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]

[Bug runtime/14026] print_ubacktrace doesn't resolve the symbol name


http://sourceware.org/bugzilla/show_bug.cgi?id=14026

--- Comment #22 from Mark Wielaard <mjw at redhat dot com> 2012-05-07 08:42:40 UTC ---
(In reply to comment #20)
> The code you quote refers to the struct uprobe->vaddr as the intended original
> address of the breakpoint.  Perhaps that address is being munged.

I guess that could be happening, but how?

This is done this way to have an architecture neutral way of setting REG_IP to
the actual probe address. But if it is unreliable then we might just have to
make this architecture specific and have some ADJUST_REG_IP_BREAKPOINT macro
that does the right thing per architecture. This might be needed anyway for the
inode based uprobe handler which currently has:

  // XXX: Can't set SET_REG_IP; we don't actually know the relocated address.

> Have we seen yet a version of the run with -DDEBUG_UPROBES and stap -p3 output?
>  Which uprobes/utrace version was being used?  (stap-report?)

No, that would be useful.

BTW. I just checked in some cleanup of the DWARF unwinder code:

commit 44c8a286464a6196bbf44fdeaea9ccee23fa4b8c
Author: Mark Wielaard <mjw@redhat.com>
Date:   Mon May 7 10:20:21 2012 +0200

    Cleanups and fixes for DWARF unwinder.

    Add explict comments and set_*_rule functions for different states.
    Do data alignment in processCFI, not afterwards in unwind_frame.
    Remove unnecessary UNW_DEFAULT_RA. Make unwind_item state union
    explicit about (unsigned) reg versus (signed) offset. Same for
    unwind_reg_state cfa.

It does fix some corner cases related to calculated offset rules, but that
isn't the issue here. But it would be nice if future tests could be done
against a tree that has this commit in, to make the results a little easier to
interpret.

-- 
Configure bugmail: http://sourceware.org/bugzilla/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the assignee for the bug.


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