This is the mail archive of the systemtap@sourceware.org mailing list for the systemtap project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

Re: [PATCH] Linux Kernel Markers 0.5 for Linux 2.6.17 (with probe management)


* Mathieu Desnoyers <compudj@krystal.dyndns.org> wrote:

> I clearly expressed my position in the previous emails, so did you. 
> You argued about a use of tracing that is not relevant to my vision of 
> reality, which is :
> 
> - Embedded systems developers won't want a breakpoint-based probe

are you arguing that i'm trying to force breakpoint-based probing on 
you? I dont. In fact i explicitly mentioned that i'd accept and support 
a 5-byte NOP in the body of the marker, in the following mails:

    "just go for [...] the 5-NOP variant"
      http://marc.theaimsgroup.com/?l=linux-kernel&m=115859771924187&w=2
        (my reply to your second proposal)

    "or at most one NOP"
      http://marc.theaimsgroup.com/?l=linux-kernel&m=115865412332230&w=2
        (my reply to your third proposal)

    "at most a NOP inserted"
      http://marc.theaimsgroup.com/?l=linux-kernel&m=115886524224874&w=2
        (my reply to your fifth proposal)

That enables the probe to be turned into a function call - not an INT3 
breakpoint. Does it take some effort to implement that on your part? 
Yes, of course, but getting code upstream is never easy, /especially/ in 
cases where most of the users wont use a particular feature.

> - High performance computing users won't want a breakpoint-based probe

I am not forcing breakpoint-based probing, at all. I dont want _static, 
build-time function call based_ probing, and there is a big difference. 
And one reason why i want to avoid "static, build-time function call 
based probing" is because high-performance computing users dont want any 
overhead at all in the kernel fastpath.

> - djprobe is far away from being in an acceptable state on 
>   architectures with very inconvenient erratas (x86).

djprobes over a NOP marker are perfectly usable and safe: just add a 
simple constraint to them to only allow a djprobes insertion if it 
replaces a 5-byte NOP.

> - kprobe and djprobe cannot access local variables in every cases

it is possible with the marker mechanism i outlined before:

  http://marc.theaimsgroup.com/?l=linux-kernel&m=115886524224874&w=2

have i missed to address any concern of yours?

	Ingo


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]