Skip to contentSkip to conversation

Remove old AES-GCM library

PATCH Applied Eric Biggers 4 participants 10 messages Jul 22 – Jul 30

Subsystems (3)
lore
Share & tools
lore

Eric Biggers

author maintainer
#0 lore raw

cover Remove old AES-GCM library

The old AES-GCM library code is being replaced as part of an overhaul that is adding support for all the common AES encryption modes with consistent conventions
(https://lore.kernel.org/linux-crypto/[email protected]/).

The SEV code is the only user of the old API. Update it to use the new API instead. Then remove the old API.

This applies to libcrypto-next. If possible I'd like to take this through libcrypto-next for 7.3. Acks are appreciated.

Eric Biggers (3):
x86/sev: Use new AES-GCM library
x86/sev: Remove obsolete virtual address check
lib/crypto: aesgcm: Remove old AES-GCM library

 arch/x86/Kconfig                        |   2 +-
 arch/x86/coco/sev/core.c                |  53 +-
 arch/x86/include/asm/sev.h              |   2 +-
 drivers/virt/coco/sev-guest/sev-guest.c |   7 +-
 include/crypto/gcm.h                    |  21 -
 lib/crypto/Kconfig                      |   6 -
 lib/crypto/Makefile                     |   3 -
 lib/crypto/aesgcm.c                     | 721 ------------------------
 8 files changed, 26 insertions(+), 789 deletions(-)
 delete mode 100644 lib/crypto/aesgcm.c

base-commit: e4f23159c8d09b6b15d47f2c95e0a3e7c62f52a2

Signature · 1 line
2.55.0
author maintainer
#1 lore raw

PATCH 1/3 x86/sev: Use new AES-GCM library

The old AES-GCM library code is being replaced as part of an overhaul that is adding support for all the common AES encryption modes with consistent conventions. The SEV code is the only user of the old AES-GCM library. Update it to use the new API instead.

Besides adjustments to the key struct name, function names, return value, and parameter order, the only notable changes are:

- Replace the direct accesses to the auth tag length field (which should be considered private until someone truly needs it) with the AUTHTAG_LEN constant which is already defined in <asm/sev.h>.

- Replace kfree() with kfree_sensitive() on lines being changed anyway.

---
 arch/x86/Kconfig                        |  2 +-
 arch/x86/coco/sev/core.c                | 44 ++++++++++++-------------
 arch/x86/include/asm/sev.h              |  2 +-
 drivers/virt/coco/sev-guest/sev-guest.c |  7 ++--
 4 files changed, 26 insertions(+), 29 deletions(-)
arch/x86/Kconfig +1 -1
index bdad90f210e4b..0b89641cac160 100644
@@ -1495,7 +1495,7 @@ config AMD_MEM_ENCRYPT
 	select ARCH_HAS_CC_PLATFORM
 	select X86_MEM_ENCRYPT
 	select UNACCEPTED_MEMORY
-	select CRYPTO_LIB_AESGCM
+	select CRYPTO_LIB_AES_GCM
 	help
 	  Say yes to enable support for the encryption of system memory.
 	  This requires an AMD processor that supports Secure Memory
arch/x86/coco/sev/core.c +21 -23

106 lines not on this page · raw message

arch/x86/include/asm/sev.h +1 -1
index 594cfa19cbd4b..9e7a077c445dc 100644
@@ -314,7 +314,7 @@ struct snp_msg_desc {
 
 	struct snp_secrets_page *secrets;
 
-	struct aesgcm_ctx *ctx;
+	struct aes_gcm_key *gcm_key;
 
 	u32 *os_area_msg_seqno;
 	u8 *vmpck;
drivers/virt/coco/sev-guest/sev-guest.c +3 -4
index d186ae55cf63b..935537a41469b 100644
@@ -17,7 +17,6 @@
 #include <linux/set_memory.h>
 #include <linux/fs.h>
 #include <linux/tsm.h>
-#include <crypto/gcm.h>
 #include <linux/psp-sev.h>
 #include <linux/sockptr.h>
 #include <linux/cleanup.h>
@@ -87,7 +86,7 @@ static int get_report(struct snp_guest_dev *snp_dev, struct snp_guest_request_io
 	 * response payload. Make sure that it has enough space to cover the
 	 * authtag.
 	 */
-	resp_len = sizeof(report_resp->data) + mdesc->ctx->authsize;
+	resp_len = sizeof(report_resp->data) + AUTHTAG_LEN;
 	report_resp = kzalloc(resp_len, GFP_KERNEL_ACCOUNT);
 	if (!report_resp)
 		return -ENOMEM;
@@ -130,7 +129,7 @@ static int get_derived_key(struct snp_guest_dev *snp_dev, struct snp_guest_reque
 	 * response payload. Make sure that it has enough space to cover the
 	 * authtag.
 	 */
-	resp_len = sizeof(derived_key_resp->data) + mdesc->ctx->authsize;
+	resp_len = sizeof(derived_key_resp->data) + AUTHTAG_LEN;
 	derived_key_resp = kzalloc(resp_len, GFP_KERNEL_ACCOUNT);
 	if (!derived_key_resp)
 		return -ENOMEM;
@@ -230,7 +229,7 @@ static int get_ext_report(struct snp_guest_dev *snp_dev, struct snp_guest_reques
 	 * response payload. Make sure that it has enough space to cover the
 	 * authtag.
 	 */
-	resp_len = sizeof(report_resp->data) + mdesc->ctx->authsize;
+	resp_len = sizeof(report_resp->data) + AUTHTAG_LEN;
 	report_resp = kzalloc(resp_len, GFP_KERNEL_ACCOUNT);
 	if (!report_resp) {
 		ret = -ENOMEM;
Signature · 1 line
2.55.0
author maintainer
#2 lore raw

PATCH 2/3 x86/sev: Remove obsolete virtual address check

In reply to Eric Biggers

Virtual addresses used with the AES-GCM library don't have to be in the linear mapping. Remove the virt_addr_valid() checks from snp_send_guest_request(), which were a workaround left over from when this code used crypto_aead, which required scatterlists.

---
 arch/x86/coco/sev/core.c | 9 ---------
 1 file changed, 9 deletions(-)
arch/x86/coco/sev/core.c +0 -9
index 54b122f6ef96a..cc292d7c6fd1e 100644
@@ -1867,15 +1867,6 @@ int snp_send_guest_request(struct snp_msg_desc *mdesc, struct snp_guest_req *req
 	u64 seqno;
 	int rc;
 
-	/*
-	 * enc_payload() calls aesgcm_encrypt(), which can potentially offload to HW.
-	 * The offload's DMA SG list of data to encrypt has to be in linear mapping.
-	 */
-	if (!virt_addr_valid(req->req_buf) || !virt_addr_valid(req->resp_buf)) {
-		pr_warn("AES-GSM buffers must be in linear mapping");
-		return -EINVAL;
-	}
-
 	guard(mutex)(&snp_cmd_mutex);
 
 	/* Check if the VMPCK is not empty */
Signature · 1 line
2.55.0
maintainer
#3 lore raw

Re: [PATCH 2/3] x86/sev: Remove obsolete virtual address check

Quoted 10 lines · Eric Biggers

Virtual addresses used with the AES-GCM library don't have to be in the linear mapping. Remove the virt_addr_valid() checks from snp_send_guest_request(), which were a workaround left over from when this code used crypto_aead, which required scatterlists.

---
 arch/x86/coco/sev/core.c | 9 ---------
 1 file changed, 9 deletions(-)
Quoted 23 lines
arch/x86/coco/sev/core.c +0 -9
index 54b122f6ef96a..cc292d7c6fd1e 100644
@@ -1867,15 +1867,6 @@ int snp_send_guest_request(struct snp_msg_desc 
 *mdesc, struct snp_guest_req *req
 	u64 seqno;
 	int rc;
 
-	/*
-	 * enc_payload() calls aesgcm_encrypt(), which can potentially offload to HW.
-	 * The offload's DMA SG list of data to encrypt has to be in linear mapping.
-	 */
-	if (!virt_addr_valid(req->req_buf) || !virt_addr_valid(req->resp_buf)) {
-		pr_warn("AES-GSM buffers must be in linear mapping");
-		return -EINVAL;
-	}
-
 	guard(mutex)(&snp_cmd_mutex);
 

/* Check if the VMPCK is not empty */

Signature · 1 line
2.55.0
author maintainer
#4 lore raw

Re: [PATCH 2/3] x86/sev: Remove obsolete virtual address check

Quoted 8 lines · Ard Biesheuvel
Quoted 5 lines · Eric Biggers

Virtual addresses used with the AES-GCM library don't have to be in the linear mapping. Remove the virt_addr_valid() checks from snp_send_guest_request(), which were a workaround left over from when this code used crypto_aead, which required scatterlists.

Quoted 8 lines
Quoted 5 lines
---
 arch/x86/coco/sev/core.c | 9 ---------
 1 file changed, 9 deletions(-)

Thanks, I completely missed that. Looks like the usual issue where people were attached to workarounds even after the underlying problem was fixed. In the end, Alexey gave Reviewed-by and Tested-by on the exact same diff, just it never got applied. Alexey, would you mind if I add those tags to this patch too? Thanks!

- Eric

author maintainer
#5 lore raw

PATCH 3/3 lib/crypto: aesgcm: Remove old AES-GCM library

In reply to Eric Biggers

The old AES-GCM library code is no longer used, so remove it.

---
 include/crypto/gcm.h |  21 --
 lib/crypto/Kconfig   |   6 -
 lib/crypto/Makefile  |   3 -
 lib/crypto/aesgcm.c  | 721 -------------------------------------------
 4 files changed, 751 deletions(-)
 delete mode 100644 lib/crypto/aesgcm.c
include/crypto/gcm.h +0 -21
index 7fd7892ad818e..154038ca73430 100644
@@ -3,9 +3,6 @@
 
 #include <linux/errno.h>
 
-#include <crypto/aes.h>
-#include <crypto/gf128hash.h>
-
 #define GCM_AES_IV_SIZE 12
 #define GCM_RFC4106_IV_SIZE 8
 #define GCM_RFC4543_IV_SIZE 8
@@ -64,22 +61,4 @@ static inline int crypto_ipsec_check_assoclen(unsigned int assoclen)
 	return 0;
 }
 
-struct aesgcm_ctx {
-	struct ghash_key	ghash_key;
-	struct aes_enckey	aes_key;
-	unsigned int		authsize;
-};
-
-int aesgcm_expandkey(struct aesgcm_ctx *ctx, const u8 *key,
-		     unsigned int keysize, unsigned int authsize);
-
-void aesgcm_encrypt(const struct aesgcm_ctx *ctx, u8 *dst, const u8 *src,
-		    int crypt_len, const u8 *assoc, int assoc_len,
-		    const u8 iv[GCM_AES_IV_SIZE], u8 *authtag);
-
-bool __must_check aesgcm_decrypt(const struct aesgcm_ctx *ctx, u8 *dst,
-				 const u8 *src, int crypt_len, const u8 *assoc,
-				 int assoc_len, const u8 iv[GCM_AES_IV_SIZE],
-				 const u8 *authtag);
-
 #endif
lib/crypto/Kconfig +0 -6
index 65a478f69715d..32c26b19127b4 100644
@@ -75,12 +75,6 @@ config CRYPTO_LIB_AES_XTS
 	help
 	  The AES-XTS library functions.
 
-config CRYPTO_LIB_AESGCM
-	tristate
-	select CRYPTO_LIB_AES
-	select CRYPTO_LIB_GF128HASH
-	select CRYPTO_LIB_UTILS
-
 config CRYPTO_LIB_ARC4
 	tristate
 
lib/crypto/Makefile +0 -3
index f1e9bf89785ff..ca068df1f71f2 100644
@@ -63,9 +63,6 @@ clean-files += powerpc/aesp8-ppc.S
 obj-$(CONFIG_CRYPTO_LIB_AESCFB)			+= libaescfb.o
 libaescfb-y					:= aescfb.o
 
-obj-$(CONFIG_CRYPTO_LIB_AESGCM)			+= libaesgcm.o
-libaesgcm-y					:= aesgcm.o
-
 obj-$(CONFIG_CRYPTO_LIB_ARC4)			+= libarc4.o
 libarc4-y					:= arc4.o
 
lib/crypto/aesgcm.c deleted +0 -721

721 lines not on this page · raw message

Signature · 1 line
2.55.0
maintainer Reviewed-by on the v1 series
#6 lore raw
In reply to Eric Biggers
Quoted 8 lines · Eric Biggers

The old AES-GCM library code is being replaced as part of an overhaul that is adding support for all the common AES encryption modes with consistent conventions
(https://lore.kernel.org/linux-crypto/[email protected]/).

The SEV code is the only user of the old API. Update it to use the new API instead. Then remove the old API.

Quoted 8 lines

This applies to libcrypto-next. If possible I'd like to take this through libcrypto-next for 7.3. Acks are appreciated.

Eric Biggers (3):
x86/sev: Use new AES-GCM library
x86/sev: Remove obsolete virtual address check
lib/crypto: aesgcm: Remove old AES-GCM library

Quoted 15 lines
 arch/x86/Kconfig                        |   2 +-
 arch/x86/coco/sev/core.c                |  53 +-
 arch/x86/include/asm/sev.h              |   2 +-
 drivers/virt/coco/sev-guest/sev-guest.c |   7 +-
 include/crypto/gcm.h                    |  21 -
 lib/crypto/Kconfig                      |   6 -
 lib/crypto/Makefile                     |   3 -
 lib/crypto/aesgcm.c                     | 721 ------------------------
 8 files changed, 26 insertions(+), 789 deletions(-)
 delete mode 100644 lib/crypto/aesgcm.c

base-commit: e4f23159c8d09b6b15d47f2c95e0a3e7c62f52a2

Signature · 1 line
2.55.0
maintainer
#7 lore raw
In reply to Eric Biggers
Quoted 10 lines · Eric Biggers

The old AES-GCM library code is being replaced as part of an overhaul that is adding support for all the common AES encryption modes with consistent conventions
(https://lore.kernel.org/linux-crypto/[email protected]/).

The SEV code is the only user of the old API. Update it to use the new API instead. Then remove the old API.

This applies to libcrypto-next. If possible I'd like to take this through libcrypto-next for 7.3. Acks are appreciated.

Fine with me as long as the additionally added AMD folks can verify everything still works with their SNP guest setups.

Thx.

Signature · 4 lines
Regards/Gruss,
    Boris.

https://people.kernel.org/tglx/notes-about-netiquette
#8 lore raw
Quoted 6 lines · Borislav Petkov
Quoted 5 lines · Eric Biggers

The old AES-GCM library code is being replaced as part of an overhaul that is adding support for all the common AES encryption modes with consistent conventions
(https://lore.kernel.org/linux-crypto/[email protected]/).

Quoted 8 lines
Quoted 5 lines

The SEV code is the only user of the old API. Update it to use the new API instead. Then remove the old API.

This applies to libcrypto-next. If possible I'd like to take this through libcrypto-next for 7.3. Acks are appreciated.

Fine with me as long as the additionally added AMD folks can verify everything still works with their SNP guest setups.

Tested with a SecureTSC-enabled SNP guest. SecureTSC guests rely on the SNP guest messaging API which uses AES-GCM library, so this exercises the new library code through that path. No issues found.

author maintainer Applied
#9 lore raw
In reply to Eric Biggers
Quoted 10 lines · Eric Biggers

The old AES-GCM library code is being replaced as part of an overhaul that is adding support for all the common AES encryption modes with consistent conventions
(https://lore.kernel.org/linux-crypto/[email protected]/).

The SEV code is the only user of the old API. Update it to use the new API instead. Then remove the old API.

This applies to libcrypto-next. If possible I'd like to take this through libcrypto-next for 7.3. Acks are appreciated.

Applied to https://git.kernel.org/pub/scm/linux/kernel/git/ebiggers/linux.git/log/?h=libcrypto-next

- Eric