RFR: 8347352: RISC-V: Cleanup bitwise AND assembler routines [v3]
Robbin Ehn
rehn at openjdk.org
Wed Jan 15 07:55:29 UTC 2025
On Tue, 14 Jan 2025 13:12:49 GMT, Fei Yang <fyang at openjdk.org> wrote:
>> Hi, Please consider this small refactoring work.
>>
>> It's a bit strange that we have `Assembler::_and_imm12` and `MacroAssembler::andi`, which is quite different from friends `Assembler::ori` and `Assembler::xori`. And it doesn't seem necessary to have this `MacroAssembler::andi` which checks the immediate range. I find the immediate is within signed 12-bit range for most of the cases. One exception is in file `sharedRuntime_riscv.cpp` where I think we can do `mv` + `andr` instead.
>>
>> Testing on linux-riscv64:
>> - [x] tier1-3 and gtest:all (release)
>> - [x] hotspot:tier1 (fastdebug)
>
> Fei Yang has updated the pull request incrementally with one additional commit since the last revision:
>
> Fix build
Ship away!
-------------
Marked as reviewed by rehn (Reviewer).
PR Review: https://git.openjdk.org/jdk/pull/23008#pullrequestreview-2551754181
More information about the hotspot-dev
mailing list