[Bug math/25487] sinl() stack corruption from crafted input (CVE-2020-10029)

[Bug math/25487] sinl() stack corruption from crafted input (CVE-2020-10029)

[cvs-commit at gcc dot gnu.org]

https://sourceware.org/bugzilla/show_bug.cgi?id=25487

--- Comment #6 from cvs-commit at gcc dot gnu.org <cvs-commit at gcc dot gnu.org> ---
The release/2.29/master branch has been updated by Patricia Franklin
<patsy@sourceware.org>:

https://sourceware.org/git/gitweb.cgi?p=glibc.git;h=0474cd5de60448f31d7b872805257092faa626e4

commit 0474cd5de60448f31d7b872805257092faa626e4
Author: Joseph Myers <joseph@codesourcery.com>
Date:   Wed Feb 12 23:31:56 2020 +0000

    Avoid ldbl-96 stack corruption from range reduction of pseudo-zero (bug
25487).

    Bug 25487 reports stack corruption in ldbl-96 sinl on a pseudo-zero
    argument (an representation where all the significand bits, including
    the explicit high bit, are zero, but the exponent is not zero, which
    is not a valid representation for the long double type).

    Although this is not a valid long double representation, existing
    practice in this area (see bug 4586, originally marked invalid but
    subsequently fixed) is that we still seek to avoid invalid memory
    accesses as a result, in case of programs that treat arbitrary binary
    data as long double representations, although the invalid
    representations of the ldbl-96 format do not need to be consistently
    handled the same as any particular valid representation.

    This patch makes the range reduction detect pseudo-zero and unnormal
    representations that would otherwise go to __kernel_rem_pio2, and
    returns a NaN for them instead of continuing with the range reduction
    process.  (Pseudo-zero and unnormal representations whose unbiased
    exponent is less than -1 have already been safely returned from the
    function before this point without going through the rest of range
    reduction.)  Pseudo-zero representations would previously result in
    the value passed to __kernel_rem_pio2 being all-zero, which is
    definitely unsafe; unnormal representations would previously result in
    a value passed whose high bit is zero, which might well be unsafe
    since that is not a form of input expected by __kernel_rem_pio2.

    Tested for x86_64.

    (cherry picked from commit 9333498794cde1d5cca518badf79533a24114b6f)

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

[Bug math/25487] sinl() stack corruption from crafted input (CVE-2020-10029)

[cvs-commit at gcc dot gnu.org]

https://sourceware.org/bugzilla/show_bug.cgi?id=25487

--- Comment #7 from cvs-commit at gcc dot gnu.org <cvs-commit at gcc dot gnu.org> ---
The release/2.29/master branch has been updated by Patricia Franklin
<patsy@sourceware.org>:

https://sourceware.org/git/gitweb.cgi?p=glibc.git;h=8e5d591b101d7d8a4628522f1e5ec24b6dfa731b

commit 8e5d591b101d7d8a4628522f1e5ec24b6dfa731b
Author: Florian Weimer <fweimer@redhat.com>
Date:   Thu Feb 13 17:01:15 2020 +0100

    math/test-sinl-pseudo: Use stack protector only if available

    This fixes commit 9333498794cde1d5cca518bad ("Avoid ldbl-96 stack
    corruption from range reduction of pseudo-zero (bug 25487).").

    (cherry picked from commit c10acd40262486dac597001aecc20ad9d3bd0e4a)

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

[Bug math/25487] sinl() stack corruption from crafted input (CVE-2020-10029)

[huzaifas at redhat dot com]

https://sourceware.org/bugzilla/show_bug.cgi?id=25487

Huzaifa Sidhpurwala <huzaifas at redhat dot com> changed:

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

--- Comment #8 from Huzaifa Sidhpurwala <huzaifas at redhat dot com> ---
Looking at the way crash is caused via pseudo-zero numbers and after running
the poc through gdb, it seems like on systems in which glibc is not compiled
with -fstack-protector-all, (which means the vuln function is not protected by
stack-canaries), all that can be achieved is overwrite the stack and the return
address with 0's.

This can only cause a crash and jumping any other location seems very difficult
to achieve if not impossible.

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

[Bug math/25487] sinl() stack corruption from crafted input (CVE-2020-10029)

[dominik.b.czarnota+bugzilla at gmail dot com]

https://sourceware.org/bugzilla/show_bug.cgi?id=25487

Disconnect3d <dominik.b.czarnota+bugzilla at gmail dot com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |dominik.b.czarnota+bugzilla
                   |                            |@gmail.com

--- Comment #9 from Disconnect3d <dominik.b.czarnota+bugzilla at gmail dot com> ---
(In reply to Guido Vranken from comment #0)
> This has been tested on x64 Linux with both the Ubuntu glibc and the
> latest git glibc.

Btw since Ubuntu 18.04 and 20.04 both seems patched as of today, if anyone
wants to play with this, it can be reproduced on e.g. gcc:9.3 docker image
(https://hub.docker.com/layers/gcc/library/gcc/9.3/images/sha256-dd7c100e12ddbf4178f5cd524a869fa54f453d35bf1b5f287ec6b70e3230c2e4?context=explore),
or by using the following docker image
(https://hub.docker.com/layers/disconnect3d/repro-cve-2020-10029/latest/images/sha256-0d7cf62eee140c9a0039945f8fe2ff3c53b7670b663cb67feef57878ab92ee06?context=explore)
and command where I compiled the example in the cve-2020-10029 directory:

docker run --rm -it --cap-drop=ALL --net=none disconnect3d/repro-cve-2020-10029
/cve-2020-10029/a.out

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

[Bug math/25487] sinl() stack corruption from crafted input (CVE-2020-10029)

[cvs-commit at gcc dot gnu.org]

https://sourceware.org/bugzilla/show_bug.cgi?id=25487

--- Comment #10 from cvs-commit at gcc dot gnu.org <cvs-commit at gcc dot gnu.org> ---
The release/2.27/master branch has been updated by Dmitry Levin
<ldv@sourceware.org>:

https://sourceware.org/git/gitweb.cgi?p=glibc.git;h=59420258afaf73dc8fab63ce186bac792613fe08

commit 59420258afaf73dc8fab63ce186bac792613fe08
Author: Joseph Myers <joseph@codesourcery.com>
Date:   Wed Feb 12 23:31:56 2020 +0000

    Avoid ldbl-96 stack corruption from range reduction of pseudo-zero (bug
25487).

    Bug 25487 reports stack corruption in ldbl-96 sinl on a pseudo-zero
    argument (an representation where all the significand bits, including
    the explicit high bit, are zero, but the exponent is not zero, which
    is not a valid representation for the long double type).

    Although this is not a valid long double representation, existing
    practice in this area (see bug 4586, originally marked invalid but
    subsequently fixed) is that we still seek to avoid invalid memory
    accesses as a result, in case of programs that treat arbitrary binary
    data as long double representations, although the invalid
    representations of the ldbl-96 format do not need to be consistently
    handled the same as any particular valid representation.

    This patch makes the range reduction detect pseudo-zero and unnormal
    representations that would otherwise go to __kernel_rem_pio2, and
    returns a NaN for them instead of continuing with the range reduction
    process.  (Pseudo-zero and unnormal representations whose unbiased
    exponent is less than -1 have already been safely returned from the
    function before this point without going through the rest of range
    reduction.)  Pseudo-zero representations would previously result in
    the value passed to __kernel_rem_pio2 being all-zero, which is
    definitely unsafe; unnormal representations would previously result in
    a value passed whose high bit is zero, which might well be unsafe
    since that is not a form of input expected by __kernel_rem_pio2.

    Tested for x86_64.

    (cherry picked from commit 9333498794cde1d5cca518badf79533a24114b6f)

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