Re: [PATCH] RISC-V: Add support for subword atomic loads/stores

On 6/13/24 12:58, Jeff Law wrote: On 6/12/24 6:10 PM, Patrick O'Neill wrote: Andrea Parri recently pointed out that we were emitting overly conservative fences for seq_cst atomic loads/stores. This adds support for the optimized fences specified in the PSABI: https://github.com/riscv-non-i

Re: [PATCH] RISC-V: Add support for subword atomic loads/stores

On 6/12/24 6:10 PM, Patrick O'Neill wrote: Andrea Parri recently pointed out that we were emitting overly conservative fences for seq_cst atomic loads/stores. This adds support for the optimized fences specified in the PSABI: https://github.com/riscv-non-isa/riscv-elf-psabi-doc/blob/2092568f78

Re: [PATCH] RISC-V: Add support for subword atomic loads/stores

On Wed, Jun 12, 2024 at 05:10:13PM -0700, Patrick O'Neill wrote: > Andrea Parri recently pointed out that we were emitting overly conservative > fences for seq_cst atomic loads/stores. This adds support for the optimized > fences specified in the PSABI: > https://github.com/riscv-non-isa/riscv-elf-

[PATCH] RISC-V: Add support for subword atomic loads/stores

Andrea Parri recently pointed out that we were emitting overly conservative fences for seq_cst atomic loads/stores. This adds support for the optimized fences specified in the PSABI: https://github.com/riscv-non-isa/riscv-elf-psabi-doc/blob/2092568f7896ceaa1ec0f02569b19eaa42cd51c9/riscv-atomic.adoc