RFR: 8346160: Fix -Wzero-as-null-pointer-constant warnings from explicit casts

[Stefan Karlsson]

On Fri, 13 Dec 2024 17:23:05 GMT, Kim Barrett <kbarrett at openjdk.org> wrote:

> Please review this change that removes -Wzero-as-null-pointer-constant
> warnings from explicit casts of a literal 0 to a pointer type.  Different
> versions of gcc seem to warn about different cases changed here, but all of
> these showed up as a warning in some testing, whether Oracle CI, GHA sanity
> tests, or local cross-builds using Oracle's devkits for that.
> 
> Some of the changes simply replace a cast of 0 with nullptr.
> 
> The UNIX_PATH_MAX macro definitions in attachListener_{aix,posix}.cpp is
> changed to use a C++11 feature for obtaining the size of a non-static data
> member without requiring an instance.
> https://www.open-std.org/jtc1/sc22/wg21/docs/papers/2007/n2253.html
> I guess I need to propose an update to the style guide to mention this feature.
> 
> In CompressedKlassPointers::initialize, casting
> CompressedClassSpaceBaseAddress to a pointer type was triggering the warning
> in release builds for some reason. That's not a literal 0, even though it *is*
> a constant 0 in a release build. Changed to avoid that cast rather than trying
> to argue with some versions of the compiler.
> 
> Although different gcc versions on different platforms seemed to vary whether
> warning about each of these, this actually seems to be all of the explict
> casts of a literal 0 to a pointer in HotSpot.  Though because of the varied
> behaviors, new ones might slip in later.
> 
> Testing: mach5 tier1, GHA sanity tests, both with and without the warning enabled.

Marked as reviewed by stefank (Reviewer).

-------------

PR Review: https://git.openjdk.org/jdk/pull/22740#pullrequestreview-2502818364