linux-kernel - [RFC PATCH 0/1] tracepoints: Use %px instead of %p in trace event definitions

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-next>] [day] [month] [year] [list]

Date:   Wed, 14 Oct 2020 17:59:10 +0900
From:   Masami Hiramatsu <mhiramat@...nel.org>
To:     LKML <linux-kernel@...r.kernel.org>
Cc:     Steven Rostedt <rostedt@...dmis.org>,
        Masami Hiramatsu <mhiramat@...nel.org>,
        Ingo Molnar <mingo@...nel.org>
Subject: [RFC PATCH 0/1] tracepoints: Use %px instead of %p in trace event definitions

Hi,

Recently I faced this "hashed pointer value" issue while tracing
percpu events.

https://lore.kernel.org/lkml/20201008172806.1591ebb538946c5ee93d372a@kernel.org/

In the mail, I explained why the base_addr and ptr was not actual value as below;

>           <idle>-0     [000] ....     0.543208: percpu_create_chunk: base_addr=000000005d5ad71c
>  [...]
>          systemd-1     [000] ....     0.568931: percpu_alloc_percpu: reserved=0 is_atomic=0 size=48 align=8 base_addr=00000000fa92a086 off=32672 ptr=000000008da0b73d
>          systemd-1     [000] ....     0.568938: xen_guest_init: Xen: alloc xen_vcpu_info ffff800011003fa0 id=000000008da0b73d
>          systemd-1     [000] ....     0.586635: xen_starting_cpu: Xen: xen_vcpu_info ffff800011003fa0, vcpup ffff00092f4ebfa0 per_cpu_offset[0] ffff80091e4e8000
> 
> (NOTE: base_addr and ptr are encoded to the ids, not actual address
>  because of "%p" printk format)

Since trace_seq_printf() use in-kernel vsprintf() at last, the %p is always
converted to the hash value instead of real address.
For the user log it maybe secure, but for the ftrace, I think it is
meaningless because

- tracefs is used by root user, so no need to hide it.
- tracefs user can access percpu/cpu*/trace_pipe_raw which stores real
  address on the trace buffer.
- external commands like perf doesn't convert to the hash value.

And when debugging kernel, we would like to know real address which tells
us the address is in which area in the kernel by comparing with kernel
memory mapping.

So I would like to replace %p with %px to print out the raw address
value on tracefs/trace file. There is %pK which can control output format,
but from above reasons, I think %px is enough.


BTW, since this is an independent reason (of printk() security conversion)
apart from each subsystems, I made it a single commit. If I should better
break it into 44 individual patches, please tell me.

Thank you,

---

Masami Hiramatsu (1):
      tracepoints: tree-wide: Replace %p with %px


 include/trace/events/afs.h           |    6 ++---
 include/trace/events/alarmtimer.h    |    2 +-
 include/trace/events/btrfs.h         |   14 ++++++-----
 include/trace/events/cachefiles.h    |   20 ++++++++--------
 include/trace/events/cma.h           |    4 ++-
 include/trace/events/filelock.h      |   10 ++++----
 include/trace/events/filemap.h       |    4 ++-
 include/trace/events/fs_dax.h        |    2 +-
 include/trace/events/fscache.h       |   28 +++++++++++------------
 include/trace/events/host1x.h        |    4 ++-
 include/trace/events/huge_memory.h   |    6 ++---
 include/trace/events/ib_mad.h        |    4 ++-
 include/trace/events/io_uring.h      |   28 +++++++++++------------
 include/trace/events/kmem.h          |   18 +++++++--------
 include/trace/events/lock.h          |    4 ++-
 include/trace/events/mlxsw.h         |   10 ++++----
 include/trace/events/mmc.h           |    4 ++-
 include/trace/events/napi.h          |    2 +-
 include/trace/events/nbd.h           |    4 ++-
 include/trace/events/net.h           |    8 +++---
 include/trace/events/nilfs2.h        |   14 ++++++-----
 include/trace/events/objagg.h        |   20 ++++++++--------
 include/trace/events/page_pool.h     |    8 +++---
 include/trace/events/page_ref.h      |    4 ++-
 include/trace/events/pagemap.h       |    4 ++-
 include/trace/events/percpu.h        |    8 +++---
 include/trace/events/pwc.h           |    4 ++-
 include/trace/events/pwm.h           |    2 +-
 include/trace/events/qdisc.h         |    2 +-
 include/trace/events/rcu.h           |   12 +++++-----
 include/trace/events/rpcgss.h        |    4 ++-
 include/trace/events/rpcrdma.h       |   42 +++++++++++++++++-----------------
 include/trace/events/rtc.h           |    2 +-
 include/trace/events/rxrpc.h         |    6 ++---
 include/trace/events/skb.h           |    6 ++---
 include/trace/events/spi.h           |    6 ++---
 include/trace/events/sunrpc.h        |   12 +++++-----
 include/trace/events/tegra_apb_dma.h |    2 +-
 include/trace/events/timer.h         |   14 ++++++-----
 include/trace/events/vb2.h           |    2 +-
 include/trace/events/vmscan.h        |    6 ++---
 include/trace/events/workqueue.h     |    8 +++---
 include/trace/events/xdp.h           |    6 ++---
 include/trace/events/xen.h           |   34 ++++++++++++++--------------
 44 files changed, 205 insertions(+), 205 deletions(-)

--
Masami Hiramatsu (Linaro) <mhiramat@...nel.org>