Lines Matching defs:Widen

4154 /// Widen a vector to a larger size with the same scalar type, with the new
4179 /// Widen a vector to a larger size with the same scalar type, with the new
4193 /// Widen a mask vector type to a minimum of v8i1/v16i1 to allow use of KSHIFT
4203 /// Widen a mask vector to a minimum of v8i1/v16i1 to allow use of KSHIFT and
4455   bool Widen = !(Subtarget.hasVLX() || VT.is512BitVector());
4458   if (Widen)
4476     if (Widen)
4483   if (Widen)
4603   // Widen the vector if needed.
8724       // Widen smaller SrcVec to match VT.
20633   // Widen to 512-bits if VLX is not supported.
21272         // Widen to 512-bits.
21372     // Widen vXi32 fp_to_uint with avx512f to 512-bit source.
21404     // Widen vXi64 fp_to_uint/fp_to_sint with avx512dq to 512-bit source.
24869   // Widen to 512-bits if VLX is not supported.
25226     // Widen the vector, cast to a v2x64 type, extract the single 64-bit element
32011     // Widen the vector in input in the case of MVT::v2i32.
32742 /// Widen a vector input to a vector of NVT.  The
33488     // Widen the result with by padding with undef.
33524   // We might have generated v2f32 FMIN/FMAX operations. Widen them to v4f32.
41120   // Widen any subvector shuffle inputs we've collected.
49774       // Widen input to v16i32 so we can truncate that.
52316       // Widen the result if its narrower than 128 bits.
52829   // Widen v2i1/v4i1 stores to v8i1.
59052   // Widen to at least 8 input elements.
59192   // Widen to at least 4 input elements.
60233         // Widen to 64 bits here to get it sign extended.