mbox series

[0/3] kallsyms: don't leak address

Message ID 1513554812-13014-1-git-send-email-me@tobin.cc
Headers show
Series kallsyms: don't leak address | expand

Message

Tobin C. Harding Dec. 17, 2017, 11:53 p.m. UTC 
This set plugs a kernel address leak that occurs if kallsyms symbol
look up fails. This set was prompted by a leaking address found using
scripts/leaking_addresses.pl on a PowerPC machine in the wild.

Patch set does not change behaviour when KALLSYMS is not defined
(suggested by Linus).

RFC has been in flight for 3 weeks with no negative response.

Patch 1 - return error code if symbol look up fails.
Patch 2 - print <no-symbol> to buffer if symbol look up returns an error.
Patch 3 - maintain current behaviour in ftrace.

Patch 3 (the ftrace stuff) is untested.

thanks,
Tobin.

Tobin C. Harding (3):
  kallsyms: don't leak address when symbol not found
  vsprintf: print <no-symbol> if symbol not found
  trace: print address if symbol not found

 include/linux/kernel.h           |  2 ++
 kernel/kallsyms.c                |  6 ++++--
 kernel/trace/trace.h             | 24 ++++++++++++++++++++++++
 kernel/trace/trace_events_hist.c |  6 +++---
 lib/vsprintf.c                   | 18 +++++++++++++++---
 5 files changed, 48 insertions(+), 8 deletions(-)

Comments

Michael Ellerman Dec. 18, 2017, 5:31 a.m. UTC | #1 
"Tobin C. Harding" <me@tobin.cc> writes:

> This set plugs a kernel address leak that occurs if kallsyms symbol
> look up fails. This set was prompted by a leaking address found using
> scripts/leaking_addresses.pl on a PowerPC machine in the wild.

Any details on that? I haven't heard about it.

cheers
Tobin C. Harding Dec. 18, 2017, 6 a.m. UTC | #2 
On Mon, Dec 18, 2017 at 04:31:25PM +1100, Michael Ellerman wrote:
> "Tobin C. Harding" <me@tobin.cc> writes:
> 
> > This set plugs a kernel address leak that occurs if kallsyms symbol
> > look up fails. This set was prompted by a leaking address found using
> > scripts/leaking_addresses.pl on a PowerPC machine in the wild.
> 
> Any details on that? I haven't heard about it.

I have an account on a server in USA. I'm not sure how much detail I
should give on the mailing list in case it is sensitive information. I
will email you off list. Responding here for the benefit of the list.

If this is not the correct way to handle this please say so.

thanks,
Tobin.
Tobin C. Harding Dec. 18, 2017, 9:17 a.m. UTC | #3 
On Mon, Dec 18, 2017 at 05:00:48PM +1100, Tobin C. Harding wrote:
> On Mon, Dec 18, 2017 at 04:31:25PM +1100, Michael Ellerman wrote:
> > "Tobin C. Harding" <me@tobin.cc> writes:
> > 
> > > This set plugs a kernel address leak that occurs if kallsyms symbol
> > > look up fails. This set was prompted by a leaking address found using
> > > scripts/leaking_addresses.pl on a PowerPC machine in the wild.
> > 
> > Any details on that? I haven't heard about it.
> 
> I have an account on a server in USA. I'm not sure how much detail I
> should give on the mailing list in case it is sensitive information. I
> will email you off list. Responding here for the benefit of the list.

Geez I'm a goose. Here is the details with a sanitized address.

/proc/8025/task/8025/stack: [<0000000000000000>] 0xc0000001XXXXXXXX

$ uname -r
4.4.0-79-powerpc64-smp

thanks,
Tobin.