On Sun, 10 Sep 2023 16:39:30 GMT, 温绍锦 <d...@openjdk.org> wrote:
> https://bugs.openjdk.org/browse/JDK-8310929 > > @TheRealMDoerr Feedback: > > > We're getting test failures on AIX: > compiler/intrinsics/Test8215792.java > compiler/intrinsics/string/TestStringIntrinsics.java > runtime/CompactStrings/TestMethodNames.java > runtime/StringIntrinsic/StringIndexOfChar.java > Is there a problem with Big Endian? src/java.base/share/classes/java/lang/StringUTF16.java line 1632: > 1630: private static int inflatePacked(int v) { > 1631: int packed = (int) StringLatin1.PACKED_DIGITS[v]; > 1632: return ((packed & 0xFF) << HI_BYTE_SHIFT) I'm not sure this is correct. Compare `StringUTF16::putChar` where these constants are used to shift _right_ to extract the equivalent byte from a value: val[index++] = (byte)(c >> HI_BYTE_SHIFT); val[index] = (byte)(c >> LO_BYTE_SHIFT); I.e., when inflating a `byte` `0xaa` to a `char` we end up with 0xaa00 on big-endian. Since `HI_BYTE_SHIFT` is 8 on big-endian and 0 on little-endian I guess this might work: ```return ((packed & 0xFF) << 16 + HI_BYTE_SHIFT) | ((packed & 0xFF00) << HI_BYTE_SHIFT)``` .. but we really need to prototype and test this out thoroughly on a big-endian system. I second @RogerRiggs notion that the best course of action right now is to back out #14699 and redo it with big-endianness issues resolved. ------------- PR Review Comment: https://git.openjdk.org/jdk/pull/15652#discussion_r1320874615
