[PATCH v3] integrity: support including firmware ".platform" keys at build time
Nayna
nayna at linux.vnet.ibm.com
Wed Nov 10 19:15:10 UTC 2021
On 11/5/21 10:05, Dimitri John Ledkov wrote:
> (Changing Seth's email to personal domain one)
>
> On 04/10/2021 15:52, Nayna Jain wrote:
>> Some firmware support secure boot by embedding static keys to verify the
>> Linux kernel during boot. However, these firmware do not expose an
>> interface for the kernel to load firmware keys onto ".platform" keyring.
>> This would prevent kernel signature verification on kexec.
>>
>> For these environments, a new function load_builtin_platform_cert() is
>> defined to load compiled in certificates onto the ".platform" keyring.
>>
>> load_certificate_list() is currently used for parsing compiled in
>> certificates to be loaded onto the .builtin or .blacklist keyrings.
>> Export load_certificate_list() allowing it to be used for parsing compiled
>> in ".platform" keyring certificates as well.
>>
>> Reported-by: kernel test robot <lkp at intel.com>(auto build test ERROR)
>> Signed-off-by: Nayna Jain <nayna at linux.ibm.com>
>> ---
>> NOTE: I am wondering if we should split this patch into two:
>> (https://lore.kernel.org/linux-integrity/be4bd13d-659d-710d-08b9-1a34a65e5c5d@linux.vnet.ibm.com/).
>> I can do so if you also prefer the same.
>>
> Yes, i think you should split the patches into refactor & new feature.
> It allows one to clearly inspect/validate the no-change refactor, and
> separately review the newly added feature with clarity.
>
> Or maybe avoid the refactor all together, see my optional suggestion at
> the end.
Thanks Dimitri for the feedback. Sorry for the delay in the response, I
was on vacation last week. Yes, I have split the patches now for next
version. Answers to other comments are inline.
>> v3:
>> * Included Jarkko's feedback
>> ** updated patch description to include approach.
>> ** removed extern for function declaration in the .h file.
>> * Included load_certificate_list() within #ifdef CONFIG_KEYS condition.
>>
>> v2:
>> * Fixed the error reported by kernel test robot
>> * Updated patch description based on Jarkko's feedback.
>>
>> certs/Makefile | 3 ++-
>> certs/blacklist.c | 1 -
>> certs/common.c | 2 +-
>> certs/common.h | 9 -------
>> certs/system_keyring.c | 1 -
>> include/keys/system_keyring.h | 6 +++++
>> security/integrity/Kconfig | 10 +++++++
>> security/integrity/Makefile | 17 +++++++++++-
>> security/integrity/digsig.c | 2 +-
>> security/integrity/integrity.h | 6 +++++
>> .../integrity/platform_certs/platform_cert.S | 23 ++++++++++++++++
>> .../platform_certs/platform_keyring.c | 26 +++++++++++++++++++
>> 12 files changed, 91 insertions(+), 15 deletions(-)
>> delete mode 100644 certs/common.h
>> create mode 100644 security/integrity/platform_certs/platform_cert.S
>>
>> diff --git a/certs/Makefile b/certs/Makefile
>> index 279433783b10..64ee37f38b85 100644
>> --- a/certs/Makefile
>> +++ b/certs/Makefile
>> @@ -3,7 +3,8 @@
>> # Makefile for the linux kernel signature checking certificates.
>> #
>>
>> -obj-$(CONFIG_SYSTEM_TRUSTED_KEYRING) += system_keyring.o system_certificates.o common.o
>> +obj-$(CONFIG_KEYS) += common.o
>> +obj-$(CONFIG_SYSTEM_TRUSTED_KEYRING) += system_keyring.o system_certificates.o
>> obj-$(CONFIG_SYSTEM_BLACKLIST_KEYRING) += blacklist.o common.o
>> obj-$(CONFIG_SYSTEM_REVOCATION_LIST) += revocation_certificates.o
>> ifneq ($(CONFIG_SYSTEM_BLACKLIST_HASH_LIST),"")
>> diff --git a/certs/blacklist.c b/certs/blacklist.c
>> index c9a435b15af4..b95e9b19c42f 100644
>> --- a/certs/blacklist.c
>> +++ b/certs/blacklist.c
>> @@ -17,7 +17,6 @@
>> #include <linux/uidgid.h>
>> #include <keys/system_keyring.h>
>> #include "blacklist.h"
>> -#include "common.h"
>>
>> static struct key *blacklist_keyring;
>>
>> diff --git a/certs/common.c b/certs/common.c
>> index 16a220887a53..41f763415a00 100644
>> --- a/certs/common.c
>> +++ b/certs/common.c
>> @@ -2,7 +2,7 @@
>>
>> #include <linux/kernel.h>
>> #include <linux/key.h>
>> -#include "common.h"
>> +#include <keys/system_keyring.h>
>>
>> int load_certificate_list(const u8 cert_list[],
>> const unsigned long list_size,
>> diff --git a/certs/common.h b/certs/common.h
>> deleted file mode 100644
>> index abdb5795936b..000000000000
>> --- a/certs/common.h
>> +++ /dev/null
>> @@ -1,9 +0,0 @@
>> -/* SPDX-License-Identifier: GPL-2.0-or-later */
>> -
>> -#ifndef _CERT_COMMON_H
>> -#define _CERT_COMMON_H
>> -
>> -int load_certificate_list(const u8 cert_list[], const unsigned long list_size,
>> - const struct key *keyring);
>> -
>> -#endif
>> diff --git a/certs/system_keyring.c b/certs/system_keyring.c
>> index 692365dee2bd..d130d5a96e09 100644
>> --- a/certs/system_keyring.c
>> +++ b/certs/system_keyring.c
>> @@ -16,7 +16,6 @@
>> #include <keys/asymmetric-type.h>
>> #include <keys/system_keyring.h>
>> #include <crypto/pkcs7.h>
>> -#include "common.h"
>>
>> static struct key *builtin_trusted_keys;
>> #ifdef CONFIG_SECONDARY_TRUSTED_KEYRING
>> diff --git a/include/keys/system_keyring.h b/include/keys/system_keyring.h
>> index 6acd3cf13a18..d3f914d9a632 100644
>> --- a/include/keys/system_keyring.h
>> +++ b/include/keys/system_keyring.h
>> @@ -10,6 +10,12 @@
>>
>> #include <linux/key.h>
>>
>> +#ifdef CONFIG_KEYS
>> +int load_certificate_list(const u8 cert_list[],
>> + const unsigned long list_size,
>> + const struct key *keyring);
>> +#endif
>> +
>> #ifdef CONFIG_SYSTEM_TRUSTED_KEYRING
>>
>> extern int restrict_link_by_builtin_trusted(struct key *keyring,
>> diff --git a/security/integrity/Kconfig b/security/integrity/Kconfig
>> index 71f0177e8716..b2009b792882 100644
>> --- a/security/integrity/Kconfig
>> +++ b/security/integrity/Kconfig
>> @@ -62,6 +62,16 @@ config INTEGRITY_PLATFORM_KEYRING
>> provided by the platform for verifying the kexec'ed kerned image
>> and, possibly, the initramfs signature.
>>
>> +config INTEGRITY_PLATFORM_BUILTIN_KEYS
>> + string "Builtin X.509 keys for .platform keyring"
>> + depends on KEYS
>> + depends on ASYMMETRIC_KEY_TYPE
>> + depends on INTEGRITY_PLATFORM_KEYRING
>> + help
>> + If set, this option should be the filename of a PEM-formatted file
>> + containing X.509 certificates to be loaded onto the ".platform"
>> + keyring.
>> +
>> config LOAD_UEFI_KEYS
>> depends on INTEGRITY_PLATFORM_KEYRING
>> depends on EFI
>> diff --git a/security/integrity/Makefile b/security/integrity/Makefile
>> index 7ee39d66cf16..a45f083589b8 100644
>> --- a/security/integrity/Makefile
>> +++ b/security/integrity/Makefile
>> @@ -3,13 +3,18 @@
>> # Makefile for caching inode integrity data (iint)
>> #
>>
>> +quiet_cmd_extract_certs = EXTRACT_CERTS $(patsubst "%",%,$(2))
>> + cmd_extract_certs = scripts/extract-cert $(2) $@
>> +$(eval $(call config_filename,INTEGRITY_PLATFORM_BUILTIN_KEYS))
>> +
>> obj-$(CONFIG_INTEGRITY) += integrity.o
>>
>> integrity-y := iint.o
>> integrity-$(CONFIG_INTEGRITY_AUDIT) += integrity_audit.o
>> integrity-$(CONFIG_INTEGRITY_SIGNATURE) += digsig.o
>> integrity-$(CONFIG_INTEGRITY_ASYMMETRIC_KEYS) += digsig_asymmetric.o
>> -integrity-$(CONFIG_INTEGRITY_PLATFORM_KEYRING) += platform_certs/platform_keyring.o
>> +integrity-$(CONFIG_INTEGRITY_PLATFORM_KEYRING) += platform_certs/platform_keyring.o \
>> + platform_certs/platform_cert.o
>> integrity-$(CONFIG_LOAD_UEFI_KEYS) += platform_certs/efi_parser.o \
>> platform_certs/load_uefi.o \
>> platform_certs/keyring_handler.o
>> @@ -19,3 +24,13 @@ integrity-$(CONFIG_LOAD_PPC_KEYS) += platform_certs/efi_parser.o \
>> platform_certs/keyring_handler.o
>> obj-$(CONFIG_IMA) += ima/
>> obj-$(CONFIG_EVM) += evm/
>> +
>> +
>> +$(obj)/platform_certs/platform_cert.o: $(obj)/platform_certs/platform_certificate_list
>> +
>> +targets += platform_certificate_list
>> +
>> +$(obj)/platform_certs/platform_certificate_list: scripts/extract-cert $(INTEGRITY_PLATFORM_BUILTIN_KEYS_FILENAME) FORCE
>> + $(call if_changed,extract_certs,$(CONFIG_INTEGRITY_PLATFORM_BUILTIN_KEYS))
>> +
>> +clean-files := platform_certs/platform_certificate_list
>> diff --git a/security/integrity/digsig.c b/security/integrity/digsig.c
>> index 3b06a01bd0fd..0ea40ed8dfcb 100644
>> --- a/security/integrity/digsig.c
>> +++ b/security/integrity/digsig.c
>> @@ -38,7 +38,7 @@ static const char * const keyring_name[INTEGRITY_KEYRING_MAX] = {
>> #define restrict_link_to_ima restrict_link_by_builtin_trusted
>> #endif
>>
>> -static struct key *integrity_keyring_from_id(const unsigned int id)
>> +struct key *integrity_keyring_from_id(const unsigned int id)
>> {
>> if (id >= INTEGRITY_KEYRING_MAX)
>> return ERR_PTR(-EINVAL);
>> diff --git a/security/integrity/integrity.h b/security/integrity/integrity.h
>> index 547425c20e11..feb84e1b1105 100644
>> --- a/security/integrity/integrity.h
>> +++ b/security/integrity/integrity.h
>> @@ -167,6 +167,7 @@ int __init integrity_init_keyring(const unsigned int id);
>> int __init integrity_load_x509(const unsigned int id, const char *path);
>> int __init integrity_load_cert(const unsigned int id, const char *source,
>> const void *data, size_t len, key_perm_t perm);
>> +struct key *integrity_keyring_from_id(const unsigned int id);
>> #else
>>
>> static inline int integrity_digsig_verify(const unsigned int id,
>> @@ -194,6 +195,11 @@ static inline int __init integrity_load_cert(const unsigned int id,
>> {
>> return 0;
>> }
>> +
>> +static inline struct key *integrity_keyring_from_id(const unsigned int id)
>> +{
>> + return ERR_PTR(-EOPNOTSUPP);
>> +}
>> #endif /* CONFIG_INTEGRITY_SIGNATURE */
>>
>> #ifdef CONFIG_INTEGRITY_ASYMMETRIC_KEYS
>> diff --git a/security/integrity/platform_certs/platform_cert.S b/security/integrity/platform_certs/platform_cert.S
>> new file mode 100644
>> index 000000000000..20bccce5dc5a
>> --- /dev/null
>> +++ b/security/integrity/platform_certs/platform_cert.S
>> @@ -0,0 +1,23 @@
>> +/* SPDX-License-Identifier: GPL-2.0 */
> My personal preference would be to have platform_cert.S next to all the
> other "extract certs and build an object out of them", as in
> /certs/platform_cert.S right next to the other similar .S files, i.e
> system_certificates.S and revocation_certificates.S it would make it
> easier to inspect the build and find all the various types of built-in keys.
>
> This is a minor comment/suggestion, that you can reject.
I think anything platform related should be in the platform_certs/
directory.
>
>> +#include <linux/export.h>
>> +#include <linux/init.h>
>> +
>> + __INITRODATA
>> +
>> + .align 8
>> +#ifdef CONFIG_INTEGRITY_PLATFORM_KEYRING
>> + .globl platform_certificate_list
>> +platform_certificate_list:
>> +__cert_list_start:
>> + .incbin "security/integrity/platform_certs/platform_certificate_list"
>> +__cert_list_end:
>> +#endif
>> +
>> + .align 8
>> + .globl platform_certificate_list_size
>> +platform_certificate_list_size:
>> +#ifdef CONFIG_64BIT
>> + .quad __cert_list_end - __cert_list_start
>> +#else
>> + .long __cert_list_end - __cert_list_start
>> +#endif
>> diff --git a/security/integrity/platform_certs/platform_keyring.c b/security/integrity/platform_certs/platform_keyring.c
>> index bcafd7387729..17535050d08d 100644
>> --- a/security/integrity/platform_certs/platform_keyring.c
>> +++ b/security/integrity/platform_certs/platform_keyring.c
>> @@ -12,8 +12,12 @@
>> #include <linux/cred.h>
>> #include <linux/err.h>
>> #include <linux/slab.h>
>> +#include <keys/system_keyring.h>
>> #include "../integrity.h"
>>
>> +extern __initconst const u8 platform_certificate_list[];
>> +extern __initconst const unsigned long platform_certificate_list_size;
>> +
>> /**
>> * add_to_platform_keyring - Add to platform keyring without validation.
>> * @source: Source of key
>> @@ -37,6 +41,28 @@ void __init add_to_platform_keyring(const char *source, const void *data,
>> pr_info("Error adding keys to platform keyring %s\n", source);
>> }
>>
> I think the below hunk should be guarded with #ifdef
> CONFIG_INTEGRITY_PLATFORM_BUILTIN_KEYS, similar to how similar call is
> ifdef'ed in case of revocation keys i.e. see ifdef
> CONFIG_SYSTEM_REVOCATION_LIST in certs/blacklist.c. Because one can
> configure a kernel with a platform keyring, but without platform builtin
> keys.
>
> I think this must be be fixed.
If there are no builtin keys configured, then size would be zero and
load_certificate_list would just return. I am not sure why it had to be
ifdef'ed for revocation, I do not see the same for loading system
certificates onto system keyring. Please let me know if you think I am
missing something.
>
>> +static __init int load_builtin_platform_cert(void)
>> +{
>> + const u8 *p;
>> + unsigned long size;
>> + int rc;
>> + struct key *keyring;
>> +
>> + p = platform_certificate_list;
>> + size = platform_certificate_list_size;
>> +
>> + keyring = integrity_keyring_from_id(INTEGRITY_KEYRING_PLATFORM);
>> + if (IS_ERR(keyring))
>> + return PTR_ERR(keyring);
>> +
>> + rc = load_certificate_list(p, size, keyring);
>> + if (rc)
>> + pr_info("Error adding keys to platform keyring %d\n", rc);
>> +
>> + return rc;
>> +}
>> +late_initcall(load_builtin_platform_cert);
>> +
>
> Some more general comment:
> I do wonder if the refactor of load_certificate_list can be somehow
> avoided. For example, when platform keyring is allocated it calls
> set_platform_trusted_keys() from certs/system-keyring.c which could add
> a call to load_builtin_platform_cert, and it already has a handle to the
> platform keyring. The set_platform_trusted_keys() could call to
> load_certificate_list there without late_initcall, alocating variables,
> and looking up the integrity_keyring_from id. It would make this feature
> patch very concise and straight forward then.
>
> That way builtin platform certs will be done before platform specific
> load_*.c things are done. But also it will make builtin certs loading be
> done in a different place from the rest of the platform certs loading
> (e.g. the uefi one). So maybe doing builtin certs loading, next to all
> other platform cert loading makes sense, despite it being away from
> other builtin certs loading into other keyrings.
>
> I am divided on this =)
Yes, I had also thought about it. I saw that set_platform_trusted_keys()
call is dependent on CONFIG_SYSTEM_TRUSTED_KEYRING. By design, it should
be possible to load platform certs onto the platform keyring even when
the system keyring is not enabled. So, I preferred to keep it separate
and in platform_keyring.c.
Thanks & Regards,
- Nayna
More information about the Linux-security-module-archive
mailing list