This is the mail archive of the
systemtap@sourceware.org
mailing list for the systemtap project.
Re: Systemtap results on power64 box against kernel version 2.6.35-rc4-git5
- From: David Smith <dsmith at redhat dot com>
- To: divya <dipraksh at linux dot vnet dot ibm dot com>
- Cc: systemtap at sources dot redhat dot com, sachinp at linux dot vnet dot ibm dot com
- Date: Mon, 19 Jul 2010 12:52:36 -0500
- Subject: Re: Systemtap results on power64 box against kernel version 2.6.35-rc4-git5
- References: <4C3C6CB0.4000900@linux.vnet.ibm.com> <4C3C7DF5.90809@redhat.com>
On 07/13/2010 09:53 AM, David Smith wrote:
> On 07/13/2010 08:40 AM, divya wrote:
>> Hi Everyone,
>> Here is the result of systemtap snap test on kernel 2.6.35-rc4-git5
>>
>> Arch - ppc64
>> elfutils version - 0.148
Divya was kind enough to send me the full log file, so I've had a look
at some of these errors. Most of the errors look to be debuginfo
problems with 1 build-id problem.
>> FAIL: systemtap.examples/general/badname build
>> FAIL: systemtap.examples/general/badname run
Hmm, it looks like the test should be a bit smarter and not try to run a
script if it fails to build. Anyway, this one fails with errors like:
semantic error: failed to retrieve location attribute for local 'child'
(dieoffset: 0xf3c95c): identifier '$child' at badname.stp:16:7
source: if ($child->d_inode || $dir->i_flags & 16) next
This looks like a debuginfo problem. Can you show me the output of:
# stap -L 'kernel.function("may_create@fs/namei.c")'
>> FAIL: systemtap.examples/io/iostat-scsi run
This one failed with:
ERROR: Build-id mismatch: "sd_mod" vs. "sd_mod.ko" byte 0 (0xc1 vs 0x00)
rc 0 -14
which means that stap doesn't think your debuginfo and the actual module
match. Other stap folk can probably expound more here.
>> FAIL: systemtap.examples/network/netdev build
>> FAIL: systemtap.examples/network/netdev run
This one failed with:
semantic error: failed to retrieve location attribute for local 'dev'
(dieoffset: 0x2daa6f2): identifier '$dev' at
/usr/local/share/systemtap/tapset/networking.stp:159:29
source: dev_name = get_netdev_name($dev)
I'd bet this one is a debuginfo problem. Can you show me the output of:
# stap -L netdev.change_rx_flag
>> FAIL: systemtap.examples/network/socktop run
This one was a testcase problem that I just fixed.
> Hmm, I'm surprised the examples failed. Can you privately email me the
> systemtap.log output for systemtap.examples/check.exp?
>
>> FAIL: buildok/nfsd-detailed.stp
This one dies with:
semantic error: not accessible at this address (0x1703c): identifier
'$filp' at /usr/local/share/systemtap/tapset/nfsd.stp:1046:29^M
source: filename = __file_filename($filp)^M
This looks like a debuginfo problem. Can you show me the output of:
# stap -L nfsd.close
>> FAIL: buildok/pr10678.stp
This one dies with:
semantic error: no match while resolving probe point
module("ne2k_pci").function("ne2k_pci_open")
This is a test problem, not a real problem.
>> FAIL: buildok/scsi-detailed.stp
This one dies errors like:
semantic error: not accessible at this address (0x10e34): identifier
'$cmd' at /usr/local/share/systemtap/tapset/scsi.stp:133:12^M
source: host_no = $cmd->device->host->host_no^M
This looks like another debuginfo problem. Can you show me the output of:
# stap -L scsi.iodone
--
David Smith
dsmith@redhat.com
Red Hat
http://www.redhat.com
256.217.0141 (direct)
256.837.0057 (fax)