kernel/bpf/verifier.c in the Linux kernel through 4.14.8 mishandles states_equal comparisons between the pointer data type and the UNKNOWN_VALUE data type, which allows local users to obtain potentially sensitive address information, aka a "pointer leak." Upstream commit: https://git.kernel.org/pub/scm/linux/kernel/git/stable/stable-queue.git/commit/?id=f3027591be9b80b66f2a2378cb67af106e7bfdee References: https://www.spinics.net/lists/stable/msg207010.html https://www.openwall.com/lists/oss-security/2017/12/24/1
Created kernel tracking bugs for this issue: Affects: fedora-all [bug 1529125]
Statement: This issue does not affect the versions of the kernel package as shipped with Red Hat Enterprise Linux 5, 6, 7 and Red Hat Enterprise MRG 2.
(In reply to Pedro Sampaio from comment #0) > kernel/bpf/verifier.c in the Linux kernel through 4.14.8 mishandles > states_equal comparisons between the pointer data type and the UNKNOWN_VALUE > data type, which allows local users to obtain potentially sensitive address > information, aka a "pointer leak." > > References: > > https://anonscm.debian.org/cgit/kernel/linux.git/tree/debian/patches/bugfix/ > all/bpf-verifier-fix-states_equal-comparison-of-pointer-and-unknown. > patch?h=stretch-security > https://anonscm.debian.org/cgit/kernel/linux.git/commit/?h=stretch- > security&id=ad775f6ff7eebb93eedc2f592bc974260e7757b0 those links no longer works, also can't see that commit upstream, could you please provide new links? thanks, jirka
Sorry I missed this needinfo. E-mail filters broke after bugzilla update. The references were updated. I believe this was only committed to 4.9 stable branch upstream, as the references point out. Please let me know if anything else is needed.
(In reply to Pedro Sampaio from comment #6) > Sorry I missed this needinfo. E-mail filters broke after bugzilla update. > > The references were updated. > > I believe this was only committed to 4.9 stable branch upstream, as the > references point out. > > Please let me know if anything else is needed. so following 3 are already in rhel8: bpf-adjust-insn_aux_data-when-patching-insns.patch bpf-fix-branch-pruning-logic.patch bpf-fix-incorrect-sign-extension-in-check_alu_op.patch but I can't upstream version of this one: bpf-reject-out-of-bounds-stack-pointer-calculation.patch any idea? thanks, jirka
This one you looking for? https://lkml.org/lkml/2017/12/25/95 https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/tag/?h=v4.9.72
(In reply to Pedro Sampaio from comment #8) > This one you looking for? > > https://lkml.org/lkml/2017/12/25/95 > https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/tag/?h=v4.9. > 72 no, the upstream version (if there is any) of this commit: bpf-reject-out-of-bounds-stack-pointer-calculation.patch this is 4.9 stable fix, which is usually backport of the upstream commit.. but in this case it looks like this was 4.9 specific fix, because I can't find the upstream (not stable) change jirka
I think the fixing commit for the upstream (mainline) was https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?id=179d1c5602997fef5a940c6ddcf31212cbfebd14.
(In reply to Nicholas Luedtke from comment #10) > I think the fixing commit for the upstream (mainline) was > https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/ > ?id=179d1c5602997fef5a940c6ddcf31212cbfebd14. this one got merged in v4.15, so we have it in RHEL8 together with the rest, so I'll close the RHEL8 BZ as CURRENTRELEASE jirka
This bug is now closed. Further updates for individual products will be reflected on the CVE page(s): https://access.redhat.com/security/cve/cve-2017-17864