Optional SIMD strchrnul (#594)

Continuing #580. `strchrnul` is used by `strchr`, `strcspn` and many
other places in libc. The implementation is a matchup between `strlen`
and `memchr`.

Author: ncruces

libc-top-half/musl/src/string/strchrnul.c

@@ -2,6 +2,10 @@
 #include <stdint.h>
 #include <limits.h>
 
+#ifdef __wasm_simd128__
+#include <wasm_simd128.h>
+#endif
+
 #define ALIGN (sizeof(size_t))
 #define ONES ((size_t)-1/UCHAR_MAX)
 #define HIGHS (ONES * (UCHAR_MAX/2+1))
@@ -12,6 +16,49 @@ char *__strchrnul(const char *s, int c)
 	c = (unsigned char)c;
 	if (!c) return (char *)s + strlen(s);
 
+#if defined(__wasm_simd128__) && defined(__wasilibc_simd_string)
+	// Skip Clang 19 and Clang 20 which have a bug (llvm/llvm-project#146574)
+	// which results in an ICE when inline assembly is used with a vector result.
+#if __clang_major__ != 19 && __clang_major__ != 20
+	// Note that reading before/after the allocation of a pointer is UB in
+	// C, so inline assembly is used to generate the exact machine
+	// instruction we want with opaque semantics to the compiler to avoid
+	// the UB.
+	uintptr_t align = (uintptr_t)s % sizeof(v128_t);
+	uintptr_t addr = (uintptr_t)s - align;
+	v128_t vc = wasm_i8x16_splat(c);
+
+	for (;;) {
+		v128_t v;
+		__asm__ (
+			"local.get %1\n"
+			"v128.load 0\n"
+			"local.set %0\n"
+			: "=r"(v)
+			: "r"(addr)
+			: "memory");
+		const v128_t cmp = wasm_i8x16_eq(v, (v128_t){}) | wasm_i8x16_eq(v, vc);
+		// Bitmask is slow on AArch64, any_true is much faster.
+		if (wasm_v128_any_true(cmp)) {
+			// Clear the bits corresponding to align (little-endian)
+			// so we can count trailing zeros.
+			int mask = wasm_i8x16_bitmask(cmp) >> align << align;
+			// At least one bit will be set, unless align cleared them.
+			// Knowing this helps the compiler if it unrolls the loop.
+			__builtin_assume(mask || align);
+			// If the mask became zero because of align,
+			// it's as if we didn't find anything.
+			if (mask) {
+				// Find the offset of the first one bit (little-endian).
+				return (char *)s + (addr - (uintptr_t)s + __builtin_ctz(mask));
+			}
+		}
+		align = 0;
+		addr += sizeof(v128_t);
+	}
+#endif
+#endif
+
 #ifdef __GNUC__
 	typedef size_t __attribute__((__may_alias__)) word;
 	const word *w;

test/src/misc/memchr.c

@@ -15,8 +15,8 @@ void test(char *ptr, size_t length, void *want) {
 int main(void) {
   char *const LIMIT = (char *)(__builtin_wasm_memory_size(0) * PAGESIZE);
 
-  for (size_t length = 0; length < 64; length++) {
-    for (size_t alignment = 0; alignment < 24; alignment++) {
+  for (ptrdiff_t length = 0; length < 64; length++) {
+    for (ptrdiff_t alignment = 0; alignment < 24; alignment++) {
       for (ptrdiff_t pos = -2; pos < length + 2; pos++) {
         // Create a buffer with the given length, at a pointer with the given
         // alignment. Using the offset LIMIT - PAGESIZE - 8 means many buffers
@@ -26,10 +26,10 @@ int main(void) {
         char *ptr = LIMIT - PAGESIZE - 8 + alignment;
         memset(LIMIT - 2 * PAGESIZE, 0, 2 * PAGESIZE);
         memset(ptr, 5, length);
-        ptr[pos] = 7;
 
         // The first instance of the character is found.
         if (pos >= 0) ptr[pos + 2] = 7;
+        ptr[pos] = 7;
 
         // The character is found if it's within range.
         test(ptr, length, 0 <= pos && pos < length ? &ptr[pos] : NULL);

test/src/misc/strchrnul.c

@@ -0,0 +1,58 @@
+//! add-flags.py(LDFLAGS): -Wl,--stack-first -Wl,--initial-memory=327680
+
+#define _GNU_SOURCE
+
+#include <__macro_PAGESIZE.h>
+#include <stddef.h>
+#include <stdio.h>
+#include <string.h>
+
+void test(char *ptr, char *want) {
+  char *got = strchrnul(ptr, 7);
+  if (got != want) {
+    printf("strchrnul(%p, 7) = %p, want %p\n", ptr, got, want);
+  }
+}
+
+int main(void) {
+  char *const LIMIT = (char *)(__builtin_wasm_memory_size(0) * PAGESIZE);
+
+  for (ptrdiff_t length = 0; length < 64; length++) {
+    for (ptrdiff_t alignment = 0; alignment < 24; alignment++) {
+      for (ptrdiff_t pos = -2; pos < length + 2; pos++) {
+        // Create a buffer with the given length, at a pointer with the given
+        // alignment. Using the offset LIMIT - PAGESIZE - 8 means many buffers
+        // will straddle a (Wasm, and likely OS) page boundary. Place the
+        // character to find at every position in the buffer, including just
+        // prior to it and after its end.
+        char *ptr = LIMIT - PAGESIZE - 8 + alignment;
+        memset(LIMIT - 2 * PAGESIZE, 0, 2 * PAGESIZE);
+        memset(ptr, 5, length);
+
+        // The first instance of the character is found.
+        if (pos >= 0) ptr[pos + 2] = 7;
+        ptr[pos] = 7;
+        ptr[length] = 0;
+
+        // The character is found if it's within range.
+        test(ptr, 0 <= pos && pos < length ? &ptr[pos] : &ptr[length]);
+      }
+    }
+
+    // We need space for the terminator.
+    if (length == 0) continue;
+
+    // Ensure we never read past the end of memory.
+    char *ptr = LIMIT - length;
+    memset(LIMIT - 2 * PAGESIZE, 0, 2 * PAGESIZE);
+    memset(ptr, 5, length);
+
+    ptr[length - 1] = 7;
+    test(ptr, &ptr[length - 1]);
+
+    ptr[length - 1] = 0;
+    test(ptr, &ptr[length - 1]);
+  }
+
+  return 0;
+}