Re: [RFC][Patch 2/2] markers: example of irq regular kernel markers

[fche at redhat dot com (Frank Ch. Eigler)]

Steven Rostedt <rostedt@goodmis.org> writes:

> [...]
>> That is not so.  They are far from panaceanic, but printf formats are
>> useful for type checked simple scalars, which we can extract and use
>> for purposes other than printf like operations.
>
> It doesn't help much if you mix a pid and prio of a task_struct, and then
> use the prio to find the actual task, or try setting another task priority
> to the pid.

Right, though the same is true for a tracer client flipping around
"next" and "prev" pointer values.

>[...]
>> > Passing in a pointer to the structure being traced should be enough
>> > for all tracers.
>>
>> On the contrary, we have explained why *this is not so*.  Using raw
>> general structure pointers in impractical for some tracers.
>
> The thing that those tracers need is something that can be stored in the
> kernel that can easily extract the needed information.

Well sure, but who is to do that storage & extraction?  Some code the
marker site maintainer needs to write for each marker?


> [...]
>> Maybe a solution could involve some restrictions on the generalities.
>> For example, can we narrow down the number of different scalar +
>> pointer types to a fixed handful?  Can we tolerate type-safety being
>> provided by families of function declarations rather than one generic
>> one?
>
> I'm all for restricting this, I even suggested something similar a while
> ago (http://lists.openwall.net/linux-kernel/2006/10/07/21). No, I'm not
> pushing that solution, that solution was only to bring out more ideas.

Parts of that approach have a lot of merit.  Perhaps we should spend
some time cataloguing the types of all the lttng/logdev/blktrace
"markers" and their nearby pointers.  Maybe it's only a few dozen
types altogether.

- FChE