[Bug runtime/16836] New: CFI expression underflow occurs during unwinding on some kernels

[Bug runtime/16836] New: CFI expression underflow occurs during unwinding on some kernels

[lberk at redhat dot com]

https://sourceware.org/bugzilla/show_bug.cgi?id=16836

            Bug ID: 16836
           Summary: CFI expression underflow occurs during unwinding on
                    some kernels
           Product: systemtap
           Version: unspecified
            Status: NEW
          Severity: normal
          Priority: P2
         Component: runtime
          Assignee: systemtap at sourceware dot org
          Reporter: lberk at redhat dot com

Running stap -e 'probe kernel.function("__run_hrtimer") { log ("hit!");
print_backtrace(); log("done"); exit(); }' on 3.13.6-100.fc19.x86_64,
3.13.7-100.fc19.x86_64 and 3.13.7-200.fc20.x86_64 causes a WARNING: DWARF
expression stack underflow in CFI (full output below)

WARNING: DWARF expression stack underflow in CFI
hit!
 0xffffffff81091ff0 : __run_hrtimer+0x0/0x1d0 [kernel]
 0xffffffff81092877 : hrtimer_interrupt+0xf7/0x240 [kernel]
 0xffffffff81042dc7 : local_apic_timer_interrupt+0x37/0x60 [kernel]
 0xffffffff8169203f : smp_apic_timer_interrupt+0x3f/0x60 [kernel]
 0xffffffff816909dd : apic_timer_interrupt+0x6d/0x80 [kernel]
done
hit!
 0xffffffff81091ff0 : __run_hrtimer+0x0/0x1d0 [kernel]
 0xffffffff81092877 : hrtimer_interrupt+0xf7/0x240 [kernel]
 0xffffffff81042dc7 : local_apic_timer_interrupt+0x37/0x60 [kernel]
 0xffffffff8169203f : smp_apic_timer_interrupt+0x3f/0x60 [kernel]
 0xffffffff816909dd : apic_timer_interrupt+0x6d/0x80 [kernel]
done
Pass 5: run completed in 10usr/70sys/679real ms.

This reportedly runs without the warning on 3.15.0-rc0.git9.1.fc21.x86_64

-- 
You are receiving this mail because:
You are the assignee for the bug.

[Bug runtime/16836] CFI expression underflow occurs during unwinding on some kernels

[lberk at redhat dot com]

https://sourceware.org/bugzilla/show_bug.cgi?id=16836

Lukas Berk <lberk at redhat dot com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |lberk at redhat dot com,
                   |                            |mjw at redhat dot com

-- 
You are receiving this mail because:
You are the assignee for the bug.

[Bug runtime/16836] CFI expression underflow occurs during unwinding on some kernels

[tromey at redhat dot com]

https://sourceware.org/bugzilla/show_bug.cgi?id=16836

Tom Tromey <tromey at redhat dot com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |tromey at redhat dot com

-- 
You are receiving this mail because:
You are the assignee for the bug.

[Bug runtime/16836] CFI expression underflow occurs during unwinding on some kernels

[lberk at redhat dot com]

https://sourceware.org/bugzilla/show_bug.cgi?id=16836

--- Comment #1 from Lukas Berk <lberk at redhat dot com> ---
I reran the script this morning on 3.15.0-rc0.git9.1.fc21.x86_64 and
3.15.0-0.rc0.git12.1.fc21.x86_64.  The warning was present on both rawhide
kernels for me.

-- 
You are receiving this mail because:
You are the assignee for the bug.

[Bug runtime/16836] CFI expression underflow occurs during unwinding on some kernels

[mark at klomp dot org]

https://sourceware.org/bugzilla/show_bug.cgi?id=16836

Mark Wielaard <mark at klomp dot org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |mark at klomp dot org

-- 
You are receiving this mail because:
You are the assignee for the bug.

[Bug runtime/16836] CFI expression underflow occurs during unwinding on some kernels

[mjw at fedoraproject dot org]

https://sourceware.org/bugzilla/show_bug.cgi?id=16836

Mark Wielaard <mjw at fedoraproject dot org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|mjw at fedoraproject dot org       |

-- 
You are receiving this mail because:
You are the assignee for the bug.

[Bug runtime/16836] CFI expression underflow occurs during unwinding on some kernels

[barrdetwix at gmail dot com]

https://sourceware.org/bugzilla/show_bug.cgi?id=16836

barrdetwix at gmail dot com changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |barrdetwix at gmail dot com

--- Comment #2 from barrdetwix at gmail dot com ---
Hit the same issue today on a freshly built v5.8.12 with SystemTap master
(effa3dbe13fa).

-- 
You are receiving this mail because:
You are the assignee for the bug.

[Bug runtime/16836] CFI expression underflow occurs during unwinding on some kernels

[fche at redhat dot com]

https://sourceware.org/bugzilla/show_bug.cgi?id=16836

Frank Ch. Eigler <fche at redhat dot com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |fche at redhat dot com

--- Comment #3 from Frank Ch. Eigler <fche at redhat dot com> ---
I believe stack -underflow- can only occur if the DWARF CFI data is invalid, or
if we interpret it in an incorrect manner.  It'd really help to extract that
somehow.  Or try running with  

    stap -DDEBUG_UNWIND=2 [...]

-- 
You are receiving this mail because:
You are the assignee for the bug.

[Bug runtime/16836] CFI expression underflow occurs during unwinding on some kernels

[barrdetwix at gmail dot com]

https://sourceware.org/bugzilla/show_bug.cgi?id=16836

--- Comment #4 from barrdetwix at gmail dot com ---
Here's 200kB of log with -DDEBUG_UNWIND=2 leading up to the "DWARF expression
stack underflow in CFI": https://paste.mozilla.org/VewSs2i7#L8,258,3146

(Note that there's a couple non-stap lines in the long becase I'm not running
stap directly, my reproducer is a small tool that starts a target process
suspended, generates a stap script for that process, and captures stap output.
And so far the CFI underflow problem reproduces perfectly only when I run my
tool under itself.)

Happy to upload vmlinux or some other binary if that would help any.

-- 
You are receiving this mail because:
You are the assignee for the bug.

[Bug runtime/16836] CFI expression underflow occurs during unwinding on some kernels

[mark at klomp dot org]

https://sourceware.org/bugzilla/show_bug.cgi?id=16836

--- Comment #5 from Mark Wielaard <mark at klomp dot org> ---
Unfortunately this warning can only show up during expression parsing and we
would have to know which expression operation is being evaluated. Which stap
only shows with -DDEBUG_UNWIND=3 which will generate even more output...

-- 
You are receiving this mail because:
You are the assignee for the bug.