International Business Machines Corporation

Paul Mackerras

Benjamin Herrenschmidt

Michael K Gschwind

Bradly George Frey

Jentje Leenstra

Guerney D H Hunt

William Eric Hall

Anthony J Bybell

William John Starke

Jeffrey A Stuecheli

Ronald Nick Kalla

Giles R Frazier

Richard Harold Boivie

Albert James Van Norstrand, Jr

Cathy May

Andrew Henry Wottreng

Timothy Hume Heil

David A Campbell

Jon K Kriegel

John Martin Ludden

Paul F Russell

The present disclosure relates to a process-based virtualization system comprising a data processing unit. The system comprises a computer readable storage media, wherein a first memory component of the computer readable storage media is configured for access by an OS, secure and non-secure applications and the firmware, and wherein a second memory component of the computer readable storage media is configured for access by the firmware and not by the OS and the non-secure application. The data processing unit is configured to operate in a first mode of operation that executes a non-secure application process using the OS, and to operate in a second mode of operation that executes the secure application using the firmware, thereby executing application code using the second memory component.

Process-based virtualization system for executing a secure application process

A system, a method, and a computer program product for secure memory implementation for secure execution of virtual machines are provided. Data is processed in a first mode and a second mode, and commands are sent to a chip interconnect bus using real addresses, wherein the chip interconnect bus transports a number of bits for the real addresses. A memory controller is operatively coupled to a memory component. A secure memory range is specified by using range registers. If the real address is detected to be in the secure memory range to match a memory component address, a real address bit is set. If the real address is in the memory address hole, a security access violation is detected. If the real address is not in the secure address range and the real address bit is set, the security access violation is detected.

Secure memory implementation for secure execution of virtual machines

Addressability of instructions and the addressing of data ranges are extended. One or more operands obtained from fields explicitly specified by an instruction are overridden (i.e., ignored), and instead, an address based on the instruction (e.g., an instruction address) is substituted for the one or more operands. This provides an address having more bits than allowed by the operand being overridden, thereby extending addressability of the instruction and extended data range addressing. Further, in one aspect, additional bits may be added to one or more immediate fields of the instruction, thereby extending addressability of the instructions and extending data range addressing.

Extending data range addressing

A system, a method, and a computer program product for secure memory implementation for secure execution of virtual machines. Data is processed in a first mode and a second mode, and commands are sent to a chip interconnect bus using real addresses, wherein the chip interconnect bus transports a number of bits for the real addresses. A memory controller is operatively coupled to a memory component. A secure memory range is specified by using range registers. If the real address is detected to be in the secure memory range to match a memory component address, a real address bit is set. If the real address is in the memory address hole, a security access violation is detected. If the real address is not in the secure address range and the real address bit is set, the security access violation is detected.

Secure memory implementation for secure execution of Virtual Machines

Hardware based isolation for secure execution of virtual machines (VMs). At least one virtual machine is executed via operation of a hypervisor and an ultravisor. A first memory component is configured for access by the hypervisor and the ultravisor, and a second memory component is configured for access by the ultravisor and not by the hypervisor. A first mode of operation is operated, such that the virtual machine is executed using the hypervisor, wherein the first memory component is accessible to the virtual machine and the second memory component is not accessible to the virtual machine. A second mode of operation is operated, such that the virtual machine is executed using the ultravisor, wherein the first memory component and the second memory component are accessible to the virtual machine, thereby executing application code and operating system code using the second memory component without code changes.

Hardware based isolation for secure execution of virtual machines

An embodiment involves secure memory implementation for secure execution of virtual machines. Data is processed in a first mode and a second mode, and commands are sent to a chip interconnect bus using real addresses, wherein the chip interconnect bus includes a number of bits for the real addresses. A memory controller is operatively coupled to a memory component. A secure memory range is specified by using range registers. If the real address is detected to be in the secure memory range to match a memory component address, a real address bit is set. If the real address is in the memory address hole, a security access violation is detected. If the real address is not in the secure address range and the real address bit is set, the security access violation is detected.

An approach is provided to dynamically select a micro-threading (MT) mode of each core of a processor based on a load on each of the respective cores while the processor is running a hypervisor. The approach sets a core's micro-threading mode to a whole-core mode (MT1) in response to identifying that the load on the selected core is at a light load level, sets the core's micro-threading mode to a two-way micro-threading mode (MT2) in response to identifying that the load on the selected core has increased above the light load level, and sets the selected core's micro-threading mode to a four-way micro-threading mode (MT4) in response to identifying that the load on the selected core is at a high load level.

Selecting processor micro-threading mode

An approach is provided to dynamically select a micro-threading (MT) mode of each core of a processor based on a load on each of the respective cores while the processor is running a hypervisor. The approach sets a core's micro-threading mode to a whole-core mode (MT) in response to identifying that the load on the selected core is at a light load level, sets the core's micro-threading mode to a two-way micro-threading mode (MT) in response to identifying that the load on the selected core has increased above the light load level, and sets the selected core's micro-threading mode to a four-way micro-threading mode (MT) in response to identifying that the load on the selected core is at a high load level.

For a current context in control of a processor requesting access to a particular address, a translation lookaside buffer (TLB) controller specifies a virtual address with a logical partition identifier value indicating a privilege setting of the current context, a process identifier value indicating whether the address is within shared address space, and an effective address comprising at least a portion of the particular address. In response to the virtual address not matching at least one entry within a TLB comprising at least one entry stored for at least one previous translation of at least one previous address, the TLB controller translates the virtual address into a real page number using at least one page table and adding a new entry to the TLB with the virtual address and the real page number, wherein each at least one entry within the TLB identifies a separate privilege setting from among a plurality of privilege settings and a separate indicator of whether the address is within the shared address space.

Managing translations across multiple contexts using a TLB with entries directed to multiple privilege levels and to multiple types of address spaces

In response to a current context, with a particular process currently in control of a processor requesting access to a shared address space, a translation lookaside buffer (TLB) controller sets a process identifier field in a virtual address to be looked up in a TLB to a clamped value different from an identifier for the process, wherein the virtual address comprises at least the process identifier field and an effective address field set to an address in the requested shared address space. In response to the TLB controller comparing the virtual address for the current context to a particular entry of at least one entry within the TLB comprising the at least one entry stored for a previous translation of a previous virtual address, the TLB controller only indicates a match between the process identifier field and a translation process identifier field within the particular entry of the TLB if the translation process identifier field is also set to the clamped value.

Managing translation of a same address across multiple contexts using a same entry in a translation lookaside buffer

An enhanced mechanism for loading entries into a translation lookaside buffer (TLB) in hardware via indirect TLB entries. In one embodiment, if no direct TLB entry associated with the given virtual address is found in the TLB, the TLB is checked for an indirect TLB entry associated with the given virtual address. Each indirect TLB entry provides the real address of a page table associated with a specified range of virtual addresses and comprises an array of page table entries. If an indirect TLB entry associated with the given virtual address is found in the TLB, a computed address is generated by combining a real address field from the indirect TLB entry and bits from the given virtual address, a page table entry (PTE) is obtained by reading a word from a memory at the computed address, and the PTE is loaded into the TLB as a direct TLB entry.

Loading entries into a TLB in hardware via indirect TLB entries

Memory allocation to multiple computing units is disclosed. A static offset for each computing unit is determined, and a portion of memory is allocated for each computing unit, and remapped into a contiguous logical region that is addressable by a pointer plus the static offset. The portion of the memory is dynamically passed out to each computing unit as the computing units need memory. Upon the initial contiguous memory being completely passed out to the computing units, a number of physically non-contiguous sections of memory are mapped into another logically contiguous section of memory. A portion of this logically contiguous section of memory is allocated for each computing unit, and is addressable by a pointer plus the static offset that was previously determined. The portion of the logically contiguous section of memory can be dynamically passed out to each computing unit as the computing units need memory.

Growing community of inventors

Weston, Australia

Paul Mackerras