deps: cherry-pick 2f5da9a from V8 upstream

Original commit message:
  Fix the uppercasing of U+00E7(ç) and U+00F7(÷)
  Due to a typo in runtime-i18n.js, 'ç'(U+00E7) was not uppercased
  while '÷'(U+00F7) was incorrectly uppercased to '×'(U+00D7).

  Add a comprehensive test for Latin-1 supplemental block
  (U+00A0 ~ U+00FF). (they're special-cased for speed-up and
  needs to have a test for the range.).

  TEST=intl/general/case-mapping
  BUG=v8:5681

  Review-Url: https://codereview.chromium.org/2533033003
  Cr-Commit-Position: refs/heads/master@{#41331}

PR-URL: #9828
Fixes: #9785
Reviewed-By: James M Snell <jasnell@gmail.com>
Reviewed-By: Ali Ijaz Sheikh <ofrobots@google.com>
Reviewed-By: Myles Borins <myles.borins@gmail.com>

deps/v8/include/v8-version.h

@@ -11,7 +11,7 @@
 #define V8_MAJOR_VERSION 5
 #define V8_MINOR_VERSION 4
 #define V8_BUILD_NUMBER 500
-#define V8_PATCH_LEVEL 47
+#define V8_PATCH_LEVEL 48
 
 // Use 1 for candidates and 0 otherwise.
 // (Boolean macro values are not supported by all preprocessors.)

deps/v8/src/runtime/runtime-i18n.cc

@@ -931,7 +931,7 @@ inline uint16_t ToASCIIUpper(uint16_t ch) {
 inline uint16_t ToLatin1Upper(uint16_t ch) {
   DCHECK(ch != 0xDF && ch != 0xB5 && ch != 0xFF);
   return ch &
-         ~(((ch >= 'a' && ch <= 'z') || (((ch & 0xE0) == 0xE0) && ch != 0xE7))
+         ~(((ch >= 'a' && ch <= 'z') || (((ch & 0xE0) == 0xE0) && ch != 0xF7))
            << 5);
 }
 

deps/v8/test/intl/general/case-mapping.js

@@ -136,3 +136,29 @@ assertEquals("\u{10CC0}", "\u{10C80}".toLocaleLowerCase());
 assertEquals("\u{10C80}", "\u{10CC0}".toLocaleUpperCase(["tr"]));
 assertEquals("\u{10C80}", "\u{10CC0}".toLocaleUpperCase(["tr"]));
 assertEquals("\u{10CC0}", "\u{10C80}".toLocaleLowerCase());
+
+// check fast path for Latin-1 supplement (U+00A0 ~ U+00FF)
+var latin1Suppl = "\u00A0¡¢£¤¥¦§¨©ª«¬\u00AD®°±²³´µ¶·¸¹º»¼½¾¿" +
+    "ÀÁÂÃÄÅÆÇÈÉÊËÌÍÎÏÐÑÒÓÔÕÖ×ØÙÚÛÜÝÞßàáâãäåæçèéêëìíîïðñòóôõö÷øùúûüýþÿ";
+var latin1SupplLowercased = "\u00A0¡¢£¤¥¦§¨©ª«¬\u00AD®°±²³´µ¶·¸¹º»¼½¾¿" +
+    "àáâãäåæçèéêëìíîïðñòóôõö×øùúûüýþßàáâãäåæçèéêëìíîïðñòóôõö÷øùúûüýþÿ";
+var latin1SupplUppercased = "\u00A0¡¢£¤¥¦§¨©ª«¬\u00AD®°±²³´\u039C¶·¸¹º»¼½¾¿" +
+    "ÀÁÂÃÄÅÆÇÈÉÊËÌÍÎÏÐÑÒÓÔÕÖ×ØÙÚÛÜÝÞSSÀÁÂÃÄÅÆÇÈÉÊËÌÍÎÏÐÑÒÓÔÕÖ÷ØÙÚÛÜÝÞ\u0178";
+
+assertEquals(latin1SupplLowercased, latin1Suppl.toLowerCase());
+assertEquals(latin1SupplUppercased, latin1Suppl.toUpperCase());
+assertEquals(latin1SupplLowercased, latin1Suppl.toLocaleLowerCase("de"));
+assertEquals(latin1SupplUppercased, latin1Suppl.toLocaleUpperCase("de"));
+assertEquals(latin1SupplLowercased, latin1Suppl.toLocaleLowerCase("el"));
+assertEquals(latin1SupplUppercased, latin1Suppl.toLocaleUpperCase("el"));
+assertEquals(latin1SupplUppercased, latin1Suppl.toLocaleUpperCase("tr"));
+assertEquals(latin1SupplLowercased, latin1Suppl.toLocaleLowerCase("tr"));
+assertEquals(latin1SupplUppercased, latin1Suppl.toLocaleUpperCase("az"));
+assertEquals(latin1SupplLowercased, latin1Suppl.toLocaleLowerCase("az"));
+assertEquals(latin1SupplUppercased, latin1Suppl.toLocaleUpperCase("lt"));
+// Lithuanian need to have a dot-above for U+00CC(Ì) and U+00CD(Í) when
+// lowercasing.
+assertEquals("\u00A0¡¢£¤¥¦§¨©ª«¬\u00AD®°±²³´µ¶·¸¹º»¼½¾¿" +
+    "àáâãäåæçèéêëi\u0307\u0300i\u0307\u0301îïðñòóôõö×øùúûüýþß" +
+    "àáâãäåæçèéêëìíîïðñòóôõö÷øùúûüýþÿ",
+    latin1Suppl.toLocaleLowerCase("lt"));