Bound SHE LoadKey auth key read to WH_SHE_KEY_SZ

[yosuke-wolfssl]

Problem

_LoadKey in the SHE extension read the key-update authorization key with the capacity set to the whole kdf input buffer (keySz = sizeof(kdfInput), 32) rather than the SHE key size (16). wh_Server_KeystoreReadKey reports the stored object's actual length, and the kdf then appends a 16-byte constant at that offset, so a slot holding more than WH_SHE_KEY_SZ bytes overruns the 32-byte stack buffer by up to 16 bytes on two memcpy calls and the subsequent _AesMp16 read.

The SHE and NVM id spaces overlap, so an authenticated client can plant an oversized object directly in a SHE slot with WH_MESSAGE_NVM_ACTION_ADDOBJECT and then reach the overflow through a normal CMD_LOAD_KEY. Requires WOLFHSM_CFG_SHE_EXTENSION.

The target-key read had a second, independent flaw: wh_Server_KeystoreReadKey returns WH_ERROR_NOSPACE before populating the caller's metadata, so an oversized target slot left meta zeroed. wh_She_Label2Meta then read all-zero flags, the WH_SHE_FLAG_WRITE_PROTECT bit appeared clear, and a write-protected slot was silently overwritten.

Fix (src/wh_server_she.c)

Both reads are pinned to WH_SHE_KEY_SZ and a slot of any other length is rejected, matching every other SHE handler.

Read	Before	After
Auth key	capacity of the full kdf buffer	WH_SHE_KEY_SZ; WH_ERROR_NOSPACE or size mismatch → WH_SHE_ERC_KEY_INVALID
Target key	inherited capacity; meta unset on NOSPACE	WH_SHE_KEY_SZ; rejected before the label is consulted, closing the write-protect bypass

wh_Client_ShePreProgramKey also rejects a keySz other than WH_SHE_KEY_SZ with WH_ERROR_BADARGS, so an ordinary client cannot create such a slot in the first place; the header documents the new precondition.

Closes 7155.

Tests (test-refactor/client-server/wh_test_she.c)

Three cases appended to whTest_She, driven only through wh_Client_*:

Case	Asserts
Pre-program key size	non-16-byte keys rejected with WH_ERROR_BADARGS, leaving no NVM object behind
Oversized auth key	32-byte object planted via wh_Client_NvmAddObject, used as the LoadKey AUTH id → WH_SHE_ERC_KEY_INVALID
Oversized target key	same object planted at a write-protected TARGET slot → WH_SHE_ERC_KEY_INVALID

Verification

• Build clean under -Werror -Wall -Wextra, C90.
• Functional: 52 passed, 22 skipped, 0 failed of 74 (make SHE=1). ASan clean.
• Negative controls: each hunk reverted individually fails only its own case. Without the auth-key bound, ASan reports a 16-byte stack-buffer-overflow past kdfInput in _LoadKey; without the target-key bound, wh_Client_SheLoadKey returns success against a write-protected slot.

[Copilot]

Pull request overview

This PR hardens the server-side SHE LOAD_KEY handler against oversized/undersized auth-key objects by bounding keystore reads to WH_SHE_KEY_SZ, ensuring consistent error mapping, and adding targeted regression tests in the refactored test suite.

Changes:

• Fix _LoadKey to read the auth key with capacity WH_SHE_KEY_SZ and reject short reads / WH_ERROR_NOSPACE as WH_SHE_ERC_KEY_INVALID.
• Pin the target-slot keystore read size to WH_SHE_KEY_SZ (instead of inheriting keySz from the auth-key read).
• Add a new refactored server test that plants 32/8/16-byte auth-key objects and validates the handler’s return codes.

Reviewed changes

Copilot reviewed 3 out of 3 changed files in this pull request and generated 1 comment.

File	Description
src/wh_server_she.c	Bounds auth-key read size, normalizes error returns for invalid auth-key storage sizes, and pins the subsequent target-slot read size.
test-refactor/server/wh_test_she_loadkey.c	Adds a regression test for LOAD_KEY auth-key length handling (oversized/undersized/valid).
test-refactor/wh_test_list.c	Registers the new server test in the refactored test suite.

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

[wolfSSL-Fenrir-bot]

Fenrir Automated Review — PR #495

Scan targets checked: wolfhsm-core-bugs, wolfhsm-src

No new issues found in the changed files. ✅

[bigbrett]

@yosuke-wolfssl merge conflicts

[yosuke-wolfssl]

Hi @bigbrett ,
I solved the merge conflict. Please check this as well.