RFR: 8332265: RISC-V: Materialize pointers faster by using a temp register [v2]

Robbin Ehn rehn at openjdk.org
Tue May 21 08:56:21 UTC 2024


On Tue, 21 May 2024 07:59:12 GMT, Ludovic Henry <luhenry at openjdk.org> wrote:

>> Robbin Ehn has updated the pull request with a new target base due to a merge or a rebase. The incremental webrev excludes the unrelated changes brought in by the merge/rebase. The pull request contains three additional commits since the last revision:
>> 
>>  - li48 -> movptr
>>  - Merge branch 'master' into 8332265
>>  - li48
>
> src/hotspot/cpu/riscv/macroAssembler_riscv.cpp line 1603:
> 
>> 1601:   } else if (NativeInstruction::is_li32_at(insn_addr)) {             // li32
>> 1602:     return get_target_of_li32(insn_addr);
>> 1603:   } else if (NativeInstruction::is_movptr2_at(insn_addr)) {          // movptr2
> 
> You could move that `else if` block right under the `NativeInstruction::is_movptr1_at` one.

Fixed

> src/hotspot/cpu/riscv/macroAssembler_riscv.cpp line 1669:
> 
>> 1667: }
>> 1668: 
>> 1669: void MacroAssembler::movptr_1(Register Rd, uint64_t imm64, int32_t &offset) {
> 
> `movptr1` instead, to make it easily searchable with `is_movptr1_at`

Fixed

> src/hotspot/cpu/riscv/macroAssembler_riscv.cpp line 1688:
> 
>> 1686: }
>> 1687: 
>> 1688: void MacroAssembler::movptr_2(Register Rd, uint64_t addr, int32_t &offset, Register tmp) {
> 
> Also `movptr2`

Fixed

-------------

PR Review Comment: https://git.openjdk.org/jdk/pull/19246#discussion_r1607911213
PR Review Comment: https://git.openjdk.org/jdk/pull/19246#discussion_r1607910772
PR Review Comment: https://git.openjdk.org/jdk/pull/19246#discussion_r1607910943


More information about the hotspot-dev mailing list