diff options
Diffstat (limited to 'arch/powerpc/platforms/pseries')
41 files changed, 2674 insertions, 574 deletions
diff --git a/arch/powerpc/platforms/pseries/Kconfig b/arch/powerpc/platforms/pseries/Kconfig index 2e57391e0778..a3b4d99567cb 100644 --- a/arch/powerpc/platforms/pseries/Kconfig +++ b/arch/powerpc/platforms/pseries/Kconfig @@ -17,21 +17,27 @@ config PPC_PSERIES select PPC_RTAS_DAEMON select RTAS_ERROR_LOGGING select PPC_UDBG_16550 - select PPC_NATIVE select PPC_DOORBELL select HOTPLUG_CPU - select ARCH_RANDOM select FORCE_SMP select SWIOTLB default y +config PARAVIRT + bool + config PARAVIRT_SPINLOCKS bool +config PARAVIRT_TIME_ACCOUNTING + select PARAVIRT + bool + config PPC_SPLPAR bool "Support for shared-processor logical partitions" depends on PPC_PSERIES select PARAVIRT_SPINLOCKS if PPC_QUEUED_SPINLOCKS + select PARAVIRT_TIME_ACCOUNTING if VIRT_CPU_ACCOUNTING_GEN default y help Enabling this option will make the kernel run more efficiently @@ -61,10 +67,6 @@ config PSERIES_ENERGY Provides: /sys/devices/system/cpu/pseries_(de)activation_hint_list and /sys/devices/system/cpu/cpuN/pseries_(de)activation_hint -config SCANLOG - tristate "Scanlog dump interface" - depends on RTAS_PROC && PPC_PSERIES - config IO_EVENT_IRQ bool "IO Event Interrupt support" depends on PPC_PSERIES @@ -147,6 +149,19 @@ config IBMEBUS help Bus device driver for GX bus based adapters. +config PSERIES_PLPKS + depends on PPC_PSERIES + bool "Support for the Platform Key Storage" + help + PowerVM provides an isolated Platform Keystore(PKS) storage + allocation for each LPAR with individually managed access + controls to store sensitive information securely. It can be + used to store asymmetric public keys or secrets as required + by different usecases. Select this config to enable + operating system interface to hypervisor to access this space. + + If unsure, select N. + config PAPR_SCM depends on PPC_PSERIES && MEMORY_HOTPLUG && LIBNVDIMM tristate "Support for the PAPR Storage Class Memory interface" diff --git a/arch/powerpc/platforms/pseries/Makefile b/arch/powerpc/platforms/pseries/Makefile index 41d8aee98da4..92310202bdd7 100644 --- a/arch/powerpc/platforms/pseries/Makefile +++ b/arch/powerpc/platforms/pseries/Makefile @@ -6,9 +6,9 @@ obj-y := lpar.o hvCall.o nvram.o reconfig.o \ of_helpers.o \ setup.o iommu.o event_sources.o ras.o \ firmware.o power.o dlpar.o mobility.o rng.o \ - pci.o pci_dlpar.o eeh_pseries.o msi.o + pci.o pci_dlpar.o eeh_pseries.o msi.o \ + papr_platform_attributes.o dtl.o obj-$(CONFIG_SMP) += smp.o -obj-$(CONFIG_SCANLOG) += scanlog.o obj-$(CONFIG_KEXEC_CORE) += kexec.o obj-$(CONFIG_PSERIES_ENERGY) += pseries_energy.o @@ -19,7 +19,6 @@ obj-$(CONFIG_HVC_CONSOLE) += hvconsole.o obj-$(CONFIG_HVCS) += hvcserver.o obj-$(CONFIG_HCALL_STATS) += hvCall_inst.o obj-$(CONFIG_CMM) += cmm.o -obj-$(CONFIG_DTL) += dtl.o obj-$(CONFIG_IO_EVENT_IRQ) += io_event_irq.o obj-$(CONFIG_LPARCFG) += lparcfg.o obj-$(CONFIG_IBMVIO) += vio.o @@ -28,8 +27,13 @@ obj-$(CONFIG_PAPR_SCM) += papr_scm.o obj-$(CONFIG_PPC_SPLPAR) += vphn.o obj-$(CONFIG_PPC_SVM) += svm.o obj-$(CONFIG_FA_DUMP) += rtas-fadump.o +obj-$(CONFIG_PSERIES_PLPKS) += plpks.o obj-$(CONFIG_SUSPEND) += suspend.o -obj-$(CONFIG_PPC_VAS) += vas.o +obj-$(CONFIG_PPC_VAS) += vas.o vas-sysfs.o obj-$(CONFIG_ARCH_HAS_CC_PLATFORM) += cc_platform.o + +# nothing that operates in real mode is safe for KASAN +KASAN_SANITIZE_ras.o := n +KASAN_SANITIZE_kexec.o := n diff --git a/arch/powerpc/platforms/pseries/cmm.c b/arch/powerpc/platforms/pseries/cmm.c index 45a3a3022a85..5f4037c1d7fe 100644 --- a/arch/powerpc/platforms/pseries/cmm.c +++ b/arch/powerpc/platforms/pseries/cmm.c @@ -19,9 +19,6 @@ #include <linux/stringify.h> #include <linux/swap.h> #include <linux/device.h> -#include <linux/mount.h> -#include <linux/pseudo_fs.h> -#include <linux/magic.h> #include <linux/balloon_compaction.h> #include <asm/firmware.h> #include <asm/hvcall.h> @@ -475,8 +472,6 @@ static struct notifier_block cmm_reboot_nb = { static int cmm_memory_cb(struct notifier_block *self, unsigned long action, void *arg) { - int ret = 0; - switch (action) { case MEM_GOING_OFFLINE: mutex_lock(&hotplug_mutex); @@ -493,7 +488,7 @@ static int cmm_memory_cb(struct notifier_block *self, break; } - return notifier_from_errno(ret); + return NOTIFY_OK; } static struct notifier_block cmm_mem_nb = { @@ -502,19 +497,6 @@ static struct notifier_block cmm_mem_nb = { }; #ifdef CONFIG_BALLOON_COMPACTION -static struct vfsmount *balloon_mnt; - -static int cmm_init_fs_context(struct fs_context *fc) -{ - return init_pseudo(fc, PPC_CMM_MAGIC) ? 0 : -ENOMEM; -} - -static struct file_system_type balloon_fs = { - .name = "ppc-cmm", - .init_fs_context = cmm_init_fs_context, - .kill_sb = kill_anon_super, -}; - static int cmm_migratepage(struct balloon_dev_info *b_dev_info, struct page *newpage, struct page *page, enum migrate_mode mode) @@ -566,47 +548,13 @@ static int cmm_migratepage(struct balloon_dev_info *b_dev_info, return MIGRATEPAGE_SUCCESS; } -static int cmm_balloon_compaction_init(void) +static void cmm_balloon_compaction_init(void) { - int rc; - balloon_devinfo_init(&b_dev_info); b_dev_info.migratepage = cmm_migratepage; - - balloon_mnt = kern_mount(&balloon_fs); - if (IS_ERR(balloon_mnt)) { - rc = PTR_ERR(balloon_mnt); - balloon_mnt = NULL; - return rc; - } - - b_dev_info.inode = alloc_anon_inode(balloon_mnt->mnt_sb); - if (IS_ERR(b_dev_info.inode)) { - rc = PTR_ERR(b_dev_info.inode); - b_dev_info.inode = NULL; - kern_unmount(balloon_mnt); - balloon_mnt = NULL; - return rc; - } - - b_dev_info.inode->i_mapping->a_ops = &balloon_aops; - return 0; -} -static void cmm_balloon_compaction_deinit(void) -{ - if (b_dev_info.inode) - iput(b_dev_info.inode); - b_dev_info.inode = NULL; - kern_unmount(balloon_mnt); - balloon_mnt = NULL; } #else /* CONFIG_BALLOON_COMPACTION */ -static int cmm_balloon_compaction_init(void) -{ - return 0; -} - -static void cmm_balloon_compaction_deinit(void) +static void cmm_balloon_compaction_init(void) { } #endif /* CONFIG_BALLOON_COMPACTION */ @@ -624,9 +572,7 @@ static int cmm_init(void) if (!firmware_has_feature(FW_FEATURE_CMO) && !simulate) return -EOPNOTSUPP; - rc = cmm_balloon_compaction_init(); - if (rc) - return rc; + cmm_balloon_compaction_init(); rc = register_oom_notifier(&cmm_oom_nb); if (rc < 0) @@ -660,7 +606,6 @@ out_reboot_notifier: out_oom_notifier: unregister_oom_notifier(&cmm_oom_nb); out_balloon_compaction: - cmm_balloon_compaction_deinit(); return rc; } @@ -679,7 +624,6 @@ static void cmm_exit(void) unregister_memory_notifier(&cmm_mem_nb); cmm_free_pages(atomic_long_read(&loaned_pages)); cmm_unregister_sysfs(&cmm_dev); - cmm_balloon_compaction_deinit(); } /** diff --git a/arch/powerpc/platforms/pseries/dlpar.c b/arch/powerpc/platforms/pseries/dlpar.c index b1f01ac0c29e..498d6efcb5ae 100644 --- a/arch/powerpc/platforms/pseries/dlpar.c +++ b/arch/powerpc/platforms/pseries/dlpar.c @@ -19,7 +19,6 @@ #include "of_helpers.h" #include "pseries.h" -#include <asm/prom.h> #include <asm/machdep.h> #include <linux/uaccess.h> #include <asm/rtas.h> @@ -389,7 +388,7 @@ static void pseries_hp_work_fn(struct work_struct *work) handle_dlpar_errorlog(hp_work->errlog); kfree(hp_work->errlog); - kfree((void *)work); + kfree(work); } void queue_hotplug_event(struct pseries_hp_errorlog *hp_errlog) diff --git a/arch/powerpc/platforms/pseries/dtl.c b/arch/powerpc/platforms/pseries/dtl.c index 352af5b14a0f..3f1cdccebc9c 100644 --- a/arch/powerpc/platforms/pseries/dtl.c +++ b/arch/powerpc/platforms/pseries/dtl.c @@ -18,6 +18,7 @@ #include <asm/plpar_wrappers.h> #include <asm/machdep.h> +#ifdef CONFIG_DTL struct dtl { struct dtl_entry *buf; int cpu; @@ -37,6 +38,15 @@ static u8 dtl_event_mask = DTL_LOG_ALL; static int dtl_buf_entries = N_DISPATCH_LOG; #ifdef CONFIG_VIRT_CPU_ACCOUNTING_NATIVE + +/* + * When CONFIG_VIRT_CPU_ACCOUNTING_NATIVE = y, the cpu accounting code controls + * reading from the dispatch trace log. If other code wants to consume + * DTL entries, it can set this pointer to a function that will get + * called once for each DTL entry that gets processed. + */ +static void (*dtl_consumer)(struct dtl_entry *entry, u64 index); + struct dtl_ring { u64 write_index; struct dtl_entry *write_ptr; @@ -355,3 +365,81 @@ static int dtl_init(void) return 0; } machine_arch_initcall(pseries, dtl_init); +#endif /* CONFIG_DTL */ + +#ifdef CONFIG_VIRT_CPU_ACCOUNTING_NATIVE +/* + * Scan the dispatch trace log and count up the stolen time. + * Should be called with interrupts disabled. + */ +static notrace u64 scan_dispatch_log(u64 stop_tb) +{ + u64 i = local_paca->dtl_ridx; + struct dtl_entry *dtl = local_paca->dtl_curr; + struct dtl_entry *dtl_end = local_paca->dispatch_log_end; + struct lppaca *vpa = local_paca->lppaca_ptr; + u64 tb_delta; + u64 stolen = 0; + u64 dtb; + + if (!dtl) + return 0; + + if (i == be64_to_cpu(vpa->dtl_idx)) + return 0; + while (i < be64_to_cpu(vpa->dtl_idx)) { + dtb = be64_to_cpu(dtl->timebase); + tb_delta = be32_to_cpu(dtl->enqueue_to_dispatch_time) + + be32_to_cpu(dtl->ready_to_enqueue_time); + barrier(); + if (i + N_DISPATCH_LOG < be64_to_cpu(vpa->dtl_idx)) { + /* buffer has overflowed */ + i = be64_to_cpu(vpa->dtl_idx) - N_DISPATCH_LOG; + dtl = local_paca->dispatch_log + (i % N_DISPATCH_LOG); + continue; + } + if (dtb > stop_tb) + break; +#ifdef CONFIG_DTL + if (dtl_consumer) + dtl_consumer(dtl, i); +#endif + stolen += tb_delta; + ++i; + ++dtl; + if (dtl == dtl_end) + dtl = local_paca->dispatch_log; + } + local_paca->dtl_ridx = i; + local_paca->dtl_curr = dtl; + return stolen; +} + +/* + * Accumulate stolen time by scanning the dispatch trace log. + * Called on entry from user mode. + */ +void notrace pseries_accumulate_stolen_time(void) +{ + u64 sst, ust; + struct cpu_accounting_data *acct = &local_paca->accounting; + + sst = scan_dispatch_log(acct->starttime_user); + ust = scan_dispatch_log(acct->starttime); + acct->stime -= sst; + acct->utime -= ust; + acct->steal_time += ust + sst; +} + +u64 pseries_calculate_stolen_time(u64 stop_tb) +{ + if (!firmware_has_feature(FW_FEATURE_SPLPAR)) + return 0; + + if (get_paca()->dtl_ridx != be64_to_cpu(get_lppaca()->dtl_idx)) + return scan_dispatch_log(stop_tb); + + return 0; +} + +#endif diff --git a/arch/powerpc/platforms/pseries/eeh_pseries.c b/arch/powerpc/platforms/pseries/eeh_pseries.c index 09fafcf2d3a0..8e40ccac0f44 100644 --- a/arch/powerpc/platforms/pseries/eeh_pseries.c +++ b/arch/powerpc/platforms/pseries/eeh_pseries.c @@ -43,6 +43,8 @@ static int ibm_get_config_addr_info; static int ibm_get_config_addr_info2; static int ibm_configure_pe; +static void pseries_eeh_init_edev(struct pci_dn *pdn); + static void pseries_pcibios_bus_add_device(struct pci_dev *pdev) { struct pci_dn *pdn = pci_get_pdn(pdev); @@ -69,7 +71,7 @@ static void pseries_pcibios_bus_add_device(struct pci_dev *pdev) if (pdev->is_virtfn) { /* * FIXME: This really should be handled by choosing the right - * parent PE in in pseries_eeh_init_edev(). + * parent PE in pseries_eeh_init_edev(). */ struct eeh_pe *physfn_pe = pci_dev_to_eeh_dev(pdev->physfn)->pe; struct eeh_dev *edev = pdn_to_eeh_dev(pdn); @@ -359,7 +361,7 @@ static struct eeh_pe *pseries_eeh_pe_get_parent(struct eeh_dev *edev) * This function takes care of the initialisation and inserts the eeh_dev * into the correct eeh_pe. If no eeh_pe exists we'll allocate one. */ -void pseries_eeh_init_edev(struct pci_dn *pdn) +static void pseries_eeh_init_edev(struct pci_dn *pdn) { struct eeh_pe pe, *parent; struct eeh_dev *edev; @@ -510,7 +512,7 @@ static int pseries_eeh_set_option(struct eeh_pe *pe, int option) int ret = 0; /* - * When we're enabling or disabling EEH functioality on + * When we're enabling or disabling EEH functionality on * the particular PE, the PE config address is possibly * unavailable. Therefore, we have to figure it out from * the FDT node. @@ -845,8 +847,7 @@ static int __init eeh_pseries_init(void) return -EINVAL; } - /* Initialize error log lock and size */ - spin_lock_init(&slot_errbuf_lock); + /* Initialize error log size */ eeh_error_buf_size = rtas_token("rtas-error-log-max"); if (eeh_error_buf_size == RTAS_UNKNOWN_SERVICE) { pr_info("%s: unknown EEH error log size\n", diff --git a/arch/powerpc/platforms/pseries/event_sources.c b/arch/powerpc/platforms/pseries/event_sources.c index be661e919c76..623dfe0d8e1c 100644 --- a/arch/powerpc/platforms/pseries/event_sources.c +++ b/arch/powerpc/platforms/pseries/event_sources.c @@ -8,7 +8,7 @@ #include "pseries.h" -void request_event_sources_irqs(struct device_node *np, +void __init request_event_sources_irqs(struct device_node *np, irq_handler_t handler, const char *name) { diff --git a/arch/powerpc/platforms/pseries/firmware.c b/arch/powerpc/platforms/pseries/firmware.c index f162156b7b68..080108d129ed 100644 --- a/arch/powerpc/platforms/pseries/firmware.c +++ b/arch/powerpc/platforms/pseries/firmware.c @@ -66,6 +66,8 @@ hypertas_fw_features_table[] = { {FW_FEATURE_BLOCK_REMOVE, "hcall-block-remove"}, {FW_FEATURE_PAPR_SCM, "hcall-scm"}, {FW_FEATURE_RPT_INVALIDATE, "hcall-rpt-invalidate"}, + {FW_FEATURE_ENERGY_SCALE_INFO, "hcall-energy-scale-info"}, + {FW_FEATURE_WATCHDOG, "hcall-watchdog"}, }; /* Build up the firmware features bitmask using the contents of diff --git a/arch/powerpc/platforms/pseries/hotplug-cpu.c b/arch/powerpc/platforms/pseries/hotplug-cpu.c index 5ab44600c8d3..e0a7ac5db15d 100644 --- a/arch/powerpc/platforms/pseries/hotplug-cpu.c +++ b/arch/powerpc/platforms/pseries/hotplug-cpu.c @@ -398,7 +398,7 @@ static int dlpar_online_cpu(struct device_node *dn) if (get_hard_smp_processor_id(cpu) != thread) continue; cpu_maps_update_done(); - find_and_online_cpu_nid(cpu); + find_and_update_cpu_nid(cpu); rc = device_online(get_cpu_device(cpu)); if (rc) { dlpar_offline_cpu(dn); @@ -619,17 +619,21 @@ static ssize_t dlpar_cpu_add(u32 drc_index) static unsigned int pseries_cpuhp_cache_use_count(const struct device_node *cachedn) { unsigned int use_count = 0; - struct device_node *dn; + struct device_node *dn, *tn; WARN_ON(!of_node_is_type(cachedn, "cache")); for_each_of_cpu_node(dn) { - if (of_find_next_cache_node(dn) == cachedn) + tn = of_find_next_cache_node(dn); + of_node_put(tn); + if (tn == cachedn) use_count++; } for_each_node_by_type(dn, "cache") { - if (of_find_next_cache_node(dn) == cachedn) + tn = of_find_next_cache_node(dn); + of_node_put(tn); + if (tn == cachedn) use_count++; } @@ -649,10 +653,13 @@ static int pseries_cpuhp_detach_nodes(struct device_node *cpudn) dn = cpudn; while ((dn = of_find_next_cache_node(dn))) { - if (pseries_cpuhp_cache_use_count(dn) > 1) + if (pseries_cpuhp_cache_use_count(dn) > 1) { + of_node_put(dn); break; + } ret = of_changeset_detach_node(&cs, dn); + of_node_put(dn); if (ret) goto out; } @@ -864,12 +871,13 @@ static int __init pseries_cpu_hotplug_init(void) /* Processors can be added/removed only on LPAR */ if (firmware_has_feature(FW_FEATURE_LPAR)) { for_each_node(node) { - alloc_bootmem_cpumask_var(&node_recorded_ids_map[node]); + if (!alloc_cpumask_var_node(&node_recorded_ids_map[node], + GFP_KERNEL, node)) + return -ENOMEM; /* Record ids of CPU added at boot time */ - cpumask_or(node_recorded_ids_map[node], - node_recorded_ids_map[node], - cpumask_of_node(node)); + cpumask_copy(node_recorded_ids_map[node], + cpumask_of_node(node)); } of_reconfig_notifier_register(&pseries_smp_nb); diff --git a/arch/powerpc/platforms/pseries/hotplug-memory.c b/arch/powerpc/platforms/pseries/hotplug-memory.c index 91cf23495ccb..2e3a317722a8 100644 --- a/arch/powerpc/platforms/pseries/hotplug-memory.c +++ b/arch/powerpc/platforms/pseries/hotplug-memory.c @@ -16,7 +16,6 @@ #include <asm/firmware.h> #include <asm/machdep.h> -#include <asm/prom.h> #include <asm/sparsemem.h> #include <asm/fadump.h> #include <asm/drmem.h> diff --git a/arch/powerpc/platforms/pseries/hvCall.S b/arch/powerpc/platforms/pseries/hvCall.S index ab9fc6506861..762eb15d3bd4 100644 --- a/arch/powerpc/platforms/pseries/hvCall.S +++ b/arch/powerpc/platforms/pseries/hvCall.S @@ -16,7 +16,7 @@ #ifdef CONFIG_TRACEPOINTS #ifndef CONFIG_JUMP_LABEL - .section ".toc","aw" + .data .globl hcall_tracepoint_refcount hcall_tracepoint_refcount: @@ -88,7 +88,7 @@ hcall_tracepoint_refcount: BEGIN_FTR_SECTION; \ b 1f; \ END_FTR_SECTION(0, 1); \ - ld r12,hcall_tracepoint_refcount@toc(r2); \ + LOAD_REG_ADDR(r12, hcall_tracepoint_refcount) ; \ std r12,32(r1); \ cmpdi r12,0; \ bne- LABEL; \ diff --git a/arch/powerpc/platforms/pseries/hvcserver.c b/arch/powerpc/platforms/pseries/hvcserver.c index 96e18d3b2fcf..d48c9c7ce10f 100644 --- a/arch/powerpc/platforms/pseries/hvcserver.c +++ b/arch/powerpc/platforms/pseries/hvcserver.c @@ -176,7 +176,7 @@ int hvcs_get_partner_info(uint32_t unit_address, struct list_head *head, = (unsigned int)last_p_partition_ID; /* copy the Null-term char too */ - strlcpy(&next_partner_info->location_code[0], + strscpy(&next_partner_info->location_code[0], (char *)&pi_buff[2], sizeof(next_partner_info->location_code)); diff --git a/arch/powerpc/platforms/pseries/ibmebus.c b/arch/powerpc/platforms/pseries/ibmebus.c index 7ee3ed7d6cc2..a870cada7acd 100644 --- a/arch/powerpc/platforms/pseries/ibmebus.c +++ b/arch/powerpc/platforms/pseries/ibmebus.c @@ -152,7 +152,11 @@ static const struct dma_map_ops ibmebus_dma_ops = { static int ibmebus_match_path(struct device *dev, const void *data) { struct device_node *dn = to_platform_device(dev)->dev.of_node; - return (of_find_node_by_path(data) == dn); + struct device_node *tn = of_find_node_by_path(data); + + of_node_put(tn); + + return (tn == dn); } static int ibmebus_match_node(struct device *dev, const void *data) diff --git a/arch/powerpc/platforms/pseries/iommu.c b/arch/powerpc/platforms/pseries/iommu.c index 8f998e55735b..561adac69022 100644 --- a/arch/powerpc/platforms/pseries/iommu.c +++ b/arch/powerpc/platforms/pseries/iommu.c @@ -666,8 +666,7 @@ static void pci_dma_bus_setup_pSeries(struct pci_bus *bus) #ifdef CONFIG_IOMMU_API static int tce_exchange_pseries(struct iommu_table *tbl, long index, unsigned - long *tce, enum dma_data_direction *direction, - bool realmode) + long *tce, enum dma_data_direction *direction) { long rc; unsigned long ioba = (unsigned long) index << tbl->it_page_shift; @@ -701,6 +700,33 @@ struct iommu_table_ops iommu_table_lpar_multi_ops = { .get = tce_get_pSeriesLP }; +/* + * Find nearest ibm,dma-window (default DMA window) or direct DMA window or + * dynamic 64bit DMA window, walking up the device tree. + */ +static struct device_node *pci_dma_find(struct device_node *dn, + const __be32 **dma_window) +{ + const __be32 *dw = NULL; + + for ( ; dn && PCI_DN(dn); dn = dn->parent) { + dw = of_get_property(dn, "ibm,dma-window", NULL); + if (dw) { + if (dma_window) + *dma_window = dw; + return dn; + } + dw = of_get_property(dn, DIRECT64_PROPNAME, NULL); + if (dw) + return dn; + dw = of_get_property(dn, DMA64_PROPNAME, NULL); + if (dw) + return dn; + } + + return NULL; +} + static void pci_dma_bus_setup_pSeriesLP(struct pci_bus *bus) { struct iommu_table *tbl; @@ -713,20 +739,10 @@ static void pci_dma_bus_setup_pSeriesLP(struct pci_bus *bus) pr_debug("pci_dma_bus_setup_pSeriesLP: setting up bus %pOF\n", dn); - /* - * Find nearest ibm,dma-window (default DMA window), walking up the - * device tree - */ - for (pdn = dn; pdn != NULL; pdn = pdn->parent) { - dma_window = of_get_property(pdn, "ibm,dma-window", NULL); - if (dma_window != NULL) - break; - } + pdn = pci_dma_find(dn, &dma_window); - if (dma_window == NULL) { + if (dma_window == NULL) pr_debug(" no ibm,dma-window property !\n"); - return; - } ppci = PCI_DN(pdn); @@ -736,11 +752,13 @@ static void pci_dma_bus_setup_pSeriesLP(struct pci_bus *bus) if (!ppci->table_group) { ppci->table_group = iommu_pseries_alloc_group(ppci->phb->node); tbl = ppci->table_group->tables[0]; - iommu_table_setparms_lpar(ppci->phb, pdn, tbl, - ppci->table_group, dma_window); + if (dma_window) { + iommu_table_setparms_lpar(ppci->phb, pdn, tbl, + ppci->table_group, dma_window); - if (!iommu_init_table(tbl, ppci->phb->node, 0, 0)) - panic("Failed to initialize iommu table"); + if (!iommu_init_table(tbl, ppci->phb->node, 0, 0)) + panic("Failed to initialize iommu table"); + } iommu_register_group(ppci->table_group, pci_domain_nr(bus), 0); pr_debug(" created table: %p\n", ppci->table_group); @@ -1022,9 +1040,6 @@ static int query_ddw(struct pci_dev *dev, const u32 *ddw_avail, ret = rtas_call(ddw_avail[DDW_QUERY_PE_DMA_WIN], 3, out_sz, query_out, cfg_addr, BUID_HI(buid), BUID_LO(buid)); - dev_info(&dev->dev, "ibm,query-pe-dma-windows(%x) %x %x %x returned %d\n", - ddw_avail[DDW_QUERY_PE_DMA_WIN], cfg_addr, BUID_HI(buid), - BUID_LO(buid), ret); switch (out_sz) { case 5: @@ -1042,6 +1057,11 @@ static int query_ddw(struct pci_dev *dev, const u32 *ddw_avail, break; } + dev_info(&dev->dev, "ibm,query-pe-dma-windows(%x) %x %x %x returned %d, lb=%llx ps=%x wn=%d\n", + ddw_avail[DDW_QUERY_PE_DMA_WIN], cfg_addr, BUID_HI(buid), + BUID_LO(buid), ret, query->largest_available_block, + query->page_size, query->windows_available); + return ret; } @@ -1233,7 +1253,7 @@ static bool enable_ddw(struct pci_dev *dev, struct device_node *pdn) bool default_win_removed = false, direct_mapping = false; bool pmem_present; struct pci_dn *pci = PCI_DN(pdn); - struct iommu_table *tbl = pci->table_group->tables[0]; + struct property *default_win = NULL; dn = of_find_node_by_type(NULL, "ibm,pmemory"); pmem_present = dn != NULL; @@ -1290,11 +1310,10 @@ static bool enable_ddw(struct pci_dev *dev, struct device_node *pdn) * for extensions presence. */ if (query.windows_available == 0) { - struct property *default_win; int reset_win_ext; /* DDW + IOMMU on single window may fail if there is any allocation */ - if (iommu_table_in_use(tbl)) { + if (iommu_table_in_use(pci->table_group->tables[0])) { dev_warn(&dev->dev, "current IOMMU table in use, can't be replaced.\n"); goto out_failed; } @@ -1430,16 +1449,18 @@ static bool enable_ddw(struct pci_dev *dev, struct device_node *pdn) pci->table_group->tables[1] = newtbl; - /* Keep default DMA window stuct if removed */ - if (default_win_removed) { - tbl->it_size = 0; - vfree(tbl->it_map); - tbl->it_map = NULL; - } - set_iommu_table_base(&dev->dev, newtbl); } + if (default_win_removed) { + iommu_tce_table_put(pci->table_group->tables[0]); + pci->table_group->tables[0] = NULL; + + /* default_win is valid here because default_win_removed == true */ + of_remove_property(pdn, default_win); + dev_info(&dev->dev, "Removed default DMA window for %pOF\n", pdn); + } + spin_lock(&dma_win_list_lock); list_add(&window->list, &dma_win_list); spin_unlock(&dma_win_list_lock); @@ -1504,13 +1525,7 @@ static void pci_dma_dev_setup_pSeriesLP(struct pci_dev *dev) dn = pci_device_to_OF_node(dev); pr_debug(" node is %pOF\n", dn); - for (pdn = dn; pdn && PCI_DN(pdn) && !PCI_DN(pdn)->table_group; - pdn = pdn->parent) { - dma_window = of_get_property(pdn, "ibm,dma-window", NULL); - if (dma_window) - break; - } - + pdn = pci_dma_find(dn, &dma_window); if (!pdn || !PCI_DN(pdn)) { printk(KERN_WARNING "pci_dma_dev_setup_pSeriesLP: " "no DMA window found for pci dev=%s dn=%pOF\n", @@ -1541,7 +1556,6 @@ static void pci_dma_dev_setup_pSeriesLP(struct pci_dev *dev) static bool iommu_bypass_supported_pSeriesLP(struct pci_dev *pdev, u64 dma_mask) { struct device_node *dn = pci_device_to_OF_node(pdev), *pdn; - const __be32 *dma_window = NULL; /* only attempt to use a new window if 64-bit DMA is requested */ if (dma_mask < DMA_BIT_MASK(64)) @@ -1555,13 +1569,7 @@ static bool iommu_bypass_supported_pSeriesLP(struct pci_dev *pdev, u64 dma_mask) * search upwards in the tree until we either hit a dma-window * property, OR find a parent with a table already allocated. */ - for (pdn = dn; pdn && PCI_DN(pdn) && !PCI_DN(pdn)->table_group; - pdn = pdn->parent) { - dma_window = of_get_property(pdn, "ibm,dma-window", NULL); - if (dma_window) - break; - } - + pdn = pci_dma_find(dn, NULL); if (pdn && PCI_DN(pdn)) return enable_ddw(pdev, pdn); @@ -1654,7 +1662,7 @@ static struct notifier_block iommu_reconfig_nb = { }; /* These are called very early. */ -void iommu_init_early_pSeries(void) +void __init iommu_init_early_pSeries(void) { if (of_chosen && of_get_property(of_chosen, "linux,iommu-off", NULL)) return; diff --git a/arch/powerpc/platforms/pseries/kexec.c b/arch/powerpc/platforms/pseries/kexec.c index 145fcfbc017f..096d09ed89f6 100644 --- a/arch/powerpc/platforms/pseries/kexec.c +++ b/arch/powerpc/platforms/pseries/kexec.c @@ -6,7 +6,7 @@ #include <linux/kernel.h> #include <linux/interrupt.h> -#include <asm/machdep.h> +#include <asm/setup.h> #include <asm/page.h> #include <asm/firmware.h> #include <asm/kexec.h> @@ -61,3 +61,11 @@ void pseries_kexec_cpu_down(int crash_shutdown, int secondary) } else xics_kexec_teardown_cpu(secondary); } + +void pseries_machine_kexec(struct kimage *image) +{ + if (firmware_has_feature(FW_FEATURE_SET_MODE)) + pseries_disable_reloc_on_exc(); + + default_machine_kexec(image); +} diff --git a/arch/powerpc/platforms/pseries/lpar.c b/arch/powerpc/platforms/pseries/lpar.c index 3df6bdfea475..97ef6499e501 100644 --- a/arch/powerpc/platforms/pseries/lpar.c +++ b/arch/powerpc/platforms/pseries/lpar.c @@ -27,11 +27,10 @@ #include <asm/processor.h> #include <asm/mmu.h> #include <asm/page.h> -#include <asm/machdep.h> +#include <asm/setup.h> #include <asm/mmu_context.h> #include <asm/iommu.h> #include <asm/tlb.h> -#include <asm/prom.h> #include <asm/cputable.h> #include <asm/udbg.h> #include <asm/smp.h> @@ -40,7 +39,6 @@ #include <asm/plpar_wrappers.h> #include <asm/kexec.h> #include <asm/fadump.h> -#include <asm/asm-prototypes.h> #include <asm/dtl.h> #include "pseries.h" @@ -58,6 +56,7 @@ EXPORT_SYMBOL(plpar_hcall); EXPORT_SYMBOL(plpar_hcall9); EXPORT_SYMBOL(plpar_hcall_norets); +#ifdef CONFIG_PPC_64S_HASH_MMU /* * H_BLOCK_REMOVE supported block size for this page size in segment who's base * page size is that page size. @@ -66,6 +65,7 @@ EXPORT_SYMBOL(plpar_hcall_norets); * page size. */ static int hblkrm_size[MMU_PAGE_COUNT][MMU_PAGE_COUNT] __ro_after_init; +#endif /* * Due to the involved complexity, and that the current hypervisor is only @@ -660,6 +660,17 @@ static int __init vcpudispatch_stats_procfs_init(void) } machine_device_initcall(pseries, vcpudispatch_stats_procfs_init); + +#ifdef CONFIG_PARAVIRT_TIME_ACCOUNTING +u64 pseries_paravirt_steal_clock(int cpu) +{ + struct lppaca *lppaca = &lppaca_of(cpu); + + return be64_to_cpu(READ_ONCE(lppaca->enqueue_dispatch_tb)) + + be64_to_cpu(READ_ONCE(lppaca->ready_enqueue_tb)); +} +#endif + #endif /* CONFIG_PPC_SPLPAR */ void vpa_init(int cpu) @@ -689,7 +700,7 @@ void vpa_init(int cpu) return; } -#ifdef CONFIG_PPC_BOOK3S_64 +#ifdef CONFIG_PPC_64S_HASH_MMU /* * PAPR says this feature is SLB-Buffer but firmware never * reports that. All SPLPAR support SLB shadow buffer. @@ -702,7 +713,7 @@ void vpa_init(int cpu) "cpu %d (hw %d) of area %lx failed with %ld\n", cpu, hwcpu, addr, ret); } -#endif /* CONFIG_PPC_BOOK3S_64 */ +#endif /* CONFIG_PPC_64S_HASH_MMU */ /* * Register dispatch trace log, if one has been allocated. @@ -712,6 +723,36 @@ void vpa_init(int cpu) #ifdef CONFIG_PPC_BOOK3S_64 +static int __init pseries_lpar_register_process_table(unsigned long base, + unsigned long page_size, unsigned long table_size) +{ + long rc; + unsigned long flags = 0; + + if (table_size) + flags |= PROC_TABLE_NEW; + if (radix_enabled()) { + flags |= PROC_TABLE_RADIX; + if (mmu_has_feature(MMU_FTR_GTSE)) + flags |= PROC_TABLE_GTSE; + } else + flags |= PROC_TABLE_HPT_SLB; + for (;;) { + rc = plpar_hcall_norets(H_REGISTER_PROC_TBL, flags, base, + page_size, table_size); + if (!H_IS_LONG_BUSY(rc)) + break; + mdelay(get_longbusy_msecs(rc)); + } + if (rc != H_SUCCESS) { + pr_err("Failed to register process table (rc=%ld)\n", rc); + BUG(); + } + return rc; +} + +#ifdef CONFIG_PPC_64S_HASH_MMU + static long pSeries_lpar_hpte_insert(unsigned long hpte_group, unsigned long vpn, unsigned long pa, unsigned long rflags, unsigned long vflags, @@ -1680,34 +1721,6 @@ static int pseries_lpar_resize_hpt(unsigned long shift) return 0; } -static int pseries_lpar_register_process_table(unsigned long base, - unsigned long page_size, unsigned long table_size) -{ - long rc; - unsigned long flags = 0; - - if (table_size) - flags |= PROC_TABLE_NEW; - if (radix_enabled()) { - flags |= PROC_TABLE_RADIX; - if (mmu_has_feature(MMU_FTR_GTSE)) - flags |= PROC_TABLE_GTSE; - } else - flags |= PROC_TABLE_HPT_SLB; - for (;;) { - rc = plpar_hcall_norets(H_REGISTER_PROC_TBL, flags, base, - page_size, table_size); - if (!H_IS_LONG_BUSY(rc)) - break; - mdelay(get_longbusy_msecs(rc)); - } - if (rc != H_SUCCESS) { - pr_err("Failed to register process table (rc=%ld)\n", rc); - BUG(); - } - return rc; -} - void __init hpte_init_pseries(void) { mmu_hash_ops.hpte_invalidate = pSeries_lpar_hpte_invalidate; @@ -1730,9 +1743,10 @@ void __init hpte_init_pseries(void) if (cpu_has_feature(CPU_FTR_ARCH_300)) pseries_lpar_register_process_table(0, 0, 0); } +#endif /* CONFIG_PPC_64S_HASH_MMU */ #ifdef CONFIG_PPC_RADIX_MMU -void radix_init_pseries(void) +void __init radix_init_pseries(void) { pr_info("Using radix MMU under hypervisor\n"); @@ -1932,7 +1946,8 @@ int h_get_mpp_x(struct hvcall_mpp_x_data *mpp_x_data) return rc; } -static unsigned long vsid_unscramble(unsigned long vsid, int ssize) +#ifdef CONFIG_PPC_64S_HASH_MMU +static unsigned long __init vsid_unscramble(unsigned long vsid, int ssize) { unsigned long protovsid; unsigned long va_bits = VA_BITS; @@ -1992,6 +2007,7 @@ static int __init reserve_vrma_context_id(void) return 0; } machine_device_initcall(pseries, reserve_vrma_context_id); +#endif #ifdef CONFIG_DEBUG_FS /* debugfs file interface for vpa data */ diff --git a/arch/powerpc/platforms/pseries/lparcfg.c b/arch/powerpc/platforms/pseries/lparcfg.c index f71eac74ea92..63fd925ccbb8 100644 --- a/arch/powerpc/platforms/pseries/lparcfg.c +++ b/arch/powerpc/platforms/pseries/lparcfg.c @@ -28,7 +28,6 @@ #include <asm/firmware.h> #include <asm/rtas.h> #include <asm/time.h> -#include <asm/prom.h> #include <asm/vdso_datapage.h> #include <asm/vio.h> #include <asm/mmu.h> @@ -36,6 +35,7 @@ #include <asm/drmem.h> #include "pseries.h" +#include "vas.h" /* pseries_vas_dlpar_cpu() */ /* * This isn't a module but we expose that to userspace @@ -311,6 +311,92 @@ static void parse_mpp_x_data(struct seq_file *m) seq_printf(m, "coalesce_pool_spurr=%ld\n", mpp_x_data.pool_spurr_cycles); } +/* + * PAPR defines, in section "7.3.16 System Parameters Option", the token 55 to + * read the LPAR name, and the largest output data to 4000 + 2 bytes length. + */ +#define SPLPAR_LPAR_NAME_TOKEN 55 +#define GET_SYS_PARM_BUF_SIZE 4002 +#if GET_SYS_PARM_BUF_SIZE > RTAS_DATA_BUF_SIZE +#error "GET_SYS_PARM_BUF_SIZE is larger than RTAS_DATA_BUF_SIZE" +#endif + +/* + * Read the lpar name using the RTAS ibm,get-system-parameter call. + * + * The name read through this call is updated if changes are made by the end + * user on the hypervisor side. + * + * Some hypervisor (like Qemu) may not provide this value. In that case, a non + * null value is returned. + */ +static int read_rtas_lpar_name(struct seq_file *m) +{ + int rc, len, token; + union { + char raw_buffer[GET_SYS_PARM_BUF_SIZE]; + struct { + __be16 len; + char name[GET_SYS_PARM_BUF_SIZE-2]; + }; + } *local_buffer; + + token = rtas_token("ibm,get-system-parameter"); + if (token == RTAS_UNKNOWN_SERVICE) + return -EINVAL; + + local_buffer = kmalloc(sizeof(*local_buffer), GFP_KERNEL); + if (!local_buffer) + return -ENOMEM; + + do { + spin_lock(&rtas_data_buf_lock); + memset(rtas_data_buf, 0, sizeof(*local_buffer)); + rc = rtas_call(token, 3, 1, NULL, SPLPAR_LPAR_NAME_TOKEN, + __pa(rtas_data_buf), sizeof(*local_buffer)); + if (!rc) + memcpy(local_buffer->raw_buffer, rtas_data_buf, + sizeof(local_buffer->raw_buffer)); + spin_unlock(&rtas_data_buf_lock); + } while (rtas_busy_delay(rc)); + + if (!rc) { + /* Force end of string */ + len = min((int) be16_to_cpu(local_buffer->len), + (int) sizeof(local_buffer->name)-1); + local_buffer->name[len] = '\0'; + + seq_printf(m, "partition_name=%s\n", local_buffer->name); + } else + rc = -ENODATA; + + kfree(local_buffer); + return rc; +} + +/* + * Read the LPAR name from the Device Tree. + * + * The value read in the DT is not updated if the end-user is touching the LPAR + * name on the hypervisor side. + */ +static int read_dt_lpar_name(struct seq_file *m) +{ + const char *name; + + if (of_property_read_string(of_root, "ibm,partition-name", &name)) + return -ENOENT; + + seq_printf(m, "partition_name=%s\n", name); + return 0; +} + +static void read_lpar_name(struct seq_file *m) +{ + if (read_rtas_lpar_name(m) && read_dt_lpar_name(m)) + pr_err_once("Error can't get the LPAR name"); +} + #define SPLPAR_CHARACTERISTICS_TOKEN 20 #define SPLPAR_MAXLENGTH 1026*(sizeof(char)) @@ -496,6 +582,7 @@ static int pseries_lparcfg_data(struct seq_file *m, void *v) if (firmware_has_feature(FW_FEATURE_SPLPAR)) { /* this call handles the ibm,get-system-parameter contents */ + read_lpar_name(m); parse_system_parameter_string(m); parse_ppp_data(m); parse_mpp_data(m); @@ -531,8 +618,9 @@ static int pseries_lparcfg_data(struct seq_file *m, void *v) seq_printf(m, "shared_processor_mode=%d\n", lppaca_shared_proc(get_lppaca())); -#ifdef CONFIG_PPC_BOOK3S_64 - seq_printf(m, "slb_size=%d\n", mmu_slb_size); +#ifdef CONFIG_PPC_64S_HASH_MMU + if (!radix_enabled()) + seq_printf(m, "slb_size=%d\n", mmu_slb_size); #endif parse_em_data(m); maxmem_data(m); @@ -661,6 +749,16 @@ static ssize_t lparcfg_write(struct file *file, const char __user * buf, return -EINVAL; retval = update_ppp(new_entitled_ptr, NULL); + + if (retval == H_SUCCESS || retval == H_CONSTRAINED) { + /* + * The hypervisor assigns VAS resources based + * on entitled capacity for shared mode. + * Reconfig VAS windows based on DLPAR CPU events. + */ + if (pseries_vas_dlpar_cpu() != 0) + retval = H_HARDWARE; + } } else if (!strcmp(kbuf, "capacity_weight")) { char *endp; *new_weight_ptr = (u8) simple_strtoul(tmp, &endp, 10); diff --git a/arch/powerpc/platforms/pseries/mobility.c b/arch/powerpc/platforms/pseries/mobility.c index 210a37a065fb..634fac5db3f9 100644 --- a/arch/powerpc/platforms/pseries/mobility.c +++ b/arch/powerpc/platforms/pseries/mobility.c @@ -26,6 +26,7 @@ #include <asm/machdep.h> #include <asm/rtas.h> #include "pseries.h" +#include "vas.h" /* vas_migration_handler() */ #include "../../kernel/cacheinfo.h" static struct kobject *mobility_kobj; @@ -47,6 +48,39 @@ struct update_props_workarea { #define MIGRATION_SCOPE (1) #define PRRN_SCOPE -2 +#ifdef CONFIG_PPC_WATCHDOG +static unsigned int nmi_wd_lpm_factor = 200; + +#ifdef CONFIG_SYSCTL +static struct ctl_table nmi_wd_lpm_factor_ctl_table[] = { + { + .procname = "nmi_wd_lpm_factor", + .data = &nmi_wd_lpm_factor, + .maxlen = sizeof(int), + .mode = 0644, + .proc_handler = proc_douintvec_minmax, + }, + {} +}; +static struct ctl_table nmi_wd_lpm_factor_sysctl_root[] = { + { + .procname = "kernel", + .mode = 0555, + .child = nmi_wd_lpm_factor_ctl_table, + }, + {} +}; + +static int __init register_nmi_wd_lpm_factor_sysctl(void) +{ + register_sysctl_table(nmi_wd_lpm_factor_sysctl_root); + + return 0; +} +device_initcall(register_nmi_wd_lpm_factor_sysctl); +#endif /* CONFIG_SYSCTL */ +#endif /* CONFIG_PPC_WATCHDOG */ + static int mobility_rtas_call(int token, char *buf, s32 scope) { int rc; @@ -182,7 +216,7 @@ static int update_dt_node(struct device_node *dn, s32 scope) nprops = be32_to_cpu(upwa->nprops); /* On the first call to ibm,update-properties for a node the - * the first property value descriptor contains an empty + * first property value descriptor contains an empty * property name, the property value length encoded as u32, * and the property value is the node path being updated. */ @@ -265,7 +299,7 @@ static int add_dt_node(struct device_node *parent_dn, __be32 drc_index) return rc; } -int pseries_devicetree_update(s32 scope) +static int pseries_devicetree_update(s32 scope) { char *rtas_buf; __be32 *data; @@ -426,6 +460,43 @@ static int wait_for_vasi_session_suspending(u64 handle) return ret; } +static void wait_for_vasi_session_completed(u64 handle) +{ + unsigned long state = 0; + int ret; + + pr_info("waiting for memory transfer to complete...\n"); + + /* + * Wait for transition from H_VASI_RESUMED to H_VASI_COMPLETED. + */ + while (true) { + ret = poll_vasi_state(handle, &state); + + /* + * If the memory transfer is already complete and the migration + * has been cleaned up by the hypervisor, H_PARAMETER is return, + * which is translate in EINVAL by poll_vasi_state(). + */ + if (ret == -EINVAL || (!ret && state == H_VASI_COMPLETED)) { + pr_info("memory transfer completed.\n"); + break; + } + + if (ret) { + pr_err("H_VASI_STATE return error (%d)\n", ret); + break; + } + + if (state != H_VASI_RESUMED) { + pr_err("unexpected H_VASI_STATE result %lu\n", state); + break; + } + + msleep(500); + } +} + static void prod_single(unsigned int target_cpu) { long hvrc; @@ -451,11 +522,15 @@ static void prod_others(void) static u16 clamp_slb_size(void) { +#ifdef CONFIG_PPC_64S_HASH_MMU u16 prev = mmu_slb_size; slb_set_size(SLB_MIN_SIZE); return prev; +#else + return 0; +#endif } static int do_suspend(void) @@ -660,17 +735,47 @@ static int pseries_suspend(u64 handle) static int pseries_migrate_partition(u64 handle) { int ret; + unsigned int factor = 0; + +#ifdef CONFIG_PPC_WATCHDOG + factor = nmi_wd_lpm_factor; +#endif + /* + * When the migration is initiated, the hypervisor changes VAS + * mappings to prepare before OS gets the notification and + * closes all VAS windows. NX generates continuous faults during + * this time and the user space can not differentiate these + * faults from the migration event. So reduce this time window + * by closing VAS windows at the beginning of this function. + */ + vas_migration_handler(VAS_SUSPEND); ret = wait_for_vasi_session_suspending(handle); if (ret) - return ret; + goto out; + + if (factor) + watchdog_nmi_set_timeout_pct(factor); ret = pseries_suspend(handle); - if (ret == 0) + if (ret == 0) { post_mobility_fixup(); - else + /* + * Wait until the memory transfer is complete, so that the user + * space process returns from the syscall after the transfer is + * complete. This allows the user hooks to be executed at the + * right time. + */ + wait_for_vasi_session_completed(handle); + } else pseries_cancel_migration(handle, ret); + if (factor) + watchdog_nmi_set_timeout_pct(0); + +out: + vas_migration_handler(VAS_RESUME); + return ret; } diff --git a/arch/powerpc/platforms/pseries/msi.c b/arch/powerpc/platforms/pseries/msi.c index 8627362f613e..a3a71d37cb9a 100644 --- a/arch/powerpc/platforms/pseries/msi.c +++ b/arch/powerpc/platforms/pseries/msi.c @@ -7,6 +7,7 @@ #include <linux/crash_dump.h> #include <linux/device.h> #include <linux/irq.h> +#include <linux/irqdomain.h> #include <linux/msi.h> #include <asm/rtas.h> @@ -321,27 +322,6 @@ out: return request; } -static int check_msix_entries(struct pci_dev *pdev) -{ - struct msi_desc *entry; - int expected; - - /* There's no way for us to express to firmware that we want - * a discontiguous, or non-zero based, range of MSI-X entries. - * So we must reject such requests. */ - - expected = 0; - for_each_pci_msi_entry(entry, pdev) { - if (entry->msi_attrib.entry_nr != expected) { - pr_debug("rtas_msi: bad MSI-X entries.\n"); - return -EINVAL; - } - expected++; - } - - return 0; -} - static void rtas_hack_32bit_msi_gen2(struct pci_dev *pdev) { u32 addr_hi, addr_lo; @@ -380,9 +360,6 @@ static int rtas_prepare_msi_irqs(struct pci_dev *pdev, int nvec_in, int type, if (quota && quota < nvec) return quota; - if (type == PCI_CAP_ID_MSIX && check_msix_entries(pdev)) - return -EINVAL; - /* * Firmware currently refuse any non power of two allocation * so we round up if the quota will allow it. @@ -448,8 +425,7 @@ static int pseries_msi_ops_prepare(struct irq_domain *domain, struct device *dev int nvec, msi_alloc_info_t *arg) { struct pci_dev *pdev = to_pci_dev(dev); - struct msi_desc *desc = first_pci_msi_entry(pdev); - int type = desc->msi_attrib.is_msix ? PCI_CAP_ID_MSIX : PCI_CAP_ID_MSI; + int type = pdev->msix_enabled ? PCI_CAP_ID_MSIX : PCI_CAP_ID_MSI; return rtas_prepare_msi_irqs(pdev, nvec, type, arg); } @@ -530,9 +506,16 @@ static struct irq_chip pseries_pci_msi_irq_chip = { .irq_write_msi_msg = pseries_msi_write_msg, }; + +/* + * Set MSI_FLAG_MSIX_CONTIGUOUS as there is no way to express to + * firmware to request a discontiguous or non-zero based range of + * MSI-X entries. Core code will reject such setup attempts. + */ static struct msi_domain_info pseries_msi_domain_info = { .flags = (MSI_FLAG_USE_DEF_DOM_OPS | MSI_FLAG_USE_DEF_CHIP_OPS | - MSI_FLAG_MULTI_PCI_MSI | MSI_FLAG_PCI_MSIX), + MSI_FLAG_MULTI_PCI_MSI | MSI_FLAG_PCI_MSIX | + MSI_FLAG_MSIX_CONTIGUOUS), .ops = &pseries_pci_msi_domain_ops, .chip = &pseries_pci_msi_irq_chip, }; @@ -580,7 +563,7 @@ static int pseries_irq_domain_alloc(struct irq_domain *domain, unsigned int virq int hwirq; int i, ret; - hwirq = rtas_query_irq_number(pci_get_pdn(pdev), desc->msi_attrib.entry_nr); + hwirq = rtas_query_irq_number(pci_get_pdn(pdev), desc->msi_index); if (hwirq < 0) { dev_err(&pdev->dev, "Failed to query HW IRQ: %d\n", hwirq); return hwirq; diff --git a/arch/powerpc/platforms/pseries/nvram.c b/arch/powerpc/platforms/pseries/nvram.c index 69db2eca367f..cbf1720eb4aa 100644 --- a/arch/powerpc/platforms/pseries/nvram.c +++ b/arch/powerpc/platforms/pseries/nvram.c @@ -13,9 +13,9 @@ #include <linux/slab.h> #include <linux/ctype.h> #include <linux/uaccess.h> +#include <linux/of.h> #include <asm/nvram.h> #include <asm/rtas.h> -#include <asm/prom.h> #include <asm/machdep.h> /* Max bytes to read/write in one go */ diff --git a/arch/powerpc/platforms/pseries/papr_platform_attributes.c b/arch/powerpc/platforms/pseries/papr_platform_attributes.c new file mode 100644 index 000000000000..526c621b098b --- /dev/null +++ b/arch/powerpc/platforms/pseries/papr_platform_attributes.c @@ -0,0 +1,362 @@ +// SPDX-License-Identifier: GPL-2.0-or-later +/* + * Platform energy and frequency attributes driver + * + * This driver creates a sys file at /sys/firmware/papr/ which encapsulates a + * directory structure containing files in keyword - value pairs that specify + * energy and frequency configuration of the system. + * + * The format of exposing the sysfs information is as follows: + * /sys/firmware/papr/energy_scale_info/ + * |-- <id>/ + * |-- desc + * |-- value + * |-- value_desc (if exists) + * |-- <id>/ + * |-- desc + * |-- value + * |-- value_desc (if exists) + * + * Copyright 2022 IBM Corp. + */ + +#include <asm/hvcall.h> +#include <asm/machdep.h> +#include <asm/firmware.h> + +#include "pseries.h" + +/* + * Flag attributes to fetch either all or one attribute from the HCALL + * flag = BE(0) => fetch all attributes with firstAttributeId = 0 + * flag = BE(1) => fetch a single attribute with firstAttributeId = id + */ +#define ESI_FLAGS_ALL 0 +#define ESI_FLAGS_SINGLE (1ull << 63) + +#define KOBJ_MAX_ATTRS 3 + +#define ESI_HDR_SIZE sizeof(struct h_energy_scale_info_hdr) +#define ESI_ATTR_SIZE sizeof(struct energy_scale_attribute) +#define CURR_MAX_ESI_ATTRS 8 + +struct energy_scale_attribute { + __be64 id; + __be64 val; + u8 desc[64]; + u8 value_desc[64]; +} __packed; + +struct h_energy_scale_info_hdr { + __be64 num_attrs; + __be64 array_offset; + u8 data_header_version; +} __packed; + +struct papr_attr { + u64 id; + struct kobj_attribute kobj_attr; +}; + +struct papr_group { + struct attribute_group pg; + struct papr_attr pgattrs[KOBJ_MAX_ATTRS]; +}; + +static struct papr_group *papr_groups; +/* /sys/firmware/papr */ +static struct kobject *papr_kobj; +/* /sys/firmware/papr/energy_scale_info */ +static struct kobject *esi_kobj; + +/* + * Energy modes can change dynamically hence making a new hcall each time the + * information needs to be retrieved + */ +static int papr_get_attr(u64 id, struct energy_scale_attribute *esi) +{ + int esi_buf_size = ESI_HDR_SIZE + (CURR_MAX_ESI_ATTRS * ESI_ATTR_SIZE); + int ret, max_esi_attrs = CURR_MAX_ESI_ATTRS; + struct energy_scale_attribute *curr_esi; + struct h_energy_scale_info_hdr *hdr; + char *buf; + + buf = kmalloc(esi_buf_size, GFP_KERNEL); + if (buf == NULL) + return -ENOMEM; + +retry: + ret = plpar_hcall_norets(H_GET_ENERGY_SCALE_INFO, ESI_FLAGS_SINGLE, + id, virt_to_phys(buf), + esi_buf_size); + + /* + * If the hcall fails with not enough memory for either the + * header or data, attempt to allocate more + */ + if (ret == H_PARTIAL || ret == H_P4) { + char *temp_buf; + + max_esi_attrs += 4; + esi_buf_size = ESI_HDR_SIZE + (CURR_MAX_ESI_ATTRS * max_esi_attrs); + + temp_buf = krealloc(buf, esi_buf_size, GFP_KERNEL); + if (temp_buf) + buf = temp_buf; + else + return -ENOMEM; + + goto retry; + } + + if (ret != H_SUCCESS) { + pr_warn("hcall failed: H_GET_ENERGY_SCALE_INFO"); + ret = -EIO; + goto out_buf; + } + + hdr = (struct h_energy_scale_info_hdr *) buf; + curr_esi = (struct energy_scale_attribute *) + (buf + be64_to_cpu(hdr->array_offset)); + + if (esi_buf_size < + be64_to_cpu(hdr->array_offset) + (be64_to_cpu(hdr->num_attrs) + * sizeof(struct energy_scale_attribute))) { + ret = -EIO; + goto out_buf; + } + + *esi = *curr_esi; + +out_buf: + kfree(buf); + + return ret; +} + +/* + * Extract and export the description of the energy scale attributes + */ +static ssize_t desc_show(struct kobject *kobj, + struct kobj_attribute *kobj_attr, + char *buf) +{ + struct papr_attr *pattr = container_of(kobj_attr, struct papr_attr, + kobj_attr); + struct energy_scale_attribute esi; + int ret; + + ret = papr_get_attr(pattr->id, &esi); + if (ret) + return ret; + + return sysfs_emit(buf, "%s\n", esi.desc); +} + +/* + * Extract and export the numeric value of the energy scale attributes + */ +static ssize_t val_show(struct kobject *kobj, + struct kobj_attribute *kobj_attr, + char *buf) +{ + struct papr_attr *pattr = container_of(kobj_attr, struct papr_attr, + kobj_attr); + struct energy_scale_attribute esi; + int ret; + + ret = papr_get_attr(pattr->id, &esi); + if (ret) + return ret; + + return sysfs_emit(buf, "%llu\n", be64_to_cpu(esi.val)); +} + +/* + * Extract and export the value description in string format of the energy + * scale attributes + */ +static ssize_t val_desc_show(struct kobject *kobj, + struct kobj_attribute *kobj_attr, + char *buf) +{ + struct papr_attr *pattr = container_of(kobj_attr, struct papr_attr, + kobj_attr); + struct energy_scale_attribute esi; + int ret; + + ret = papr_get_attr(pattr->id, &esi); + if (ret) + return ret; + + return sysfs_emit(buf, "%s\n", esi.value_desc); +} + +static struct papr_ops_info { + const char *attr_name; + ssize_t (*show)(struct kobject *kobj, struct kobj_attribute *kobj_attr, + char *buf); +} ops_info[KOBJ_MAX_ATTRS] = { + { "desc", desc_show }, + { "value", val_show }, + { "value_desc", val_desc_show }, +}; + +static void add_attr(u64 id, int index, struct papr_attr *attr) +{ + attr->id = id; + sysfs_attr_init(&attr->kobj_attr.attr); + attr->kobj_attr.attr.name = ops_info[index].attr_name; + attr->kobj_attr.attr.mode = 0444; + attr->kobj_attr.show = ops_info[index].show; +} + +static int add_attr_group(u64 id, struct papr_group *pg, bool show_val_desc) +{ + int i; + + for (i = 0; i < KOBJ_MAX_ATTRS; i++) { + if (!strcmp(ops_info[i].attr_name, "value_desc") && + !show_val_desc) { + continue; + } + add_attr(id, i, &pg->pgattrs[i]); + pg->pg.attrs[i] = &pg->pgattrs[i].kobj_attr.attr; + } + + return sysfs_create_group(esi_kobj, &pg->pg); +} + + +static int __init papr_init(void) +{ + int esi_buf_size = ESI_HDR_SIZE + (CURR_MAX_ESI_ATTRS * ESI_ATTR_SIZE); + int ret, idx, i, max_esi_attrs = CURR_MAX_ESI_ATTRS; + struct h_energy_scale_info_hdr *esi_hdr; + struct energy_scale_attribute *esi_attrs; + uint64_t num_attrs; + char *esi_buf; + + if (!firmware_has_feature(FW_FEATURE_LPAR) || + !firmware_has_feature(FW_FEATURE_ENERGY_SCALE_INFO)) { + return -ENXIO; + } + + esi_buf = kmalloc(esi_buf_size, GFP_KERNEL); + if (esi_buf == NULL) + return -ENOMEM; + /* + * hcall( + * uint64 H_GET_ENERGY_SCALE_INFO, // Get energy scale info + * uint64 flags, // Per the flag request + * uint64 firstAttributeId, // The attribute id + * uint64 bufferAddress, // Guest physical address of the output buffer + * uint64 bufferSize); // The size in bytes of the output buffer + */ +retry: + + ret = plpar_hcall_norets(H_GET_ENERGY_SCALE_INFO, ESI_FLAGS_ALL, 0, + virt_to_phys(esi_buf), esi_buf_size); + + /* + * If the hcall fails with not enough memory for either the + * header or data, attempt to allocate more + */ + if (ret == H_PARTIAL || ret == H_P4) { + char *temp_esi_buf; + + max_esi_attrs += 4; + esi_buf_size = ESI_HDR_SIZE + (CURR_MAX_ESI_ATTRS * max_esi_attrs); + + temp_esi_buf = krealloc(esi_buf, esi_buf_size, GFP_KERNEL); + if (temp_esi_buf) + esi_buf = temp_esi_buf; + else + return -ENOMEM; + + goto retry; + } + + if (ret != H_SUCCESS) { + pr_warn("hcall failed: H_GET_ENERGY_SCALE_INFO, ret: %d\n", ret); + goto out_free_esi_buf; + } + + esi_hdr = (struct h_energy_scale_info_hdr *) esi_buf; + num_attrs = be64_to_cpu(esi_hdr->num_attrs); + esi_attrs = (struct energy_scale_attribute *) + (esi_buf + be64_to_cpu(esi_hdr->array_offset)); + + if (esi_buf_size < + be64_to_cpu(esi_hdr->array_offset) + + (num_attrs * sizeof(struct energy_scale_attribute))) { + goto out_free_esi_buf; + } + + papr_groups = kcalloc(num_attrs, sizeof(*papr_groups), GFP_KERNEL); + if (!papr_groups) + goto out_free_esi_buf; + + papr_kobj = kobject_create_and_add("papr", firmware_kobj); + if (!papr_kobj) { + pr_warn("kobject_create_and_add papr failed\n"); + goto out_papr_groups; + } + + esi_kobj = kobject_create_and_add("energy_scale_info", papr_kobj); + if (!esi_kobj) { + pr_warn("kobject_create_and_add energy_scale_info failed\n"); + goto out_kobj; + } + + /* Allocate the groups before registering */ + for (idx = 0; idx < num_attrs; idx++) { + papr_groups[idx].pg.attrs = kcalloc(KOBJ_MAX_ATTRS + 1, + sizeof(*papr_groups[idx].pg.attrs), + GFP_KERNEL); + if (!papr_groups[idx].pg.attrs) + goto out_pgattrs; + + papr_groups[idx].pg.name = kasprintf(GFP_KERNEL, "%lld", + be64_to_cpu(esi_attrs[idx].id)); + if (papr_groups[idx].pg.name == NULL) + goto out_pgattrs; + } + + for (idx = 0; idx < num_attrs; idx++) { + bool show_val_desc = true; + + /* Do not add the value desc attr if it does not exist */ + if (strnlen(esi_attrs[idx].value_desc, + sizeof(esi_attrs[idx].value_desc)) == 0) + show_val_desc = false; + + if (add_attr_group(be64_to_cpu(esi_attrs[idx].id), + &papr_groups[idx], + show_val_desc)) { + pr_warn("Failed to create papr attribute group %s\n", + papr_groups[idx].pg.name); + idx = num_attrs; + goto out_pgattrs; + } + } + + kfree(esi_buf); + return 0; +out_pgattrs: + for (i = 0; i < idx ; i++) { + kfree(papr_groups[i].pg.attrs); + kfree(papr_groups[i].pg.name); + } + kobject_put(esi_kobj); +out_kobj: + kobject_put(papr_kobj); +out_papr_groups: + kfree(papr_groups); +out_free_esi_buf: + kfree(esi_buf); + + return -ENOMEM; +} + +machine_device_initcall(pseries, papr_init); diff --git a/arch/powerpc/platforms/pseries/papr_scm.c b/arch/powerpc/platforms/pseries/papr_scm.c index f48e87ac89c9..2f8385523a13 100644 --- a/arch/powerpc/platforms/pseries/papr_scm.c +++ b/arch/powerpc/platforms/pseries/papr_scm.c @@ -19,6 +19,7 @@ #include <asm/papr_pdsm.h> #include <asm/mce.h> #include <asm/unaligned.h> +#include <linux/perf_event.h> #define BIND_ANY_ADDR (~0ul) @@ -28,7 +29,7 @@ (1ul << ND_CMD_SET_CONFIG_DATA) | \ (1ul << ND_CMD_CALL)) -/* DIMM health bitmap bitmap indicators */ +/* DIMM health bitmap indicators */ /* SCM device is unable to persist memory contents */ #define PAPR_PMEM_UNARMED (1ULL << (63 - 0)) /* SCM device failed to persist memory contents */ @@ -120,6 +121,9 @@ struct papr_scm_priv { /* length of the stat buffer as expected by phyp */ size_t stat_buffer_len; + + /* The bits which needs to be overridden */ + u64 health_bitmap_inject_mask; }; static int papr_scm_pmem_flush(struct nd_region *nd_region, @@ -340,6 +344,191 @@ static ssize_t drc_pmem_query_stats(struct papr_scm_priv *p, return 0; } +#ifdef CONFIG_PERF_EVENTS +#define to_nvdimm_pmu(_pmu) container_of(_pmu, struct nvdimm_pmu, pmu) + +static const char * const nvdimm_events_map[] = { + [1] = "CtlResCt", + [2] = "CtlResTm", + [3] = "PonSecs ", + [4] = "MemLife ", + [5] = "CritRscU", + [6] = "HostLCnt", + [7] = "HostSCnt", + [8] = "HostSDur", + [9] = "HostLDur", + [10] = "MedRCnt ", + [11] = "MedWCnt ", + [12] = "MedRDur ", + [13] = "MedWDur ", + [14] = "CchRHCnt", + [15] = "CchWHCnt", + [16] = "FastWCnt", +}; + +static int papr_scm_pmu_get_value(struct perf_event *event, struct device *dev, u64 *count) +{ + struct papr_scm_perf_stat *stat; + struct papr_scm_perf_stats *stats; + struct papr_scm_priv *p = dev_get_drvdata(dev); + int rc, size; + + /* Invalid eventcode */ + if (event->attr.config == 0 || event->attr.config >= ARRAY_SIZE(nvdimm_events_map)) + return -EINVAL; + + /* Allocate request buffer enough to hold single performance stat */ + size = sizeof(struct papr_scm_perf_stats) + + sizeof(struct papr_scm_perf_stat); + + if (!p) + return -EINVAL; + + stats = kzalloc(size, GFP_KERNEL); + if (!stats) + return -ENOMEM; + + stat = &stats->scm_statistic[0]; + memcpy(&stat->stat_id, + nvdimm_events_map[event->attr.config], + sizeof(stat->stat_id)); + stat->stat_val = 0; + + rc = drc_pmem_query_stats(p, stats, 1); + if (rc < 0) { + kfree(stats); + return rc; + } + + *count = be64_to_cpu(stat->stat_val); + kfree(stats); + return 0; +} + +static int papr_scm_pmu_event_init(struct perf_event *event) +{ + struct nvdimm_pmu *nd_pmu = to_nvdimm_pmu(event->pmu); + struct papr_scm_priv *p; + + if (!nd_pmu) + return -EINVAL; + + /* test the event attr type for PMU enumeration */ + if (event->attr.type != event->pmu->type) + return -ENOENT; + + /* it does not support event sampling mode */ + if (is_sampling_event(event)) + return -EOPNOTSUPP; + + /* no branch sampling */ + if (has_branch_stack(event)) + return -EOPNOTSUPP; + + p = (struct papr_scm_priv *)nd_pmu->dev->driver_data; + if (!p) + return -EINVAL; + + /* Invalid eventcode */ + if (event->attr.config == 0 || event->attr.config > 16) + return -EINVAL; + + return 0; +} + +static int papr_scm_pmu_add(struct perf_event *event, int flags) +{ + u64 count; + int rc; + struct nvdimm_pmu *nd_pmu = to_nvdimm_pmu(event->pmu); + + if (!nd_pmu) + return -EINVAL; + + if (flags & PERF_EF_START) { + rc = papr_scm_pmu_get_value(event, nd_pmu->dev, &count); + if (rc) + return rc; + + local64_set(&event->hw.prev_count, count); + } + + return 0; +} + +static void papr_scm_pmu_read(struct perf_event *event) +{ + u64 prev, now; + int rc; + struct nvdimm_pmu *nd_pmu = to_nvdimm_pmu(event->pmu); + + if (!nd_pmu) + return; + + rc = papr_scm_pmu_get_value(event, nd_pmu->dev, &now); + if (rc) + return; + + prev = local64_xchg(&event->hw.prev_count, now); + local64_add(now - prev, &event->count); +} + +static void papr_scm_pmu_del(struct perf_event *event, int flags) +{ + papr_scm_pmu_read(event); +} + +static void papr_scm_pmu_register(struct papr_scm_priv *p) +{ + struct nvdimm_pmu *nd_pmu; + int rc, nodeid; + + nd_pmu = kzalloc(sizeof(*nd_pmu), GFP_KERNEL); + if (!nd_pmu) { + rc = -ENOMEM; + goto pmu_err_print; + } + + if (!p->stat_buffer_len) { + rc = -ENOENT; + goto pmu_check_events_err; + } + + nd_pmu->pmu.task_ctx_nr = perf_invalid_context; + nd_pmu->pmu.name = nvdimm_name(p->nvdimm); + nd_pmu->pmu.event_init = papr_scm_pmu_event_init; + nd_pmu->pmu.read = papr_scm_pmu_read; + nd_pmu->pmu.add = papr_scm_pmu_add; + nd_pmu->pmu.del = papr_scm_pmu_del; + + nd_pmu->pmu.capabilities = PERF_PMU_CAP_NO_INTERRUPT | + PERF_PMU_CAP_NO_EXCLUDE; + + /*updating the cpumask variable */ + nodeid = numa_map_to_online_node(dev_to_node(&p->pdev->dev)); + nd_pmu->arch_cpumask = *cpumask_of_node(nodeid); + + rc = register_nvdimm_pmu(nd_pmu, p->pdev); + if (rc) + goto pmu_check_events_err; + + /* + * Set archdata.priv value to nvdimm_pmu structure, to handle the + * unregistering of pmu device. + */ + p->pdev->archdata.priv = nd_pmu; + return; + +pmu_check_events_err: + kfree(nd_pmu); +pmu_err_print: + dev_info(&p->pdev->dev, "nvdimm pmu didn't register rc=%d\n", rc); +} + +#else +static void papr_scm_pmu_register(struct papr_scm_priv *p) { } +#endif + /* * Issue hcall to retrieve dimm health info and populate papr_scm_priv with the * health information. @@ -347,19 +536,29 @@ static ssize_t drc_pmem_query_stats(struct papr_scm_priv *p, static int __drc_pmem_query_health(struct papr_scm_priv *p) { unsigned long ret[PLPAR_HCALL_BUFSIZE]; + u64 bitmap = 0; long rc; /* issue the hcall */ rc = plpar_hcall(H_SCM_HEALTH, ret, p->drc_index); - if (rc != H_SUCCESS) { + if (rc == H_SUCCESS) + bitmap = ret[0] & ret[1]; + else if (rc == H_FUNCTION) + dev_info_once(&p->pdev->dev, + "Hcall H_SCM_HEALTH not implemented, assuming empty health bitmap"); + else { + dev_err(&p->pdev->dev, "Failed to query health information, Err:%ld\n", rc); return -ENXIO; } p->lasthealth_jiffies = jiffies; - p->health_bitmap = ret[0] & ret[1]; - + /* Allow injecting specific health bits via inject mask. */ + if (p->health_bitmap_inject_mask) + bitmap = (bitmap & ~p->health_bitmap_inject_mask) | + p->health_bitmap_inject_mask; + WRITE_ONCE(p->health_bitmap, bitmap); dev_dbg(&p->pdev->dev, "Queried dimm health info. Bitmap:0x%016lx Mask:0x%016lx\n", ret[0], ret[1]); @@ -669,6 +868,56 @@ out: return rc; } +/* Inject a smart error Add the dirty-shutdown-counter value to the pdsm */ +static int papr_pdsm_smart_inject(struct papr_scm_priv *p, + union nd_pdsm_payload *payload) +{ + int rc; + u32 supported_flags = 0; + u64 inject_mask = 0, clear_mask = 0; + u64 mask; + + /* Check for individual smart error flags and update inject/clear masks */ + if (payload->smart_inject.flags & PDSM_SMART_INJECT_HEALTH_FATAL) { + supported_flags |= PDSM_SMART_INJECT_HEALTH_FATAL; + if (payload->smart_inject.fatal_enable) + inject_mask |= PAPR_PMEM_HEALTH_FATAL; + else + clear_mask |= PAPR_PMEM_HEALTH_FATAL; + } + + if (payload->smart_inject.flags & PDSM_SMART_INJECT_BAD_SHUTDOWN) { + supported_flags |= PDSM_SMART_INJECT_BAD_SHUTDOWN; + if (payload->smart_inject.unsafe_shutdown_enable) + inject_mask |= PAPR_PMEM_SHUTDOWN_DIRTY; + else + clear_mask |= PAPR_PMEM_SHUTDOWN_DIRTY; + } + + dev_dbg(&p->pdev->dev, "[Smart-inject] inject_mask=%#llx clear_mask=%#llx\n", + inject_mask, clear_mask); + + /* Prevent concurrent access to dimm health bitmap related members */ + rc = mutex_lock_interruptible(&p->health_mutex); + if (rc) + return rc; + + /* Use inject/clear masks to set health_bitmap_inject_mask */ + mask = READ_ONCE(p->health_bitmap_inject_mask); + mask = (mask & ~clear_mask) | inject_mask; + WRITE_ONCE(p->health_bitmap_inject_mask, mask); + + /* Invalidate cached health bitmap */ + p->lasthealth_jiffies = 0; + + mutex_unlock(&p->health_mutex); + + /* Return the supported flags back to userspace */ + payload->smart_inject.flags = supported_flags; + + return sizeof(struct nd_papr_pdsm_health); +} + /* * 'struct pdsm_cmd_desc' * Identifies supported PDSMs' expected length of in/out payloads @@ -702,6 +951,12 @@ static const struct pdsm_cmd_desc __pdsm_cmd_descriptors[] = { .size_out = sizeof(struct nd_papr_pdsm_health), .service = papr_pdsm_health, }, + + [PAPR_PDSM_SMART_INJECT] = { + .size_in = sizeof(struct nd_papr_pdsm_smart_inject), + .size_out = sizeof(struct nd_papr_pdsm_smart_inject), + .service = papr_pdsm_smart_inject, + }, /* Empty */ [PAPR_PDSM_MAX] = { .size_in = 0, @@ -838,6 +1093,19 @@ static int papr_scm_ndctl(struct nvdimm_bus_descriptor *nd_desc, return 0; } +static ssize_t health_bitmap_inject_show(struct device *dev, + struct device_attribute *attr, + char *buf) +{ + struct nvdimm *dimm = to_nvdimm(dev); + struct papr_scm_priv *p = nvdimm_provider_data(dimm); + + return sprintf(buf, "%#llx\n", + READ_ONCE(p->health_bitmap_inject_mask)); +} + +static DEVICE_ATTR_ADMIN_RO(health_bitmap_inject); + static ssize_t perf_stats_show(struct device *dev, struct device_attribute *attr, char *buf) { @@ -952,10 +1220,11 @@ static struct attribute *papr_nd_attributes[] = { &dev_attr_flags.attr, &dev_attr_perf_stats.attr, &dev_attr_dirty_shutdown.attr, + &dev_attr_health_bitmap_inject.attr, NULL, }; -static struct attribute_group papr_nd_attribute_group = { +static const struct attribute_group papr_nd_attribute_group = { .name = "papr", .is_visible = papr_nd_attribute_visible, .attrs = papr_nd_attributes, @@ -1236,6 +1505,7 @@ static int papr_scm_probe(struct platform_device *pdev) goto err2; platform_set_drvdata(pdev, p); + papr_scm_pmu_register(p); return 0; @@ -1254,6 +1524,11 @@ static int papr_scm_remove(struct platform_device *pdev) nvdimm_bus_unregister(p->bus); drc_pmem_unbind(p); + + if (pdev->archdata.priv) + unregister_nvdimm_pmu(pdev->archdata.priv); + + pdev->archdata.priv = NULL; kfree(p->bus_desc.provider_name); kfree(p); diff --git a/arch/powerpc/platforms/pseries/pci.c b/arch/powerpc/platforms/pseries/pci.c index 3b6800f774c2..6e671c3809ec 100644 --- a/arch/powerpc/platforms/pseries/pci.c +++ b/arch/powerpc/platforms/pseries/pci.c @@ -14,7 +14,6 @@ #include <asm/eeh.h> #include <asm/pci-bridge.h> -#include <asm/prom.h> #include <asm/ppc-pci.h> #include <asm/pci.h> #include "pseries.h" diff --git a/arch/powerpc/platforms/pseries/pci_dlpar.c b/arch/powerpc/platforms/pseries/pci_dlpar.c index 90c9d3531694..4ba824568119 100644 --- a/arch/powerpc/platforms/pseries/pci_dlpar.c +++ b/arch/powerpc/platforms/pseries/pci_dlpar.c @@ -78,6 +78,9 @@ int remove_phb_dynamic(struct pci_controller *phb) pseries_msi_free_domains(phb); + /* Keep a reference so phb isn't freed yet */ + get_device(&host_bridge->dev); + /* Remove the PCI bus and unregister the bridge device from sysfs */ phb->bus = NULL; pci_remove_bus(b); @@ -101,6 +104,7 @@ int remove_phb_dynamic(struct pci_controller *phb) * the pcibios_free_controller_deferred() callback; * see pseries_root_bridge_prepare(). */ + put_device(&host_bridge->dev); return 0; } diff --git a/arch/powerpc/platforms/pseries/plpks.c b/arch/powerpc/platforms/pseries/plpks.c new file mode 100644 index 000000000000..f4b5b5a64db3 --- /dev/null +++ b/arch/powerpc/platforms/pseries/plpks.c @@ -0,0 +1,461 @@ +// SPDX-License-Identifier: GPL-2.0-only +/* + * POWER LPAR Platform KeyStore(PLPKS) + * Copyright (C) 2022 IBM Corporation + * Author: Nayna Jain <nayna@linux.ibm.com> + * + * Provides access to variables stored in Power LPAR Platform KeyStore(PLPKS). + */ + +#define pr_fmt(fmt) "plpks: " fmt + +#include <linux/delay.h> +#include <linux/errno.h> +#include <linux/io.h> +#include <linux/printk.h> +#include <linux/slab.h> +#include <linux/string.h> +#include <linux/types.h> +#include <asm/hvcall.h> +#include <asm/machdep.h> + +#include "plpks.h" + +#define PKS_FW_OWNER 0x1 +#define PKS_BOOTLOADER_OWNER 0x2 +#define PKS_OS_OWNER 0x3 + +#define LABEL_VERSION 0 +#define MAX_LABEL_ATTR_SIZE 16 +#define MAX_NAME_SIZE 239 +#define MAX_DATA_SIZE 4000 + +#define PKS_FLUSH_MAX_TIMEOUT 5000 //msec +#define PKS_FLUSH_SLEEP 10 //msec +#define PKS_FLUSH_SLEEP_RANGE 400 + +static u8 *ospassword; +static u16 ospasswordlength; + +// Retrieved with H_PKS_GET_CONFIG +static u16 maxpwsize; +static u16 maxobjsize; + +struct plpks_auth { + u8 version; + u8 consumer; + __be64 rsvd0; + __be32 rsvd1; + __be16 passwordlength; + u8 password[]; +} __packed __aligned(16); + +struct label_attr { + u8 prefix[8]; + u8 version; + u8 os; + u8 length; + u8 reserved[5]; +}; + +struct label { + struct label_attr attr; + u8 name[MAX_NAME_SIZE]; + size_t size; +}; + +static int pseries_status_to_err(int rc) +{ + int err; + + switch (rc) { + case H_SUCCESS: + err = 0; + break; + case H_FUNCTION: + err = -ENXIO; + break; + case H_P1: + case H_P2: + case H_P3: + case H_P4: + case H_P5: + case H_P6: + err = -EINVAL; + break; + case H_NOT_FOUND: + err = -ENOENT; + break; + case H_BUSY: + err = -EBUSY; + break; + case H_AUTHORITY: + err = -EPERM; + break; + case H_NO_MEM: + err = -ENOMEM; + break; + case H_RESOURCE: + err = -EEXIST; + break; + case H_TOO_BIG: + err = -EFBIG; + break; + case H_STATE: + err = -EIO; + break; + case H_R_STATE: + err = -EIO; + break; + case H_IN_USE: + err = -EEXIST; + break; + case H_ABORTED: + err = -EINTR; + break; + default: + err = -EINVAL; + } + + return err; +} + +static int plpks_gen_password(void) +{ + unsigned long retbuf[PLPAR_HCALL_BUFSIZE] = { 0 }; + u8 *password, consumer = PKS_OS_OWNER; + int rc; + + password = kzalloc(maxpwsize, GFP_KERNEL); + if (!password) + return -ENOMEM; + + rc = plpar_hcall(H_PKS_GEN_PASSWORD, retbuf, consumer, 0, + virt_to_phys(password), maxpwsize); + + if (!rc) { + ospasswordlength = maxpwsize; + ospassword = kzalloc(maxpwsize, GFP_KERNEL); + if (!ospassword) { + kfree(password); + return -ENOMEM; + } + memcpy(ospassword, password, ospasswordlength); + } else { + if (rc == H_IN_USE) { + pr_warn("Password is already set for POWER LPAR Platform KeyStore\n"); + rc = 0; + } else { + goto out; + } + } +out: + kfree(password); + + return pseries_status_to_err(rc); +} + +static struct plpks_auth *construct_auth(u8 consumer) +{ + struct plpks_auth *auth; + + if (consumer > PKS_OS_OWNER) + return ERR_PTR(-EINVAL); + + auth = kmalloc(struct_size(auth, password, maxpwsize), GFP_KERNEL); + if (!auth) + return ERR_PTR(-ENOMEM); + + auth->version = 1; + auth->consumer = consumer; + auth->rsvd0 = 0; + auth->rsvd1 = 0; + + if (consumer == PKS_FW_OWNER || consumer == PKS_BOOTLOADER_OWNER) { + auth->passwordlength = 0; + return auth; + } + + memcpy(auth->password, ospassword, ospasswordlength); + + auth->passwordlength = cpu_to_be16(ospasswordlength); + + return auth; +} + +/** + * Label is combination of label attributes + name. + * Label attributes are used internally by kernel and not exposed to the user. + */ +static struct label *construct_label(char *component, u8 varos, u8 *name, + u16 namelen) +{ + struct label *label; + size_t slen; + + if (!name || namelen > MAX_NAME_SIZE) + return ERR_PTR(-EINVAL); + + slen = strlen(component); + if (component && slen > sizeof(label->attr.prefix)) + return ERR_PTR(-EINVAL); + + label = kzalloc(sizeof(*label), GFP_KERNEL); + if (!label) + return ERR_PTR(-ENOMEM); + + if (component) + memcpy(&label->attr.prefix, component, slen); + + label->attr.version = LABEL_VERSION; + label->attr.os = varos; + label->attr.length = MAX_LABEL_ATTR_SIZE; + memcpy(&label->name, name, namelen); + + label->size = sizeof(struct label_attr) + namelen; + + return label; +} + +static int _plpks_get_config(void) +{ + unsigned long retbuf[PLPAR_HCALL_BUFSIZE] = { 0 }; + struct { + u8 version; + u8 flags; + __be32 rsvd0; + __be16 maxpwsize; + __be16 maxobjlabelsize; + __be16 maxobjsize; + __be32 totalsize; + __be32 usedspace; + __be32 supportedpolicies; + __be64 rsvd1; + } __packed config; + size_t size; + int rc; + + size = sizeof(config); + + rc = plpar_hcall(H_PKS_GET_CONFIG, retbuf, virt_to_phys(&config), size); + + if (rc != H_SUCCESS) + return pseries_status_to_err(rc); + + maxpwsize = be16_to_cpu(config.maxpwsize); + maxobjsize = be16_to_cpu(config.maxobjsize); + + return 0; +} + +static int plpks_confirm_object_flushed(struct label *label, + struct plpks_auth *auth) +{ + unsigned long retbuf[PLPAR_HCALL_BUFSIZE] = { 0 }; + u64 timeout = 0; + u8 status; + int rc; + + do { + rc = plpar_hcall(H_PKS_CONFIRM_OBJECT_FLUSHED, retbuf, + virt_to_phys(auth), virt_to_phys(label), + label->size); + + status = retbuf[0]; + if (rc) { + if (rc == H_NOT_FOUND && status == 1) + rc = 0; + break; + } + + if (!rc && status == 1) + break; + + usleep_range(PKS_FLUSH_SLEEP, + PKS_FLUSH_SLEEP + PKS_FLUSH_SLEEP_RANGE); + timeout = timeout + PKS_FLUSH_SLEEP; + } while (timeout < PKS_FLUSH_MAX_TIMEOUT); + + rc = pseries_status_to_err(rc); + + return rc; +} + +int plpks_write_var(struct plpks_var var) +{ + unsigned long retbuf[PLPAR_HCALL_BUFSIZE] = { 0 }; + struct plpks_auth *auth; + struct label *label; + int rc; + + if (!var.component || !var.data || var.datalen <= 0 || + var.namelen > MAX_NAME_SIZE || var.datalen > MAX_DATA_SIZE) + return -EINVAL; + + if (var.policy & SIGNEDUPDATE) + return -EINVAL; + + auth = construct_auth(PKS_OS_OWNER); + if (IS_ERR(auth)) + return PTR_ERR(auth); + + label = construct_label(var.component, var.os, var.name, var.namelen); + if (IS_ERR(label)) { + rc = PTR_ERR(label); + goto out; + } + + rc = plpar_hcall(H_PKS_WRITE_OBJECT, retbuf, virt_to_phys(auth), + virt_to_phys(label), label->size, var.policy, + virt_to_phys(var.data), var.datalen); + + if (!rc) + rc = plpks_confirm_object_flushed(label, auth); + + if (rc) + pr_err("Failed to write variable %s for component %s with error %d\n", + var.name, var.component, rc); + + rc = pseries_status_to_err(rc); + kfree(label); +out: + kfree(auth); + + return rc; +} + +int plpks_remove_var(char *component, u8 varos, struct plpks_var_name vname) +{ + unsigned long retbuf[PLPAR_HCALL_BUFSIZE] = { 0 }; + struct plpks_auth *auth; + struct label *label; + int rc; + + if (!component || vname.namelen > MAX_NAME_SIZE) + return -EINVAL; + + auth = construct_auth(PKS_OS_OWNER); + if (IS_ERR(auth)) + return PTR_ERR(auth); + + label = construct_label(component, varos, vname.name, vname.namelen); + if (IS_ERR(label)) { + rc = PTR_ERR(label); + goto out; + } + + rc = plpar_hcall(H_PKS_REMOVE_OBJECT, retbuf, virt_to_phys(auth), + virt_to_phys(label), label->size); + + if (!rc) + rc = plpks_confirm_object_flushed(label, auth); + + if (rc) + pr_err("Failed to remove variable %s for component %s with error %d\n", + vname.name, component, rc); + + rc = pseries_status_to_err(rc); + kfree(label); +out: + kfree(auth); + + return rc; +} + +static int plpks_read_var(u8 consumer, struct plpks_var *var) +{ + unsigned long retbuf[PLPAR_HCALL_BUFSIZE] = { 0 }; + struct plpks_auth *auth; + struct label *label; + u8 *output; + int rc; + + if (var->namelen > MAX_NAME_SIZE) + return -EINVAL; + + auth = construct_auth(PKS_OS_OWNER); + if (IS_ERR(auth)) + return PTR_ERR(auth); + + label = construct_label(var->component, var->os, var->name, + var->namelen); + if (IS_ERR(label)) { + rc = PTR_ERR(label); + goto out_free_auth; + } + + output = kzalloc(maxobjsize, GFP_KERNEL); + if (!output) { + rc = -ENOMEM; + goto out_free_label; + } + + rc = plpar_hcall(H_PKS_READ_OBJECT, retbuf, virt_to_phys(auth), + virt_to_phys(label), label->size, virt_to_phys(output), + maxobjsize); + + if (rc != H_SUCCESS) { + pr_err("Failed to read variable %s for component %s with error %d\n", + var->name, var->component, rc); + rc = pseries_status_to_err(rc); + goto out_free_output; + } + + if (var->datalen == 0 || var->datalen > retbuf[0]) + var->datalen = retbuf[0]; + + var->data = kzalloc(var->datalen, GFP_KERNEL); + if (!var->data) { + rc = -ENOMEM; + goto out_free_output; + } + var->policy = retbuf[1]; + + memcpy(var->data, output, var->datalen); + rc = 0; + +out_free_output: + kfree(output); +out_free_label: + kfree(label); +out_free_auth: + kfree(auth); + + return rc; +} + +int plpks_read_os_var(struct plpks_var *var) +{ + return plpks_read_var(PKS_OS_OWNER, var); +} + +int plpks_read_fw_var(struct plpks_var *var) +{ + return plpks_read_var(PKS_FW_OWNER, var); +} + +int plpks_read_bootloader_var(struct plpks_var *var) +{ + return plpks_read_var(PKS_BOOTLOADER_OWNER, var); +} + +static __init int pseries_plpks_init(void) +{ + int rc; + + rc = _plpks_get_config(); + + if (rc) { + pr_err("POWER LPAR Platform KeyStore is not supported or enabled\n"); + return rc; + } + + rc = plpks_gen_password(); + if (rc) + pr_err("Failed setting POWER LPAR Platform KeyStore Password\n"); + else + pr_info("POWER LPAR Platform KeyStore initialized successfully\n"); + + return rc; +} +machine_arch_initcall(pseries, pseries_plpks_init); diff --git a/arch/powerpc/platforms/pseries/plpks.h b/arch/powerpc/platforms/pseries/plpks.h new file mode 100644 index 000000000000..c6a291367bb1 --- /dev/null +++ b/arch/powerpc/platforms/pseries/plpks.h @@ -0,0 +1,71 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +/* + * Copyright (C) 2022 IBM Corporation + * Author: Nayna Jain <nayna@linux.ibm.com> + * + * Platform keystore for pseries LPAR(PLPKS). + */ + +#ifndef _PSERIES_PLPKS_H +#define _PSERIES_PLPKS_H + +#include <linux/types.h> +#include <linux/list.h> + +#define OSSECBOOTAUDIT 0x40000000 +#define OSSECBOOTENFORCE 0x20000000 +#define WORLDREADABLE 0x08000000 +#define SIGNEDUPDATE 0x01000000 + +#define PLPKS_VAR_LINUX 0x01 +#define PLPKS_VAR_COMMON 0x04 + +struct plpks_var { + char *component; + u8 *name; + u8 *data; + u32 policy; + u16 namelen; + u16 datalen; + u8 os; +}; + +struct plpks_var_name { + u8 *name; + u16 namelen; +}; + +struct plpks_var_name_list { + u32 varcount; + struct plpks_var_name varlist[]; +}; + +/** + * Writes the specified var and its data to PKS. + * Any caller of PKS driver should present a valid component type for + * their variable. + */ +int plpks_write_var(struct plpks_var var); + +/** + * Removes the specified var and its data from PKS. + */ +int plpks_remove_var(char *component, u8 varos, + struct plpks_var_name vname); + +/** + * Returns the data for the specified os variable. + */ +int plpks_read_os_var(struct plpks_var *var); + +/** + * Returns the data for the specified firmware variable. + */ +int plpks_read_fw_var(struct plpks_var *var); + +/** + * Returns the data for the specified bootloader variable. + */ +int plpks_read_bootloader_var(struct plpks_var *var); + +#endif diff --git a/arch/powerpc/platforms/pseries/pmem.c b/arch/powerpc/platforms/pseries/pmem.c index 439ac72c2470..3c290b9ed01b 100644 --- a/arch/powerpc/platforms/pseries/pmem.c +++ b/arch/powerpc/platforms/pseries/pmem.c @@ -15,7 +15,6 @@ #include <linux/of.h> #include <linux/of_platform.h> #include <linux/slab.h> -#include <asm/prom.h> #include <asm/rtas.h> #include <asm/firmware.h> #include <asm/machdep.h> diff --git a/arch/powerpc/platforms/pseries/power.c b/arch/powerpc/platforms/pseries/power.c index ee343ec6ab94..3676cb297767 100644 --- a/arch/powerpc/platforms/pseries/power.c +++ b/arch/powerpc/platforms/pseries/power.c @@ -51,7 +51,7 @@ static struct attribute *g[] = { NULL, }; -static struct attribute_group attr_group = { +static const struct attribute_group attr_group = { .attrs = g, }; diff --git a/arch/powerpc/platforms/pseries/pseries.h b/arch/powerpc/platforms/pseries/pseries.h index 3544778e06d0..1d75b7742ef0 100644 --- a/arch/powerpc/platforms/pseries/pseries.h +++ b/arch/powerpc/platforms/pseries/pseries.h @@ -11,7 +11,7 @@ struct device_node; -extern void request_event_sources_irqs(struct device_node *np, +void __init request_event_sources_irqs(struct device_node *np, irq_handler_t handler, const char *name); #include <linux/of.h> @@ -21,6 +21,7 @@ struct pt_regs; extern int pSeries_system_reset_exception(struct pt_regs *regs); extern int pSeries_machine_check_exception(struct pt_regs *regs); extern long pseries_machine_check_realmode(struct pt_regs *regs); +void pSeries_machine_check_log_err(void); #ifdef CONFIG_SMP extern void smp_init_pseries(void); @@ -37,6 +38,7 @@ static inline void smp_init_pseries(void) { } #endif extern void pseries_kexec_cpu_down(int crash_shutdown, int secondary); +void pseries_machine_kexec(struct kimage *image); extern void pSeries_final_fixup(void); @@ -113,6 +115,13 @@ int dlpar_workqueue_init(void); extern u32 pseries_security_flavor; void pseries_setup_security_mitigations(void); + +#ifdef CONFIG_PPC_64S_HASH_MMU void pseries_lpar_read_hblkrm_characteristics(void); +#else +static inline void pseries_lpar_read_hblkrm_characteristics(void) { } +#endif + +void pseries_rng_init(void); #endif /* _PSERIES_PSERIES_H */ diff --git a/arch/powerpc/platforms/pseries/ras.c b/arch/powerpc/platforms/pseries/ras.c index 56092dccfdb8..f12516c3998c 100644 --- a/arch/powerpc/platforms/pseries/ras.c +++ b/arch/powerpc/platforms/pseries/ras.c @@ -23,11 +23,6 @@ static DEFINE_SPINLOCK(ras_log_buf_lock); static int ras_check_exception_token; -static void mce_process_errlog_event(struct irq_work *work); -static struct irq_work mce_errlog_process_work = { - .func = mce_process_errlog_event, -}; - #define EPOW_SENSOR_TOKEN 9 #define EPOW_SENSOR_INDEX 0 @@ -60,11 +55,17 @@ struct pseries_mc_errorlog { * XX 2: Reserved. * XXX 3: Type of UE error. * - * For error_type != MC_ERROR_TYPE_UE + * For error_type == MC_ERROR_TYPE_SLB/ERAT/TLB * XXXXXXXX * X 1: Effective address provided. * XXXXX 5: Reserved. * XX 2: Type of SLB/ERAT/TLB error. + * + * For error_type == MC_ERROR_TYPE_CTRL_MEM_ACCESS + * XXXXXXXX + * X 1: Error causing address provided. + * XXX 3: Type of error. + * XXXX 4: Reserved. */ u8 sub_err_type; u8 reserved_1[6]; @@ -80,6 +81,7 @@ struct pseries_mc_errorlog { #define MC_ERROR_TYPE_TLB 0x04 #define MC_ERROR_TYPE_D_CACHE 0x05 #define MC_ERROR_TYPE_I_CACHE 0x07 +#define MC_ERROR_TYPE_CTRL_MEM_ACCESS 0x08 /* RTAS pseries MCE error sub types */ #define MC_ERROR_UE_INDETERMINATE 0 @@ -90,6 +92,7 @@ struct pseries_mc_errorlog { #define UE_EFFECTIVE_ADDR_PROVIDED 0x40 #define UE_LOGICAL_ADDR_PROVIDED 0x20 +#define MC_EFFECTIVE_ADDR_PROVIDED 0x80 #define MC_ERROR_SLB_PARITY 0 #define MC_ERROR_SLB_MULTIHIT 1 @@ -103,6 +106,9 @@ struct pseries_mc_errorlog { #define MC_ERROR_TLB_MULTIHIT 2 #define MC_ERROR_TLB_INDETERMINATE 3 +#define MC_ERROR_CTRL_MEM_ACCESS_PTABLE_WALK 0 +#define MC_ERROR_CTRL_MEM_ACCESS_OP_ACCESS 1 + static inline u8 rtas_mc_error_sub_type(const struct pseries_mc_errorlog *mlog) { switch (mlog->error_type) { @@ -112,6 +118,8 @@ static inline u8 rtas_mc_error_sub_type(const struct pseries_mc_errorlog *mlog) case MC_ERROR_TYPE_ERAT: case MC_ERROR_TYPE_TLB: return (mlog->sub_err_type & 0x03); + case MC_ERROR_TYPE_CTRL_MEM_ACCESS: + return (mlog->sub_err_type & 0x70) >> 4; default: return 0; } @@ -526,6 +534,7 @@ static int mce_handle_err_realmode(int disposition, u8 error_type) disposition = RTAS_DISP_FULLY_RECOVERED; break; case MC_ERROR_TYPE_SLB: +#ifdef CONFIG_PPC_64S_HASH_MMU /* * Store the old slb content in paca before flushing. * Print this when we go to virtual mode. @@ -538,6 +547,7 @@ static int mce_handle_err_realmode(int disposition, u8 error_type) slb_save_contents(local_paca->mce_faulty_slbs); flush_and_reload_slb(); disposition = RTAS_DISP_FULLY_RECOVERED; +#endif break; default: break; @@ -656,7 +666,7 @@ static int mce_handle_err_virtmode(struct pt_regs *regs, mce_err.u.slb_error_type = MCE_SLB_ERROR_INDETERMINATE; break; } - if (mce_log->sub_err_type & 0x80) + if (mce_log->sub_err_type & MC_EFFECTIVE_ADDR_PROVIDED) eaddr = be64_to_cpu(mce_log->effective_address); break; case MC_ERROR_TYPE_ERAT: @@ -673,7 +683,7 @@ static int mce_handle_err_virtmode(struct pt_regs *regs, mce_err.u.erat_error_type = MCE_ERAT_ERROR_INDETERMINATE; break; } - if (mce_log->sub_err_type & 0x80) + if (mce_log->sub_err_type & MC_EFFECTIVE_ADDR_PROVIDED) eaddr = be64_to_cpu(mce_log->effective_address); break; case MC_ERROR_TYPE_TLB: @@ -690,7 +700,7 @@ static int mce_handle_err_virtmode(struct pt_regs *regs, mce_err.u.tlb_error_type = MCE_TLB_ERROR_INDETERMINATE; break; } - if (mce_log->sub_err_type & 0x80) + if (mce_log->sub_err_type & MC_EFFECTIVE_ADDR_PROVIDED) eaddr = be64_to_cpu(mce_log->effective_address); break; case MC_ERROR_TYPE_D_CACHE: @@ -699,6 +709,21 @@ static int mce_handle_err_virtmode(struct pt_regs *regs, case MC_ERROR_TYPE_I_CACHE: mce_err.error_type = MCE_ERROR_TYPE_ICACHE; break; + case MC_ERROR_TYPE_CTRL_MEM_ACCESS: + mce_err.error_type = MCE_ERROR_TYPE_RA; + switch (err_sub_type) { + case MC_ERROR_CTRL_MEM_ACCESS_PTABLE_WALK: + mce_err.u.ra_error_type = + MCE_RA_ERROR_PAGE_TABLE_WALK_LOAD_STORE_FOREIGN; + break; + case MC_ERROR_CTRL_MEM_ACCESS_OP_ACCESS: + mce_err.u.ra_error_type = + MCE_RA_ERROR_LOAD_STORE_FOREIGN; + break; + } + if (mce_log->sub_err_type & MC_EFFECTIVE_ADDR_PROVIDED) + eaddr = be64_to_cpu(mce_log->effective_address); + break; case MC_ERROR_TYPE_UNKNOWN: default: mce_err.error_type = MCE_ERROR_TYPE_UNKNOWN; @@ -715,7 +740,6 @@ static int mce_handle_error(struct pt_regs *regs, struct rtas_error_log *errp) struct pseries_errorlog *pseries_log; struct pseries_mc_errorlog *mce_log = NULL; int disposition = rtas_error_disposition(errp); - unsigned long msr; u8 error_type; if (!rtas_error_extended(errp)) @@ -729,40 +753,16 @@ static int mce_handle_error(struct pt_regs *regs, struct rtas_error_log *errp) error_type = mce_log->error_type; disposition = mce_handle_err_realmode(disposition, error_type); - - /* - * Enable translation as we will be accessing per-cpu variables - * in save_mce_event() which may fall outside RMO region, also - * leave it enabled because subsequently we will be queuing work - * to workqueues where again per-cpu variables accessed, besides - * fwnmi_release_errinfo() crashes when called in realmode on - * pseries. - * Note: All the realmode handling like flushing SLB entries for - * SLB multihit is done by now. - */ out: - msr = mfmsr(); - mtmsr(msr | MSR_IR | MSR_DR); - disposition = mce_handle_err_virtmode(regs, errp, mce_log, disposition); - - /* - * Queue irq work to log this rtas event later. - * irq_work_queue uses per-cpu variables, so do this in virt - * mode as well. - */ - irq_work_queue(&mce_errlog_process_work); - - mtmsr(msr); - return disposition; } /* * Process MCE rtas errlog event. */ -static void mce_process_errlog_event(struct irq_work *work) +void pSeries_machine_check_log_err(void) { struct rtas_error_log *err; diff --git a/arch/powerpc/platforms/pseries/reconfig.c b/arch/powerpc/platforms/pseries/reconfig.c index 7f7369fec46b..599bd2c78514 100644 --- a/arch/powerpc/platforms/pseries/reconfig.c +++ b/arch/powerpc/platforms/pseries/reconfig.c @@ -10,10 +10,10 @@ #include <linux/kernel.h> #include <linux/notifier.h> #include <linux/proc_fs.h> +#include <linux/security.h> #include <linux/slab.h> #include <linux/of.h> -#include <asm/prom.h> #include <asm/machdep.h> #include <linux/uaccess.h> #include <asm/mmu.h> @@ -362,6 +362,10 @@ static ssize_t ofdt_write(struct file *file, const char __user *buf, size_t coun char *kbuf; char *tmp; + rv = security_locked_down(LOCKDOWN_DEVICE_TREE); + if (rv) + return rv; + kbuf = memdup_user_nul(buf, count); if (IS_ERR(kbuf)) return PTR_ERR(kbuf); diff --git a/arch/powerpc/platforms/pseries/rng.c b/arch/powerpc/platforms/pseries/rng.c index 6268545947b8..6ddfdeaace9e 100644 --- a/arch/powerpc/platforms/pseries/rng.c +++ b/arch/powerpc/platforms/pseries/rng.c @@ -10,6 +10,7 @@ #include <asm/archrandom.h> #include <asm/machdep.h> #include <asm/plpar_wrappers.h> +#include "pseries.h" static int pseries_get_random_long(unsigned long *v) @@ -24,19 +25,13 @@ static int pseries_get_random_long(unsigned long *v) return 0; } -static __init int rng_init(void) +void __init pseries_rng_init(void) { struct device_node *dn; dn = of_find_compatible_node(NULL, NULL, "ibm,random"); if (!dn) - return -ENODEV; - - pr_info("Registering arch random hook.\n"); - + return; ppc_md.get_random_seed = pseries_get_random_long; - of_node_put(dn); - return 0; } -machine_subsys_initcall(pseries, rng_init); diff --git a/arch/powerpc/platforms/pseries/rtas-fadump.c b/arch/powerpc/platforms/pseries/rtas-fadump.c index f8f73b47b107..b5853e9fcc3c 100644 --- a/arch/powerpc/platforms/pseries/rtas-fadump.c +++ b/arch/powerpc/platforms/pseries/rtas-fadump.c @@ -13,9 +13,10 @@ #include <linux/delay.h> #include <linux/seq_file.h> #include <linux/crash_dump.h> +#include <linux/of.h> +#include <linux/of_fdt.h> #include <asm/page.h> -#include <asm/prom.h> #include <asm/rtas.h> #include <asm/fadump.h> #include <asm/fadump-internal.h> @@ -39,7 +40,7 @@ static void rtas_fadump_update_config(struct fw_dump *fadump_conf, * This function is called in the capture kernel to get configuration details * setup in the first kernel and passed to the f/w. */ -static void rtas_fadump_get_config(struct fw_dump *fadump_conf, +static void __init rtas_fadump_get_config(struct fw_dump *fadump_conf, const struct rtas_fadump_mem_struct *fdm) { fadump_conf->boot_mem_addr[0] = @@ -108,6 +109,12 @@ static u64 rtas_fadump_init_mem_struct(struct fw_dump *fadump_conf) fdm.hpte_region.destination_address = cpu_to_be64(addr); addr += fadump_conf->hpte_region_size; + /* + * Align boot memory area destination address to page boundary to + * be able to mmap read this area in the vmcore. + */ + addr = PAGE_ALIGN(addr); + /* RMA region section */ fdm.rmr_region.request_flag = cpu_to_be32(RTAS_FADUMP_REQUEST_FLAG); fdm.rmr_region.source_data_type = @@ -247,7 +254,7 @@ static inline int rtas_fadump_gpr_index(u64 id) return i; } -static void rtas_fadump_set_regval(struct pt_regs *regs, u64 reg_id, u64 reg_val) +static void __init rtas_fadump_set_regval(struct pt_regs *regs, u64 reg_id, u64 reg_val) { int i; @@ -272,7 +279,7 @@ static void rtas_fadump_set_regval(struct pt_regs *regs, u64 reg_id, u64 reg_val regs->dsisr = (unsigned long)reg_val; } -static struct rtas_fadump_reg_entry* +static struct rtas_fadump_reg_entry* __init rtas_fadump_read_regs(struct rtas_fadump_reg_entry *reg_entry, struct pt_regs *regs) { @@ -351,7 +358,7 @@ static int __init rtas_fadump_build_cpu_notes(struct fw_dump *fadump_conf) /* Lower 4 bytes of reg_value contains logical cpu id */ cpu = (be64_to_cpu(reg_entry->reg_value) & RTAS_FADUMP_CPU_ID_MASK); - if (fdh && !cpumask_test_cpu(cpu, &fdh->online_mask)) { + if (fdh && !cpumask_test_cpu(cpu, &fdh->cpu_mask)) { RTAS_FADUMP_SKIP_TO_NEXT_CPU(reg_entry); continue; } @@ -462,10 +469,10 @@ static void rtas_fadump_region_show(struct fw_dump *fadump_conf, be64_to_cpu(fdm_ptr->rmr_region.source_len), be64_to_cpu(fdm_ptr->rmr_region.bytes_dumped)); - /* Dump is active. Show reserved area start address. */ + /* Dump is active. Show preserved area start address. */ if (fdm_active) { - seq_printf(m, "\nMemory above %#016lx is reserved for saving crash dump\n", - fadump_conf->reserve_dump_area_start); + seq_printf(m, "\nMemory above %#016llx is reserved for saving crash dump\n", + fadump_conf->boot_mem_top); } } diff --git a/arch/powerpc/platforms/pseries/scanlog.c b/arch/powerpc/platforms/pseries/scanlog.c deleted file mode 100644 index 2879c4f0ceb7..000000000000 --- a/arch/powerpc/platforms/pseries/scanlog.c +++ /dev/null @@ -1,195 +0,0 @@ -// SPDX-License-Identifier: GPL-2.0-or-later -/* - * c 2001 PPC 64 Team, IBM Corp - * - * scan-log-data driver for PPC64 Todd Inglett <tinglett@vnet.ibm.com> - * - * When ppc64 hardware fails the service processor dumps internal state - * of the system. After a reboot the operating system can access a dump - * of this data using this driver. A dump exists if the device-tree - * /chosen/ibm,scan-log-data property exists. - * - * This driver exports /proc/powerpc/scan-log-dump which can be read. - * The driver supports only sequential reads. - * - * The driver looks at a write to the driver for the single word "reset". - * If given, the driver will reset the scanlog so the platform can free it. - */ - -#include <linux/module.h> -#include <linux/types.h> -#include <linux/errno.h> -#include <linux/proc_fs.h> -#include <linux/init.h> -#include <linux/delay.h> -#include <linux/slab.h> -#include <linux/uaccess.h> -#include <asm/rtas.h> -#include <asm/prom.h> - -#define MODULE_VERS "1.0" -#define MODULE_NAME "scanlog" - -/* Status returns from ibm,scan-log-dump */ -#define SCANLOG_COMPLETE 0 -#define SCANLOG_HWERROR -1 -#define SCANLOG_CONTINUE 1 - - -static unsigned int ibm_scan_log_dump; /* RTAS token */ -static unsigned int *scanlog_buffer; /* The data buffer */ - -static ssize_t scanlog_read(struct file *file, char __user *buf, - size_t count, loff_t *ppos) -{ - unsigned int *data = scanlog_buffer; - int status; - unsigned long len, off; - unsigned int wait_time; - - if (count > RTAS_DATA_BUF_SIZE) - count = RTAS_DATA_BUF_SIZE; - - if (count < 1024) { - /* This is the min supported by this RTAS call. Rather - * than do all the buffering we insist the user code handle - * larger reads. As long as cp works... :) - */ - printk(KERN_ERR "scanlog: cannot perform a small read (%ld)\n", count); - return -EINVAL; - } - - if (!access_ok(buf, count)) - return -EFAULT; - - for (;;) { - wait_time = 500; /* default wait if no data */ - spin_lock(&rtas_data_buf_lock); - memcpy(rtas_data_buf, data, RTAS_DATA_BUF_SIZE); - status = rtas_call(ibm_scan_log_dump, 2, 1, NULL, - (u32) __pa(rtas_data_buf), (u32) count); - memcpy(data, rtas_data_buf, RTAS_DATA_BUF_SIZE); - spin_unlock(&rtas_data_buf_lock); - - pr_debug("scanlog: status=%d, data[0]=%x, data[1]=%x, " \ - "data[2]=%x\n", status, data[0], data[1], data[2]); - switch (status) { - case SCANLOG_COMPLETE: - pr_debug("scanlog: hit eof\n"); - return 0; - case SCANLOG_HWERROR: - pr_debug("scanlog: hardware error reading data\n"); - return -EIO; - case SCANLOG_CONTINUE: - /* We may or may not have data yet */ - len = data[1]; - off = data[2]; - if (len > 0) { - if (copy_to_user(buf, ((char *)data)+off, len)) - return -EFAULT; - return len; - } - /* Break to sleep default time */ - break; - default: - /* Assume extended busy */ - wait_time = rtas_busy_delay_time(status); - if (!wait_time) { - printk(KERN_ERR "scanlog: unknown error " \ - "from rtas: %d\n", status); - return -EIO; - } - } - /* Apparently no data yet. Wait and try again. */ - msleep_interruptible(wait_time); - } - /*NOTREACHED*/ -} - -static ssize_t scanlog_write(struct file * file, const char __user * buf, - size_t count, loff_t *ppos) -{ - char stkbuf[20]; - int status; - - if (count > 19) count = 19; - if (copy_from_user (stkbuf, buf, count)) { - return -EFAULT; - } - stkbuf[count] = 0; - - if (buf) { - if (strncmp(stkbuf, "reset", 5) == 0) { - pr_debug("scanlog: reset scanlog\n"); - status = rtas_call(ibm_scan_log_dump, 2, 1, NULL, 0, 0); - pr_debug("scanlog: rtas returns %d\n", status); - } - } - return count; -} - -static int scanlog_open(struct inode * inode, struct file * file) -{ - unsigned int *data = scanlog_buffer; - - if (data[0] != 0) { - /* This imperfect test stops a second copy of the - * data (or a reset while data is being copied) - */ - return -EBUSY; - } - - data[0] = 0; /* re-init so we restart the scan */ - - return 0; -} - -static int scanlog_release(struct inode * inode, struct file * file) -{ - unsigned int *data = scanlog_buffer; - - data[0] = 0; - return 0; -} - -static const struct proc_ops scanlog_proc_ops = { - .proc_read = scanlog_read, - .proc_write = scanlog_write, - .proc_open = scanlog_open, - .proc_release = scanlog_release, - .proc_lseek = noop_llseek, -}; - -static int __init scanlog_init(void) -{ - struct proc_dir_entry *ent; - int err = -ENOMEM; - - ibm_scan_log_dump = rtas_token("ibm,scan-log-dump"); - if (ibm_scan_log_dump == RTAS_UNKNOWN_SERVICE) - return -ENODEV; - - /* Ideally we could allocate a buffer < 4G */ - scanlog_buffer = kzalloc(RTAS_DATA_BUF_SIZE, GFP_KERNEL); - if (!scanlog_buffer) - goto err; - - ent = proc_create("powerpc/rtas/scan-log-dump", 0400, NULL, - &scanlog_proc_ops); - if (!ent) - goto err; - return 0; -err: - kfree(scanlog_buffer); - return err; -} - -static void __exit scanlog_cleanup(void) -{ - remove_proc_entry("powerpc/rtas/scan-log-dump", NULL); - kfree(scanlog_buffer); -} - -module_init(scanlog_init); -module_exit(scanlog_cleanup); -MODULE_LICENSE("GPL"); diff --git a/arch/powerpc/platforms/pseries/setup.c b/arch/powerpc/platforms/pseries/setup.c index 8a62af5b9c24..8ef3270515a9 100644 --- a/arch/powerpc/platforms/pseries/setup.c +++ b/arch/powerpc/platforms/pseries/setup.c @@ -14,6 +14,7 @@ #include <linux/cpu.h> #include <linux/errno.h> +#include <linux/platform_device.h> #include <linux/sched.h> #include <linux/kernel.h> #include <linux/mm.h> @@ -36,14 +37,15 @@ #include <linux/seq_file.h> #include <linux/root_dev.h> #include <linux/of.h> +#include <linux/of_irq.h> #include <linux/of_pci.h> #include <linux/memblock.h> #include <linux/swiotlb.h> +#include <linux/seq_buf.h> #include <asm/mmu.h> #include <asm/processor.h> #include <asm/io.h> -#include <asm/prom.h> #include <asm/rtas.h> #include <asm/pci-bridge.h> #include <asm/iommu.h> @@ -72,12 +74,27 @@ #include <asm/svm.h> #include <asm/dtl.h> #include <asm/hvconsole.h> +#include <asm/setup.h> #include "pseries.h" DEFINE_STATIC_KEY_FALSE(shared_processor); EXPORT_SYMBOL(shared_processor); +#ifdef CONFIG_PARAVIRT_TIME_ACCOUNTING +struct static_key paravirt_steal_enabled; +struct static_key paravirt_steal_rq_enabled; + +static bool steal_acc = true; +static int __init parse_no_stealacc(char *arg) +{ + steal_acc = false; + return 0; +} + +early_param("no-steal-acc", parse_no_stealacc); +#endif + int CMO_PrPSP = -1; int CMO_SecPSP = -1; unsigned long CMO_PageSize = (ASM_CONST(1) << IOMMU_PAGE_SHIFT_4K); @@ -112,7 +129,7 @@ static void __init fwnmi_init(void) u8 *mce_data_buf; unsigned int i; int nr_cpus = num_possible_cpus(); -#ifdef CONFIG_PPC_BOOK3S_64 +#ifdef CONFIG_PPC_64S_HASH_MMU struct slb_entry *slb_ptr; size_t size; #endif @@ -152,7 +169,7 @@ static void __init fwnmi_init(void) (RTAS_ERROR_LOG_MAX * i); } -#ifdef CONFIG_PPC_BOOK3S_64 +#ifdef CONFIG_PPC_64S_HASH_MMU if (!radix_enabled()) { /* Allocate per cpu area to save old slb contents during MCE */ size = sizeof(struct slb_entry) * mmu_slb_size * nr_cpus; @@ -169,6 +186,18 @@ static void __init fwnmi_init(void) #endif } +/* + * Affix a device for the first timer to the platform bus if + * we have firmware support for the H_WATCHDOG hypercall. + */ +static __init int pseries_wdt_init(void) +{ + if (firmware_has_feature(FW_FEATURE_WATCHDOG)) + platform_device_register_simple("pseries-wdt", 0, NULL, 0); + return 0; +} +machine_subsys_initcall(pseries, pseries_wdt_init); + static void pseries_8259_cascade(struct irq_desc *desc) { struct irq_chip *chip = irq_desc_get_chip(desc); @@ -353,6 +382,14 @@ static void pseries_lpar_idle(void) pseries_idle_epilog(); } +static bool pseries_reloc_on_exception_enabled; + +bool pseries_reloc_on_exception(void) +{ + return pseries_reloc_on_exception_enabled; +} +EXPORT_SYMBOL_GPL(pseries_reloc_on_exception); + /* * Enable relocation on during exceptions. This has partition wide scope and * may take a while to complete, if it takes longer than one second we will @@ -377,6 +414,7 @@ bool pseries_enable_reloc_on_exc(void) " on exceptions: %ld\n", rc); return false; } + pseries_reloc_on_exception_enabled = true; return true; } @@ -404,22 +442,14 @@ void pseries_disable_reloc_on_exc(void) break; mdelay(get_longbusy_msecs(rc)); } - if (rc != H_SUCCESS) + if (rc == H_SUCCESS) + pseries_reloc_on_exception_enabled = false; + else pr_warn("Warning: Failed to disable relocation on exceptions: %ld\n", rc); } EXPORT_SYMBOL(pseries_disable_reloc_on_exc); -#ifdef CONFIG_KEXEC_CORE -static void pSeries_machine_kexec(struct kimage *image) -{ - if (firmware_has_feature(FW_FEATURE_SET_MODE)) - pseries_disable_reloc_on_exc(); - - default_machine_kexec(image); -} -#endif - #ifdef __LITTLE_ENDIAN__ void pseries_big_endian_exceptions(void) { @@ -447,7 +477,7 @@ void pseries_big_endian_exceptions(void) panic("Could not enable big endian exceptions"); } -void pseries_little_endian_exceptions(void) +void __init pseries_little_endian_exceptions(void) { long rc; @@ -647,7 +677,7 @@ static resource_size_t pseries_get_iov_fw_value(struct pci_dev *dev, int resno, */ num_res = of_read_number(&indexes[NUM_RES_PROPERTY], 1); if (resno >= num_res) - return 0; /* or an errror */ + return 0; /* or an error */ i = START_OF_ENTRIES + NEXT_ENTRY * resno; switch (value) { @@ -751,7 +781,7 @@ static void pseries_pci_fixup_iov_resources(struct pci_dev *pdev) if (!pdev->is_physfn) return; - /*Firmware must support open sriov otherwise dont configure*/ + /*Firmware must support open sriov otherwise don't configure*/ indexes = of_get_property(dn, "ibm,open-sriov-vf-bar-info", NULL); if (indexes) of_pci_parse_iov_addrs(pdev, indexes); @@ -801,7 +831,8 @@ static void __init pSeries_setup_arch(void) fwnmi_init(); pseries_setup_security_mitigations(); - pseries_lpar_read_hblkrm_characteristics(); + if (!radix_enabled()) + pseries_lpar_read_hblkrm_characteristics(); /* By default, only probe PCI (can be overridden by rtas_pci) */ pci_add_flags(PCI_PROBE_ONLY); @@ -818,6 +849,11 @@ static void __init pSeries_setup_arch(void) if (lppaca_shared_proc(get_lppaca())) { static_branch_enable(&shared_processor); pv_spinlocks_init(); +#ifdef CONFIG_PARAVIRT_TIME_ACCOUNTING + static_key_slow_inc(¶virt_steal_enabled); + if (steal_acc) + static_key_slow_inc(¶virt_steal_rq_enabled); +#endif } ppc_md.power_save = pseries_lpar_idle; @@ -836,9 +872,7 @@ static void __init pSeries_setup_arch(void) } ppc_md.pcibios_root_bridge_prepare = pseries_root_bridge_prepare; - - if (swiotlb_force == SWIOTLB_FORCE) - ppc_swiotlb_enable = 1; + pseries_rng_init(); } static void pseries_panic(char *str) @@ -905,7 +939,7 @@ void pSeries_coalesce_init(void) * fw_cmo_feature_init - FW_FEATURE_CMO is not stored in ibm,hypertas-functions, * handle that here. (Stolen from parse_system_parameter_string) */ -static void pSeries_cmo_feature_init(void) +static void __init pSeries_cmo_feature_init(void) { char *ptr, *key, *value, *end; int call_status; @@ -978,6 +1012,33 @@ static void pSeries_cmo_feature_init(void) pr_debug(" <- fw_cmo_feature_init()\n"); } +static void __init pseries_add_hw_description(void) +{ + struct device_node *dn; + const char *s; + + dn = of_find_node_by_path("/openprom"); + if (dn) { + if (of_property_read_string(dn, "model", &s) == 0) + seq_buf_printf(&ppc_hw_desc, "of:%s ", s); + + of_node_put(dn); + } + + dn = of_find_node_by_path("/hypervisor"); + if (dn) { + if (of_property_read_string(dn, "compatible", &s) == 0) + seq_buf_printf(&ppc_hw_desc, "hv:%s ", s); + + of_node_put(dn); + return; + } + + if (of_property_read_bool(of_root, "ibm,powervm-partition") || + of_property_read_bool(of_root, "ibm,fw-net-version")) + seq_buf_printf(&ppc_hw_desc, "hv:phyp "); +} + /* * Early initialization. Relocation is on but do not reference unbolted pages */ @@ -985,6 +1046,8 @@ static void __init pseries_init(void) { pr_debug(" -> pseries_init()\n"); + pseries_add_hw_description(); + #ifdef CONFIG_HVC_CONSOLE if (firmware_has_feature(FW_FEATURE_LPAR)) hvc_vio_init_early(); @@ -1084,8 +1147,9 @@ define_machine(pseries) { .system_reset_exception = pSeries_system_reset_exception, .machine_check_early = pseries_machine_check_realmode, .machine_check_exception = pSeries_machine_check_exception, + .machine_check_log_err = pSeries_machine_check_log_err, #ifdef CONFIG_KEXEC_CORE - .machine_kexec = pSeries_machine_kexec, + .machine_kexec = pseries_machine_kexec, .kexec_cpu_down = pseries_kexec_cpu_down, #endif #ifdef CONFIG_MEMORY_HOTPLUG diff --git a/arch/powerpc/platforms/pseries/smp.c b/arch/powerpc/platforms/pseries/smp.c index f47429323eee..fd2174edfa1d 100644 --- a/arch/powerpc/platforms/pseries/smp.c +++ b/arch/powerpc/platforms/pseries/smp.c @@ -27,7 +27,6 @@ #include <asm/irq.h> #include <asm/page.h> #include <asm/io.h> -#include <asm/prom.h> #include <asm/smp.h> #include <asm/paca.h> #include <asm/machdep.h> diff --git a/arch/powerpc/platforms/pseries/svm.c b/arch/powerpc/platforms/pseries/svm.c index c5228f4969eb..3b4045d508ec 100644 --- a/arch/powerpc/platforms/pseries/svm.c +++ b/arch/powerpc/platforms/pseries/svm.c @@ -28,7 +28,7 @@ static int __init init_svm(void) * need to use the SWIOTLB buffer for DMA even if dma_capable() says * otherwise. */ - swiotlb_force = SWIOTLB_FORCE; + ppc_swiotlb_flags |= SWIOTLB_ANY | SWIOTLB_FORCE; /* Share the SWIOTLB buffer with the host. */ swiotlb_update_mem_attributes(); @@ -37,30 +37,6 @@ static int __init init_svm(void) } machine_early_initcall(pseries, init_svm); -/* - * Initialize SWIOTLB. Essentially the same as swiotlb_init(), except that it - * can allocate the buffer anywhere in memory. Since the hypervisor doesn't have - * any addressing limitation, we don't need to allocate it in low addresses. - */ -void __init svm_swiotlb_init(void) -{ - unsigned char *vstart; - unsigned long bytes, io_tlb_nslabs; - - io_tlb_nslabs = (swiotlb_size_or_default() >> IO_TLB_SHIFT); - io_tlb_nslabs = ALIGN(io_tlb_nslabs, IO_TLB_SEGSIZE); - - bytes = io_tlb_nslabs << IO_TLB_SHIFT; - - vstart = memblock_alloc(PAGE_ALIGN(bytes), PAGE_SIZE); - if (vstart && !swiotlb_init_with_tbl(vstart, io_tlb_nslabs, false)) - return; - - - memblock_free(vstart, PAGE_ALIGN(io_tlb_nslabs << IO_TLB_SHIFT)); - panic("SVM: Cannot allocate SWIOTLB buffer"); -} - int set_memory_encrypted(unsigned long addr, int numpages) { if (!cc_platform_has(CC_ATTR_MEM_ENCRYPT)) diff --git a/arch/powerpc/platforms/pseries/vas-sysfs.c b/arch/powerpc/platforms/pseries/vas-sysfs.c new file mode 100644 index 000000000000..f9f682724e77 --- /dev/null +++ b/arch/powerpc/platforms/pseries/vas-sysfs.c @@ -0,0 +1,281 @@ +// SPDX-License-Identifier: GPL-2.0-or-later +/* + * Copyright 2022-23 IBM Corp. + */ + +#define pr_fmt(fmt) "vas: " fmt + +#include <linux/module.h> +#include <linux/kernel.h> +#include <linux/miscdevice.h> +#include <linux/kobject.h> +#include <linux/slab.h> +#include <linux/mm.h> + +#include "vas.h" + +#ifdef CONFIG_SYSFS +static struct kobject *pseries_vas_kobj; +static struct kobject *gzip_caps_kobj; + +struct vas_caps_entry { + struct kobject kobj; + struct vas_cop_feat_caps *caps; +}; + +#define to_caps_entry(entry) container_of(entry, struct vas_caps_entry, kobj) + +/* + * This function is used to get the notification from the drmgr when + * QoS credits are changed. + */ +static ssize_t update_total_credits_store(struct vas_cop_feat_caps *caps, + const char *buf, size_t count) +{ + int err; + u16 creds; + + err = kstrtou16(buf, 0, &creds); + /* + * The user space interface from the management console + * notifies OS with the new QoS credits and then the + * hypervisor. So OS has to use this new credits value + * and reconfigure VAS windows (close or reopen depends + * on the credits available) instead of depending on VAS + * QoS capabilities from the hypervisor. + */ + if (!err) + err = vas_reconfig_capabilties(caps->win_type, creds); + + if (err) + return -EINVAL; + + pr_info("Set QoS total credits %u\n", creds); + + return count; +} + +#define sysfs_caps_entry_read(_name) \ +static ssize_t _name##_show(struct vas_cop_feat_caps *caps, char *buf) \ +{ \ + return sprintf(buf, "%d\n", atomic_read(&caps->_name)); \ +} + +struct vas_sysfs_entry { + struct attribute attr; + ssize_t (*show)(struct vas_cop_feat_caps *, char *); + ssize_t (*store)(struct vas_cop_feat_caps *, const char *, size_t); +}; + +#define VAS_ATTR_RO(_name) \ + sysfs_caps_entry_read(_name); \ + static struct vas_sysfs_entry _name##_attribute = __ATTR(_name, \ + 0444, _name##_show, NULL); + +/* + * Create sysfs interface: + * /sys/devices/virtual/misc/vas/vas0/gzip/default_capabilities + * This directory contains the following VAS GZIP capabilities + * for the default credit type. + * /sys/devices/virtual/misc/vas/vas0/gzip/default_capabilities/nr_total_credits + * Total number of default credits assigned to the LPAR which + * can be changed with DLPAR operation. + * /sys/devices/virtual/misc/vas/vas0/gzip/default_capabilities/nr_used_credits + * Number of credits used by the user space. One credit will + * be assigned for each window open. + * + * /sys/devices/virtual/misc/vas/vas0/gzip/qos_capabilities + * This directory contains the following VAS GZIP capabilities + * for the Quality of Service (QoS) credit type. + * /sys/devices/virtual/misc/vas/vas0/gzip/qos_capabilities/nr_total_credits + * Total number of QoS credits assigned to the LPAR. The user + * has to define this value using HMC interface. It can be + * changed dynamically by the user. + * /sys/devices/virtual/misc/vas/vas0/gzip/qos_capabilities/nr_used_credits + * Number of credits used by the user space. + * /sys/devices/virtual/misc/vas/vas0/gzip/qos_capabilities/update_total_credits + * Update total QoS credits dynamically + */ + +VAS_ATTR_RO(nr_total_credits); +VAS_ATTR_RO(nr_used_credits); + +static struct vas_sysfs_entry update_total_credits_attribute = + __ATTR(update_total_credits, 0200, NULL, update_total_credits_store); + +static struct attribute *vas_def_capab_attrs[] = { + &nr_total_credits_attribute.attr, + &nr_used_credits_attribute.attr, + NULL, +}; +ATTRIBUTE_GROUPS(vas_def_capab); + +static struct attribute *vas_qos_capab_attrs[] = { + &nr_total_credits_attribute.attr, + &nr_used_credits_attribute.attr, + &update_total_credits_attribute.attr, + NULL, +}; +ATTRIBUTE_GROUPS(vas_qos_capab); + +static ssize_t vas_type_show(struct kobject *kobj, struct attribute *attr, + char *buf) +{ + struct vas_caps_entry *centry; + struct vas_cop_feat_caps *caps; + struct vas_sysfs_entry *entry; + + centry = to_caps_entry(kobj); + caps = centry->caps; + entry = container_of(attr, struct vas_sysfs_entry, attr); + + if (!entry->show) + return -EIO; + + return entry->show(caps, buf); +} + +static ssize_t vas_type_store(struct kobject *kobj, struct attribute *attr, + const char *buf, size_t count) +{ + struct vas_caps_entry *centry; + struct vas_cop_feat_caps *caps; + struct vas_sysfs_entry *entry; + + centry = to_caps_entry(kobj); + caps = centry->caps; + entry = container_of(attr, struct vas_sysfs_entry, attr); + if (!entry->store) + return -EIO; + + return entry->store(caps, buf, count); +} + +static void vas_type_release(struct kobject *kobj) +{ + struct vas_caps_entry *centry = to_caps_entry(kobj); + kfree(centry); +} + +static const struct sysfs_ops vas_sysfs_ops = { + .show = vas_type_show, + .store = vas_type_store, +}; + +static struct kobj_type vas_def_attr_type = { + .release = vas_type_release, + .sysfs_ops = &vas_sysfs_ops, + .default_groups = vas_def_capab_groups, +}; + +static struct kobj_type vas_qos_attr_type = { + .release = vas_type_release, + .sysfs_ops = &vas_sysfs_ops, + .default_groups = vas_qos_capab_groups, +}; + +static char *vas_caps_kobj_name(struct vas_caps_entry *centry, + struct kobject **kobj) +{ + struct vas_cop_feat_caps *caps = centry->caps; + + if (caps->descriptor == VAS_GZIP_QOS_CAPABILITIES) { + kobject_init(¢ry->kobj, &vas_qos_attr_type); + *kobj = gzip_caps_kobj; + return "qos_capabilities"; + } else if (caps->descriptor == VAS_GZIP_DEFAULT_CAPABILITIES) { + kobject_init(¢ry->kobj, &vas_def_attr_type); + *kobj = gzip_caps_kobj; + return "default_capabilities"; + } else + return "Unknown"; +} + +/* + * Add feature specific capability dir entry. + * Ex: VDefGzip or VQosGzip + */ +int sysfs_add_vas_caps(struct vas_cop_feat_caps *caps) +{ + struct vas_caps_entry *centry; + struct kobject *kobj = NULL; + int ret = 0; + char *name; + + centry = kzalloc(sizeof(*centry), GFP_KERNEL); + if (!centry) + return -ENOMEM; + + centry->caps = caps; + name = vas_caps_kobj_name(centry, &kobj); + + if (kobj) { + ret = kobject_add(¢ry->kobj, kobj, "%s", name); + + if (ret) { + pr_err("VAS: sysfs kobject add / event failed %d\n", + ret); + kobject_put(¢ry->kobj); + } + } + + return ret; +} + +static struct miscdevice vas_miscdev = { + .minor = MISC_DYNAMIC_MINOR, + .name = "vas", +}; + +/* + * Add VAS and VasCaps (overall capabilities) dir entries. + */ +int __init sysfs_pseries_vas_init(struct vas_all_caps *vas_caps) +{ + int ret; + + ret = misc_register(&vas_miscdev); + if (ret < 0) { + pr_err("%s: register vas misc device failed\n", __func__); + return ret; + } + + /* + * The hypervisor does not expose multiple VAS instances, but can + * see multiple VAS instances on PowerNV. So create 'vas0' directory + * on pseries. + */ + pseries_vas_kobj = kobject_create_and_add("vas0", + &vas_miscdev.this_device->kobj); + if (!pseries_vas_kobj) { + misc_deregister(&vas_miscdev); + pr_err("Failed to create VAS sysfs entry\n"); + return -ENOMEM; + } + + if ((vas_caps->feat_type & VAS_GZIP_QOS_FEAT_BIT) || + (vas_caps->feat_type & VAS_GZIP_DEF_FEAT_BIT)) { + gzip_caps_kobj = kobject_create_and_add("gzip", + pseries_vas_kobj); + if (!gzip_caps_kobj) { + pr_err("Failed to create VAS GZIP capability entry\n"); + kobject_put(pseries_vas_kobj); + misc_deregister(&vas_miscdev); + return -ENOMEM; + } + } + + return 0; +} + +#else +int sysfs_add_vas_caps(struct vas_cop_feat_caps *caps) +{ + return 0; +} + +int __init sysfs_pseries_vas_init(struct vas_all_caps *vas_caps) +{ + return 0; +} +#endif diff --git a/arch/powerpc/platforms/pseries/vas.c b/arch/powerpc/platforms/pseries/vas.c index b043e3936d21..4ad6e510d405 100644 --- a/arch/powerpc/platforms/pseries/vas.c +++ b/arch/powerpc/platforms/pseries/vas.c @@ -16,6 +16,7 @@ #include <asm/machdep.h> #include <asm/hvcall.h> #include <asm/plpar_wrappers.h> +#include <asm/firmware.h> #include <asm/vas.h> #include "vas.h" @@ -26,9 +27,11 @@ static struct vas_all_caps caps_all; static bool copypaste_feat; +static struct hv_vas_cop_feat_caps hv_cop_caps; static struct vas_caps vascaps[VAS_MAX_FEAT_TYPE]; static DEFINE_MUTEX(vas_pseries_mutex); +static bool migration_in_progress; static long hcall_return_busy_check(long rc) { @@ -107,7 +110,6 @@ static int h_deallocate_vas_window(u64 winid) static int h_modify_vas_window(struct pseries_vas_window *win) { long rc; - u32 lpid = mfspr(SPRN_PID); /* * AMR value is not supported in Linux VAS implementation. @@ -115,7 +117,7 @@ static int h_modify_vas_window(struct pseries_vas_window *win) */ do { rc = plpar_hcall_norets(H_MODIFY_VAS_WINDOW, - win->vas_win.winid, lpid, 0, + win->vas_win.winid, win->pid, 0, VAS_MOD_WIN_FLAGS, 0); rc = hcall_return_busy_check(rc); @@ -124,8 +126,8 @@ static int h_modify_vas_window(struct pseries_vas_window *win) if (rc == H_SUCCESS) return 0; - pr_err("H_MODIFY_VAS_WINDOW error: %ld, winid %u lpid %u\n", - rc, win->vas_win.winid, lpid); + pr_err("H_MODIFY_VAS_WINDOW error: %ld, winid %u pid %u\n", + rc, win->vas_win.winid, win->pid); return -EIO; } @@ -151,8 +153,15 @@ int h_query_vas_capabilities(const u64 hcall, u8 query_type, u64 result) if (rc == H_SUCCESS) return 0; - pr_err("HCALL(%llx) error %ld, query_type %u, result buffer 0x%llx\n", - hcall, rc, query_type, result); + /* H_FUNCTION means HV does not support VAS so don't print an error */ + if (rc != H_FUNCTION) { + pr_err("%s error %ld, query_type %u, result buffer 0x%llx\n", + (hcall == H_QUERY_VAS_CAPABILITIES) ? + "H_QUERY_VAS_CAPABILITIES" : + "H_QUERY_NX_CAPABILITIES", + rc, query_type, result); + } + return -EIO; } EXPORT_SYMBOL_GPL(h_query_vas_capabilities); @@ -191,17 +200,42 @@ static irqreturn_t pseries_vas_fault_thread_fn(int irq, void *data) struct vas_user_win_ref *tsk_ref; int rc; - rc = h_get_nx_fault(txwin->vas_win.winid, (u64)virt_to_phys(&crb)); - if (!rc) { - tsk_ref = &txwin->vas_win.task_ref; - vas_dump_crb(&crb); - vas_update_csb(&crb, tsk_ref); + while (atomic_read(&txwin->pending_faults)) { + rc = h_get_nx_fault(txwin->vas_win.winid, (u64)virt_to_phys(&crb)); + if (!rc) { + tsk_ref = &txwin->vas_win.task_ref; + vas_dump_crb(&crb); + vas_update_csb(&crb, tsk_ref); + } + atomic_dec(&txwin->pending_faults); } return IRQ_HANDLED; } /* + * irq_default_primary_handler() can be used only with IRQF_ONESHOT + * which disables IRQ before executing the thread handler and enables + * it after. But this disabling interrupt sets the VAS IRQ OFF + * state in the hypervisor. If the NX generates fault interrupt + * during this window, the hypervisor will not deliver this + * interrupt to the LPAR. So use VAS specific IRQ handler instead + * of calling the default primary handler. + */ +static irqreturn_t pseries_vas_irq_handler(int irq, void *data) +{ + struct pseries_vas_window *txwin = data; + + /* + * The thread hanlder will process this interrupt if it is + * already running. + */ + atomic_inc(&txwin->pending_faults); + + return IRQ_WAKE_THREAD; +} + +/* * Allocate window and setup IRQ mapping. */ static int allocate_setup_window(struct pseries_vas_window *txwin, @@ -231,8 +265,9 @@ static int allocate_setup_window(struct pseries_vas_window *txwin, goto out_irq; } - rc = request_threaded_irq(txwin->fault_virq, NULL, - pseries_vas_fault_thread_fn, IRQF_ONESHOT, + rc = request_threaded_irq(txwin->fault_virq, + pseries_vas_irq_handler, + pseries_vas_fault_thread_fn, 0, txwin->name, txwin); if (rc) { pr_err("VAS-Window[%d]: Request IRQ(%u) failed with %d\n", @@ -303,8 +338,8 @@ static struct vas_window *vas_allocate_window(int vas_id, u64 flags, cop_feat_caps = &caps->caps; - if (atomic_inc_return(&cop_feat_caps->used_lpar_creds) > - atomic_read(&cop_feat_caps->target_lpar_creds)) { + if (atomic_inc_return(&cop_feat_caps->nr_used_credits) > + atomic_read(&cop_feat_caps->nr_total_credits)) { pr_err("Credits are not available to allocate window\n"); rc = -EINVAL; goto out; @@ -324,13 +359,15 @@ static struct vas_window *vas_allocate_window(int vas_id, u64 flags, * So no unpacking needs to be done. */ rc = plpar_hcall9(H_HOME_NODE_ASSOCIATIVITY, domain, - VPHN_FLAG_VCPU, smp_processor_id()); + VPHN_FLAG_VCPU, hard_smp_processor_id()); if (rc != H_SUCCESS) { pr_err("H_HOME_NODE_ASSOCIATIVITY error: %d\n", rc); goto out; } } + txwin->pid = mfspr(SPRN_PID); + /* * Allocate / Deallocate window hcalls and setup / free IRQs * have to be protected with mutex. @@ -347,7 +384,10 @@ static struct vas_window *vas_allocate_window(int vas_id, u64 flags, * same fault IRQ is not freed by the OS before. */ mutex_lock(&vas_pseries_mutex); - rc = allocate_setup_window(txwin, (u64 *)&domain[0], + if (migration_in_progress) + rc = -EBUSY; + else + rc = allocate_setup_window(txwin, (u64 *)&domain[0], cop_feat_caps->win_type); mutex_unlock(&vas_pseries_mutex); if (rc) @@ -362,13 +402,28 @@ static struct vas_window *vas_allocate_window(int vas_id, u64 flags, if (rc) goto out_free; - vas_user_win_add_mm_context(&txwin->vas_win.task_ref); txwin->win_type = cop_feat_caps->win_type; mutex_lock(&vas_pseries_mutex); - list_add(&txwin->win_list, &caps->list); + /* + * Possible to lose the acquired credit with DLPAR core + * removal after the window is opened. So if there are any + * closed windows (means with lost credits), do not give new + * window to user space. New windows will be opened only + * after the existing windows are reopened when credits are + * available. + */ + if (!caps->nr_close_wins) { + list_add(&txwin->win_list, &caps->list); + caps->nr_open_windows++; + mutex_unlock(&vas_pseries_mutex); + vas_user_win_add_mm_context(&txwin->vas_win.task_ref); + return &txwin->vas_win; + } mutex_unlock(&vas_pseries_mutex); - return &txwin->vas_win; + put_vas_user_win_ref(&txwin->vas_win.task_ref); + rc = -EBUSY; + pr_err("No credit is available to allocate window\n"); out_free: /* @@ -378,7 +433,7 @@ out_free: free_irq_setup(txwin); h_deallocate_vas_window(txwin->vas_win.winid); out: - atomic_dec(&cop_feat_caps->used_lpar_creds); + atomic_dec(&cop_feat_caps->nr_used_credits); kfree(txwin); return ERR_PTR(rc); } @@ -431,14 +486,25 @@ static int vas_deallocate_window(struct vas_window *vwin) caps = &vascaps[win->win_type].caps; mutex_lock(&vas_pseries_mutex); - rc = deallocate_free_window(win); - if (rc) { - mutex_unlock(&vas_pseries_mutex); - return rc; - } + /* + * VAS window is already closed in the hypervisor when + * lost the credit or with migration. So just remove the entry + * from the list, remove task references and free vas_window + * struct. + */ + if (!(win->vas_win.status & VAS_WIN_NO_CRED_CLOSE) && + !(win->vas_win.status & VAS_WIN_MIGRATE_CLOSE)) { + rc = deallocate_free_window(win); + if (rc) { + mutex_unlock(&vas_pseries_mutex); + return rc; + } + } else + vascaps[win->win_type].nr_close_wins--; list_del(&win->win_list); - atomic_dec(&caps->used_lpar_creds); + atomic_dec(&caps->nr_used_credits); + vascaps[win->win_type].nr_open_windows--; mutex_unlock(&vas_pseries_mutex); put_vas_user_win_ref(&vwin->task_ref); @@ -461,14 +527,10 @@ static const struct vas_user_win_ops vops_pseries = { int vas_register_api_pseries(struct module *mod, enum vas_cop_type cop_type, const char *name) { - int rc; - if (!copypaste_feat) return -ENOTSUPP; - rc = vas_register_coproc_api(mod, cop_type, name, &vops_pseries); - - return rc; + return vas_register_coproc_api(mod, cop_type, name, &vops_pseries); } EXPORT_SYMBOL_GPL(vas_register_api_pseries); @@ -482,7 +544,7 @@ EXPORT_SYMBOL_GPL(vas_unregister_api_pseries); * Get the specific capabilities based on the feature type. * Right now supports GZIP default and GZIP QoS capabilities. */ -static int get_vas_capabilities(u8 feat, enum vas_cop_feat_type type, +static int __init get_vas_capabilities(u8 feat, enum vas_cop_feat_type type, struct hv_vas_cop_feat_caps *hv_caps) { struct vas_cop_feat_caps *caps; @@ -493,6 +555,7 @@ static int get_vas_capabilities(u8 feat, enum vas_cop_feat_type type, memset(vcaps, 0, sizeof(*vcaps)); INIT_LIST_HEAD(&vcaps->list); + vcaps->feat = feat; caps = &vcaps->caps; rc = h_query_vas_capabilities(H_QUERY_VAS_CAPABILITIES, feat, @@ -514,7 +577,7 @@ static int get_vas_capabilities(u8 feat, enum vas_cop_feat_type type, } caps->max_lpar_creds = be16_to_cpu(hv_caps->max_lpar_creds); caps->max_win_creds = be16_to_cpu(hv_caps->max_win_creds); - atomic_set(&caps->target_lpar_creds, + atomic_set(&caps->nr_total_credits, be16_to_cpu(hv_caps->target_lpar_creds)); if (feat == VAS_GZIP_DEF_FEAT) { caps->def_lpar_creds = be16_to_cpu(hv_caps->def_lpar_creds); @@ -526,16 +589,425 @@ static int get_vas_capabilities(u8 feat, enum vas_cop_feat_type type, } } + rc = sysfs_add_vas_caps(caps); + if (rc) + return rc; + copypaste_feat = true; return 0; } +/* + * VAS windows can be closed due to lost credits when the core is + * removed. So reopen them if credits are available due to DLPAR + * core add and set the window active status. When NX sees the page + * fault on the unmapped paste address, the kernel handles the fault + * by setting the remapping to new paste address if the window is + * active. + */ +static int reconfig_open_windows(struct vas_caps *vcaps, int creds, + bool migrate) +{ + long domain[PLPAR_HCALL9_BUFSIZE] = {VAS_DEFAULT_DOMAIN_ID}; + struct vas_cop_feat_caps *caps = &vcaps->caps; + struct pseries_vas_window *win = NULL, *tmp; + int rc, mv_ents = 0; + int flag; + + /* + * Nothing to do if there are no closed windows. + */ + if (!vcaps->nr_close_wins) + return 0; + + /* + * For the core removal, the hypervisor reduces the credits + * assigned to the LPAR and the kernel closes VAS windows + * in the hypervisor depends on reduced credits. The kernel + * uses LIFO (the last windows that are opened will be closed + * first) and expects to open in the same order when credits + * are available. + * For example, 40 windows are closed when the LPAR lost 2 cores + * (dedicated). If 1 core is added, this LPAR can have 20 more + * credits. It means the kernel can reopen 20 windows. So move + * 20 entries in the VAS windows lost and reopen next 20 windows. + * For partition migration, reopen all windows that are closed + * during resume. + */ + if ((vcaps->nr_close_wins > creds) && !migrate) + mv_ents = vcaps->nr_close_wins - creds; + + list_for_each_entry_safe(win, tmp, &vcaps->list, win_list) { + if (!mv_ents) + break; + + mv_ents--; + } + + /* + * Open windows if they are closed only with migration or + * DLPAR (lost credit) before. + */ + if (migrate) + flag = VAS_WIN_MIGRATE_CLOSE; + else + flag = VAS_WIN_NO_CRED_CLOSE; + + list_for_each_entry_safe_from(win, tmp, &vcaps->list, win_list) { + /* + * This window is closed with DLPAR and migration events. + * So reopen the window with the last event. + * The user space is not suspended with the current + * migration notifier. So the user space can issue DLPAR + * CPU hotplug while migration in progress. In this case + * this window will be opened with the last event. + */ + if ((win->vas_win.status & VAS_WIN_NO_CRED_CLOSE) && + (win->vas_win.status & VAS_WIN_MIGRATE_CLOSE)) { + win->vas_win.status &= ~flag; + continue; + } + + /* + * Nothing to do on this window if it is not closed + * with this flag + */ + if (!(win->vas_win.status & flag)) + continue; + + rc = allocate_setup_window(win, (u64 *)&domain[0], + caps->win_type); + if (rc) + return rc; + + rc = h_modify_vas_window(win); + if (rc) + goto out; + + mutex_lock(&win->vas_win.task_ref.mmap_mutex); + /* + * Set window status to active + */ + win->vas_win.status &= ~flag; + mutex_unlock(&win->vas_win.task_ref.mmap_mutex); + win->win_type = caps->win_type; + if (!--vcaps->nr_close_wins) + break; + } + + return 0; +out: + /* + * Window modify HCALL failed. So close the window to the + * hypervisor and return. + */ + free_irq_setup(win); + h_deallocate_vas_window(win->vas_win.winid); + return rc; +} + +/* + * The hypervisor reduces the available credits if the LPAR lost core. It + * means the excessive windows should not be active and the user space + * should not be using these windows to send compression requests to NX. + * So the kernel closes the excessive windows and unmap the paste address + * such that the user space receives paste instruction failure. Then up to + * the user space to fall back to SW compression and manage with the + * existing windows. + */ +static int reconfig_close_windows(struct vas_caps *vcap, int excess_creds, + bool migrate) +{ + struct pseries_vas_window *win, *tmp; + struct vas_user_win_ref *task_ref; + struct vm_area_struct *vma; + int rc = 0, flag; + + if (migrate) + flag = VAS_WIN_MIGRATE_CLOSE; + else + flag = VAS_WIN_NO_CRED_CLOSE; + + list_for_each_entry_safe(win, tmp, &vcap->list, win_list) { + /* + * This window is already closed due to lost credit + * or for migration before. Go for next window. + * For migration, nothing to do since this window + * closed for DLPAR and will be reopened even on + * the destination system with other DLPAR operation. + */ + if ((win->vas_win.status & VAS_WIN_MIGRATE_CLOSE) || + (win->vas_win.status & VAS_WIN_NO_CRED_CLOSE)) { + win->vas_win.status |= flag; + continue; + } + + task_ref = &win->vas_win.task_ref; + mutex_lock(&task_ref->mmap_mutex); + vma = task_ref->vma; + /* + * Number of available credits are reduced, So select + * and close windows. + */ + win->vas_win.status |= flag; + + mmap_write_lock(task_ref->mm); + /* + * vma is set in the original mapping. But this mapping + * is done with mmap() after the window is opened with ioctl. + * so we may not see the original mapping if the core remove + * is done before the original mmap() and after the ioctl. + */ + if (vma) + zap_page_range(vma, vma->vm_start, + vma->vm_end - vma->vm_start); + + mmap_write_unlock(task_ref->mm); + mutex_unlock(&task_ref->mmap_mutex); + /* + * Close VAS window in the hypervisor, but do not + * free vas_window struct since it may be reused + * when the credit is available later (DLPAR with + * adding cores). This struct will be used + * later when the process issued with close(FD). + */ + rc = deallocate_free_window(win); + /* + * This failure is from the hypervisor. + * No way to stop migration for these failures. + * So ignore error and continue closing other windows. + */ + if (rc && !migrate) + return rc; + + vcap->nr_close_wins++; + + /* + * For migration, do not depend on lpar_creds in case if + * mismatch with the hypervisor value (should not happen). + * So close all active windows in the list and will be + * reopened windows based on the new lpar_creds on the + * destination system during resume. + */ + if (!migrate && !--excess_creds) + break; + } + + return 0; +} + +/* + * Get new VAS capabilities when the core add/removal configuration + * changes. Reconfig window configurations based on the credits + * availability from this new capabilities. + */ +int vas_reconfig_capabilties(u8 type, int new_nr_creds) +{ + struct vas_cop_feat_caps *caps; + int old_nr_creds; + struct vas_caps *vcaps; + int rc = 0, nr_active_wins; + + if (type >= VAS_MAX_FEAT_TYPE) { + pr_err("Invalid credit type %d\n", type); + return -EINVAL; + } + + vcaps = &vascaps[type]; + caps = &vcaps->caps; + + mutex_lock(&vas_pseries_mutex); + + old_nr_creds = atomic_read(&caps->nr_total_credits); + + atomic_set(&caps->nr_total_credits, new_nr_creds); + /* + * The total number of available credits may be decreased or + * increased with DLPAR operation. Means some windows have to be + * closed / reopened. Hold the vas_pseries_mutex so that the + * user space can not open new windows. + */ + if (old_nr_creds < new_nr_creds) { + /* + * If the existing target credits is less than the new + * target, reopen windows if they are closed due to + * the previous DLPAR (core removal). + */ + rc = reconfig_open_windows(vcaps, new_nr_creds - old_nr_creds, + false); + } else { + /* + * # active windows is more than new LPAR available + * credits. So close the excessive windows. + * On pseries, each window will have 1 credit. + */ + nr_active_wins = vcaps->nr_open_windows - vcaps->nr_close_wins; + if (nr_active_wins > new_nr_creds) + rc = reconfig_close_windows(vcaps, + nr_active_wins - new_nr_creds, + false); + } + + mutex_unlock(&vas_pseries_mutex); + return rc; +} + +int pseries_vas_dlpar_cpu(void) +{ + int new_nr_creds, rc; + + rc = h_query_vas_capabilities(H_QUERY_VAS_CAPABILITIES, + vascaps[VAS_GZIP_DEF_FEAT_TYPE].feat, + (u64)virt_to_phys(&hv_cop_caps)); + if (!rc) { + new_nr_creds = be16_to_cpu(hv_cop_caps.target_lpar_creds); + rc = vas_reconfig_capabilties(VAS_GZIP_DEF_FEAT_TYPE, new_nr_creds); + } + + if (rc) + pr_err("Failed reconfig VAS capabilities with DLPAR\n"); + + return rc; +} + +/* + * Total number of default credits available (target_credits) + * in LPAR depends on number of cores configured. It varies based on + * whether processors are in shared mode or dedicated mode. + * Get the notifier when CPU configuration is changed with DLPAR + * operation so that get the new target_credits (vas default capabilities) + * and then update the existing windows usage if needed. + */ +static int pseries_vas_notifier(struct notifier_block *nb, + unsigned long action, void *data) +{ + struct of_reconfig_data *rd = data; + struct device_node *dn = rd->dn; + const __be32 *intserv = NULL; + int len; + + /* + * For shared CPU partition, the hypervisor assigns total credits + * based on entitled core capacity. So updating VAS windows will + * be called from lparcfg_write(). + */ + if (is_shared_processor()) + return NOTIFY_OK; + + if ((action == OF_RECONFIG_ATTACH_NODE) || + (action == OF_RECONFIG_DETACH_NODE)) + intserv = of_get_property(dn, "ibm,ppc-interrupt-server#s", + &len); + /* + * Processor config is not changed + */ + if (!intserv) + return NOTIFY_OK; + + return pseries_vas_dlpar_cpu(); +} + +static struct notifier_block pseries_vas_nb = { + .notifier_call = pseries_vas_notifier, +}; + +/* + * For LPM, all windows have to be closed on the source partition + * before migration and reopen them on the destination partition + * after migration. So closing windows during suspend and + * reopen them during resume. + */ +int vas_migration_handler(int action) +{ + struct vas_cop_feat_caps *caps; + int old_nr_creds, new_nr_creds = 0; + struct vas_caps *vcaps; + int i, rc = 0; + + /* + * NX-GZIP is not enabled. Nothing to do for migration. + */ + if (!copypaste_feat) + return rc; + + mutex_lock(&vas_pseries_mutex); + + if (action == VAS_SUSPEND) + migration_in_progress = true; + else + migration_in_progress = false; + + for (i = 0; i < VAS_MAX_FEAT_TYPE; i++) { + vcaps = &vascaps[i]; + caps = &vcaps->caps; + old_nr_creds = atomic_read(&caps->nr_total_credits); + + rc = h_query_vas_capabilities(H_QUERY_VAS_CAPABILITIES, + vcaps->feat, + (u64)virt_to_phys(&hv_cop_caps)); + if (!rc) { + new_nr_creds = be16_to_cpu(hv_cop_caps.target_lpar_creds); + /* + * Should not happen. But incase print messages, close + * all windows in the list during suspend and reopen + * windows based on new lpar_creds on the destination + * system. + */ + if (old_nr_creds != new_nr_creds) { + pr_err("Target credits mismatch with the hypervisor\n"); + pr_err("state(%d): lpar creds: %d HV lpar creds: %d\n", + action, old_nr_creds, new_nr_creds); + pr_err("Used creds: %d, Active creds: %d\n", + atomic_read(&caps->nr_used_credits), + vcaps->nr_open_windows - vcaps->nr_close_wins); + } + } else { + pr_err("state(%d): Get VAS capabilities failed with %d\n", + action, rc); + /* + * We can not stop migration with the current lpm + * implementation. So continue closing all windows in + * the list (during suspend) and return without + * opening windows (during resume) if VAS capabilities + * HCALL failed. + */ + if (action == VAS_RESUME) + goto out; + } + + switch (action) { + case VAS_SUSPEND: + rc = reconfig_close_windows(vcaps, vcaps->nr_open_windows, + true); + break; + case VAS_RESUME: + atomic_set(&caps->nr_total_credits, new_nr_creds); + rc = reconfig_open_windows(vcaps, new_nr_creds, true); + break; + default: + /* should not happen */ + pr_err("Invalid migration action %d\n", action); + rc = -EINVAL; + goto out; + } + + /* + * Ignore errors during suspend and return for resume. + */ + if (rc && (action == VAS_RESUME)) + goto out; + } + +out: + mutex_unlock(&vas_pseries_mutex); + return rc; +} + static int __init pseries_vas_init(void) { - struct hv_vas_cop_feat_caps *hv_cop_caps; struct hv_vas_all_caps *hv_caps; - int rc; + int rc = 0; /* * Linux supports user space COPY/PASTE only with Radix @@ -559,35 +1031,39 @@ static int __init pseries_vas_init(void) caps_all.descriptor = be64_to_cpu(hv_caps->descriptor); caps_all.feat_type = be64_to_cpu(hv_caps->feat_type); - hv_cop_caps = kmalloc(sizeof(*hv_cop_caps), GFP_KERNEL); - if (!hv_cop_caps) { - rc = -ENOMEM; - goto out; - } + sysfs_pseries_vas_init(&caps_all); + /* * QOS capabilities available */ if (caps_all.feat_type & VAS_GZIP_QOS_FEAT_BIT) { rc = get_vas_capabilities(VAS_GZIP_QOS_FEAT, - VAS_GZIP_QOS_FEAT_TYPE, hv_cop_caps); + VAS_GZIP_QOS_FEAT_TYPE, &hv_cop_caps); if (rc) - goto out_cop; + goto out; } /* * Default capabilities available */ - if (caps_all.feat_type & VAS_GZIP_DEF_FEAT_BIT) { + if (caps_all.feat_type & VAS_GZIP_DEF_FEAT_BIT) rc = get_vas_capabilities(VAS_GZIP_DEF_FEAT, - VAS_GZIP_DEF_FEAT_TYPE, hv_cop_caps); - if (rc) - goto out_cop; - } + VAS_GZIP_DEF_FEAT_TYPE, &hv_cop_caps); + + if (!rc && copypaste_feat) { + if (firmware_has_feature(FW_FEATURE_LPAR)) + of_reconfig_notifier_register(&pseries_vas_nb); - pr_info("GZIP feature is available\n"); + pr_info("GZIP feature is available\n"); + } else { + /* + * Should not happen, but only when get default + * capabilities HCALL failed. So disable copy paste + * feature. + */ + copypaste_feat = false; + } -out_cop: - kfree(hv_cop_caps); out: kfree(hv_caps); return rc; diff --git a/arch/powerpc/platforms/pseries/vas.h b/arch/powerpc/platforms/pseries/vas.h index 4ecb3fcabd10..7115043ec488 100644 --- a/arch/powerpc/platforms/pseries/vas.h +++ b/arch/powerpc/platforms/pseries/vas.h @@ -30,6 +30,14 @@ #define VAS_COPY_PASTE_USER_MODE 0x00000001 #define VAS_COP_OP_USER_MODE 0x00000010 +#define VAS_GZIP_QOS_CAPABILITIES 0x56516F73477A6970 +#define VAS_GZIP_DEFAULT_CAPABILITIES 0x56446566477A6970 + +enum vas_migrate_action { + VAS_SUSPEND, + VAS_RESUME, +}; + /* * Co-processor feature - GZIP QoS windows or GZIP default windows */ @@ -72,9 +80,8 @@ struct vas_cop_feat_caps { }; /* Total LPAR available credits. Can be different from max LPAR */ /* credits due to DLPAR operation */ - atomic_t target_lpar_creds; - atomic_t used_lpar_creds; /* Used credits so far */ - u16 avail_lpar_creds; /* Remaining available credits */ + atomic_t nr_total_credits; /* Total credits assigned to LPAR */ + atomic_t nr_used_credits; /* Used credits so far */ }; /* @@ -84,6 +91,9 @@ struct vas_cop_feat_caps { struct vas_caps { struct vas_cop_feat_caps caps; struct list_head list; /* List of open windows */ + int nr_close_wins; /* closed windows in the hypervisor for DLPAR */ + int nr_open_windows; /* Number of successful open windows */ + u8 feat; /* Feature type */ }; /* @@ -115,11 +125,31 @@ struct pseries_vas_window { u64 domain[6]; /* Associativity domain Ids */ /* this window is allocated */ u64 util; + u32 pid; /* PID associated with this window */ /* List of windows opened which is used for LPM */ struct list_head win_list; u64 flags; char *name; int fault_virq; + atomic_t pending_faults; /* Number of pending faults */ }; + +int sysfs_add_vas_caps(struct vas_cop_feat_caps *caps); +int vas_reconfig_capabilties(u8 type, int new_nr_creds); +int __init sysfs_pseries_vas_init(struct vas_all_caps *vas_caps); + +#ifdef CONFIG_PPC_VAS +int vas_migration_handler(int action); +int pseries_vas_dlpar_cpu(void); +#else +static inline int vas_migration_handler(int action) +{ + return 0; +} +static inline int pseries_vas_dlpar_cpu(void) +{ + return 0; +} +#endif #endif /* _VAS_H */ diff --git a/arch/powerpc/platforms/pseries/vio.c b/arch/powerpc/platforms/pseries/vio.c index feafcb582e1b..00ecac2c205b 100644 --- a/arch/powerpc/platforms/pseries/vio.c +++ b/arch/powerpc/platforms/pseries/vio.c @@ -23,6 +23,7 @@ #include <linux/dma-map-ops.h> #include <linux/kobject.h> #include <linux/kexec.h> +#include <linux/of_irq.h> #include <asm/iommu.h> #include <asm/dma.h> @@ -1061,7 +1062,7 @@ static struct attribute *vio_bus_attrs[] = { }; ATTRIBUTE_GROUPS(vio_bus); -static void vio_cmo_sysfs_init(void) +static void __init vio_cmo_sysfs_init(void) { vio_bus_type.dev_groups = vio_cmo_dev_groups; vio_bus_type.bus_groups = vio_bus_groups; @@ -1073,7 +1074,7 @@ static int vio_cmo_bus_probe(struct vio_dev *viodev) { return 0; } static void vio_cmo_bus_remove(struct vio_dev *viodev) {} static void vio_cmo_set_dma_ops(struct vio_dev *viodev) {} static void vio_cmo_bus_init(void) {} -static void vio_cmo_sysfs_init(void) { } +static void __init vio_cmo_sysfs_init(void) { } #endif /* CONFIG_PPC_SMLPAR */ EXPORT_SYMBOL(vio_cmo_entitlement_update); EXPORT_SYMBOL(vio_cmo_set_dev_desired); @@ -1479,7 +1480,7 @@ EXPORT_SYMBOL(vio_register_device_node); * Starting from the root node provide, register the device node for * each child beneath the root. */ -static void vio_bus_scan_register_devices(char *root_name) +static void __init vio_bus_scan_register_devices(char *root_name) { struct device_node *node_root, *node_child; |