public inbox for systemtap@sourceware.org
 help / color / mirror / Atom feed
From: Wenji Huang <wenji.huang@oracle.com>
To: SystemTAP <systemtap@sources.redhat.com>
Subject: [PATCH 3/3] PR4886: Check build-id in module_init function if able.
Date: Sun, 28 Sep 2008 02:26:00 -0000	[thread overview]
Message-ID: <48DEEA31.9010004@oracle.com> (raw)

There are several cases to be considered, whether build-id exists in 
debuginfo file or not, whether module is loaded or not and whether 
build-id exists in loaded module/kernel or not.
---
  runtime/sym.c |   66 
+++++++++++++++++++++++++++++++++++++++++++++++++++++++++
  translate.cxx |    1 +
  2 files changed, 67 insertions(+), 0 deletions(-)

diff --git a/runtime/sym.c b/runtime/sym.c
index 1a9e26b..38323c9 100644
--- a/runtime/sym.c
+++ b/runtime/sym.c
@@ -160,6 +160,72 @@ static const char *_stp_kallsyms_lookup(unsigned 
long addr, unsigned long *symbo
         return NULL;
  }

+/* Validate module/kernel based on build-id if there
+*  The completed case is the following combination:
+*         Debuginfo             Module                          Kernel
+*                         X                            X
+*      has build-id/not        unloaded                      has 
build-id/not
+*                              loaded && (has build-id/not)
+*
+*  NB: build-id exists only if ld>=2.18 and kernel>= 2.6.23
+*/
+static int _stp_module_check(void)
+{
+       struct _stp_module *m = NULL;
+       unsigned long notes_addr, base_addr;
+       unsigned i;
+
+       for (i = 0; i < _stp_num_modules; i++)
+       {
+               m = _stp_modules[i];
+
+               /* unloaded module */
+               if (m->notes_sect == 0) {
+                           _stp_warn("skip checking %s\n", m->name);
+                    continue;
+                }
+               if (m->build_id_len > 0) { /* build-id in debuginfo file */
+                   dbug_sym(1, "validate %s based on build-id\n", m->name);
+
+                   /* loaded module/kernel, but without build-id */
+                   if (m->notes_sect == 1) {
+                       _stp_error("missing build-id in %s\n", m->name);
+                       return 1;
+                   }
+                   /* notes end address */
+                   if (!strcmp(m->name, "kernel")) {
+                         notes_addr = m->build_id_offset;
+                         base_addr = _stp_module_relocate("kernel",
+                                                          "_stext", 0);
+                    } else {
+                         notes_addr = m->notes_sect + m->build_id_offset;
+                         base_addr = m->notes_sect;
+                   }
+                   /* notes start address */
+                   notes_addr -= m->build_id_len;
+                   if (notes_addr > base_addr) {
+                       if (memcmp(m->build_id_bits,
+                                   (unsigned char *) notes_addr,
+                                   m->build_id_len))
+                       {
+                          _stp_error("inconsistent build-id in %s with 
debuginfo\n", m->name);
+                          return 1;
+                       }
+                   } else { /* bug, shouldn't come here */
+                            _stp_error("unknown failure in checking %s\n",
+                                                               m->name);
+                            return 1;
+                          } /* end comparing */
+               } else {
+                         /* build-id in module/kernel, absent in 
debuginfo */
+                         if (m->notes_sect > 1) {
+                           _stp_error("unexpected build-id in %s\n", 
m->name);
+                           return 1;
+                         }
+                } /* end checking */
+       } /* end loop */
+       return 0;
+}

  /** Print an address symbolically.
   * @param address The address to lookup.
diff --git a/translate.cxx b/translate.cxx
index a9695f1..0ee5179 100644
--- a/translate.cxx
+++ b/translate.cxx
@@ -1113,6 +1113,7 @@ c_unparser::emit_module_init ()
    o->newline(-1) << "}";

    // XXX: perform buildid-based checking if able
+  o->newline() << "if (_stp_module_check()) rc = -EINVAL;";

    o->newline(-1) << "}";
    o->newline() << "if (rc) goto out;";
-- 
1.5.6

             reply	other threads:[~2008-09-28  2:26 UTC|newest]

Thread overview: 6+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2008-09-28  2:26 Wenji Huang [this message]
2008-09-28 13:38 ` Frank Ch. Eigler
2008-09-30 16:29   ` Elena Zannoni
2008-10-01 21:41 ` Masami Hiramatsu
2008-10-01 21:44   ` Roland McGrath
2008-10-01 21:45   ` Frank Ch. Eigler

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=48DEEA31.9010004@oracle.com \
    --to=wenji.huang@oracle.com \
    --cc=systemtap@sources.redhat.com \
    /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).