Re: dtrace/systemtap options

2008-06-02 Thread olafBuddenhagen
Hi, On Fri, May 30, 2008 at 01:14:34AM -0400, Andrei Barbu wrote: > SystemTap itself doesn't get to sit in the Linux kernel. Linux exposes > an in-kernel API called kprobes. That allows you to do something like > (conceptually, not real code) set_entry_probe("some_function", myfun), > that will c

Re: dtrace/systemtap options

2008-05-29 Thread Andrei Barbu
> I don't get this: Why implement module loading just for this? Can't > SystemTap be just statically linked in gnumach? Or are the actual test > functions compiled and loaded as modules in SystemTap?... > SystemTap itself doesn't get to sit in the Linux kernel. Linux exposes an in-kernel API calle

Re: dtrace/systemtap options

2008-05-29 Thread olafBuddenhagen
Hi again, On Fri, May 23, 2008 at 04:20:12PM -0400, Andrei Barbu wrote: > 2 This one is ugly, and I'd rather not do it. It involves adding > module loading support to do the same thing that SystemTap does on > Linux. Compile code, add a prelude so that it becomes a regular > module, load it in an

Re: dtrace/systemtap options

2008-05-27 Thread Samuel Thibault
Hello, Andrei Barbu, le Mon 26 May 2008 17:57:04 -0400, a écrit : > On Mon, May 26, 2008 at 5:27 PM, Samuel Thibault > > BTW, I'm not sure to understand: what does RW means? Being able to > > tinker with kernel parameters? > > > > Yup, you could potentially do so in both dtrace and systemtap; >

Re: dtrace/systemtap options

2008-05-26 Thread olafBuddenhagen
Hi, On Fri, May 23, 2008 at 04:20:12PM -0400, Andrei Barbu wrote: > I've been thinking about all of the various ways to get either Dtrace > or SystemTap working. I've come up with 4 options: > [...] On the downside this is inherently R/O, although that does > increase security significantly. I

Re: dtrace/systemtap options

2008-05-26 Thread Andrei Barbu
On Mon, May 26, 2008 at 5:27 PM, Samuel Thibault <[EMAIL PROTECTED]> wrote: > [EMAIL PROTECTED], le Mon 26 May 2008 21:25:33 +0200, a écrit : >> Maybe you actually mean that RW increases the risk of crashing?... >> (Well, what fun would there be in system development, without the >> prospect of reg

Re: dtrace/systemtap options

2008-05-26 Thread Samuel Thibault
[EMAIL PROTECTED], le Mon 26 May 2008 21:25:33 +0200, a écrit : > Maybe you actually mean that RW increases the risk of crashing?... > (Well, what fun would there be in system development, without the > prospect of regular crashes? ;-) ) BTW, I'm not sure to understand: what does RW means? Being

dtrace/systemtap options

2008-05-23 Thread Andrei Barbu
Hi, I've been thinking about all of the various ways to get either Dtrace or SystemTap working. I've come up with 4 options: 1 Dtrace would need to sit entirely out of the kernel. About 3k lines of simple code to rewrite, FSF lawyers are ok with this option. Since it must sit in userland and it