linux-kernel - Re: [BUG] Kprobes fails on 2.6.25-rc3-mm1 (x86) systems, if CONFIG_DEBUG

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-Id: <200803061903.09003.srinivasa@in.ibm.com>
Date:	Thu, 6 Mar 2008 19:03:08 +0530
From:	Srinivasa DS <srinivasa@...ibm.com>
To:	Mathieu Desnoyers <mathieu.desnoyers@...ymtl.ca>
Cc:	Andrew Morton <akpm@...ux-foundation.org>,
	linux-kernel@...r.kernel.org, ananth@...ibm.com,
	Jim Keniston <jkenisto@...ibm.com>, srikar@...ux.vnet.ibm.com,
	SystemTAP <systemtap@...rces.redhat.com>
Subject: Re: [BUG] Kprobes fails on 2.6.25-rc3-mm1 (x86) systems, if CONFIG_DEBUG_RODATA is set.

On Thursday 06 March 2008 06:25:51 pm Mathieu Desnoyers wrote:
> * Srinivasa DS (srinivasa@...ibm.com) wrote:
> > Insertion of kprobe module causes oops on 2.6.25-rc3-mm1 (x86) systems if
> > we enable CONFIG_DEBUG_RODATA.  While registering the probe, memcpy()
> > fails to copy breakpoint instruction to the instruction address and
> > generates the oops.
> >
> > BUG: unable to handle kernel paging request at ffffffff8047d1a7
> > IP: [<ffffffff8047ff62>] text_poke+0xa/0x10
> > PGD 203067 PUD 207063 PMD 7e191163 PTE 47d161
> > Oops: 0003 [1] PREEMPT SMP
> > ................................................
> >
> > This is because, Mathieu's patch (http://lkml.org/lkml/2008/2/2/226)
> > makes entire text segment as READONLY, if we enable CONFIG_DEBUG_RODATA.
> >
> > So reverting this patch or new patch, which considers kprobes while
> > deciding boundary for setting readonly pages might solve the probelm.
>
> Hi,
>
> Are you running under Xen, lguest, kvm.. ?

No, on a bare x86_64 system.   That shouldn't matter for this bug, A quick 
look at the call path shows me this.
     start_kernel()->rest_init()->kernel_init()->init_post()->mark_rodata_ro()

Srinivasa DS
>
> Mathieu
>
> > Thanks
> >  Srinivasa DS
> >  LTC-IBM


--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/