Re: [PATCH -fixes] dma-mapping: add default implementation to arch_dma_{set|clear}_uncached

[Yangyu Chen]

> On Jul 9, 2024, at 23:52, Robin Murphy <robin.murphy@xxxxxxx> wrote:
> On 09/07/2024 1:22 pm, Yangyu Chen wrote:
>> The only failed symbol on the kernel with `ARCH=riscv defconfig`
>> is `arch_dma_set_uncached` since the compiler requires all possible
>> values to be known. I think a pattern like in kernel/dma/direct.c:349
>> for symbol `arch_dma_clear_uncached`, which uses `if
>> (IS_ENABLE(CONFIG_ARCH_HAS_xxx)) xxx` is acceptable. But for the
>> symbol `arch_dma_set_uncached`, a complex analysis is needed for a
>> value set in a block of branches. I think we should not rely on such
>> compiler optimization in such a complex pattern.
> I'm not a compiler guy, but is it really that complex when the variable is only ever written with the same compile-time-constant value that it's already initialised with? If the optimisation pass is so focused on being able to use a conditional store instruction that it would rather emit one which has no effect either way than elide it entirely, that doesn't strike me as a particularly good optimisation :/

I confirm this is a regression for the compiler. However, I think
both the compiler and the kernel should be fixed, as I mentioned
the complexity of this pattern in the last email. Also, here is the
only failed symbol in the kernel with RISC-V defconfig.

Thanks,
Yangyu Chen