Lines Matching refs:NEON

4 …C++ code as we can from the SSE family of instruction sets on x86 or from NEON on ARM or from MIPS…
6 …rawing routine might be specialized for NEON but not for SSE, or might have a MIPS DSP implementat…
13 …ill write to this interface _once_, which then compiles to efficient SSE, NEON, or portable code (…
20 …NEON is by far the easiest task involved here.  Both SSE and NEON naturally work with 128-bit vect…
32 -   math written with either SSE or NEON instrinsics is still very hard to read; and
37 …oes nothing, so all `SkNf<N>` recurse down to the default `SkNf<1>`;  the NEON backend specializes…
51 … implementations.  The 3 `Sk4f` transfermodes replaced portable, SSE, and NEON implementations whi…
58 …NEON and SSE again have some overlap, and they could probably be implemented in terms of each othe…
113 …NEON when working in fixed point is that SSE works most naturally with 4 interlaced pixels at a ti…
115 …NEON backend works with 8 planar pixels, loading them with `vld4_u8` into an `uint8x8x4_t` struct …
119 Keeping `Alpha` as a single dense `uint8x8_t` plane allows the NEON backend to be much more efficie…
126 …4` and restricts the layout to always use interlaced pixels: i.e. fine for SSE, not great for NEON.
139 I'm also confident that if you're looking to use floats, `SkNf` is ready.  Do not write NEON or SSE…
141 …the stage where we should try it aware that we might need to fall back on hand-written SSE or NEON.