Re: git: bdc0f7678257 - main - hwpstate_intel: Fix i386 build

Hi,

Thanks for the correction! I swear that I've tried to build on i386 and 
clang doesn't feel unhappy on this. It might be the problem of the cache.

I try to do a clean build and it looks great on my new patch.

Besides, https://reviews.freebsd.org/D58301 is required to do a DEFAULT 
i386 build. If nobody is unhappy about this. I will commit this tomorrow.

  7/19/26 21:42, Jessica Clarke wrote:
> On 19 Jul 2026, at 14:40, Jessica Clarke <jrtc27@freebsd.org> wrote:
>> On 19 Jul 2026, at 14:36, ShengYi Hung <aokblast@FreeBSD.org> wrote:
>>> The branch main has been updated by aokblast:
>>>
>>> URL: https://cgit.FreeBSD.org/src/commit/?id=bdc0f7678257eaa739b9c816285504470e71e3de
>>>
>>> commit bdc0f7678257eaa739b9c816285504470e71e3de
>>> Author:     Harry Schmalzb <freebsd@omnilan.de>
>>> AuthorDate: 2026-07-13 05:29:34 +0000
>>> Commit:     ShengYi Hung <aokblast@FreeBSD.org>
>>> CommitDate: 2026-07-19 13:36:00 +0000
>>>
>>>    hwpstate_intel: Fix i386 build
>>>
>>>    Reviewed by:    olce
>>>    Fixes:  7b26353a59d6
>>>    MFC after:      3 days
>>>    Sponsored by:   The FreeBSD Foundation
>>>    Differential Revision: https://reviews.freebsd.org/D58208
>>> ---
>>> sys/x86/cpufreq/hwpstate_intel.c | 8 ++++++++
>>> 1 file changed, 8 insertions(+)
>>>
>>> diff --git a/sys/x86/cpufreq/hwpstate_intel.c b/sys/x86/cpufreq/hwpstate_intel.c
>>> index 59d7a155fda8..56a33fa88e66 100644
>>> --- a/sys/x86/cpufreq/hwpstate_intel.c
>>> +++ b/sys/x86/cpufreq/hwpstate_intel.c
>>> @@ -352,13 +352,19 @@ intel_hwpstate_hybrid_cb(void *ctx)
>>> {
>>> uint32_t *small_cores = ctx;
>>>
>>> +#ifdef __i386__
>>> + (void)small_cores;
>>> +#else
>>> atomic_add_32(small_cores, PCPU_GET(small_core));
>>> +#endif
>>> }
>> Hi,
>> The whole point of my suggestion was that this would be the only part
>> of the diff needed. By stubbing out the function, small_cores is never
>> incremented, and so the code ifndef’ed out below is dead without the
>> need for as many ifndefs.
> (And if that doesn’t work then you should go back to your original
> diff. But this version here is the worst of both worlds, since now
> intel_hwpstate_hybrid_cb is an unused static function, I believe.)