aaron/ispc
1
0
Fork 0
Code Issues Pull Requests Releases Wiki Activity

Add 8-bit and 16-bit specialized NEON targets.

Browse Source 
Like SSE4-8 and SSE4-16, these use 8-bit and 16-bit values for mask
elements, respectively, and thus should generate the best code when used
for computation with datatypes of those sizes.
This commit is contained in:
Matt Pharr
2013-07-29 16:14:58 -07:00
parent b6df447b55
commit ab3b633733
12 changed files with 1561 additions and 373 deletions
Download Patch File Download Diff File Expand all files Collapse all files

3
ispc.h
View File

@@ -175,7 +175,8 @@ public:
flexible/performant of them will apear last in the enumerant. Note
also that __best_available_isa() needs to be updated if ISAs are
added or the enumerant values are reordered. */
enum ISA { NEON, SSE2, SSE4, AVX, AVX11, AVX2, GENERIC, NUM_ISAS };
enum ISA { NEON32, NEON16, NEON8, SSE2, SSE4, AVX, AVX11, AVX2, GENERIC,
NUM_ISAS };
/** Initializes the given Target pointer for a target of the given
name, if the name is a known target. Returns true if the