Re: [PATCH] Implement timekeeping for rumprun/hw (x86)

  • From: Antti Kantee <pooka@xxxxxx>
  • To: rumpkernel-users@xxxxxxxxxxxxx
  • Date: Sun, 05 Jul 2015 12:14:10 +0000

On 05/07/15 08:44, Martin Lucina wrote:

Not sure if evident from the data: that program also runs slow over
the wall clock in both modes. (not sure if expected or not)

I'm not sure what you mean by that. The wall clock value is informative in
that the test program always sleeps a fixed amount and doesn't attempt to
account for overhead, so the wall clock will diverge if the overheads are

I mean that the gettimeofday() prints don't match the physical clock on my wall as the test progresses. In other words, the test runs exhibit the problem.

Incidentally, does remote debugging on QEMU/x86_64 work for you? On my
system it seems completely broken (endless series of "program interrupted
with SIGTRAP" during BIOS boot).

Yes, but you have to attach gdb after the CPU has switched from 32 to 64bit mode (or, if you want to debug the 32bit bootstrap, you have to "set arch i386"). As far as I recall, the alternatives are to fix gdb to not get confused when the remote packet format changes from 32bit to 64bit due to the switch to long mode, or define a new remote debugging protocol and make qemu/gdb use it. Neither seemed like a particularly compelling way to get sidetracked.

Also, generally speaking, debugging under kvm has been mostly useless for me. breakpoints are not triggered, etc.

Other related posts: