Re: [PATCH] ACPI: NUMA: Only parse CFMWS at boot when CXL_ACPI is on

[Huang, Kai]

On Wed, 2026-03-04 at 19:01 -0800, dan.j.williams@xxxxxxxxx wrote:
> Gregory Price wrote:
> > On Wed, Mar 04, 2026 at 06:08:36PM -0800, dan.j.williams@xxxxxxxxx wrote:
> > > Huang, Kai wrote:
> > > 
> > > A small 2 line patch to save a few megabytes of memory, sure. A
> > > paragraph of context that almost nobody will ever care about on a go
> > > forward basis? No, thank you.
> > 
> > eh throw it in the changelog *shrug*, i think it's useful context.
> 
> Definitely great changelog fodder and yes, useful context.

Thanks for all the feedback.

So if I follow you guys correctly, the context I provided in the changelog
is still useful.  And the "save a few megabytes of memory" is also covered
since I provided some data about "Slab" in /proc/meminfo.  So to me I'll
keep the changelog as-is in v2.

Btw, I tried to build with W=1 when CXL_ACPI is off to verify what Dan said
about the acpi_parse_cfmws() will get compiled out (otherwise we will see '-
Wunused-function' warning):

  The call to acpi_table_parse_cedt() will get skipped and the code for
  acpi_parse_cfmws() will get automatically compiled out of the file.

I tried both -O2 (CC_OPTIMIZE_FOR_PERFORMANCE) and -Os
(CC_OPTIMIZE_FOR_SIZE) with W=1 and no warning was found.  I also tried both
gcc and clang and no warning was found either.

So I'll follow Dan's suggestion to make the diff smaller.

Hi Alison, Gregory, thanks for your review, please let me know if you have
any concern.

Thanks.