Commander1024/ESP32_Serial_Swiss_Army_Knife
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Harden wolfSSL and wolfSSH validation

Browse Source 
Enable validated ECC imports and X25519 all-zero rejection through
PUBLIC build policy. Tighten wolfSSH parser bounds, overflow handling,
and signature framing with guard-page and crypto vector contracts.
This commit is contained in:
Commander1024
2026-09-15 23:54:39 +02:00
parent c010e1a1d5
commit 4d3bb490c9
17 changed files with 1495 additions and 99 deletions
Download Patch File Download Diff File Expand all files Collapse all files
docs/agent/code-map.md
+5 -2
View File
@@ -20,8 +20,11 @@ This is a semantic map, not a complete file inventory. Start here, then read the
## Source-pinned dependency corrections (Phases 9C–9D)
- Files: root `CMakeLists.txt` (after `project()`), `cmake/security_overrides.cmake`, `tools/security_overrides.py`; tests: `tests/sdk_security_overrides/run.py --build-dir .pio/build/esp32-s3-devkitc-1-n16r8`.
- Build input is the exact-hash original **plus checked-in edits**, not installed source alone. Generated copies replace seven target sources without modifying SDK/managed components: HTTPS cleanup/private-key release, HTTPD scratch lifetime/null first read, ESP-TLS server-only protocol list, wolfSSH password bounds/payload wiping, DHCP option bounds (CVE-2026-45160), TLS 1.2 EMS error propagation (CVE-2026-50581), and X.509 OID allocation failure (CVE-2026-34874). The mbedTLS entries explicitly select validated nested `mbedtls`/`mbedx509` targets, not the component wrapper. Original notices and compile properties retained; all seven copies carry the 2026-09-15 modification notice; source/hash/target ambiguity fails configuration. Never hand-edit generated copies or silently repin.
- Xtensa crypto policy: root `CMakeLists.txt` sets `CURVE25519_SMALL`/`ED25519_SMALL` before component parsing; `cmake/wolf_crypto_policy.cmake` PUBLIC-propagates `cmake/wolf_crypto_policy.h` to wolfSSL consumers. The resolved-settings guard requires both small implementations and rejects X25519 blinding and unreviewed Curve448/Ed448 enablement. Tests: `tests/wolf_crypto_policy/run.py`. [Wolf review](../wolf_security_review.md) distinguishes the implemented mitigation from pending ordering/parser/ECC review; [IDF review](../idf_security_review.md) records the three backports and remaining findings.
- Build input is the exact-hash original **plus checked-in edits**, not installed source alone. Generated copies replace seven target sources without modifying SDK/managed components: HTTPS cleanup/private-key release, HTTPD scratch lifetime/null first read, ESP-TLS server-only protocol list, wolfSSH password bounds/payload wiping plus bounded IGNORE/service/string parsing, channel-window overflow rejection and ECC/Ed25519 label/exact-signature framing, DHCP option bounds (CVE-2026-45160), TLS 1.2 EMS error propagation (CVE-2026-50581), and X.509 OID allocation failure (CVE-2026-34874). The mbedTLS entries explicitly select validated nested `mbedtls`/`mbedx509` targets, not the component wrapper. Original notices and compile properties retained; all seven copies carry the 2026-09-15 modification notice; source/hash/target ambiguity fails configuration. Never hand-edit generated copies or silently repin.
- Xtensa crypto policy: root `CMakeLists.txt` sets `CURVE25519_SMALL`/`ED25519_SMALL` before component parsing; `cmake/wolf_crypto_policy.cmake` PUBLIC-propagates `cmake/wolf_crypto_policy.h` to wolfSSL consumers. PUBLIC `WOLFSSL_VALIDATE_ECC_IMPORT` and `WOLFSSL_ECDHX_SHARED_NOT_ZERO` enable existing P-256 import and X25519 all-zero-result checks. The resolved-settings guard requires these checks and both small implementations, rejects reviewed ECC validator-disabling/hardware-stub configurations, X25519 blinding and unreviewed Curve448/Ed448 enablement. Tests: `tests/wolf_crypto_policy/run.py` (strict actual production flags, real vendor crypto/ASN vectors and independently specified exact source deltas; candidate injection is not production evidence). [Key-validation review](../ssh_key_validation_review.md) records effective flags, caller/API limits and unmeasured validation cost. [Wolf review](../wolf_security_review.md) distinguishes implemented mitigations from unresolved ordering/deferred parsers; [IDF review](../idf_security_review.md) records the three backports and remaining findings.
- Parser tests: `tests/wolfssh_parser_contract/run.py`, [scope and exclusions](../../tests/wolfssh_parser_contract/README.md): 3,124 extracted-function cases per each of two stack modes, guard pages/UBSan traps and six rejected guard-removal mutations. Crypto doubles test gating, not arithmetic. PR892/881/880 subsets plus local signature framing only; no PR899 or ordering changes. Password/dispatch/deferred-source fences preserve prior contracts.
- Latest supplied parent build PASS: 94,340 B linked RAM / 1,768,949 B flash (+1,732 B versus 1,767,217 B). Strict crypto, parser, auth (135 cases), protocol policy and SDK override `--build-dir .pio/build/esp32-s3-devkitc-1-n16r8` commands PASS; independent review found no scoped blocker. Not target/runtime-reserve evidence.
- Ordering remains open: temporary PR793/819/840/855/921 attempt retained no changes; `SendNewKeys` WANT_WRITE/`SendExtInfo` continuation, `extInfoSent` rekey semantics and manual prerequisites unresolved. Registry 1.5.0/5.9.2 queries returned 404 on 2026-09-15 despite upstream tags. [Next strategy/commit pins](../wolf_security_review.md#ordering-blocker-and-actionable-next-strategy): isolated packaging/compatibility evaluation or prerequisite-audited source/header backport, with nonblocking/initial-KEX/rekey ordering tests before closure. No upgrade/device operation.
- Policy/evidence/limits: [library review](../security_library_review.md), [Phase 9C](../security_hardening.md#9c-library-cleanup-and-protocol-policy). Source-contract tests must locate and verify actual generated compilation inputs, not assume original vendor paths.
## Secure randomness