Skip to contentSkip to conversation

Allow guest access to EFI confidential computing secret area

PATCH In review Dov Murik 2 participants 6 messages Apr 12, 2022 – Apr 13, 2022

Subsystems (4)
lore
Share & tools
lore

Dov Murik

author
#0 lore raw

v10 cover Allow guest access to EFI confidential computing secret area

Confidential computing (coco) hardware such as AMD SEV (Secure Encrypted Virtualization) allows guest owners to inject secrets into the VMs memory without the host/hypervisor being able to read them. In SEV, secret injection is performed early in the VM launch process, before the guest starts running.

OVMF already reserves designated area for secret injection (in its AmdSev package; see edk2 commit 01726b6d23d4 "OvmfPkg/AmdSev: Expose the Sev Secret area using a configuration table" [1]), but the secrets were not available in the guest kernel.

The patch series keeps the address of the EFI-provided memory for injected secrets, and exposes the secrets to userspace via securityfs using a new efi_secret kernel module. The module is autoloaded (by the EFI driver) if the secret area is populated.

The first patch in EFI keeps the address of the secret area as passed in the EFI configuration table. The second patch introduces the new efi_secret module that exposes the content of the secret entries as securityfs files, and allows clearing out secrets with a file unlink interface. The third patch auto-loads the efi_secret module during startup if the injected secrets area is populated. The last patch documents the data flow of confidential computing secret injection.

As a usage example, consider a guest performing computations on encrypted files. The Guest Owner provides the decryption key (= secret) using the secret injection mechanism. The guest application reads the secret from the efi_secret filesystem and proceeds to decrypt the files into memory and then performs the needed computations on the content.

In this example, the host can't read the files from the disk image because they are encrypted. Host can't read the decryption key because it is passed using the secret injection mechanism (= secure channel). Host can't read the decrypted content from memory because it's a confidential (memory-encrypted) guest.

This has been tested with AMD SEV and SEV-ES guests, but the kernel side of handling the secret area has no SEV-specific dependencies, and therefore might be usable (perhaps with minor changes) for any confidential computing hardware that can publish the secret area via the standard EFI config table entry.

To enable this functionality, set CONFIG_EFI_SECRET=m when building the guest kernel.

Here is a simple example for usage of the efi_secret module in a guest to which an EFI secret area with 4 secrets was injected during launch:

...

[    0.000000] efi: EFI v2.70 by EDK II
[    0.000000] efi: CocoSecret=0x7f222680 SMBIOS=0x7f541000 ACPI=0x7f77e000 ACPI 2.0=0x7f77e014 MEMATTR=0x7ea16418
...
[    1.127627] Run /init as init process
Loading, please wait...
Starting version 245.4-4ubuntu3.15
...
[    0.763204] efi_secret efi_secret.0: Created 4 entries in securityfs secrets/coco

...

# ls -la /sys/kernel/security/secrets/coco
total 0
drwxr-xr-x 2 root root 0 Jun 28 11:54 .
drwxr-xr-x 3 root root 0 Jun 28 11:54 ..
-r--r----- 1 root root 0 Jun 28 11:54 736870e5-84f0-4973-92ec-06879ce3da0b
-r--r----- 1 root root 0 Jun 28 11:54 83c83f7f-1356-4975-8b7e-d3a0b54312c6
-r--r----- 1 root root 0 Jun 28 11:54 9553f55d-3da2-43ee-ab5d-ff17f78864d2
-r--r----- 1 root root 0 Jun 28 11:54 e6f5a162-d67f-4750-a67c-5d065f2a9910
# hd /sys/kernel/security/secrets/coco/e6f5a162-d67f-4750-a67c-5d065f2a9910
00000000  74 68 65 73 65 2d 61 72  65 2d 74 68 65 2d 6b 61  |these-are-the-ka|
00000010  74 61 2d 73 65 63 72 65  74 73 00 01 02 03 04 05  |ta-secrets......|
00000020  06 07                                             |..|
00000022
# rm /sys/kernel/security/secrets/coco/e6f5a162-d67f-4750-a67c-5d065f2a9910
# ls -la /sys/kernel/security/secrets/coco
total 0
drwxr-xr-x 2 root root 0 Jun 28 11:55 .
drwxr-xr-x 3 root root 0 Jun 28 11:54 ..
-r--r----- 1 root root 0 Jun 28 11:54 736870e5-84f0-4973-92ec-06879ce3da0b
-r--r----- 1 root root 0 Jun 28 11:54 83c83f7f-1356-4975-8b7e-d3a0b54312c6
-r--r----- 1 root root 0 Jun 28 11:54 9553f55d-3da2-43ee-ab5d-ff17f78864d2

[1] https://github.com/tianocore/edk2/commit/01726b6d23d4

---

v10 changes:
- Add #ifdef CONFIG_EFI_COCO_SECRET to platform device registration in patch 3.

v9: https://lore.kernel.org/linux-coco/[email protected]/
v9 changes:
- Change the module into a platform driver (thanks Ard)
- Remove special auto-loading code in efi; instead register a platform device (udev will load the efi_secret module) (thanks Ard)
- Change logging in the efi_secret module to dev_err() etc.
- efi_secret: first check that the secret area header is valid; only then start creating securityfs dirs.

v8: https://lore.kernel.org/linux-coco/[email protected]/
v8 changes:
- Change path of filesystem to <securityfs>/secrets/coco and fix the documentation accordingly (Thanks Gerd, Matthew)
- Remove patch 2/5 (of v7) because the latest OVMF release (edk2-stable202202) already contains the fix to mark the launch secret page as EFI_RESERVED_TYPE.

v7: https://lore.kernel.org/linux-coco/[email protected]/
v7 changes:
- Improve description of efi_secret module in Kconfig.
- Fix sparse warnings on pointer address space mismatch
(Reported-by: kernel test robot <[email protected]>)

v6: https://lore.kernel.org/linux-coco/[email protected]/
v6 changes:
- Autoload the efi_secret module if the secret area is populated
(thanks Greg KH).
- efi_secret: Depend on X86_64 because we use ioremap_encrypted() which is only defined for this arch.
- efi_secret.c: Remove unneeded tableheader_guid local variable.
- Documentation fixes.

v5: https://lore.kernel.org/linux-coco/[email protected]/
v5 changes:
- Simplify EFI code: instead of copying the secret area, the firmware marks the secret area as EFI_RESERVED_TYPE, and then the uefi_init() code just keeps the pointer as it appears in the EFI configuration table. The use of reserved pages is similar to the AMD SEV-SNP patches for handling SNP-Secrets and SNP-CPUID pages.
- In order to handle OVMF releases out there which mark the confidential computing secrets page as EFI_BOOT_SERVICES_DATA, add efi/libstub code that detects this and fixes the E820 map to reserve this page.
- In the efi_secret module code, map the secrets page using ioremap_encrypted (again, similar to the AMD SEV-SNP guest patches for accessing SNP-Secrets and SNP-CPUID pages).
- Add documentation in Documentation/security/coco/efi_secret.

v4: https://lore.kernel.org/linux-coco/[email protected]/
v4 changes:
- Guard all the new EFI and efi-stub code (patches 1+2) with #ifdef CONFIG_EFI_COCO_SECRET (thanks Greg KH). Selecting CONFIG_EFI_SECRET=m (patch 3) will enable the EFI parts as well.
- Guard call to clflush_cache_range() with #ifdef CONFIG_X86 (Reported-by: kernel test robot <[email protected]>)

v3: https://lore.kernel.org/linux-coco/[email protected]/
v3 changes:
- Rename the module to efi_secret
- Remove the exporting of clean_cache_range
- Use clflush_cache_range in wipe_memory
- Document function wipe_memory
- Initialize efi.coco_secret to EFI_INVALID_TABLE_ADDR to correctly detect when there's no secret area published in the EFI configuration tables

v2: https://lore.kernel.org/linux-coco/[email protected]
v2 changes:
- Export clean_cache_range()
- When deleteing a secret, call clean_cache_range() after explicit_memzero
- Add Documentation/ABI/testing/securityfs-coco-sev_secret

v1: https://lore.kernel.org/linux-coco/[email protected]/

RFC: https://lore.kernel.org/linux-coco/[email protected]/

Dov Murik (4):
efi: Save location of EFI confidential computing area
virt: Add efi_secret module to expose confidential computing secrets
efi: Register efi_secret platform device if EFI secret area is
declared
docs: security: Add secrets/coco documentation

 Documentation/ABI/testing/securityfs-secrets-coco |  51 +++
 Documentation/security/index.rst                  |   1 +
 Documentation/security/secrets/coco.rst           | 103 ++++++
 Documentation/security/secrets/index.rst          |   9 +
 arch/x86/platform/efi/efi.c                       |   3 +
 drivers/firmware/efi/Kconfig                      |  16 +
 drivers/firmware/efi/efi.c                        |  11 +
 drivers/virt/Kconfig                              |   3 +
 drivers/virt/Makefile                             |   1 +
 drivers/virt/coco/efi_secret/Kconfig              |  16 +
 drivers/virt/coco/efi_secret/Makefile             |   2 +
 drivers/virt/coco/efi_secret/efi_secret.c         | 349 ++++++++++++++++++++
 include/linux/efi.h                               |  10 +
 13 files changed, 575 insertions(+)
 create mode 100644 Documentation/ABI/testing/securityfs-secrets-coco
 create mode 100644 Documentation/security/secrets/coco.rst
 create mode 100644 Documentation/security/secrets/index.rst
 create mode 100644 drivers/virt/coco/efi_secret/Kconfig
 create mode 100644 drivers/virt/coco/efi_secret/Makefile
 create mode 100644 drivers/virt/coco/efi_secret/efi_secret.c

base-commit: 7e57714cd0ad2d5bb90e50b5096a0e671dec1ef3

Signature · 1 line
2.25.1
author
#1 lore raw

PATCH v10 1/4 efi: Save location of EFI confidential computing area

Confidential computing (coco) hardware such as AMD SEV (Secure Encrypted Virtualization) allows a guest owner to inject secrets into the VMs memory without the host/hypervisor being able to read them.

Firmware support for secret injection is available in OVMF, which reserves a memory area for secret injection and includes a pointer to it the in EFI config table entry LINUX_EFI_COCO_SECRET_TABLE_GUID.

If EFI exposes such a table entry, uefi_init() will keep a pointer to the EFI config table entry in efi.coco_secret, so it can be used later by the kernel (specifically drivers/virt/coco/efi_secret). It will also appear in the kernel log as "CocoSecret=ADDRESS"; for example:

    [    0.000000] efi: EFI v2.70 by EDK II
    [    0.000000] efi: CocoSecret=0x7f22e680 SMBIOS=0x7f541000 ACPI=0x7f77e000 ACPI 2.0=0x7f77e014 MEMATTR=0x7ea0c018

The new functionality can be enabled with CONFIG_EFI_COCO_SECRET=y.

---
 arch/x86/platform/efi/efi.c  |  3 +++
 drivers/firmware/efi/Kconfig | 16 ++++++++++++++++
 drivers/firmware/efi/efi.c   |  6 ++++++
 include/linux/efi.h          | 10 ++++++++++
 4 files changed, 35 insertions(+)
arch/x86/platform/efi/efi.c +3 -0
index 147c30a81f15..1591d67e0bcd 100644
@@ -93,6 +93,9 @@ static const unsigned long * const efi_tables[] = {
 #ifdef CONFIG_LOAD_UEFI_KEYS
 	&efi.mokvar_table,
 #endif
+#ifdef CONFIG_EFI_COCO_SECRET
+	&efi.coco_secret,
+#endif
 };
 
 u64 efi_setup;		/* efi setup_data physical address */
drivers/firmware/efi/Kconfig +16 -0
index 2c3dac5ecb36..6fa251b3709f 100644
@@ -284,3 +284,19 @@ config EFI_CUSTOM_SSDT_OVERLAYS
 
 	  See Documentation/admin-guide/acpi/ssdt-overlays.rst for more
 	  information.
+
+config EFI_COCO_SECRET
+	bool "EFI Confidential Computing Secret Area Support"
+	depends on EFI
+	help
+	  Confidential Computing platforms (such as AMD SEV) allow the
+	  Guest Owner to securely inject secrets during guest VM launch.
+	  The secrets are placed in a designated EFI reserved memory area.
+
+	  In order to use the secrets in the kernel, the location of the secret
+	  area (as published in the EFI config table) must be kept.
+
+	  If you say Y here, the address of the EFI secret area will be kept
+	  for usage inside the kernel.  This will allow the
+	  virt/coco/efi_secret module to access the secrets, which in turn
+	  allows userspace programs to access the injected secrets.
drivers/firmware/efi/efi.c +6 -0
index 7de3f5b6e8d0..378d044b2463 100644
@@ -46,6 +46,9 @@ struct efi __read_mostly efi = {
 #ifdef CONFIG_LOAD_UEFI_KEYS
 	.mokvar_table		= EFI_INVALID_TABLE_ADDR,
 #endif
+#ifdef CONFIG_EFI_COCO_SECRET
+	.coco_secret		= EFI_INVALID_TABLE_ADDR,
+#endif
 };
 EXPORT_SYMBOL(efi);
 
@@ -528,6 +531,9 @@ static const efi_config_table_type_t common_tables[] __initconst = {
 #endif
 #ifdef CONFIG_LOAD_UEFI_KEYS
 	{LINUX_EFI_MOK_VARIABLE_TABLE_GUID,	&efi.mokvar_table,	"MOKvar"	},
+#endif
+#ifdef CONFIG_EFI_COCO_SECRET
+	{LINUX_EFI_COCO_SECRET_AREA_GUID,	&efi.coco_secret,	"CocoSecret"	},
 #endif
 	{},
 };
include/linux/efi.h +10 -0
index ccd4d3f91c98..771d4cd06b56 100644
@@ -405,6 +405,7 @@ void efi_native_runtime_setup(void);
 #define LINUX_EFI_MEMRESERVE_TABLE_GUID		EFI_GUID(0x888eb0c6, 0x8ede, 0x4ff5,  0xa8, 0xf0, 0x9a, 0xee, 0x5c, 0xb9, 0x77, 0xc2)
 #define LINUX_EFI_INITRD_MEDIA_GUID		EFI_GUID(0x5568e427, 0x68fc, 0x4f3d,  0xac, 0x74, 0xca, 0x55, 0x52, 0x31, 0xcc, 0x68)
 #define LINUX_EFI_MOK_VARIABLE_TABLE_GUID	EFI_GUID(0xc451ed2b, 0x9694, 0x45d3,  0xba, 0xba, 0xed, 0x9f, 0x89, 0x88, 0xa3, 0x89)
+#define LINUX_EFI_COCO_SECRET_AREA_GUID		EFI_GUID(0xadf956ad, 0xe98c, 0x484c,  0xae, 0x11, 0xb5, 0x1c, 0x7d, 0x33, 0x64, 0x47)
 
 /* OEM GUIDs */
 #define DELLEMC_EFI_RCI2_TABLE_GUID		EFI_GUID(0x2d9f28a2, 0xa886, 0x456a,  0x97, 0xa8, 0xf1, 0x1e, 0xf2, 0x4f, 0xf4, 0x55)
@@ -596,6 +597,7 @@ extern struct efi {
 	unsigned long			tpm_log;		/* TPM2 Event Log table */
 	unsigned long			tpm_final_log;		/* TPM2 Final Events Log table */
 	unsigned long			mokvar_table;		/* MOK variable config table */
+	unsigned long			coco_secret;		/* Confidential computing secret table */
 
 	efi_get_time_t			*get_time;
 	efi_set_time_t			*set_time;
@@ -1335,4 +1337,12 @@ extern void efifb_setup_from_dmi(struct screen_info *si, const char *opt);
 static inline void efifb_setup_from_dmi(struct screen_info *si, const char *opt) { }
 #endif
 
+struct linux_efi_coco_secret_area {
+	u64	base_pa;
+	u64	size;
+};
+
+/* Header of a populated EFI secret area */
+#define EFI_SECRET_TABLE_HEADER_GUID	EFI_GUID(0x1e74f542, 0x71dd, 0x4d66,  0x96, 0x3e, 0xef, 0x42, 0x87, 0xff, 0x17, 0x3b)
+
 #endif /* _LINUX_EFI_H */
Signature · 1 line
2.25.1
author
#2 lore raw

PATCH v10 2/4 virt: Add efi_secret module to expose confidential computing secrets

In reply to Dov Murik

The new efi_secret module exposes the confidential computing (coco) EFI secret area via securityfs interface.

When the module is loaded (and securityfs is mounted, typically under /sys/kernel/security), a "secrets/coco" directory is created in securityfs. In it, a file is created for each secret entry. The name of each such file is the GUID of the secret entry, and its content is the secret data.

This allows applications running in a confidential computing setting to read secrets provided by the guest owner via a secure secret injection mechanism (such as AMD SEV's LAUNCH_SECRET command).

Removing (unlinking) files in the "secrets/coco" directory will zero out the secret in memory, and remove the filesystem entry. If the module is removed and loaded again, that secret will not appear in the filesystem.

---
 Documentation/ABI/testing/securityfs-secrets-coco |  51 +++
 drivers/virt/Kconfig                              |   3 +
 drivers/virt/Makefile                             |   1 +
 drivers/virt/coco/efi_secret/Kconfig              |  16 +
 drivers/virt/coco/efi_secret/Makefile             |   2 +
 drivers/virt/coco/efi_secret/efi_secret.c         | 349 ++++++++++++++++++++
 6 files changed, 422 insertions(+)
Documentation/ABI/testing/securityfs-secrets-coco new file +51 -0
new file mode 100644
index 000000000000..f2b6909155f9
@@ -0,0 +1,51 @@
+What:		security/secrets/coco
+Date:		February 2022
+Contact:	Dov Murik <[email protected]>
+Description:
+		Exposes confidential computing (coco) EFI secrets to
+		userspace via securityfs.
+
+		EFI can declare memory area used by confidential computing
+		platforms (such as AMD SEV and SEV-ES) for secret injection by
+		the Guest Owner during VM's launch.  The secrets are encrypted
+		by the Guest Owner and decrypted inside the trusted enclave,
+		and therefore are not readable by the untrusted host.
+
+		The efi_secret module exposes the secrets to userspace.  Each
+		secret appears as a file under <securityfs>/secrets/coco,
+		where the filename is the GUID of the entry in the secrets
+		table.  This module is loaded automatically by the EFI driver
+		if the EFI secret area is populated.
+
+		Two operations are supported for the files: read and unlink.
+		Reading the file returns the content of secret entry.
+		Unlinking the file overwrites the secret data with zeroes and
+		removes the entry from the filesystem.  A secret cannot be read
+		after it has been unlinked.
+
+		For example, listing the available secrets::
+
+		  # modprobe efi_secret
+		  # ls -l /sys/kernel/security/secrets/coco
+		  -r--r----- 1 root root 0 Jun 28 11:54 736870e5-84f0-4973-92ec-06879ce3da0b
+		  -r--r----- 1 root root 0 Jun 28 11:54 83c83f7f-1356-4975-8b7e-d3a0b54312c6
+		  -r--r----- 1 root root 0 Jun 28 11:54 9553f55d-3da2-43ee-ab5d-ff17f78864d2
+		  -r--r----- 1 root root 0 Jun 28 11:54 e6f5a162-d67f-4750-a67c-5d065f2a9910
+
+		Reading the secret data by reading a file::
+
+		  # cat /sys/kernel/security/secrets/coco/e6f5a162-d67f-4750-a67c-5d065f2a9910
+		  the-content-of-the-secret-data
+
+		Wiping a secret by unlinking a file::
+
+		  # rm /sys/kernel/security/secrets/coco/e6f5a162-d67f-4750-a67c-5d065f2a9910
+		  # ls -l /sys/kernel/security/secrets/coco
+		  -r--r----- 1 root root 0 Jun 28 11:54 736870e5-84f0-4973-92ec-06879ce3da0b
+		  -r--r----- 1 root root 0 Jun 28 11:54 83c83f7f-1356-4975-8b7e-d3a0b54312c6
+		  -r--r----- 1 root root 0 Jun 28 11:54 9553f55d-3da2-43ee-ab5d-ff17f78864d2
+
+		Note: The binary format of the secrets table injected by the
+		Guest Owner is described in
+		drivers/virt/coco/efi_secret/efi_secret.c under "Structure of
+		the EFI secret area".
drivers/virt/Kconfig +3 -0
index 8061e8ef449f..fe7a6579b974 100644
@@ -36,4 +36,7 @@ source "drivers/virt/vboxguest/Kconfig"
 source "drivers/virt/nitro_enclaves/Kconfig"
 
 source "drivers/virt/acrn/Kconfig"
+
+source "drivers/virt/coco/efi_secret/Kconfig"
+
 endif
drivers/virt/Makefile +1 -0
index 3e272ea60cd9..efdb015783f9 100644
@@ -8,3 +8,4 @@ obj-y				+= vboxguest/
 
 obj-$(CONFIG_NITRO_ENCLAVES)	+= nitro_enclaves/
 obj-$(CONFIG_ACRN_HSM)		+= acrn/
+obj-$(CONFIG_EFI_SECRET)	+= coco/efi_secret/
drivers/virt/coco/efi_secret/Kconfig new file +16 -0
new file mode 100644
index 000000000000..4404d198f3b2
@@ -0,0 +1,16 @@
+# SPDX-License-Identifier: GPL-2.0-only
+config EFI_SECRET
+	tristate "EFI secret area securityfs support"
+	depends on EFI && X86_64
+	select EFI_COCO_SECRET
+	select SECURITYFS
+	help
+	  This is a driver for accessing the EFI secret area via securityfs.
+	  The EFI secret area is a memory area designated by the firmware for
+	  confidential computing secret injection (for example for AMD SEV
+	  guests).  The driver exposes the secrets as files in
+	  <securityfs>/secrets/coco.  Files can be read and deleted (deleting
+	  a file wipes the secret from memory).
+
+	  To compile this driver as a module, choose M here.
+	  The module will be called efi_secret.
drivers/virt/coco/efi_secret/Makefile new file +2 -0
new file mode 100644
index 000000000000..c7047ce804f7
@@ -0,0 +1,2 @@
+# SPDX-License-Identifier: GPL-2.0-only
+obj-$(CONFIG_EFI_SECRET) += efi_secret.o
drivers/virt/coco/efi_secret/efi_secret.c new file +349 -0

349 lines not on this page · raw message

Signature · 1 line
2.25.1
author
#3 lore raw

PATCH v10 3/4 efi: Register efi_secret platform device if EFI secret area is declared

In reply to Dov Murik

During efi initialization, check if coco_secret is defined in the EFI configuration table; in such case, register platform device "efi_secret". This allows udev to automatically load the efi_secret module (platform driver), which in turn will populate the <securityfs>/secrets/coco directory in guests into which secrets were injected.

Note that a declared address of an EFI secret area doesn't mean that secrets where indeed injected to that area; if the secret area is not populated, the driver will not load (but the platform device will still be registered).

---
 drivers/firmware/efi/efi.c | 5 +++++
 1 file changed, 5 insertions(+)
drivers/firmware/efi/efi.c +5 -0
index 378d044b2463..b8e68a121119 100644
@@ -425,6 +425,11 @@ static int __init efisubsys_init(void)
 	if (efi_enabled(EFI_DBG) && efi_enabled(EFI_PRESERVE_BS_REGIONS))
 		efi_debugfs_init();
 
+#ifdef CONFIG_EFI_COCO_SECRET
+	if (efi.coco_secret != EFI_INVALID_TABLE_ADDR)
+		platform_device_register_simple("efi_secret", 0, NULL, 0);
+#endif
+
 	return 0;
 
 err_remove_group:
Signature · 1 line
2.25.1
author
#4 lore raw

PATCH v10 4/4 docs: security: Add secrets/coco documentation

In reply to Dov Murik

Add documentation for the efi_secret module which allows access to Confidential Computing injected secrets.

---
 Documentation/security/index.rst         |   1 +
 Documentation/security/secrets/coco.rst  | 103 ++++++++++++++++++++
 Documentation/security/secrets/index.rst |   9 ++
 3 files changed, 113 insertions(+)
Documentation/security/index.rst +1 -0
index 16335de04e8c..6ed8d2fa6f9e 100644
@@ -17,3 +17,4 @@ Security Documentation
    tpm/index
    digsig
    landlock
+   secrets/index
Documentation/security/secrets/coco.rst new file +103 -0

103 lines not on this page · raw message

Documentation/security/secrets/index.rst new file +9 -0
new file mode 100644
index 000000000000..ced34e9c43bd
@@ -0,0 +1,9 @@
+.. SPDX-License-Identifier: GPL-2.0
+
+=====================
+Secrets documentation
+=====================
+
+.. toctree::
+
+   coco
Signature · 1 line
2.25.1
maintainer
#5 lore raw
In reply to Dov Murik

(cc Jon for the Documentation/ changes)

Quoted 86 lines · Dov Murik

Confidential computing (coco) hardware such as AMD SEV (Secure Encrypted Virtualization) allows guest owners to inject secrets into the VMs memory without the host/hypervisor being able to read them. In SEV, secret injection is performed early in the VM launch process, before the guest starts running.

OVMF already reserves designated area for secret injection (in its AmdSev package; see edk2 commit 01726b6d23d4 "OvmfPkg/AmdSev: Expose the Sev Secret area using a configuration table" [1]), but the secrets were not available in the guest kernel.

The patch series keeps the address of the EFI-provided memory for injected secrets, and exposes the secrets to userspace via securityfs using a new efi_secret kernel module. The module is autoloaded (by the EFI driver) if the secret area is populated.

The first patch in EFI keeps the address of the secret area as passed in the EFI configuration table. The second patch introduces the new efi_secret module that exposes the content of the secret entries as securityfs files, and allows clearing out secrets with a file unlink interface. The third patch auto-loads the efi_secret module during startup if the injected secrets area is populated. The last patch documents the data flow of confidential computing secret injection.

As a usage example, consider a guest performing computations on encrypted files. The Guest Owner provides the decryption key (= secret) using the secret injection mechanism. The guest application reads the secret from the efi_secret filesystem and proceeds to decrypt the files into memory and then performs the needed computations on the content.

In this example, the host can't read the files from the disk image because they are encrypted. Host can't read the decryption key because it is passed using the secret injection mechanism (= secure channel). Host can't read the decrypted content from memory because it's a confidential (memory-encrypted) guest.

This has been tested with AMD SEV and SEV-ES guests, but the kernel side of handling the secret area has no SEV-specific dependencies, and therefore might be usable (perhaps with minor changes) for any confidential computing hardware that can publish the secret area via the standard EFI config table entry.

To enable this functionality, set CONFIG_EFI_SECRET=m when building the guest kernel.

Here is a simple example for usage of the efi_secret module in a guest to which an EFI secret area with 4 secrets was injected during launch:

...

[    0.000000] efi: EFI v2.70 by EDK II
[    0.000000] efi: CocoSecret=0x7f222680 SMBIOS=0x7f541000 ACPI=0x7f77e000 ACPI 2.0=0x7f77e014 MEMATTR=0x7ea16418
...
[    1.127627] Run /init as init process
Loading, please wait...
Starting version 245.4-4ubuntu3.15
...
[    0.763204] efi_secret efi_secret.0: Created 4 entries in securityfs secrets/coco

...

# ls -la /sys/kernel/security/secrets/coco
total 0
drwxr-xr-x 2 root root 0 Jun 28 11:54 .
drwxr-xr-x 3 root root 0 Jun 28 11:54 ..
(quoted diff) +0 -4
-r--r----- 1 root root 0 Jun 28 11:54 736870e5-84f0-4973-92ec-06879ce3da0b
-r--r----- 1 root root 0 Jun 28 11:54 83c83f7f-1356-4975-8b7e-d3a0b54312c6
-r--r----- 1 root root 0 Jun 28 11:54 9553f55d-3da2-43ee-ab5d-ff17f78864d2
-r--r----- 1 root root 0 Jun 28 11:54 e6f5a162-d67f-4750-a67c-5d065f2a9910
# hd /sys/kernel/security/secrets/coco/e6f5a162-d67f-4750-a67c-5d065f2a9910
00000000  74 68 65 73 65 2d 61 72  65 2d 74 68 65 2d 6b 61  |these-are-the-ka|
00000010  74 61 2d 73 65 63 72 65  74 73 00 01 02 03 04 05  |ta-secrets......|
00000020  06 07                                             |..|
00000022
# rm /sys/kernel/security/secrets/coco/e6f5a162-d67f-4750-a67c-5d065f2a9910
# ls -la /sys/kernel/security/secrets/coco
total 0
drwxr-xr-x 2 root root 0 Jun 28 11:55 .
drwxr-xr-x 3 root root 0 Jun 28 11:54 ..
(quoted diff) +0 -3
-r--r----- 1 root root 0 Jun 28 11:54 736870e5-84f0-4973-92ec-06879ce3da0b
-r--r----- 1 root root 0 Jun 28 11:54 83c83f7f-1356-4975-8b7e-d3a0b54312c6
-r--r----- 1 root root 0 Jun 28 11:54 9553f55d-3da2-43ee-ab5d-ff17f78864d2
Quoted 9 lines

[1] https://github.com/tianocore/edk2/commit/01726b6d23d4

---

v10 changes:
- Add #ifdef CONFIG_EFI_COCO_SECRET to platform device registration in patch 3.

I have queued this up in efi/next. Jon, is that ok with you?

Thanks all,