摘要:
A system, computer program product and method for booting to a partition in a non-volatile storage unit without a local operator. In one embodiment, one or more bits in a BOOT register may be set by an operating system indicating if the BIOS should boot to the partition. The BIOS may then read the BOOT register to determine if the BIOS is to boot to the partition as well as any activities to perform if the BIOS is to boot to the partition. In another embodiment, a network interface card may insert directive information received from a packet in a register within the network interface card. The BIOS may then read the register within the network interface card to determine if the BIOS is to boot to the partition as well as any activities to perform if the BIOS is to boot to the partition.
摘要:
A method, system and computer readable medium containing programming instructions for tracking a secure boot in a computer system having a plurality of devices is disclosed. The method, system and computer readable medium include providing an embedded security system (ESS) in the computer system, wherein the ESS includes at least one boot platform configuration register (PCR) and a shadow PCR for each of the at least one boot PCRs, initiating a platform reset to boot the computer system via BIOS, and, for a device booted, generating a measurement value for the device and extending that value to one of the at least one boot PCRs and its corresponding shadow PCR. The system, method and computer readable medium of the present invention also includes comparing the measurement values of the boot PCRs to their corresponding shadow PCRs, whereby the computer system is trusted if the measurement values match.
摘要:
In a computer network including a plurality of interconnected computers, one of the computers being a sleeping computer in a power down state, the sleeping computer listening for a packet associated with the sleeping computer, a method and system of waking the sleeping computer from the computer network. An incoming packet of data is transmitted from one of the computers in the network to the sleeping computer. When the sleeping computer detects the incoming packet, it determines if the incoming packet contains a data sequence associated with the sleeping computer. Further, the sleeping computer compares a transit value in the incoming packet to a predetermined value stored at the sleeping computer. The transit value indicates how far the data packet has traveled through the network, indicating the approximate origin of the data packet. Knowing the approximate origin of the data packet allows the client system to identify if the data packet originated from an external network. The predetermined value represents an origin within the internal network. Accordingly, if the incoming packet matches the particular data sequence associated with the sleeping computer, and the transit value in the packet matches the predetermined value stored at the sleeping computer, then a signal is issued to wake the sleeping computer. Otherwise, the incoming packet is discarded and the sleeping computer is not awaken.
摘要:
A method, system, and program for selecting and implementing a basic input/output system (“BIOS”) configuration among various BIOS configurations for a data processing system are disclosed. Different BIOS configurations are defined for various types of users, such as a home user, a commercial user, and a network user. Each of the BIOS configurations includes a different set of BIOS characteristics, such as program setup features security features, and network server features, under which the data processing system is able to run. The different BIOS configurations are stored into a memory device for the data processing system. A designation is set within the memory device that directs a processor of the data processing system to select and execute a desired one of the BIOS configurations for a particular type of user.
摘要:
A method and system for enabling security attestation for a computing device during a return from an S4 sleep state. When the computing device enters into the S4 state following a successful boot up, the attestation log is appended to the TPM tick count and the log is signed (with a security signature). When the device is awaken from S4 state, the BIOS obtains and verifies the log created during the previous boot. The CRTM maintains a set of virtual PCRs and references these virtual PCRs against the log. If the values do not match, the return from S4 state fails and the device is rebooted.
摘要:
A system and method for access control of a hardfile responsive to a computer system having an operating system is disclosed. The method includes detecting a special boot condition during a pre-boot test of the computer system; and altering, in response to the special boot condition, an operating system access configuration of the hardfile. The system includes a computer system that adjusts an operating system access to a hardfile based upon various boot conditions.
摘要:
During power up initialization, security data such as passwords and other sensitive data which are stored in a lockable memory device are read and copied to protected system management interrupt (SMI) memory space, subject to verification by code running in the SMI memory space that the call to write the security data originates with a trusted entity. Once copied to SMI memory space, the security data is erased from regular system memory and the lockable storage device is hard locked (requiring a reset to unlock) against direct access prior to starting the operating system. The copy of the security data within the SMI memory space is invisible to the operating system. However, the operating system may initiate a call to code running in the SMI memory space to check a password entered by the user, with the SMI code returning a “match” or “no match” indication. The security data may thus be employed after the lockable memory device is hard locked and the operating system is started.
摘要:
In a computer network including a plurality of interconnected computers, one of the computers being a sleeping computer in a power down state, the sleeping computer listening for a packet associated with the sleeping computer, a method of waking the sleeping computer from the computer network. An incoming packet of data is transmitted from an administration system in the network to the sleeping computer. When the sleeping computer detects the incoming packet, it determines if the incoming packet contains a data sequence associated with the sleeping computer. If the incoming packet matches the particular data sequence associated with the sleeping computer, the sleeping computer transmits a reply message to the administration system. Upon receiving the reply, the administration system modifies the reply message in a predetermined manner and transmits the modified reply to the sleeping computer. If the sleeping computer determines the reply message was modified in the predetermined manner, then a signal is issued to wake the sleeping computer. Otherwise, the incoming packet is discarded and the sleeping computer is not awakened.
摘要:
A computer system processor incorporates a special S-latch which may only be set by secure signals. One state of the S-latch sets the processor into a secure mode where it only executes instructions and not commands from an In Circuit Emulator (ICE) unit. A second state of the S-latch sets the processor into a non-secure mode. A non-volatile random access memory (NVRAM) is written with secure data which can only be read by boot block code stored in a BIOS storage device. The boot block code is operable to read the secure data in the NVRAM and set the S-latch to an appropriate security state. If the boot block code cannot set the S-latch, then remaining boot up with BIOS data is stopped. On boot up the boot block code reads the NVRAM and sets the S-latch into the appropriate security state.
摘要:
A periodic system “wake-up” is implemented during S1, S2 or S3 states utilizing a hardware timer. A memory scrubbing routine is initiated that reads out all memory locations and writes back any memory locations that have single bit (correctable) Error Correction Code errors. This procedure minimizes the chances of a multiple bit error build up over time that may cause an unrecoverable error. The scrubbing routine is invoked whenever the system is brought out of S1, S2, or S3 state to insure that there are no single bit errors present when full system operation is resumed.