摘要:
Embodiments of techniques and systems for out-of-band verification of host OS components are described. In embodiments, a out-of-band host OS boot sequence verification system (“BSVS”) may access system memory without detection by a host OS process, or “out of band.” The BSVS may access host OS components in the system memory and may generate signatures from memory footprints of the host OS components. These signatures may then be compared to trusted signatures to verify integrity of the host OS components. In embodiments, this verification may be performed during a boot of a host OS or on demand. In embodiments, the trusted signatures may be pre-stored by the BSVS before a boot; in some embodiments, the trusted signatures may be previously-computed and then stored by the BSVS. Other embodiments may be described and claimed.
摘要:
Embodiments of techniques and systems for out-of-band verification of host OS components are described. In embodiments, a out-of-band host OS boot sequence verification system (“BSVS”) may access system memory without detection by a host OS process, or “out of band.” The BSVS may access host OS components in the system memory and may generate signatures from memory footprints of the host OS components. These signatures may then be compared to trusted signatures to verify integrity of the host OS components. In embodiments, this verification may be performed during a boot of a host OS or on demand. In embodiments, the trusted signatures may be pre-stored by the BSVS before a boot; in some embodiments, the trusted signatures may be previously-computed and then stored by the BSVS. Other embodiments may be described and claimed.
摘要:
An embodiment may include a storage processor that may be comprised, at least in part, in a host. The host may include at least one host central processing unit (CPU) to execute at least one host operating system (OS). The storage processor may execute at least one operation in isolation from interference from and control by the at least one host CPU and the at least one host OS. The at least one operation may facilitate, at least in part: (1) prevention, at least in part, of unauthorized access to storage, (2) prevention, at least in part, of execution by the at least one host CPU of at least one unauthorized instruction, (3) detection, at least in part, of the at least one unauthorized instruction, and/or (4) remediation, at least in part, of at least one condition associated, at least in part, with the at least unauthorized instruction.
摘要:
A modular BIOS update mechanism provides a standardized method to update options ROMs and to provide video and processor microcode upgrades in a computer system without requiring a complete replacement of the system BIOS. The MBU mechanism provides several advantages. First, new features and BIOS bugs from earlier release may be delivered to an installed base of end-user systems even if direct OEM support cannot be identified. Also, BIOS components may be provided as a validated set of revisions. With resort to a validation matrix, BIOS updates may be managed easily. The modular BIOS update is particularly useful in systems having several independent BIOS's stored within unitary firmware.
摘要:
Methods of reliably allocating, writing, reading, de-allocating, re-allocating, and reclaiming space within a nonvolatile memory having a bifurcated storage architecture are described. Allocation, writing, reading, de-allocating, re-allocating, and reclamation are handled by a memory manager. The memory manager tracks the progress of each process during execution in order to detect whether a selected process was interrupted for purposes of recovery. The nonvolatile memory is recovered to a known state during initialization. Initialization includes the step of determining a recovery state from a recovery state lookup table. A selected recovery process is selected in accordance with the recovery state lookup table. A restart level for the selected process is determined from a corresponding restart state lookup table. The selected process is then restarted at the restart level. In one embodiment, a method of managing a nonvolatile memory includes the step of identifying an interrupted process from at least one of an allocation, a reclamation, a configuration header reclaim, and a re-allocation process initiated on the nonvolatile memory. A recovery process is selected for the interrupted process. An entry point into the recovery process is determined. The selected recovery process is then restarted at the entry point.
摘要:
A system including a host coupled to a memory device and a peripheral controller device. The host is coupled to the peripheral controller device via a bus having a plurality of general purpose signal lines to carry time-multiplexed address, data, and control information. The peripheral controller device performs direct memory access (DMA) transactions with the memory device via the host and the bus.
摘要:
Methods of allocating, writing, reading, de-allocating, re-allocating, and reclaiming space within a nonvolatile memory having a bifurcated storage architecture are described. A method of reliably re-allocating a first object includes the step of storing a location of a first object in a first data structure. A location of the first data structure is stored in a second data structure. A duplicate of the first object is formed by initiating a copy of the first object. An erase of the first object is initiated. A write of a second object to the location of the first object is then initiated. The duplicate object is invalidated. The status of copying, erasing, and writing is tracked. The copy status, erase status, write status, and a restoration status are used to determine a recovery state upon initialization of the nonvolatile memory. The duplicate object is invalidated, if the writing status indicates that the writing of the second object has been completed. The first object is erased, if a restoration status indicates copying of the duplicate object was initiated but not completed. The erasing of the first object is completed, if the erase status indicates that erasure of the first object is not completed. A restoration of the duplicate object to the location of the first object is initiated, if the copying status indicates that copying of the first object was completed. The copying of the duplicate object is tracked as a restoration status.
摘要:
Methods of allocating, writing, reading, de-allocating, re-allocating, and reclaiming space within a nonvolatile memory having a bifurcated storage architecture are described. A method of reliably re-allocating a first object includes the step of storing a location of a first object in a first data structure. A location of the first data structure is stored in a second data structure. A duplicate of the first object is formed by initiating a copy of the first object. An erase of the first object is initiated. A write of a second object to the location of the first object is then initiated. The duplicate object is invalidated. The status of copying, erasing, and writing is tracked. The copy status, erase status, write status, and a restoration status are used to determine a recovery state upon initialization of the nonvolatile memory. The duplicate object is invalidated , if the writing status indicates that the writing of the second object has been completed. The first object is erased, if a restoration status indicates copying of the duplicate object was initiated but not completed. The erasing of the first object is completed, if the erase status indicates that erasure of the first object is not completed. A restoration of the duplicate object to the location of the first object is initiated, if the copying status indicates that copying of the first object was completed. The copying of the duplicate object is tracked as a restoration status.
摘要:
An arrangement for accessing a non-volatile memory array including providing a signal having a first condition if an access is a read and a second condition if an access is for any other operation; reading data directly from an address in the non-volatile memory array if the signal is a first condition; and performing any other access of the non-volatile memory array utilizing a command-centric interface if the signal is a second condition.
摘要:
A protected boot sequence in a computer system. A reset vector directs the system to a boot program including a protected program. This protected program verifies the integrity of the BIOS contents before branching to the BIOS for execution of normal bootstrap functions. The protected program can also lock down various blocks of bootstrap code to prevent them from being changed after a certain point in the boot sequence. The protected boot sequence can proceed in layers, with each layer providing some level of validation or security for succeeding layers.