ext/standard: Skip mbrlen() for ASCII bytes in php_mblen()

UPGRADING

@@ -96,6 +96,9 @@ PHP 8.7 UPGRADE NOTES
 ========================================
 
 - Standard:
+  . Improved performance of fgetcsv(), str_getcsv(), escapeshellarg() and
+    escapeshellcmd() in locales where ASCII characters are single bytes
+    (C, UTF-8 and single-byte locales).
   . Improved performance of strcspn() and strpbrk() when searching longer strings
     with a single-byte character mask.
   . Improved performance of array_chunk() when not preserving keys.

UPGRADING.INTERNALS

@@ -75,6 +75,9 @@ PHP 8.7 INTERNALS UPGRADE NOTES
 - Standard:
   . php_explode() now takes a HashTable* instead of a zval* to store the
     resulting parts. It already expected the zval* to be IS_ARRAY.
+  . php_mblen() and php_mb_reset() are inline functions. php_mb_reset() must be
+    called before the first php_mblen() of a string; the conversion state is
+    kept in BG(mblen_state) in NTS builds too.
 
 ========================
 4. OpCode changes

ext/standard/basic_functions.h

@@ -98,10 +98,9 @@ typedef struct _php_basic_globals {
 	HashTable url_adapt_output_hosts_ht;
 	HashTable *user_filter_map;
 
-	/* file.c */
-#if defined(_REENTRANT)
+	/* php_mblen() */
 	mbstate_t mblen_state;
-#endif
+	bool mblen_ascii_singletons;
 
 	int umask;
 	zend_long unserialize_max_depth;

ext/standard/exec.c

@@ -287,11 +287,13 @@ PHPAPI zend_string *php_escape_shell_cmd(const zend_string *unescaped_cmd)
 
 	cmd = zend_string_safe_alloc(2, l, 0, 0);
 
+	php_mb_reset();
 	for (x = 0, y = 0; x < l; x++) {
 		int mb_len = php_mblen(str + x, (l - x));
 
 		/* skip non-valid multibyte characters */
 		if (mb_len < 0) {
+			php_mb_reset();
 			continue;
 		} else if (mb_len > 1) {
 			memcpy(ZSTR_VAL(cmd) + y, str + x, mb_len);
@@ -400,11 +402,13 @@ PHPAPI zend_string *php_escape_shell_arg(const zend_string *unescaped_arg)
 	ZSTR_VAL(cmd)[y++] = '\'';
 #endif
 
+	php_mb_reset();
 	for (x = 0; x < l; x++) {
 		int mb_len = php_mblen(str + x, (l - x));
 
 		/* skip non-valid multibyte characters */
 		if (mb_len < 0) {
+			php_mb_reset();
 			continue;
 		} else if (mb_len > 1) {
 			memcpy(ZSTR_VAL(cmd) + y, str + x, mb_len);

ext/standard/php_string.h

@@ -57,16 +57,31 @@ PHPAPI size_t php_strcspn(const char *s1, const char *s2, const char *s1_end, co
 
 PHPAPI bool php_binary_string_shuffle(php_random_algo_with_state engine, char *str, zend_long len);
 
-#ifdef _REENTRANT
-# ifdef PHP_WIN32
-#  include <wchar.h>
-# endif
-# define php_mblen(ptr, len) ((int) mbrlen(ptr, len, &BG(mblen_state)))
-# define php_mb_reset() memset(&BG(mblen_state), 0, sizeof(BG(mblen_state)))
-#else
-# define php_mblen(ptr, len) mblen(ptr, len)
-# define php_mb_reset() php_ignore_value(mblen(NULL, 0))
-#endif
+#include <wchar.h>
+/* Starts a new string. A byte below 0x80 is one character when the locale uses ASCII
+ * characters as singletons */
+static zend_always_inline void php_mb_reset_ex(mbstate_t *state, bool *ascii_singletons)
+{
+	memset(state, 0, sizeof(*state));
+	*ascii_singletons = CG(ascii_compatible_locale);
+}
+
+static zend_always_inline int php_mblen_ex(const char *ptr, size_t len, mbstate_t *state, bool ascii_singletons)
+{
+	if (len > 0 && *ptr != '\0' && (unsigned char) *ptr < 0x80 && ascii_singletons) {
+		return 1;
+	}
+	int result = (int) mbrlen(ptr, len, state);
+	if (result == 0 && *ptr != '\0') {
+		/* A buffered character was flushed without consuming the byte; decode it from the initial state */
+		memset(state, 0, sizeof(*state));
+		result = (int) mbrlen(ptr, len, state);
+	}
+	return result;
+}
+
+#define php_mblen(ptr, len) php_mblen_ex(ptr, len, &BG(mblen_state), BG(mblen_ascii_singletons))
+#define php_mb_reset() php_mb_reset_ex(&BG(mblen_state), &BG(mblen_ascii_singletons))
 
 #define PHP_STR_PAD_LEFT		0
 #define PHP_STR_PAD_RIGHT		1

ext/standard/string.c

@@ -1482,6 +1482,7 @@ PHPAPI zend_string *php_basename(const char *s, size_t len, const char *suffix,
 		 * State 1 is everything else. */
 		int state = 0;
 
+		php_mb_reset();
 		basename_start = s;
 		basename_end = s;
 		while (len > 0) {