_si128, "sse2");
 helpers::unsafe_map_binary_op!(i16x8, std::ops::BitOr, bitor, _mm_or_si128, "sse2");
@@ -125,4 +132,5 @@ mod test_x86_i16 {
 
     // Bit ops
     test_utils::ops::test_bitops!(i16x8, 0x2dd8796bda1cb89d, V4::new_checked_uncached());
+    test_utils::ops::test_popcount!(i16x8, 0x6bc496e226618794, V4::new_checked_uncached());
 }
diff --git a/diskann-wide/src/arch/x86_64/v4/i32x16_.rs b/diskann-wide/src/arch/x86_64/v4/i32x16_.rs
@@ -16,7 +16,10 @@ use crate::{
     },
     constant::Const,
     helpers,
-    traits::{SIMDAbs, SIMDDotProduct, SIMDMask, SIMDMulAdd, SIMDSelect, SIMDSumTree, SIMDVector},
+    traits::{
+        SIMDAbs, SIMDDotProduct, SIMDMask, SIMDMulAdd, SIMDPopcount, SIMDSelect, SIMDSumTree,
+        SIMDVector,
+    },
 };
 
 /////
@@ -32,6 +35,13 @@ helpers::unsafe_map_binary_op!(i32x16, std::ops::Add, add, _mm512_add_epi32, "av
 helpers::unsafe_map_binary_op!(i32x16, std::ops::Sub, sub, _mm512_sub_epi32, "avx512f");
 helpers::unsafe_map_binary_op!(i32x16, std::ops::Mul, mul, _mm512_mullo_epi32, "avx512f");
 helpers::unsafe_map_unary_op!(i32x16, SIMDAbs, abs_simd, _mm512_abs_epi32, "avx512f");
+helpers::unsafe_map_unary_op!(
+    i32x16,
+    SIMDPopcount,
+    popcount_simd,
+    _mm512_popcnt_epi32,
+    "avx512vpopcntdq"
+);
 
 helpers::unsafe_map_binary_op!(
     i32x16,
@@ -170,6 +180,7 @@ mod test_x86_i32 {
 
     // Bit ops
     test_utils::ops::test_bitops!(i32x16, 0xc5f7d8d8df0b7b6c, V4::new_checked_uncached());
+    test_utils::ops::test_popcount!(i32x16, 0xc8068253462939e5, V4::new_checked_uncached());
 
     // Dot Products
     test_utils::dot_product::test_dot_product!(

diff --git a/diskann-wide/src/arch/x86_64/v4/i32x4_.rs b/diskann-wide/src/arch/x86_64/v4/i32x4_.rs
@@ -16,7 +16,7 @@ use crate::{
     },
     constant::Const,
     helpers,
-    traits::{SIMDAbs, SIMDMask, SIMDMulAdd, SIMDVector},
+    traits::{SIMDAbs, SIMDMask, SIMDMulAdd, SIMDPopcount, SIMDVector},
 };
 
 /////
@@ -32,6 +32,13 @@ helpers::unsafe_map_binary_op!(i32x4, std::ops::Add, add, _mm_add_epi32, "sse2")
 helpers::unsafe_map_binary_op!(i32x4, std::ops::Sub, sub, _mm_sub_epi32, "sse2");
 helpers::unsafe_map_binary_op!(i32x4, std::ops::Mul, mul, _mm_mullo_epi32, "sse4.1");
 helpers::unsafe_map_unary_op!(i32x4, SIMDAbs, abs_simd, _mm_abs_epi32, "ssse3");
+helpers::unsafe_map_unary_op!(
+    i32x4,
+    SIMDPopcount,
+    popcount_simd,
+    _mm_popcnt_epi32,
+    "avx512vpopcntdq,avx512vl"
+);
 
 helpers::unsafe_map_binary_op!(i32x4, std::ops::BitAnd, bitand, _mm_and_si128, "sse2");
 helpers::unsafe_map_binary_op!(i32x4, std::ops::BitOr, bitor, _mm_or_si128, "sse2");
@@ -109,4 +116,5 @@ mod test_x86_i32 {
 
     // Bit ops
     test_utils::ops::test_bitops!(i32x4, 0x763fc44f8f7cd40c, V4::new_checked_uncached());
+    test_utils::ops::test_popcount!(i32x4, 0x6b756a0a75b4f2d6, V4::new_checked_uncached());
 }
diff --git a/diskann-wide/src/arch/x86_64/v4/i32x8_.rs b/diskann-wide/src/arch/x86_64/v4/i32x8_.rs
@@ -17,7 +17,10 @@ use crate::{
     },
     constant::Const,
     helpers,
-    traits::{SIMDAbs, SIMDDotProduct, SIMDMask, SIMDMulAdd, SIMDSelect, SIMDSumTree, SIMDVector},
+    traits::{
+        SIMDAbs, SIMDDotProduct, SIMDMask, SIMDMulAdd, SIMDPopcount, SIMDSelect, SIMDSumTree,
+        SIMDVector,
+    },
 };
 
 /////
@@ -41,6 +44,13 @@ helpers::unsafe_map_binary_op!(i32x8, std::ops::Add, add, _mm256_add_epi32, "avx
 helpers::unsafe_map_binary_op!(i32x8, std::ops::Sub, sub, _mm256_sub_epi32, "avx2");
 helpers::unsafe_map_binary_op!(i32x8, std::ops::Mul, mul, _mm256_mullo_epi32, "avx2");
 helpers::unsafe_map_unary_op!(i32x8, SIMDAbs, abs_simd, _mm256_abs_epi32, "avx2");
+helpers::unsafe_map_unary_op!(
+    i32x8,
+    SIMDPopcount,
+    popcount_simd,
+    _mm256_popcnt_epi32,
+    "avx512vpopcntdq,avx512vl"
+);
 
 helpers::unsafe_map_binary_op!(i32x8, std::ops::BitAnd, bitand, _mm256_and_si256, "avx2");
 helpers::unsafe_map_binary_op!(i32x8, std::ops::BitOr, bitor, _mm256_or_si256, "avx2");
@@ -167,6 +177,7 @@ mod test_x86_i32 {
 
     // Bit ops
     test_utils::ops::test_bitops!(i32x8, 0xc5f7d8d8df0b7b6c, V4::new_checked_uncached());
+    test_utils::ops::test_popcount!(i32x8, 0xb24fb02ef70f7345, V4::new_checked_uncached());
 
     // Dot Products
     test_utils::dot_product::test_dot_product!(

diff --git a/diskann-wide/src/arch/x86_64/v4/i8x16_.rs b/diskann-wide/src/arch/x86_64/v4/i8x16_.rs
@@ -16,7 +16,7 @@ use crate::{
     },
     constant::Const,
     helpers,
-    traits::{SIMDAbs, SIMDMask, SIMDMulAdd, SIMDVector},
+    traits::{SIMDAbs, SIMDMask, SIMDMulAdd, SIMDPopcount, SIMDVector},
 };
 
 //////////////////
@@ -31,6 +31,13 @@ macros::x86_retarget!(i8x16 => v3::i8x16);
 helpers::unsafe_map_binary_op!(i8x16, std::ops::Add, add, _mm_add_epi8, "sse2");
 helpers::unsafe_map_binary_op!(i8x16, std::ops::Sub, sub, _mm_sub_epi8, "sse2");
 helpers::unsafe_map_unary_op!(i8x16, SIMDAbs, abs_simd, _mm_abs_epi8, "ssse3");
+helpers::unsafe_map_unary_op!(
+    i8x16,
+    SIMDPopcount,
+    popcount_simd,
+    _mm_popcnt_epi8,
+    "avx512bitalg,avx512vl"
+);
 
 impl std::ops::Mul for i8x16 {
     type Output = Self;
@@ -134,4 +141,5 @@ mod test_x86_i8 {
 
     // Bit ops
     test_utils::ops::test_bitops!(i8x16, 0xeb7bb4da5b84ebbe, V4::new_checked_uncached());
+    test_utils::ops::test_popcount!(i8x16, 0x9d61a6aa25b322cc, V4::new_checked_uncached());
 }
diff --git a/diskann-wide/src/arch/x86_64/v4/i8x32_.rs b/diskann-wide/src/arch/x86_64/v4/i8x32_.rs
@@ -16,7 +16,7 @@ use crate::{
     },
     constant::Const,
     helpers,
-    traits::{SIMDAbs, SIMDMask, SIMDMulAdd, SIMDVector},
+    traits::{SIMDAbs, SIMDMask, SIMDMulAdd, SIMDPopcount, SIMDVector},
 };
 
 //////////////////
@@ -59,6 +59,13 @@ impl std::ops::Mul for i8x32 {
 helpers::unsafe_map_binary_op!(i8x32, std::ops::Add, add, _mm256_add_epi8, "avx2");
 helpers::unsafe_map_binary_op!(i8x32, std::ops::Sub, sub, _mm256_sub_epi8, "avx2");
 helpers::unsafe_map_unary_op!(i8x32, SIMDAbs, abs_simd, _mm256_abs_epi8, "avx2");
+helpers::unsafe_map_unary_op!(
+    i8x32,
+    SIMDPopcount,
+    popcount_simd,
+    _mm256_popcnt_epi8,
+    "avx512bitalg,avx512vl"
+);
 
 helpers::unsafe_map_binary_op!(i8x32, std::ops::BitAnd, bitand, _mm256_and_si256, "avx2");
 helpers::unsafe_map_binary_op!(i8x32, std::ops::BitOr, bitor, _mm256_or_si256, "avx2");
@@ -158,4 +165,5 @@ mod test_x86_i8 {
 
     // Bit ops
     test_utils::ops::test_bitops!(i8x32, 0xd62d8de09f82ed4e, V4::new_checked_uncached());
+    test_utils::ops::test_popcount!(i8x32, 0xdc54b21bedc25f46, V4::new_checked_uncached());
 }
diff --git a/diskann-wide/src/arch/x86_64/v4/i8x64_.rs b/diskann-wide/src/arch/x86_64/v4/i8x64_.rs
@@ -16,7 +16,7 @@ use crate::{
     },
     constant::Const,
     helpers,
-    traits::{SIMDAbs, SIMDMask, SIMDMulAdd, SIMDVector},
+    traits::{SIMDAbs, SIMDMask, SIMDMulAdd, SIMDPopcount, SIMDVector},
 };
 
 //////////////////
@@ -39,6 +39,13 @@ impl std::ops::Mul for i8x64 {
 helpers::unsafe_map_binary_op!(i8x64, std::ops::Add, add, _mm512_add_epi8, "avx512bw");
 helpers::unsafe_map_binary_op!(i8x64, std::ops::Sub, sub, _mm512_sub_epi8, "avx512bw");
 helpers::unsafe_map_unary_op!(i8x64, SIMDAbs, abs_simd, _mm512_abs_epi8, "avx512bw");
+helpers::unsafe_map_unary_op!(
+    i8x64,
+    SIMDPopcount,
+    popcount_simd,
+    _mm512_popcnt_epi8,
+    "avx512bitalg"
+);
 
 helpers::unsafe_map_binary_op!(i8x64, std::ops::BitAnd, bitand, _mm512_and_si512, "avx512f");
 helpers::unsafe_map_binary_op!(i8x64, std::ops::BitOr, bitor, _mm512_or_si512, "avx512f");
@@ -131,4 +138,5 @@ mod test_x86_i8 {
 
     // Bit ops
     test_utils::ops::test_bitops!(i8x64, 0xd62d8de09f82ed4e, V4::new_checked_uncached());
+    test_utils::ops::test_popcount!(i8x64, 0x50971ea8ed24cedd, V4::new_checked_uncached());
 }