Bitcast of 64-bit immediates may need to split the immediate, not a var.

Bitcast of 64-bit immediates may need to split the immediate, not a var.

Currently, the integer immediate is legalized to a
64-bit integer register first, and then the lower/upper
parts of that register are used for the bitcast.
However, mov(64_bit_reg, imm) done by the legalization
isn't legal.

Similarly, trunc of 64-bit immediates need to take the
lower half of the immediate, not legalize to a var first.

This shifts the legalization code around.

Other cases where immediates are illegal and legalized
are idiv/div, but for those cases 64-bit operands are
handled separately via a function call. The function
call code properly splits up immediate arguments.

BUG=none
R=stichnot@chromium.org

Committed: https://gerrit.chromium.org/gerrit/gitweb?p=native_client/pnacl-subzero.git;a=commit;h=1ee3416

[Jim Stichnoth, 2014-06-23 22:40:58]

LGTM

https://codereview.chromium.org/348373005/diff/1/src/IceTargetLoweringX8632.cpp
File src/IceTargetLoweringX8632.cpp (right):

https://codereview.chromium.org/348373005/diff/1/src/IceTargetLoweringX8632.c...
src/IceTargetLoweringX8632.cpp:1638: //   FakeDef(s.f64)
I realize this comment was changed to match the code.  But I think it might be
slightly better to put the FakeDef after the "lo(s.f64) = t_lo.i32" because it
looks correct and it should reduce register pressure.  (At least a TODO to
investigate.)

[jvoung (off chromium), 2014-06-24 00:13:27]

https://codereview.chromium.org/348373005/diff/1/src/IceTargetLoweringX8632.cpp
File src/IceTargetLoweringX8632.cpp (right):

https://codereview.chromium.org/348373005/diff/1/src/IceTargetLoweringX8632.c...
src/IceTargetLoweringX8632.cpp:1638: //   FakeDef(s.f64)
On 2014/06/23 22:40:58, stichnot wrote:
> I realize this comment was changed to match the code.  But I think it might be
> slightly better to put the FakeDef after the "lo(s.f64) = t_lo.i32" because it
> looks correct and it should reduce register pressure.  (At least a TODO to
> investigate.)


I moved it to right before the _store(). The store considers Spill to be a
source/use through SpillLo, so if I move the define to after the store, the
store's use will be considered outside of Spill's live range.

[jvoung (off chromium), 2014-06-24 20:43:37]

Committed patchset #2 manually as r1ee3416 (presubmit successful).