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/17126] tracepoints.exp testcase causing stalls/hang on ppc64


https://sourceware.org/bugzilla/show_bug.cgi?id=17126

hemkshaw at in dot ibm.com <hemkshaw at in dot ibm.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |hemkshaw at in dot ibm.com

--- Comment #5 from hemkshaw at in dot ibm.com <hemkshaw at in dot ibm.com> ---
Hi,
I am trying to reproduce this bug on a RHEL7 system. But due to some reasons,
stap is not able to find the trace points related to hcall.
# uname -r
3.10.0-123.el7.ppc64

And the debug and devel pkgs for this kernel are also installed :
# rpm -qa | grep kernel-de
kernel-devel-3.10.0-123.el7.ppc64
kernel-debug-3.10.0-123.el7.ppc64

But,
# stap -l 'kernel.trace("hcall_*")'
shows nothing!

Although, perf is able to show them,
# perf list tracepoints | grep hcall
  powerpc:hcall_entry                                [Tracepoint event]
  powerpc:hcall_exit                                 [Tracepoint event]

However, stap is able to show the events related to hrtimer and is also able to
probe into these events, viz.

# stap -l 'kernel.trace("hrtimer_*")'
kernel.trace("hrtimer_cancel")
kernel.trace("hrtimer_expire_entry")
kernel.trace("hrtimer_expire_exit")
kernel.trace("hrtimer_init")
kernel.trace("hrtimer_start")

Since, the problem is due to the combination of probing into hrtimer_* and
hcall_*, I am unable to produce this bug as systemtap is failing to probe into
hcall_* :

# stap -v -e 'probe kernel.trace("hcall_*") { printf("hcall hit!"); }'
Pass 1: parsed user script and 97 library script(s) using
153600virt/35968res/7936shr/24640data kb, in 220usr/10sys/233real ms.
semantic error: while resolving probe point: identifier 'kernel' at <input>:1:7
        source: probe kernel.trace("hcall_*") { printf("hcall hit!"); }
                      ^

semantic error: no match (similar functions: kfree, kmalloc, console, kvm_fpu,
cpu_idle)
Pass 2: analyzed script: 0 probe(s), 0 function(s), 0 embed(s), 0 global(s)
using 164416virt/46144res/11648shr/28608data kb, in 60usr/180sys/243real ms.
Pass 2: analysis failed.  [man error::pass2]

Thanks,
Hemant

-- 
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]