Re: [PATCH] ld.so.8: Document LD_AUDIT and LD_POINTER_GUARD

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



Hi Petr,

Thanks for this work.

One general comment: since there are two logically independent pieces
here, two patches would have been best.  Thgerefore, I'll respond to
this mail in two separate threads.

Cheers,

Michael

On Sun, Dec 7, 2008 at 1:28 PM, Petr Baudis <pasky@xxxxxxx> wrote:
> I was using existing comments, mailing list posts, and mainly the glibc
> source code as a reference. I'm not sure if LD_AUDIT is 100% compatible
> with Solaris but aside of the header names, it appears to be, based on
> quick comparison.
>
> Signed-off-by: Petr Baudis <pasky@xxxxxxx>
>
> diff --git a/man8/ld.so.8 b/man8/ld.so.8
> index 6ca9a8c..69827f9 100644
> --- a/man8/ld.so.8
> +++ b/man8/ld.so.8
> @@ -1,5 +1,9 @@
>  .\" This is in the public domain
> -.TH LD.SO 8 2008-10-27 "GNU" "Linux Programmer's Manual"
> +.\"
> +.\" 2008-12-07 Petr Baudis <pasky@xxxxxxx>
> +.\"    Document LD_AUDIT and LD_POINTER_GUARD
> +.\"
> +.TH LD.SO 8 2008-12-07 "GNU" "Linux Programmer's Manual"
>  .SH NAME
>  ld.so, ld-linux.so* \- dynamic linker/loader
>  .SH SYNOPSIS
> @@ -177,13 +181,25 @@ Version of
>  for a.out binaries only.
>  Old versions of ld\-linux.so.1 also supported
>  .BR LD_ELF_PRELOAD .
> -.\" FIXME
> -.\" Document LD_AUDIT ("Install audit libraries for glibc")
> -.\" new in glibc 2.4
> -.\" ignored in set-user-ID and set-group-ID programs
> -.\"
> -.\" For some info, see Solaris Linker and Libraries Guide,
> -.\" "Runtime Linker Auditing Interface"
> +.TP
> +.B LD_AUDIT
> +(glibc since 2.4)
> +A colon-separated list of additional, user-specified, ELF shared libraries
> +to be loaded before all others in a separate linker namespace.
> +.B LD_AUDIT
> +is ignored for set-user-ID/set-group-ID binaries.
> +
> +The dynamic linker will notify the audit
> +libraries at so-called auditing checkpoints, like loading a new library,
> +resolving a symbol or calling a symbol from another shared object, by
> +calling an appropriate function within the audit library (the functions
> +start all with an
> +.B la_
> +prefix and are prototyped in
> +.BR "<link.h> " and " <bits/link.h>").
> +The auditing interface is compatible with Solaris as described in its
> +.IR "Linker and Libraries Guide" ,
> +chapter Runtime Linker Auditing Interface.
>  .TP
>  .B LD_BIND_NOT
>  (glibc since 2.1.95)
> @@ -240,18 +256,18 @@ For security reasons, since glibc 2.4,
>  is ignored for set-user-ID/set-group-ID binaries.
>  .\" Only used if $ORIGIN can't be determined by normal means
>  .\" (from the origin path saved at load time, or from /proc/self/exe)?
> -.\"
> -.\" FIXME
> -.\" Document LD_POINTER_GUARD
> -.\" Since glibc 2.4
> -.\" Set to 0 to disable pointer guarding
> -.\" Any other value enables pointer guarding, which is also the default.
> -.\" Pointer guarding is a security mechanism(?) to minimize the
> -.\" the risk of having usable pointer in the event of a buffer overrun
> -.\" or stack smashing attack(?).
> -.\" http://article.gmane.org/gmane.comp.gdb.patches/41147/match=ld_pointer_guard
> -.\" http://sourceware.org/ml/gdb-patches/2008-04/msg00252.html
> -.\" http://www.cygwin.com/ml/libc-alpha/2006-01/msg00011.html
> +.TP
> +.B LD_POINTER_GUARD
> +(glibc since 2.4)
> +Set to 0 to disable pointer guarding.
> +Any other value enables pointer guarding, which is also the default.
> +Pointer guarding is a security mechanism where some pointers to code
> +stored in writable program memory (return address saved by
> +.BR setjmp (3)
> +or function pointers used by various glibc internals) are mangled
> +semi-randomly to make it more difficult for an attacker to hijack
> +these to their own code in the event of a buffer overrun or stack
> +smashing attack.
>  .TP
>  .B LD_PROFILE
>  (glibc since 2.1)
>



-- 
Michael Kerrisk
Linux man-pages maintainer; http://www.kernel.org/doc/man-pages/
git://git.kernel.org/pub/scm/docs/man-pages/man-pages.git
man-pages online: http://www.kernel.org/doc/man-pages/online_pages.html
Found a bug? http://www.kernel.org/doc/man-pages/reporting_bugs.html
--
To unsubscribe from this list: send the line "unsubscribe linux-man" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html

[Index of Archives]     [Kernel Documentation]     [Netdev]     [Linux Ethernet Bridging]     [Linux Wireless]     [Kernel Newbies]     [Security]     [Linux for Hams]     [Netfilter]     [Bugtraq]     [Yosemite News]     [MIPS Linux]     [ARM Linux]     [Linux RAID]     [Linux Admin]     [Samba]

  Powered by Linux