Re: [PATCH] vect: Verify that GET_MODE_NUNITS is greater than one.

2023-04-21 Thread Jeff Law via Gcc-patches
On 3/14/23 15:52, Michael Collison wrote: While working on autovectorizing for the RISCV port I encountered an issue where can_duplicate_and_interleave_p assumes that GET_MODE_NUNITS is a evenly divisible by two. The RISC-V target has vector modes (e.g. VNx1DImode), where GET_MODE_NUNITS is

Re: [PATCH] vect: Verify that GET_MODE_NUNITS is greater than one.

2023-03-31 Thread Richard Sandiford via Gcc-patches
Michael Collison writes: > While working on autovectorizing for the RISCV port I encountered an issue > where can_duplicate_and_interleave_p assumes that GET_MODE_NUNITS is a > evenly divisible by two. The RISC-V target has vector modes (e.g. VNx1DImode), > where GET_MODE_NUNITS is equal to one.

Re: [PATCH] vect: Verify that GET_MODE_NUNITS is greater than one.

2023-03-19 Thread Jeff Law via Gcc-patches
On 3/14/23 15:52, Michael Collison wrote: While working on autovectorizing for the RISCV port I encountered an issue where can_duplicate_and_interleave_p assumes that GET_MODE_NUNITS is a evenly divisible by two. The RISC-V target has vector modes (e.g. VNx1DImode), where GET_MODE_NUNITS is

[PATCH] vect: Verify that GET_MODE_NUNITS is greater than one.

2023-03-14 Thread Michael Collison
While working on autovectorizing for the RISCV port I encountered an issue where can_duplicate_and_interleave_p assumes that GET_MODE_NUNITS is a evenly divisible by two. The RISC-V target has vector modes (e.g. VNx1DImode), where GET_MODE_NUNITS is equal to one. Tested on RISCV and