Lines Matching full:and

6  * This library is free software; you can redistribute it and/or
804 /* Integer 8 and 16-bit dot-product.
842      * for vector lengths beyond that this must be SVE and we know        \
1181  * Note that EQ doesn't signal InvalidOp for QNaNs but GE and GT do.
1405  * non-fused multiply-and-subtract.
1801  * the fused ops below they assume accumulate both from and into Vd.
2153  * Convert float16 to float32, raising no exceptions and
3034      * mode and denormal-flushing, and we do unfused multiplies and
3035      * additions with intermediate rounding of all products and sums.
3036      * For EBF = 1, we honour FPCR rounding mode and denormal-flushing bits,
3037      * and we perform a fused two-way sum-of-products without intermediate
3066      * Extract each BFloat16 from the element pair, and shift
3125          * and the add with a single rounding operation.  Emulate this
3127          * the second multiply as fused multiply-add, and rounding to