[Bug dynamic-link/21942] _dl_dst_substitute incorrectly handles $ORIGIN: with AT_SECURE=1

["carlos at redhat dot com" <sourceware-bugzilla at sourceware dot org>]

https://sourceware.org/bugzilla/show_bug.cgi?id=21942

Carlos O'Donell <carlos at redhat dot com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |carlos at redhat dot com

--- Comment #1 from Carlos O'Donell <carlos at redhat dot com> ---
(In reply to Florian Weimer from comment #0)
> I'm not sure if there is a trivial fix for this.  The execution flow is very
> complicated.

I can't reproduce this issue with my testcases for this, and the 'is_path' code
was removed by recent refactoring to simplify this code.

For example:

Test 5  [SUID]: Test that RPATH of $ORIGIN:./ ignores $ORIGIN and uses ./.
     25758:     expand_dynamic_string_token name=$ORIGIN, start=$ORIGIN
     25758:     _dl_dst_count $ORIGIN (7fbbbf87ad20)
     25758:     is_dst name=ORIGIN (begin)
     25758:     is_dst nlen=6 name=ORIGIN (7fbbbf87ad21) (str=ORIGIN)
     25758:     _dl_dst_count next
     25758:     _dl_dst_count cnt 1
     25758:     DL_DST_REQUIRED l_origin 7fbbbf87ad50
     25758:     _dl_dst_substitute name=$ORIGIN start=$ORIGIN
     25758:     is_dst name=ORIGIN (begin)
     25758:     is_dst nlen=6 name=ORIGIN (7fbbbf87ad21) (str=ORIGIN)
     25758:     _dl_dst_substitute 7fbbbf87ad50
     25758:     is_trusted_path_normalize: /root 5
     25758:     is_trusted_path_normalize: Transformed path /root/
     25758:     is_trusted_path_normalize: Path is NOT trusted.
     25758:     expand_dynamic_string_token name=./, start=$ORIGIN
     25758:     security = 1
     25758:     security = 1
     25758:     security = 1
     25758:     security = 1
PASS: Correct primary library.

You see that the RPATH splitting happens at expand_dynamic_string_token after
my fixes, and thus is_trusted_path_normalize is never passed multiple path
sequences.

I'm going to mark that my patches fix this also, and you can review upstream.

-- 
You are receiving this mail because:
You are on the CC list for the bug.