brintos

brintos / linux-shallow public Read only

0
0
Text · 14.4 KiB · 80d2c0a Raw
303 lines · plain
1.. SPDX-License-Identifier: GPL-2.02 3===========================4Hypercall Op-codes (hcalls)5===========================6 7Overview8=========9 10Virtualization on 64-bit Power Book3S Platforms is based on the PAPR11specification [1]_ which describes the run-time environment for a guest12operating system and how it should interact with the hypervisor for13privileged operations. Currently there are two PAPR compliant hypervisors:14 15- **IBM PowerVM (PHYP)**: IBM's proprietary hypervisor that supports AIX,16  IBM-i and  Linux as supported guests (termed as Logical Partitions17  or LPARS). It supports the full PAPR specification.18 19- **Qemu/KVM**: Supports PPC64 linux guests running on a PPC64 linux host.20  Though it only implements a subset of PAPR specification called LoPAPR [2]_.21 22On PPC64 arch a guest kernel running on top of a PAPR hypervisor is called23a *pSeries guest*. A pseries guest runs in a supervisor mode (HV=0) and must24issue hypercalls to the hypervisor whenever it needs to perform an action25that is hypervisor privileged [3]_ or for other services managed by the26hypervisor.27 28Hence a Hypercall (hcall) is essentially a request by the pseries guest29asking hypervisor to perform a privileged operation on behalf of the guest. The30guest issues a with necessary input operands. The hypervisor after performing31the privilege operation returns a status code and output operands back to the32guest.33 34HCALL ABI35=========36The ABI specification for a hcall between a pseries guest and PAPR hypervisor37is covered in section 14.5.3 of ref [2]_. Switch to the  Hypervisor context is38done via the instruction **HVCS** that expects the Opcode for hcall is set in *r3*39and any in-arguments for the hcall are provided in registers *r4-r12*. If values40have to be passed through a memory buffer, the data stored in that buffer should be41in Big-endian byte order.42 43Once control returns back to the guest after hypervisor has serviced the44'HVCS' instruction the return value of the hcall is available in *r3* and any45out values are returned in registers *r4-r12*. Again like in case of in-arguments,46any out values stored in a memory buffer will be in Big-endian byte order.47 48Powerpc arch code provides convenient wrappers named **plpar_hcall_xxx** defined49in a arch specific header [4]_ to issue hcalls from the linux kernel50running as pseries guest.51 52Register Conventions53====================54 55Any hcall should follow same register convention as described in section 2.2.1.156of "64-Bit ELF V2 ABI Specification: Power Architecture"[5]_. Table below57summarizes these conventions:58 59+----------+----------+-------------------------------------------+60| Register |Volatile  |  Purpose                                  |61| Range    |(Y/N)     |                                           |62+==========+==========+===========================================+63|   r0     |    Y     |  Optional-usage                           |64+----------+----------+-------------------------------------------+65|   r1     |    N     |  Stack Pointer                            |66+----------+----------+-------------------------------------------+67|   r2     |    N     |  TOC                                      |68+----------+----------+-------------------------------------------+69|   r3     |    Y     |  hcall opcode/return value                |70+----------+----------+-------------------------------------------+71|  r4-r10  |    Y     |  in and out values                        |72+----------+----------+-------------------------------------------+73|   r11    |    Y     |  Optional-usage/Environmental pointer     |74+----------+----------+-------------------------------------------+75|   r12    |    Y     |  Optional-usage/Function entry address at |76|          |          |  global entry point                       |77+----------+----------+-------------------------------------------+78|   r13    |    N     |  Thread-Pointer                           |79+----------+----------+-------------------------------------------+80|  r14-r31 |    N     |  Local Variables                          |81+----------+----------+-------------------------------------------+82|    LR    |    Y     |  Link Register                            |83+----------+----------+-------------------------------------------+84|   CTR    |    Y     |  Loop Counter                             |85+----------+----------+-------------------------------------------+86|   XER    |    Y     |  Fixed-point exception register.          |87+----------+----------+-------------------------------------------+88|  CR0-1   |    Y     |  Condition register fields.               |89+----------+----------+-------------------------------------------+90|  CR2-4   |    N     |  Condition register fields.               |91+----------+----------+-------------------------------------------+92|  CR5-7   |    Y     |  Condition register fields.               |93+----------+----------+-------------------------------------------+94|  Others  |    N     |                                           |95+----------+----------+-------------------------------------------+96 97DRC & DRC Indexes98=================99::100 101     DR1                                  Guest102     +--+        +------------+         +---------+103     |  | <----> |            |         |  User   |104     +--+  DRC1  |            |   DRC   |  Space  |105                 |    PAPR    |  Index  +---------+106     DR2         | Hypervisor |         |         |107     +--+        |            | <-----> |  Kernel |108     |  | <----> |            |  Hcall  |         |109     +--+  DRC2  +------------+         +---------+110 111PAPR hypervisor terms shared hardware resources like PCI devices, NVDIMMs etc112available for use by LPARs as Dynamic Resource (DR). When a DR is allocated to113an LPAR, PHYP creates a data-structure called Dynamic Resource Connector (DRC)114to manage LPAR access. An LPAR refers to a DRC via an opaque 32-bit number115called DRC-Index. The DRC-index value is provided to the LPAR via device-tree116where its present as an attribute in the device tree node associated with the117DR.118 119HCALL Return-values120===================121 122After servicing the hcall, hypervisor sets the return-value in *r3* indicating123success or failure of the hcall. In case of a failure an error code indicates124the cause for error. These codes are defined and documented in arch specific125header [4]_.126 127In some cases a hcall can potentially take a long time and need to be issued128multiple times in order to be completely serviced. These hcalls will usually129accept an opaque value *continue-token* within there argument list and a130return value of *H_CONTINUE* indicates that hypervisor hasn't still finished131servicing the hcall yet.132 133To make such hcalls the guest need to set *continue-token == 0* for the134initial call and use the hypervisor returned value of *continue-token*135for each subsequent hcall until hypervisor returns a non *H_CONTINUE*136return value.137 138HCALL Op-codes139==============140 141Below is a partial list of HCALLs that are supported by PHYP. For the142corresponding opcode values please look into the arch specific header [4]_:143 144**H_SCM_READ_METADATA**145 146| Input: *drcIndex, offset, buffer-address, numBytesToRead*147| Out: *numBytesRead*148| Return Value: *H_Success, H_Parameter, H_P2, H_P3, H_Hardware*149 150Given a DRC Index of an NVDIMM, read N-bytes from the metadata area151associated with it, at a specified offset and copy it to provided buffer.152The metadata area stores configuration information such as label information,153bad-blocks etc. The metadata area is located out-of-band of NVDIMM storage154area hence a separate access semantics is provided.155 156**H_SCM_WRITE_METADATA**157 158| Input: *drcIndex, offset, data, numBytesToWrite*159| Out: *None*160| Return Value: *H_Success, H_Parameter, H_P2, H_P4, H_Hardware*161 162Given a DRC Index of an NVDIMM, write N-bytes to the metadata area163associated with it, at the specified offset and from the provided buffer.164 165**H_SCM_BIND_MEM**166 167| Input: *drcIndex, startingScmBlockIndex, numScmBlocksToBind,*168| *targetLogicalMemoryAddress, continue-token*169| Out: *continue-token, targetLogicalMemoryAddress, numScmBlocksToBound*170| Return Value: *H_Success, H_Parameter, H_P2, H_P3, H_P4, H_Overlap,*171| *H_Too_Big, H_P5, H_Busy*172 173Given a DRC-Index of an NVDIMM, map a continuous SCM blocks range174*(startingScmBlockIndex, startingScmBlockIndex+numScmBlocksToBind)* to the guest175at *targetLogicalMemoryAddress* within guest physical address space. In176case *targetLogicalMemoryAddress == 0xFFFFFFFF_FFFFFFFF* then hypervisor177assigns a target address to the guest. The HCALL can fail if the Guest has178an active PTE entry to the SCM block being bound.179 180**H_SCM_UNBIND_MEM**181| Input: drcIndex, startingScmLogicalMemoryAddress, numScmBlocksToUnbind182| Out: numScmBlocksUnbound183| Return Value: *H_Success, H_Parameter, H_P2, H_P3, H_In_Use, H_Overlap,*184| *H_Busy, H_LongBusyOrder1mSec, H_LongBusyOrder10mSec*185 186Given a DRC-Index of an NVDimm, unmap *numScmBlocksToUnbind* SCM blocks starting187at *startingScmLogicalMemoryAddress* from guest physical address space. The188HCALL can fail if the Guest has an active PTE entry to the SCM block being189unbound.190 191**H_SCM_QUERY_BLOCK_MEM_BINDING**192 193| Input: *drcIndex, scmBlockIndex*194| Out: *Guest-Physical-Address*195| Return Value: *H_Success, H_Parameter, H_P2, H_NotFound*196 197Given a DRC-Index and an SCM Block index return the guest physical address to198which the SCM block is mapped to.199 200**H_SCM_QUERY_LOGICAL_MEM_BINDING**201 202| Input: *Guest-Physical-Address*203| Out: *drcIndex, scmBlockIndex*204| Return Value: *H_Success, H_Parameter, H_P2, H_NotFound*205 206Given a guest physical address return which DRC Index and SCM block is mapped207to that address.208 209**H_SCM_UNBIND_ALL**210 211| Input: *scmTargetScope, drcIndex*212| Out: *None*213| Return Value: *H_Success, H_Parameter, H_P2, H_P3, H_In_Use, H_Busy,*214| *H_LongBusyOrder1mSec, H_LongBusyOrder10mSec*215 216Depending on the Target scope unmap all SCM blocks belonging to all NVDIMMs217or all SCM blocks belonging to a single NVDIMM identified by its drcIndex218from the LPAR memory.219 220**H_SCM_HEALTH**221 222| Input: drcIndex223| Out: *health-bitmap (r4), health-bit-valid-bitmap (r5)*224| Return Value: *H_Success, H_Parameter, H_Hardware*225 226Given a DRC Index return the info on predictive failure and overall health of227the PMEM device. The asserted bits in the health-bitmap indicate one or more states228(described in table below) of the PMEM device and health-bit-valid-bitmap indicate229which bits in health-bitmap are valid. The bits are reported in230reverse bit ordering for example a value of 0xC400000000000000231indicates bits 0, 1, and 5 are valid.232 233Health Bitmap Flags:234 235+------+-----------------------------------------------------------------------+236|  Bit |               Definition                                              |237+======+=======================================================================+238|  00  | PMEM device is unable to persist memory contents.                     |239|      | If the system is powered down, nothing will be saved.                 |240+------+-----------------------------------------------------------------------+241|  01  | PMEM device failed to persist memory contents. Either contents were   |242|      | not saved successfully on power down or were not restored properly on |243|      | power up.                                                             |244+------+-----------------------------------------------------------------------+245|  02  | PMEM device contents are persisted from previous IPL. The data from   |246|      | the last boot were successfully restored.                             |247+------+-----------------------------------------------------------------------+248|  03  | PMEM device contents are not persisted from previous IPL. There was no|249|      | data to restore from the last boot.                                   |250+------+-----------------------------------------------------------------------+251|  04  | PMEM device memory life remaining is critically low                   |252+------+-----------------------------------------------------------------------+253|  05  | PMEM device will be garded off next IPL due to failure                |254+------+-----------------------------------------------------------------------+255|  06  | PMEM device contents cannot persist due to current platform health    |256|      | status. A hardware failure may prevent data from being saved or       |257|      | restored.                                                             |258+------+-----------------------------------------------------------------------+259|  07  | PMEM device is unable to persist memory contents in certain conditions|260+------+-----------------------------------------------------------------------+261|  08  | PMEM device is encrypted                                              |262+------+-----------------------------------------------------------------------+263|  09  | PMEM device has successfully completed a requested erase or secure    |264|      | erase procedure.                                                      |265+------+-----------------------------------------------------------------------+266|10:63 | Reserved / Unused                                                     |267+------+-----------------------------------------------------------------------+268 269**H_SCM_PERFORMANCE_STATS**270 271| Input: drcIndex, resultBuffer Addr272| Out: None273| Return Value:  *H_Success, H_Parameter, H_Unsupported, H_Hardware, H_Authority, H_Privilege*274 275Given a DRC Index collect the performance statistics for NVDIMM and copy them276to the resultBuffer.277 278**H_SCM_FLUSH**279 280| Input: *drcIndex, continue-token*281| Out: *continue-token*282| Return Value: *H_SUCCESS, H_Parameter, H_P2, H_BUSY*283 284Given a DRC Index Flush the data to backend NVDIMM device.285 286The hcall returns H_BUSY when the flush takes longer time and the hcall needs287to be issued multiple times in order to be completely serviced. The288*continue-token* from the output to be passed in the argument list of289subsequent hcalls to the hypervisor until the hcall is completely serviced290at which point H_SUCCESS or other error is returned by the hypervisor.291 292References293==========294.. [1] "Power Architecture Platform Reference"295       https://en.wikipedia.org/wiki/Power_Architecture_Platform_Reference296.. [2] "Linux on Power Architecture Platform Reference"297       https://members.openpowerfoundation.org/document/dl/469298.. [3] "Definitions and Notation" Book III-Section 14.5.3299       https://openpowerfoundation.org/?resource_lib=power-isa-version-3-0300.. [4] arch/powerpc/include/asm/hvcall.h301.. [5] "64-Bit ELF V2 ABI Specification: Power Architecture"302       https://openpowerfoundation.org/?resource_lib=64-bit-elf-v2-abi-specification-power-architecture303