Bug 931 - Kprobes: separate stacks for kprobes handlers
Summary: Kprobes: separate stacks for kprobes handlers
Status: RESOLVED WONTFIX
Alias: None
Product: systemtap
Classification: Unclassified
Component: kprobes (show other bugs)
Version: unspecified
: P3 enhancement
Target Milestone: ---
Assignee: Jim Keniston
URL:
Keywords:
Depends on:
Blocks: 912
  Show dependency tree / graph
 
Reported: 2005-05-05 17:33 UTC by Frank Ch. Eigler
Modified: 2007-05-08 19:24 UTC (History)
0 users

See Also:
Host:
Target:
Build:
Last reconfirmed:

Attachments
Run kprobes handlers on a separate stack - i386 (1017 bytes, patch)
2005-05-11 19:49 UTC, Jim Keniston 		Details | Diff
View All Add an attachment (proposed patch, testcase, etc.)

Note You need to log in before you can comment on or make changes to this bug.
Description Frank Ch. Eigler 2005-05-05 17:33:44 UTC 
separate per-cpu stacks for executing kprobes handlers
arrange to disable interrupts during handler execution
Comment 1 Jim Keniston 2005-05-11 19:49:13 UTC 
Created attachment 484 [details]
Run kprobes handlers on a separate stack - i386

This patch is described in my SystemTap posting of 5/11/05 and in CVS:
patches/kprobe_stack (sic).
Comment 2 Jim Keniston 2005-05-11 19:50:52 UTC 
I've attached a working prototype for i386.
Comment 3 Jim Keniston 2005-11-17 22:58:12 UTC 
The necessity for this feature is diminished significantly by the SystemTap
translator's practice of statically allocating space for the handler's "local"
variables.
Comment 4 Frank Ch. Eigler 2005-11-18 17:47:38 UTC 
That practice was factored in at the time this suggestion was initially raised.
 We still use probably a few hundred bytes of stack, just with kprobes and
translator artifacts.  The question was whether even those few hundred could
pose a problem.
Comment 5 Frank Ch. Eigler 2007-05-08 20:24:20 UTC 
Systemtap probe handlers now check for a minimum of available stack space upon
entry, so maybe we don't need to pursue this any more.