public inbox for systemtap@sourceware.org
 help / color / mirror / Atom feed
From: David Smith <dsmith@redhat.com>
To: "Frank Ch. Eigler" <fche@redhat.com>
Cc: "Theodore Ts'o" <tytso@mit.edu>, systemtap@sources.redhat.com
Subject: Re: Latest systemtap fails when used with latest Linux kernel
Date: Tue, 06 Jan 2009 20:30:00 -0000	[thread overview]
Message-ID: <4963BF60.2050902@redhat.com> (raw)
In-Reply-To: <y0m8wpojn3k.fsf@ton.toronto.redhat.com>

Frank Ch. Eigler wrote:
> "Theodore Ts'o" <tytso@mit.edu> writes:
> 
>> Using the latest Systemtap (commit be21f2db) compiling a 2.6.28-git7
>> kernel, I get this failure....
>> [...]
> 
> Initial results (using an O=DIR x86_64 build) indicate that the kbuild
> machinery was broken with respect to the include2/ symlink directory
> in the build tree.  It was not updated when commit
> bb8985586b7a906e116db835c64773b7a7d51663 moved a bunch of files from
> include/asm/x86/ to arch/include/asm/.

I tested this on the latest rawhide kernel,
2.6.29-0.12.rc0.git7.fc11.x86_64.  I'm seeing 2 classes of problems:

1) autoconf problems - the autoconf tests are failing to find headers
included like "#include <linux/namei.h>".  (Here the problem *might* be
that the module sources are compiled with '-Iinclude' specified, but the
autoconf tests don't get compiled that way for some reason.)

To work around this one, you can specify the needed autoconfs on the
stap command line, like this:

# stap -DSTAPCONF_X86_UNIREGS -DSTAPCONF_INODE_PRIVATE
-DSTAPCONF_NAMEIDATA_CLEANUP -v -e 'probe begin { printf("hello\n") }'

(obviously this isn't even a short term solution, but it does kill lots
of compile errors)

2) with the -DSTAPCONF* added as above, then a real kernel change causes
us problems:

In file included from /usr/local/share/systemtap/runtime/print.c:17,
                 from /usr/local/share/systemtap/runtime/runtime.h:84,
                 from /tmp/stapnALnka/stap_30776.c:41:
/usr/local/share/systemtap/runtime/transport/transport.c: In function
‘_stp_transport_init’:
/usr/local/share/systemtap/runtime/transport/transport.c:214: error:
‘struct task_struct’ has no member named ‘uid’
/usr/local/share/systemtap/runtime/transport/transport.c:215: error:
‘struct task_struct’ has no member named ‘gid’

Handling this might entail adding another autoconf test looking for the
'cred' struct in task_struct (as mentioned by bug #9699).

-- 
David Smith
dsmith@redhat.com
Red Hat
http://www.redhat.com
256.217.0141 (direct)
256.837.0057 (fax)

      reply	other threads:[~2009-01-06 20:30 UTC|newest]

Thread overview: 7+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2009-01-06  7:46 Theodore Ts'o
2009-01-06  8:50 ` Srikar Dronamraju
2009-01-06 17:49 ` Frank Ch. Eigler
2009-01-06 20:45   ` Masami Hiramatsu
2009-01-06 23:09     ` Masami Hiramatsu
2009-01-06 18:39 ` Frank Ch. Eigler
2009-01-06 20:30   ` David Smith [this message]

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=4963BF60.2050902@redhat.com \
    --to=dsmith@redhat.com \
    --cc=fche@redhat.com \
    --cc=systemtap@sources.redhat.com \
    --cc=tytso@mit.edu \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).