Add instruction alignment tests to unit tests.

Add instruction alignment tests to unit tests.

BUG=None
R=jvoung@chromium.org, stichnot@chromium.org

Committed: https://gerrit.chromium.org/gerrit/gitweb?p=native_client/pnacl-subzero.git;a=commit;h=af238b25224c684aac5b434c45d6f1c1cf42a1ee

[Karl, 2015-01-09 23:06:03]

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

[jvoung (off chromium), 2015-01-12 18:10:07]

https://codereview.chromium.org/848473002/diff/40001/src/PNaClTranslator.cpp
File src/PNaClTranslator.cpp (right):

https://codereview.chromium.org/848473002/diff/40001/src/PNaClTranslator.cpp#...
src/PNaClTranslator.cpp:1188: << ". Found: 2**" << AlignPower;
"2**... AlignPower - 1"

https://codereview.chromium.org/848473002/diff/40001/unittest/IceParseInstsTe...
File unittest/IceParseInstsTest.cpp (right):

https://codereview.chromium.org/848473002/diff/40001/unittest/IceParseInstsTe...
unittest/IceParseInstsTest.cpp:143: // Note: alignment stored as 0 or
log2(Alignment)+1.
Nit: probably don't need to repeat the note about alignment everywhere, or you
could move this to a helper function like

getEncodedAlignment...(30)?

https://codereview.chromium.org/848473002/diff/40001/unittest/IceParseInstsTe...
unittest/IceParseInstsTest.cpp:407: const uint64_t ReplaceIndex = 10; // index
for FUNC_CODE_INST_LOAD
STORE

https://codereview.chromium.org/848473002/diff/40001/unittest/IceParseInstsTe...
unittest/IceParseInstsTest.cpp:457:
EXPECT_TRUE(DumpMunger.runTestForAssembly("Bad store alignment 4", Align4,
Shouldn't be "Bad" if it returns true

https://codereview.chromium.org/848473002/diff/40001/unittest/IceParseInstsTe...
unittest/IceParseInstsTest.cpp:473: "Error(62:4): store: Illegal alignment for
float. Expects: 4\n",
I guess technically 4 or 1 =)

[Karl, 2015-01-14 22:03:08]

https://codereview.chromium.org/848473002/diff/40001/src/PNaClTranslator.cpp
File src/PNaClTranslator.cpp (right):

https://codereview.chromium.org/848473002/diff/40001/src/PNaClTranslator.cpp#...
src/PNaClTranslator.cpp:1188: << ". Found: 2**" << AlignPower;
On 2015/01/12 18:10:06, jvoung wrote:
> "2**... AlignPower - 1"

Done.

https://codereview.chromium.org/848473002/diff/40001/unittest/IceParseInstsTe...
File unittest/IceParseInstsTest.cpp (right):

https://codereview.chromium.org/848473002/diff/40001/unittest/IceParseInstsTe...
unittest/IceParseInstsTest.cpp:143: // Note: alignment stored as 0 or
log2(Alignment)+1.
On 2015/01/12 18:10:06, jvoung wrote:
> Nit: probably don't need to repeat the note about alignment everywhere, or you
> could move this to a helper function like
> 
> getEncodedAlignment...(30)?

Done.

https://codereview.chromium.org/848473002/diff/40001/unittest/IceParseInstsTe...
unittest/IceParseInstsTest.cpp:407: const uint64_t ReplaceIndex = 10; // index
for FUNC_CODE_INST_LOAD
On 2015/01/12 18:10:06, jvoung wrote:
> STORE

Done.

https://codereview.chromium.org/848473002/diff/40001/unittest/IceParseInstsTe...
unittest/IceParseInstsTest.cpp:457:
EXPECT_TRUE(DumpMunger.runTestForAssembly("Bad store alignment 4", Align4,
On 2015/01/12 18:10:06, jvoung wrote:
> Shouldn't be "Bad" if it returns true

Done.

https://codereview.chromium.org/848473002/diff/40001/unittest/IceParseInstsTe...
unittest/IceParseInstsTest.cpp:473: "Error(62:4): store: Illegal alignment for
float. Expects: 4\n",
On 2015/01/12 18:10:06, jvoung wrote:
> I guess technically 4 or 1 =)

Acknowledged.

This problem needs to be fixed in LLVM, so I will patch separately.

[jvoung (off chromium), 2015-01-14 22:23:18]

lgtm

[Jim Stichnoth, 2015-01-14 22:23:56]

also lgtm (sorry that I had nothing to add...)

[Karl, 2015-01-20 20:19:24]

Committed patchset #5 (id:80001) manually as
af238b25224c684aac5b434c45d6f1c1cf42a1ee (presubmit successful).