diff options
Diffstat (limited to 'Documentation/arm64')
21 files changed, 5204 insertions, 0 deletions
diff --git a/Documentation/arm64/acpi_object_usage.rst b/Documentation/arm64/acpi_object_usage.rst new file mode 100644 index 000000000..0609da739 --- /dev/null +++ b/Documentation/arm64/acpi_object_usage.rst @@ -0,0 +1,738 @@ +=========== +ACPI Tables +=========== + +The expectations of individual ACPI tables are discussed in the list that +follows. + +If a section number is used, it refers to a section number in the ACPI +specification where the object is defined. If "Signature Reserved" is used, +the table signature (the first four bytes of the table) is the only portion +of the table recognized by the specification, and the actual table is defined +outside of the UEFI Forum (see Section 5.2.6 of the specification). + +For ACPI on arm64, tables also fall into the following categories: + + - Required: DSDT, FADT, GTDT, MADT, MCFG, RSDP, SPCR, XSDT + + - Recommended: BERT, EINJ, ERST, HEST, PCCT, SSDT + + - Optional: BGRT, CPEP, CSRT, DBG2, DRTM, ECDT, FACS, FPDT, IBFT, + IORT, MCHI, MPST, MSCT, NFIT, PMTT, RASF, SBST, SLIT, SPMI, SRAT, + STAO, TCPA, TPM2, UEFI, XENV + + - Not supported: BOOT, DBGP, DMAR, ETDT, HPET, IVRS, LPIT, MSDM, OEMx, + PSDT, RSDT, SLIC, WAET, WDAT, WDRT, WPBT + +====== ======================================================================== +Table Usage for ARMv8 Linux +====== ======================================================================== +BERT Section 18.3 (signature == "BERT") + + **Boot Error Record Table** + + Must be supplied if RAS support is provided by the platform. It + is recommended this table be supplied. + +BOOT Signature Reserved (signature == "BOOT") + + **simple BOOT flag table** + + Microsoft only table, will not be supported. + +BGRT Section 5.2.22 (signature == "BGRT") + + **Boot Graphics Resource Table** + + Optional, not currently supported, with no real use-case for an + ARM server. + +CPEP Section 5.2.18 (signature == "CPEP") + + **Corrected Platform Error Polling table** + + Optional, not currently supported, and not recommended until such + time as ARM-compatible hardware is available, and the specification + suitably modified. + +CSRT Signature Reserved (signature == "CSRT") + + **Core System Resources Table** + + Optional, not currently supported. + +DBG2 Signature Reserved (signature == "DBG2") + + **DeBuG port table 2** + + License has changed and should be usable. Optional if used instead + of earlycon=<device> on the command line. + +DBGP Signature Reserved (signature == "DBGP") + + **DeBuG Port table** + + Microsoft only table, will not be supported. + +DSDT Section 5.2.11.1 (signature == "DSDT") + + **Differentiated System Description Table** + + A DSDT is required; see also SSDT. + + ACPI tables contain only one DSDT but can contain one or more SSDTs, + which are optional. Each SSDT can only add to the ACPI namespace, + but cannot modify or replace anything in the DSDT. + +DMAR Signature Reserved (signature == "DMAR") + + **DMA Remapping table** + + x86 only table, will not be supported. + +DRTM Signature Reserved (signature == "DRTM") + + **Dynamic Root of Trust for Measurement table** + + Optional, not currently supported. + +ECDT Section 5.2.16 (signature == "ECDT") + + **Embedded Controller Description Table** + + Optional, not currently supported, but could be used on ARM if and + only if one uses the GPE_BIT field to represent an IRQ number, since + there are no GPE blocks defined in hardware reduced mode. This would + need to be modified in the ACPI specification. + +EINJ Section 18.6 (signature == "EINJ") + + **Error Injection table** + + This table is very useful for testing platform response to error + conditions; it allows one to inject an error into the system as + if it had actually occurred. However, this table should not be + shipped with a production system; it should be dynamically loaded + and executed with the ACPICA tools only during testing. + +ERST Section 18.5 (signature == "ERST") + + **Error Record Serialization Table** + + On a platform supports RAS, this table must be supplied if it is not + UEFI-based; if it is UEFI-based, this table may be supplied. When this + table is not present, UEFI run time service will be utilized to save + and retrieve hardware error information to and from a persistent store. + +ETDT Signature Reserved (signature == "ETDT") + + **Event Timer Description Table** + + Obsolete table, will not be supported. + +FACS Section 5.2.10 (signature == "FACS") + + **Firmware ACPI Control Structure** + + It is unlikely that this table will be terribly useful. If it is + provided, the Global Lock will NOT be used since it is not part of + the hardware reduced profile, and only 64-bit address fields will + be considered valid. + +FADT Section 5.2.9 (signature == "FACP") + + **Fixed ACPI Description Table** + Required for arm64. + + + The HW_REDUCED_ACPI flag must be set. All of the fields that are + to be ignored when HW_REDUCED_ACPI is set are expected to be set to + zero. + + If an FACS table is provided, the X_FIRMWARE_CTRL field is to be + used, not FIRMWARE_CTRL. + + If PSCI is used (as is recommended), make sure that ARM_BOOT_ARCH is + filled in properly - that the PSCI_COMPLIANT flag is set and that + PSCI_USE_HVC is set or unset as needed (see table 5-37). + + For the DSDT that is also required, the X_DSDT field is to be used, + not the DSDT field. + +FPDT Section 5.2.23 (signature == "FPDT") + + **Firmware Performance Data Table** + + Optional, not currently supported. + +GTDT Section 5.2.24 (signature == "GTDT") + + **Generic Timer Description Table** + + Required for arm64. + +HEST Section 18.3.2 (signature == "HEST") + + **Hardware Error Source Table** + + ARM-specific error sources have been defined; please use those or the + PCI types such as type 6 (AER Root Port), 7 (AER Endpoint), or 8 (AER + Bridge), or use type 9 (Generic Hardware Error Source). Firmware first + error handling is possible if and only if Trusted Firmware is being + used on arm64. + + Must be supplied if RAS support is provided by the platform. It + is recommended this table be supplied. + +HPET Signature Reserved (signature == "HPET") + + **High Precision Event timer Table** + + x86 only table, will not be supported. + +IBFT Signature Reserved (signature == "IBFT") + + **iSCSI Boot Firmware Table** + + Microsoft defined table, support TBD. + +IORT Signature Reserved (signature == "IORT") + + **Input Output Remapping Table** + + arm64 only table, required in order to describe IO topology, SMMUs, + and GIC ITSs, and how those various components are connected together, + such as identifying which components are behind which SMMUs/ITSs. + This table will only be required on certain SBSA platforms (e.g., + when using GICv3-ITS and an SMMU); on SBSA Level 0 platforms, it + remains optional. + +IVRS Signature Reserved (signature == "IVRS") + + **I/O Virtualization Reporting Structure** + + x86_64 (AMD) only table, will not be supported. + +LPIT Signature Reserved (signature == "LPIT") + + **Low Power Idle Table** + + x86 only table as of ACPI 5.1; starting with ACPI 6.0, processor + descriptions and power states on ARM platforms should use the DSDT + and define processor container devices (_HID ACPI0010, Section 8.4, + and more specifically 8.4.3 and 8.4.4). + +MADT Section 5.2.12 (signature == "APIC") + + **Multiple APIC Description Table** + + Required for arm64. Only the GIC interrupt controller structures + should be used (types 0xA - 0xF). + +MCFG Signature Reserved (signature == "MCFG") + + **Memory-mapped ConFiGuration space** + + If the platform supports PCI/PCIe, an MCFG table is required. + +MCHI Signature Reserved (signature == "MCHI") + + **Management Controller Host Interface table** + + Optional, not currently supported. + +MPST Section 5.2.21 (signature == "MPST") + + **Memory Power State Table** + + Optional, not currently supported. + +MSCT Section 5.2.19 (signature == "MSCT") + + **Maximum System Characteristic Table** + + Optional, not currently supported. + +MSDM Signature Reserved (signature == "MSDM") + + **Microsoft Data Management table** + + Microsoft only table, will not be supported. + +NFIT Section 5.2.25 (signature == "NFIT") + + **NVDIMM Firmware Interface Table** + + Optional, not currently supported. + +OEMx Signature of "OEMx" only + + **OEM Specific Tables** + + All tables starting with a signature of "OEM" are reserved for OEM + use. Since these are not meant to be of general use but are limited + to very specific end users, they are not recommended for use and are + not supported by the kernel for arm64. + +PCCT Section 14.1 (signature == "PCCT) + + **Platform Communications Channel Table** + + Recommend for use on arm64; use of PCC is recommended when using CPPC + to control performance and power for platform processors. + +PMTT Section 5.2.21.12 (signature == "PMTT") + + **Platform Memory Topology Table** + + Optional, not currently supported. + +PSDT Section 5.2.11.3 (signature == "PSDT") + + **Persistent System Description Table** + + Obsolete table, will not be supported. + +RASF Section 5.2.20 (signature == "RASF") + + **RAS Feature table** + + Optional, not currently supported. + +RSDP Section 5.2.5 (signature == "RSD PTR") + + **Root System Description PoinTeR** + + Required for arm64. + +RSDT Section 5.2.7 (signature == "RSDT") + + **Root System Description Table** + + Since this table can only provide 32-bit addresses, it is deprecated + on arm64, and will not be used. If provided, it will be ignored. + +SBST Section 5.2.14 (signature == "SBST") + + **Smart Battery Subsystem Table** + + Optional, not currently supported. + +SLIC Signature Reserved (signature == "SLIC") + + **Software LIcensing table** + + Microsoft only table, will not be supported. + +SLIT Section 5.2.17 (signature == "SLIT") + + **System Locality distance Information Table** + + Optional in general, but required for NUMA systems. + +SPCR Signature Reserved (signature == "SPCR") + + **Serial Port Console Redirection table** + + Required for arm64. + +SPMI Signature Reserved (signature == "SPMI") + + **Server Platform Management Interface table** + + Optional, not currently supported. + +SRAT Section 5.2.16 (signature == "SRAT") + + **System Resource Affinity Table** + + Optional, but if used, only the GICC Affinity structures are read. + To support arm64 NUMA, this table is required. + +SSDT Section 5.2.11.2 (signature == "SSDT") + + **Secondary System Description Table** + + These tables are a continuation of the DSDT; these are recommended + for use with devices that can be added to a running system, but can + also serve the purpose of dividing up device descriptions into more + manageable pieces. + + An SSDT can only ADD to the ACPI namespace. It cannot modify or + replace existing device descriptions already in the namespace. + + These tables are optional, however. ACPI tables should contain only + one DSDT but can contain many SSDTs. + +STAO Signature Reserved (signature == "STAO") + + **_STA Override table** + + Optional, but only necessary in virtualized environments in order to + hide devices from guest OSs. + +TCPA Signature Reserved (signature == "TCPA") + + **Trusted Computing Platform Alliance table** + + Optional, not currently supported, and may need changes to fully + interoperate with arm64. + +TPM2 Signature Reserved (signature == "TPM2") + + **Trusted Platform Module 2 table** + + Optional, not currently supported, and may need changes to fully + interoperate with arm64. + +UEFI Signature Reserved (signature == "UEFI") + + **UEFI ACPI data table** + + Optional, not currently supported. No known use case for arm64, + at present. + +WAET Signature Reserved (signature == "WAET") + + **Windows ACPI Emulated devices Table** + + Microsoft only table, will not be supported. + +WDAT Signature Reserved (signature == "WDAT") + + **Watch Dog Action Table** + + Microsoft only table, will not be supported. + +WDRT Signature Reserved (signature == "WDRT") + + **Watch Dog Resource Table** + + Microsoft only table, will not be supported. + +WPBT Signature Reserved (signature == "WPBT") + + **Windows Platform Binary Table** + + Microsoft only table, will not be supported. + +XENV Signature Reserved (signature == "XENV") + + **Xen project table** + + Optional, used only by Xen at present. + +XSDT Section 5.2.8 (signature == "XSDT") + + **eXtended System Description Table** + + Required for arm64. +====== ======================================================================== + +ACPI Objects +------------ +The expectations on individual ACPI objects that are likely to be used are +shown in the list that follows; any object not explicitly mentioned below +should be used as needed for a particular platform or particular subsystem, +such as power management or PCI. + +===== ================ ======================================================== +Name Section Usage for ARMv8 Linux +===== ================ ======================================================== +_CCA 6.2.17 This method must be defined for all bus masters + on arm64 - there are no assumptions made about + whether such devices are cache coherent or not. + The _CCA value is inherited by all descendants of + these devices so it does not need to be repeated. + Without _CCA on arm64, the kernel does not know what + to do about setting up DMA for the device. + + NB: this method provides default cache coherency + attributes; the presence of an SMMU can be used to + modify that, however. For example, a master could + default to non-coherent, but be made coherent with + the appropriate SMMU configuration (see Table 17 of + the IORT specification, ARM Document DEN 0049B). + +_CID 6.1.2 Use as needed, see also _HID. + +_CLS 6.1.3 Use as needed, see also _HID. + +_CPC 8.4.7.1 Use as needed, power management specific. CPPC is + recommended on arm64. + +_CRS 6.2.2 Required on arm64. + +_CSD 8.4.2.2 Use as needed, used only in conjunction with _CST. + +_CST 8.4.2.1 Low power idle states (8.4.4) are recommended instead + of C-states. + +_DDN 6.1.4 This field can be used for a device name. However, + it is meant for DOS device names (e.g., COM1), so be + careful of its use across OSes. + +_DSD 6.2.5 To be used with caution. If this object is used, try + to use it within the constraints already defined by the + Device Properties UUID. Only in rare circumstances + should it be necessary to create a new _DSD UUID. + + In either case, submit the _DSD definition along with + any driver patches for discussion, especially when + device properties are used. A driver will not be + considered complete without a corresponding _DSD + description. Once approved by kernel maintainers, + the UUID or device properties must then be registered + with the UEFI Forum; this may cause some iteration as + more than one OS will be registering entries. + +_DSM 9.1.1 Do not use this method. It is not standardized, the + return values are not well documented, and it is + currently a frequent source of error. + +\_GL 5.7.1 This object is not to be used in hardware reduced + mode, and therefore should not be used on arm64. + +_GLK 6.5.7 This object requires a global lock be defined; there + is no global lock on arm64 since it runs in hardware + reduced mode. Hence, do not use this object on arm64. + +\_GPE 5.3.1 This namespace is for x86 use only. Do not use it + on arm64. + +_HID 6.1.5 This is the primary object to use in device probing, + though _CID and _CLS may also be used. + +_INI 6.5.1 Not required, but can be useful in setting up devices + when UEFI leaves them in a state that may not be what + the driver expects before it starts probing. + +_LPI 8.4.4.3 Recommended for use with processor definitions (_HID + ACPI0010) on arm64. See also _RDI. + +_MLS 6.1.7 Highly recommended for use in internationalization. + +_OFF 7.2.2 It is recommended to define this method for any device + that can be turned on or off. + +_ON 7.2.3 It is recommended to define this method for any device + that can be turned on or off. + +\_OS 5.7.3 This method will return "Linux" by default (this is + the value of the macro ACPI_OS_NAME on Linux). The + command line parameter acpi_os=<string> can be used + to set it to some other value. + +_OSC 6.2.11 This method can be a global method in ACPI (i.e., + \_SB._OSC), or it may be associated with a specific + device (e.g., \_SB.DEV0._OSC), or both. When used + as a global method, only capabilities published in + the ACPI specification are allowed. When used as + a device-specific method, the process described for + using _DSD MUST be used to create an _OSC definition; + out-of-process use of _OSC is not allowed. That is, + submit the device-specific _OSC usage description as + part of the kernel driver submission, get it approved + by the kernel community, then register it with the + UEFI Forum. + +\_OSI 5.7.2 Deprecated on ARM64. As far as ACPI firmware is + concerned, _OSI is not to be used to determine what + sort of system is being used or what functionality + is provided. The _OSC method is to be used instead. + +_PDC 8.4.1 Deprecated, do not use on arm64. + +\_PIC 5.8.1 The method should not be used. On arm64, the only + interrupt model available is GIC. + +\_PR 5.3.1 This namespace is for x86 use only on legacy systems. + Do not use it on arm64. + +_PRT 6.2.13 Required as part of the definition of all PCI root + devices. + +_PRx 7.3.8-11 Use as needed; power management specific. If _PR0 is + defined, _PR3 must also be defined. + +_PSx 7.3.2-5 Use as needed; power management specific. If _PS0 is + defined, _PS3 must also be defined. If clocks or + regulators need adjusting to be consistent with power + usage, change them in these methods. + +_RDI 8.4.4.4 Recommended for use with processor definitions (_HID + ACPI0010) on arm64. This should only be used in + conjunction with _LPI. + +\_REV 5.7.4 Always returns the latest version of ACPI supported. + +\_SB 5.3.1 Required on arm64; all devices must be defined in this + namespace. + +_SLI 6.2.15 Use is recommended when SLIT table is in use. + +_STA 6.3.7, It is recommended to define this method for any device + 7.2.4 that can be turned on or off. See also the STAO table + that provides overrides to hide devices in virtualized + environments. + +_SRS 6.2.16 Use as needed; see also _PRS. + +_STR 6.1.10 Recommended for conveying device names to end users; + this is preferred over using _DDN. + +_SUB 6.1.9 Use as needed; _HID or _CID are preferred. + +_SUN 6.1.11 Use as needed, but recommended. + +_SWS 7.4.3 Use as needed; power management specific; this may + require specification changes for use on arm64. + +_UID 6.1.12 Recommended for distinguishing devices of the same + class; define it if at all possible. +===== ================ ======================================================== + + + + +ACPI Event Model +---------------- +Do not use GPE block devices; these are not supported in the hardware reduced +profile used by arm64. Since there are no GPE blocks defined for use on ARM +platforms, ACPI events must be signaled differently. + +There are two options: GPIO-signaled interrupts (Section 5.6.5), and +interrupt-signaled events (Section 5.6.9). Interrupt-signaled events are a +new feature in the ACPI 6.1 specification. Either - or both - can be used +on a given platform, and which to use may be dependent of limitations in any +given SoC. If possible, interrupt-signaled events are recommended. + + +ACPI Processor Control +---------------------- +Section 8 of the ACPI specification changed significantly in version 6.0. +Processors should now be defined as Device objects with _HID ACPI0007; do +not use the deprecated Processor statement in ASL. All multiprocessor systems +should also define a hierarchy of processors, done with Processor Container +Devices (see Section 8.4.3.1, _HID ACPI0010); do not use processor aggregator +devices (Section 8.5) to describe processor topology. Section 8.4 of the +specification describes the semantics of these object definitions and how +they interrelate. + +Most importantly, the processor hierarchy defined also defines the low power +idle states that are available to the platform, along with the rules for +determining which processors can be turned on or off and the circumstances +that control that. Without this information, the processors will run in +whatever power state they were left in by UEFI. + +Note too, that the processor Device objects defined and the entries in the +MADT for GICs are expected to be in synchronization. The _UID of the Device +object must correspond to processor IDs used in the MADT. + +It is recommended that CPPC (8.4.5) be used as the primary model for processor +performance control on arm64. C-states and P-states may become available at +some point in the future, but most current design work appears to favor CPPC. + +Further, it is essential that the ARMv8 SoC provide a fully functional +implementation of PSCI; this will be the only mechanism supported by ACPI +to control CPU power state. Booting of secondary CPUs using the ACPI +parking protocol is possible, but discouraged, since only PSCI is supported +for ARM servers. + + +ACPI System Address Map Interfaces +---------------------------------- +In Section 15 of the ACPI specification, several methods are mentioned as +possible mechanisms for conveying memory resource information to the kernel. +For arm64, we will only support UEFI for booting with ACPI, hence the UEFI +GetMemoryMap() boot service is the only mechanism that will be used. + + +ACPI Platform Error Interfaces (APEI) +------------------------------------- +The APEI tables supported are described above. + +APEI requires the equivalent of an SCI and an NMI on ARMv8. The SCI is used +to notify the OSPM of errors that have occurred but can be corrected and the +system can continue correct operation, even if possibly degraded. The NMI is +used to indicate fatal errors that cannot be corrected, and require immediate +attention. + +Since there is no direct equivalent of the x86 SCI or NMI, arm64 handles +these slightly differently. The SCI is handled as a high priority interrupt; +given that these are corrected (or correctable) errors being reported, this +is sufficient. The NMI is emulated as the highest priority interrupt +possible. This implies some caution must be used since there could be +interrupts at higher privilege levels or even interrupts at the same priority +as the emulated NMI. In Linux, this should not be the case but one should +be aware it could happen. + + +ACPI Objects Not Supported on ARM64 +----------------------------------- +While this may change in the future, there are several classes of objects +that can be defined, but are not currently of general interest to ARM servers. +Some of these objects have x86 equivalents, and may actually make sense in ARM +servers. However, there is either no hardware available at present, or there +may not even be a non-ARM implementation yet. Hence, they are not currently +supported. + +The following classes of objects are not supported: + + - Section 9.2: ambient light sensor devices + + - Section 9.3: battery devices + + - Section 9.4: lids (e.g., laptop lids) + + - Section 9.8.2: IDE controllers + + - Section 9.9: floppy controllers + + - Section 9.10: GPE block devices + + - Section 9.15: PC/AT RTC/CMOS devices + + - Section 9.16: user presence detection devices + + - Section 9.17: I/O APIC devices; all GICs must be enumerable via MADT + + - Section 9.18: time and alarm devices (see 9.15) + + - Section 10: power source and power meter devices + + - Section 11: thermal management + + - Section 12: embedded controllers interface + + - Section 13: SMBus interfaces + + +This also means that there is no support for the following objects: + +==== =========================== ==== ========== +Name Section Name Section +==== =========================== ==== ========== +_ALC 9.3.4 _FDM 9.10.3 +_ALI 9.3.2 _FIX 6.2.7 +_ALP 9.3.6 _GAI 10.4.5 +_ALR 9.3.5 _GHL 10.4.7 +_ALT 9.3.3 _GTM 9.9.2.1.1 +_BCT 10.2.2.10 _LID 9.5.1 +_BDN 6.5.3 _PAI 10.4.4 +_BIF 10.2.2.1 _PCL 10.3.2 +_BIX 10.2.2.1 _PIF 10.3.3 +_BLT 9.2.3 _PMC 10.4.1 +_BMA 10.2.2.4 _PMD 10.4.8 +_BMC 10.2.2.12 _PMM 10.4.3 +_BMD 10.2.2.11 _PRL 10.3.4 +_BMS 10.2.2.5 _PSR 10.3.1 +_BST 10.2.2.6 _PTP 10.4.2 +_BTH 10.2.2.7 _SBS 10.1.3 +_BTM 10.2.2.9 _SHL 10.4.6 +_BTP 10.2.2.8 _STM 9.9.2.1.1 +_DCK 6.5.2 _UPD 9.16.1 +_EC 12.12 _UPP 9.16.2 +_FDE 9.10.1 _WPC 10.5.2 +_FDI 9.10.2 _WPP 10.5.3 +==== =========================== ==== ========== diff --git a/Documentation/arm64/amu.rst b/Documentation/arm64/amu.rst new file mode 100644 index 000000000..01f2de2b0 --- /dev/null +++ b/Documentation/arm64/amu.rst @@ -0,0 +1,119 @@ +.. _amu_index: + +======================================================= +Activity Monitors Unit (AMU) extension in AArch64 Linux +======================================================= + +Author: Ionela Voinescu <ionela.voinescu@arm.com> + +Date: 2019-09-10 + +This document briefly describes the provision of Activity Monitors Unit +support in AArch64 Linux. + + +Architecture overview +--------------------- + +The activity monitors extension is an optional extension introduced by the +ARMv8.4 CPU architecture. + +The activity monitors unit, implemented in each CPU, provides performance +counters intended for system management use. The AMU extension provides a +system register interface to the counter registers and also supports an +optional external memory-mapped interface. + +Version 1 of the Activity Monitors architecture implements a counter group +of four fixed and architecturally defined 64-bit event counters. + + - CPU cycle counter: increments at the frequency of the CPU. + - Constant counter: increments at the fixed frequency of the system + clock. + - Instructions retired: increments with every architecturally executed + instruction. + - Memory stall cycles: counts instruction dispatch stall cycles caused by + misses in the last level cache within the clock domain. + +When in WFI or WFE these counters do not increment. + +The Activity Monitors architecture provides space for up to 16 architected +event counters. Future versions of the architecture may use this space to +implement additional architected event counters. + +Additionally, version 1 implements a counter group of up to 16 auxiliary +64-bit event counters. + +On cold reset all counters reset to 0. + + +Basic support +------------- + +The kernel can safely run a mix of CPUs with and without support for the +activity monitors extension. Therefore, when CONFIG_ARM64_AMU_EXTN is +selected we unconditionally enable the capability to allow any late CPU +(secondary or hotplugged) to detect and use the feature. + +When the feature is detected on a CPU, we flag the availability of the +feature but this does not guarantee the correct functionality of the +counters, only the presence of the extension. + +Firmware (code running at higher exception levels, e.g. arm-tf) support is +needed to: + + - Enable access for lower exception levels (EL2 and EL1) to the AMU + registers. + - Enable the counters. If not enabled these will read as 0. + - Save/restore the counters before/after the CPU is being put/brought up + from the 'off' power state. + +When using kernels that have this feature enabled but boot with broken +firmware the user may experience panics or lockups when accessing the +counter registers. Even if these symptoms are not observed, the values +returned by the register reads might not correctly reflect reality. Most +commonly, the counters will read as 0, indicating that they are not +enabled. + +If proper support is not provided in firmware it's best to disable +CONFIG_ARM64_AMU_EXTN. To be noted that for security reasons, this does not +bypass the setting of AMUSERENR_EL0 to trap accesses from EL0 (userspace) to +EL1 (kernel). Therefore, firmware should still ensure accesses to AMU registers +are not trapped in EL2/EL3. + +The fixed counters of AMUv1 are accessible though the following system +register definitions: + + - SYS_AMEVCNTR0_CORE_EL0 + - SYS_AMEVCNTR0_CONST_EL0 + - SYS_AMEVCNTR0_INST_RET_EL0 + - SYS_AMEVCNTR0_MEM_STALL_EL0 + +Auxiliary platform specific counters can be accessed using +SYS_AMEVCNTR1_EL0(n), where n is a value between 0 and 15. + +Details can be found in: arch/arm64/include/asm/sysreg.h. + + +Userspace access +---------------- + +Currently, access from userspace to the AMU registers is disabled due to: + + - Security reasons: they might expose information about code executed in + secure mode. + - Purpose: AMU counters are intended for system management use. + +Also, the presence of the feature is not visible to userspace. + + +Virtualization +-------------- + +Currently, access from userspace (EL0) and kernelspace (EL1) on the KVM +guest side is disabled due to: + + - Security reasons: they might expose information about code executed + by other guests or the host. + +Any attempt to access the AMU registers will result in an UNDEFINED +exception being injected into the guest. diff --git a/Documentation/arm64/arm-acpi.rst b/Documentation/arm64/arm-acpi.rst new file mode 100644 index 000000000..47ecb9930 --- /dev/null +++ b/Documentation/arm64/arm-acpi.rst @@ -0,0 +1,528 @@ +===================== +ACPI on ARMv8 Servers +===================== + +ACPI can be used for ARMv8 general purpose servers designed to follow +the ARM SBSA (Server Base System Architecture) [0] and SBBR (Server +Base Boot Requirements) [1] specifications. Please note that the SBBR +can be retrieved simply by visiting [1], but the SBSA is currently only +available to those with an ARM login due to ARM IP licensing concerns. + +The ARMv8 kernel implements the reduced hardware model of ACPI version +5.1 or later. Links to the specification and all external documents +it refers to are managed by the UEFI Forum. The specification is +available at http://www.uefi.org/specifications and documents referenced +by the specification can be found via http://www.uefi.org/acpi. + +If an ARMv8 system does not meet the requirements of the SBSA and SBBR, +or cannot be described using the mechanisms defined in the required ACPI +specifications, then ACPI may not be a good fit for the hardware. + +While the documents mentioned above set out the requirements for building +industry-standard ARMv8 servers, they also apply to more than one operating +system. The purpose of this document is to describe the interaction between +ACPI and Linux only, on an ARMv8 system -- that is, what Linux expects of +ACPI and what ACPI can expect of Linux. + + +Why ACPI on ARM? +---------------- +Before examining the details of the interface between ACPI and Linux, it is +useful to understand why ACPI is being used. Several technologies already +exist in Linux for describing non-enumerable hardware, after all. In this +section we summarize a blog post [2] from Grant Likely that outlines the +reasoning behind ACPI on ARMv8 servers. Actually, we snitch a good portion +of the summary text almost directly, to be honest. + +The short form of the rationale for ACPI on ARM is: + +- ACPI’s byte code (AML) allows the platform to encode hardware behavior, + while DT explicitly does not support this. For hardware vendors, being + able to encode behavior is a key tool used in supporting operating + system releases on new hardware. + +- ACPI’s OSPM defines a power management model that constrains what the + platform is allowed to do into a specific model, while still providing + flexibility in hardware design. + +- In the enterprise server environment, ACPI has established bindings (such + as for RAS) which are currently used in production systems. DT does not. + Such bindings could be defined in DT at some point, but doing so means ARM + and x86 would end up using completely different code paths in both firmware + and the kernel. + +- Choosing a single interface to describe the abstraction between a platform + and an OS is important. Hardware vendors would not be required to implement + both DT and ACPI if they want to support multiple operating systems. And, + agreeing on a single interface instead of being fragmented into per OS + interfaces makes for better interoperability overall. + +- The new ACPI governance process works well and Linux is now at the same + table as hardware vendors and other OS vendors. In fact, there is no + longer any reason to feel that ACPI only belongs to Windows or that + Linux is in any way secondary to Microsoft in this arena. The move of + ACPI governance into the UEFI forum has significantly opened up the + specification development process, and currently, a large portion of the + changes being made to ACPI are being driven by Linux. + +Key to the use of ACPI is the support model. For servers in general, the +responsibility for hardware behaviour cannot solely be the domain of the +kernel, but rather must be split between the platform and the kernel, in +order to allow for orderly change over time. ACPI frees the OS from needing +to understand all the minute details of the hardware so that the OS doesn’t +need to be ported to each and every device individually. It allows the +hardware vendors to take responsibility for power management behaviour without +depending on an OS release cycle which is not under their control. + +ACPI is also important because hardware and OS vendors have already worked +out the mechanisms for supporting a general purpose computing ecosystem. The +infrastructure is in place, the bindings are in place, and the processes are +in place. DT does exactly what Linux needs it to when working with vertically +integrated devices, but there are no good processes for supporting what the +server vendors need. Linux could potentially get there with DT, but doing so +really just duplicates something that already works. ACPI already does what +the hardware vendors need, Microsoft won’t collaborate on DT, and hardware +vendors would still end up providing two completely separate firmware +interfaces -- one for Linux and one for Windows. + + +Kernel Compatibility +-------------------- +One of the primary motivations for ACPI is standardization, and using that +to provide backward compatibility for Linux kernels. In the server market, +software and hardware are often used for long periods. ACPI allows the +kernel and firmware to agree on a consistent abstraction that can be +maintained over time, even as hardware or software change. As long as the +abstraction is supported, systems can be updated without necessarily having +to replace the kernel. + +When a Linux driver or subsystem is first implemented using ACPI, it by +definition ends up requiring a specific version of the ACPI specification +-- it's baseline. ACPI firmware must continue to work, even though it may +not be optimal, with the earliest kernel version that first provides support +for that baseline version of ACPI. There may be a need for additional drivers, +but adding new functionality (e.g., CPU power management) should not break +older kernel versions. Further, ACPI firmware must also work with the most +recent version of the kernel. + + +Relationship with Device Tree +----------------------------- +ACPI support in drivers and subsystems for ARMv8 should never be mutually +exclusive with DT support at compile time. + +At boot time the kernel will only use one description method depending on +parameters passed from the boot loader (including kernel bootargs). + +Regardless of whether DT or ACPI is used, the kernel must always be capable +of booting with either scheme (in kernels with both schemes enabled at compile +time). + + +Booting using ACPI tables +------------------------- +The only defined method for passing ACPI tables to the kernel on ARMv8 +is via the UEFI system configuration table. Just so it is explicit, this +means that ACPI is only supported on platforms that boot via UEFI. + +When an ARMv8 system boots, it can either have DT information, ACPI tables, +or in some very unusual cases, both. If no command line parameters are used, +the kernel will try to use DT for device enumeration; if there is no DT +present, the kernel will try to use ACPI tables, but only if they are present. +In neither is available, the kernel will not boot. If acpi=force is used +on the command line, the kernel will attempt to use ACPI tables first, but +fall back to DT if there are no ACPI tables present. The basic idea is that +the kernel will not fail to boot unless it absolutely has no other choice. + +Processing of ACPI tables may be disabled by passing acpi=off on the kernel +command line; this is the default behavior. + +In order for the kernel to load and use ACPI tables, the UEFI implementation +MUST set the ACPI_20_TABLE_GUID to point to the RSDP table (the table with +the ACPI signature "RSD PTR "). If this pointer is incorrect and acpi=force +is used, the kernel will disable ACPI and try to use DT to boot instead; the +kernel has, in effect, determined that ACPI tables are not present at that +point. + +If the pointer to the RSDP table is correct, the table will be mapped into +the kernel by the ACPI core, using the address provided by UEFI. + +The ACPI core will then locate and map in all other ACPI tables provided by +using the addresses in the RSDP table to find the XSDT (eXtended System +Description Table). The XSDT in turn provides the addresses to all other +ACPI tables provided by the system firmware; the ACPI core will then traverse +this table and map in the tables listed. + +The ACPI core will ignore any provided RSDT (Root System Description Table). +RSDTs have been deprecated and are ignored on arm64 since they only allow +for 32-bit addresses. + +Further, the ACPI core will only use the 64-bit address fields in the FADT +(Fixed ACPI Description Table). Any 32-bit address fields in the FADT will +be ignored on arm64. + +Hardware reduced mode (see Section 4.1 of the ACPI 6.1 specification) will +be enforced by the ACPI core on arm64. Doing so allows the ACPI core to +run less complex code since it no longer has to provide support for legacy +hardware from other architectures. Any fields that are not to be used for +hardware reduced mode must be set to zero. + +For the ACPI core to operate properly, and in turn provide the information +the kernel needs to configure devices, it expects to find the following +tables (all section numbers refer to the ACPI 6.1 specification): + + - RSDP (Root System Description Pointer), section 5.2.5 + + - XSDT (eXtended System Description Table), section 5.2.8 + + - FADT (Fixed ACPI Description Table), section 5.2.9 + + - DSDT (Differentiated System Description Table), section + 5.2.11.1 + + - MADT (Multiple APIC Description Table), section 5.2.12 + + - GTDT (Generic Timer Description Table), section 5.2.24 + + - If PCI is supported, the MCFG (Memory mapped ConFiGuration + Table), section 5.2.6, specifically Table 5-31. + + - If booting without a console=<device> kernel parameter is + supported, the SPCR (Serial Port Console Redirection table), + section 5.2.6, specifically Table 5-31. + + - If necessary to describe the I/O topology, SMMUs and GIC ITSs, + the IORT (Input Output Remapping Table, section 5.2.6, specifically + Table 5-31). + + - If NUMA is supported, the SRAT (System Resource Affinity Table) + and SLIT (System Locality distance Information Table), sections + 5.2.16 and 5.2.17, respectively. + +If the above tables are not all present, the kernel may or may not be +able to boot properly since it may not be able to configure all of the +devices available. This list of tables is not meant to be all inclusive; +in some environments other tables may be needed (e.g., any of the APEI +tables from section 18) to support specific functionality. + + +ACPI Detection +-------------- +Drivers should determine their probe() type by checking for a null +value for ACPI_HANDLE, or checking .of_node, or other information in +the device structure. This is detailed further in the "Driver +Recommendations" section. + +In non-driver code, if the presence of ACPI needs to be detected at +run time, then check the value of acpi_disabled. If CONFIG_ACPI is not +set, acpi_disabled will always be 1. + + +Device Enumeration +------------------ +Device descriptions in ACPI should use standard recognized ACPI interfaces. +These may contain less information than is typically provided via a Device +Tree description for the same device. This is also one of the reasons that +ACPI can be useful -- the driver takes into account that it may have less +detailed information about the device and uses sensible defaults instead. +If done properly in the driver, the hardware can change and improve over +time without the driver having to change at all. + +Clocks provide an excellent example. In DT, clocks need to be specified +and the drivers need to take them into account. In ACPI, the assumption +is that UEFI will leave the device in a reasonable default state, including +any clock settings. If for some reason the driver needs to change a clock +value, this can be done in an ACPI method; all the driver needs to do is +invoke the method and not concern itself with what the method needs to do +to change the clock. Changing the hardware can then take place over time +by changing what the ACPI method does, and not the driver. + +In DT, the parameters needed by the driver to set up clocks as in the example +above are known as "bindings"; in ACPI, these are known as "Device Properties" +and provided to a driver via the _DSD object. + +ACPI tables are described with a formal language called ASL, the ACPI +Source Language (section 19 of the specification). This means that there +are always multiple ways to describe the same thing -- including device +properties. For example, device properties could use an ASL construct +that looks like this: Name(KEY0, "value0"). An ACPI device driver would +then retrieve the value of the property by evaluating the KEY0 object. +However, using Name() this way has multiple problems: (1) ACPI limits +names ("KEY0") to four characters unlike DT; (2) there is no industry +wide registry that maintains a list of names, minimizing re-use; (3) +there is also no registry for the definition of property values ("value0"), +again making re-use difficult; and (4) how does one maintain backward +compatibility as new hardware comes out? The _DSD method was created +to solve precisely these sorts of problems; Linux drivers should ALWAYS +use the _DSD method for device properties and nothing else. + +The _DSM object (ACPI Section 9.14.1) could also be used for conveying +device properties to a driver. Linux drivers should only expect it to +be used if _DSD cannot represent the data required, and there is no way +to create a new UUID for the _DSD object. Note that there is even less +regulation of the use of _DSM than there is of _DSD. Drivers that depend +on the contents of _DSM objects will be more difficult to maintain over +time because of this; as of this writing, the use of _DSM is the cause +of quite a few firmware problems and is not recommended. + +Drivers should look for device properties in the _DSD object ONLY; the _DSD +object is described in the ACPI specification section 6.2.5, but this only +describes how to define the structure of an object returned via _DSD, and +how specific data structures are defined by specific UUIDs. Linux should +only use the _DSD Device Properties UUID [5]: + + - UUID: daffd814-6eba-4d8c-8a91-bc9bbf4aa301 + + - https://www.uefi.org/sites/default/files/resources/_DSD-device-properties-UUID.pdf + +The UEFI Forum provides a mechanism for registering device properties [4] +so that they may be used across all operating systems supporting ACPI. +Device properties that have not been registered with the UEFI Forum should +not be used. + +Before creating new device properties, check to be sure that they have not +been defined before and either registered in the Linux kernel documentation +as DT bindings, or the UEFI Forum as device properties. While we do not want +to simply move all DT bindings into ACPI device properties, we can learn from +what has been previously defined. + +If it is necessary to define a new device property, or if it makes sense to +synthesize the definition of a binding so it can be used in any firmware, +both DT bindings and ACPI device properties for device drivers have review +processes. Use them both. When the driver itself is submitted for review +to the Linux mailing lists, the device property definitions needed must be +submitted at the same time. A driver that supports ACPI and uses device +properties will not be considered complete without their definitions. Once +the device property has been accepted by the Linux community, it must be +registered with the UEFI Forum [4], which will review it again for consistency +within the registry. This may require iteration. The UEFI Forum, though, +will always be the canonical site for device property definitions. + +It may make sense to provide notice to the UEFI Forum that there is the +intent to register a previously unused device property name as a means of +reserving the name for later use. Other operating system vendors will +also be submitting registration requests and this may help smooth the +process. + +Once registration and review have been completed, the kernel provides an +interface for looking up device properties in a manner independent of +whether DT or ACPI is being used. This API should be used [6]; it can +eliminate some duplication of code paths in driver probing functions and +discourage divergence between DT bindings and ACPI device properties. + + +Programmable Power Control Resources +------------------------------------ +Programmable power control resources include such resources as voltage/current +providers (regulators) and clock sources. + +With ACPI, the kernel clock and regulator framework is not expected to be used +at all. + +The kernel assumes that power control of these resources is represented with +Power Resource Objects (ACPI section 7.1). The ACPI core will then handle +correctly enabling and disabling resources as they are needed. In order to +get that to work, ACPI assumes each device has defined D-states and that these +can be controlled through the optional ACPI methods _PS0, _PS1, _PS2, and _PS3; +in ACPI, _PS0 is the method to invoke to turn a device full on, and _PS3 is for +turning a device full off. + +There are two options for using those Power Resources. They can: + + - be managed in a _PSx method which gets called on entry to power + state Dx. + + - be declared separately as power resources with their own _ON and _OFF + methods. They are then tied back to D-states for a particular device + via _PRx which specifies which power resources a device needs to be on + while in Dx. Kernel then tracks number of devices using a power resource + and calls _ON/_OFF as needed. + +The kernel ACPI code will also assume that the _PSx methods follow the normal +ACPI rules for such methods: + + - If either _PS0 or _PS3 is implemented, then the other method must also + be implemented. + + - If a device requires usage or setup of a power resource when on, the ASL + should organize that it is allocated/enabled using the _PS0 method. + + - Resources allocated or enabled in the _PS0 method should be disabled + or de-allocated in the _PS3 method. + + - Firmware will leave the resources in a reasonable state before handing + over control to the kernel. + +Such code in _PSx methods will of course be very platform specific. But, +this allows the driver to abstract out the interface for operating the device +and avoid having to read special non-standard values from ACPI tables. Further, +abstracting the use of these resources allows the hardware to change over time +without requiring updates to the driver. + + +Clocks +------ +ACPI makes the assumption that clocks are initialized by the firmware -- +UEFI, in this case -- to some working value before control is handed over +to the kernel. This has implications for devices such as UARTs, or SoC-driven +LCD displays, for example. + +When the kernel boots, the clocks are assumed to be set to reasonable +working values. If for some reason the frequency needs to change -- e.g., +throttling for power management -- the device driver should expect that +process to be abstracted out into some ACPI method that can be invoked +(please see the ACPI specification for further recommendations on standard +methods to be expected). The only exceptions to this are CPU clocks where +CPPC provides a much richer interface than ACPI methods. If the clocks +are not set, there is no direct way for Linux to control them. + +If an SoC vendor wants to provide fine-grained control of the system clocks, +they could do so by providing ACPI methods that could be invoked by Linux +drivers. However, this is NOT recommended and Linux drivers should NOT use +such methods, even if they are provided. Such methods are not currently +standardized in the ACPI specification, and using them could tie a kernel +to a very specific SoC, or tie an SoC to a very specific version of the +kernel, both of which we are trying to avoid. + + +Driver Recommendations +---------------------- +DO NOT remove any DT handling when adding ACPI support for a driver. The +same device may be used on many different systems. + +DO try to structure the driver so that it is data-driven. That is, set up +a struct containing internal per-device state based on defaults and whatever +else must be discovered by the driver probe function. Then, have the rest +of the driver operate off of the contents of that struct. Doing so should +allow most divergence between ACPI and DT functionality to be kept local to +the probe function instead of being scattered throughout the driver. For +example:: + + static int device_probe_dt(struct platform_device *pdev) + { + /* DT specific functionality */ + ... + } + + static int device_probe_acpi(struct platform_device *pdev) + { + /* ACPI specific functionality */ + ... + } + + static int device_probe(struct platform_device *pdev) + { + ... + struct device_node node = pdev->dev.of_node; + ... + + if (node) + ret = device_probe_dt(pdev); + else if (ACPI_HANDLE(&pdev->dev)) + ret = device_probe_acpi(pdev); + else + /* other initialization */ + ... + /* Continue with any generic probe operations */ + ... + } + +DO keep the MODULE_DEVICE_TABLE entries together in the driver to make it +clear the different names the driver is probed for, both from DT and from +ACPI:: + + static struct of_device_id virtio_mmio_match[] = { + { .compatible = "virtio,mmio", }, + { } + }; + MODULE_DEVICE_TABLE(of, virtio_mmio_match); + + static const struct acpi_device_id virtio_mmio_acpi_match[] = { + { "LNRO0005", }, + { } + }; + MODULE_DEVICE_TABLE(acpi, virtio_mmio_acpi_match); + + +ASWG +---- +The ACPI specification changes regularly. During the year 2014, for instance, +version 5.1 was released and version 6.0 substantially completed, with most of +the changes being driven by ARM-specific requirements. Proposed changes are +presented and discussed in the ASWG (ACPI Specification Working Group) which +is a part of the UEFI Forum. The current version of the ACPI specification +is 6.1 release in January 2016. + +Participation in this group is open to all UEFI members. Please see +http://www.uefi.org/workinggroup for details on group membership. + +It is the intent of the ARMv8 ACPI kernel code to follow the ACPI specification +as closely as possible, and to only implement functionality that complies with +the released standards from UEFI ASWG. As a practical matter, there will be +vendors that provide bad ACPI tables or violate the standards in some way. +If this is because of errors, quirks and fix-ups may be necessary, but will +be avoided if possible. If there are features missing from ACPI that preclude +it from being used on a platform, ECRs (Engineering Change Requests) should be +submitted to ASWG and go through the normal approval process; for those that +are not UEFI members, many other members of the Linux community are and would +likely be willing to assist in submitting ECRs. + + +Linux Code +---------- +Individual items specific to Linux on ARM, contained in the Linux +source code, are in the list that follows: + +ACPI_OS_NAME + This macro defines the string to be returned when + an ACPI method invokes the _OS method. On ARM64 + systems, this macro will be "Linux" by default. + The command line parameter acpi_os=<string> + can be used to set it to some other value. The + default value for other architectures is "Microsoft + Windows NT", for example. + +ACPI Objects +------------ +Detailed expectations for ACPI tables and object are listed in the file +Documentation/arm64/acpi_object_usage.rst. + + +References +---------- +[0] http://silver.arm.com + document ARM-DEN-0029, or newer: + "Server Base System Architecture", version 2.3, dated 27 Mar 2014 + +[1] http://infocenter.arm.com/help/topic/com.arm.doc.den0044a/Server_Base_Boot_Requirements.pdf + Document ARM-DEN-0044A, or newer: "Server Base Boot Requirements, System + Software on ARM Platforms", dated 16 Aug 2014 + +[2] http://www.secretlab.ca/archives/151, + 10 Jan 2015, Copyright (c) 2015, + Linaro Ltd., written by Grant Likely. + +[3] AMD ACPI for Seattle platform documentation + http://amd-dev.wpengine.netdna-cdn.com/wordpress/media/2012/10/Seattle_ACPI_Guide.pdf + + +[4] http://www.uefi.org/acpi + please see the link for the "ACPI _DSD Device + Property Registry Instructions" + +[5] http://www.uefi.org/acpi + please see the link for the "_DSD (Device + Specific Data) Implementation Guide" + +[6] Kernel code for the unified device + property interface can be found in + include/linux/property.h and drivers/base/property.c. + + +Authors +------- +- Al Stone <al.stone@linaro.org> +- Graeme Gregory <graeme.gregory@linaro.org> +- Hanjun Guo <hanjun.guo@linaro.org> + +- Grant Likely <grant.likely@linaro.org>, for the "Why ACPI on ARM?" section diff --git a/Documentation/arm64/asymmetric-32bit.rst b/Documentation/arm64/asymmetric-32bit.rst new file mode 100644 index 000000000..64a0b505d --- /dev/null +++ b/Documentation/arm64/asymmetric-32bit.rst @@ -0,0 +1,155 @@ +====================== +Asymmetric 32-bit SoCs +====================== + +Author: Will Deacon <will@kernel.org> + +This document describes the impact of asymmetric 32-bit SoCs on the +execution of 32-bit (``AArch32``) applications. + +Date: 2021-05-17 + +Introduction +============ + +Some Armv9 SoCs suffer from a big.LITTLE misfeature where only a subset +of the CPUs are capable of executing 32-bit user applications. On such +a system, Linux by default treats the asymmetry as a "mismatch" and +disables support for both the ``PER_LINUX32`` personality and +``execve(2)`` of 32-bit ELF binaries, with the latter returning +``-ENOEXEC``. If the mismatch is detected during late onlining of a +64-bit-only CPU, then the onlining operation fails and the new CPU is +unavailable for scheduling. + +Surprisingly, these SoCs have been produced with the intention of +running legacy 32-bit binaries. Unsurprisingly, that doesn't work very +well with the default behaviour of Linux. + +It seems inevitable that future SoCs will drop 32-bit support +altogether, so if you're stuck in the unenviable position of needing to +run 32-bit code on one of these transitionary platforms then you would +be wise to consider alternatives such as recompilation, emulation or +retirement. If neither of those options are practical, then read on. + +Enabling kernel support +======================= + +Since the kernel support is not completely transparent to userspace, +allowing 32-bit tasks to run on an asymmetric 32-bit system requires an +explicit "opt-in" and can be enabled by passing the +``allow_mismatched_32bit_el0`` parameter on the kernel command-line. + +For the remainder of this document we will refer to an *asymmetric +system* to mean an asymmetric 32-bit SoC running Linux with this kernel +command-line option enabled. + +Userspace impact +================ + +32-bit tasks running on an asymmetric system behave in mostly the same +way as on a homogeneous system, with a few key differences relating to +CPU affinity. + +sysfs +----- + +The subset of CPUs capable of running 32-bit tasks is described in +``/sys/devices/system/cpu/aarch32_el0`` and is documented further in +``Documentation/ABI/testing/sysfs-devices-system-cpu``. + +**Note:** CPUs are advertised by this file as they are detected and so +late-onlining of 32-bit-capable CPUs can result in the file contents +being modified by the kernel at runtime. Once advertised, CPUs are never +removed from the file. + +``execve(2)`` +------------- + +On a homogeneous system, the CPU affinity of a task is preserved across +``execve(2)``. This is not always possible on an asymmetric system, +specifically when the new program being executed is 32-bit yet the +affinity mask contains 64-bit-only CPUs. In this situation, the kernel +determines the new affinity mask as follows: + + 1. If the 32-bit-capable subset of the affinity mask is not empty, + then the affinity is restricted to that subset and the old affinity + mask is saved. This saved mask is inherited over ``fork(2)`` and + preserved across ``execve(2)`` of 32-bit programs. + + **Note:** This step does not apply to ``SCHED_DEADLINE`` tasks. + See `SCHED_DEADLINE`_. + + 2. Otherwise, the cpuset hierarchy of the task is walked until an + ancestor is found containing at least one 32-bit-capable CPU. The + affinity of the task is then changed to match the 32-bit-capable + subset of the cpuset determined by the walk. + + 3. On failure (i.e. out of memory), the affinity is changed to the set + of all 32-bit-capable CPUs of which the kernel is aware. + +A subsequent ``execve(2)`` of a 64-bit program by the 32-bit task will +invalidate the affinity mask saved in (1) and attempt to restore the CPU +affinity of the task using the saved mask if it was previously valid. +This restoration may fail due to intervening changes to the deadline +policy or cpuset hierarchy, in which case the ``execve(2)`` continues +with the affinity unchanged. + +Calls to ``sched_setaffinity(2)`` for a 32-bit task will consider only +the 32-bit-capable CPUs of the requested affinity mask. On success, the +affinity for the task is updated and any saved mask from a prior +``execve(2)`` is invalidated. + +``SCHED_DEADLINE`` +------------------ + +Explicit admission of a 32-bit deadline task to the default root domain +(e.g. by calling ``sched_setattr(2)``) is rejected on an asymmetric +32-bit system unless admission control is disabled by writing -1 to +``/proc/sys/kernel/sched_rt_runtime_us``. + +``execve(2)`` of a 32-bit program from a 64-bit deadline task will +return ``-ENOEXEC`` if the root domain for the task contains any +64-bit-only CPUs and admission control is enabled. Concurrent offlining +of 32-bit-capable CPUs may still necessitate the procedure described in +`execve(2)`_, in which case step (1) is skipped and a warning is +emitted on the console. + +**Note:** It is recommended that a set of 32-bit-capable CPUs are placed +into a separate root domain if ``SCHED_DEADLINE`` is to be used with +32-bit tasks on an asymmetric system. Failure to do so is likely to +result in missed deadlines. + +Cpusets +------- + +The affinity of a 32-bit task on an asymmetric system may include CPUs +that are not explicitly allowed by the cpuset to which it is attached. +This can occur as a result of the following two situations: + + - A 64-bit task attached to a cpuset which allows only 64-bit CPUs + executes a 32-bit program. + + - All of the 32-bit-capable CPUs allowed by a cpuset containing a + 32-bit task are offlined. + +In both of these cases, the new affinity is calculated according to step +(2) of the process described in `execve(2)`_ and the cpuset hierarchy is +unchanged irrespective of the cgroup version. + +CPU hotplug +----------- + +On an asymmetric system, the first detected 32-bit-capable CPU is +prevented from being offlined by userspace and any such attempt will +return ``-EPERM``. Note that suspend is still permitted even if the +primary CPU (i.e. CPU 0) is 64-bit-only. + +KVM +--- + +Although KVM will not advertise 32-bit EL0 support to any vCPUs on an +asymmetric system, a broken guest at EL1 could still attempt to execute +32-bit code at EL0. In this case, an exit from a vCPU thread in 32-bit +mode will return to host userspace with an ``exit_reason`` of +``KVM_EXIT_FAIL_ENTRY`` and will remain non-runnable until successfully +re-initialised by a subsequent ``KVM_ARM_VCPU_INIT`` operation. diff --git a/Documentation/arm64/booting.rst b/Documentation/arm64/booting.rst new file mode 100644 index 000000000..8c324ad63 --- /dev/null +++ b/Documentation/arm64/booting.rst @@ -0,0 +1,420 @@ +===================== +Booting AArch64 Linux +===================== + +Author: Will Deacon <will.deacon@arm.com> + +Date : 07 September 2012 + +This document is based on the ARM booting document by Russell King and +is relevant to all public releases of the AArch64 Linux kernel. + +The AArch64 exception model is made up of a number of exception levels +(EL0 - EL3), with EL0, EL1 and EL2 having a secure and a non-secure +counterpart. EL2 is the hypervisor level, EL3 is the highest priority +level and exists only in secure mode. Both are architecturally optional. + +For the purposes of this document, we will use the term `boot loader` +simply to define all software that executes on the CPU(s) before control +is passed to the Linux kernel. This may include secure monitor and +hypervisor code, or it may just be a handful of instructions for +preparing a minimal boot environment. + +Essentially, the boot loader should provide (as a minimum) the +following: + +1. Setup and initialise the RAM +2. Setup the device tree +3. Decompress the kernel image +4. Call the kernel image + + +1. Setup and initialise RAM +--------------------------- + +Requirement: MANDATORY + +The boot loader is expected to find and initialise all RAM that the +kernel will use for volatile data storage in the system. It performs +this in a machine dependent manner. (It may use internal algorithms +to automatically locate and size all RAM, or it may use knowledge of +the RAM in the machine, or any other method the boot loader designer +sees fit.) + + +2. Setup the device tree +------------------------- + +Requirement: MANDATORY + +The device tree blob (dtb) must be placed on an 8-byte boundary and must +not exceed 2 megabytes in size. Since the dtb will be mapped cacheable +using blocks of up to 2 megabytes in size, it must not be placed within +any 2M region which must be mapped with any specific attributes. + +NOTE: versions prior to v4.2 also require that the DTB be placed within +the 512 MB region starting at text_offset bytes below the kernel Image. + +3. Decompress the kernel image +------------------------------ + +Requirement: OPTIONAL + +The AArch64 kernel does not currently provide a decompressor and +therefore requires decompression (gzip etc.) to be performed by the boot +loader if a compressed Image target (e.g. Image.gz) is used. For +bootloaders that do not implement this requirement, the uncompressed +Image target is available instead. + + +4. Call the kernel image +------------------------ + +Requirement: MANDATORY + +The decompressed kernel image contains a 64-byte header as follows:: + + u32 code0; /* Executable code */ + u32 code1; /* Executable code */ + u64 text_offset; /* Image load offset, little endian */ + u64 image_size; /* Effective Image size, little endian */ + u64 flags; /* kernel flags, little endian */ + u64 res2 = 0; /* reserved */ + u64 res3 = 0; /* reserved */ + u64 res4 = 0; /* reserved */ + u32 magic = 0x644d5241; /* Magic number, little endian, "ARM\x64" */ + u32 res5; /* reserved (used for PE COFF offset) */ + + +Header notes: + +- As of v3.17, all fields are little endian unless stated otherwise. + +- code0/code1 are responsible for branching to stext. + +- when booting through EFI, code0/code1 are initially skipped. + res5 is an offset to the PE header and the PE header has the EFI + entry point (efi_stub_entry). When the stub has done its work, it + jumps to code0 to resume the normal boot process. + +- Prior to v3.17, the endianness of text_offset was not specified. In + these cases image_size is zero and text_offset is 0x80000 in the + endianness of the kernel. Where image_size is non-zero image_size is + little-endian and must be respected. Where image_size is zero, + text_offset can be assumed to be 0x80000. + +- The flags field (introduced in v3.17) is a little-endian 64-bit field + composed as follows: + + ============= =============================================================== + Bit 0 Kernel endianness. 1 if BE, 0 if LE. + Bit 1-2 Kernel Page size. + + * 0 - Unspecified. + * 1 - 4K + * 2 - 16K + * 3 - 64K + Bit 3 Kernel physical placement + + 0 + 2MB aligned base should be as close as possible + to the base of DRAM, since memory below it is not + accessible via the linear mapping + 1 + 2MB aligned base may be anywhere in physical + memory + Bits 4-63 Reserved. + ============= =============================================================== + +- When image_size is zero, a bootloader should attempt to keep as much + memory as possible free for use by the kernel immediately after the + end of the kernel image. The amount of space required will vary + depending on selected features, and is effectively unbound. + +The Image must be placed text_offset bytes from a 2MB aligned base +address anywhere in usable system RAM and called there. The region +between the 2 MB aligned base address and the start of the image has no +special significance to the kernel, and may be used for other purposes. +At least image_size bytes from the start of the image must be free for +use by the kernel. +NOTE: versions prior to v4.6 cannot make use of memory below the +physical offset of the Image so it is recommended that the Image be +placed as close as possible to the start of system RAM. + +If an initrd/initramfs is passed to the kernel at boot, it must reside +entirely within a 1 GB aligned physical memory window of up to 32 GB in +size that fully covers the kernel Image as well. + +Any memory described to the kernel (even that below the start of the +image) which is not marked as reserved from the kernel (e.g., with a +memreserve region in the device tree) will be considered as available to +the kernel. + +Before jumping into the kernel, the following conditions must be met: + +- Quiesce all DMA capable devices so that memory does not get + corrupted by bogus network packets or disk data. This will save + you many hours of debug. + +- Primary CPU general-purpose register settings: + + - x0 = physical address of device tree blob (dtb) in system RAM. + - x1 = 0 (reserved for future use) + - x2 = 0 (reserved for future use) + - x3 = 0 (reserved for future use) + +- CPU mode + + All forms of interrupts must be masked in PSTATE.DAIF (Debug, SError, + IRQ and FIQ). + The CPU must be in non-secure state, either in EL2 (RECOMMENDED in order + to have access to the virtualisation extensions), or in EL1. + +- Caches, MMUs + + The MMU must be off. + + The instruction cache may be on or off, and must not hold any stale + entries corresponding to the loaded kernel image. + + The address range corresponding to the loaded kernel image must be + cleaned to the PoC. In the presence of a system cache or other + coherent masters with caches enabled, this will typically require + cache maintenance by VA rather than set/way operations. + System caches which respect the architected cache maintenance by VA + operations must be configured and may be enabled. + System caches which do not respect architected cache maintenance by VA + operations (not recommended) must be configured and disabled. + +- Architected timers + + CNTFRQ must be programmed with the timer frequency and CNTVOFF must + be programmed with a consistent value on all CPUs. If entering the + kernel at EL1, CNTHCTL_EL2 must have EL1PCTEN (bit 0) set where + available. + +- Coherency + + All CPUs to be booted by the kernel must be part of the same coherency + domain on entry to the kernel. This may require IMPLEMENTATION DEFINED + initialisation to enable the receiving of maintenance operations on + each CPU. + +- System registers + + All writable architected system registers at or below the exception + level where the kernel image will be entered must be initialised by + software at a higher exception level to prevent execution in an UNKNOWN + state. + + For all systems: + - If EL3 is present: + + - SCR_EL3.FIQ must have the same value across all CPUs the kernel is + executing on. + - The value of SCR_EL3.FIQ must be the same as the one present at boot + time whenever the kernel is executing. + + - If EL3 is present and the kernel is entered at EL2: + + - SCR_EL3.HCE (bit 8) must be initialised to 0b1. + + For systems with a GICv3 interrupt controller to be used in v3 mode: + - If EL3 is present: + + - ICC_SRE_EL3.Enable (bit 3) must be initialiased to 0b1. + - ICC_SRE_EL3.SRE (bit 0) must be initialised to 0b1. + - ICC_CTLR_EL3.PMHE (bit 6) must be set to the same value across + all CPUs the kernel is executing on, and must stay constant + for the lifetime of the kernel. + + - If the kernel is entered at EL1: + + - ICC.SRE_EL2.Enable (bit 3) must be initialised to 0b1 + - ICC_SRE_EL2.SRE (bit 0) must be initialised to 0b1. + + - The DT or ACPI tables must describe a GICv3 interrupt controller. + + For systems with a GICv3 interrupt controller to be used in + compatibility (v2) mode: + + - If EL3 is present: + + ICC_SRE_EL3.SRE (bit 0) must be initialised to 0b0. + + - If the kernel is entered at EL1: + + ICC_SRE_EL2.SRE (bit 0) must be initialised to 0b0. + + - The DT or ACPI tables must describe a GICv2 interrupt controller. + + For CPUs with pointer authentication functionality: + + - If EL3 is present: + + - SCR_EL3.APK (bit 16) must be initialised to 0b1 + - SCR_EL3.API (bit 17) must be initialised to 0b1 + + - If the kernel is entered at EL1: + + - HCR_EL2.APK (bit 40) must be initialised to 0b1 + - HCR_EL2.API (bit 41) must be initialised to 0b1 + + For CPUs with Activity Monitors Unit v1 (AMUv1) extension present: + + - If EL3 is present: + + - CPTR_EL3.TAM (bit 30) must be initialised to 0b0 + - CPTR_EL2.TAM (bit 30) must be initialised to 0b0 + - AMCNTENSET0_EL0 must be initialised to 0b1111 + - AMCNTENSET1_EL0 must be initialised to a platform specific value + having 0b1 set for the corresponding bit for each of the auxiliary + counters present. + + - If the kernel is entered at EL1: + + - AMCNTENSET0_EL0 must be initialised to 0b1111 + - AMCNTENSET1_EL0 must be initialised to a platform specific value + having 0b1 set for the corresponding bit for each of the auxiliary + counters present. + + For CPUs with the Fine Grained Traps (FEAT_FGT) extension present: + + - If EL3 is present and the kernel is entered at EL2: + + - SCR_EL3.FGTEn (bit 27) must be initialised to 0b1. + + For CPUs with support for HCRX_EL2 (FEAT_HCX) present: + + - If EL3 is present and the kernel is entered at EL2: + + - SCR_EL3.HXEn (bit 38) must be initialised to 0b1. + + For CPUs with Advanced SIMD and floating point support: + + - If EL3 is present: + + - CPTR_EL3.TFP (bit 10) must be initialised to 0b0. + + - If EL2 is present and the kernel is entered at EL1: + + - CPTR_EL2.TFP (bit 10) must be initialised to 0b0. + + For CPUs with the Scalable Vector Extension (FEAT_SVE) present: + + - if EL3 is present: + + - CPTR_EL3.EZ (bit 8) must be initialised to 0b1. + + - ZCR_EL3.LEN must be initialised to the same value for all CPUs the + kernel is executed on. + + - If the kernel is entered at EL1 and EL2 is present: + + - CPTR_EL2.TZ (bit 8) must be initialised to 0b0. + + - CPTR_EL2.ZEN (bits 17:16) must be initialised to 0b11. + + - ZCR_EL2.LEN must be initialised to the same value for all CPUs the + kernel will execute on. + + For CPUs with the Scalable Matrix Extension (FEAT_SME): + + - If EL3 is present: + + - CPTR_EL3.ESM (bit 12) must be initialised to 0b1. + + - SCR_EL3.EnTP2 (bit 41) must be initialised to 0b1. + + - SMCR_EL3.LEN must be initialised to the same value for all CPUs the + kernel will execute on. + + - If the kernel is entered at EL1 and EL2 is present: + + - CPTR_EL2.TSM (bit 12) must be initialised to 0b0. + + - CPTR_EL2.SMEN (bits 25:24) must be initialised to 0b11. + + - SCTLR_EL2.EnTP2 (bit 60) must be initialised to 0b1. + + - SMCR_EL2.LEN must be initialised to the same value for all CPUs the + kernel will execute on. + + - HWFGRTR_EL2.nTPIDR2_EL0 (bit 55) must be initialised to 0b01. + + - HWFGWTR_EL2.nTPIDR2_EL0 (bit 55) must be initialised to 0b01. + + - HWFGRTR_EL2.nSMPRI_EL1 (bit 54) must be initialised to 0b01. + + - HWFGWTR_EL2.nSMPRI_EL1 (bit 54) must be initialised to 0b01. + + For CPUs with the Scalable Matrix Extension FA64 feature (FEAT_SME_FA64) + + - If EL3 is present: + + - SMCR_EL3.FA64 (bit 31) must be initialised to 0b1. + + - If the kernel is entered at EL1 and EL2 is present: + + - SMCR_EL2.FA64 (bit 31) must be initialised to 0b1. + + For CPUs with the Memory Tagging Extension feature (FEAT_MTE2): + + - If EL3 is present: + + - SCR_EL3.ATA (bit 26) must be initialised to 0b1. + + - If the kernel is entered at EL1 and EL2 is present: + + - HCR_EL2.ATA (bit 56) must be initialised to 0b1. + +The requirements described above for CPU mode, caches, MMUs, architected +timers, coherency and system registers apply to all CPUs. All CPUs must +enter the kernel in the same exception level. Where the values documented +disable traps it is permissible for these traps to be enabled so long as +those traps are handled transparently by higher exception levels as though +the values documented were set. + +The boot loader is expected to enter the kernel on each CPU in the +following manner: + +- The primary CPU must jump directly to the first instruction of the + kernel image. The device tree blob passed by this CPU must contain + an 'enable-method' property for each cpu node. The supported + enable-methods are described below. + + It is expected that the bootloader will generate these device tree + properties and insert them into the blob prior to kernel entry. + +- CPUs with a "spin-table" enable-method must have a 'cpu-release-addr' + property in their cpu node. This property identifies a + naturally-aligned 64-bit zero-initalised memory location. + + These CPUs should spin outside of the kernel in a reserved area of + memory (communicated to the kernel by a /memreserve/ region in the + device tree) polling their cpu-release-addr location, which must be + contained in the reserved region. A wfe instruction may be inserted + to reduce the overhead of the busy-loop and a sev will be issued by + the primary CPU. When a read of the location pointed to by the + cpu-release-addr returns a non-zero value, the CPU must jump to this + value. The value will be written as a single 64-bit little-endian + value, so CPUs must convert the read value to their native endianness + before jumping to it. + +- CPUs with a "psci" enable method should remain outside of + the kernel (i.e. outside of the regions of memory described to the + kernel in the memory node, or in a reserved area of memory described + to the kernel by a /memreserve/ region in the device tree). The + kernel will issue CPU_ON calls as described in ARM document number ARM + DEN 0022A ("Power State Coordination Interface System Software on ARM + processors") to bring CPUs into the kernel. + + The device tree should contain a 'psci' node, as described in + Documentation/devicetree/bindings/arm/psci.yaml. + +- Secondary CPU general-purpose register settings + + - x0 = 0 (reserved for future use) + - x1 = 0 (reserved for future use) + - x2 = 0 (reserved for future use) + - x3 = 0 (reserved for future use) diff --git a/Documentation/arm64/cpu-feature-registers.rst b/Documentation/arm64/cpu-feature-registers.rst new file mode 100644 index 000000000..c7adc7897 --- /dev/null +++ b/Documentation/arm64/cpu-feature-registers.rst @@ -0,0 +1,400 @@ +=========================== +ARM64 CPU Feature Registers +=========================== + +Author: Suzuki K Poulose <suzuki.poulose@arm.com> + + +This file describes the ABI for exporting the AArch64 CPU ID/feature +registers to userspace. The availability of this ABI is advertised +via the HWCAP_CPUID in HWCAPs. + +1. Motivation +------------- + +The ARM architecture defines a set of feature registers, which describe +the capabilities of the CPU/system. Access to these system registers is +restricted from EL0 and there is no reliable way for an application to +extract this information to make better decisions at runtime. There is +limited information available to the application via HWCAPs, however +there are some issues with their usage. + + a) Any change to the HWCAPs requires an update to userspace (e.g libc) + to detect the new changes, which can take a long time to appear in + distributions. Exposing the registers allows applications to get the + information without requiring updates to the toolchains. + + b) Access to HWCAPs is sometimes limited (e.g prior to libc, or + when ld is initialised at startup time). + + c) HWCAPs cannot represent non-boolean information effectively. The + architecture defines a canonical format for representing features + in the ID registers; this is well defined and is capable of + representing all valid architecture variations. + + +2. Requirements +--------------- + + a) Safety: + + Applications should be able to use the information provided by the + infrastructure to run safely across the system. This has greater + implications on a system with heterogeneous CPUs. + The infrastructure exports a value that is safe across all the + available CPU on the system. + + e.g, If at least one CPU doesn't implement CRC32 instructions, while + others do, we should report that the CRC32 is not implemented. + Otherwise an application could crash when scheduled on the CPU + which doesn't support CRC32. + + b) Security: + + Applications should only be able to receive information that is + relevant to the normal operation in userspace. Hence, some of the + fields are masked out(i.e, made invisible) and their values are set to + indicate the feature is 'not supported'. See Section 4 for the list + of visible features. Also, the kernel may manipulate the fields + based on what it supports. e.g, If FP is not supported by the + kernel, the values could indicate that the FP is not available + (even when the CPU provides it). + + c) Implementation Defined Features + + The infrastructure doesn't expose any register which is + IMPLEMENTATION DEFINED as per ARMv8-A Architecture. + + d) CPU Identification: + + MIDR_EL1 is exposed to help identify the processor. On a + heterogeneous system, this could be racy (just like getcpu()). The + process could be migrated to another CPU by the time it uses the + register value, unless the CPU affinity is set. Hence, there is no + guarantee that the value reflects the processor that it is + currently executing on. The REVIDR is not exposed due to this + constraint, as REVIDR makes sense only in conjunction with the + MIDR. Alternately, MIDR_EL1 and REVIDR_EL1 are exposed via sysfs + at:: + + /sys/devices/system/cpu/cpu$ID/regs/identification/ + \- midr + \- revidr + +3. Implementation +-------------------- + +The infrastructure is built on the emulation of the 'MRS' instruction. +Accessing a restricted system register from an application generates an +exception and ends up in SIGILL being delivered to the process. +The infrastructure hooks into the exception handler and emulates the +operation if the source belongs to the supported system register space. + +The infrastructure emulates only the following system register space:: + + Op0=3, Op1=0, CRn=0, CRm=0,2,3,4,5,6,7 + +(See Table C5-6 'System instruction encodings for non-Debug System +register accesses' in ARMv8 ARM DDI 0487A.h, for the list of +registers). + +The following rules are applied to the value returned by the +infrastructure: + + a) The value of an 'IMPLEMENTATION DEFINED' field is set to 0. + b) The value of a reserved field is populated with the reserved + value as defined by the architecture. + c) The value of a 'visible' field holds the system wide safe value + for the particular feature (except for MIDR_EL1, see section 4). + d) All other fields (i.e, invisible fields) are set to indicate + the feature is missing (as defined by the architecture). + +4. List of registers with visible features +------------------------------------------- + + 1) ID_AA64ISAR0_EL1 - Instruction Set Attribute Register 0 + + +------------------------------+---------+---------+ + | Name | bits | visible | + +------------------------------+---------+---------+ + | RNDR | [63-60] | y | + +------------------------------+---------+---------+ + | TS | [55-52] | y | + +------------------------------+---------+---------+ + | FHM | [51-48] | y | + +------------------------------+---------+---------+ + | DP | [47-44] | y | + +------------------------------+---------+---------+ + | SM4 | [43-40] | y | + +------------------------------+---------+---------+ + | SM3 | [39-36] | y | + +------------------------------+---------+---------+ + | SHA3 | [35-32] | y | + +------------------------------+---------+---------+ + | RDM | [31-28] | y | + +------------------------------+---------+---------+ + | ATOMICS | [23-20] | y | + +------------------------------+---------+---------+ + | CRC32 | [19-16] | y | + +------------------------------+---------+---------+ + | SHA2 | [15-12] | y | + +------------------------------+---------+---------+ + | SHA1 | [11-8] | y | + +------------------------------+---------+---------+ + | AES | [7-4] | y | + +------------------------------+---------+---------+ + + + 2) ID_AA64PFR0_EL1 - Processor Feature Register 0 + + +------------------------------+---------+---------+ + | Name | bits | visible | + +------------------------------+---------+---------+ + | DIT | [51-48] | y | + +------------------------------+---------+---------+ + | SVE | [35-32] | y | + +------------------------------+---------+---------+ + | GIC | [27-24] | n | + +------------------------------+---------+---------+ + | AdvSIMD | [23-20] | y | + +------------------------------+---------+---------+ + | FP | [19-16] | y | + +------------------------------+---------+---------+ + | EL3 | [15-12] | n | + +------------------------------+---------+---------+ + | EL2 | [11-8] | n | + +------------------------------+---------+---------+ + | EL1 | [7-4] | n | + +------------------------------+---------+---------+ + | EL0 | [3-0] | n | + +------------------------------+---------+---------+ + + + 3) ID_AA64PFR1_EL1 - Processor Feature Register 1 + + +------------------------------+---------+---------+ + | Name | bits | visible | + +------------------------------+---------+---------+ + | MTE | [11-8] | y | + +------------------------------+---------+---------+ + | SSBS | [7-4] | y | + +------------------------------+---------+---------+ + | BT | [3-0] | y | + +------------------------------+---------+---------+ + + + 4) MIDR_EL1 - Main ID Register + + +------------------------------+---------+---------+ + | Name | bits | visible | + +------------------------------+---------+---------+ + | Implementer | [31-24] | y | + +------------------------------+---------+---------+ + | Variant | [23-20] | y | + +------------------------------+---------+---------+ + | Architecture | [19-16] | y | + +------------------------------+---------+---------+ + | PartNum | [15-4] | y | + +------------------------------+---------+---------+ + | Revision | [3-0] | y | + +------------------------------+---------+---------+ + + NOTE: The 'visible' fields of MIDR_EL1 will contain the value + as available on the CPU where it is fetched and is not a system + wide safe value. + + 5) ID_AA64ISAR1_EL1 - Instruction set attribute register 1 + + +------------------------------+---------+---------+ + | Name | bits | visible | + +------------------------------+---------+---------+ + | I8MM | [55-52] | y | + +------------------------------+---------+---------+ + | DGH | [51-48] | y | + +------------------------------+---------+---------+ + | BF16 | [47-44] | y | + +------------------------------+---------+---------+ + | SB | [39-36] | y | + +------------------------------+---------+---------+ + | FRINTTS | [35-32] | y | + +------------------------------+---------+---------+ + | GPI | [31-28] | y | + +------------------------------+---------+---------+ + | GPA | [27-24] | y | + +------------------------------+---------+---------+ + | LRCPC | [23-20] | y | + +------------------------------+---------+---------+ + | FCMA | [19-16] | y | + +------------------------------+---------+---------+ + | JSCVT | [15-12] | y | + +------------------------------+---------+---------+ + | API | [11-8] | y | + +------------------------------+---------+---------+ + | APA | [7-4] | y | + +------------------------------+---------+---------+ + | DPB | [3-0] | y | + +------------------------------+---------+---------+ + + 6) ID_AA64MMFR0_EL1 - Memory model feature register 0 + + +------------------------------+---------+---------+ + | Name | bits | visible | + +------------------------------+---------+---------+ + | ECV | [63-60] | y | + +------------------------------+---------+---------+ + + 7) ID_AA64MMFR2_EL1 - Memory model feature register 2 + + +------------------------------+---------+---------+ + | Name | bits | visible | + +------------------------------+---------+---------+ + | AT | [35-32] | y | + +------------------------------+---------+---------+ + + 8) ID_AA64ZFR0_EL1 - SVE feature ID register 0 + + +------------------------------+---------+---------+ + | Name | bits | visible | + +------------------------------+---------+---------+ + | F64MM | [59-56] | y | + +------------------------------+---------+---------+ + | F32MM | [55-52] | y | + +------------------------------+---------+---------+ + | I8MM | [47-44] | y | + +------------------------------+---------+---------+ + | SM4 | [43-40] | y | + +------------------------------+---------+---------+ + | SHA3 | [35-32] | y | + +------------------------------+---------+---------+ + | BF16 | [23-20] | y | + +------------------------------+---------+---------+ + | BitPerm | [19-16] | y | + +------------------------------+---------+---------+ + | AES | [7-4] | y | + +------------------------------+---------+---------+ + | SVEVer | [3-0] | y | + +------------------------------+---------+---------+ + + 8) ID_AA64MMFR1_EL1 - Memory model feature register 1 + + +------------------------------+---------+---------+ + | Name | bits | visible | + +------------------------------+---------+---------+ + | AFP | [47-44] | y | + +------------------------------+---------+---------+ + + 9) ID_AA64ISAR2_EL1 - Instruction set attribute register 2 + + +------------------------------+---------+---------+ + | Name | bits | visible | + +------------------------------+---------+---------+ + | RPRES | [7-4] | y | + +------------------------------+---------+---------+ + | WFXT | [3-0] | y | + +------------------------------+---------+---------+ + + 10) MVFR0_EL1 - AArch32 Media and VFP Feature Register 0 + + +------------------------------+---------+---------+ + | Name | bits | visible | + +------------------------------+---------+---------+ + | FPDP | [11-8] | y | + +------------------------------+---------+---------+ + + 11) MVFR1_EL1 - AArch32 Media and VFP Feature Register 1 + + +------------------------------+---------+---------+ + | Name | bits | visible | + +------------------------------+---------+---------+ + | SIMDFMAC | [31-28] | y | + +------------------------------+---------+---------+ + | SIMDSP | [19-16] | y | + +------------------------------+---------+---------+ + | SIMDInt | [15-12] | y | + +------------------------------+---------+---------+ + | SIMDLS | [11-8] | y | + +------------------------------+---------+---------+ + + 12) ID_ISAR5_EL1 - AArch32 Instruction Set Attribute Register 5 + + +------------------------------+---------+---------+ + | Name | bits | visible | + +------------------------------+---------+---------+ + | CRC32 | [19-16] | y | + +------------------------------+---------+---------+ + | SHA2 | [15-12] | y | + +------------------------------+---------+---------+ + | SHA1 | [11-8] | y | + +------------------------------+---------+---------+ + | AES | [7-4] | y | + +------------------------------+---------+---------+ + + +Appendix I: Example +------------------- + +:: + + /* + * Sample program to demonstrate the MRS emulation ABI. + * + * Copyright (C) 2015-2016, ARM Ltd + * + * Author: Suzuki K Poulose <suzuki.poulose@arm.com> + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + */ + + #include <asm/hwcap.h> + #include <stdio.h> + #include <sys/auxv.h> + + #define get_cpu_ftr(id) ({ \ + unsigned long __val; \ + asm("mrs %0, "#id : "=r" (__val)); \ + printf("%-20s: 0x%016lx\n", #id, __val); \ + }) + + int main(void) + { + + if (!(getauxval(AT_HWCAP) & HWCAP_CPUID)) { + fputs("CPUID registers unavailable\n", stderr); + return 1; + } + + get_cpu_ftr(ID_AA64ISAR0_EL1); + get_cpu_ftr(ID_AA64ISAR1_EL1); + get_cpu_ftr(ID_AA64MMFR0_EL1); + get_cpu_ftr(ID_AA64MMFR1_EL1); + get_cpu_ftr(ID_AA64PFR0_EL1); + get_cpu_ftr(ID_AA64PFR1_EL1); + get_cpu_ftr(ID_AA64DFR0_EL1); + get_cpu_ftr(ID_AA64DFR1_EL1); + + get_cpu_ftr(MIDR_EL1); + get_cpu_ftr(MPIDR_EL1); + get_cpu_ftr(REVIDR_EL1); + + #if 0 + /* Unexposed register access causes SIGILL */ + get_cpu_ftr(ID_MMFR0_EL1); + #endif + + return 0; + } diff --git a/Documentation/arm64/elf_hwcaps.rst b/Documentation/arm64/elf_hwcaps.rst new file mode 100644 index 000000000..bb34287c8 --- /dev/null +++ b/Documentation/arm64/elf_hwcaps.rst @@ -0,0 +1,282 @@ +.. _elf_hwcaps_index: + +================ +ARM64 ELF hwcaps +================ + +This document describes the usage and semantics of the arm64 ELF hwcaps. + + +1. Introduction +--------------- + +Some hardware or software features are only available on some CPU +implementations, and/or with certain kernel configurations, but have no +architected discovery mechanism available to userspace code at EL0. The +kernel exposes the presence of these features to userspace through a set +of flags called hwcaps, exposed in the auxilliary vector. + +Userspace software can test for features by acquiring the AT_HWCAP or +AT_HWCAP2 entry of the auxiliary vector, and testing whether the relevant +flags are set, e.g.:: + + bool floating_point_is_present(void) + { + unsigned long hwcaps = getauxval(AT_HWCAP); + if (hwcaps & HWCAP_FP) + return true; + + return false; + } + +Where software relies on a feature described by a hwcap, it should check +the relevant hwcap flag to verify that the feature is present before +attempting to make use of the feature. + +Features cannot be probed reliably through other means. When a feature +is not available, attempting to use it may result in unpredictable +behaviour, and is not guaranteed to result in any reliable indication +that the feature is unavailable, such as a SIGILL. + + +2. Interpretation of hwcaps +--------------------------- + +The majority of hwcaps are intended to indicate the presence of features +which are described by architected ID registers inaccessible to +userspace code at EL0. These hwcaps are defined in terms of ID register +fields, and should be interpreted with reference to the definition of +these fields in the ARM Architecture Reference Manual (ARM ARM). + +Such hwcaps are described below in the form:: + + Functionality implied by idreg.field == val. + +Such hwcaps indicate the availability of functionality that the ARM ARM +defines as being present when idreg.field has value val, but do not +indicate that idreg.field is precisely equal to val, nor do they +indicate the absence of functionality implied by other values of +idreg.field. + +Other hwcaps may indicate the presence of features which cannot be +described by ID registers alone. These may be described without +reference to ID registers, and may refer to other documentation. + + +3. The hwcaps exposed in AT_HWCAP +--------------------------------- + +HWCAP_FP + Functionality implied by ID_AA64PFR0_EL1.FP == 0b0000. + +HWCAP_ASIMD + Functionality implied by ID_AA64PFR0_EL1.AdvSIMD == 0b0000. + +HWCAP_EVTSTRM + The generic timer is configured to generate events at a frequency of + approximately 10KHz. + +HWCAP_AES + Functionality implied by ID_AA64ISAR0_EL1.AES == 0b0001. + +HWCAP_PMULL + Functionality implied by ID_AA64ISAR0_EL1.AES == 0b0010. + +HWCAP_SHA1 + Functionality implied by ID_AA64ISAR0_EL1.SHA1 == 0b0001. + +HWCAP_SHA2 + Functionality implied by ID_AA64ISAR0_EL1.SHA2 == 0b0001. + +HWCAP_CRC32 + Functionality implied by ID_AA64ISAR0_EL1.CRC32 == 0b0001. + +HWCAP_ATOMICS + Functionality implied by ID_AA64ISAR0_EL1.Atomic == 0b0010. + +HWCAP_FPHP + Functionality implied by ID_AA64PFR0_EL1.FP == 0b0001. + +HWCAP_ASIMDHP + Functionality implied by ID_AA64PFR0_EL1.AdvSIMD == 0b0001. + +HWCAP_CPUID + EL0 access to certain ID registers is available, to the extent + described by Documentation/arm64/cpu-feature-registers.rst. + + These ID registers may imply the availability of features. + +HWCAP_ASIMDRDM + Functionality implied by ID_AA64ISAR0_EL1.RDM == 0b0001. + +HWCAP_JSCVT + Functionality implied by ID_AA64ISAR1_EL1.JSCVT == 0b0001. + +HWCAP_FCMA + Functionality implied by ID_AA64ISAR1_EL1.FCMA == 0b0001. + +HWCAP_LRCPC + Functionality implied by ID_AA64ISAR1_EL1.LRCPC == 0b0001. + +HWCAP_DCPOP + Functionality implied by ID_AA64ISAR1_EL1.DPB == 0b0001. + +HWCAP_SHA3 + Functionality implied by ID_AA64ISAR0_EL1.SHA3 == 0b0001. + +HWCAP_SM3 + Functionality implied by ID_AA64ISAR0_EL1.SM3 == 0b0001. + +HWCAP_SM4 + Functionality implied by ID_AA64ISAR0_EL1.SM4 == 0b0001. + +HWCAP_ASIMDDP + Functionality implied by ID_AA64ISAR0_EL1.DP == 0b0001. + +HWCAP_SHA512 + Functionality implied by ID_AA64ISAR0_EL1.SHA2 == 0b0010. + +HWCAP_SVE + Functionality implied by ID_AA64PFR0_EL1.SVE == 0b0001. + +HWCAP_ASIMDFHM + Functionality implied by ID_AA64ISAR0_EL1.FHM == 0b0001. + +HWCAP_DIT + Functionality implied by ID_AA64PFR0_EL1.DIT == 0b0001. + +HWCAP_USCAT + Functionality implied by ID_AA64MMFR2_EL1.AT == 0b0001. + +HWCAP_ILRCPC + Functionality implied by ID_AA64ISAR1_EL1.LRCPC == 0b0010. + +HWCAP_FLAGM + Functionality implied by ID_AA64ISAR0_EL1.TS == 0b0001. + +HWCAP_SSBS + Functionality implied by ID_AA64PFR1_EL1.SSBS == 0b0010. + +HWCAP_SB + Functionality implied by ID_AA64ISAR1_EL1.SB == 0b0001. + +HWCAP_PACA + Functionality implied by ID_AA64ISAR1_EL1.APA == 0b0001 or + ID_AA64ISAR1_EL1.API == 0b0001, as described by + Documentation/arm64/pointer-authentication.rst. + +HWCAP_PACG + Functionality implied by ID_AA64ISAR1_EL1.GPA == 0b0001 or + ID_AA64ISAR1_EL1.GPI == 0b0001, as described by + Documentation/arm64/pointer-authentication.rst. + +HWCAP2_DCPODP + Functionality implied by ID_AA64ISAR1_EL1.DPB == 0b0010. + +HWCAP2_SVE2 + Functionality implied by ID_AA64ZFR0_EL1.SVEVer == 0b0001. + +HWCAP2_SVEAES + Functionality implied by ID_AA64ZFR0_EL1.AES == 0b0001. + +HWCAP2_SVEPMULL + Functionality implied by ID_AA64ZFR0_EL1.AES == 0b0010. + +HWCAP2_SVEBITPERM + Functionality implied by ID_AA64ZFR0_EL1.BitPerm == 0b0001. + +HWCAP2_SVESHA3 + Functionality implied by ID_AA64ZFR0_EL1.SHA3 == 0b0001. + +HWCAP2_SVESM4 + Functionality implied by ID_AA64ZFR0_EL1.SM4 == 0b0001. + +HWCAP2_FLAGM2 + Functionality implied by ID_AA64ISAR0_EL1.TS == 0b0010. + +HWCAP2_FRINT + Functionality implied by ID_AA64ISAR1_EL1.FRINTTS == 0b0001. + +HWCAP2_SVEI8MM + Functionality implied by ID_AA64ZFR0_EL1.I8MM == 0b0001. + +HWCAP2_SVEF32MM + Functionality implied by ID_AA64ZFR0_EL1.F32MM == 0b0001. + +HWCAP2_SVEF64MM + Functionality implied by ID_AA64ZFR0_EL1.F64MM == 0b0001. + +HWCAP2_SVEBF16 + Functionality implied by ID_AA64ZFR0_EL1.BF16 == 0b0001. + +HWCAP2_I8MM + Functionality implied by ID_AA64ISAR1_EL1.I8MM == 0b0001. + +HWCAP2_BF16 + Functionality implied by ID_AA64ISAR1_EL1.BF16 == 0b0001. + +HWCAP2_DGH + Functionality implied by ID_AA64ISAR1_EL1.DGH == 0b0001. + +HWCAP2_RNG + Functionality implied by ID_AA64ISAR0_EL1.RNDR == 0b0001. + +HWCAP2_BTI + Functionality implied by ID_AA64PFR0_EL1.BT == 0b0001. + +HWCAP2_MTE + Functionality implied by ID_AA64PFR1_EL1.MTE == 0b0010, as described + by Documentation/arm64/memory-tagging-extension.rst. + +HWCAP2_ECV + Functionality implied by ID_AA64MMFR0_EL1.ECV == 0b0001. + +HWCAP2_AFP + Functionality implied by ID_AA64MFR1_EL1.AFP == 0b0001. + +HWCAP2_RPRES + Functionality implied by ID_AA64ISAR2_EL1.RPRES == 0b0001. + +HWCAP2_MTE3 + Functionality implied by ID_AA64PFR1_EL1.MTE == 0b0011, as described + by Documentation/arm64/memory-tagging-extension.rst. + +HWCAP2_SME + Functionality implied by ID_AA64PFR1_EL1.SME == 0b0001, as described + by Documentation/arm64/sme.rst. + +HWCAP2_SME_I16I64 + Functionality implied by ID_AA64SMFR0_EL1.I16I64 == 0b1111. + +HWCAP2_SME_F64F64 + Functionality implied by ID_AA64SMFR0_EL1.F64F64 == 0b1. + +HWCAP2_SME_I8I32 + Functionality implied by ID_AA64SMFR0_EL1.I8I32 == 0b1111. + +HWCAP2_SME_F16F32 + Functionality implied by ID_AA64SMFR0_EL1.F16F32 == 0b1. + +HWCAP2_SME_B16F32 + Functionality implied by ID_AA64SMFR0_EL1.B16F32 == 0b1. + +HWCAP2_SME_F32F32 + Functionality implied by ID_AA64SMFR0_EL1.F32F32 == 0b1. + +HWCAP2_SME_FA64 + Functionality implied by ID_AA64SMFR0_EL1.FA64 == 0b1. + +HWCAP2_WFXT + Functionality implied by ID_AA64ISAR2_EL1.WFXT == 0b0010. + +HWCAP2_EBF16 + Functionality implied by ID_AA64ISAR1_EL1.BF16 == 0b0010. + +HWCAP2_SVE_EBF16 + Functionality implied by ID_AA64ZFR0_EL1.BF16 == 0b0010. + +4. Unused AT_HWCAP bits +----------------------- + +For interoperation with userspace, the kernel guarantees that bits 62 +and 63 of AT_HWCAP will always be returned as 0. diff --git a/Documentation/arm64/features.rst b/Documentation/arm64/features.rst new file mode 100644 index 000000000..dfa4cb3cd --- /dev/null +++ b/Documentation/arm64/features.rst @@ -0,0 +1,3 @@ +.. SPDX-License-Identifier: GPL-2.0 + +.. kernel-feat:: $srctree/Documentation/features arm64 diff --git a/Documentation/arm64/hugetlbpage.rst b/Documentation/arm64/hugetlbpage.rst new file mode 100644 index 000000000..a110124c1 --- /dev/null +++ b/Documentation/arm64/hugetlbpage.rst @@ -0,0 +1,43 @@ +.. _hugetlbpage_index: + +==================== +HugeTLBpage on ARM64 +==================== + +Hugepage relies on making efficient use of TLBs to improve performance of +address translations. The benefit depends on both - + + - the size of hugepages + - size of entries supported by the TLBs + +The ARM64 port supports two flavours of hugepages. + +1) Block mappings at the pud/pmd level +-------------------------------------- + +These are regular hugepages where a pmd or a pud page table entry points to a +block of memory. Regardless of the supported size of entries in TLB, block +mappings reduce the depth of page table walk needed to translate hugepage +addresses. + +2) Using the Contiguous bit +--------------------------- + +The architecture provides a contiguous bit in the translation table entries +(D4.5.3, ARM DDI 0487C.a) that hints to the MMU to indicate that it is one of a +contiguous set of entries that can be cached in a single TLB entry. + +The contiguous bit is used in Linux to increase the mapping size at the pmd and +pte (last) level. The number of supported contiguous entries varies by page size +and level of the page table. + + +The following hugepage sizes are supported - + + ====== ======== ==== ======== === + - CONT PTE PMD CONT PMD PUD + ====== ======== ==== ======== === + 4K: 64K 2M 32M 1G + 16K: 2M 32M 1G + 64K: 2M 512M 16G + ====== ======== ==== ======== === diff --git a/Documentation/arm64/index.rst b/Documentation/arm64/index.rst new file mode 100644 index 000000000..ae21f8118 --- /dev/null +++ b/Documentation/arm64/index.rst @@ -0,0 +1,36 @@ +.. _arm64_index: + +================== +ARM64 Architecture +================== + +.. toctree:: + :maxdepth: 1 + + acpi_object_usage + amu + arm-acpi + asymmetric-32bit + booting + cpu-feature-registers + elf_hwcaps + hugetlbpage + legacy_instructions + memory + memory-tagging-extension + perf + pointer-authentication + silicon-errata + sme + sve + tagged-address-abi + tagged-pointers + + features + +.. only:: subproject and html + + Indices + ======= + + * :ref:`genindex` diff --git a/Documentation/arm64/kasan-offsets.sh b/Documentation/arm64/kasan-offsets.sh new file mode 100644 index 000000000..2dc5f9e18 --- /dev/null +++ b/Documentation/arm64/kasan-offsets.sh @@ -0,0 +1,26 @@ +#!/bin/sh + +# Print out the KASAN_SHADOW_OFFSETS required to place the KASAN SHADOW +# start address at the top of the linear region + +print_kasan_offset () { + printf "%02d\t" $1 + printf "0x%08x00000000\n" $(( (0xffffffff & (-1 << ($1 - 1 - 32))) \ + - (1 << (64 - 32 - $2)) )) +} + +echo KASAN_SHADOW_SCALE_SHIFT = 3 +printf "VABITS\tKASAN_SHADOW_OFFSET\n" +print_kasan_offset 48 3 +print_kasan_offset 47 3 +print_kasan_offset 42 3 +print_kasan_offset 39 3 +print_kasan_offset 36 3 +echo +echo KASAN_SHADOW_SCALE_SHIFT = 4 +printf "VABITS\tKASAN_SHADOW_OFFSET\n" +print_kasan_offset 48 4 +print_kasan_offset 47 4 +print_kasan_offset 42 4 +print_kasan_offset 39 4 +print_kasan_offset 36 4 diff --git a/Documentation/arm64/legacy_instructions.rst b/Documentation/arm64/legacy_instructions.rst new file mode 100644 index 000000000..54401b22c --- /dev/null +++ b/Documentation/arm64/legacy_instructions.rst @@ -0,0 +1,68 @@ +=================== +Legacy instructions +=================== + +The arm64 port of the Linux kernel provides infrastructure to support +emulation of instructions which have been deprecated, or obsoleted in +the architecture. The infrastructure code uses undefined instruction +hooks to support emulation. Where available it also allows turning on +the instruction execution in hardware. + +The emulation mode can be controlled by writing to sysctl nodes +(/proc/sys/abi). The following explains the different execution +behaviours and the corresponding values of the sysctl nodes - + +* Undef + Value: 0 + + Generates undefined instruction abort. Default for instructions that + have been obsoleted in the architecture, e.g., SWP + +* Emulate + Value: 1 + + Uses software emulation. To aid migration of software, in this mode + usage of emulated instruction is traced as well as rate limited + warnings are issued. This is the default for deprecated + instructions, .e.g., CP15 barriers + +* Hardware Execution + Value: 2 + + Although marked as deprecated, some implementations may support the + enabling/disabling of hardware support for the execution of these + instructions. Using hardware execution generally provides better + performance, but at the loss of ability to gather runtime statistics + about the use of the deprecated instructions. + +The default mode depends on the status of the instruction in the +architecture. Deprecated instructions should default to emulation +while obsolete instructions must be undefined by default. + +Note: Instruction emulation may not be possible in all cases. See +individual instruction notes for further information. + +Supported legacy instructions +----------------------------- +* SWP{B} + +:Node: /proc/sys/abi/swp +:Status: Obsolete +:Default: Undef (0) + +* CP15 Barriers + +:Node: /proc/sys/abi/cp15_barrier +:Status: Deprecated +:Default: Emulate (1) + +* SETEND + +:Node: /proc/sys/abi/setend +:Status: Deprecated +:Default: Emulate (1)* + + Note: All the cpus on the system must have mixed endian support at EL0 + for this feature to be enabled. If a new CPU - which doesn't support mixed + endian - is hotplugged in after this feature has been enabled, there could + be unexpected results in the application. diff --git a/Documentation/arm64/memory-tagging-extension.rst b/Documentation/arm64/memory-tagging-extension.rst new file mode 100644 index 000000000..dbae47bba --- /dev/null +++ b/Documentation/arm64/memory-tagging-extension.rst @@ -0,0 +1,375 @@ +=============================================== +Memory Tagging Extension (MTE) in AArch64 Linux +=============================================== + +Authors: Vincenzo Frascino <vincenzo.frascino@arm.com> + Catalin Marinas <catalin.marinas@arm.com> + +Date: 2020-02-25 + +This document describes the provision of the Memory Tagging Extension +functionality in AArch64 Linux. + +Introduction +============ + +ARMv8.5 based processors introduce the Memory Tagging Extension (MTE) +feature. MTE is built on top of the ARMv8.0 virtual address tagging TBI +(Top Byte Ignore) feature and allows software to access a 4-bit +allocation tag for each 16-byte granule in the physical address space. +Such memory range must be mapped with the Normal-Tagged memory +attribute. A logical tag is derived from bits 59-56 of the virtual +address used for the memory access. A CPU with MTE enabled will compare +the logical tag against the allocation tag and potentially raise an +exception on mismatch, subject to system registers configuration. + +Userspace Support +================= + +When ``CONFIG_ARM64_MTE`` is selected and Memory Tagging Extension is +supported by the hardware, the kernel advertises the feature to +userspace via ``HWCAP2_MTE``. + +PROT_MTE +-------- + +To access the allocation tags, a user process must enable the Tagged +memory attribute on an address range using a new ``prot`` flag for +``mmap()`` and ``mprotect()``: + +``PROT_MTE`` - Pages allow access to the MTE allocation tags. + +The allocation tag is set to 0 when such pages are first mapped in the +user address space and preserved on copy-on-write. ``MAP_SHARED`` is +supported and the allocation tags can be shared between processes. + +**Note**: ``PROT_MTE`` is only supported on ``MAP_ANONYMOUS`` and +RAM-based file mappings (``tmpfs``, ``memfd``). Passing it to other +types of mapping will result in ``-EINVAL`` returned by these system +calls. + +**Note**: The ``PROT_MTE`` flag (and corresponding memory type) cannot +be cleared by ``mprotect()``. + +**Note**: ``madvise()`` memory ranges with ``MADV_DONTNEED`` and +``MADV_FREE`` may have the allocation tags cleared (set to 0) at any +point after the system call. + +Tag Check Faults +---------------- + +When ``PROT_MTE`` is enabled on an address range and a mismatch between +the logical and allocation tags occurs on access, there are three +configurable behaviours: + +- *Ignore* - This is the default mode. The CPU (and kernel) ignores the + tag check fault. + +- *Synchronous* - The kernel raises a ``SIGSEGV`` synchronously, with + ``.si_code = SEGV_MTESERR`` and ``.si_addr = <fault-address>``. The + memory access is not performed. If ``SIGSEGV`` is ignored or blocked + by the offending thread, the containing process is terminated with a + ``coredump``. + +- *Asynchronous* - The kernel raises a ``SIGSEGV``, in the offending + thread, asynchronously following one or multiple tag check faults, + with ``.si_code = SEGV_MTEAERR`` and ``.si_addr = 0`` (the faulting + address is unknown). + +- *Asymmetric* - Reads are handled as for synchronous mode while writes + are handled as for asynchronous mode. + +The user can select the above modes, per thread, using the +``prctl(PR_SET_TAGGED_ADDR_CTRL, flags, 0, 0, 0)`` system call where ``flags`` +contains any number of the following values in the ``PR_MTE_TCF_MASK`` +bit-field: + +- ``PR_MTE_TCF_NONE``  - *Ignore* tag check faults + (ignored if combined with other options) +- ``PR_MTE_TCF_SYNC`` - *Synchronous* tag check fault mode +- ``PR_MTE_TCF_ASYNC`` - *Asynchronous* tag check fault mode + +If no modes are specified, tag check faults are ignored. If a single +mode is specified, the program will run in that mode. If multiple +modes are specified, the mode is selected as described in the "Per-CPU +preferred tag checking modes" section below. + +The current tag check fault configuration can be read using the +``prctl(PR_GET_TAGGED_ADDR_CTRL, 0, 0, 0, 0)`` system call. If +multiple modes were requested then all will be reported. + +Tag checking can also be disabled for a user thread by setting the +``PSTATE.TCO`` bit with ``MSR TCO, #1``. + +**Note**: Signal handlers are always invoked with ``PSTATE.TCO = 0``, +irrespective of the interrupted context. ``PSTATE.TCO`` is restored on +``sigreturn()``. + +**Note**: There are no *match-all* logical tags available for user +applications. + +**Note**: Kernel accesses to the user address space (e.g. ``read()`` +system call) are not checked if the user thread tag checking mode is +``PR_MTE_TCF_NONE`` or ``PR_MTE_TCF_ASYNC``. If the tag checking mode is +``PR_MTE_TCF_SYNC``, the kernel makes a best effort to check its user +address accesses, however it cannot always guarantee it. Kernel accesses +to user addresses are always performed with an effective ``PSTATE.TCO`` +value of zero, regardless of the user configuration. + +Excluding Tags in the ``IRG``, ``ADDG`` and ``SUBG`` instructions +----------------------------------------------------------------- + +The architecture allows excluding certain tags to be randomly generated +via the ``GCR_EL1.Exclude`` register bit-field. By default, Linux +excludes all tags other than 0. A user thread can enable specific tags +in the randomly generated set using the ``prctl(PR_SET_TAGGED_ADDR_CTRL, +flags, 0, 0, 0)`` system call where ``flags`` contains the tags bitmap +in the ``PR_MTE_TAG_MASK`` bit-field. + +**Note**: The hardware uses an exclude mask but the ``prctl()`` +interface provides an include mask. An include mask of ``0`` (exclusion +mask ``0xffff``) results in the CPU always generating tag ``0``. + +Per-CPU preferred tag checking mode +----------------------------------- + +On some CPUs the performance of MTE in stricter tag checking modes +is similar to that of less strict tag checking modes. This makes it +worthwhile to enable stricter checks on those CPUs when a less strict +checking mode is requested, in order to gain the error detection +benefits of the stricter checks without the performance downsides. To +support this scenario, a privileged user may configure a stricter +tag checking mode as the CPU's preferred tag checking mode. + +The preferred tag checking mode for each CPU is controlled by +``/sys/devices/system/cpu/cpu<N>/mte_tcf_preferred``, to which a +privileged user may write the value ``async``, ``sync`` or ``asymm``. The +default preferred mode for each CPU is ``async``. + +To allow a program to potentially run in the CPU's preferred tag +checking mode, the user program may set multiple tag check fault mode +bits in the ``flags`` argument to the ``prctl(PR_SET_TAGGED_ADDR_CTRL, +flags, 0, 0, 0)`` system call. If both synchronous and asynchronous +modes are requested then asymmetric mode may also be selected by the +kernel. If the CPU's preferred tag checking mode is in the task's set +of provided tag checking modes, that mode will be selected. Otherwise, +one of the modes in the task's mode will be selected by the kernel +from the task's mode set using the preference order: + + 1. Asynchronous + 2. Asymmetric + 3. Synchronous + +Note that there is no way for userspace to request multiple modes and +also disable asymmetric mode. + +Initial process state +--------------------- + +On ``execve()``, the new process has the following configuration: + +- ``PR_TAGGED_ADDR_ENABLE`` set to 0 (disabled) +- No tag checking modes are selected (tag check faults ignored) +- ``PR_MTE_TAG_MASK`` set to 0 (all tags excluded) +- ``PSTATE.TCO`` set to 0 +- ``PROT_MTE`` not set on any of the initial memory maps + +On ``fork()``, the new process inherits the parent's configuration and +memory map attributes with the exception of the ``madvise()`` ranges +with ``MADV_WIPEONFORK`` which will have the data and tags cleared (set +to 0). + +The ``ptrace()`` interface +-------------------------- + +``PTRACE_PEEKMTETAGS`` and ``PTRACE_POKEMTETAGS`` allow a tracer to read +the tags from or set the tags to a tracee's address space. The +``ptrace()`` system call is invoked as ``ptrace(request, pid, addr, +data)`` where: + +- ``request`` - one of ``PTRACE_PEEKMTETAGS`` or ``PTRACE_POKEMTETAGS``. +- ``pid`` - the tracee's PID. +- ``addr`` - address in the tracee's address space. +- ``data`` - pointer to a ``struct iovec`` where ``iov_base`` points to + a buffer of ``iov_len`` length in the tracer's address space. + +The tags in the tracer's ``iov_base`` buffer are represented as one +4-bit tag per byte and correspond to a 16-byte MTE tag granule in the +tracee's address space. + +**Note**: If ``addr`` is not aligned to a 16-byte granule, the kernel +will use the corresponding aligned address. + +``ptrace()`` return value: + +- 0 - tags were copied, the tracer's ``iov_len`` was updated to the + number of tags transferred. This may be smaller than the requested + ``iov_len`` if the requested address range in the tracee's or the + tracer's space cannot be accessed or does not have valid tags. +- ``-EPERM`` - the specified process cannot be traced. +- ``-EIO`` - the tracee's address range cannot be accessed (e.g. invalid + address) and no tags copied. ``iov_len`` not updated. +- ``-EFAULT`` - fault on accessing the tracer's memory (``struct iovec`` + or ``iov_base`` buffer) and no tags copied. ``iov_len`` not updated. +- ``-EOPNOTSUPP`` - the tracee's address does not have valid tags (never + mapped with the ``PROT_MTE`` flag). ``iov_len`` not updated. + +**Note**: There are no transient errors for the requests above, so user +programs should not retry in case of a non-zero system call return. + +``PTRACE_GETREGSET`` and ``PTRACE_SETREGSET`` with ``addr == +``NT_ARM_TAGGED_ADDR_CTRL`` allow ``ptrace()`` access to the tagged +address ABI control and MTE configuration of a process as per the +``prctl()`` options described in +Documentation/arm64/tagged-address-abi.rst and above. The corresponding +``regset`` is 1 element of 8 bytes (``sizeof(long))``). + +Core dump support +----------------- + +The allocation tags for user memory mapped with ``PROT_MTE`` are dumped +in the core file as additional ``PT_AARCH64_MEMTAG_MTE`` segments. The +program header for such segment is defined as: + +:``p_type``: ``PT_AARCH64_MEMTAG_MTE`` +:``p_flags``: 0 +:``p_offset``: segment file offset +:``p_vaddr``: segment virtual address, same as the corresponding + ``PT_LOAD`` segment +:``p_paddr``: 0 +:``p_filesz``: segment size in file, calculated as ``p_mem_sz / 32`` + (two 4-bit tags cover 32 bytes of memory) +:``p_memsz``: segment size in memory, same as the corresponding + ``PT_LOAD`` segment +:``p_align``: 0 + +The tags are stored in the core file at ``p_offset`` as two 4-bit tags +in a byte. With the tag granule of 16 bytes, a 4K page requires 128 +bytes in the core file. + +Example of correct usage +======================== + +*MTE Example code* + +.. code-block:: c + + /* + * To be compiled with -march=armv8.5-a+memtag + */ + #include <errno.h> + #include <stdint.h> + #include <stdio.h> + #include <stdlib.h> + #include <unistd.h> + #include <sys/auxv.h> + #include <sys/mman.h> + #include <sys/prctl.h> + + /* + * From arch/arm64/include/uapi/asm/hwcap.h + */ + #define HWCAP2_MTE (1 << 18) + + /* + * From arch/arm64/include/uapi/asm/mman.h + */ + #define PROT_MTE 0x20 + + /* + * From include/uapi/linux/prctl.h + */ + #define PR_SET_TAGGED_ADDR_CTRL 55 + #define PR_GET_TAGGED_ADDR_CTRL 56 + # define PR_TAGGED_ADDR_ENABLE (1UL << 0) + # define PR_MTE_TCF_SHIFT 1 + # define PR_MTE_TCF_NONE (0UL << PR_MTE_TCF_SHIFT) + # define PR_MTE_TCF_SYNC (1UL << PR_MTE_TCF_SHIFT) + # define PR_MTE_TCF_ASYNC (2UL << PR_MTE_TCF_SHIFT) + # define PR_MTE_TCF_MASK (3UL << PR_MTE_TCF_SHIFT) + # define PR_MTE_TAG_SHIFT 3 + # define PR_MTE_TAG_MASK (0xffffUL << PR_MTE_TAG_SHIFT) + + /* + * Insert a random logical tag into the given pointer. + */ + #define insert_random_tag(ptr) ({ \ + uint64_t __val; \ + asm("irg %0, %1" : "=r" (__val) : "r" (ptr)); \ + __val; \ + }) + + /* + * Set the allocation tag on the destination address. + */ + #define set_tag(tagged_addr) do { \ + asm volatile("stg %0, [%0]" : : "r" (tagged_addr) : "memory"); \ + } while (0) + + int main() + { + unsigned char *a; + unsigned long page_sz = sysconf(_SC_PAGESIZE); + unsigned long hwcap2 = getauxval(AT_HWCAP2); + + /* check if MTE is present */ + if (!(hwcap2 & HWCAP2_MTE)) + return EXIT_FAILURE; + + /* + * Enable the tagged address ABI, synchronous or asynchronous MTE + * tag check faults (based on per-CPU preference) and allow all + * non-zero tags in the randomly generated set. + */ + if (prctl(PR_SET_TAGGED_ADDR_CTRL, + PR_TAGGED_ADDR_ENABLE | PR_MTE_TCF_SYNC | PR_MTE_TCF_ASYNC | + (0xfffe << PR_MTE_TAG_SHIFT), + 0, 0, 0)) { + perror("prctl() failed"); + return EXIT_FAILURE; + } + + a = mmap(0, page_sz, PROT_READ | PROT_WRITE, + MAP_PRIVATE | MAP_ANONYMOUS, -1, 0); + if (a == MAP_FAILED) { + perror("mmap() failed"); + return EXIT_FAILURE; + } + + /* + * Enable MTE on the above anonymous mmap. The flag could be passed + * directly to mmap() and skip this step. + */ + if (mprotect(a, page_sz, PROT_READ | PROT_WRITE | PROT_MTE)) { + perror("mprotect() failed"); + return EXIT_FAILURE; + } + + /* access with the default tag (0) */ + a[0] = 1; + a[1] = 2; + + printf("a[0] = %hhu a[1] = %hhu\n", a[0], a[1]); + + /* set the logical and allocation tags */ + a = (unsigned char *)insert_random_tag(a); + set_tag(a); + + printf("%p\n", a); + + /* non-zero tag access */ + a[0] = 3; + printf("a[0] = %hhu a[1] = %hhu\n", a[0], a[1]); + + /* + * If MTE is enabled correctly the next instruction will generate an + * exception. + */ + printf("Expecting SIGSEGV...\n"); + a[16] = 0xdd; + + /* this should not be printed in the PR_MTE_TCF_SYNC mode */ + printf("...haven't got one\n"); + + return EXIT_FAILURE; + } diff --git a/Documentation/arm64/memory.rst b/Documentation/arm64/memory.rst new file mode 100644 index 000000000..2a641ba7b --- /dev/null +++ b/Documentation/arm64/memory.rst @@ -0,0 +1,167 @@ +============================== +Memory Layout on AArch64 Linux +============================== + +Author: Catalin Marinas <catalin.marinas@arm.com> + +This document describes the virtual memory layout used by the AArch64 +Linux kernel. The architecture allows up to 4 levels of translation +tables with a 4KB page size and up to 3 levels with a 64KB page size. + +AArch64 Linux uses either 3 levels or 4 levels of translation tables +with the 4KB page configuration, allowing 39-bit (512GB) or 48-bit +(256TB) virtual addresses, respectively, for both user and kernel. With +64KB pages, only 2 levels of translation tables, allowing 42-bit (4TB) +virtual address, are used but the memory layout is the same. + +ARMv8.2 adds optional support for Large Virtual Address space. This is +only available when running with a 64KB page size and expands the +number of descriptors in the first level of translation. + +User addresses have bits 63:48 set to 0 while the kernel addresses have +the same bits set to 1. TTBRx selection is given by bit 63 of the +virtual address. The swapper_pg_dir contains only kernel (global) +mappings while the user pgd contains only user (non-global) mappings. +The swapper_pg_dir address is written to TTBR1 and never written to +TTBR0. + + +AArch64 Linux memory layout with 4KB pages + 4 levels (48-bit):: + + Start End Size Use + ----------------------------------------------------------------------- + 0000000000000000 0000ffffffffffff 256TB user + ffff000000000000 ffff7fffffffffff 128TB kernel logical memory map + [ffff600000000000 ffff7fffffffffff] 32TB [kasan shadow region] + ffff800000000000 ffff800007ffffff 128MB modules + ffff800008000000 fffffbffefffffff 124TB vmalloc + fffffbfff0000000 fffffbfffdffffff 224MB fixed mappings (top down) + fffffbfffe000000 fffffbfffe7fffff 8MB [guard region] + fffffbfffe800000 fffffbffff7fffff 16MB PCI I/O space + fffffbffff800000 fffffbffffffffff 8MB [guard region] + fffffc0000000000 fffffdffffffffff 2TB vmemmap + fffffe0000000000 ffffffffffffffff 2TB [guard region] + + +AArch64 Linux memory layout with 64KB pages + 3 levels (52-bit with HW support):: + + Start End Size Use + ----------------------------------------------------------------------- + 0000000000000000 000fffffffffffff 4PB user + fff0000000000000 ffff7fffffffffff ~4PB kernel logical memory map + [fffd800000000000 ffff7fffffffffff] 512TB [kasan shadow region] + ffff800000000000 ffff800007ffffff 128MB modules + ffff800008000000 fffffbffefffffff 124TB vmalloc + fffffbfff0000000 fffffbfffdffffff 224MB fixed mappings (top down) + fffffbfffe000000 fffffbfffe7fffff 8MB [guard region] + fffffbfffe800000 fffffbffff7fffff 16MB PCI I/O space + fffffbffff800000 fffffbffffffffff 8MB [guard region] + fffffc0000000000 ffffffdfffffffff ~4TB vmemmap + ffffffe000000000 ffffffffffffffff 128GB [guard region] + + +Translation table lookup with 4KB pages:: + + +--------+--------+--------+--------+--------+--------+--------+--------+ + |63 56|55 48|47 40|39 32|31 24|23 16|15 8|7 0| + +--------+--------+--------+--------+--------+--------+--------+--------+ + | | | | | | + | | | | | v + | | | | | [11:0] in-page offset + | | | | +-> [20:12] L3 index + | | | +-----------> [29:21] L2 index + | | +---------------------> [38:30] L1 index + | +-------------------------------> [47:39] L0 index + +-------------------------------------------------> [63] TTBR0/1 + + +Translation table lookup with 64KB pages:: + + +--------+--------+--------+--------+--------+--------+--------+--------+ + |63 56|55 48|47 40|39 32|31 24|23 16|15 8|7 0| + +--------+--------+--------+--------+--------+--------+--------+--------+ + | | | | | + | | | | v + | | | | [15:0] in-page offset + | | | +----------> [28:16] L3 index + | | +--------------------------> [41:29] L2 index + | +-------------------------------> [47:42] L1 index (48-bit) + | [51:42] L1 index (52-bit) + +-------------------------------------------------> [63] TTBR0/1 + + +When using KVM without the Virtualization Host Extensions, the +hypervisor maps kernel pages in EL2 at a fixed (and potentially +random) offset from the linear mapping. See the kern_hyp_va macro and +kvm_update_va_mask function for more details. MMIO devices such as +GICv2 gets mapped next to the HYP idmap page, as do vectors when +ARM64_SPECTRE_V3A is enabled for particular CPUs. + +When using KVM with the Virtualization Host Extensions, no additional +mappings are created, since the host kernel runs directly in EL2. + +52-bit VA support in the kernel +------------------------------- +If the ARMv8.2-LVA optional feature is present, and we are running +with a 64KB page size; then it is possible to use 52-bits of address +space for both userspace and kernel addresses. However, any kernel +binary that supports 52-bit must also be able to fall back to 48-bit +at early boot time if the hardware feature is not present. + +This fallback mechanism necessitates the kernel .text to be in the +higher addresses such that they are invariant to 48/52-bit VAs. Due +to the kasan shadow being a fraction of the entire kernel VA space, +the end of the kasan shadow must also be in the higher half of the +kernel VA space for both 48/52-bit. (Switching from 48-bit to 52-bit, +the end of the kasan shadow is invariant and dependent on ~0UL, +whilst the start address will "grow" towards the lower addresses). + +In order to optimise phys_to_virt and virt_to_phys, the PAGE_OFFSET +is kept constant at 0xFFF0000000000000 (corresponding to 52-bit), +this obviates the need for an extra variable read. The physvirt +offset and vmemmap offsets are computed at early boot to enable +this logic. + +As a single binary will need to support both 48-bit and 52-bit VA +spaces, the VMEMMAP must be sized large enough for 52-bit VAs and +also must be sized large enough to accommodate a fixed PAGE_OFFSET. + +Most code in the kernel should not need to consider the VA_BITS, for +code that does need to know the VA size the variables are +defined as follows: + +VA_BITS constant the *maximum* VA space size + +VA_BITS_MIN constant the *minimum* VA space size + +vabits_actual variable the *actual* VA space size + + +Maximum and minimum sizes can be useful to ensure that buffers are +sized large enough or that addresses are positioned close enough for +the "worst" case. + +52-bit userspace VAs +-------------------- +To maintain compatibility with software that relies on the ARMv8.0 +VA space maximum size of 48-bits, the kernel will, by default, +return virtual addresses to userspace from a 48-bit range. + +Software can "opt-in" to receiving VAs from a 52-bit space by +specifying an mmap hint parameter that is larger than 48-bit. + +For example: + +.. code-block:: c + + maybe_high_address = mmap(~0UL, size, prot, flags,...); + +It is also possible to build a debug kernel that returns addresses +from a 52-bit space by enabling the following kernel config options: + +.. code-block:: sh + + CONFIG_EXPERT=y && CONFIG_ARM64_FORCE_52BIT=y + +Note that this option is only intended for debugging applications +and should not be used in production. diff --git a/Documentation/arm64/perf.rst b/Documentation/arm64/perf.rst new file mode 100644 index 000000000..1f87b57c2 --- /dev/null +++ b/Documentation/arm64/perf.rst @@ -0,0 +1,166 @@ +.. SPDX-License-Identifier: GPL-2.0 + +.. _perf_index: + +==== +Perf +==== + +Perf Event Attributes +===================== + +:Author: Andrew Murray <andrew.murray@arm.com> +:Date: 2019-03-06 + +exclude_user +------------ + +This attribute excludes userspace. + +Userspace always runs at EL0 and thus this attribute will exclude EL0. + + +exclude_kernel +-------------- + +This attribute excludes the kernel. + +The kernel runs at EL2 with VHE and EL1 without. Guest kernels always run +at EL1. + +For the host this attribute will exclude EL1 and additionally EL2 on a VHE +system. + +For the guest this attribute will exclude EL1. Please note that EL2 is +never counted within a guest. + + +exclude_hv +---------- + +This attribute excludes the hypervisor. + +For a VHE host this attribute is ignored as we consider the host kernel to +be the hypervisor. + +For a non-VHE host this attribute will exclude EL2 as we consider the +hypervisor to be any code that runs at EL2 which is predominantly used for +guest/host transitions. + +For the guest this attribute has no effect. Please note that EL2 is +never counted within a guest. + + +exclude_host / exclude_guest +---------------------------- + +These attributes exclude the KVM host and guest, respectively. + +The KVM host may run at EL0 (userspace), EL1 (non-VHE kernel) and EL2 (VHE +kernel or non-VHE hypervisor). + +The KVM guest may run at EL0 (userspace) and EL1 (kernel). + +Due to the overlapping exception levels between host and guests we cannot +exclusively rely on the PMU's hardware exception filtering - therefore we +must enable/disable counting on the entry and exit to the guest. This is +performed differently on VHE and non-VHE systems. + +For non-VHE systems we exclude EL2 for exclude_host - upon entering and +exiting the guest we disable/enable the event as appropriate based on the +exclude_host and exclude_guest attributes. + +For VHE systems we exclude EL1 for exclude_guest and exclude both EL0,EL2 +for exclude_host. Upon entering and exiting the guest we modify the event +to include/exclude EL0 as appropriate based on the exclude_host and +exclude_guest attributes. + +The statements above also apply when these attributes are used within a +non-VHE guest however please note that EL2 is never counted within a guest. + + +Accuracy +-------- + +On non-VHE hosts we enable/disable counters on the entry/exit of host/guest +transition at EL2 - however there is a period of time between +enabling/disabling the counters and entering/exiting the guest. We are +able to eliminate counters counting host events on the boundaries of guest +entry/exit when counting guest events by filtering out EL2 for +exclude_host. However when using !exclude_hv there is a small blackout +window at the guest entry/exit where host events are not captured. + +On VHE systems there are no blackout windows. + +Perf Userspace PMU Hardware Counter Access +========================================== + +Overview +-------- +The perf userspace tool relies on the PMU to monitor events. It offers an +abstraction layer over the hardware counters since the underlying +implementation is cpu-dependent. +Arm64 allows userspace tools to have access to the registers storing the +hardware counters' values directly. + +This targets specifically self-monitoring tasks in order to reduce the overhead +by directly accessing the registers without having to go through the kernel. + +How-to +------ +The focus is set on the armv8 PMUv3 which makes sure that the access to the pmu +registers is enabled and that the userspace has access to the relevant +information in order to use them. + +In order to have access to the hardware counters, the global sysctl +kernel/perf_user_access must first be enabled: + +.. code-block:: sh + + echo 1 > /proc/sys/kernel/perf_user_access + +It is necessary to open the event using the perf tool interface with config1:1 +attr bit set: the sys_perf_event_open syscall returns a fd which can +subsequently be used with the mmap syscall in order to retrieve a page of memory +containing information about the event. The PMU driver uses this page to expose +to the user the hardware counter's index and other necessary data. Using this +index enables the user to access the PMU registers using the `mrs` instruction. +Access to the PMU registers is only valid while the sequence lock is unchanged. +In particular, the PMSELR_EL0 register is zeroed each time the sequence lock is +changed. + +The userspace access is supported in libperf using the perf_evsel__mmap() +and perf_evsel__read() functions. See `tools/lib/perf/tests/test-evsel.c`_ for +an example. + +About heterogeneous systems +--------------------------- +On heterogeneous systems such as big.LITTLE, userspace PMU counter access can +only be enabled when the tasks are pinned to a homogeneous subset of cores and +the corresponding PMU instance is opened by specifying the 'type' attribute. +The use of generic event types is not supported in this case. + +Have a look at `tools/perf/arch/arm64/tests/user-events.c`_ for an example. It +can be run using the perf tool to check that the access to the registers works +correctly from userspace: + +.. code-block:: sh + + perf test -v user + +About chained events and counter sizes +-------------------------------------- +The user can request either a 32-bit (config1:0 == 0) or 64-bit (config1:0 == 1) +counter along with userspace access. The sys_perf_event_open syscall will fail +if a 64-bit counter is requested and the hardware doesn't support 64-bit +counters. Chained events are not supported in conjunction with userspace counter +access. If a 32-bit counter is requested on hardware with 64-bit counters, then +userspace must treat the upper 32-bits read from the counter as UNKNOWN. The +'pmc_width' field in the user page will indicate the valid width of the counter +and should be used to mask the upper bits as needed. + +.. Links +.. _tools/perf/arch/arm64/tests/user-events.c: + https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/tree/tools/perf/arch/arm64/tests/user-events.c +.. _tools/lib/perf/tests/test-evsel.c: + https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/tree/tools/lib/perf/tests/test-evsel.c diff --git a/Documentation/arm64/pointer-authentication.rst b/Documentation/arm64/pointer-authentication.rst new file mode 100644 index 000000000..e5dad2e40 --- /dev/null +++ b/Documentation/arm64/pointer-authentication.rst @@ -0,0 +1,142 @@ +======================================= +Pointer authentication in AArch64 Linux +======================================= + +Author: Mark Rutland <mark.rutland@arm.com> + +Date: 2017-07-19 + +This document briefly describes the provision of pointer authentication +functionality in AArch64 Linux. + + +Architecture overview +--------------------- + +The ARMv8.3 Pointer Authentication extension adds primitives that can be +used to mitigate certain classes of attack where an attacker can corrupt +the contents of some memory (e.g. the stack). + +The extension uses a Pointer Authentication Code (PAC) to determine +whether pointers have been modified unexpectedly. A PAC is derived from +a pointer, another value (such as the stack pointer), and a secret key +held in system registers. + +The extension adds instructions to insert a valid PAC into a pointer, +and to verify/remove the PAC from a pointer. The PAC occupies a number +of high-order bits of the pointer, which varies dependent on the +configured virtual address size and whether pointer tagging is in use. + +A subset of these instructions have been allocated from the HINT +encoding space. In the absence of the extension (or when disabled), +these instructions behave as NOPs. Applications and libraries using +these instructions operate correctly regardless of the presence of the +extension. + +The extension provides five separate keys to generate PACs - two for +instruction addresses (APIAKey, APIBKey), two for data addresses +(APDAKey, APDBKey), and one for generic authentication (APGAKey). + + +Basic support +------------- + +When CONFIG_ARM64_PTR_AUTH is selected, and relevant HW support is +present, the kernel will assign random key values to each process at +exec*() time. The keys are shared by all threads within the process, and +are preserved across fork(). + +Presence of address authentication functionality is advertised via +HWCAP_PACA, and generic authentication functionality via HWCAP_PACG. + +The number of bits that the PAC occupies in a pointer is 55 minus the +virtual address size configured by the kernel. For example, with a +virtual address size of 48, the PAC is 7 bits wide. + +When ARM64_PTR_AUTH_KERNEL is selected, the kernel will be compiled +with HINT space pointer authentication instructions protecting +function returns. Kernels built with this option will work on hardware +with or without pointer authentication support. + +In addition to exec(), keys can also be reinitialized to random values +using the PR_PAC_RESET_KEYS prctl. A bitmask of PR_PAC_APIAKEY, +PR_PAC_APIBKEY, PR_PAC_APDAKEY, PR_PAC_APDBKEY and PR_PAC_APGAKEY +specifies which keys are to be reinitialized; specifying 0 means "all +keys". + + +Debugging +--------- + +When CONFIG_ARM64_PTR_AUTH is selected, and HW support for address +authentication is present, the kernel will expose the position of TTBR0 +PAC bits in the NT_ARM_PAC_MASK regset (struct user_pac_mask), which +userspace can acquire via PTRACE_GETREGSET. + +The regset is exposed only when HWCAP_PACA is set. Separate masks are +exposed for data pointers and instruction pointers, as the set of PAC +bits can vary between the two. Note that the masks apply to TTBR0 +addresses, and are not valid to apply to TTBR1 addresses (e.g. kernel +pointers). + +Additionally, when CONFIG_CHECKPOINT_RESTORE is also set, the kernel +will expose the NT_ARM_PACA_KEYS and NT_ARM_PACG_KEYS regsets (struct +user_pac_address_keys and struct user_pac_generic_keys). These can be +used to get and set the keys for a thread. + + +Virtualization +-------------- + +Pointer authentication is enabled in KVM guest when each virtual cpu is +initialised by passing flags KVM_ARM_VCPU_PTRAUTH_[ADDRESS/GENERIC] and +requesting these two separate cpu features to be enabled. The current KVM +guest implementation works by enabling both features together, so both +these userspace flags are checked before enabling pointer authentication. +The separate userspace flag will allow to have no userspace ABI changes +if support is added in the future to allow these two features to be +enabled independently of one another. + +As Arm Architecture specifies that Pointer Authentication feature is +implemented along with the VHE feature so KVM arm64 ptrauth code relies +on VHE mode to be present. + +Additionally, when these vcpu feature flags are not set then KVM will +filter out the Pointer Authentication system key registers from +KVM_GET/SET_REG_* ioctls and mask those features from cpufeature ID +register. Any attempt to use the Pointer Authentication instructions will +result in an UNDEFINED exception being injected into the guest. + + +Enabling and disabling keys +--------------------------- + +The prctl PR_PAC_SET_ENABLED_KEYS allows the user program to control which +PAC keys are enabled in a particular task. It takes two arguments, the +first being a bitmask of PR_PAC_APIAKEY, PR_PAC_APIBKEY, PR_PAC_APDAKEY +and PR_PAC_APDBKEY specifying which keys shall be affected by this prctl, +and the second being a bitmask of the same bits specifying whether the key +should be enabled or disabled. For example:: + + prctl(PR_PAC_SET_ENABLED_KEYS, + PR_PAC_APIAKEY | PR_PAC_APIBKEY | PR_PAC_APDAKEY | PR_PAC_APDBKEY, + PR_PAC_APIBKEY, 0, 0); + +disables all keys except the IB key. + +The main reason why this is useful is to enable a userspace ABI that uses PAC +instructions to sign and authenticate function pointers and other pointers +exposed outside of the function, while still allowing binaries conforming to +the ABI to interoperate with legacy binaries that do not sign or authenticate +pointers. + +The idea is that a dynamic loader or early startup code would issue this +prctl very early after establishing that a process may load legacy binaries, +but before executing any PAC instructions. + +For compatibility with previous kernel versions, processes start up with IA, +IB, DA and DB enabled, and are reset to this state on exec(). Processes created +via fork() and clone() inherit the key enabled state from the calling process. + +It is recommended to avoid disabling the IA key, as this has higher performance +overhead than disabling any of the other keys. diff --git a/Documentation/arm64/silicon-errata.rst b/Documentation/arm64/silicon-errata.rst new file mode 100644 index 000000000..d9fce65b2 --- /dev/null +++ b/Documentation/arm64/silicon-errata.rst @@ -0,0 +1,223 @@ +======================================= +Silicon Errata and Software Workarounds +======================================= + +Author: Will Deacon <will.deacon@arm.com> + +Date : 27 November 2015 + +It is an unfortunate fact of life that hardware is often produced with +so-called "errata", which can cause it to deviate from the architecture +under specific circumstances. For hardware produced by ARM, these +errata are broadly classified into the following categories: + + ========== ======================================================== + Category A A critical error without a viable workaround. + Category B A significant or critical error with an acceptable + workaround. + Category C A minor error that is not expected to occur under normal + operation. + ========== ======================================================== + +For more information, consult one of the "Software Developers Errata +Notice" documents available on infocenter.arm.com (registration +required). + +As far as Linux is concerned, Category B errata may require some special +treatment in the operating system. For example, avoiding a particular +sequence of code, or configuring the processor in a particular way. A +less common situation may require similar actions in order to declassify +a Category A erratum into a Category C erratum. These are collectively +known as "software workarounds" and are only required in the minority of +cases (e.g. those cases that both require a non-secure workaround *and* +can be triggered by Linux). + +For software workarounds that may adversely impact systems unaffected by +the erratum in question, a Kconfig entry is added under "Kernel +Features" -> "ARM errata workarounds via the alternatives framework". +These are enabled by default and patched in at runtime when an affected +CPU is detected. For less-intrusive workarounds, a Kconfig option is not +available and the code is structured (preferably with a comment) in such +a way that the erratum will not be hit. + +This approach can make it slightly onerous to determine exactly which +errata are worked around in an arbitrary kernel source tree, so this +file acts as a registry of software workarounds in the Linux Kernel and +will be updated when new workarounds are committed and backported to +stable kernels. + ++----------------+-----------------+-----------------+-----------------------------+ +| Implementor | Component | Erratum ID | Kconfig | ++================+=================+=================+=============================+ +| Allwinner | A64/R18 | UNKNOWN1 | SUN50I_ERRATUM_UNKNOWN1 | ++----------------+-----------------+-----------------+-----------------------------+ ++----------------+-----------------+-----------------+-----------------------------+ +| Ampere | AmpereOne | AC03_CPU_38 | AMPERE_ERRATUM_AC03_CPU_38 | ++----------------+-----------------+-----------------+-----------------------------+ ++----------------+-----------------+-----------------+-----------------------------+ +| ARM | Cortex-A510 | #2457168 | ARM64_ERRATUM_2457168 | ++----------------+-----------------+-----------------+-----------------------------+ +| ARM | Cortex-A510 | #2064142 | ARM64_ERRATUM_2064142 | ++----------------+-----------------+-----------------+-----------------------------+ +| ARM | Cortex-A510 | #2038923 | ARM64_ERRATUM_2038923 | ++----------------+-----------------+-----------------+-----------------------------+ +| ARM | Cortex-A510 | #1902691 | ARM64_ERRATUM_1902691 | ++----------------+-----------------+-----------------+-----------------------------+ +| ARM | Cortex-A520 | #2966298 | ARM64_ERRATUM_2966298 | ++----------------+-----------------+-----------------+-----------------------------+ +| ARM | Cortex-A53 | #826319 | ARM64_ERRATUM_826319 | ++----------------+-----------------+-----------------+-----------------------------+ +| ARM | Cortex-A53 | #827319 | ARM64_ERRATUM_827319 | ++----------------+-----------------+-----------------+-----------------------------+ +| ARM | Cortex-A53 | #824069 | ARM64_ERRATUM_824069 | ++----------------+-----------------+-----------------+-----------------------------+ +| ARM | Cortex-A53 | #819472 | ARM64_ERRATUM_819472 | ++----------------+-----------------+-----------------+-----------------------------+ +| ARM | Cortex-A53 | #845719 | ARM64_ERRATUM_845719 | ++----------------+-----------------+-----------------+-----------------------------+ +| ARM | Cortex-A53 | #843419 | ARM64_ERRATUM_843419 | ++----------------+-----------------+-----------------+-----------------------------+ +| ARM | Cortex-A55 | #1024718 | ARM64_ERRATUM_1024718 | ++----------------+-----------------+-----------------+-----------------------------+ +| ARM | Cortex-A55 | #1530923 | ARM64_ERRATUM_1530923 | ++----------------+-----------------+-----------------+-----------------------------+ +| ARM | Cortex-A55 | #2441007 | ARM64_ERRATUM_2441007 | ++----------------+-----------------+-----------------+-----------------------------+ +| ARM | Cortex-A57 | #832075 | ARM64_ERRATUM_832075 | ++----------------+-----------------+-----------------+-----------------------------+ +| ARM | Cortex-A57 | #852523 | N/A | ++----------------+-----------------+-----------------+-----------------------------+ +| ARM | Cortex-A57 | #834220 | ARM64_ERRATUM_834220 | ++----------------+-----------------+-----------------+-----------------------------+ +| ARM | Cortex-A57 | #1319537 | ARM64_ERRATUM_1319367 | ++----------------+-----------------+-----------------+-----------------------------+ +| ARM | Cortex-A57 | #1742098 | ARM64_ERRATUM_1742098 | ++----------------+-----------------+-----------------+-----------------------------+ +| ARM | Cortex-A72 | #853709 | N/A | ++----------------+-----------------+-----------------+-----------------------------+ +| ARM | Cortex-A72 | #1319367 | ARM64_ERRATUM_1319367 | ++----------------+-----------------+-----------------+-----------------------------+ +| ARM | Cortex-A72 | #1655431 | ARM64_ERRATUM_1742098 | ++----------------+-----------------+-----------------+-----------------------------+ +| ARM | Cortex-A73 | #858921 | ARM64_ERRATUM_858921 | ++----------------+-----------------+-----------------+-----------------------------+ +| ARM | Cortex-A76 | #1188873,1418040| ARM64_ERRATUM_1418040 | ++----------------+-----------------+-----------------+-----------------------------+ +| ARM | Cortex-A76 | #1165522 | ARM64_ERRATUM_1165522 | ++----------------+-----------------+-----------------+-----------------------------+ +| ARM | Cortex-A76 | #1286807 | ARM64_ERRATUM_1286807 | ++----------------+-----------------+-----------------+-----------------------------+ +| ARM | Cortex-A76 | #1463225 | ARM64_ERRATUM_1463225 | ++----------------+-----------------+-----------------+-----------------------------+ +| ARM | Cortex-A77 | #1508412 | ARM64_ERRATUM_1508412 | ++----------------+-----------------+-----------------+-----------------------------+ +| ARM | Cortex-A510 | #2051678 | ARM64_ERRATUM_2051678 | ++----------------+-----------------+-----------------+-----------------------------+ +| ARM | Cortex-A510 | #2077057 | ARM64_ERRATUM_2077057 | ++----------------+-----------------+-----------------+-----------------------------+ +| ARM | Cortex-A510 | #2441009 | ARM64_ERRATUM_2441009 | ++----------------+-----------------+-----------------+-----------------------------+ +| ARM | Cortex-A510 | #2658417 | ARM64_ERRATUM_2658417 | ++----------------+-----------------+-----------------+-----------------------------+ +| ARM | Cortex-A710 | #2119858 | ARM64_ERRATUM_2119858 | ++----------------+-----------------+-----------------+-----------------------------+ +| ARM | Cortex-A710 | #2054223 | ARM64_ERRATUM_2054223 | ++----------------+-----------------+-----------------+-----------------------------+ +| ARM | Cortex-A710 | #2224489 | ARM64_ERRATUM_2224489 | ++----------------+-----------------+-----------------+-----------------------------+ +| ARM | Cortex-X2 | #2119858 | ARM64_ERRATUM_2119858 | ++----------------+-----------------+-----------------+-----------------------------+ +| ARM | Cortex-X2 | #2224489 | ARM64_ERRATUM_2224489 | ++----------------+-----------------+-----------------+-----------------------------+ +| ARM | Neoverse-N1 | #1188873,1418040| ARM64_ERRATUM_1418040 | ++----------------+-----------------+-----------------+-----------------------------+ +| ARM | Neoverse-N1 | #1349291 | N/A | ++----------------+-----------------+-----------------+-----------------------------+ +| ARM | Neoverse-N1 | #1542419 | ARM64_ERRATUM_1542419 | ++----------------+-----------------+-----------------+-----------------------------+ +| ARM | Neoverse-N2 | #2139208 | ARM64_ERRATUM_2139208 | ++----------------+-----------------+-----------------+-----------------------------+ +| ARM | Neoverse-N2 | #2067961 | ARM64_ERRATUM_2067961 | ++----------------+-----------------+-----------------+-----------------------------+ +| ARM | Neoverse-N2 | #2253138 | ARM64_ERRATUM_2253138 | ++----------------+-----------------+-----------------+-----------------------------+ +| ARM | MMU-500 | #841119,826419 | N/A | ++----------------+-----------------+-----------------+-----------------------------+ +| ARM | MMU-600 | #1076982,1209401| N/A | ++----------------+-----------------+-----------------+-----------------------------+ +| ARM | MMU-700 | #2268618,2812531| N/A | ++----------------+-----------------+-----------------+-----------------------------+ ++----------------+-----------------+-----------------+-----------------------------+ +| Broadcom | Brahma-B53 | N/A | ARM64_ERRATUM_845719 | ++----------------+-----------------+-----------------+-----------------------------+ +| Broadcom | Brahma-B53 | N/A | ARM64_ERRATUM_843419 | ++----------------+-----------------+-----------------+-----------------------------+ ++----------------+-----------------+-----------------+-----------------------------+ +| Cavium | ThunderX ITS | #22375,24313 | CAVIUM_ERRATUM_22375 | ++----------------+-----------------+-----------------+-----------------------------+ +| Cavium | ThunderX ITS | #23144 | CAVIUM_ERRATUM_23144 | ++----------------+-----------------+-----------------+-----------------------------+ +| Cavium | ThunderX GICv3 | #23154,38545 | CAVIUM_ERRATUM_23154 | ++----------------+-----------------+-----------------+-----------------------------+ +| Cavium | ThunderX GICv3 | #38539 | N/A | ++----------------+-----------------+-----------------+-----------------------------+ +| Cavium | ThunderX Core | #27456 | CAVIUM_ERRATUM_27456 | ++----------------+-----------------+-----------------+-----------------------------+ +| Cavium | ThunderX Core | #30115 | CAVIUM_ERRATUM_30115 | ++----------------+-----------------+-----------------+-----------------------------+ +| Cavium | ThunderX SMMUv2 | #27704 | N/A | ++----------------+-----------------+-----------------+-----------------------------+ +| Cavium | ThunderX2 SMMUv3| #74 | N/A | ++----------------+-----------------+-----------------+-----------------------------+ +| Cavium | ThunderX2 SMMUv3| #126 | N/A | ++----------------+-----------------+-----------------+-----------------------------+ +| Cavium | ThunderX2 Core | #219 | CAVIUM_TX2_ERRATUM_219 | ++----------------+-----------------+-----------------+-----------------------------+ ++----------------+-----------------+-----------------+-----------------------------+ +| Marvell | ARM-MMU-500 | #582743 | N/A | ++----------------+-----------------+-----------------+-----------------------------+ ++----------------+-----------------+-----------------+-----------------------------+ +| NVIDIA | Carmel Core | N/A | NVIDIA_CARMEL_CNP_ERRATUM | ++----------------+-----------------+-----------------+-----------------------------+ +| NVIDIA | T241 GICv3/4.x | T241-FABRIC-4 | N/A | ++----------------+-----------------+-----------------+-----------------------------+ ++----------------+-----------------+-----------------+-----------------------------+ +| Freescale/NXP | LS2080A/LS1043A | A-008585 | FSL_ERRATUM_A008585 | ++----------------+-----------------+-----------------+-----------------------------+ ++----------------+-----------------+-----------------+-----------------------------+ +| Hisilicon | Hip0{5,6,7} | #161010101 | HISILICON_ERRATUM_161010101 | ++----------------+-----------------+-----------------+-----------------------------+ +| Hisilicon | Hip0{6,7} | #161010701 | N/A | ++----------------+-----------------+-----------------+-----------------------------+ +| Hisilicon | Hip0{6,7} | #161010803 | N/A | ++----------------+-----------------+-----------------+-----------------------------+ +| Hisilicon | Hip07 | #161600802 | HISILICON_ERRATUM_161600802 | ++----------------+-----------------+-----------------+-----------------------------+ +| Hisilicon | Hip08 SMMU PMCG | #162001800 | N/A | ++----------------+-----------------+-----------------+-----------------------------+ +| Hisilicon | Hip08 SMMU PMCG | #162001900 | N/A | +| | Hip09 SMMU PMCG | | | ++----------------+-----------------+-----------------+-----------------------------+ ++----------------+-----------------+-----------------+-----------------------------+ +| Qualcomm Tech. | Kryo/Falkor v1 | E1003 | QCOM_FALKOR_ERRATUM_1003 | ++----------------+-----------------+-----------------+-----------------------------+ +| Qualcomm Tech. | Kryo/Falkor v1 | E1009 | QCOM_FALKOR_ERRATUM_1009 | ++----------------+-----------------+-----------------+-----------------------------+ +| Qualcomm Tech. | QDF2400 ITS | E0065 | QCOM_QDF2400_ERRATUM_0065 | ++----------------+-----------------+-----------------+-----------------------------+ +| Qualcomm Tech. | Falkor v{1,2} | E1041 | QCOM_FALKOR_ERRATUM_1041 | ++----------------+-----------------+-----------------+-----------------------------+ +| Qualcomm Tech. | Kryo4xx Gold | N/A | ARM64_ERRATUM_1463225 | ++----------------+-----------------+-----------------+-----------------------------+ +| Qualcomm Tech. | Kryo4xx Gold | N/A | ARM64_ERRATUM_1418040 | ++----------------+-----------------+-----------------+-----------------------------+ +| Qualcomm Tech. | Kryo4xx Silver | N/A | ARM64_ERRATUM_1530923 | ++----------------+-----------------+-----------------+-----------------------------+ +| Qualcomm Tech. | Kryo4xx Silver | N/A | ARM64_ERRATUM_1024718 | ++----------------+-----------------+-----------------+-----------------------------+ +| Qualcomm Tech. | Kryo4xx Gold | N/A | ARM64_ERRATUM_1286807 | ++----------------+-----------------+-----------------+-----------------------------+ + ++----------------+-----------------+-----------------+-----------------------------+ +| Fujitsu | A64FX | E#010001 | FUJITSU_ERRATUM_010001 | ++----------------+-----------------+-----------------+-----------------------------+ diff --git a/Documentation/arm64/sme.rst b/Documentation/arm64/sme.rst new file mode 100644 index 000000000..16d2db4c2 --- /dev/null +++ b/Documentation/arm64/sme.rst @@ -0,0 +1,431 @@ +=================================================== +Scalable Matrix Extension support for AArch64 Linux +=================================================== + +This document outlines briefly the interface provided to userspace by Linux in +order to support use of the ARM Scalable Matrix Extension (SME). + +This is an outline of the most important features and issues only and not +intended to be exhaustive. It should be read in conjunction with the SVE +documentation in sve.rst which provides details on the Streaming SVE mode +included in SME. + +This document does not aim to describe the SME architecture or programmer's +model. To aid understanding, a minimal description of relevant programmer's +model features for SME is included in Appendix A. + + +1. General +----------- + +* PSTATE.SM, PSTATE.ZA, the streaming mode vector length, the ZA + register state and TPIDR2_EL0 are tracked per thread. + +* The presence of SME is reported to userspace via HWCAP2_SME in the aux vector + AT_HWCAP2 entry. Presence of this flag implies the presence of the SME + instructions and registers, and the Linux-specific system interfaces + described in this document. SME is reported in /proc/cpuinfo as "sme". + +* Support for the execution of SME instructions in userspace can also be + detected by reading the CPU ID register ID_AA64PFR1_EL1 using an MRS + instruction, and checking that the value of the SME field is nonzero. [3] + + It does not guarantee the presence of the system interfaces described in the + following sections: software that needs to verify that those interfaces are + present must check for HWCAP2_SME instead. + +* There are a number of optional SME features, presence of these is reported + through AT_HWCAP2 through: + + HWCAP2_SME_I16I64 + HWCAP2_SME_F64F64 + HWCAP2_SME_I8I32 + HWCAP2_SME_F16F32 + HWCAP2_SME_B16F32 + HWCAP2_SME_F32F32 + HWCAP2_SME_FA64 + + This list may be extended over time as the SME architecture evolves. + + These extensions are also reported via the CPU ID register ID_AA64SMFR0_EL1, + which userspace can read using an MRS instruction. See elf_hwcaps.txt and + cpu-feature-registers.txt for details. + +* Debuggers should restrict themselves to interacting with the target via the + NT_ARM_SVE, NT_ARM_SSVE and NT_ARM_ZA regsets. The recommended way + of detecting support for these regsets is to connect to a target process + first and then attempt a + + ptrace(PTRACE_GETREGSET, pid, NT_ARM_<regset>, &iov). + +* Whenever ZA register values are exchanged in memory between userspace and + the kernel, the register value is encoded in memory as a series of horizontal + vectors from 0 to VL/8-1 stored in the same endianness invariant format as is + used for SVE vectors. + +* On thread creation TPIDR2_EL0 is preserved unless CLONE_SETTLS is specified, + in which case it is set to 0. + +2. Vector lengths +------------------ + +SME defines a second vector length similar to the SVE vector length which is +controls the size of the streaming mode SVE vectors and the ZA matrix array. +The ZA matrix is square with each side having as many bytes as a streaming +mode SVE vector. + + +3. Sharing of streaming and non-streaming mode SVE state +--------------------------------------------------------- + +It is implementation defined which if any parts of the SVE state are shared +between streaming and non-streaming modes. When switching between modes +via software interfaces such as ptrace if no register content is provided as +part of switching no state will be assumed to be shared and everything will +be zeroed. + + +4. System call behaviour +------------------------- + +* On syscall PSTATE.ZA is preserved, if PSTATE.ZA==1 then the contents of the + ZA matrix are preserved. + +* On syscall PSTATE.SM will be cleared and the SVE registers will be handled + as per the standard SVE ABI. + +* Neither the SVE registers nor ZA are used to pass arguments to or receive + results from any syscall. + +* On process creation (eg, clone()) the newly created process will have + PSTATE.SM cleared. + +* All other SME state of a thread, including the currently configured vector + length, the state of the PR_SME_VL_INHERIT flag, and the deferred vector + length (if any), is preserved across all syscalls, subject to the specific + exceptions for execve() described in section 6. + + +5. Signal handling +------------------- + +* Signal handlers are invoked with streaming mode and ZA disabled. + +* A new signal frame record za_context encodes the ZA register contents on + signal delivery. [1] + +* The signal frame record for ZA always contains basic metadata, in particular + the thread's vector length (in za_context.vl). + +* The ZA matrix may or may not be included in the record, depending on + the value of PSTATE.ZA. The registers are present if and only if: + za_context.head.size >= ZA_SIG_CONTEXT_SIZE(sve_vq_from_vl(za_context.vl)) + in which case PSTATE.ZA == 1. + +* If matrix data is present, the remainder of the record has a vl-dependent + size and layout. Macros ZA_SIG_* are defined [1] to facilitate access to + them. + +* The matrix is stored as a series of horizontal vectors in the same format as + is used for SVE vectors. + +* If the ZA context is too big to fit in sigcontext.__reserved[], then extra + space is allocated on the stack, an extra_context record is written in + __reserved[] referencing this space. za_context is then written in the + extra space. Refer to [1] for further details about this mechanism. + + +5. Signal return +----------------- + +When returning from a signal handler: + +* If there is no za_context record in the signal frame, or if the record is + present but contains no register data as described in the previous section, + then ZA is disabled. + +* If za_context is present in the signal frame and contains matrix data then + PSTATE.ZA is set to 1 and ZA is populated with the specified data. + +* The vector length cannot be changed via signal return. If za_context.vl in + the signal frame does not match the current vector length, the signal return + attempt is treated as illegal, resulting in a forced SIGSEGV. + + +6. prctl extensions +-------------------- + +Some new prctl() calls are added to allow programs to manage the SME vector +length: + +prctl(PR_SME_SET_VL, unsigned long arg) + + Sets the vector length of the calling thread and related flags, where + arg == vl | flags. Other threads of the calling process are unaffected. + + vl is the desired vector length, where sve_vl_valid(vl) must be true. + + flags: + + PR_SME_VL_INHERIT + + Inherit the current vector length across execve(). Otherwise, the + vector length is reset to the system default at execve(). (See + Section 9.) + + PR_SME_SET_VL_ONEXEC + + Defer the requested vector length change until the next execve() + performed by this thread. + + The effect is equivalent to implicit execution of the following + call immediately after the next execve() (if any) by the thread: + + prctl(PR_SME_SET_VL, arg & ~PR_SME_SET_VL_ONEXEC) + + This allows launching of a new program with a different vector + length, while avoiding runtime side effects in the caller. + + Without PR_SME_SET_VL_ONEXEC, the requested change takes effect + immediately. + + + Return value: a nonnegative on success, or a negative value on error: + EINVAL: SME not supported, invalid vector length requested, or + invalid flags. + + + On success: + + * Either the calling thread's vector length or the deferred vector length + to be applied at the next execve() by the thread (dependent on whether + PR_SME_SET_VL_ONEXEC is present in arg), is set to the largest value + supported by the system that is less than or equal to vl. If vl == + SVE_VL_MAX, the value set will be the largest value supported by the + system. + + * Any previously outstanding deferred vector length change in the calling + thread is cancelled. + + * The returned value describes the resulting configuration, encoded as for + PR_SME_GET_VL. The vector length reported in this value is the new + current vector length for this thread if PR_SME_SET_VL_ONEXEC was not + present in arg; otherwise, the reported vector length is the deferred + vector length that will be applied at the next execve() by the calling + thread. + + * Changing the vector length causes all of ZA, P0..P15, FFR and all bits of + Z0..Z31 except for Z0 bits [127:0] .. Z31 bits [127:0] to become + unspecified, including both streaming and non-streaming SVE state. + Calling PR_SME_SET_VL with vl equal to the thread's current vector + length, or calling PR_SME_SET_VL with the PR_SVE_SET_VL_ONEXEC flag, + does not constitute a change to the vector length for this purpose. + + * Changing the vector length causes PSTATE.ZA and PSTATE.SM to be cleared. + Calling PR_SME_SET_VL with vl equal to the thread's current vector + length, or calling PR_SME_SET_VL with the PR_SVE_SET_VL_ONEXEC flag, + does not constitute a change to the vector length for this purpose. + + +prctl(PR_SME_GET_VL) + + Gets the vector length of the calling thread. + + The following flag may be OR-ed into the result: + + PR_SME_VL_INHERIT + + Vector length will be inherited across execve(). + + There is no way to determine whether there is an outstanding deferred + vector length change (which would only normally be the case between a + fork() or vfork() and the corresponding execve() in typical use). + + To extract the vector length from the result, bitwise and it with + PR_SME_VL_LEN_MASK. + + Return value: a nonnegative value on success, or a negative value on error: + EINVAL: SME not supported. + + +7. ptrace extensions +--------------------- + +* A new regset NT_ARM_SSVE is defined for access to streaming mode SVE + state via PTRACE_GETREGSET and PTRACE_SETREGSET, this is documented in + sve.rst. + +* A new regset NT_ARM_ZA is defined for ZA state for access to ZA state via + PTRACE_GETREGSET and PTRACE_SETREGSET. + + Refer to [2] for definitions. + +The regset data starts with struct user_za_header, containing: + + size + + Size of the complete regset, in bytes. + This depends on vl and possibly on other things in the future. + + If a call to PTRACE_GETREGSET requests less data than the value of + size, the caller can allocate a larger buffer and retry in order to + read the complete regset. + + max_size + + Maximum size in bytes that the regset can grow to for the target + thread. The regset won't grow bigger than this even if the target + thread changes its vector length etc. + + vl + + Target thread's current streaming vector length, in bytes. + + max_vl + + Maximum possible streaming vector length for the target thread. + + flags + + Zero or more of the following flags, which have the same + meaning and behaviour as the corresponding PR_SET_VL_* flags: + + SME_PT_VL_INHERIT + + SME_PT_VL_ONEXEC (SETREGSET only). + +* The effects of changing the vector length and/or flags are equivalent to + those documented for PR_SME_SET_VL. + + The caller must make a further GETREGSET call if it needs to know what VL is + actually set by SETREGSET, unless is it known in advance that the requested + VL is supported. + +* The size and layout of the payload depends on the header fields. The + SME_PT_ZA_*() macros are provided to facilitate access to the data. + +* In either case, for SETREGSET it is permissible to omit the payload, in which + case the vector length and flags are changed and PSTATE.ZA is set to 0 + (along with any consequences of those changes). If a payload is provided + then PSTATE.ZA will be set to 1. + +* For SETREGSET, if the requested VL is not supported, the effect will be the + same as if the payload were omitted, except that an EIO error is reported. + No attempt is made to translate the payload data to the correct layout + for the vector length actually set. It is up to the caller to translate the + payload layout for the actual VL and retry. + +* The effect of writing a partial, incomplete payload is unspecified. + + +8. ELF coredump extensions +--------------------------- + +* NT_ARM_SSVE notes will be added to each coredump for + each thread of the dumped process. The contents will be equivalent to the + data that would have been read if a PTRACE_GETREGSET of the corresponding + type were executed for each thread when the coredump was generated. + +* A NT_ARM_ZA note will be added to each coredump for each thread of the + dumped process. The contents will be equivalent to the data that would have + been read if a PTRACE_GETREGSET of NT_ARM_ZA were executed for each thread + when the coredump was generated. + +* The NT_ARM_TLS note will be extended to two registers, the second register + will contain TPIDR2_EL0 on systems that support SME and will be read as + zero with writes ignored otherwise. + +9. System runtime configuration +-------------------------------- + +* To mitigate the ABI impact of expansion of the signal frame, a policy + mechanism is provided for administrators, distro maintainers and developers + to set the default vector length for userspace processes: + +/proc/sys/abi/sme_default_vector_length + + Writing the text representation of an integer to this file sets the system + default vector length to the specified value, unless the value is greater + than the maximum vector length supported by the system in which case the + default vector length is set to that maximum. + + The result can be determined by reopening the file and reading its + contents. + + At boot, the default vector length is initially set to 32 or the maximum + supported vector length, whichever is smaller and supported. This + determines the initial vector length of the init process (PID 1). + + Reading this file returns the current system default vector length. + +* At every execve() call, the new vector length of the new process is set to + the system default vector length, unless + + * PR_SME_VL_INHERIT (or equivalently SME_PT_VL_INHERIT) is set for the + calling thread, or + + * a deferred vector length change is pending, established via the + PR_SME_SET_VL_ONEXEC flag (or SME_PT_VL_ONEXEC). + +* Modifying the system default vector length does not affect the vector length + of any existing process or thread that does not make an execve() call. + + +Appendix A. SME programmer's model (informative) +================================================= + +This section provides a minimal description of the additions made by SME to the +ARMv8-A programmer's model that are relevant to this document. + +Note: This section is for information only and not intended to be complete or +to replace any architectural specification. + +A.1. Registers +--------------- + +In A64 state, SME adds the following: + +* A new mode, streaming mode, in which a subset of the normal FPSIMD and SVE + features are available. When supported EL0 software may enter and leave + streaming mode at any time. + + For best system performance it is strongly encouraged for software to enable + streaming mode only when it is actively being used. + +* A new vector length controlling the size of ZA and the Z registers when in + streaming mode, separately to the vector length used for SVE when not in + streaming mode. There is no requirement that either the currently selected + vector length or the set of vector lengths supported for the two modes in + a given system have any relationship. The streaming mode vector length + is referred to as SVL. + +* A new ZA matrix register. This is a square matrix of SVLxSVL bits. Most + operations on ZA require that streaming mode be enabled but ZA can be + enabled without streaming mode in order to load, save and retain data. + + For best system performance it is strongly encouraged for software to enable + ZA only when it is actively being used. + +* Two new 1 bit fields in PSTATE which may be controlled via the SMSTART and + SMSTOP instructions or by access to the SVCR system register: + + * PSTATE.ZA, if this is 1 then the ZA matrix is accessible and has valid + data while if it is 0 then ZA can not be accessed. When PSTATE.ZA is + changed from 0 to 1 all bits in ZA are cleared. + + * PSTATE.SM, if this is 1 then the PE is in streaming mode. When the value + of PSTATE.SM is changed then it is implementation defined if the subset + of the floating point register bits valid in both modes may be retained. + Any other bits will be cleared. + + +References +========== + +[1] arch/arm64/include/uapi/asm/sigcontext.h + AArch64 Linux signal ABI definitions + +[2] arch/arm64/include/uapi/asm/ptrace.h + AArch64 Linux ptrace ABI definitions + +[3] Documentation/arm64/cpu-feature-registers.rst diff --git a/Documentation/arm64/sve.rst b/Documentation/arm64/sve.rst new file mode 100644 index 000000000..f338ee2df --- /dev/null +++ b/Documentation/arm64/sve.rst @@ -0,0 +1,615 @@ +=================================================== +Scalable Vector Extension support for AArch64 Linux +=================================================== + +Author: Dave Martin <Dave.Martin@arm.com> + +Date: 4 August 2017 + +This document outlines briefly the interface provided to userspace by Linux in +order to support use of the ARM Scalable Vector Extension (SVE), including +interactions with Streaming SVE mode added by the Scalable Matrix Extension +(SME). + +This is an outline of the most important features and issues only and not +intended to be exhaustive. + +This document does not aim to describe the SVE architecture or programmer's +model. To aid understanding, a minimal description of relevant programmer's +model features for SVE is included in Appendix A. + + +1. General +----------- + +* SVE registers Z0..Z31, P0..P15 and FFR and the current vector length VL, are + tracked per-thread. + +* In streaming mode FFR is not accessible unless HWCAP2_SME_FA64 is present + in the system, when it is not supported and these interfaces are used to + access streaming mode FFR is read and written as zero. + +* The presence of SVE is reported to userspace via HWCAP_SVE in the aux vector + AT_HWCAP entry. Presence of this flag implies the presence of the SVE + instructions and registers, and the Linux-specific system interfaces + described in this document. SVE is reported in /proc/cpuinfo as "sve". + +* Support for the execution of SVE instructions in userspace can also be + detected by reading the CPU ID register ID_AA64PFR0_EL1 using an MRS + instruction, and checking that the value of the SVE field is nonzero. [3] + + It does not guarantee the presence of the system interfaces described in the + following sections: software that needs to verify that those interfaces are + present must check for HWCAP_SVE instead. + +* On hardware that supports the SVE2 extensions, HWCAP2_SVE2 will also + be reported in the AT_HWCAP2 aux vector entry. In addition to this, + optional extensions to SVE2 may be reported by the presence of: + + HWCAP2_SVE2 + HWCAP2_SVEAES + HWCAP2_SVEPMULL + HWCAP2_SVEBITPERM + HWCAP2_SVESHA3 + HWCAP2_SVESM4 + + This list may be extended over time as the SVE architecture evolves. + + These extensions are also reported via the CPU ID register ID_AA64ZFR0_EL1, + which userspace can read using an MRS instruction. See elf_hwcaps.txt and + cpu-feature-registers.txt for details. + +* On hardware that supports the SME extensions, HWCAP2_SME will also be + reported in the AT_HWCAP2 aux vector entry. Among other things SME adds + streaming mode which provides a subset of the SVE feature set using a + separate SME vector length and the same Z/V registers. See sme.rst + for more details. + +* Debuggers should restrict themselves to interacting with the target via the + NT_ARM_SVE regset. The recommended way of detecting support for this regset + is to connect to a target process first and then attempt a + ptrace(PTRACE_GETREGSET, pid, NT_ARM_SVE, &iov). Note that when SME is + present and streaming SVE mode is in use the FPSIMD subset of registers + will be read via NT_ARM_SVE and NT_ARM_SVE writes will exit streaming mode + in the target. + +* Whenever SVE scalable register values (Zn, Pn, FFR) are exchanged in memory + between userspace and the kernel, the register value is encoded in memory in + an endianness-invariant layout, with bits [(8 * i + 7) : (8 * i)] encoded at + byte offset i from the start of the memory representation. This affects for + example the signal frame (struct sve_context) and ptrace interface + (struct user_sve_header) and associated data. + + Beware that on big-endian systems this results in a different byte order than + for the FPSIMD V-registers, which are stored as single host-endian 128-bit + values, with bits [(127 - 8 * i) : (120 - 8 * i)] of the register encoded at + byte offset i. (struct fpsimd_context, struct user_fpsimd_state). + + +2. Vector length terminology +----------------------------- + +The size of an SVE vector (Z) register is referred to as the "vector length". + +To avoid confusion about the units used to express vector length, the kernel +adopts the following conventions: + +* Vector length (VL) = size of a Z-register in bytes + +* Vector quadwords (VQ) = size of a Z-register in units of 128 bits + +(So, VL = 16 * VQ.) + +The VQ convention is used where the underlying granularity is important, such +as in data structure definitions. In most other situations, the VL convention +is used. This is consistent with the meaning of the "VL" pseudo-register in +the SVE instruction set architecture. + + +3. System call behaviour +------------------------- + +* On syscall, V0..V31 are preserved (as without SVE). Thus, bits [127:0] of + Z0..Z31 are preserved. All other bits of Z0..Z31, and all of P0..P15 and FFR + become zero on return from a syscall. + +* The SVE registers are not used to pass arguments to or receive results from + any syscall. + +* In practice the affected registers/bits will be preserved or will be replaced + with zeros on return from a syscall, but userspace should not make + assumptions about this. The kernel behaviour may vary on a case-by-case + basis. + +* All other SVE state of a thread, including the currently configured vector + length, the state of the PR_SVE_VL_INHERIT flag, and the deferred vector + length (if any), is preserved across all syscalls, subject to the specific + exceptions for execve() described in section 6. + + In particular, on return from a fork() or clone(), the parent and new child + process or thread share identical SVE configuration, matching that of the + parent before the call. + + +4. Signal handling +------------------- + +* A new signal frame record sve_context encodes the SVE registers on signal + delivery. [1] + +* This record is supplementary to fpsimd_context. The FPSR and FPCR registers + are only present in fpsimd_context. For convenience, the content of V0..V31 + is duplicated between sve_context and fpsimd_context. + +* The record contains a flag field which includes a flag SVE_SIG_FLAG_SM which + if set indicates that the thread is in streaming mode and the vector length + and register data (if present) describe the streaming SVE data and vector + length. + +* The signal frame record for SVE always contains basic metadata, in particular + the thread's vector length (in sve_context.vl). + +* The SVE registers may or may not be included in the record, depending on + whether the registers are live for the thread. The registers are present if + and only if: + sve_context.head.size >= SVE_SIG_CONTEXT_SIZE(sve_vq_from_vl(sve_context.vl)). + +* If the registers are present, the remainder of the record has a vl-dependent + size and layout. Macros SVE_SIG_* are defined [1] to facilitate access to + the members. + +* Each scalable register (Zn, Pn, FFR) is stored in an endianness-invariant + layout, with bits [(8 * i + 7) : (8 * i)] stored at byte offset i from the + start of the register's representation in memory. + +* If the SVE context is too big to fit in sigcontext.__reserved[], then extra + space is allocated on the stack, an extra_context record is written in + __reserved[] referencing this space. sve_context is then written in the + extra space. Refer to [1] for further details about this mechanism. + + +5. Signal return +----------------- + +When returning from a signal handler: + +* If there is no sve_context record in the signal frame, or if the record is + present but contains no register data as desribed in the previous section, + then the SVE registers/bits become non-live and take unspecified values. + +* If sve_context is present in the signal frame and contains full register + data, the SVE registers become live and are populated with the specified + data. However, for backward compatibility reasons, bits [127:0] of Z0..Z31 + are always restored from the corresponding members of fpsimd_context.vregs[] + and not from sve_context. The remaining bits are restored from sve_context. + +* Inclusion of fpsimd_context in the signal frame remains mandatory, + irrespective of whether sve_context is present or not. + +* The vector length cannot be changed via signal return. If sve_context.vl in + the signal frame does not match the current vector length, the signal return + attempt is treated as illegal, resulting in a forced SIGSEGV. + +* It is permitted to enter or leave streaming mode by setting or clearing + the SVE_SIG_FLAG_SM flag but applications should take care to ensure that + when doing so sve_context.vl and any register data are appropriate for the + vector length in the new mode. + + +6. prctl extensions +-------------------- + +Some new prctl() calls are added to allow programs to manage the SVE vector +length: + +prctl(PR_SVE_SET_VL, unsigned long arg) + + Sets the vector length of the calling thread and related flags, where + arg == vl | flags. Other threads of the calling process are unaffected. + + vl is the desired vector length, where sve_vl_valid(vl) must be true. + + flags: + + PR_SVE_VL_INHERIT + + Inherit the current vector length across execve(). Otherwise, the + vector length is reset to the system default at execve(). (See + Section 9.) + + PR_SVE_SET_VL_ONEXEC + + Defer the requested vector length change until the next execve() + performed by this thread. + + The effect is equivalent to implicit exceution of the following + call immediately after the next execve() (if any) by the thread: + + prctl(PR_SVE_SET_VL, arg & ~PR_SVE_SET_VL_ONEXEC) + + This allows launching of a new program with a different vector + length, while avoiding runtime side effects in the caller. + + + Without PR_SVE_SET_VL_ONEXEC, the requested change takes effect + immediately. + + + Return value: a nonnegative on success, or a negative value on error: + EINVAL: SVE not supported, invalid vector length requested, or + invalid flags. + + + On success: + + * Either the calling thread's vector length or the deferred vector length + to be applied at the next execve() by the thread (dependent on whether + PR_SVE_SET_VL_ONEXEC is present in arg), is set to the largest value + supported by the system that is less than or equal to vl. If vl == + SVE_VL_MAX, the value set will be the largest value supported by the + system. + + * Any previously outstanding deferred vector length change in the calling + thread is cancelled. + + * The returned value describes the resulting configuration, encoded as for + PR_SVE_GET_VL. The vector length reported in this value is the new + current vector length for this thread if PR_SVE_SET_VL_ONEXEC was not + present in arg; otherwise, the reported vector length is the deferred + vector length that will be applied at the next execve() by the calling + thread. + + * Changing the vector length causes all of P0..P15, FFR and all bits of + Z0..Z31 except for Z0 bits [127:0] .. Z31 bits [127:0] to become + unspecified. Calling PR_SVE_SET_VL with vl equal to the thread's current + vector length, or calling PR_SVE_SET_VL with the PR_SVE_SET_VL_ONEXEC + flag, does not constitute a change to the vector length for this purpose. + + +prctl(PR_SVE_GET_VL) + + Gets the vector length of the calling thread. + + The following flag may be OR-ed into the result: + + PR_SVE_VL_INHERIT + + Vector length will be inherited across execve(). + + There is no way to determine whether there is an outstanding deferred + vector length change (which would only normally be the case between a + fork() or vfork() and the corresponding execve() in typical use). + + To extract the vector length from the result, bitwise and it with + PR_SVE_VL_LEN_MASK. + + Return value: a nonnegative value on success, or a negative value on error: + EINVAL: SVE not supported. + + +7. ptrace extensions +--------------------- + +* New regsets NT_ARM_SVE and NT_ARM_SSVE are defined for use with + PTRACE_GETREGSET and PTRACE_SETREGSET. NT_ARM_SSVE describes the + streaming mode SVE registers and NT_ARM_SVE describes the + non-streaming mode SVE registers. + + In this description a register set is referred to as being "live" when + the target is in the appropriate streaming or non-streaming mode and is + using data beyond the subset shared with the FPSIMD Vn registers. + + Refer to [2] for definitions. + +The regset data starts with struct user_sve_header, containing: + + size + + Size of the complete regset, in bytes. + This depends on vl and possibly on other things in the future. + + If a call to PTRACE_GETREGSET requests less data than the value of + size, the caller can allocate a larger buffer and retry in order to + read the complete regset. + + max_size + + Maximum size in bytes that the regset can grow to for the target + thread. The regset won't grow bigger than this even if the target + thread changes its vector length etc. + + vl + + Target thread's current vector length, in bytes. + + max_vl + + Maximum possible vector length for the target thread. + + flags + + at most one of + + SVE_PT_REGS_FPSIMD + + SVE registers are not live (GETREGSET) or are to be made + non-live (SETREGSET). + + The payload is of type struct user_fpsimd_state, with the same + meaning as for NT_PRFPREG, starting at offset + SVE_PT_FPSIMD_OFFSET from the start of user_sve_header. + + Extra data might be appended in the future: the size of the + payload should be obtained using SVE_PT_FPSIMD_SIZE(vq, flags). + + vq should be obtained using sve_vq_from_vl(vl). + + or + + SVE_PT_REGS_SVE + + SVE registers are live (GETREGSET) or are to be made live + (SETREGSET). + + The payload contains the SVE register data, starting at offset + SVE_PT_SVE_OFFSET from the start of user_sve_header, and with + size SVE_PT_SVE_SIZE(vq, flags); + + ... OR-ed with zero or more of the following flags, which have the same + meaning and behaviour as the corresponding PR_SET_VL_* flags: + + SVE_PT_VL_INHERIT + + SVE_PT_VL_ONEXEC (SETREGSET only). + + If neither FPSIMD nor SVE flags are provided then no register + payload is available, this is only possible when SME is implemented. + + +* The effects of changing the vector length and/or flags are equivalent to + those documented for PR_SVE_SET_VL. + + The caller must make a further GETREGSET call if it needs to know what VL is + actually set by SETREGSET, unless is it known in advance that the requested + VL is supported. + +* In the SVE_PT_REGS_SVE case, the size and layout of the payload depends on + the header fields. The SVE_PT_SVE_*() macros are provided to facilitate + access to the members. + +* In either case, for SETREGSET it is permissible to omit the payload, in which + case only the vector length and flags are changed (along with any + consequences of those changes). + +* In systems supporting SME when in streaming mode a GETREGSET for + NT_REG_SVE will return only the user_sve_header with no register data, + similarly a GETREGSET for NT_REG_SSVE will not return any register data + when not in streaming mode. + +* A GETREGSET for NT_ARM_SSVE will never return SVE_PT_REGS_FPSIMD. + +* For SETREGSET, if an SVE_PT_REGS_SVE payload is present and the + requested VL is not supported, the effect will be the same as if the + payload were omitted, except that an EIO error is reported. No + attempt is made to translate the payload data to the correct layout + for the vector length actually set. The thread's FPSIMD state is + preserved, but the remaining bits of the SVE registers become + unspecified. It is up to the caller to translate the payload layout + for the actual VL and retry. + +* Where SME is implemented it is not possible to GETREGSET the register + state for normal SVE when in streaming mode, nor the streaming mode + register state when in normal mode, regardless of the implementation defined + behaviour of the hardware for sharing data between the two modes. + +* Any SETREGSET of NT_ARM_SVE will exit streaming mode if the target was in + streaming mode and any SETREGSET of NT_ARM_SSVE will enter streaming mode + if the target was not in streaming mode. + +* The effect of writing a partial, incomplete payload is unspecified. + + +8. ELF coredump extensions +--------------------------- + +* NT_ARM_SVE and NT_ARM_SSVE notes will be added to each coredump for + each thread of the dumped process. The contents will be equivalent to the + data that would have been read if a PTRACE_GETREGSET of the corresponding + type were executed for each thread when the coredump was generated. + +9. System runtime configuration +-------------------------------- + +* To mitigate the ABI impact of expansion of the signal frame, a policy + mechanism is provided for administrators, distro maintainers and developers + to set the default vector length for userspace processes: + +/proc/sys/abi/sve_default_vector_length + + Writing the text representation of an integer to this file sets the system + default vector length to the specified value, unless the value is greater + than the maximum vector length supported by the system in which case the + default vector length is set to that maximum. + + The result can be determined by reopening the file and reading its + contents. + + At boot, the default vector length is initially set to 64 or the maximum + supported vector length, whichever is smaller. This determines the initial + vector length of the init process (PID 1). + + Reading this file returns the current system default vector length. + +* At every execve() call, the new vector length of the new process is set to + the system default vector length, unless + + * PR_SVE_VL_INHERIT (or equivalently SVE_PT_VL_INHERIT) is set for the + calling thread, or + + * a deferred vector length change is pending, established via the + PR_SVE_SET_VL_ONEXEC flag (or SVE_PT_VL_ONEXEC). + +* Modifying the system default vector length does not affect the vector length + of any existing process or thread that does not make an execve() call. + +10. Perf extensions +-------------------------------- + +* The arm64 specific DWARF standard [5] added the VG (Vector Granule) register + at index 46. This register is used for DWARF unwinding when variable length + SVE registers are pushed onto the stack. + +* Its value is equivalent to the current SVE vector length (VL) in bits divided + by 64. + +* The value is included in Perf samples in the regs[46] field if + PERF_SAMPLE_REGS_USER is set and the sample_regs_user mask has bit 46 set. + +* The value is the current value at the time the sample was taken, and it can + change over time. + +* If the system doesn't support SVE when perf_event_open is called with these + settings, the event will fail to open. + +Appendix A. SVE programmer's model (informative) +================================================= + +This section provides a minimal description of the additions made by SVE to the +ARMv8-A programmer's model that are relevant to this document. + +Note: This section is for information only and not intended to be complete or +to replace any architectural specification. + +A.1. Registers +--------------- + +In A64 state, SVE adds the following: + +* 32 8VL-bit vector registers Z0..Z31 + For each Zn, Zn bits [127:0] alias the ARMv8-A vector register Vn. + + A register write using a Vn register name zeros all bits of the corresponding + Zn except for bits [127:0]. + +* 16 VL-bit predicate registers P0..P15 + +* 1 VL-bit special-purpose predicate register FFR (the "first-fault register") + +* a VL "pseudo-register" that determines the size of each vector register + + The SVE instruction set architecture provides no way to write VL directly. + Instead, it can be modified only by EL1 and above, by writing appropriate + system registers. + +* The value of VL can be configured at runtime by EL1 and above: + 16 <= VL <= VLmax, where VL must be a multiple of 16. + +* The maximum vector length is determined by the hardware: + 16 <= VLmax <= 256. + + (The SVE architecture specifies 256, but permits future architecture + revisions to raise this limit.) + +* FPSR and FPCR are retained from ARMv8-A, and interact with SVE floating-point + operations in a similar way to the way in which they interact with ARMv8 + floating-point operations:: + + 8VL-1 128 0 bit index + +---- //// -----------------+ + Z0 | : V0 | + : : + Z7 | : V7 | + Z8 | : * V8 | + : : : + Z15 | : *V15 | + Z16 | : V16 | + : : + Z31 | : V31 | + +---- //// -----------------+ + 31 0 + VL-1 0 +-------+ + +---- //// --+ FPSR | | + P0 | | +-------+ + : | | *FPCR | | + P15 | | +-------+ + +---- //// --+ + FFR | | +-----+ + +---- //// --+ VL | | + +-----+ + +(*) callee-save: + This only applies to bits [63:0] of Z-/V-registers. + FPCR contains callee-save and caller-save bits. See [4] for details. + + +A.2. Procedure call standard +----------------------------- + +The ARMv8-A base procedure call standard is extended as follows with respect to +the additional SVE register state: + +* All SVE register bits that are not shared with FP/SIMD are caller-save. + +* Z8 bits [63:0] .. Z15 bits [63:0] are callee-save. + + This follows from the way these bits are mapped to V8..V15, which are caller- + save in the base procedure call standard. + + +Appendix B. ARMv8-A FP/SIMD programmer's model +=============================================== + +Note: This section is for information only and not intended to be complete or +to replace any architectural specification. + +Refer to [4] for more information. + +ARMv8-A defines the following floating-point / SIMD register state: + +* 32 128-bit vector registers V0..V31 +* 2 32-bit status/control registers FPSR, FPCR + +:: + + 127 0 bit index + +---------------+ + V0 | | + : : : + V7 | | + * V8 | | + : : : : + *V15 | | + V16 | | + : : : + V31 | | + +---------------+ + + 31 0 + +-------+ + FPSR | | + +-------+ + *FPCR | | + +-------+ + +(*) callee-save: + This only applies to bits [63:0] of V-registers. + FPCR contains a mixture of callee-save and caller-save bits. + + +References +========== + +[1] arch/arm64/include/uapi/asm/sigcontext.h + AArch64 Linux signal ABI definitions + +[2] arch/arm64/include/uapi/asm/ptrace.h + AArch64 Linux ptrace ABI definitions + +[3] Documentation/arm64/cpu-feature-registers.rst + +[4] ARM IHI0055C + http://infocenter.arm.com/help/topic/com.arm.doc.ihi0055c/IHI0055C_beta_aapcs64.pdf + http://infocenter.arm.com/help/topic/com.arm.doc.subset.swdev.abi/index.html + Procedure Call Standard for the ARM 64-bit Architecture (AArch64) + +[5] https://github.com/ARM-software/abi-aa/blob/main/aadwarf64/aadwarf64.rst diff --git a/Documentation/arm64/tagged-address-abi.rst b/Documentation/arm64/tagged-address-abi.rst new file mode 100644 index 000000000..540a1d4fc --- /dev/null +++ b/Documentation/arm64/tagged-address-abi.rst @@ -0,0 +1,179 @@ +========================== +AArch64 TAGGED ADDRESS ABI +========================== + +Authors: Vincenzo Frascino <vincenzo.frascino@arm.com> + Catalin Marinas <catalin.marinas@arm.com> + +Date: 21 August 2019 + +This document describes the usage and semantics of the Tagged Address +ABI on AArch64 Linux. + +1. Introduction +--------------- + +On AArch64 the ``TCR_EL1.TBI0`` bit is set by default, allowing +userspace (EL0) to perform memory accesses through 64-bit pointers with +a non-zero top byte. This document describes the relaxation of the +syscall ABI that allows userspace to pass certain tagged pointers to +kernel syscalls. + +2. AArch64 Tagged Address ABI +----------------------------- + +From the kernel syscall interface perspective and for the purposes of +this document, a "valid tagged pointer" is a pointer with a potentially +non-zero top-byte that references an address in the user process address +space obtained in one of the following ways: + +- ``mmap()`` syscall where either: + + - flags have the ``MAP_ANONYMOUS`` bit set or + - the file descriptor refers to a regular file (including those + returned by ``memfd_create()``) or ``/dev/zero`` + +- ``brk()`` syscall (i.e. the heap area between the initial location of + the program break at process creation and its current location). + +- any memory mapped by the kernel in the address space of the process + during creation and with the same restrictions as for ``mmap()`` above + (e.g. data, bss, stack). + +The AArch64 Tagged Address ABI has two stages of relaxation depending on +how the user addresses are used by the kernel: + +1. User addresses not accessed by the kernel but used for address space + management (e.g. ``mprotect()``, ``madvise()``). The use of valid + tagged pointers in this context is allowed with these exceptions: + + - ``brk()``, ``mmap()`` and the ``new_address`` argument to + ``mremap()`` as these have the potential to alias with existing + user addresses. + + NOTE: This behaviour changed in v5.6 and so some earlier kernels may + incorrectly accept valid tagged pointers for the ``brk()``, + ``mmap()`` and ``mremap()`` system calls. + + - The ``range.start``, ``start`` and ``dst`` arguments to the + ``UFFDIO_*`` ``ioctl()``s used on a file descriptor obtained from + ``userfaultfd()``, as fault addresses subsequently obtained by reading + the file descriptor will be untagged, which may otherwise confuse + tag-unaware programs. + + NOTE: This behaviour changed in v5.14 and so some earlier kernels may + incorrectly accept valid tagged pointers for this system call. + +2. User addresses accessed by the kernel (e.g. ``write()``). This ABI + relaxation is disabled by default and the application thread needs to + explicitly enable it via ``prctl()`` as follows: + + - ``PR_SET_TAGGED_ADDR_CTRL``: enable or disable the AArch64 Tagged + Address ABI for the calling thread. + + The ``(unsigned int) arg2`` argument is a bit mask describing the + control mode used: + + - ``PR_TAGGED_ADDR_ENABLE``: enable AArch64 Tagged Address ABI. + Default status is disabled. + + Arguments ``arg3``, ``arg4``, and ``arg5`` must be 0. + + - ``PR_GET_TAGGED_ADDR_CTRL``: get the status of the AArch64 Tagged + Address ABI for the calling thread. + + Arguments ``arg2``, ``arg3``, ``arg4``, and ``arg5`` must be 0. + + The ABI properties described above are thread-scoped, inherited on + clone() and fork() and cleared on exec(). + + Calling ``prctl(PR_SET_TAGGED_ADDR_CTRL, PR_TAGGED_ADDR_ENABLE, 0, 0, 0)`` + returns ``-EINVAL`` if the AArch64 Tagged Address ABI is globally + disabled by ``sysctl abi.tagged_addr_disabled=1``. The default + ``sysctl abi.tagged_addr_disabled`` configuration is 0. + +When the AArch64 Tagged Address ABI is enabled for a thread, the +following behaviours are guaranteed: + +- All syscalls except the cases mentioned in section 3 can accept any + valid tagged pointer. + +- The syscall behaviour is undefined for invalid tagged pointers: it may + result in an error code being returned, a (fatal) signal being raised, + or other modes of failure. + +- The syscall behaviour for a valid tagged pointer is the same as for + the corresponding untagged pointer. + + +A definition of the meaning of tagged pointers on AArch64 can be found +in Documentation/arm64/tagged-pointers.rst. + +3. AArch64 Tagged Address ABI Exceptions +----------------------------------------- + +The following system call parameters must be untagged regardless of the +ABI relaxation: + +- ``prctl()`` other than pointers to user data either passed directly or + indirectly as arguments to be accessed by the kernel. + +- ``ioctl()`` other than pointers to user data either passed directly or + indirectly as arguments to be accessed by the kernel. + +- ``shmat()`` and ``shmdt()``. + +- ``brk()`` (since kernel v5.6). + +- ``mmap()`` (since kernel v5.6). + +- ``mremap()``, the ``new_address`` argument (since kernel v5.6). + +Any attempt to use non-zero tagged pointers may result in an error code +being returned, a (fatal) signal being raised, or other modes of +failure. + +4. Example of correct usage +--------------------------- +.. code-block:: c + + #include <stdlib.h> + #include <string.h> + #include <unistd.h> + #include <sys/mman.h> + #include <sys/prctl.h> + + #define PR_SET_TAGGED_ADDR_CTRL 55 + #define PR_TAGGED_ADDR_ENABLE (1UL << 0) + + #define TAG_SHIFT 56 + + int main(void) + { + int tbi_enabled = 0; + unsigned long tag = 0; + char *ptr; + + /* check/enable the tagged address ABI */ + if (!prctl(PR_SET_TAGGED_ADDR_CTRL, PR_TAGGED_ADDR_ENABLE, 0, 0, 0)) + tbi_enabled = 1; + + /* memory allocation */ + ptr = mmap(NULL, sysconf(_SC_PAGE_SIZE), PROT_READ | PROT_WRITE, + MAP_PRIVATE | MAP_ANONYMOUS, -1, 0); + if (ptr == MAP_FAILED) + return 1; + + /* set a non-zero tag if the ABI is available */ + if (tbi_enabled) + tag = rand() & 0xff; + ptr = (char *)((unsigned long)ptr | (tag << TAG_SHIFT)); + + /* memory access to a tagged address */ + strcpy(ptr, "tagged pointer\n"); + + /* syscall with a tagged pointer */ + write(1, ptr, strlen(ptr)); + + return 0; + } diff --git a/Documentation/arm64/tagged-pointers.rst b/Documentation/arm64/tagged-pointers.rst new file mode 100644 index 000000000..19d284b70 --- /dev/null +++ b/Documentation/arm64/tagged-pointers.rst @@ -0,0 +1,88 @@ +========================================= +Tagged virtual addresses in AArch64 Linux +========================================= + +Author: Will Deacon <will.deacon@arm.com> + +Date : 12 June 2013 + +This document briefly describes the provision of tagged virtual +addresses in the AArch64 translation system and their potential uses +in AArch64 Linux. + +The kernel configures the translation tables so that translations made +via TTBR0 (i.e. userspace mappings) have the top byte (bits 63:56) of +the virtual address ignored by the translation hardware. This frees up +this byte for application use. + + +Passing tagged addresses to the kernel +-------------------------------------- + +All interpretation of userspace memory addresses by the kernel assumes +an address tag of 0x00, unless the application enables the AArch64 +Tagged Address ABI explicitly +(Documentation/arm64/tagged-address-abi.rst). + +This includes, but is not limited to, addresses found in: + + - pointer arguments to system calls, including pointers in structures + passed to system calls, + + - the stack pointer (sp), e.g. when interpreting it to deliver a + signal, + + - the frame pointer (x29) and frame records, e.g. when interpreting + them to generate a backtrace or call graph. + +Using non-zero address tags in any of these locations when the +userspace application did not enable the AArch64 Tagged Address ABI may +result in an error code being returned, a (fatal) signal being raised, +or other modes of failure. + +For these reasons, when the AArch64 Tagged Address ABI is disabled, +passing non-zero address tags to the kernel via system calls is +forbidden, and using a non-zero address tag for sp is strongly +discouraged. + +Programs maintaining a frame pointer and frame records that use non-zero +address tags may suffer impaired or inaccurate debug and profiling +visibility. + + +Preserving tags +--------------- + +When delivering signals, non-zero tags are not preserved in +siginfo.si_addr unless the flag SA_EXPOSE_TAGBITS was set in +sigaction.sa_flags when the signal handler was installed. This means +that signal handlers in applications making use of tags cannot rely +on the tag information for user virtual addresses being maintained +in these fields unless the flag was set. + +Due to architecture limitations, bits 63:60 of the fault address +are not preserved in response to synchronous tag check faults +(SEGV_MTESERR) even if SA_EXPOSE_TAGBITS was set. Applications should +treat the values of these bits as undefined in order to accommodate +future architecture revisions which may preserve the bits. + +For signals raised in response to watchpoint debug exceptions, the +tag information will be preserved regardless of the SA_EXPOSE_TAGBITS +flag setting. + +Non-zero tags are never preserved in sigcontext.fault_address +regardless of the SA_EXPOSE_TAGBITS flag setting. + +The architecture prevents the use of a tagged PC, so the upper byte will +be set to a sign-extension of bit 55 on exception return. + +This behaviour is maintained when the AArch64 Tagged Address ABI is +enabled. + + +Other considerations +-------------------- + +Special care should be taken when using tagged pointers, since it is +likely that C compilers will not hazard two virtual addresses differing +only in the upper byte. |