Re: [PATCH v2 2.39 0/7] Cherry-pick CVE-2026-5435 and CVE-2026-6238 fixes to release/2.39/master
Jaipaul Cheernam <[email protected]>
| Newsgroups | gmane.comp.lib.glibc.alpha |
|---|---|
| Message-ID | <DU7PPF66507B2D7153D9C1E595566A322EAF3F12@DU7PPF66507B2D7.EURP189.PROD.OUTLOOK.COM> |
Hi Florian, Andreas, My v2 series ([PATCH v2 2.39 XXX]) was marked "Not Applicable” in patchwork. Could you clarify — is this being handled internally, or is there something I need to change in the submission? Happy to adjust if needed. Regards, Jaipaul Cheernam 🔗 EST Website<https://www.est.tech/> 🔗 EST LinkedIn<https://www.linkedin.com/company/ericsson-software-technology/> From: Jaipaul Cheernam <[email protected]> Date: Thursday, 2 July 2026 at 22:40 To: [email protected] <[email protected]> Cc: [email protected] <[email protected]>; Jaipaul Cheernam <[email protected]> Subject: [PATCH v2 2.39 0/7] Cherry-pick CVE-2026-5435 and CVE-2026-6238 fixes to release/2.39/master Hi, v2: Backport the full series including the 3 preparatory commits (type/class formatting fixes) needed for the test case to pass. All 7 commits cherry-pick cleanly with no conflicts. v1 only included 4 patches and had a trivial conflict. Florian pointed out the test would fail without the type formatting changes. This series cherry-picks the fixes for CVE-2026-5435 (bug 34033) and CVE-2026-6238 (bug 34069) to the release/2.39/master branch. Patches 1-3 are preparatory fixes for ns_sprintrrf type/class formatting (bug 34289), required for the test in patch 7 to pass. Patches 4-6 are the actual security fixes. Patch 7 is the regression test. The same fixes were already cherry-picked to release/2.43/master as commits e64ae5a591f0, 3a418da6a393, and 8759917de589. All patches apply cleanly on top of current release/2.39/master HEAD (be1e627cd7). glibc 2.39 is actively used in Yocto Project's scarthgap (LTS) release, so this branch continues to have downstream consumers relying on it for security fixes. Testing: All 56 resolv tests pass, including the new tst-ns_sprintrr. No regressions. Thanks, Jaipaul Florian Weimer (7): resolv: Declare __p_class_syms, __p_type_syms for internal use resolv: Fix ns_sprintrrf formatting of class, type values (bug 34289) resolv: Improve formatting of unknown records in ns_sprintrrf resolv: Check for inet_ntop failure in ns_sprintrrf resolv: More types as unknown in ns_sprintrrf (CVE-2026-5435) resolv: Fix buffer overreads in ns_sprintrrf (CVE-2026-6238) resolv: Add test case tst-ns_sprintrr (bug 34033, bug 34069) include/resolv.h | 5 + resolv/Makefile | 2 + resolv/ns_print.c | 170 +++++++------------- resolv/res_debug.c | 4 - resolv/tst-ns_sprintrr.c | 329 +++++++++++++++++++++++++++++++++++++++ 5 files changed, 389 insertions(+), 121 deletions(-) create mode 100644 resolv/tst-ns_sprintrr.c -- 2.34.1