mirror of
https://mirrors.bfsu.edu.cn/git/linux.git
synced 2024-12-21 10:05:00 +08:00
c82c618650
The whole point of SGX is to create a hardware protected place to do “stuff”. But, before someone is willing to hand over the keys to the castle , an enclave must often prove that it is running on an SGX-protected processor. Provisioning enclaves play a key role in providing proof. There are actually three different enclaves in play in order to make this happen: 1. The application enclave. The familiar one we know and love that runs the actual code that’s doing real work. There can be many of these on a single system, or even in a single application. 2. The quoting enclave (QE). The QE is mentioned in lots of silly whitepapers, but, for the purposes of kernel enabling, just pretend they do not exist. 3. The provisioning enclave. There is typically only one of these enclaves per system. Provisioning enclaves have access to a special hardware key. They can use this key to help to generate certificates which serve as proof that enclaves are running on trusted SGX hardware. These certificates can be passed around without revealing the special key. Any user who can create a provisioning enclave can access the processor-unique Provisioning Certificate Key which has privacy and fingerprinting implications. Even if a user is permitted to create normal application enclaves (via /dev/sgx_enclave), they should not be able to create provisioning enclaves. That means a separate permissions scheme is needed to control provisioning enclave privileges. Implement a separate device file (/dev/sgx_provision) which allows creating provisioning enclaves. This device will typically have more strict permissions than the plain enclave device. The actual device “driver” is an empty stub. Open file descriptors for this device will represent a token which allows provisioning enclave duty. This file descriptor can be passed around and ultimately given as an argument to the /dev/sgx_enclave driver ioctl(). [ bp: Touchups. ] Suggested-by: Andy Lutomirski <luto@kernel.org> Signed-off-by: Jarkko Sakkinen <jarkko@kernel.org> Signed-off-by: Borislav Petkov <bp@suse.de> Cc: linux-security-module@vger.kernel.org Link: https://lkml.kernel.org/r/20201112220135.165028-16-jarkko@kernel.org
30 lines
739 B
C
30 lines
739 B
C
/* SPDX-License-Identifier: GPL-2.0 */
|
|
#ifndef __ARCH_SGX_DRIVER_H__
|
|
#define __ARCH_SGX_DRIVER_H__
|
|
|
|
#include <crypto/hash.h>
|
|
#include <linux/kref.h>
|
|
#include <linux/mmu_notifier.h>
|
|
#include <linux/radix-tree.h>
|
|
#include <linux/rwsem.h>
|
|
#include <linux/sched.h>
|
|
#include <linux/workqueue.h>
|
|
#include <uapi/asm/sgx.h>
|
|
#include "sgx.h"
|
|
|
|
#define SGX_EINIT_SPIN_COUNT 20
|
|
#define SGX_EINIT_SLEEP_COUNT 50
|
|
#define SGX_EINIT_SLEEP_TIME 20
|
|
|
|
extern u64 sgx_attributes_reserved_mask;
|
|
extern u64 sgx_xfrm_reserved_mask;
|
|
extern u32 sgx_misc_reserved_mask;
|
|
|
|
extern const struct file_operations sgx_provision_fops;
|
|
|
|
long sgx_ioctl(struct file *filep, unsigned int cmd, unsigned long arg);
|
|
|
|
int sgx_drv_init(void);
|
|
|
|
#endif /* __ARCH_X86_SGX_DRIVER_H__ */
|