https://bugs.kde.org/show_bug.cgi?id=372195
Carl Love <c...@us.ibm.com> changed:
What |Removed |Added
----------------------------------------------------------------------------
Status|UNCONFIRMED |CONFIRMED
Ever confirmed|0 |1
--- Comment #1 from Carl Love <c...@us.ibm.com> ---
The issue is the opc2 field is used by the xxsel instruction to specify an
additional operand. That leaves a subset of the opc2 field for the actual
opcode. The function get_VSX60_opc2(UInt opc2_full) in
VEX/priv/guest_ppc_toIR.c is used to determine the instruction in opc2 field
for instructions that have an opc1 field of 0x60. The 10-bit opc2 field is
masked to get the actual opcode. There are several different masks that are
used for the various forms. The issue is that a match was being made on the
xxsel instruction depending on the value of the additional opcode. The
function was updated to fix the issue.
--
You are receiving this mail because:
You are watching all bug changes.
