[PATCH] D11564: [X86][AVX512bw] add convert i16 to i8 and unpack intrinsics

[Michael Kuperstein]

mkuper added inline comments.

================
Comment at: lib/Headers/avx512bwintrin.h:1297
@@ -1296,3 +1296,3 @@
 
-static __inline__ __m512i __attribute__((__always_inline__, __nodebug__))
+static __inline__ __m512i  __DEFAULT_FN_ATTRS
 _mm512_maddubs_epi16 (__m512i __X, __m512i __Y) {
----------------
Can these fixes be a separate commit?

================
Comment at: lib/Headers/avx512bwintrin.h:1369
@@ +1368,3 @@
+               (__v32qi)
+               _mm256_setzero_si256 (),
+               __M);
----------------
Sometimes the call is on the same line as the cast, sometimes not. Looks odd, style-wise.

================
Comment at: lib/Headers/avx512bwintrin.h:1391
@@ +1390,3 @@
+                (__v32qi)
+                _mm256_setzero_si256 (),
+                __M);
----------------
Extra space before the ()

================
Comment at: lib/Headers/avx512bwintrin.h:1412
@@ +1411,3 @@
+              (__v32qi)
+              _mm256_setzero_si256 (),
+              __M);
----------------
Same here.

================
Comment at: lib/Headers/avx512bwintrin.h:1421
@@ +1420,3 @@
+                 (__v64qi)
+                 _mm512_setzero_qi (),
+                 (__mmask64) -1);
----------------
Same here.

================
Comment at: lib/Headers/avx512bwintrin.h:1448
@@ +1447,3 @@
+                 (__v32hi)
+                 _mm512_setzero_hi (),
+                 (__mmask32) -1);
----------------
Same here.
There are more below, can you just find them all and remove the extra spaces?


http://reviews.llvm.org/D11564