diff --git a/mocks/DDrppi_mocks/Makefile b/mocks/DDrppi_mocks/Makefile
index 8c36b65f..f6e286dc 100644
--- a/mocks/DDrppi_mocks/Makefile
+++ b/mocks/DDrppi_mocks/Makefile
@@ -13,7 +13,7 @@ LIBRARY := lib$(LIBNAME).a
 LIBSRC  := countpairs_rp_pi_mocks.c countpairs_rp_pi_mocks_impl_double.c countpairs_rp_pi_mocks_impl_float.c \
            $(UTILS_DIR)/gridlink_mocks_impl_float.c $(UTILS_DIR)/gridlink_mocks_impl_double.c \
            $(UTILS_DIR)/gridlink_utils_float.c $(UTILS_DIR)/gridlink_utils_double.c \
-           $(UTILS_DIR)/utils.c $(UTILS_DIR)/progressbar.c $(UTILS_DIR)/cpu_features.c $(UTILS_DIR)/avx512_calls.c \
+           $(UTILS_DIR)/utils.c $(UTILS_DIR)/progressbar.c $(UTILS_DIR)/cpu_features.c \
 	   $(UTILS_DIR)/set_cosmo_dist.c $(UTILS_DIR)/cosmology_params.c
 LIBRARY_HEADERS := $(LIBNAME).h
 
diff --git a/mocks/DDrppi_mocks/countpairs_rp_pi_mocks_kernels.c.src b/mocks/DDrppi_mocks/countpairs_rp_pi_mocks_kernels.c.src
index 407b2ef9..a61ab06c 100644
--- a/mocks/DDrppi_mocks/countpairs_rp_pi_mocks_kernels.c.src
+++ b/mocks/DDrppi_mocks/countpairs_rp_pi_mocks_kernels.c.src
@@ -178,7 +178,7 @@ static inline int countpairs_rp_pi_mocks_avx512_intrinsics_DOUBLE(const int64_t
         const AVX512_FLOATS m_sqr_rpmin  = AVX512_SET_FLOAT(sqr_rpmin);
 
         for(int64_t j=n_off;j<N1;j+=AVX512_NVEC){
-            AVX512_MASK m_mask_left = (N1 - j) >= AVX512_NVEC ? ~0:masks_per_misalignment_value_DOUBLE[N1-j];
+            AVX512_MASK m_mask_left = (N1 - j) >= AVX512_NVEC ? ~0U:((1U << (unsigned)(N1 - j)) - 1U);
             const AVX512_FLOATS m_x2 = AVX512_MASKZ_LOAD_FLOATS_UNALIGNED(m_mask_left, localx1);
             const AVX512_FLOATS m_y2 = AVX512_MASKZ_LOAD_FLOATS_UNALIGNED(m_mask_left, localy1);
             const AVX512_FLOATS m_z2 = AVX512_MASKZ_LOAD_FLOATS_UNALIGNED(m_mask_left, localz1);
diff --git a/mocks/DDsmu_mocks/Makefile b/mocks/DDsmu_mocks/Makefile
index 3642b932..f5ca84f1 100644
--- a/mocks/DDsmu_mocks/Makefile
+++ b/mocks/DDsmu_mocks/Makefile
@@ -13,7 +13,7 @@ LIBRARY := lib$(LIBNAME).a
 LIBSRC  := countpairs_s_mu_mocks.c countpairs_s_mu_mocks_impl_double.c countpairs_s_mu_mocks_impl_float.c \
            $(UTILS_DIR)/gridlink_mocks_impl_float.c $(UTILS_DIR)/gridlink_mocks_impl_double.c \
            $(UTILS_DIR)/gridlink_utils_float.c $(UTILS_DIR)/gridlink_utils_double.c \
-           $(UTILS_DIR)/utils.c $(UTILS_DIR)/progressbar.c $(UTILS_DIR)/cpu_features.c $(UTILS_DIR)/avx512_calls.c \
+           $(UTILS_DIR)/utils.c $(UTILS_DIR)/progressbar.c $(UTILS_DIR)/cpu_features.c \
 	   $(UTILS_DIR)/set_cosmo_dist.c $(UTILS_DIR)/cosmology_params.c
 LIBRARY_HEADERS := $(LIBNAME).h
 
diff --git a/mocks/DDsmu_mocks/countpairs_s_mu_mocks_kernels.c.src b/mocks/DDsmu_mocks/countpairs_s_mu_mocks_kernels.c.src
index 643fe83b..c6b0e4bf 100644
--- a/mocks/DDsmu_mocks/countpairs_s_mu_mocks_kernels.c.src
+++ b/mocks/DDsmu_mocks/countpairs_s_mu_mocks_kernels.c.src
@@ -172,7 +172,7 @@ static inline int countpairs_s_mu_mocks_avx512_intrinsics_DOUBLE(const int64_t N
         const AVX512_FLOATS m_sqr_smin = AVX512_SET_FLOAT(sqr_smin);
         const AVX512_FLOATS m_sqr_mumax = AVX512_SET_FLOAT(sqr_mumax);
         for(int64_t j=n_off;j<N1;j+=AVX512_NVEC){
-            AVX512_MASK m_mask_left = (N1 - j) >= AVX512_NVEC ? ~0:masks_per_misalignment_value_DOUBLE[N1-j];
+            AVX512_MASK m_mask_left = (N1 - j) >= AVX512_NVEC ? ~0U:((1U << (unsigned)(N1 - j)) - 1U);
             const AVX512_FLOATS m_x1 = AVX512_MASKZ_LOAD_FLOATS_UNALIGNED(m_mask_left, localx1);
             const AVX512_FLOATS m_y1 = AVX512_MASKZ_LOAD_FLOATS_UNALIGNED(m_mask_left, localy1);
             const AVX512_FLOATS m_z1 = AVX512_MASKZ_LOAD_FLOATS_UNALIGNED(m_mask_left, localz1);
diff --git a/mocks/DDtheta_mocks/Makefile b/mocks/DDtheta_mocks/Makefile
index 0ef90bbf..61a31924 100644
--- a/mocks/DDtheta_mocks/Makefile
+++ b/mocks/DDtheta_mocks/Makefile
@@ -13,7 +13,7 @@ LIBRARY := lib$(LIBNAME).a
 LIBSRC  := countpairs_theta_mocks.c countpairs_theta_mocks_impl_float.c countpairs_theta_mocks_impl_double.c \
            $(UTILS_DIR)/gridlink_mocks_impl_float.c $(UTILS_DIR)/gridlink_mocks_impl_double.c \
            $(UTILS_DIR)/gridlink_utils_float.c $(UTILS_DIR)/gridlink_utils_double.c \
-           $(UTILS_DIR)/utils.c $(UTILS_DIR)/progressbar.c $(UTILS_DIR)/cpu_features.c $(UTILS_DIR)/avx512_calls.c
+           $(UTILS_DIR)/utils.c $(UTILS_DIR)/progressbar.c $(UTILS_DIR)/cpu_features.c
 LIBRARY_HEADERS := $(LIBNAME).h
 
 TARGET := DDtheta_mocks
diff --git a/mocks/DDtheta_mocks/countpairs_theta_mocks_kernels.c.src b/mocks/DDtheta_mocks/countpairs_theta_mocks_kernels.c.src
index acac3954..13691d4e 100644
--- a/mocks/DDtheta_mocks/countpairs_theta_mocks_kernels.c.src
+++ b/mocks/DDtheta_mocks/countpairs_theta_mocks_kernels.c.src
@@ -1027,7 +1027,7 @@ static inline int countpairs_theta_mocks_avx512_intrinsics_DOUBLE(const int64_t
 
         const AVX512_FLOATS m_max_dz = AVX512_SET_FLOAT(max_dz);
         for(int64_t j=n_off;j<N1;j+=AVX512_NVEC) {
-            AVX512_MASK m_mask_left = (N1 - j) >= AVX512_NVEC ? ~0:masks_per_misalignment_value_DOUBLE[N1-j];
+            AVX512_MASK m_mask_left = (N1 - j) >= AVX512_NVEC ? ~0U:((1U << (unsigned)(N1 - j)) - 1U);
             const AVX512_FLOATS m_x1 = AVX512_MASKZ_LOAD_FLOATS_UNALIGNED(m_mask_left, localx1);
             const AVX512_FLOATS m_y1 = AVX512_MASKZ_LOAD_FLOATS_UNALIGNED(m_mask_left, localy1);
             const AVX512_FLOATS m_z1 = AVX512_MASKZ_LOAD_FLOATS_UNALIGNED(m_mask_left, localz1);
@@ -1099,7 +1099,7 @@ static inline int countpairs_theta_mocks_avx512_intrinsics_DOUBLE(const int64_t
             // at the beginning of the loop.
             for(int kbin=nthetabin-1;kbin>=1;kbin--) {
                 const AVX512_MASK m_bin_mask = AVX512_MASK_COMPARE_FLOATS(m_mask_left, m_costheta,m_costheta_upp[kbin-1],_CMP_LE_OQ);
-                npairs[kbin] += bits_set_in_avx512_mask_DOUBLE[m_bin_mask];
+                npairs[kbin] += AVX512_BIT_COUNT_UNSIGNED_INT(m_bin_mask);
                 if(need_thetaavg || need_weightavg) {
                     m_thetabin = AVX512_BLEND_INTS_WITH_MASK(m_bin_mask, m_thetabin, AVX512_SET_INT(kbin));
                 }
diff --git a/mocks/vpf_mocks/Makefile b/mocks/vpf_mocks/Makefile
index 808cc938..40df0a99 100644
--- a/mocks/vpf_mocks/Makefile
+++ b/mocks/vpf_mocks/Makefile
@@ -13,7 +13,7 @@ LIBRARY:=lib$(LIBNAME).a
 LIBSRC:=countspheres_mocks.c countspheres_mocks_impl_float.c countspheres_mocks_impl_double.c \
         $(UTILS_DIR)/gridlink_impl_float.c $(UTILS_DIR)/gridlink_impl_double.c \
         $(UTILS_DIR)/gridlink_utils_float.c $(UTILS_DIR)/gridlink_utils_double.c \
-        $(UTILS_DIR)/utils.c $(UTILS_DIR)/progressbar.c $(UTILS_DIR)/cpu_features.c $(UTILS_DIR)/avx512_calls.c \
+        $(UTILS_DIR)/utils.c $(UTILS_DIR)/progressbar.c $(UTILS_DIR)/cpu_features.c \
 	$(UTILS_DIR)/set_cosmo_dist.c $(UTILS_DIR)/cosmology_params.c
 LIBRARY_HEADERS := $(LIBNAME).h
 
diff --git a/mocks/vpf_mocks/vpf_mocks_kernels.c.src b/mocks/vpf_mocks/vpf_mocks_kernels.c.src
index b559ef76..bf153df1 100644
--- a/mocks/vpf_mocks/vpf_mocks_kernels.c.src
+++ b/mocks/vpf_mocks/vpf_mocks_kernels.c.src
@@ -53,7 +53,7 @@ static inline int vpf_mocks_avx512_intrinsics_DOUBLE(const int64_t np, DOUBLE *X
     DOUBLE *localz2 = (DOUBLE *) Z;
 
     for(int64_t j=0;j<np;j+=AVX512_NVEC) {
-        AVX512_MASK m_mask_left = (np - j) >= AVX512_NVEC ? ~0:masks_per_misalignment_value_DOUBLE[np-j];
+        AVX512_MASK m_mask_left = (np - j) >= AVX512_NVEC ? ~0U:((1U << (unsigned)(np - j)) - 1U);
 
         const AVX512_FLOATS m_x1 = AVX512_MASKZ_LOAD_FLOATS_UNALIGNED(m_mask_left, localx2);
         const AVX512_FLOATS m_y1 = AVX512_MASKZ_LOAD_FLOATS_UNALIGNED(m_mask_left, localy2);
@@ -79,12 +79,12 @@ static inline int vpf_mocks_avx512_intrinsics_DOUBLE(const int64_t np, DOUBLE *X
             const AVX512_MASK m_mask1 = AVX512_MASK_COMPARE_FLOATS(m_mask_left, m_r2,m_rupp_sqr[k],_CMP_LT_OS);
             const AVX512_MASK m_mask2 = AVX512_MASK_COMPARE_FLOATS(m_mask_left, m_r2,m_rupp_sqr[k-1],_CMP_GE_OS);
             const AVX512_MASK m_bin_mask = AVX512_MASK_BITWISE_AND(m_mask1,m_mask2);
-            counts[k] += bits_set_in_avx512_mask_DOUBLE[m_bin_mask];
+            counts[k] += AVX512_BIT_COUNT_UNSIGNED_INT(m_bin_mask);
             m_mask_left = AVX512_MASK_BITWISE_AND_NOT(m_bin_mask, m_mask_left);//ANDNOT(X, Y) -> NOT X AND Y
             if(m_mask_left == 0) {
                 break;
             } else if(k==1){
-                counts[0] += bits_set_in_avx512_mask_DOUBLE[m_mask_left];
+                counts[0] += AVX512_BIT_COUNT_UNSIGNED_INT(m_mask_left);
             }
         }
     }
diff --git a/theory/DD/Makefile b/theory/DD/Makefile
index 330b9770..d4e0985b 100644
--- a/theory/DD/Makefile
+++ b/theory/DD/Makefile
@@ -13,7 +13,7 @@ LIBRARY := lib$(LIBNAME).a
 LIBSRC  := countpairs.c countpairs_impl_double.c countpairs_impl_float.c \
          $(UTILS_DIR)/gridlink_impl_double.c $(UTILS_DIR)/gridlink_impl_float.c \
          $(UTILS_DIR)/gridlink_utils_double.c $(UTILS_DIR)/gridlink_utils_float.c \
-         $(UTILS_DIR)/utils.c $(UTILS_DIR)/progressbar.c $(UTILS_DIR)/cpu_features.c $(UTILS_DIR)/avx512_calls.c
+         $(UTILS_DIR)/utils.c $(UTILS_DIR)/progressbar.c $(UTILS_DIR)/cpu_features.c
 LIBRARY_HEADERS := $(LIBNAME).h
 
 TARGET := DD
diff --git a/theory/DD/countpairs_kernels.c.src b/theory/DD/countpairs_kernels.c.src
index c841fab8..83dcc0c9 100644
--- a/theory/DD/countpairs_kernels.c.src
+++ b/theory/DD/countpairs_kernels.c.src
@@ -147,7 +147,7 @@ static inline int countpairs_avx512_intrinsics_DOUBLE(const int64_t N0, DOUBLE *
         const AVX512_FLOATS m_zpos  = AVX512_SET_FLOAT(zpos);
 
         for(int64_t j=n_off;j<N1;j+=AVX512_NVEC) {
-            AVX512_MASK m_mask_left = (N1 - j) >= AVX512_NVEC ? ~0:masks_per_misalignment_value_DOUBLE[N1-j];
+            AVX512_MASK m_mask_left = (N1 - j) >= AVX512_NVEC ? ~0U:((1U << (unsigned)(N1 - j)) - 1U);
             const AVX512_FLOATS m_x1 = AVX512_MASKZ_LOAD_FLOATS_UNALIGNED(m_mask_left, localx1);
             const AVX512_FLOATS m_y1 = AVX512_MASKZ_LOAD_FLOATS_UNALIGNED(m_mask_left, localy1);
             const AVX512_FLOATS m_z1 = AVX512_MASKZ_LOAD_FLOATS_UNALIGNED(m_mask_left, localz1);
@@ -222,7 +222,7 @@ static inline int countpairs_avx512_intrinsics_DOUBLE(const int64_t N0, DOUBLE *
             // at the beginning of the loop.
             for(int kbin=nbin-1;kbin>=1;kbin--) {
                 const AVX512_MASK m_bin_mask = AVX512_MASK_COMPARE_FLOATS(m_mask_left, r2,m_rupp_sqr[kbin-1],_CMP_GE_OS);
-                npairs[kbin] += bits_set_in_avx512_mask_DOUBLE[m_bin_mask];
+                npairs[kbin] += AVX512_BIT_COUNT_UNSIGNED_INT(m_bin_mask);
                 if(need_rpavg || need_weightavg) {
                     m_rpbin = AVX512_BLEND_INTS_WITH_MASK(m_bin_mask, m_rpbin, AVX512_SET_INT(kbin));
                 }
diff --git a/theory/DDrppi/Makefile b/theory/DDrppi/Makefile
index 0f6f5c16..7555a8db 100644
--- a/theory/DDrppi/Makefile
+++ b/theory/DDrppi/Makefile
@@ -13,7 +13,7 @@ LIBRARY := libcountpairs_rp_pi.a
 LIBSRC  := countpairs_rp_pi.c countpairs_rp_pi_impl_double.c countpairs_rp_pi_impl_float.c \
          $(UTILS_DIR)/gridlink_impl_double.c $(UTILS_DIR)/gridlink_impl_float.c \
          $(UTILS_DIR)/gridlink_utils_double.c $(UTILS_DIR)/gridlink_utils_float.c \
-         $(UTILS_DIR)/utils.c $(UTILS_DIR)/progressbar.c $(UTILS_DIR)/cpu_features.c $(UTILS_DIR)/avx512_calls.c
+         $(UTILS_DIR)/utils.c $(UTILS_DIR)/progressbar.c $(UTILS_DIR)/cpu_features.c
 LIBRARY_HEADERS := countpairs_rp_pi.h
 
 TARGETSRC := DDrppi.c $(IO_DIR)/ftread.c $(IO_DIR)/io.c $(LIBSRC)
diff --git a/theory/DDrppi/countpairs_rp_pi_kernels.c.src b/theory/DDrppi/countpairs_rp_pi_kernels.c.src
index b0af4624..45a5a62f 100644
--- a/theory/DDrppi/countpairs_rp_pi_kernels.c.src
+++ b/theory/DDrppi/countpairs_rp_pi_kernels.c.src
@@ -162,7 +162,7 @@ static inline int countpairs_rp_pi_avx512_intrinsics_DOUBLE(const int64_t N0, DO
         }
 
         for(int64_t j=n_off;j<N1;j+=AVX512_NVEC){
-            AVX512_MASK m_mask_left = (N1 - j) >= AVX512_NVEC ? ~0:masks_per_misalignment_value_DOUBLE[N1-j];
+            AVX512_MASK m_mask_left = (N1 - j) >= AVX512_NVEC ? ~0U:((1U << (unsigned)(N1 - j)) - 1U);
             const AVX512_FLOATS m_x1 = AVX512_MASKZ_LOAD_FLOATS_UNALIGNED(m_mask_left, localx1);
             const AVX512_FLOATS m_y1 = AVX512_MASKZ_LOAD_FLOATS_UNALIGNED(m_mask_left, localy1);
             const AVX512_FLOATS m_z1 = AVX512_MASKZ_LOAD_FLOATS_UNALIGNED(m_mask_left, localz1);
diff --git a/theory/DDsmu/Makefile b/theory/DDsmu/Makefile
index d38d0947..76168c66 100644
--- a/theory/DDsmu/Makefile
+++ b/theory/DDsmu/Makefile
@@ -13,7 +13,7 @@ LIBRARY := libcountpairs_s_mu.a
 LIBSRC  := countpairs_s_mu.c countpairs_s_mu_impl_double.c countpairs_s_mu_impl_float.c \
          $(UTILS_DIR)/gridlink_impl_double.c $(UTILS_DIR)/gridlink_impl_float.c \
          $(UTILS_DIR)/gridlink_utils_float.c $(UTILS_DIR)/gridlink_utils_double.c \
-         $(UTILS_DIR)/utils.c $(UTILS_DIR)/progressbar.c $(UTILS_DIR)/cpu_features.c $(UTILS_DIR)/avx512_calls.c
+         $(UTILS_DIR)/utils.c $(UTILS_DIR)/progressbar.c $(UTILS_DIR)/cpu_features.c
 LIBRARY_HEADERS := countpairs_s_mu.h
 
 TARGETSRC := DDsmu.c $(IO_DIR)/ftread.c $(IO_DIR)/io.c $(LIBSRC)
diff --git a/theory/DDsmu/countpairs_s_mu_kernels.c.src b/theory/DDsmu/countpairs_s_mu_kernels.c.src
index cd590cca..9e567b24 100644
--- a/theory/DDsmu/countpairs_s_mu_kernels.c.src
+++ b/theory/DDsmu/countpairs_s_mu_kernels.c.src
@@ -178,7 +178,7 @@ static inline int countpairs_s_mu_avx512_intrinsics_DOUBLE(const int64_t N0, DOU
         }
 
         for(int64_t j=n_off;j<N1;j+=AVX512_NVEC){
-            AVX512_MASK m_mask_left = (N1 - j) >= AVX512_NVEC ? ~0:masks_per_misalignment_value_DOUBLE[N1-j];
+            AVX512_MASK m_mask_left = (N1 - j) >= AVX512_NVEC ? ~0U:((1U << (unsigned)(N1 - j)) - 1U);
             const AVX512_FLOATS m_x1 = AVX512_MASKZ_LOAD_FLOATS_UNALIGNED(m_mask_left, localx1);
             const AVX512_FLOATS m_y1 = AVX512_MASKZ_LOAD_FLOATS_UNALIGNED(m_mask_left, localy1);
             const AVX512_FLOATS m_z1 = AVX512_MASKZ_LOAD_FLOATS_UNALIGNED(m_mask_left, localz1);
diff --git a/theory/vpf/Makefile b/theory/vpf/Makefile
index ad3de9fc..c02c3dab 100644
--- a/theory/vpf/Makefile
+++ b/theory/vpf/Makefile
@@ -14,7 +14,7 @@ LIBSRC := countspheres.c countspheres_impl_double.c countspheres_impl_float.c \
           $(UTILS_DIR)/gridlink_impl_double.c $(UTILS_DIR)/gridlink_impl_float.c \
           $(UTILS_DIR)/gridlink_utils_float.c $(UTILS_DIR)/gridlink_utils_double.c \
           $(UTILS_DIR)/utils.c $(UTILS_DIR)/progressbar.c \
-          $(UTILS_DIR)/cpu_features.c $(UTILS_DIR)/avx512_calls.c
+          $(UTILS_DIR)/cpu_features.c
 
 TARGET := vpf
 TARGETSRC := $(TARGET).c $(IO_DIR)/ftread.c $(IO_DIR)/io.c $(LIBSRC)
diff --git a/theory/vpf/vpf_kernels.c.src b/theory/vpf/vpf_kernels.c.src
index 6e0f74c3..b1bc5a73 100644
--- a/theory/vpf/vpf_kernels.c.src
+++ b/theory/vpf/vpf_kernels.c.src
@@ -48,7 +48,7 @@ static inline int vpf_avx512_intrinsics_DOUBLE(const int64_t np,  DOUBLE * restr
     DOUBLE *localz2 = (DOUBLE *) Z;
 
     for(int64_t j=0;j<np;j+=AVX512_NVEC) {
-        AVX512_MASK m_mask_left = (np - j) >= AVX512_NVEC ? ~0:masks_per_misalignment_value_DOUBLE[np-j];
+        AVX512_MASK m_mask_left = (np - j) >= AVX512_NVEC ? ~0U:((1U << (unsigned)(np - j)) - 1U);
 
         const AVX512_FLOATS m_x1 = AVX512_MASKZ_LOAD_FLOATS_UNALIGNED(m_mask_left, localx2);
         const AVX512_FLOATS m_y1 = AVX512_MASKZ_LOAD_FLOATS_UNALIGNED(m_mask_left, localy2);
@@ -75,12 +75,12 @@ static inline int vpf_avx512_intrinsics_DOUBLE(const int64_t np,  DOUBLE * restr
           const AVX512_MASK m_mask1 = AVX512_MASK_COMPARE_FLOATS(m_mask_left, m_r2,m_rupp_sqr[k],_CMP_LT_OS);
           const AVX512_MASK m_mask2 = AVX512_MASK_COMPARE_FLOATS(m_mask_left, m_r2,m_rupp_sqr[k-1],_CMP_GE_OS);
           const AVX512_MASK m_bin_mask = AVX512_MASK_BITWISE_AND(m_mask1,m_mask2);
-          counts[k] += bits_set_in_avx512_mask_DOUBLE[m_bin_mask];
+          counts[k] += AVX512_BIT_COUNT_UNSIGNED_INT(m_bin_mask);
           m_mask_left = AVX512_MASK_BITWISE_AND_NOT(m_bin_mask, m_mask_left);//ANDNOT(X, Y) -> NOT X AND Y
           if(m_mask_left == 0) {
               break;
           } else if(k==1){
-              counts[0] += bits_set_in_avx512_mask_DOUBLE[m_mask_left];
+              counts[0] += AVX512_BIT_COUNT_UNSIGNED_INT(m_mask_left);
           }
         }
     }
diff --git a/theory/wp/Makefile b/theory/wp/Makefile
index cba558a4..0a6bbc1c 100644
--- a/theory/wp/Makefile
+++ b/theory/wp/Makefile
@@ -11,7 +11,7 @@ LIBRARY := lib$(LIBNAME).a
 LIBSRC := countpairs_wp.c countpairs_wp_impl_double.c countpairs_wp_impl_float.c \
          $(UTILS_DIR)/gridlink_impl_double.c $(UTILS_DIR)/gridlink_impl_float.c \
          $(UTILS_DIR)/gridlink_utils_float.c $(UTILS_DIR)/gridlink_utils_double.c \
-         $(UTILS_DIR)/utils.c $(UTILS_DIR)/progressbar.c $(UTILS_DIR)/cpu_features.c $(UTILS_DIR)/avx512_calls.c
+         $(UTILS_DIR)/utils.c $(UTILS_DIR)/progressbar.c $(UTILS_DIR)/cpu_features.c
 LIBRARY_HEADERS := $(LIBNAME).h
 
 TARGET := wp
diff --git a/theory/wp/wp_kernels.c.src b/theory/wp/wp_kernels.c.src
index 06b6d172..265791ce 100644
--- a/theory/wp/wp_kernels.c.src
+++ b/theory/wp/wp_kernels.c.src
@@ -156,7 +156,7 @@ static inline int wp_avx512_intrinsics_DOUBLE(DOUBLE *x0, DOUBLE *y0, DOUBLE *z0
         const AVX512_FLOATS m_zpos  = AVX512_SET_FLOAT(zpos);
         
         for(int64_t j=n_off;j<N1;j+=AVX512_NVEC) {
-            AVX512_MASK m_mask_left = (N1 - j) >= AVX512_NVEC ? ~0:masks_per_misalignment_value_DOUBLE[N1-j];
+            AVX512_MASK m_mask_left = (N1 - j) >= AVX512_NVEC ? ~0U:((1U << (unsigned)(N1 - j)) - 1U);
             const AVX512_FLOATS m_x1 = AVX512_MASKZ_LOAD_FLOATS_UNALIGNED(m_mask_left, localx1);
             const AVX512_FLOATS m_y1 = AVX512_MASKZ_LOAD_FLOATS_UNALIGNED(m_mask_left, localy1);
             const AVX512_FLOATS m_z1 = AVX512_MASKZ_LOAD_FLOATS_UNALIGNED(m_mask_left, localz1);
@@ -244,7 +244,7 @@ static inline int wp_avx512_intrinsics_DOUBLE(DOUBLE *x0, DOUBLE *y0, DOUBLE *z0
             // at the beginning of the loop. 
             for(int kbin=nbin-1;kbin>=1;kbin--) {
                 const AVX512_MASK m_bin_mask = AVX512_MASK_COMPARE_FLOATS(m_mask_left, r2,m_rupp_sqr[kbin-1],_CMP_GE_OS);
-                npairs[kbin] += bits_set_in_avx512_mask_DOUBLE[m_bin_mask];
+                npairs[kbin] += AVX512_BIT_COUNT_UNSIGNED_INT(m_bin_mask);
                 if(need_rpavg || need_weightavg) {
                     m_rpbin = AVX512_BLEND_INTS_WITH_MASK(m_bin_mask, m_rpbin, AVX512_SET_INT(kbin));
                 }
diff --git a/theory/xi/Makefile b/theory/xi/Makefile
index 153cefa8..1a4e1926 100644
--- a/theory/xi/Makefile
+++ b/theory/xi/Makefile
@@ -13,7 +13,7 @@ LIBRARY_HEADERS := $(LIBNAME).h
 LIBSRC := countpairs_xi.c countpairs_xi_impl_double.c countpairs_xi_impl_float.c  \
           $(UTILS_DIR)/gridlink_impl_double.c $(UTILS_DIR)/gridlink_impl_float.c \
           $(UTILS_DIR)/gridlink_utils_float.c $(UTILS_DIR)/gridlink_utils_double.c \
-          $(UTILS_DIR)/utils.c $(UTILS_DIR)/progressbar.c $(UTILS_DIR)/cpu_features.c $(UTILS_DIR)/avx512_calls.c
+          $(UTILS_DIR)/utils.c $(UTILS_DIR)/progressbar.c $(UTILS_DIR)/cpu_features.c
 
 TARGET := xi
 TARGETSRC := $(TARGET).c $(IO_DIR)/ftread.c $(IO_DIR)/io.c $(LIBSRC)
diff --git a/theory/xi/xi_kernels.c.src b/theory/xi/xi_kernels.c.src
index a3c404d6..71372fc1 100644
--- a/theory/xi/xi_kernels.c.src
+++ b/theory/xi/xi_kernels.c.src
@@ -146,7 +146,7 @@ static inline int xi_avx512_intrinsics_DOUBLE(DOUBLE *x1, DOUBLE *y1, DOUBLE *z1
         const AVX512_FLOATS m_zpos  = AVX512_SET_FLOAT(zpos);
 
         for(int64_t j=n_off;j<N2;j+=AVX512_NVEC) {
-            AVX512_MASK m_mask_left = (N2 - j) >= AVX512_NVEC ? ~0:masks_per_misalignment_value_DOUBLE[N2-j];
+            AVX512_MASK m_mask_left = (N2 - j) >= AVX512_NVEC ? ~0U:((1U << (unsigned)(N2 - j)) - 1U);
             const AVX512_FLOATS m_x2 = AVX512_MASKZ_LOAD_FLOATS_UNALIGNED(m_mask_left, localx2);
             const AVX512_FLOATS m_y2 = AVX512_MASKZ_LOAD_FLOATS_UNALIGNED(m_mask_left, localy2);
             const AVX512_FLOATS m_z2 = AVX512_MASKZ_LOAD_FLOATS_UNALIGNED(m_mask_left, localz2);
@@ -218,7 +218,7 @@ static inline int xi_avx512_intrinsics_DOUBLE(DOUBLE *x1, DOUBLE *y1, DOUBLE *z1
             // at the beginning of the loop.
             for(int kbin=nbin-1;kbin>=1;kbin--) {
                 const AVX512_MASK m_bin_mask = AVX512_MASK_COMPARE_FLOATS(m_mask_left, r2,m_rupp_sqr[kbin-1],_CMP_GE_OQ);
-                npairs[kbin] += bits_set_in_avx512_mask_DOUBLE[m_bin_mask];
+                npairs[kbin] += AVX512_BIT_COUNT_UNSIGNED_INT(m_bin_mask);
                 if(need_rpavg || need_weightavg) {
                     m_rpbin = AVX512_BLEND_INTS_WITH_MASK(m_bin_mask, m_rpbin, AVX512_SET_INT(kbin));
                 }
diff --git a/utils/Makefile b/utils/Makefile
index 182aa247..2b48145f 100644
--- a/utils/Makefile
+++ b/utils/Makefile
@@ -8,7 +8,7 @@ ROOT_DIR := ..
 include $(ROOT_DIR)/common.mk
 TARGETSRC   := cosmology_params.c gridlink_impl_double.c gridlink_impl_float.c gridlink_mocks_impl_float.c \
                gridlink_mocks_impl_double.c gridlink_utils_double.c gridlink_utils_float.c \
-               progressbar.c set_cosmo_dist.c utils.c cpu_features.c avx512_calls.c
+               progressbar.c set_cosmo_dist.c utils.c cpu_features.c
 TARGETOBJS  := $(TARGETSRC:.c=.o)
 INCL  := avx512_calls.h avx2_calls.h avx_calls.h sse_calls.h defs.h defs.h function_precision.h cosmology_params.h \
          cellarray_float.h cellarray_double.h cellarray.h.src \
diff --git a/utils/avx512_calls.c b/utils/avx512_calls.c
deleted file mode 100644
index 6f3f4f6e..00000000
--- a/utils/avx512_calls.c
+++ /dev/null
@@ -1,40 +0,0 @@
-/* File: avx512_calls.c */
-/*
-  This file is a part of the Corrfunc package
-  Copyright (C) 2015-- Manodeep Sinha (manodeep@gmail.com)
-  License: MIT LICENSE. See LICENSE file under the top-level
-  directory at https://github.com/manodeep/Corrfunc/
-*/
-#if defined(__AVX512F__)
-#include "avx512_calls.h"
-
-
-const int64_t bits_set_in_avx512_mask_float[] = { B16(0) };
-const uint16_t masks_per_misalignment_value_float[] = {0b1111111111111111,
-                                                       0b0000000000000001,
-                                                       0b0000000000000011,
-                                                       0b0000000000000111,
-                                                       0b0000000000001111,
-                                                       0b0000000000011111,
-                                                       0b0000000000111111,
-                                                       0b0000000001111111,
-                                                       0b0000000011111111,
-                                                       0b0000000111111111,
-                                                       0b0000001111111111,
-                                                       0b0000011111111111,
-                                                       0b0000111111111111,
-                                                       0b0001111111111111,
-                                                       0b0011111111111111,
-                                                       0b0111111111111111};
-
-const int64_t bits_set_in_avx512_mask_double[] = { B8(0) };
-const uint8_t masks_per_misalignment_value_double[] = {0b11111111, 
-                                                       0b00000001,
-                                                       0b00000011,
-                                                       0b00000111,
-                                                       0b00001111,
-                                                       0b00011111,
-                                                       0b00111111,
-                                                       0b01111111};
-
-#endif
diff --git a/utils/avx512_calls.h b/utils/avx512_calls.h
index 5ff60f73..2b81296c 100644
--- a/utils/avx512_calls.h
+++ b/utils/avx512_calls.h
@@ -42,18 +42,6 @@ extern "C" {
 #define AVX512_MASK_BITWISE_NOT(X,Y)              _mm512_knot(X) //~X
 
     
-
-  /* For setting up the array that contains the number bits set in a mask */
-  // Can be used to generate up to 16 bit lookup tables
-#   define B2(n)  n,      n+1,      n+1,      n+2
-#   define B4(n)  B2(n),  B2(n+1),  B2(n+1),  B2(n+2)
-#   define B6(n)  B4(n),  B4(n+1),  B4(n+1),  B4(n+2)
-#   define B8(n)  B6(n),  B6(n+1),  B6(n+1),  B6(n+2)
-#   define B10(n) B8(n),  B8(n+1),  B8(n+1),  B8(n+2)
-#   define B12(n) B10(n), B10(n+1), B10(n+1), B10(n+2)
-#   define B14(n) B12(n), B12(n+1), B12(n+1), B12(n+2)
-#   define B16(n) B14(0),B14(1), B14(1),   B14(2)
-
 #ifndef DOUBLE_PREC
 
 #define DOUBLE                              float  
@@ -366,12 +354,6 @@ static inline AVX512_FLOATS inv_cosine_avx512(const AVX512_FLOATS X, const int o
 
 #endif
 
-  extern const int64_t bits_set_in_avx512_mask_float[];
-  extern const uint16_t masks_per_misalignment_value_float[];
-
-  extern const int64_t bits_set_in_avx512_mask_double[];
-  extern const uint8_t masks_per_misalignment_value_double[];
-
   union int16 {
     AVX512_INTS m_ibin;
     int ibin[AVX512_NVEC];