Commit Graph

8 Commits

Author SHA1 Message Date
Andreas Kling
79e79ad1d2 Kernel: Don't blindly dereference Process::executable()
When setting up profiling, don't assume that the profiled process has
a valid executable.
2020-02-27 10:41:32 +01:00
Andreas Kling
d7a13dbaa7 Kernel: Reset profiling state on exec() (but keep it going)
We now log the new executable on exec() and throw away all the samples
we've accumulated so far. But profiling keeps going.
2020-02-22 10:54:50 +01:00
Andreas Kling
983b4bd9f2 Kernel+ProfileViewer: Move symbolication to userspace for time profiles
This makes the time profiles look like the memory profiles so we can
use the userspace symbolication code in ProfileViewer.
2020-02-22 10:09:54 +01:00
Andreas Kling
f9a138aa4b Kernel: Commit the profiling sample buffer memory up front
This avoids getting page faults while storing samples in the timer IRQ.
2020-02-21 15:49:37 +01:00
Andreas Kling
a356e48150 Kernel: Move all code into the Kernel namespace 2020-02-16 01:27:42 +01:00
Andreas Kling
94ca55cefd Meta: Add license header to source files
As suggested by Joshua, this commit adds the 2-clause BSD license as a
comment block to the top of every source file.

For the first pass, I've just added myself for simplicity. I encourage
everyone to add themselves as copyright holders of any file they've
added or modified in some significant way. If I've added myself in
error somewhere, feel free to replace it with the appropriate copyright
holder instead.

Going forward, all new source files should include a license header.
2020-01-18 09:45:54 +01:00
Andreas Kling
0f393148da Kernel: Separate out the symbol offsets in profile output
Instead of saying "main +39" and "main +57" etc, we now have a separate
field in /proc/profile for the offset-into-the-symbol.
2019-12-12 21:59:47 +01:00
Andreas Kling
b32e961a84 Kernel: Implement a simple process time profiler
The kernel now supports basic profiling of all the threads in a process
by calling profiling_enable(pid_t). You finish the profiling by calling
profiling_disable(pid_t).

This all works by recording thread stacks when the timer interrupt
fires and the current thread is in a process being profiled.
Note that symbolication is deferred until profiling_disable() to avoid
adding more noise than necessary to the profile.

A simple "/bin/profile" command is included here that can be used to
start/stop profiling like so:

    $ profile 10 on
    ... wait ...
    $ profile 10 off

After a profile has been recorded, it can be fetched in /proc/profile

There are various limits (or "bugs") on this mechanism at the moment:

- Only one process can be profiled at a time.
- We allocate 8MB for the samples, if you use more space, things will
  not work, and probably break a bit.
- Things will probably fall apart if the profiled process dies during
  profiling, or while extracing /proc/profile
2019-12-11 20:36:56 +01:00