Re: [PATCH v3 00/16] Introduce and use generic parity16/32/64 helper

Hi Yury,

On Fri, Mar 07, 2025 at 10:55:13AM -0500, Yury Norov wrote:
> On Fri, Mar 07, 2025 at 07:57:48AM +0100, Jiri Slaby wrote:
> > On 06. 03. 25, 17:25, Kuan-Wei Chiu wrote:
> > > Several parts of the kernel contain redundant implementations of parity
> > > calculations for 16/32/64-bit values. Introduces generic
> > > parity16/32/64() helpers in bitops.h, providing a standardized
> > > and optimized implementation.
> > > 
> > > Subsequent patches refactor various kernel components to replace
> > > open-coded parity calculations with the new helpers, reducing code
> > > duplication and improving maintainability.
> > > 
> > > Co-developed-by: Yu-Chun Lin <[email protected]>
> > > Signed-off-by: Yu-Chun Lin <[email protected]>
> > > Signed-off-by: Kuan-Wei Chiu <[email protected]>
> > > ---
> > > In v3, I use parityXX() instead of the parity() macro since the
> > > parity() macro may generate suboptimal code and requires special hacks
> > > to make GCC happy. If anyone still prefers a single parity() macro,
> > > please let me know.
> > 
> > What is suboptimal and where exactly it matters? Have you actually measured
> > it?
> 
> I asked exactly this question at least 3 times, and have never
> received perf tests or asm listings - nothing. I've never received
> any comments from driver maintainers about how performance of the
> parity() is important for them, as well.
> 
To be clear, I use parityXX() was mainly because you dislike the >>
16 >> 16 hack, and I dislike the #if gcc #else hack—not due to
performance or generated code considerations.

> With the absence of _any_ feedback, I'm not going to take this series,
> of course, for the reason: overengineering.
> 
I'm quite surprised that three separate one-line functions are
considered overengineering compared to a multi-line approach that
requires special handling to satisfy gcc.

> With that said, the simplest way would be replacing parity8(u8) with
> parity(u64) 'one size fits all' thing. I even made a one extra step,
> suggesting a macro that would generate a better code for smaller types
> with almost no extra maintenance burden. This is another acceptable
> option to me.
> 
I'm fine with unifying everything to a single parity(u64) function.
Before I submit the next version, please let me know if anyone has
objections to this approach.

Regards,
Kuan-Wei