On 11/09/18 21:38, Jeffrey Hugo wrote:
On 9/11/2018 2:16 PM, Jeremy Linton wrote:
If you look at the next line of code following this comment its going
to update the cache type for fully populated PPTT nodes. Although with
the suggested change its only going to activate if someone completely
fills out the node and fails to set the valid flag on the cache type.
Yes, however that case doesn't apply to the scenario we are concerned
about, doesn't seem to be fully following the PPTT spec, and seems odd
that Linux just assumes that a "fully specified" cache is unified.
Agreed, but adding code that will never get used is also not so good.
Do you have systems where this is needed ?
What I suspect is happening in the reported case is that the nodes in
the PPTT table are missing fields we consider to be important. Since
that data isn't being filled out anywhere else, so we leave the cache
type alone too. This has the effect of hiding sysfs nodes with
Also, the lack of the DATA/INST fields is based on the assumption that
the only nodes which need their type field updated are outside of the
CPU core itself so they are pretty much guaranteed to be UNIFIED. Are
you hitting this case?
Yes.Â Without this change, we hit the lscpu error in the commit message,
and get zero output about the system.Â We don't even get information
about the caches which are architecturally specified or how many cpus
are present.Â With this change, we get what we expect out of lscpu (and
also lstopo) including the cache(s) which are not architecturally
lscpu and lstopo are so broken. They just assume everything on CPU0.
If you hotplug them out, you start seeing issues. So reading and file
that doesn't exist and then bail out on other essential info though they
are present, hmmm ...
I guess I still don't understand why its important for PPTT to list, for
example, the sets/ways of a cache in all scenarios.Â In the case of a
"transparent" cache (implementation defined as not reported per section
D3.4.2 of the ARM ARM where the cache cannot be managed by SW), there
may not be valid values for sets/ways.Â I would argue its better to not
report that information, rather than provide bogus information just to
make Linux happy, which may break other OSes and provide means for which
a user to hang themselves.
While I agree presenting wrong info is not good, but one of the reasons
the cache info is present is to provide those info for some applications
to do optimization based on that(I am told so and not sure if just type
and size will be good enough) and you seem to agree with that below.
However, in the case of a transparent cache, the size/type/level/write
policy/etc (whatever the firmware provider deems relevant) might be
valuable information for the OS to make scheduling decisions, and is
certainly valuable for user space utilities for cross-machine/data
center level job scheduling.