Add icmp and fcmp instructions to Subzero bitcode reader.

Add icmp and fcmp instructions to Subzero bitcode reader.

BUG= https://code.google.com/p/nativeclient/issues/detail?id=3894
R=jvoung@chromium.org, stichnot@chromium.org

Committed: https://gerrit.chromium.org/gerrit/gitweb?p=native_client/pnacl-subzero.git;a=commit;h=83f9f0c

[Karl, 2014-09-04 19:39:12]

kschimpf@google.com changed reviewers:
+ jvoung@chromium.org, stichnot@chromium.org

[Jim Stichnoth, 2014-09-04 20:30:33]

lgtm

https://codereview.chromium.org/543793003/diff/1/src/IceTypes.def
File src/IceTypes.def (right):

https://codereview.chromium.org/543793003/diff/1/src/IceTypes.def#newcode43
src/IceTypes.def:43: //   CR - Result type of compare instruction for argument
type
I'm not thrilled about adding a new column whose values could be synthesized
from other columns (maybe also requiring the other table).  But I don't see a
good and efficient way to implement it...

[jvoung (off chromium), 2014-09-04 21:14:39]

Small nit, otherwise LGTM

https://codereview.chromium.org/543793003/diff/1/src/IceTypes.cpp
File src/IceTypes.cpp (right):

https://codereview.chromium.org/543793003/diff/1/src/IceTypes.cpp#newcode48
src/IceTypes.cpp:48: #define X(tag, IsVec, IsInt, IsFloat, IsIntArith,
ComareResult)                \
CompareResult

[Karl, 2014-09-05 15:31:02]

Committed patchset #2 (id:20001) manually as 83f9f0c (presubmit successful).

[Karl, 2014-09-05 15:31:10]

https://codereview.chromium.org/543793003/diff/1/src/IceTypes.cpp
File src/IceTypes.cpp (right):

https://codereview.chromium.org/543793003/diff/1/src/IceTypes.cpp#newcode48
src/IceTypes.cpp:48: #define X(tag, IsVec, IsInt, IsFloat, IsIntArith,
ComareResult)                \
On 2014/09/04 21:14:39, jvoung wrote:
> CompareResult

Done.

https://codereview.chromium.org/543793003/diff/1/src/IceTypes.def
File src/IceTypes.def (right):

https://codereview.chromium.org/543793003/diff/1/src/IceTypes.def#newcode43
src/IceTypes.def:43: //   CR - Result type of compare instruction for argument
type
On 2014/09/04 20:30:33, stichnot wrote:
> I'm not thrilled about adding a new column whose values could be synthesized
> from other columns (maybe also requiring the other table).  But I don't see a
> good and efficient way to implement it...

Acknowledged.