Re: [PATCH v3] rs6000: Fix vector parity support [PR108699]

2023-04-03 Thread Segher Boessenkool
On Mon, Mar 20, 2023 at 02:31:31PM +0800, Kewen.Lin wrote: > The failures on the original failed case builtin-bitops-1.c > and the associated test case pr108699.c here show that the > current support of parity vector mode is wrong on Power. > The hardware insns vprtyb[wdq] which operate on the leas

[PATCH v3] rs6000: Fix vector parity support [PR108699]

2023-03-19 Thread Kewen.Lin via Gcc-patches
Hi, The failures on the original failed case builtin-bitops-1.c and the associated test case pr108699.c here show that the current support of parity vector mode is wrong on Power. The hardware insns vprtyb[wdq] which operate on the least significant bit of each byte per element, they doesn't match