Fix typos

8 files changed, 11 insertions, 11 deletions

diff --git a/c/README.md b/c/README.md
index 259af2a..965d8c7 100644
--- a/c/README.md
+++ b/c/README.md
@@ -256,7 +256,7 @@ gcc -shared -O3 -o libblake3.so blake3.c blake3_dispatch.c blake3_portable.c \
 Note above that building `blake3_avx512.c` requires both `-mavx512f` and
 `-mavx512vl` under GCC and Clang. Under MSVC, the single `/arch:AVX512`
 flag is sufficient. The MSVC equivalent of `-mavx2` is `/arch:AVX2`.
-MSVC enables SSE2 and SSE4.1 by defaut, and it doesn't have a
+MSVC enables SSE2 and SSE4.1 by default, and it doesn't have a
 corresponding flag.
 
 If you want to omit SIMD code entirely, you need to explicitly disable
diff --git a/c/blake3_avx2.c b/c/blake3_avx2.c
index e76aa1a..381e7c4 100644
--- a/c/blake3_avx2.c
+++ b/c/blake3_avx2.c
@@ -167,7 +167,7 @@ INLINE void transpose_vecs(__m256i vecs[DEGREE]) {
   __m256i gh_0145 = _mm256_unpacklo_epi32(vecs[6], vecs[7]);
   __m256i gh_2367 = _mm256_unpackhi_epi32(vecs[6], vecs[7]);
 
-  // Interleave 64-bit lates. The low unpack is lanes 00/22 and the high is
+  // Interleave 64-bit lanes. The low unpack is lanes 00/22 and the high is
   // 11/33.
   __m256i abcd_04 = _mm256_unpacklo_epi64(ab_0145, cd_0145);
   __m256i abcd_15 = _mm256_unpackhi_epi64(ab_0145, cd_0145);
diff --git a/c/blake3_avx512.c b/c/blake3_avx512.c
index 334d82d..d6b1ae9 100644
--- a/c/blake3_avx512.c
+++ b/c/blake3_avx512.c
@@ -429,7 +429,7 @@ INLINE void round_fn4(__m128i v[16], __m128i m[16], size_t r) {
 }
 
 INLINE void transpose_vecs_128(__m128i vecs[4]) {
-  // Interleave 32-bit lates. The low unpack is lanes 00/11 and the high is
+  // Interleave 32-bit lanes. The low unpack is lanes 00/11 and the high is
   // 22/33. Note that this doesn't split the vector into two lanes, as the
   // AVX2 counterparts do.
   __m128i ab_01 = _mm_unpacklo_epi32(vecs[0], vecs[1]);
@@ -684,7 +684,7 @@ INLINE void transpose_vecs_256(__m256i vecs[8]) {
   __m256i gh_0145 = _mm256_unpacklo_epi32(vecs[6], vecs[7]);
   __m256i gh_2367 = _mm256_unpackhi_epi32(vecs[6], vecs[7]);
 
-  // Interleave 64-bit lates. The low unpack is lanes 00/22 and the high is
+  // Interleave 64-bit lanes. The low unpack is lanes 00/22 and the high is
   // 11/33.
   __m256i abcd_04 = _mm256_unpacklo_epi64(ab_0145, cd_0145);
   __m256i abcd_15 = _mm256_unpackhi_epi64(ab_0145, cd_0145);
@@ -959,7 +959,7 @@ INLINE void transpose_vecs_512(__m512i vecs[16]) {
   __m512i op_0 = _mm512_unpacklo_epi32(vecs[14], vecs[15]);
   __m512i op_2 = _mm512_unpackhi_epi32(vecs[14], vecs[15]);
 
-  // Interleave 64-bit lates. The _0 unpack is lanes
+  // Interleave 64-bit lanes. The _0 unpack is lanes
   // 0/0/0/0/4/4/4/4/8/8/8/8/12/12/12/12, the _1 unpack is lanes
   // 1/1/1/1/5/5/5/5/9/9/9/9/13/13/13/13, the _2 unpack is lanes
   // 2/2/2/2/6/6/6/6/10/10/10/10/14/14/14/14, and the _3 unpack is lanes
diff --git a/c/blake3_c_rust_bindings/build.rs b/c/blake3_c_rust_bindings/build.rs
index 98f8396..624dbb9 100644
--- a/c/blake3_c_rust_bindings/build.rs
+++ b/c/blake3_c_rust_bindings/build.rs
@@ -110,7 +110,7 @@ fn main() -> Result<(), Box<dyn std::error::Error>> {
         if is_windows_msvc() {
             // /arch:SSE2 is the default on x86 and undefined on x86_64:
             // https://docs.microsoft.com/en-us/cpp/build/reference/arch-x86
-            // It also includes SSE4.1 intrisincs:
+            // It also includes SSE4.1 intrinsics:
             // https://stackoverflow.com/a/32183222/823869
         } else {
             sse2_build.flag("-msse2");
@@ -122,7 +122,7 @@ fn main() -> Result<(), Box<dyn std::error::Error>> {
         if is_windows_msvc() {
             // /arch:SSE2 is the default on x86 and undefined on x86_64:
             // https://docs.microsoft.com/en-us/cpp/build/reference/arch-x86
-            // It also includes SSE4.1 intrisincs:
+            // It also includes SSE4.1 intrinsics:
             // https://stackoverflow.com/a/32183222/823869
         } else {
             sse41_build.flag("-msse4.1");
diff --git a/c/blake3_c_rust_bindings/src/test.rs b/c/blake3_c_rust_bindings/src/test.rs
index 28ec34b..1fc077c 100644
--- a/c/blake3_c_rust_bindings/src/test.rs
+++ b/c/blake3_c_rust_bindings/src/test.rs
@@ -63,7 +63,7 @@ pub const TEST_KEY_WORDS: [u32; 8] = [
 ];
 
 // Paint the input with a repeating byte pattern. We use a cycle length of 251,
-// because that's the largets prime number less than 256. This makes it
+// because that's the largest prime number less than 256. This makes it
 // unlikely to swapping any two adjacent input blocks or chunks will give the
 // same answer.
 fn paint_test_input(buf: &mut [u8]) {
diff --git a/c/blake3_sse2.c b/c/blake3_sse2.c
index f4449ac..691e1c6 100644
--- a/c/blake3_sse2.c
+++ b/c/blake3_sse2.c
@@ -396,7 +396,7 @@ INLINE void round_fn(__m128i v[16], __m128i m[16], size_t r) {
 }
 
 INLINE void transpose_vecs(__m128i vecs[DEGREE]) {
-  // Interleave 32-bit lates. The low unpack is lanes 00/11 and the high is
+  // Interleave 32-bit lanes. The low unpack is lanes 00/11 and the high is
   // 22/33. Note that this doesn't split the vector into two lanes, as the
   // AVX2 counterparts do.
   __m128i ab_01 = _mm_unpacklo_epi32(vecs[0], vecs[1]);
diff --git a/c/blake3_sse41.c b/c/blake3_sse41.c
index 87a8dae..4653a85 100644
--- a/c/blake3_sse41.c
+++ b/c/blake3_sse41.c
@@ -390,7 +390,7 @@ INLINE void round_fn(__m128i v[16], __m128i m[16], size_t r) {
 }
 
 INLINE void transpose_vecs(__m128i vecs[DEGREE]) {
-  // Interleave 32-bit lates. The low unpack is lanes 00/11 and the high is
+  // Interleave 32-bit lanes. The low unpack is lanes 00/11 and the high is
   // 22/33. Note that this doesn't split the vector into two lanes, as the
   // AVX2 counterparts do.
   __m128i ab_01 = _mm_unpacklo_epi32(vecs[0], vecs[1]);
diff --git a/c/test.py b/c/test.py
index b0b1929..98b1c3d 100755
--- a/c/test.py
+++ b/c/test.py
@@ -19,7 +19,7 @@ def run_blake3(args, input):
 
 
 # Fill the input with a repeating byte pattern. We use a cycle length of 251,
-# because that's the largets prime number less than 256. This makes it unlikely
+# because that's the largest prime number less than 256. This makes it unlikely
 # to swapping any two adjacent input blocks or chunks will give the same
 # answer.
 def make_test_input(length):