发明名称 Activating a subphysical driver on failure of hypervisor for operating an I/O device shared by hypervisor and guest OS and virtual computer system
摘要 A control method for an I/O device, the I/O device being shared by a hypervisor and a first guest OS, the hypervisor comprising a physical driver for using a physical function, the first guest OS comprising a virtual driver for using a virtual function, the control method comprising: a first step of acquiring, by the hypervisor, a state of the I/O device via the physical driver; a second step of monitoring, by the first guest OS, the hypervisor, thereby determining whether the hypervisor has been brought into a predetermined state or not; a third step of activating, by the first guest OS, a sub physical driver for operating the I/O device when the first guest OS determines that the hypervisor has been brought into the predetermined state; and a fourth step of carrying out, by the first guest OS, transmission/reception via a queue set in advance on the memory.
申请公布号 US9430266(B2) 申请公布日期 2016.08.30
申请号 US201314083493 申请日期 2013.11.19
申请人 Hitachi, Ltd. 发明人 Takeuchi Tadashi
分类号 G06F9/455;G06F9/50;G06F11/14;G06F11/20 主分类号 G06F9/455
代理机构 Mattingly & Malur, PC 代理人 Mattingly & Malur, PC
主权项 1. A control method for a physical I/O device, in a computer comprising a processor, a memory, and the I/O device, the I/O device being shared by a hypervisor and a first guest OS, the I/O device comprising a physical function and a virtual function, the hypervisor comprising a physical driver for using the physical function, the first guest OS comprising a virtual driver for using the virtual function, the control method comprising: a first step of acquiring, by the hypervisor, a state of the I/O device via the physical driver; a second step of monitoring, by the first guest OS, the hypervisor, thereby determining whether the hypervisor has been brought into a predetermined state or not; a third step of activating, by the first guest OS, a sub physical driver for operating the I/O device when the first guest OS determines that the hypervisor has been brought into the predetermined state; and a fourth step of carrying out, by the first guest OS, transmission of data and reception of data via a queue previously stored on the memory, wherein the third step comprises acquiring, by the first guest OS, a state of the I/O device from the virtual driver via the sub physical driver, the state of the I/O device comprising a link state and a bandwidth state for each of the physical function and the virtual function, and then activating the sub physical driver.
地址 Tokyo JP