https://gcc.gnu.org/bugzilla/show_bug.cgi?id=66382
Bug ID: 66382 Summary: POWER8 Vector optimized implementation of __float128 (IEEE754 128-bit Binary Floating Point) Product: gcc Version: 6.0 Status: UNCONFIRMED Severity: enhancement Priority: P3 Component: libgcc Assignee: unassigned at gcc dot gnu.org Reporter: dje at gcc dot gnu.org Target Milestone: --- Using the enhanced Vector integer capabilities of the POWER8 ISA provide an all vector implementation of Soft-float __float128. The current implementation assumes that using 32- or 64-bit integer operations to implement __float128. POWER8 should be able to provide all the required operations using only Vector instruction. This should eliminate the any need to use GMP and additional local memory the non-volatile Vector Registers should provide all the storage capacity needed.