RFR: JDK-8051990 - Uninitialised memory in jdk/src/share/native/java/lang/fdlibm/src/k_standard.c

[Alan Bateman]

On 10/11/2014 19:09, Mark Sheppard wrote:
> Hi
>
> please oblige and review the follow minor change
>
> diff -r c0d1026bff6f src/java.base/share/native/libfdlibm/k_standard.c
> --- a/src/java.base/share/native/libfdlibm/k_standard.c Tue Nov 04 
> 15:10:38 2014 +0000
> +++ b/src/java.base/share/native/libfdlibm/k_standard.c Mon Nov 10 
> 19:02:57 2014 +0000
> @@ -1,6 +1,6 @@
>
>  /*
> - * Copyright (c) 1998, 2001, Oracle and/or its affiliates. All rights 
> reserved.
> + * Copyright (c) 1998, 2014, Oracle and/or its affiliates. All rights 
> reserved.
>   * DO NOT ALTER OR REMOVE COPYRIGHT NOTICES OR THIS FILE HEADER.
>   *
>   * This code is free software; you can redistribute it and/or modify it
> @@ -106,6 +106,7 @@
>  #endif
>          exc.arg1 = x;
>          exc.arg2 = y;
> +        exc.retval = zero;
>          switch(type) {
>              case 1:
>                  /* acos(|x|>1) */
I wonder if we should be feeding changes to fdlibm upstream, Joe Darcy 
might know.

So is the real issue here that the switch doesn't have a default 
statement to catch a possible mis-use of this function? I just wonder if 
that might be better than setting retval twice (although it might get 
optimized away sometimes).

-Alan