I would add a local like const struct asmc_model *m;, and set it to the found 'model'. Then goto to the end of the function, which would do

free(model);
return (m);

This way you have free() only in one place, and all returns would be guaranteed to free the memory.

In D55395#1267017, @kib wrote:

I would add a local like const struct asmc_model *m;, and set it to the found 'model'. Then goto to the end of the function, which would do

free(model);
return (m);

This way you have free() only in one place, and all returns would be guaranteed to free the memory.

Yup -- agreed.

This is awesome, this is the similar concept i was planning on proposing,

I was trying to gather enough hardware samples from the community to better understand the patterns across different models before suggesting a generic approach. (probed from machines from 2009 to 2017)

For example, on some laptops I have worked with, I had to upgrade them to the latest macOS version supported by the hardware in order to get the SMC firmware updated.
Only after that would FreeBSD behave properly. That suggests there are firmware-level differences that may influence how far a generic model can safely go. (We can't downgrade the versions, so i can't retry)

For example, some Mac mini and laptop models expose WOL support (see D54439), while others do not. It is not totally clear to me whether capabilities are consistent enough across all models within a given family string (ex. Macmini) to safely assume the same feature set.