Re: igc, macOS avoiding signals

[Gerd Möllmann]

Eli Zaretskii <eliz@gnu.org> writes:

>> From: Gerd Möllmann <gerd.moellmann@gmail.com>
>> Cc: eller.helmut@gmail.com,  pipcet@protonmail.com,  spd@toadstyle.org,
>>   emacs-devel@gnu.org
>> Date: Sat, 04 Jan 2025 10:02:07 +0100
>> 
>> Eli Zaretskii <eliz@gnu.org> writes:
>> 
>> > Why does time matter?  A profile shows the percentage of the samples
>> > that we found ourselves in every function.  Time is never of any
>> > importance, AFAIU.  That's why the frequency of SIGPROF does not
>> > matter very much, except for very low percent values.  IOW, the
>> > frequency of SIGPROF can explain why we miss some functions, but
>> > that's all.
>> 
>> Well, when I read "42% so-and-so" in the profile result, I always read
>> that as "42% of the time spent in so-and-so".
>
> But that's not an accurate interpretation, is it?  My interpretation
> of 42% is "we found 42% of samples to be inside the function
> so-and-so."

Strictly speaking it's not accurate, but practically it is. But be that
as it may, I just wanted to plant that thought.