Re: [musl] [PATCH v3] sysconf: add _SC_NPROCESSORS_CONF support

[Vincent Donnefort <vincent.donnefort@arm.com>]

[...]

> > > If they're mandated stable interfaces that "happen" to have the exact
> > > data we need, I don't see what the objection to using them is. A
> > > better question would be whether access to them is restricted in any
> > > hardening profiles. At least they don't seem to be hidden by the
> > > hidepid mount option.
> > 
> > Indeed the function hasn't changed for 10y, which is I guess somewhat stable
> > and it is currently walking through all the possible CPUs (which is what we want
> > to count). Also, this file is currently always present whenever procfs is
> > mounted.
> > 
> > Nonetheless, as this is human-readable, nothing mandates the format. And as an
> > example, on my desktop machine, with 448 allocated CPUS, the first line of
> > /proc/softirqs (the line that contains all the CPUs) is 4949 long. The
> > "possible" mask describes same information with only 6 characters.
> 
> It's not just a matter of age and de facto stability. Kernel policy is
> that the procfs files are stable interfaces. Despite being "human
> readable", they're also intended to be readable by, and actually read
> by, utilities such as the procps suite of tools, "top"-type programs,
> etc.
>

What I wanted to emphasis is the cost of reading that interface vs the CPU mask
in the sysfs. The only gain is for a system with a _CONF user, where some CPUs
have been hotunplug'ed, and where the sysfs isn't mounted while the procfs is.
This can surely happen, but compared with the cost of parsing a long string
(see the ~x1000 in the example above) it doesn't seem a good trade-off to me.