Age | Commit message (Expand) | Author |
---|---|---|
2013-01-19 | On Sandybridge split unaligned 256bit stores into two xmm-sized stores. | Nadav Rotem |
2013-01-18 | On Sandybridge loading unaligned 256bits using two XMM loads (vmovups and vin... | Nadav Rotem |
2013-01-02 | AVX: Fix a bug in WidenMaskArithmetic. | Nadav Rotem |
2012-12-27 | On AVX/AVX2 the type v8i1 is legalized to v8i16, which is an XMM sized | Nadav Rotem |