x86/amd_node: Fix PCI device reference counting in amd_smn_init()grafted

The local "root" pointer is a temporary variable used during the device
search. Therefore, refcount related to the search iterators should be cleaned
up after the search is complete.

Use the __free() cleanup macro to ensure the refcount is decremented when the
temporary pointer goes out of scope.

Additionally, increment the refcount when caching a root pointer. This ensures
the in-use refcount is separate from the temporary search refcounting.

Finally, drop the redundant "root = NULL" before the second search loop. The
pci_get_class() iterator always decrements the refcount of its "from"
argument, so the first loop can only fall through with "root" already NULL.

Fixes: 0a4b61d9c2e4 ("x86/amd_node: Fix AMD root device caching")
Closes: https://sashiko.dev/#/patchset/20260806160159.230453-1-jason.andryuk%40amd.com
Reported-by: Sashiko <sashiko-bot@kernel.org>
Assisted-by: LLM
Signed-off-by: Yazen Ghannam <yazen.ghannam@amd.com>
Signed-off-by: Borislav Petkov (AMD) <bp@alien8.de>
Reviewed-by: Mario Limonciello (AMD) <superm1@kernel.org>
Cc: <stable@kernel.org>
Link: https://patch.msgid.link/20260903154325.74343-1-yazen.ghannam@amd.com

1 files changed, 56 insertions, 0 deletions

diff --git a/lib/crypto/arm/aes.h b/lib/crypto/arm/aes.h
new file mode 100644
index 000000000..1dd7dfa65
--- /dev/null
+++ b/lib/crypto/arm/aes.h
@@ -0,0 +1,56 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
+/*
+ * AES block cipher, optimized for ARM
+ *
+ * Copyright (C) 2017 Linaro Ltd.
+ * Copyright 2026 Google LLC
+ */
+
+asmlinkage void __aes_arm_encrypt(const u32 rk[], int rounds,
+				  const u8 in[AES_BLOCK_SIZE],
+				  u8 out[AES_BLOCK_SIZE]);
+asmlinkage void __aes_arm_decrypt(const u32 inv_rk[], int rounds,
+				  const u8 in[AES_BLOCK_SIZE],
+				  u8 out[AES_BLOCK_SIZE]);
+
+static void aes_preparekey_arch(union aes_enckey_arch *k,
+				union aes_invkey_arch *inv_k,
+				const u8 *in_key, int key_len, int nrounds)
+{
+	aes_expandkey_generic(k->rndkeys, inv_k ? inv_k->inv_rndkeys : NULL,
+			      in_key, key_len);
+}
+
+static void aes_encrypt_arch(const struct aes_enckey *key,
+			     u8 out[AES_BLOCK_SIZE],
+			     const u8 in[AES_BLOCK_SIZE])
+{
+	if (!IS_ENABLED(CONFIG_HAVE_EFFICIENT_UNALIGNED_ACCESS) &&
+	    !IS_ALIGNED((uintptr_t)out | (uintptr_t)in, 4)) {
+		u8 bounce_buf[AES_BLOCK_SIZE] __aligned(4);
+
+		memcpy(bounce_buf, in, AES_BLOCK_SIZE);
+		__aes_arm_encrypt(key->k.rndkeys, key->nrounds, bounce_buf,
+				  bounce_buf);
+		memcpy(out, bounce_buf, AES_BLOCK_SIZE);
+		return;
+	}
+	__aes_arm_encrypt(key->k.rndkeys, key->nrounds, in, out);
+}
+
+static void aes_decrypt_arch(const struct aes_key *key,
+			     u8 out[AES_BLOCK_SIZE],
+			     const u8 in[AES_BLOCK_SIZE])
+{
+	if (!IS_ENABLED(CONFIG_HAVE_EFFICIENT_UNALIGNED_ACCESS) &&
+	    !IS_ALIGNED((uintptr_t)out | (uintptr_t)in, 4)) {
+		u8 bounce_buf[AES_BLOCK_SIZE] __aligned(4);
+
+		memcpy(bounce_buf, in, AES_BLOCK_SIZE);
+		__aes_arm_decrypt(key->inv_k.inv_rndkeys, key->nrounds,
+				  bounce_buf, bounce_buf);
+		memcpy(out, bounce_buf, AES_BLOCK_SIZE);
+		return;
+	}
+	__aes_arm_decrypt(key->inv_k.inv_rndkeys, key->nrounds, in, out);
+}