Re: [PATCH v7] RISCV: Inline subword atomic ops

[Patrick O'Neill <patrick@rivosinc.com>]

On 4/24/23 22:52, Jeff Law wrote:

>
> On 4/18/23 15:41, Patrick O'Neill wrote:
>> RISC-V has no support for subword atomic operations; code currently
>> generates libatomic library calls.
>>
>> This patch changes the default behavior to inline subword atomic calls
>> (using the same logic as the existing library call).
>> Behavior can be specified using the -minline-atomics and
>> -mno-inline-atomics command line flags.
>>
>> gcc/libgcc/config/riscv/atomic.c has the same logic implemented in asm.
>> This will need to stay for backwards compatibility and the
>> -mno-inline-atomics flag.
>>
>> 2023-04-18 Patrick O'Neill <patrick@rivosinc.com>
>>
>>     PR target/104338
>>     * riscv-protos.h: Add helper function stubs.
>>     * riscv.cc: Add helper functions for subword masking.
>>     * riscv.opt: Add command-line flag.
>>     * sync.md: Add masking logic and inline asm for fetch_and_op,
>>     fetch_and_nand, CAS, and exchange ops.
>>     * invoke.texi: Add blurb regarding command-line flag.
>>     * inline-atomics-1.c: New test.
>>     * inline-atomics-2.c: Likewise.
>>     * inline-atomics-3.c: Likewise.
>>     * inline-atomics-4.c: Likewise.
>>     * inline-atomics-5.c: Likewise.
>>     * inline-atomics-6.c: Likewise.
>>     * inline-atomics-7.c: Likewise.
>>     * inline-atomics-8.c: Likewise.
>>     * atomic.c: Add reference to duplicate logic.
> This is OK for the trunk.  I think the only question is whether or not 
> you're going to contribute to GCC regularly.  If so, we should go 
> ahead and get you write access so you can commit ACK's changes.  If 
> you're not going to be making regular contributions, then I can go 
> ahead and commit it for you.
>
> Jeff

I should be contributing regularly, so write access would be great.

Is this the correct form?
https://sourceware.org/cgi-bin/pdw/ps_form.cgi

Patrick