[PATCH 1/1] riscv: Enable ARCH_HAS_FAST_MULTIPLIER for RV64I

[Chenxi Mao]

Hi Palmer:

Thanks for your reply.

Frankly, I didn't test ARCH_HAS_FAST_MULTIPLIER on RV32,

so I cannot put it in RISCV platform.

I am trying to comparing ARM64 with Riscv to find out more optimization

configurations.

I suggest to enable ARCH_HAS_FAST_MULTIPLIER on RV64 first.

If someone else evaluate this on RV32, we could move it to RISCV platform.


Chenxi


On 2020/7/21 上午9:17, Palmer Dabbelt wrote:
> On Wed, 08 Jul 2020 22:19:22 PDT (-0700), maochenxi at eswin.com wrote:
>> Enable ARCH_HAS_FAST_MULTIPLIER on RV64I
>> which works fine on GCC-9.3 and GCC-10.1
>>
>> PS2: remove ARCH_SUPPORTS_INT128 because of RV64I already enabled.
>>
>> Signed-off-by: Chenxi Mao <maochenxi at eswin.com>
>> ---
>>  arch/riscv/Kconfig | 1 +
>>  1 file changed, 1 insertion(+)
>>
>> diff --git a/arch/riscv/Kconfig b/arch/riscv/Kconfig
>> index 128192e14ff2..84e6777fecad 100644
>> --- a/arch/riscv/Kconfig
>> +++ b/arch/riscv/Kconfig
>> @@ -202,6 +202,7 @@ config ARCH_RV64I
>>      bool "RV64I"
>>      select 64BIT
>>      select ARCH_SUPPORTS_INT128 if CC_HAS_INT128 && GCC_VERSION >= 50000
>> +    select ARCH_HAS_FAST_MULTIPLIER
>>      select HAVE_DYNAMIC_FTRACE if MMU
>>      select HAVE_DYNAMIC_FTRACE_WITH_REGS if HAVE_DYNAMIC_FTRACE
>>      select HAVE_FTRACE_MCOUNT_RECORD
>
> Ah, thanks -- this one didn't show up when I was looking at the last one.  I
> think we can put the fast multiplier on rv32 and rv64, there shouldn't be any
> difference there.  I guess in theory we should be sticking this all in some
> sort of "platform type" optimization flags, but that's probably bit much for
> now.