RFR 8162458 Buffer view implementations use incorrect offset for Unsafe access
Alan.Bateman at oracle.com
Thu Jul 28 10:40:14 UTC 2016
On 27/07/2016 12:47, Paul Sandoz wrote:
> I made an embarrassing mistake in the fix for
> All Buffer implementations should leverage Unsafe unaligned accessors
> The offset calculation for Unsafe access was incorrect, it’s easy to get confused because for heap buffers the offset is relative to the array, and for direct buffers the address (which can update for slices/duplicates). Disturbingly all existing tests were passing both for core and hotspot when i pushed to hs.
> As a penance i wrote a combinatorial test for buffer views to navigate the twisty maze of heap/direct, aligned/unaligned, little/big endian for accessing binary data and views from the source buffer.
> Please review:
> (This may be a duplicate of ).
> Test has been verified to fail with the existing code. Focused JPRT runs pass, but i will kick off core/hotspot runs later on today.
It is disturbing that it wasn't caught by the original tests. The new
test does appear to overlap with existing tests but I think that is
okay. One small request is to trim down the really long lines as they
are annoying when doing side-by-side diffs.
In any case, the offsets in the updated code look right for me but tests
is the only way to catch issues.
More information about the core-libs-dev