diff options
author | 2017-09-04 13:08:27 -0700 | |
---|---|---|
committer | 2017-09-04 13:08:27 -0700 | |
commit | 93cc1228b4a60584f3dadbd84f19f365bb945acb (patch) | |
tree | 9b50949e72f6df7116727e898a137f0881cb1b43 /drivers/irqchip | |
parent | Merge branch 'timers-core-for-linus' of git://git.kernel.org/pub/scm/linux/ke... (diff) | |
parent | Merge tag 'irqchip-4.14' of git://git.kernel.org/pub/scm/linux/kernel/git/maz... (diff) | |
download | linux-93cc1228b4a60584f3dadbd84f19f365bb945acb.tar.gz linux-93cc1228b4a60584f3dadbd84f19f365bb945acb.tar.bz2 linux-93cc1228b4a60584f3dadbd84f19f365bb945acb.zip |
Merge branch 'irq-core-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip
Pull irq updates from Thomas Gleixner:
"The interrupt subsystem delivers this time:
- Refactoring of the GIC-V3 driver to prepare for the GIC-V4 support
- Initial GIC-V4 support
- Consolidation of the FSL MSI support
- Utilize the effective affinity interface in various ARM irqchip
drivers
- Yet another interrupt chip driver (UniPhier AIDET)
- Bulk conversion of the irq chip driver to use %pOF
- The usual small fixes and improvements all over the place"
* 'irq-core-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: (77 commits)
irqchip/ls-scfg-msi: Add MSI affinity support
irqchip/ls-scfg-msi: Add LS1043a v1.1 MSI support
irqchip/ls-scfg-msi: Add LS1046a MSI support
arm64: dts: ls1046a: Add MSI dts node
arm64: dts: ls1043a: Share all MSIs
arm: dts: ls1021a: Share all MSIs
arm64: dts: ls1043a: Fix typo of MSI compatible string
arm: dts: ls1021a: Fix typo of MSI compatible string
irqchip/ls-scfg-msi: Fix typo of MSI compatible strings
irqchip/irq-bcm7120-l2: Use correct I/O accessors for irq_fwd_mask
irqchip/mmp: Make mmp_intc_conf const
irqchip/gic: Make irq_chip const
irqchip/gic-v3: Advertise GICv4 support to KVM
irqchip/gic-v4: Enable low-level GICv4 operations
irqchip/gic-v4: Add some basic documentation
irqchip/gic-v4: Add VLPI configuration interface
irqchip/gic-v4: Add VPE command interface
irqchip/gic-v4: Add per-VM VPE domain creation
irqchip/gic-v3-its: Set implementation defined bit to enable VLPIs
irqchip/gic-v3-its: Allow doorbell interrupts to be injected/cleared
...
Diffstat (limited to 'drivers/irqchip')
31 files changed, 2266 insertions, 241 deletions
diff --git a/drivers/irqchip/Kconfig b/drivers/irqchip/Kconfig index f1fd5f44d1d4..9d8a1dd2e2c2 100644 --- a/drivers/irqchip/Kconfig +++ b/drivers/irqchip/Kconfig @@ -7,6 +7,7 @@ config ARM_GIC select IRQ_DOMAIN select IRQ_DOMAIN_HIERARCHY select MULTI_IRQ_HANDLER + select GENERIC_IRQ_EFFECTIVE_AFF_MASK config ARM_GIC_PM bool @@ -34,6 +35,7 @@ config ARM_GIC_V3 select MULTI_IRQ_HANDLER select IRQ_DOMAIN_HIERARCHY select PARTITION_PERCPU + select GENERIC_IRQ_EFFECTIVE_AFF_MASK config ARM_GIC_V3_ITS bool @@ -64,6 +66,7 @@ config ARMADA_370_XP_IRQ bool select GENERIC_IRQ_CHIP select PCI_MSI if PCI + select GENERIC_IRQ_EFFECTIVE_AFF_MASK config ALPINE_MSI bool @@ -93,11 +96,13 @@ config BCM6345_L1_IRQ bool select GENERIC_IRQ_CHIP select IRQ_DOMAIN + select GENERIC_IRQ_EFFECTIVE_AFF_MASK config BCM7038_L1_IRQ bool select GENERIC_IRQ_CHIP select IRQ_DOMAIN + select GENERIC_IRQ_EFFECTIVE_AFF_MASK config BCM7120_L2_IRQ bool @@ -136,6 +141,7 @@ config IRQ_MIPS_CPU select GENERIC_IRQ_IPI if SYS_SUPPORTS_MULTITHREADING select IRQ_DOMAIN select IRQ_DOMAIN_HIERARCHY if GENERIC_IRQ_IPI + select GENERIC_IRQ_EFFECTIVE_AFF_MASK config CLPS711X_IRQCHIP bool @@ -217,6 +223,7 @@ config VERSATILE_FPGA_IRQ_NR config XTENSA_MX bool select IRQ_DOMAIN + select GENERIC_IRQ_EFFECTIVE_AFF_MASK config XILINX_INTC bool @@ -306,3 +313,11 @@ config QCOM_IRQ_COMBINER help Say yes here to add support for the IRQ combiner devices embedded in Qualcomm Technologies chips. + +config IRQ_UNIPHIER_AIDET + bool "UniPhier AIDET support" if COMPILE_TEST + depends on ARCH_UNIPHIER || COMPILE_TEST + default ARCH_UNIPHIER + select IRQ_DOMAIN_HIERARCHY + help + Support for the UniPhier AIDET (ARM Interrupt Detector). diff --git a/drivers/irqchip/Makefile b/drivers/irqchip/Makefile index e88d856cc09c..845abc107ad5 100644 --- a/drivers/irqchip/Makefile +++ b/drivers/irqchip/Makefile @@ -28,7 +28,7 @@ obj-$(CONFIG_ARM_GIC_PM) += irq-gic-pm.o obj-$(CONFIG_ARCH_REALVIEW) += irq-gic-realview.o obj-$(CONFIG_ARM_GIC_V2M) += irq-gic-v2m.o obj-$(CONFIG_ARM_GIC_V3) += irq-gic-v3.o irq-gic-common.o -obj-$(CONFIG_ARM_GIC_V3_ITS) += irq-gic-v3-its.o irq-gic-v3-its-pci-msi.o irq-gic-v3-its-platform-msi.o +obj-$(CONFIG_ARM_GIC_V3_ITS) += irq-gic-v3-its.o irq-gic-v3-its-pci-msi.o irq-gic-v3-its-platform-msi.o irq-gic-v4.o obj-$(CONFIG_PARTITION_PERCPU) += irq-partition-percpu.o obj-$(CONFIG_HISILICON_IRQ_MBIGEN) += irq-mbigen.o obj-$(CONFIG_ARM_NVIC) += irq-nvic.o @@ -78,3 +78,4 @@ obj-$(CONFIG_EZNPS_GIC) += irq-eznps.o obj-$(CONFIG_ARCH_ASPEED) += irq-aspeed-vic.o irq-aspeed-i2c-ic.o obj-$(CONFIG_STM32_EXTI) += irq-stm32-exti.o obj-$(CONFIG_QCOM_IRQ_COMBINER) += qcom-irq-combiner.o +obj-$(CONFIG_IRQ_UNIPHIER_AIDET) += irq-uniphier-aidet.o diff --git a/drivers/irqchip/irq-armada-370-xp.c b/drivers/irqchip/irq-armada-370-xp.c index b207b2c3aa55..c9bdc5221b82 100644 --- a/drivers/irqchip/irq-armada-370-xp.c +++ b/drivers/irqchip/irq-armada-370-xp.c @@ -203,7 +203,7 @@ static struct irq_chip armada_370_xp_msi_irq_chip = { static struct msi_domain_info armada_370_xp_msi_domain_info = { .flags = (MSI_FLAG_USE_DEF_DOM_OPS | MSI_FLAG_USE_DEF_CHIP_OPS | - MSI_FLAG_MULTI_PCI_MSI), + MSI_FLAG_MULTI_PCI_MSI | MSI_FLAG_PCI_MSIX), .chip = &armada_370_xp_msi_irq_chip, }; @@ -330,6 +330,8 @@ static int armada_xp_set_affinity(struct irq_data *d, writel(reg, main_int_base + ARMADA_370_XP_INT_SOURCE_CTL(hwirq)); raw_spin_unlock(&irq_controller_lock); + irq_data_update_effective_affinity(d, cpumask_of(cpu)); + return IRQ_SET_MASK_OK; } #endif @@ -363,6 +365,7 @@ static int armada_370_xp_mpic_irq_map(struct irq_domain *h, } else { irq_set_chip_and_handler(virq, &armada_370_xp_irq_chip, handle_level_irq); + irqd_set_single_target(irq_desc_get_irq_data(irq_to_desc(virq))); } irq_set_probe(virq); diff --git a/drivers/irqchip/irq-bcm2835.c b/drivers/irqchip/irq-bcm2835.c index 44d7c38dde47..d2da8a1e6b1b 100644 --- a/drivers/irqchip/irq-bcm2835.c +++ b/drivers/irqchip/irq-bcm2835.c @@ -147,13 +147,12 @@ static int __init armctrl_of_init(struct device_node *node, base = of_iomap(node, 0); if (!base) - panic("%s: unable to map IC registers\n", - node->full_name); + panic("%pOF: unable to map IC registers\n", node); intc.domain = irq_domain_add_linear(node, MAKE_HWIRQ(NR_BANKS, 0), &armctrl_ops, NULL); if (!intc.domain) - panic("%s: unable to create IRQ domain\n", node->full_name); + panic("%pOF: unable to create IRQ domain\n", node); for (b = 0; b < NR_BANKS; b++) { intc.pending[b] = base + reg_pending[b]; @@ -173,8 +172,8 @@ static int __init armctrl_of_init(struct device_node *node, int parent_irq = irq_of_parse_and_map(node, 0); if (!parent_irq) { - panic("%s: unable to get parent interrupt.\n", - node->full_name); + panic("%pOF: unable to get parent interrupt.\n", + node); } irq_set_chained_handler(parent_irq, bcm2836_chained_handle_irq); } else { diff --git a/drivers/irqchip/irq-bcm2836.c b/drivers/irqchip/irq-bcm2836.c index e7463e3c0814..dc8c1e3eafe7 100644 --- a/drivers/irqchip/irq-bcm2836.c +++ b/drivers/irqchip/irq-bcm2836.c @@ -282,8 +282,7 @@ static int __init bcm2836_arm_irqchip_l1_intc_of_init(struct device_node *node, { intc.base = of_iomap(node, 0); if (!intc.base) { - panic("%s: unable to map local interrupt registers\n", - node->full_name); + panic("%pOF: unable to map local interrupt registers\n", node); } bcm2835_init_local_timer_frequency(); @@ -292,7 +291,7 @@ static int __init bcm2836_arm_irqchip_l1_intc_of_init(struct device_node *node, &bcm2836_arm_irqchip_intc_ops, NULL); if (!intc.domain) - panic("%s: unable to create IRQ domain\n", node->full_name); + panic("%pOF: unable to create IRQ domain\n", node); bcm2836_arm_irqchip_register_irq(LOCAL_IRQ_CNTPSIRQ, &bcm2836_arm_irqchip_timer); diff --git a/drivers/irqchip/irq-bcm6345-l1.c b/drivers/irqchip/irq-bcm6345-l1.c index daa4ae89e466..43f8abe40878 100644 --- a/drivers/irqchip/irq-bcm6345-l1.c +++ b/drivers/irqchip/irq-bcm6345-l1.c @@ -231,6 +231,8 @@ static int bcm6345_l1_set_affinity(struct irq_data *d, } raw_spin_unlock_irqrestore(&intc->lock, flags); + irq_data_update_effective_affinity(d, cpumask_of(new_cpu)); + return IRQ_SET_MASK_OK_NOCOPY; } @@ -291,6 +293,7 @@ static int bcm6345_l1_map(struct irq_domain *d, unsigned int virq, irq_set_chip_and_handler(virq, &bcm6345_l1_irq_chip, handle_percpu_irq); irq_set_chip_data(virq, d->host_data); + irqd_set_single_target(irq_desc_get_irq_data(irq_to_desc(virq))); return 0; } diff --git a/drivers/irqchip/irq-bcm7038-l1.c b/drivers/irqchip/irq-bcm7038-l1.c index c2662a1bfdd3..55cfb986225b 100644 --- a/drivers/irqchip/irq-bcm7038-l1.c +++ b/drivers/irqchip/irq-bcm7038-l1.c @@ -212,6 +212,8 @@ static int bcm7038_l1_set_affinity(struct irq_data *d, __bcm7038_l1_unmask(d, first_cpu); raw_spin_unlock_irqrestore(&intc->lock, flags); + irq_data_update_effective_affinity(d, cpumask_of(first_cpu)); + return 0; } @@ -299,6 +301,7 @@ static int bcm7038_l1_map(struct irq_domain *d, unsigned int virq, { irq_set_chip_and_handler(virq, &bcm7038_l1_irq_chip, handle_level_irq); irq_set_chip_data(virq, d->host_data); + irqd_set_single_target(irq_desc_get_irq_data(irq_to_desc(virq))); return 0; } diff --git a/drivers/irqchip/irq-bcm7120-l2.c b/drivers/irqchip/irq-bcm7120-l2.c index 64c2692070ef..983640eba418 100644 --- a/drivers/irqchip/irq-bcm7120-l2.c +++ b/drivers/irqchip/irq-bcm7120-l2.c @@ -250,12 +250,6 @@ static int __init bcm7120_l2_intc_probe(struct device_node *dn, if (ret < 0) goto out_free_l1_data; - for (idx = 0; idx < data->n_words; idx++) { - __raw_writel(data->irq_fwd_mask[idx], - data->pair_base[idx] + - data->en_offset[idx]); - } - for (irq = 0; irq < data->num_parent_irqs; irq++) { ret = bcm7120_l2_intc_init_one(dn, data, irq, valid_mask); if (ret) @@ -297,6 +291,10 @@ static int __init bcm7120_l2_intc_probe(struct device_node *dn, gc->reg_base = data->pair_base[idx]; ct->regs.mask = data->en_offset[idx]; + /* gc->reg_base is defined and so is gc->writel */ + irq_reg_writel(gc, data->irq_fwd_mask[idx], + data->en_offset[idx]); + ct->chip.irq_mask = irq_gc_mask_clr_bit; ct->chip.irq_unmask = irq_gc_mask_set_bit; ct->chip.irq_ack = irq_gc_noop; diff --git a/drivers/irqchip/irq-crossbar.c b/drivers/irqchip/irq-crossbar.c index f96601268f71..99d97d7e3fd7 100644 --- a/drivers/irqchip/irq-crossbar.c +++ b/drivers/irqchip/irq-crossbar.c @@ -341,13 +341,13 @@ static int __init irqcrossbar_init(struct device_node *node, int err; if (!parent) { - pr_err("%s: no parent, giving up\n", node->full_name); + pr_err("%pOF: no parent, giving up\n", node); return -ENODEV; } parent_domain = irq_find_host(parent); if (!parent_domain) { - pr_err("%s: unable to obtain parent domain\n", node->full_name); + pr_err("%pOF: unable to obtain parent domain\n", node); return -ENXIO; } @@ -360,7 +360,7 @@ static int __init irqcrossbar_init(struct device_node *node, node, &crossbar_domain_ops, NULL); if (!domain) { - pr_err("%s: failed to allocated domain\n", node->full_name); + pr_err("%pOF: failed to allocated domain\n", node); return -ENOMEM; } diff --git a/drivers/irqchip/irq-digicolor.c b/drivers/irqchip/irq-digicolor.c index 3aae015469a5..fc38d2da11b9 100644 --- a/drivers/irqchip/irq-digicolor.c +++ b/drivers/irqchip/irq-digicolor.c @@ -78,7 +78,7 @@ static int __init digicolor_of_init(struct device_node *node, reg_base = of_iomap(node, 0); if (!reg_base) { - pr_err("%s: unable to map IC registers\n", node->full_name); + pr_err("%pOF: unable to map IC registers\n", node); return -ENXIO; } @@ -88,7 +88,7 @@ static int __init digicolor_of_init(struct device_node *node, ucregs = syscon_regmap_lookup_by_phandle(node, "syscon"); if (IS_ERR(ucregs)) { - pr_err("%s: unable to map UC registers\n", node->full_name); + pr_err("%pOF: unable to map UC registers\n", node); return PTR_ERR(ucregs); } /* channel 1, regular IRQs */ @@ -97,7 +97,7 @@ static int __init digicolor_of_init(struct device_node *node, digicolor_irq_domain = irq_domain_add_linear(node, 64, &irq_generic_chip_ops, NULL); if (!digicolor_irq_domain) { - pr_err("%s: unable to create IRQ domain\n", node->full_name); + pr_err("%pOF: unable to create IRQ domain\n", node); return -ENOMEM; } @@ -105,7 +105,7 @@ static int __init digicolor_of_init(struct device_node *node, "digicolor_irq", handle_level_irq, clr, 0, 0); if (ret) { - pr_err("%s: unable to allocate IRQ gc\n", node->full_name); + pr_err("%pOF: unable to allocate IRQ gc\n", node); return ret; } diff --git a/drivers/irqchip/irq-dw-apb-ictl.c b/drivers/irqchip/irq-dw-apb-ictl.c index 052f266364c0..0a19618ce2c8 100644 --- a/drivers/irqchip/irq-dw-apb-ictl.c +++ b/drivers/irqchip/irq-dw-apb-ictl.c @@ -79,24 +79,24 @@ static int __init dw_apb_ictl_init(struct device_node *np, /* Map the parent interrupt for the chained handler */ irq = irq_of_parse_and_map(np, 0); if (irq <= 0) { - pr_err("%s: unable to parse irq\n", np->full_name); + pr_err("%pOF: unable to parse irq\n", np); return -EINVAL; } ret = of_address_to_resource(np, 0, &r); if (ret) { - pr_err("%s: unable to get resource\n", np->full_name); + pr_err("%pOF: unable to get resource\n", np); return ret; } if (!request_mem_region(r.start, resource_size(&r), np->full_name)) { - pr_err("%s: unable to request mem region\n", np->full_name); + pr_err("%pOF: unable to request mem region\n", np); return -ENOMEM; } iobase = ioremap(r.start, resource_size(&r)); if (!iobase) { - pr_err("%s: unable to map resource\n", np->full_name); + pr_err("%pOF: unable to map resource\n", np); ret = -ENOMEM; goto err_release; } @@ -123,7 +123,7 @@ static int __init dw_apb_ictl_init(struct device_node *np, domain = irq_domain_add_linear(np, nrirqs, &irq_generic_chip_ops, NULL); if (!domain) { - pr_err("%s: unable to add irq domain\n", np->full_name); + pr_err("%pOF: unable to add irq domain\n", np); ret = -ENOMEM; goto err_unmap; } @@ -132,7 +132,7 @@ static int __init dw_apb_ictl_init(struct device_node *np, handle_level_irq, clr, 0, IRQ_GC_INIT_MASK_CACHE); if (ret) { - pr_err("%s: unable to alloc irq domain gc\n", np->full_name); + pr_err("%pOF: unable to alloc irq domain gc\n", np); goto err_unmap; } diff --git a/drivers/irqchip/irq-gic-v3-its-pci-msi.c b/drivers/irqchip/irq-gic-v3-its-pci-msi.c index 77931214d954..14a8c0a7e095 100644 --- a/drivers/irqchip/irq-gic-v3-its-pci-msi.c +++ b/drivers/irqchip/irq-gic-v3-its-pci-msi.c @@ -138,7 +138,7 @@ static int __init its_pci_of_msi_init(void) if (its_pci_msi_init_one(of_node_to_fwnode(np), np->full_name)) continue; - pr_info("PCI/MSI: %s domain created\n", np->full_name); + pr_info("PCI/MSI: %pOF domain created\n", np); } return 0; diff --git a/drivers/irqchip/irq-gic-v3-its.c b/drivers/irqchip/irq-gic-v3-its.c index 284738add89b..e8d89343d613 100644 --- a/drivers/irqchip/irq-gic-v3-its.c +++ b/drivers/irqchip/irq-gic-v3-its.c @@ -1,5 +1,5 @@ /* - * Copyright (C) 2013, 2014 ARM Limited, All Rights Reserved. + * Copyright (C) 2013-2017 ARM Limited, All Rights Reserved. * Author: Marc Zyngier <marc.zyngier@arm.com> * * This program is free software; you can redistribute it and/or modify @@ -36,6 +36,7 @@ #include <linux/irqchip.h> #include <linux/irqchip/arm-gic-v3.h> +#include <linux/irqchip/arm-gic-v4.h> #include <asm/cputype.h> #include <asm/exception.h> @@ -48,6 +49,19 @@ #define RDIST_FLAGS_PROPBASE_NEEDS_FLUSHING (1 << 0) +static u32 lpi_id_bits; + +/* + * We allocate memory for PROPBASE to cover 2 ^ lpi_id_bits LPIs to + * deal with (one configuration byte per interrupt). PENDBASE has to + * be 64kB aligned (one bit per LPI, plus 8192 bits for SPI/PPI/SGI). + */ +#define LPI_NRBITS lpi_id_bits +#define LPI_PROPBASE_SZ ALIGN(BIT(LPI_NRBITS), SZ_64K) +#define LPI_PENDBASE_SZ ALIGN(BIT(LPI_NRBITS) / 8, SZ_64K) + +#define LPI_PROP_DEFAULT_PRIO 0xa0 + /* * Collection structure - just an ID, and a redistributor address to * ping. We use one per CPU as a bag of interrupts assigned to this @@ -88,6 +102,7 @@ struct its_node { u32 ite_size; u32 device_ids; int numa_node; + bool is_v4; }; #define ITS_ITT_ALIGN SZ_256 @@ -100,11 +115,17 @@ struct event_lpi_map { u16 *col_map; irq_hw_number_t lpi_base; int nr_lpis; + struct mutex vlpi_lock; + struct its_vm *vm; + struct its_vlpi_map *vlpi_maps; + int nr_vlpis; }; /* - * The ITS view of a device - belongs to an ITS, a collection, owns an - * interrupt translation table, and a list of interrupts. + * The ITS view of a device - belongs to an ITS, owns an interrupt + * translation table, and a list of interrupts. If it some of its + * LPIs are injected into a guest (GICv4), the event_map.vm field + * indicates which one. */ struct its_device { struct list_head entry; @@ -115,13 +136,33 @@ struct its_device { u32 device_id; }; +static struct { + raw_spinlock_t lock; + struct its_device *dev; + struct its_vpe **vpes; + int next_victim; +} vpe_proxy; + static LIST_HEAD(its_nodes); static DEFINE_SPINLOCK(its_lock); static struct rdists *gic_rdists; static struct irq_domain *its_parent; +/* + * We have a maximum number of 16 ITSs in the whole system if we're + * using the ITSList mechanism + */ +#define ITS_LIST_MAX 16 + +static unsigned long its_list_map; +static u16 vmovp_seq_num; +static DEFINE_RAW_SPINLOCK(vmovp_lock); + +static DEFINE_IDA(its_vpeid_ida); + #define gic_data_rdist() (raw_cpu_ptr(gic_rdists->rdist)) #define gic_data_rdist_rd_base() (gic_data_rdist()->rd_base) +#define gic_data_rdist_vlpi_base() (gic_data_rdist_rd_base() + SZ_128K) static struct its_collection *dev_event_to_col(struct its_device *its_dev, u32 event) @@ -145,6 +186,11 @@ struct its_cmd_desc { struct { struct its_device *dev; u32 event_id; + } its_clear_cmd; + + struct { + struct its_device *dev; + u32 event_id; } its_int_cmd; struct { @@ -177,6 +223,38 @@ struct its_cmd_desc { struct { struct its_collection *col; } its_invall_cmd; + + struct { + struct its_vpe *vpe; + } its_vinvall_cmd; + + struct { + struct its_vpe *vpe; + struct its_collection *col; + bool valid; + } its_vmapp_cmd; + + struct { + struct its_vpe *vpe; + struct its_device *dev; + u32 virt_id; + u32 event_id; + bool db_enabled; + } its_vmapti_cmd; + + struct { + struct its_vpe *vpe; + struct its_device *dev; + u32 event_id; + bool db_enabled; + } its_vmovi_cmd; + + struct { + struct its_vpe *vpe; + struct its_collection *col; + u16 seq_num; + u16 its_list; + } its_vmovp_cmd; }; }; @@ -193,6 +271,9 @@ struct its_cmd_block { typedef struct its_collection *(*its_cmd_builder_t)(struct its_cmd_block *, struct its_cmd_desc *); +typedef struct its_vpe *(*its_cmd_vbuilder_t)(struct its_cmd_block *, + struct its_cmd_desc *); + static void its_mask_encode(u64 *raw_cmd, u64 val, int h, int l) { u64 mask = GENMASK_ULL(h, l); @@ -245,6 +326,46 @@ static void its_encode_collection(struct its_cmd_block *cmd, u16 col) its_mask_encode(&cmd->raw_cmd[2], col, 15, 0); } +static void its_encode_vpeid(struct its_cmd_block *cmd, u16 vpeid) +{ + its_mask_encode(&cmd->raw_cmd[1], vpeid, 47, 32); +} + +static void its_encode_virt_id(struct its_cmd_block *cmd, u32 virt_id) +{ + its_mask_encode(&cmd->raw_cmd[2], virt_id, 31, 0); +} + +static void its_encode_db_phys_id(struct its_cmd_block *cmd, u32 db_phys_id) +{ + its_mask_encode(&cmd->raw_cmd[2], db_phys_id, 63, 32); +} + +static void its_encode_db_valid(struct its_cmd_block *cmd, bool db_valid) +{ + its_mask_encode(&cmd->raw_cmd[2], db_valid, 0, 0); +} + +static void its_encode_seq_num(struct its_cmd_block *cmd, u16 seq_num) +{ + its_mask_encode(&cmd->raw_cmd[0], seq_num, 47, 32); +} + +static void its_encode_its_list(struct its_cmd_block *cmd, u16 its_list) +{ + its_mask_encode(&cmd->raw_cmd[1], its_list, 15, 0); +} + +static void its_encode_vpt_addr(struct its_cmd_block *cmd, u64 vpt_pa) +{ + its_mask_encode(&cmd->raw_cmd[3], vpt_pa >> 16, 50, 16); +} + +static void its_encode_vpt_size(struct its_cmd_block *cmd, u8 vpt_size) +{ + its_mask_encode(&cmd->raw_cmd[3], vpt_size, 4, 0); +} + static inline void its_fixup_cmd(struct its_cmd_block *cmd) { /* Let's fixup BE commands */ @@ -358,6 +479,40 @@ static struct its_collection *its_build_inv_cmd(struct its_cmd_block *cmd, return col; } +static struct its_collection *its_build_int_cmd(struct its_cmd_block *cmd, + struct its_cmd_desc *desc) +{ + struct its_collection *col; + + col = dev_event_to_col(desc->its_int_cmd.dev, + desc->its_int_cmd.event_id); + + its_encode_cmd(cmd, GITS_CMD_INT); + its_encode_devid(cmd, desc->its_int_cmd.dev->device_id); + its_encode_event_id(cmd, desc->its_int_cmd.event_id); + + its_fixup_cmd(cmd); + + return col; +} + +static struct its_collection *its_build_clear_cmd(struct its_cmd_block *cmd, + struct its_cmd_desc *desc) +{ + struct its_collection *col; + + col = dev_event_to_col(desc->its_clear_cmd.dev, + desc->its_clear_cmd.event_id); + + its_encode_cmd(cmd, GITS_CMD_CLEAR); + its_encode_devid(cmd, desc->its_clear_cmd.dev->device_id); + its_encode_event_id(cmd, desc->its_clear_cmd.event_id); + + its_fixup_cmd(cmd); + + return col; +} + static struct its_collection *its_build_invall_cmd(struct its_cmd_block *cmd, struct its_cmd_desc *desc) { @@ -369,6 +524,94 @@ static struct its_collection *its_build_invall_cmd(struct its_cmd_block *cmd, return NULL; } +static struct its_vpe *its_build_vinvall_cmd(struct its_cmd_block *cmd, + struct its_cmd_desc *desc) +{ + its_encode_cmd(cmd, GITS_CMD_VINVALL); + its_encode_vpeid(cmd, desc->its_vinvall_cmd.vpe->vpe_id); + + its_fixup_cmd(cmd); + + return desc->its_vinvall_cmd.vpe; +} + +static struct its_vpe *its_build_vmapp_cmd(struct its_cmd_block *cmd, + struct its_cmd_desc *desc) +{ + unsigned long vpt_addr; + + vpt_addr = virt_to_phys(page_address(desc->its_vmapp_cmd.vpe->vpt_page)); + + its_encode_cmd(cmd, GITS_CMD_VMAPP); + its_encode_vpeid(cmd, desc->its_vmapp_cmd.vpe->vpe_id); + its_encode_valid(cmd, desc->its_vmapp_cmd.valid); + its_encode_target(cmd, desc->its_vmapp_cmd.col->target_address); + its_encode_vpt_addr(cmd, vpt_addr); + its_encode_vpt_size(cmd, LPI_NRBITS - 1); + + its_fixup_cmd(cmd); + + return desc->its_vmapp_cmd.vpe; +} + +static struct its_vpe *its_build_vmapti_cmd(struct its_cmd_block *cmd, + struct its_cmd_desc *desc) +{ + u32 db; + + if (desc->its_vmapti_cmd.db_enabled) + db = desc->its_vmapti_cmd.vpe->vpe_db_lpi; + else + db = 1023; + + its_encode_cmd(cmd, GITS_CMD_VMAPTI); + its_encode_devid(cmd, desc->its_vmapti_cmd.dev->device_id); + its_encode_vpeid(cmd, desc->its_vmapti_cmd.vpe->vpe_id); + its_encode_event_id(cmd, desc->its_vmapti_cmd.event_id); + its_encode_db_phys_id(cmd, db); + its_encode_virt_id(cmd, desc->its_vmapti_cmd.virt_id); + + its_fixup_cmd(cmd); + + return desc->its_vmapti_cmd.vpe; +} + +static struct its_vpe *its_build_vmovi_cmd(struct its_cmd_block *cmd, + struct its_cmd_desc *desc) +{ + u32 db; + + if (desc->its_vmovi_cmd.db_enabled) + db = desc->its_vmovi_cmd.vpe->vpe_db_lpi; + else + db = 1023; + + its_encode_cmd(cmd, GITS_CMD_VMOVI); + its_encode_devid(cmd, desc->its_vmovi_cmd.dev->device_id); + its_encode_vpeid(cmd, desc->its_vmovi_cmd.vpe->vpe_id); + its_encode_event_id(cmd, desc->its_vmovi_cmd.event_id); + its_encode_db_phys_id(cmd, db); + its_encode_db_valid(cmd, true); + + its_fixup_cmd(cmd); + + return desc->its_vmovi_cmd.vpe; +} + +static struct its_vpe *its_build_vmovp_cmd(struct its_cmd_block *cmd, + struct its_cmd_desc *desc) +{ + its_encode_cmd(cmd, GITS_CMD_VMOVP); + its_encode_seq_num(cmd, desc->its_vmovp_cmd.seq_num); + its_encode_its_list(cmd, desc->its_vmovp_cmd.its_list); + its_encode_vpeid(cmd, desc->its_vmovp_cmd.vpe->vpe_id); + its_encode_target(cmd, desc->its_vmovp_cmd.col->target_address); + + its_fixup_cmd(cmd); + + return desc->its_vmovp_cmd.vpe; +} + static u64 its_cmd_ptr_to_offset(struct its_node *its, struct its_cmd_block *ptr) { @@ -453,7 +696,13 @@ static void its_wait_for_range_completion(struct its_node *its, while (1) { rd_idx = readl_relaxed(its->base + GITS_CREADR); - if (rd_idx >= to_idx || rd_idx < from_idx) + + /* Direct case */ + if (from_idx < to_idx && rd_idx >= to_idx) + break; + + /* Wrapped case */ + if (from_idx >= to_idx && rd_idx >= to_idx && rd_idx < from_idx) break; count--; @@ -466,42 +715,84 @@ static void its_wait_for_range_completion(struct its_node *its, } } -static void its_send_single_command(struct its_node *its, - its_cmd_builder_t builder, - struct its_cmd_desc *desc) +/* Warning, macro hell follows */ +#define BUILD_SINGLE_CMD_FUNC(name, buildtype, synctype, buildfn) \ +void name(struct its_node *its, \ + buildtype builder, \ + struct its_cmd_desc *desc) \ +{ \ + struct its_cmd_block *cmd, *sync_cmd, *next_cmd; \ + synctype *sync_obj; \ + unsigned long flags; \ + \ + raw_spin_lock_irqsave(&its->lock, flags); \ + \ + cmd = its_allocate_entry(its); \ + if (!cmd) { /* We're soooooo screewed... */ \ + raw_spin_unlock_irqrestore(&its->lock, flags); \ + return; \ + } \ + sync_obj = builder(cmd, desc); \ + its_flush_cmd(its, cmd); \ + \ + if (sync_obj) { \ + sync_cmd = its_allocate_entry(its); \ + if (!sync_cmd) \ + goto post; \ + \ + buildfn(sync_cmd, sync_obj); \ + its_flush_cmd(its, sync_cmd); \ + } \ + \ +post: \ + next_cmd = its_post_commands(its); \ + raw_spin_unlock_irqrestore(&its->lock, flags); \ + \ + its_wait_for_range_completion(its, cmd, next_cmd); \ +} + +static void its_build_sync_cmd(struct its_cmd_block *sync_cmd, + struct its_collection *sync_col) +{ + its_encode_cmd(sync_cmd, GITS_CMD_SYNC); + its_encode_target(sync_cmd, sync_col->target_address); + + its_fixup_cmd(sync_cmd); +} + +static BUILD_SINGLE_CMD_FUNC(its_send_single_command, its_cmd_builder_t, + struct its_collection, its_build_sync_cmd) + +static void its_build_vsync_cmd(struct its_cmd_block *sync_cmd, + struct its_vpe *sync_vpe) +{ + its_encode_cmd(sync_cmd, GITS_CMD_VSYNC); + its_encode_vpeid(sync_cmd, sync_vpe->vpe_id); + + its_fixup_cmd(sync_cmd); +} + +static BUILD_SINGLE_CMD_FUNC(its_send_single_vcommand, its_cmd_vbuilder_t, + struct its_vpe, its_build_vsync_cmd) + +static void its_send_int(struct its_device *dev, u32 event_id) { - struct its_cmd_block *cmd, *sync_cmd, *next_cmd; - struct its_collection *sync_col; - unsigned long flags; + struct its_cmd_desc desc; - raw_spin_lock_irqsave(&its->lock, flags); + desc.its_int_cmd.dev = dev; + desc.its_int_cmd.event_id = event_id; - cmd = its_allocate_entry(its); - if (!cmd) { /* We're soooooo screewed... */ - pr_err_ratelimited("ITS can't allocate, dropping command\n"); - raw_spin_unlock_irqrestore(&its->lock, flags); - return; - } - sync_col = builder(cmd, desc); - its_flush_cmd(its, cmd); + its_send_single_command(dev->its, its_build_int_cmd, &desc); +} - if (sync_col) { - sync_cmd = its_allocate_entry(its); - if (!sync_cmd) { - pr_err_ratelimited("ITS can't SYNC, skipping\n"); - goto post; - } - its_encode_cmd(sync_cmd, GITS_CMD_SYNC); - its_encode_target(sync_cmd, sync_col->target_address); - its_fixup_cmd(sync_cmd); - its_flush_cmd(its, sync_cmd); - } +static void its_send_clear(struct its_device *dev, u32 event_id) +{ + struct its_cmd_desc desc; -post: - next_cmd = its_post_commands(its); - raw_spin_unlock_irqrestore(&its->lock, flags); + desc.its_clear_cmd.dev = dev; + desc.its_clear_cmd.event_id = event_id; - its_wait_for_range_completion(its, cmd, next_cmd); + its_send_single_command(dev->its, its_build_clear_cmd, &desc); } static void its_send_inv(struct its_device *dev, u32 event_id) @@ -577,6 +868,106 @@ static void its_send_invall(struct its_node *its, struct its_collection *col) its_send_single_command(its, its_build_invall_cmd, &desc); } +static void its_send_vmapti(struct its_device *dev, u32 id) +{ + struct its_vlpi_map *map = &dev->event_map.vlpi_maps[id]; + struct its_cmd_desc desc; + + desc.its_vmapti_cmd.vpe = map->vpe; + desc.its_vmapti_cmd.dev = dev; + desc.its_vmapti_cmd.virt_id = map->vintid; + desc.its_vmapti_cmd.event_id = id; + desc.its_vmapti_cmd.db_enabled = map->db_enabled; + + its_send_single_vcommand(dev->its, its_build_vmapti_cmd, &desc); +} + +static void its_send_vmovi(struct its_device *dev, u32 id) +{ + struct its_vlpi_map *map = &dev->event_map.vlpi_maps[id]; + struct its_cmd_desc desc; + + desc.its_vmovi_cmd.vpe = map->vpe; + desc.its_vmovi_cmd.dev = dev; + desc.its_vmovi_cmd.event_id = id; + desc.its_vmovi_cmd.db_enabled = map->db_enabled; + + its_send_single_vcommand(dev->its, its_build_vmovi_cmd, &desc); +} + +static void its_send_vmapp(struct its_vpe *vpe, bool valid) +{ + struct its_cmd_desc desc; + struct its_node *its; + + desc.its_vmapp_cmd.vpe = vpe; + desc.its_vmapp_cmd.valid = valid; + + list_for_each_entry(its, &its_nodes, entry) { + if (!its->is_v4) + continue; + + desc.its_vmapp_cmd.col = &its->collections[vpe->col_idx]; + its_send_single_vcommand(its, its_build_vmapp_cmd, &desc); + } +} + +static void its_send_vmovp(struct its_vpe *vpe) +{ + struct its_cmd_desc desc; + struct its_node *its; + unsigned long flags; + int col_id = vpe->col_idx; + + desc.its_vmovp_cmd.vpe = vpe; + desc.its_vmovp_cmd.its_list = (u16)its_list_map; + + if (!its_list_map) { + its = list_first_entry(&its_nodes, struct its_node, entry); + desc.its_vmovp_cmd.seq_num = 0; + desc.its_vmovp_cmd.col = &its->collections[col_id]; + its_send_single_vcommand(its, its_build_vmovp_cmd, &desc); + return; + } + + /* + * Yet another marvel of the architecture. If using the + * its_list "feature", we need to make sure that all ITSs + * receive all VMOVP commands in the same order. The only way + * to guarantee this is to make vmovp a serialization point. + * + * Wall <-- Head. + */ + raw_spin_lock_irqsave(&vmovp_lock, flags); + + desc.its_vmovp_cmd.seq_num = vmovp_seq_num++; + + /* Emit VMOVPs */ + list_for_each_entry(its, &its_nodes, entry) { + if (!its->is_v4) + continue; + + desc.its_vmovp_cmd.col = &its->collections[col_id]; + its_send_single_vcommand(its, its_build_vmovp_cmd, &desc); + } + + raw_spin_unlock_irqrestore(&vmovp_lock, flags); +} + +static void its_send_vinvall(struct its_vpe *vpe) +{ + struct its_cmd_desc desc; + struct its_node *its; + + desc.its_vinvall_cmd.vpe = vpe; + + list_for_each_entry(its, &its_nodes, entry) { + if (!its->is_v4) + continue; + its_send_single_vcommand(its, its_build_vinvall_cmd, &desc); + } +} + /* * irqchip functions - assumes MSI, mostly. */ @@ -587,17 +978,26 @@ static inline u32 its_get_event_id(struct irq_data *d) return d->hwirq - its_dev->event_map.lpi_base; } -static void lpi_set_config(struct irq_data *d, bool enable) +static void lpi_write_config(struct irq_data *d, u8 clr, u8 set) { - struct its_device *its_dev = irq_data_get_irq_chip_data(d); - irq_hw_number_t hwirq = d->hwirq; - u32 id = its_get_event_id(d); - u8 *cfg = page_address(gic_rdists->prop_page) + hwirq - 8192; + irq_hw_number_t hwirq; + struct page *prop_page; + u8 *cfg; - if (enable) - *cfg |= LPI_PROP_ENABLED; - else - *cfg &= ~LPI_PROP_ENABLED; + if (irqd_is_forwarded_to_vcpu(d)) { + struct its_device *its_dev = irq_data_get_irq_chip_data(d); + u32 event = its_get_event_id(d); + + prop_page = its_dev->event_map.vm->vprop_page; + hwirq = its_dev->event_map.vlpi_maps[event].vintid; + } else { + prop_page = gic_rdists->prop_page; + hwirq = d->hwirq; + } + + cfg = page_address(prop_page) + hwirq - 8192; + *cfg &= ~clr; + *cfg |= set | LPI_PROP_GROUP1; /* * Make the above write visible to the redistributors. @@ -608,17 +1008,53 @@ static void lpi_set_config(struct irq_data *d, bool enable) gic_flush_dcache_to_poc(cfg, sizeof(*cfg)); else dsb(ishst); - its_send_inv(its_dev, id); +} + +static void lpi_update_config(struct irq_data *d, u8 clr, u8 set) +{ + struct its_device *its_dev = irq_data_get_irq_chip_data(d); + + lpi_write_config(d, clr, set); + its_send_inv(its_dev, its_get_event_id(d)); +} + +static void its_vlpi_set_doorbell(struct irq_data *d, bool enable) +{ + struct its_device *its_dev = irq_data_get_irq_chip_data(d); + u32 event = its_get_event_id(d); + + if (its_dev->event_map.vlpi_maps[event].db_enabled == enable) + return; + + its_dev->event_map.vlpi_maps[event].db_enabled = enable; + + /* + * More fun with the architecture: + * + * Ideally, we'd issue a VMAPTI to set the doorbell to its LPI + * value or to 1023, depending on the enable bit. But that + * would be issueing a mapping for an /existing/ DevID+EventID + * pair, which is UNPREDICTABLE. Instead, let's issue a VMOVI + * to the /same/ vPE, using this opportunity to adjust the + * doorbell. Mouahahahaha. We loves it, Precious. + */ + its_send_vmovi(its_dev, event); } static void its_mask_irq(struct irq_data *d) { - lpi_set_config(d, false); + if (irqd_is_forwarded_to_vcpu(d)) + its_vlpi_set_doorbell(d, false); + + lpi_update_config(d, LPI_PROP_ENABLED, 0); } static void its_unmask_irq(struct irq_data *d) { - lpi_set_config(d, true); + if (irqd_is_forwarded_to_vcpu(d)) + its_vlpi_set_doorbell(d, true); + + lpi_update_config(d, 0, LPI_PROP_ENABLED); } static int its_set_affinity(struct irq_data *d, const struct cpumask *mask_val, @@ -630,6 +1066,10 @@ static int its_set_affinity(struct irq_data *d, const struct cpumask *mask_val, struct its_collection *target_col; u32 id = its_get_event_id(d); + /* A forwarded interrupt should use irq_set_vcpu_affinity */ + if (irqd_is_forwarded_to_vcpu(d)) + return -EINVAL; + /* lpi cannot be routed to a redistributor that is on a foreign node */ if (its_dev->its->flags & ITS_FLAGS_WORKAROUND_CAVIUM_23144) { if (its_dev->its->numa_node >= 0) { @@ -649,6 +1089,7 @@ static int its_set_affinity(struct irq_data *d, const struct cpumask *mask_val, target_col = &its_dev->its->collections[cpu]; its_send_movi(its_dev, target_col, id); its_dev->event_map.col_map[id] = cpu; + irq_data_update_effective_affinity(d, cpumask_of(cpu)); } return IRQ_SET_MASK_OK_DONE; @@ -670,6 +1111,179 @@ static void its_irq_compose_msi_msg(struct irq_data *d, struct msi_msg *msg) iommu_dma_map_msi_msg(d->irq, msg); } +static int its_irq_set_irqchip_state(struct irq_data *d, + enum irqchip_irq_state which, + bool state) +{ + struct its_device *its_dev = irq_data_get_irq_chip_data(d); + u32 event = its_get_event_id(d); + + if (which != IRQCHIP_STATE_PENDING) + return -EINVAL; + + if (state) + its_send_int(its_dev, event); + else + its_send_clear(its_dev, event); + + return 0; +} + +static int its_vlpi_map(struct irq_data *d, struct its_cmd_info *info) +{ + struct its_device *its_dev = irq_data_get_irq_chip_data(d); + u32 event = its_get_event_id(d); + int ret = 0; + + if (!info->map) + return -EINVAL; + + mutex_lock(&its_dev->event_map.vlpi_lock); + + if (!its_dev->event_map.vm) { + struct its_vlpi_map *maps; + + maps = kzalloc(sizeof(*maps) * its_dev->event_map.nr_lpis, + GFP_KERNEL); + if (!maps) { + ret = -ENOMEM; + goto out; + } + + its_dev->event_map.vm = info->map->vm; + its_dev->event_map.vlpi_maps = maps; + } else if (its_dev->event_map.vm != info->map->vm) { + ret = -EINVAL; + goto out; + } + + /* Get our private copy of the mapping information */ + its_dev->event_map.vlpi_maps[event] = *info->map; + + if (irqd_is_forwarded_to_vcpu(d)) { + /* Already mapped, move it around */ + its_send_vmovi(its_dev, event); + } else { + /* Drop the physical mapping */ + its_send_discard(its_dev, event); + + /* and install the virtual one */ + its_send_vmapti(its_dev, event); + irqd_set_forwarded_to_vcpu(d); + + /* Increment the number of VLPIs */ + its_dev->event_map.nr_vlpis++; + } + +out: + mutex_unlock(&its_dev->event_map.vlpi_lock); + return ret; +} + +static int its_vlpi_get(struct irq_data *d, struct its_cmd_info *info) +{ + struct its_device *its_dev = irq_data_get_irq_chip_data(d); + u32 event = its_get_event_id(d); + int ret = 0; + + mutex_lock(&its_dev->event_map.vlpi_lock); + + if (!its_dev->event_map.vm || + !its_dev->event_map.vlpi_maps[event].vm) { + ret = -EINVAL; + goto out; + } + + /* Copy our mapping information to the incoming request */ + *info->map = its_dev->event_map.vlpi_maps[event]; + +out: + mutex_unlock(&its_dev->event_map.vlpi_lock); + return ret; +} + +static int its_vlpi_unmap(struct irq_data *d) +{ + struct its_device *its_dev = irq_data_get_irq_chip_data(d); + u32 event = its_get_event_id(d); + int ret = 0; + + mutex_lock(&its_dev->event_map.vlpi_lock); + + if (!its_dev->event_map.vm || !irqd_is_forwarded_to_vcpu(d)) { + ret = -EINVAL; + goto out; + } + + /* Drop the virtual mapping */ + its_send_discard(its_dev, event); + + /* and restore the physical one */ + irqd_clr_forwarded_to_vcpu(d); + its_send_mapti(its_dev, d->hwirq, event); + lpi_update_config(d, 0xff, (LPI_PROP_DEFAULT_PRIO | + LPI_PROP_ENABLED | + LPI_PROP_GROUP1)); + + /* + * Drop the refcount and make the device available again if + * this was the last VLPI. + */ + if (!--its_dev->event_map.nr_vlpis) { + its_dev->event_map.vm = NULL; + kfree(its_dev->event_map.vlpi_maps); + } + +out: + mutex_unlock(&its_dev->event_map.vlpi_lock); + return ret; +} + +static int its_vlpi_prop_update(struct irq_data *d, struct its_cmd_info *info) +{ + struct its_device *its_dev = irq_data_get_irq_chip_data(d); + + if (!its_dev->event_map.vm || !irqd_is_forwarded_to_vcpu(d)) + return -EINVAL; + + if (info->cmd_type == PROP_UPDATE_AND_INV_VLPI) + lpi_update_config(d, 0xff, info->config); + else + lpi_write_config(d, 0xff, info->config); + its_vlpi_set_doorbell(d, !!(info->config & LPI_PROP_ENABLED)); + + return 0; +} + +static int its_irq_set_vcpu_affinity(struct irq_data *d, void *vcpu_info) +{ + struct its_device *its_dev = irq_data_get_irq_chip_data(d); + struct its_cmd_info *info = vcpu_info; + + /* Need a v4 ITS */ + if (!its_dev->its->is_v4) + return -EINVAL; + + /* Unmap request? */ + if (!info) + return its_vlpi_unmap(d); + + switch (info->cmd_type) { + case MAP_VLPI: + return its_vlpi_map(d, info); + + case GET_VLPI: + return its_vlpi_get(d, info); + + case PROP_UPDATE_VLPI: + case PROP_UPDATE_AND_INV_VLPI: + return its_vlpi_prop_update(d, info); + + default: + return -EINVAL; + } +} + static struct irq_chip its_irq_chip = { .name = "ITS", .irq_mask = its_mask_irq, @@ -677,6 +1291,8 @@ static struct irq_chip its_irq_chip = { .irq_eoi = irq_chip_eoi_parent, .irq_set_affinity = its_set_affinity, .irq_compose_msi_msg = its_irq_compose_msi_msg, + .irq_set_irqchip_state = its_irq_set_irqchip_state, + .irq_set_vcpu_affinity = its_irq_set_vcpu_affinity, }; /* @@ -695,7 +1311,6 @@ static struct irq_chip its_irq_chip = { static unsigned long *lpi_bitmap; static u32 lpi_chunks; -static u32 lpi_id_bits; static DEFINE_SPINLOCK(lpi_lock); static int its_lpi_to_chunk(int lpi) @@ -766,16 +1381,15 @@ out: return bitmap; } -static void its_lpi_free(struct event_lpi_map *map) +static void its_lpi_free_chunks(unsigned long *bitmap, int base, int nr_ids) { - int base = map->lpi_base; - int nr_ids = map->nr_lpis; int lpi; spin_lock(&lpi_lock); for (lpi = base; lpi < (base + nr_ids); lpi += IRQS_PER_CHUNK) { int chunk = its_lpi_to_chunk(lpi); + BUG_ON(chunk > lpi_chunks); if (test_bit(chunk, lpi_bitmap)) { clear_bit(chunk, lpi_bitmap); @@ -786,28 +1400,40 @@ static void its_lpi_free(struct event_lpi_map *map) spin_unlock(&lpi_lock); - kfree(map->lpi_map); - kfree(map->col_map); + kfree(bitmap); } -/* - * We allocate memory for PROPBASE to cover 2 ^ lpi_id_bits LPIs to - * deal with (one configuration byte per interrupt). PENDBASE has to - * be 64kB aligned (one bit per LPI, plus 8192 bits for SPI/PPI/SGI). - */ -#define LPI_NRBITS lpi_id_bits -#define LPI_PROPBASE_SZ ALIGN(BIT(LPI_NRBITS), SZ_64K) -#define LPI_PENDBASE_SZ ALIGN(BIT(LPI_NRBITS) / 8, SZ_64K) +static struct page *its_allocate_prop_table(gfp_t gfp_flags) +{ + struct page *prop_page; -#define LPI_PROP_DEFAULT_PRIO 0xa0 + prop_page = alloc_pages(gfp_flags, get_order(LPI_PROPBASE_SZ)); + if (!prop_page) + return NULL; + + /* Priority 0xa0, Group-1, disabled */ + memset(page_address(prop_page), + LPI_PROP_DEFAULT_PRIO | LPI_PROP_GROUP1, + LPI_PROPBASE_SZ); + + /* Make sure the GIC will observe the written configuration */ + gic_flush_dcache_to_poc(page_address(prop_page), LPI_PROPBASE_SZ); + + return prop_page; +} + +static void its_free_prop_table(struct page *prop_page) +{ + free_pages((unsigned long)page_address(prop_page), + get_order(LPI_PROPBASE_SZ)); +} static int __init its_alloc_lpi_tables(void) { phys_addr_t paddr; lpi_id_bits = min_t(u32, gic_rdists->id_bits, ITS_MAX_LPI_NRBITS); - gic_rdists->prop_page = alloc_pages(GFP_NOWAIT, - get_order(LPI_PROPBASE_SZ)); + gic_rdists->prop_page = its_allocate_prop_table(GFP_NOWAIT); if (!gic_rdists->prop_page) { pr_err("Failed to allocate PROPBASE\n"); return -ENOMEM; @@ -816,14 +1442,6 @@ static int __init its_alloc_lpi_tables(void) paddr = page_to_phys(gic_rdists->prop_page); pr_info("GIC: using LPI property table @%pa\n", &paddr); - /* Priority 0xa0, Group-1, disabled */ - memset(page_address(gic_rdists->prop_page), - LPI_PROP_DEFAULT_PRIO | LPI_PROP_GROUP1, - LPI_PROPBASE_SZ); - - /* Make sure the GIC will observe the written configuration */ - gic_flush_dcache_to_poc(page_address(gic_rdists->prop_page), LPI_PROPBASE_SZ); - return its_lpi_init(lpi_id_bits); } @@ -962,10 +1580,13 @@ retry_baser: return 0; } -static bool its_parse_baser_device(struct its_node *its, struct its_baser *baser, - u32 psz, u32 *order) +static bool its_parse_indirect_baser(struct its_node *its, + struct its_baser *baser, + u32 psz, u32 *order) { - u64 esz = GITS_BASER_ENTRY_SIZE(its_read_baser(its, baser)); + u64 tmp = its_read_baser(its, baser); + u64 type = GITS_BASER_TYPE(tmp); + u64 esz = GITS_BASER_ENTRY_SIZE(tmp); u64 val = GITS_BASER_InnerShareable | GITS_BASER_RaWaWb; u32 ids = its->device_ids; u32 new_order = *order; @@ -1004,8 +1625,9 @@ static bool its_parse_baser_device(struct its_node *its, struct its_baser *baser if (new_order >= MAX_ORDER) { new_order = MAX_ORDER - 1; ids = ilog2(PAGE_ORDER_TO_SIZE(new_order) / (int)esz); - pr_warn("ITS@%pa: Device Table too large, reduce ids %u->%u\n", - &its->phys_base, its->device_ids, ids); + pr_warn("ITS@%pa: %s Table too large, reduce ids %u->%u\n", + &its->phys_base, its_base_type_string[type], + its->device_ids, ids); } *order = new_order; @@ -1053,11 +1675,16 @@ static int its_alloc_tables(struct its_node *its) u32 order = get_order(psz); bool indirect = false; - if (type == GITS_BASER_TYPE_NONE) + switch (type) { + case GITS_BASER_TYPE_NONE: continue; - if (type == GITS_BASER_TYPE_DEVICE) - indirect = its_parse_baser_device(its, baser, psz, &order); + case GITS_BASER_TYPE_DEVICE: + case GITS_BASER_TYPE_VCPU: + indirect = its_parse_indirect_baser(its, baser, + psz, &order); + break; + } err = its_setup_baser(its, baser, cache, shr, psz, order, indirect); if (err < 0) { @@ -1084,6 +1711,30 @@ static int its_alloc_collections(struct its_node *its) return 0; } +static struct page *its_allocate_pending_table(gfp_t gfp_flags) +{ + struct page *pend_page; + /* + * The pending pages have to be at least 64kB aligned, + * hence the 'max(LPI_PENDBASE_SZ, SZ_64K)' below. + */ + pend_page = alloc_pages(gfp_flags | __GFP_ZERO, + get_order(max_t(u32, LPI_PENDBASE_SZ, SZ_64K))); + if (!pend_page) + return NULL; + + /* Make sure the GIC will observe the zero-ed page */ + gic_flush_dcache_to_poc(page_address(pend_page), LPI_PENDBASE_SZ); + + return pend_page; +} + +static void its_free_pending_table(struct page *pt) +{ + free_pages((unsigned long)page_address(pt), + get_order(max_t(u32, LPI_PENDBASE_SZ, SZ_64K))); +} + static void its_cpu_init_lpis(void) { void __iomem *rbase = gic_data_rdist_rd_base(); @@ -1094,21 +1745,14 @@ static void its_cpu_init_lpis(void) pend_page = gic_data_rdist()->pend_page; if (!pend_page) { phys_addr_t paddr; - /* - * The pending pages have to be at least 64kB aligned, - * hence the 'max(LPI_PENDBASE_SZ, SZ_64K)' below. - */ - pend_page = alloc_pages(GFP_NOWAIT | __GFP_ZERO, - get_order(max_t(u32, LPI_PENDBASE_SZ, SZ_64K))); + + pend_page = its_allocate_pending_table(GFP_NOWAIT); if (!pend_page) { pr_err("Failed to allocate PENDBASE for CPU%d\n", smp_processor_id()); return; } - /* Make sure the GIC will observe the zero-ed page */ - gic_flush_dcache_to_poc(page_address(pend_page), LPI_PENDBASE_SZ); - paddr = page_to_phys(pend_page); pr_info("CPU%d: using LPI pending table @%pa\n", smp_processor_id(), &paddr); @@ -1259,26 +1903,19 @@ static struct its_baser *its_get_baser(struct its_node *its, u32 type) return NULL; } -static bool its_alloc_device_table(struct its_node *its, u32 dev_id) +static bool its_alloc_table_entry(struct its_baser *baser, u32 id) { - struct its_baser *baser; struct page *page; u32 esz, idx; __le64 *table; - baser = its_get_baser(its, GITS_BASER_TYPE_DEVICE); - - /* Don't allow device id that exceeds ITS hardware limit */ - if (!baser) - return (ilog2(dev_id) < its->device_ids); - /* Don't allow device id that exceeds single, flat table limit */ esz = GITS_BASER_ENTRY_SIZE(baser->val); if (!(baser->val & GITS_BASER_INDIRECT)) - return (dev_id < (PAGE_ORDER_TO_SIZE(baser->order) / esz)); + return (id < (PAGE_ORDER_TO_SIZE(baser->order) / esz)); /* Compute 1st level table index & check if that exceeds table limit */ - idx = dev_id >> ilog2(baser->psz / esz); + idx = id >> ilog2(baser->psz / esz); if (idx >= (PAGE_ORDER_TO_SIZE(baser->order) / GITS_LVL1_ENTRY_SIZE)) return false; @@ -1307,11 +1944,52 @@ static bool its_alloc_device_table(struct its_node *its, u32 dev_id) return true; } +static bool its_alloc_device_table(struct its_node *its, u32 dev_id) +{ + struct its_baser *baser; + + baser = its_get_baser(its, GITS_BASER_TYPE_DEVICE); + + /* Don't allow device id that exceeds ITS hardware limit */ + if (!baser) + return (ilog2(dev_id) < its->device_ids); + + return its_alloc_table_entry(baser, dev_id); +} + +static bool its_alloc_vpe_table(u32 vpe_id) +{ + struct its_node *its; + + /* + * Make sure the L2 tables are allocated on *all* v4 ITSs. We + * could try and only do it on ITSs corresponding to devices + * that have interrupts targeted at this VPE, but the + * complexity becomes crazy (and you have tons of memory + * anyway, right?). + */ + list_for_each_entry(its, &its_nodes, entry) { + struct its_baser *baser; + + if (!its->is_v4) + continue; + + baser = its_get_baser(its, GITS_BASER_TYPE_VCPU); + if (!baser) + return false; + + if (!its_alloc_table_entry(baser, vpe_id)) + return false; + } + + return true; +} + static struct its_device *its_create_device(struct its_node *its, u32 dev_id, - int nvecs) + int nvecs, bool alloc_lpis) { struct its_device *dev; - unsigned long *lpi_map; + unsigned long *lpi_map = NULL; unsigned long flags; u16 *col_map = NULL; void *itt; @@ -1333,11 +2011,18 @@ static struct its_device *its_create_device(struct its_node *its, u32 dev_id, sz = nr_ites * its->ite_size; sz = max(sz, ITS_ITT_ALIGN) + ITS_ITT_ALIGN - 1; itt = kzalloc(sz, GFP_KERNEL); - lpi_map = its_lpi_alloc_chunks(nvecs, &lpi_base, &nr_lpis); - if (lpi_map) - col_map = kzalloc(sizeof(*col_map) * nr_lpis, GFP_KERNEL); + if (alloc_lpis) { + lpi_map = its_lpi_alloc_chunks(nvecs, &lpi_base, &nr_lpis); + if (lpi_map) + col_map = kzalloc(sizeof(*col_map) * nr_lpis, + GFP_KERNEL); + } else { + col_map = kzalloc(sizeof(*col_map) * nr_ites, GFP_KERNEL); + nr_lpis = 0; + lpi_base = 0; + } - if (!dev || !itt || !lpi_map || !col_map) { + if (!dev || !itt || !col_map || (!lpi_map && alloc_lpis)) { kfree(dev); kfree(itt); kfree(lpi_map); @@ -1354,6 +2039,7 @@ static struct its_device *its_create_device(struct its_node *its, u32 dev_id, dev->event_map.col_map = col_map; dev->event_map.lpi_base = lpi_base; dev->event_map.nr_lpis = nr_lpis; + mutex_init(&dev->event_map.vlpi_lock); dev->device_id = dev_id; INIT_LIST_HEAD(&dev->entry); @@ -1412,6 +2098,16 @@ static int its_msi_prepare(struct irq_domain *domain, struct device *dev, msi_info = msi_get_domain_info(domain); its = msi_info->data; + if (!gic_rdists->has_direct_lpi && + vpe_proxy.dev && + vpe_proxy.dev->its == its && + dev_id == vpe_proxy.dev->device_id) { + /* Bad luck. Get yourself a better implementation */ + WARN_ONCE(1, "DevId %x clashes with GICv4 VPE proxy device\n", + dev_id); + return -EINVAL; + } + its_dev = its_find_device(its, dev_id); if (its_dev) { /* @@ -1423,7 +2119,7 @@ static int its_msi_prepare(struct irq_domain *domain, struct device *dev, goto out; } - its_dev = its_create_device(its, dev_id, nvec); + its_dev = its_create_device(its, dev_id, nvec, true); if (!its_dev) return -ENOMEM; @@ -1481,6 +2177,7 @@ static int its_irq_domain_alloc(struct irq_domain *domain, unsigned int virq, irq_domain_set_hwirq_and_chip(domain, virq + i, hwirq, &its_irq_chip, its_dev); + irqd_set_single_target(irq_desc_get_irq_data(irq_to_desc(virq + i))); pr_debug("ID:%d pID:%d vID:%d\n", (int)(hwirq - its_dev->event_map.lpi_base), (int) hwirq, virq + i); @@ -1495,13 +2192,16 @@ static void its_irq_domain_activate(struct irq_domain *domain, struct its_device *its_dev = irq_data_get_irq_chip_data(d); u32 event = its_get_event_id(d); const struct cpumask *cpu_mask = cpu_online_mask; + int cpu; /* get the cpu_mask of local node */ if (its_dev->its->numa_node >= 0) cpu_mask = cpumask_of_node(its_dev->its->numa_node); /* Bind the LPI to the first possible CPU */ - its_dev->event_map.col_map[event] = cpumask_first(cpu_mask); + cpu = cpumask_first(cpu_mask); + its_dev->event_map.col_map[event] = cpu; + irq_data_update_effective_affinity(d, cpumask_of(cpu)); /* Map the GIC IRQ and event to the device */ its_send_mapti(its_dev, d->hwirq, event); @@ -1539,7 +2239,10 @@ static void its_irq_domain_free(struct irq_domain *domain, unsigned int virq, /* If all interrupts have been freed, start mopping the floor */ if (bitmap_empty(its_dev->event_map.lpi_map, its_dev->event_map.nr_lpis)) { - its_lpi_free(&its_dev->event_map); + its_lpi_free_chunks(its_dev->event_map.lpi_map, + its_dev->event_map.lpi_base, + its_dev->event_map.nr_lpis); + kfree(its_dev->event_map.col_map); /* Unmap device/itt */ its_send_mapd(its_dev, 0); @@ -1556,6 +2259,451 @@ static const struct irq_domain_ops its_domain_ops = { .deactivate = its_irq_domain_deactivate, }; +/* + * This is insane. + * + * If a GICv4 doesn't implement Direct LPIs (which is extremely + * likely), the only way to perform an invalidate is to use a fake + * device to issue an INV command, implying that the LPI has first + * been mapped to some event on that device. Since this is not exactly + * cheap, we try to keep that mapping around as long as possible, and + * only issue an UNMAP if we're short on available slots. + * + * Broken by design(tm). + */ +static void its_vpe_db_proxy_unmap_locked(struct its_vpe *vpe) +{ + /* Already unmapped? */ + if (vpe->vpe_proxy_event == -1) + return; + + its_send_discard(vpe_proxy.dev, vpe->vpe_proxy_event); + vpe_proxy.vpes[vpe->vpe_proxy_event] = NULL; + + /* + * We don't track empty slots at all, so let's move the + * next_victim pointer if we can quickly reuse that slot + * instead of nuking an existing entry. Not clear that this is + * always a win though, and this might just generate a ripple + * effect... Let's just hope VPEs don't migrate too often. + */ + if (vpe_proxy.vpes[vpe_proxy.next_victim]) + vpe_proxy.next_victim = vpe->vpe_proxy_event; + + vpe->vpe_proxy_event = -1; +} + +static void its_vpe_db_proxy_unmap(struct its_vpe *vpe) +{ + if (!gic_rdists->has_direct_lpi) { + unsigned long flags; + + raw_spin_lock_irqsave(&vpe_proxy.lock, flags); + its_vpe_db_proxy_unmap_locked(vpe); + raw_spin_unlock_irqrestore(&vpe_proxy.lock, flags); + } +} + +static void its_vpe_db_proxy_map_locked(struct its_vpe *vpe) +{ + /* Already mapped? */ + if (vpe->vpe_proxy_event != -1) + return; + + /* This slot was already allocated. Kick the other VPE out. */ + if (vpe_proxy.vpes[vpe_proxy.next_victim]) + its_vpe_db_proxy_unmap_locked(vpe_proxy.vpes[vpe_proxy.next_victim]); + + /* Map the new VPE instead */ + vpe_proxy.vpes[vpe_proxy.next_victim] = vpe; + vpe->vpe_proxy_event = vpe_proxy.next_victim; + vpe_proxy.next_victim = (vpe_proxy.next_victim + 1) % vpe_proxy.dev->nr_ites; + + vpe_proxy.dev->event_map.col_map[vpe->vpe_proxy_event] = vpe->col_idx; + its_send_mapti(vpe_proxy.dev, vpe->vpe_db_lpi, vpe->vpe_proxy_event); +} + +static void its_vpe_db_proxy_move(struct its_vpe *vpe, int from, int to) +{ + unsigned long flags; + struct its_collection *target_col; + + if (gic_rdists->has_direct_lpi) { + void __iomem *rdbase; + + rdbase = per_cpu_ptr(gic_rdists->rdist, from)->rd_base; + gic_write_lpir(vpe->vpe_db_lpi, rdbase + GICR_CLRLPIR); + while (gic_read_lpir(rdbase + GICR_SYNCR) & 1) + cpu_relax(); + + return; + } + + raw_spin_lock_irqsave(&vpe_proxy.lock, flags); + + its_vpe_db_proxy_map_locked(vpe); + + target_col = &vpe_proxy.dev->its->collections[to]; + its_send_movi(vpe_proxy.dev, target_col, vpe->vpe_proxy_event); + vpe_proxy.dev->event_map.col_map[vpe->vpe_proxy_event] = to; + + raw_spin_unlock_irqrestore(&vpe_proxy.lock, flags); +} + +static int its_vpe_set_affinity(struct irq_data *d, + const struct cpumask *mask_val, + bool force) +{ + struct its_vpe *vpe = irq_data_get_irq_chip_data(d); + int cpu = cpumask_first(mask_val); + + /* + * Changing affinity is mega expensive, so let's be as lazy as + * we can and only do it if we really have to. Also, if mapped + * into the proxy device, we need to move the doorbell + * interrupt to its new location. + */ + if (vpe->col_idx != cpu) { + int from = vpe->col_idx; + + vpe->col_idx = cpu; + its_send_vmovp(vpe); + its_vpe_db_proxy_move(vpe, from, cpu); + } + + return IRQ_SET_MASK_OK_DONE; +} + +static void its_vpe_schedule(struct its_vpe *vpe) +{ + void * __iomem vlpi_base = gic_data_rdist_vlpi_base(); + u64 val; + + /* Schedule the VPE */ + val = virt_to_phys(page_address(vpe->its_vm->vprop_page)) & + GENMASK_ULL(51, 12); + val |= (LPI_NRBITS - 1) & GICR_VPROPBASER_IDBITS_MASK; + val |= GICR_VPROPBASER_RaWb; + val |= GICR_VPROPBASER_InnerShareable; + gits_write_vpropbaser(val, vlpi_base + GICR_VPROPBASER); + + val = virt_to_phys(page_address(vpe->vpt_page)) & + GENMASK_ULL(51, 16); + val |= GICR_VPENDBASER_RaWaWb; + val |= GICR_VPENDBASER_NonShareable; + /* + * There is no good way of finding out if the pending table is + * empty as we can race against the doorbell interrupt very + * easily. So in the end, vpe->pending_last is only an + * indication that the vcpu has something pending, not one + * that the pending table is empty. A good implementation + * would be able to read its coarse map pretty quickly anyway, + * making this a tolerable issue. + */ + val |= GICR_VPENDBASER_PendingLast; + val |= vpe->idai ? GICR_VPENDBASER_IDAI : 0; + val |= GICR_VPENDBASER_Valid; + gits_write_vpendbaser(val, vlpi_base + GICR_VPENDBASER); +} + +static void its_vpe_deschedule(struct its_vpe *vpe) +{ + void * __iomem vlpi_base = gic_data_rdist_vlpi_base(); + u32 count = 1000000; /* 1s! */ + bool clean; + u64 val; + + /* We're being scheduled out */ + val = gits_read_vpendbaser(vlpi_base + GICR_VPENDBASER); + val &= ~GICR_VPENDBASER_Valid; + gits_write_vpendbaser(val, vlpi_base + GICR_VPENDBASER); + + do { + val = gits_read_vpendbaser(vlpi_base + GICR_VPENDBASER); + clean = !(val & GICR_VPENDBASER_Dirty); + if (!clean) { + count--; + cpu_relax(); + udelay(1); + } + } while (!clean && count); + + if (unlikely(!clean && !count)) { + pr_err_ratelimited("ITS virtual pending table not cleaning\n"); + vpe->idai = false; + vpe->pending_last = true; + } else { + vpe->idai = !!(val & GICR_VPENDBASER_IDAI); + vpe->pending_last = !!(val & GICR_VPENDBASER_PendingLast); + } +} + +static int its_vpe_set_vcpu_affinity(struct irq_data *d, void *vcpu_info) +{ + struct its_vpe *vpe = irq_data_get_irq_chip_data(d); + struct its_cmd_info *info = vcpu_info; + + switch (info->cmd_type) { + case SCHEDULE_VPE: + its_vpe_schedule(vpe); + return 0; + + case DESCHEDULE_VPE: + its_vpe_deschedule(vpe); + return 0; + + case INVALL_VPE: + its_send_vinvall(vpe); + return 0; + + default: + return -EINVAL; + } +} + +static void its_vpe_send_cmd(struct its_vpe *vpe, + void (*cmd)(struct its_device *, u32)) +{ + unsigned long flags; + + raw_spin_lock_irqsave(&vpe_proxy.lock, flags); + + its_vpe_db_proxy_map_locked(vpe); + cmd(vpe_proxy.dev, vpe->vpe_proxy_event); + + raw_spin_unlock_irqrestore(&vpe_proxy.lock, flags); +} + +static void its_vpe_send_inv(struct irq_data *d) +{ + struct its_vpe *vpe = irq_data_get_irq_chip_data(d); + + if (gic_rdists->has_direct_lpi) { + void __iomem *rdbase; + + rdbase = per_cpu_ptr(gic_rdists->rdist, vpe->col_idx)->rd_base; + gic_write_lpir(vpe->vpe_db_lpi, rdbase + GICR_INVLPIR); + while (gic_read_lpir(rdbase + GICR_SYNCR) & 1) + cpu_relax(); + } else { + its_vpe_send_cmd(vpe, its_send_inv); + } +} + +static void its_vpe_mask_irq(struct irq_data *d) +{ + /* + * We need to unmask the LPI, which is described by the parent + * irq_data. Instead of calling into the parent (which won't + * exactly do the right thing, let's simply use the + * parent_data pointer. Yes, I'm naughty. + */ + lpi_write_config(d->parent_data, LPI_PROP_ENABLED, 0); + its_vpe_send_inv(d); +} + +static void its_vpe_unmask_irq(struct irq_data *d) +{ + /* Same hack as above... */ + lpi_write_config(d->parent_data, 0, LPI_PROP_ENABLED); + its_vpe_send_inv(d); +} + +static int its_vpe_set_irqchip_state(struct irq_data *d, + enum irqchip_irq_state which, + bool state) +{ + struct its_vpe *vpe = irq_data_get_irq_chip_data(d); + + if (which != IRQCHIP_STATE_PENDING) + return -EINVAL; + + if (gic_rdists->has_direct_lpi) { + void __iomem *rdbase; + + rdbase = per_cpu_ptr(gic_rdists->rdist, vpe->col_idx)->rd_base; + if (state) { + gic_write_lpir(vpe->vpe_db_lpi, rdbase + GICR_SETLPIR); + } else { + gic_write_lpir(vpe->vpe_db_lpi, rdbase + GICR_CLRLPIR); + while (gic_read_lpir(rdbase + GICR_SYNCR) & 1) + cpu_relax(); + } + } else { + if (state) + its_vpe_send_cmd(vpe, its_send_int); + else + its_vpe_send_cmd(vpe, its_send_clear); + } + + return 0; +} + +static struct irq_chip its_vpe_irq_chip = { + .name = "GICv4-vpe", + .irq_mask = its_vpe_mask_irq, + .irq_unmask = its_vpe_unmask_irq, + .irq_eoi = irq_chip_eoi_parent, + .irq_set_affinity = its_vpe_set_affinity, + .irq_set_irqchip_state = its_vpe_set_irqchip_state, + .irq_set_vcpu_affinity = its_vpe_set_vcpu_affinity, +}; + +static int its_vpe_id_alloc(void) +{ + return ida_simple_get(&its_vpeid_ida, 0, 1 << 16, GFP_KERNEL); +} + +static void its_vpe_id_free(u16 id) +{ + ida_simple_remove(&its_vpeid_ida, id); +} + +static int its_vpe_init(struct its_vpe *vpe) +{ + struct page *vpt_page; + int vpe_id; + + /* Allocate vpe_id */ + vpe_id = its_vpe_id_alloc(); + if (vpe_id < 0) + return vpe_id; + + /* Allocate VPT */ + vpt_page = its_allocate_pending_table(GFP_KERNEL); + if (!vpt_page) { + its_vpe_id_free(vpe_id); + return -ENOMEM; + } + + if (!its_alloc_vpe_table(vpe_id)) { + its_vpe_id_free(vpe_id); + its_free_pending_table(vpe->vpt_page); + return -ENOMEM; + } + + vpe->vpe_id = vpe_id; + vpe->vpt_page = vpt_page; + vpe->vpe_proxy_event = -1; + + return 0; +} + +static void its_vpe_teardown(struct its_vpe *vpe) +{ + its_vpe_db_proxy_unmap(vpe); + its_vpe_id_free(vpe->vpe_id); + its_free_pending_table(vpe->vpt_page); +} + +static void its_vpe_irq_domain_free(struct irq_domain *domain, + unsigned int virq, + unsigned int nr_irqs) +{ + struct its_vm *vm = domain->host_data; + int i; + + irq_domain_free_irqs_parent(domain, virq, nr_irqs); + + for (i = 0; i < nr_irqs; i++) { + struct irq_data *data = irq_domain_get_irq_data(domain, + virq + i); + struct its_vpe *vpe = irq_data_get_irq_chip_data(data); + + BUG_ON(vm != vpe->its_vm); + + clear_bit(data->hwirq, vm->db_bitmap); + its_vpe_teardown(vpe); + irq_domain_reset_irq_data(data); + } + + if (bitmap_empty(vm->db_bitmap, vm->nr_db_lpis)) { + its_lpi_free_chunks(vm->db_bitmap, vm->db_lpi_base, vm->nr_db_lpis); + its_free_prop_table(vm->vprop_page); + } +} + +static int its_vpe_irq_domain_alloc(struct irq_domain *domain, unsigned int virq, + unsigned int nr_irqs, void *args) +{ + struct its_vm *vm = args; + unsigned long *bitmap; + struct page *vprop_page; + int base, nr_ids, i, err = 0; + + BUG_ON(!vm); + + bitmap = its_lpi_alloc_chunks(nr_irqs, &base, &nr_ids); + if (!bitmap) + return -ENOMEM; + + if (nr_ids < nr_irqs) { + its_lpi_free_chunks(bitmap, base, nr_ids); + return -ENOMEM; + } + + vprop_page = its_allocate_prop_table(GFP_KERNEL); + if (!vprop_page) { + its_lpi_free_chunks(bitmap, base, nr_ids); + return -ENOMEM; + } + + vm->db_bitmap = bitmap; + vm->db_lpi_base = base; + vm->nr_db_lpis = nr_ids; + vm->vprop_page = vprop_page; + + for (i = 0; i < nr_irqs; i++) { + vm->vpes[i]->vpe_db_lpi = base + i; + err = its_vpe_init(vm->vpes[i]); + if (err) + break; + err = its_irq_gic_domain_alloc(domain, virq + i, + vm->vpes[i]->vpe_db_lpi); + if (err) + break; + irq_domain_set_hwirq_and_chip(domain, virq + i, i, + &its_vpe_irq_chip, vm->vpes[i]); + set_bit(i, bitmap); + } + + if (err) { + if (i > 0) + its_vpe_irq_domain_free(domain, virq, i - 1); + + its_lpi_free_chunks(bitmap, base, nr_ids); + its_free_prop_table(vprop_page); + } + + return err; +} + +static void its_vpe_irq_domain_activate(struct irq_domain *domain, + struct irq_data *d) +{ + struct its_vpe *vpe = irq_data_get_irq_chip_data(d); + + /* Map the VPE to the first possible CPU */ + vpe->col_idx = cpumask_first(cpu_online_mask); + its_send_vmapp(vpe, true); + its_send_vinvall(vpe); +} + +static void its_vpe_irq_domain_deactivate(struct irq_domain *domain, + struct irq_data *d) +{ + struct its_vpe *vpe = irq_data_get_irq_chip_data(d); + + its_send_vmapp(vpe, false); +} + +static const struct irq_domain_ops its_vpe_domain_ops = { + .alloc = its_vpe_irq_domain_alloc, + .free = its_vpe_irq_domain_free, + .activate = its_vpe_irq_domain_activate, + .deactivate = its_vpe_irq_domain_deactivate, +}; + static int its_force_quiescent(void __iomem *base) { u32 count = 1000000; /* 1s */ @@ -1571,7 +2719,7 @@ static int its_force_quiescent(void __iomem *base) return 0; /* Disable the generation of all interrupts to this ITS */ - val &= ~GITS_CTLR_ENABLE; + val &= ~(GITS_CTLR_ENABLE | GITS_CTLR_ImDe); writel_relaxed(val, base + GITS_CTLR); /* Poll GITS_CTLR and wait until ITS becomes quiescent */ @@ -1672,13 +2820,92 @@ static int its_init_domain(struct fwnode_handle *handle, struct its_node *its) return 0; } +static int its_init_vpe_domain(void) +{ + struct its_node *its; + u32 devid; + int entries; + + if (gic_rdists->has_direct_lpi) { + pr_info("ITS: Using DirectLPI for VPE invalidation\n"); + return 0; + } + + /* Any ITS will do, even if not v4 */ + its = list_first_entry(&its_nodes, struct its_node, entry); + + entries = roundup_pow_of_two(nr_cpu_ids); + vpe_proxy.vpes = kzalloc(sizeof(*vpe_proxy.vpes) * entries, + GFP_KERNEL); + if (!vpe_proxy.vpes) { + pr_err("ITS: Can't allocate GICv4 proxy device array\n"); + return -ENOMEM; + } + + /* Use the last possible DevID */ + devid = GENMASK(its->device_ids - 1, 0); + vpe_proxy.dev = its_create_device(its, devid, entries, false); + if (!vpe_proxy.dev) { + kfree(vpe_proxy.vpes); + pr_err("ITS: Can't allocate GICv4 proxy device\n"); + return -ENOMEM; + } + + BUG_ON(entries != vpe_proxy.dev->nr_ites); + + raw_spin_lock_init(&vpe_proxy.lock); + vpe_proxy.next_victim = 0; + pr_info("ITS: Allocated DevID %x as GICv4 proxy device (%d slots)\n", + devid, vpe_proxy.dev->nr_ites); + + return 0; +} + +static int __init its_compute_its_list_map(struct resource *res, + void __iomem *its_base) +{ + int its_number; + u32 ctlr; + + /* + * This is assumed to be done early enough that we're + * guaranteed to be single-threaded, hence no + * locking. Should this change, we should address + * this. + */ + its_number = find_first_zero_bit(&its_list_map, ITS_LIST_MAX); + if (its_number >= ITS_LIST_MAX) { + pr_err("ITS@%pa: No ITSList entry available!\n", + &res->start); + return -EINVAL; + } + + ctlr = readl_relaxed(its_base + GITS_CTLR); + ctlr &= ~GITS_CTLR_ITS_NUMBER; + ctlr |= its_number << GITS_CTLR_ITS_NUMBER_SHIFT; + writel_relaxed(ctlr, its_base + GITS_CTLR); + ctlr = readl_relaxed(its_base + GITS_CTLR); + if ((ctlr & GITS_CTLR_ITS_NUMBER) != (its_number << GITS_CTLR_ITS_NUMBER_SHIFT)) { + its_number = ctlr & GITS_CTLR_ITS_NUMBER; + its_number >>= GITS_CTLR_ITS_NUMBER_SHIFT; + } + + if (test_and_set_bit(its_number, &its_list_map)) { + pr_err("ITS@%pa: Duplicate ITSList entry %d\n", + &res->start, its_number); + return -EINVAL; + } + + return its_number; +} + static int __init its_probe_one(struct resource *res, struct fwnode_handle *handle, int numa_node) { struct its_node *its; void __iomem *its_base; - u32 val; - u64 baser, tmp; + u32 val, ctlr; + u64 baser, tmp, typer; int err; its_base = ioremap(res->start, resource_size(res)); @@ -1711,9 +2938,24 @@ static int __init its_probe_one(struct resource *res, raw_spin_lock_init(&its->lock); INIT_LIST_HEAD(&its->entry); INIT_LIST_HEAD(&its->its_device_list); + typer = gic_read_typer(its_base + GITS_TYPER); its->base = its_base; its->phys_base = res->start; - its->ite_size = ((gic_read_typer(its_base + GITS_TYPER) >> 4) & 0xf) + 1; + its->ite_size = GITS_TYPER_ITT_ENTRY_SIZE(typer); + its->is_v4 = !!(typer & GITS_TYPER_VLPIS); + if (its->is_v4) { + if (!(typer & GITS_TYPER_VMOVP)) { + err = its_compute_its_list_map(res, its_base); + if (err < 0) + goto out_free_its; + + pr_info("ITS@%pa: Using ITS number %d\n", + &res->start, err); + } else { + pr_info("ITS@%pa: Single VMOVP capable\n", &res->start); + } + } + its->numa_node = numa_node; its->cmd_base = (void *)__get_free_pages(GFP_KERNEL | __GFP_ZERO, @@ -1760,7 +3002,11 @@ static int __init its_probe_one(struct resource *res, } gits_write_cwriter(0, its->base + GITS_CWRITER); - writel_relaxed(GITS_CTLR_ENABLE, its->base + GITS_CTLR); + ctlr = readl_relaxed(its->base + GITS_CTLR); + ctlr |= GITS_CTLR_ENABLE; + if (its->is_v4) + ctlr |= GITS_CTLR_ImDe; + writel_relaxed(ctlr, its->base + GITS_CTLR); err = its_init_domain(handle, its); if (err) @@ -1816,13 +3062,13 @@ static int __init its_of_probe(struct device_node *node) for (np = of_find_matching_node(node, its_device_id); np; np = of_find_matching_node(np, its_device_id)) { if (!of_property_read_bool(np, "msi-controller")) { - pr_warn("%s: no msi-controller property, ITS ignored\n", - np->full_name); + pr_warn("%pOF: no msi-controller property, ITS ignored\n", + np); continue; } if (of_address_to_resource(np, 0, &res)) { - pr_warn("%s: no regs?\n", np->full_name); + pr_warn("%pOF: no regs?\n", np); continue; } @@ -1984,6 +3230,9 @@ int __init its_init(struct fwnode_handle *handle, struct rdists *rdists, struct irq_domain *parent_domain) { struct device_node *of_node; + struct its_node *its; + bool has_v4 = false; + int err; its_parent = parent_domain; of_node = to_of_node(handle); @@ -1998,5 +3247,20 @@ int __init its_init(struct fwnode_handle *handle, struct rdists *rdists, } gic_rdists = rdists; - return its_alloc_lpi_tables(); + err = its_alloc_lpi_tables(); + if (err) + return err; + + list_for_each_entry(its, &its_nodes, entry) + has_v4 |= its->is_v4; + + if (has_v4 & rdists->has_vlpis) { + if (its_init_vpe_domain() || + its_init_v4(parent_domain, &its_vpe_domain_ops)) { + rdists->has_vlpis = false; + pr_err("ITS: Disabling GICv4 support\n"); + } + } + + return 0; } diff --git a/drivers/irqchip/irq-gic-v3.c b/drivers/irqchip/irq-gic-v3.c index 984c3ecfd22c..519149ec9053 100644 --- a/drivers/irqchip/irq-gic-v3.c +++ b/drivers/irqchip/irq-gic-v3.c @@ -1,5 +1,5 @@ /* - * Copyright (C) 2013, 2014 ARM Limited, All Rights Reserved. + * Copyright (C) 2013-2017 ARM Limited, All Rights Reserved. * Author: Marc Zyngier <marc.zyngier@arm.com> * * This program is free software; you can redistribute it and/or modify @@ -423,24 +423,14 @@ static void __init gic_dist_init(void) gic_write_irouter(affinity, base + GICD_IROUTER + i * 8); } -static int gic_populate_rdist(void) +static int gic_iterate_rdists(int (*fn)(struct redist_region *, void __iomem *)) { - unsigned long mpidr = cpu_logical_map(smp_processor_id()); - u64 typer; - u32 aff; + int ret = -ENODEV; int i; - /* - * Convert affinity to a 32bit value that can be matched to - * GICR_TYPER bits [63:32]. - */ - aff = (MPIDR_AFFINITY_LEVEL(mpidr, 3) << 24 | - MPIDR_AFFINITY_LEVEL(mpidr, 2) << 16 | - MPIDR_AFFINITY_LEVEL(mpidr, 1) << 8 | - MPIDR_AFFINITY_LEVEL(mpidr, 0)); - for (i = 0; i < gic_data.nr_redist_regions; i++) { void __iomem *ptr = gic_data.redist_regions[i].redist_base; + u64 typer; u32 reg; reg = readl_relaxed(ptr + GICR_PIDR2) & GIC_PIDR2_ARCH_MASK; @@ -452,15 +442,9 @@ static int gic_populate_rdist(void) do { typer = gic_read_typer(ptr + GICR_TYPER); - if ((typer >> 32) == aff) { - u64 offset = ptr - gic_data.redist_regions[i].redist_base; - gic_data_rdist_rd_base() = ptr; - gic_data_rdist()->phys_base = gic_data.redist_regions[i].phys_base + offset; - pr_info("CPU%d: found redistributor %lx region %d:%pa\n", - smp_processor_id(), mpidr, i, - &gic_data_rdist()->phys_base); + ret = fn(gic_data.redist_regions + i, ptr); + if (!ret) return 0; - } if (gic_data.redist_regions[i].single_redist) break; @@ -475,12 +459,71 @@ static int gic_populate_rdist(void) } while (!(typer & GICR_TYPER_LAST)); } + return ret ? -ENODEV : 0; +} + +static int __gic_populate_rdist(struct redist_region *region, void __iomem *ptr) +{ + unsigned long mpidr = cpu_logical_map(smp_processor_id()); + u64 typer; + u32 aff; + + /* + * Convert affinity to a 32bit value that can be matched to + * GICR_TYPER bits [63:32]. + */ + aff = (MPIDR_AFFINITY_LEVEL(mpidr, 3) << 24 | + MPIDR_AFFINITY_LEVEL(mpidr, 2) << 16 | + MPIDR_AFFINITY_LEVEL(mpidr, 1) << 8 | + MPIDR_AFFINITY_LEVEL(mpidr, 0)); + + typer = gic_read_typer(ptr + GICR_TYPER); + if ((typer >> 32) == aff) { + u64 offset = ptr - region->redist_base; + gic_data_rdist_rd_base() = ptr; + gic_data_rdist()->phys_base = region->phys_base + offset; + + pr_info("CPU%d: found redistributor %lx region %d:%pa\n", + smp_processor_id(), mpidr, + (int)(region - gic_data.redist_regions), + &gic_data_rdist()->phys_base); + return 0; + } + + /* Try next one */ + return 1; +} + +static int gic_populate_rdist(void) +{ + if (gic_iterate_rdists(__gic_populate_rdist) == 0) + return 0; + /* We couldn't even deal with ourselves... */ WARN(true, "CPU%d: mpidr %lx has no re-distributor!\n", - smp_processor_id(), mpidr); + smp_processor_id(), + (unsigned long)cpu_logical_map(smp_processor_id())); return -ENODEV; } +static int __gic_update_vlpi_properties(struct redist_region *region, + void __iomem *ptr) +{ + u64 typer = gic_read_typer(ptr + GICR_TYPER); + gic_data.rdists.has_vlpis &= !!(typer & GICR_TYPER_VLPIS); + gic_data.rdists.has_direct_lpi &= !!(typer & GICR_TYPER_DirectLPIS); + + return 1; +} + +static void gic_update_vlpi_properties(void) +{ + gic_iterate_rdists(__gic_update_vlpi_properties); + pr_info("%sVLPI support, %sdirect LPI support\n", + !gic_data.rdists.has_vlpis ? "no " : "", + !gic_data.rdists.has_direct_lpi ? "no " : ""); +} + static void gic_cpu_sys_reg_init(void) { /* @@ -677,6 +720,8 @@ static int gic_set_affinity(struct irq_data *d, const struct cpumask *mask_val, else gic_dist_wait_for_rwp(); + irq_data_update_effective_affinity(d, cpumask_of(cpu)); + return IRQ_SET_MASK_OK_DONE; } #else @@ -775,6 +820,7 @@ static int gic_irq_domain_map(struct irq_domain *d, unsigned int irq, irq_domain_set_info(d, irq, hw, chip, d->host_data, handle_fasteoi_irq, NULL, NULL); irq_set_probe(irq); + irqd_set_single_target(irq_desc_get_irq_data(irq_to_desc(irq))); } /* LPIs */ if (hw >= 8192 && hw < GIC_ID_NR) { @@ -953,6 +999,8 @@ static int __init gic_init_bases(void __iomem *dist_base, gic_data.domain = irq_domain_create_tree(handle, &gic_irq_domain_ops, &gic_data); gic_data.rdists.rdist = alloc_percpu(typeof(*gic_data.rdists.rdist)); + gic_data.rdists.has_vlpis = true; + gic_data.rdists.has_direct_lpi = true; if (WARN_ON(!gic_data.domain) || WARN_ON(!gic_data.rdists.rdist)) { err = -ENOMEM; @@ -961,6 +1009,8 @@ static int __init gic_init_bases(void __iomem *dist_base, set_handle_irq(gic_handle_irq); + gic_update_vlpi_properties(); + if (IS_ENABLED(CONFIG_ARM_GIC_V3_ITS) && gic_dist_supports_lpis()) its_init(handle, &gic_data.rdists, gic_data.domain); @@ -1067,7 +1117,7 @@ static void __init gic_populate_ppi_partitions(struct device_node *gic_node) if (WARN_ON(cpu == -1)) continue; - pr_cont("%s[%d] ", cpu_node->full_name, cpu); + pr_cont("%pOF[%d] ", cpu_node, cpu); cpumask_set_cpu(cpu, &part->mask); } @@ -1122,6 +1172,7 @@ static void __init gic_of_setup_kvm_info(struct device_node *node) if (!ret) gic_v3_kvm_info.vcpu = r; + gic_v3_kvm_info.has_v4 = gic_data.rdists.has_vlpis; gic_set_kvm_info(&gic_v3_kvm_info); } @@ -1135,15 +1186,13 @@ static int __init gic_of_init(struct device_node *node, struct device_node *pare dist_base = of_iomap(node, 0); if (!dist_base) { - pr_err("%s: unable to map gic dist registers\n", - node->full_name); + pr_err("%pOF: unable to map gic dist registers\n", node); return -ENXIO; } err = gic_validate_dist_version(dist_base); if (err) { - pr_err("%s: no distributor detected, giving up\n", - node->full_name); + pr_err("%pOF: no distributor detected, giving up\n", node); goto out_unmap_dist; } @@ -1163,8 +1212,7 @@ static int __init gic_of_init(struct device_node *node, struct device_node *pare ret = of_address_to_resource(node, 1 + i, &res); rdist_regs[i].redist_base = of_iomap(node, 1 + i); if (ret || !rdist_regs[i].redist_base) { - pr_err("%s: couldn't map region %d\n", - node->full_name, i); + pr_err("%pOF: couldn't map region %d\n", node, i); err = -ENODEV; goto out_unmap_rdist; } @@ -1418,6 +1466,7 @@ static void __init gic_acpi_setup_kvm_info(void) vcpu->end = vcpu->start + ACPI_GICV2_VCPU_MEM_SIZE - 1; } + gic_v3_kvm_info.has_v4 = gic_data.rdists.has_vlpis; gic_set_kvm_info(&gic_v3_kvm_info); } diff --git a/drivers/irqchip/irq-gic-v4.c b/drivers/irqchip/irq-gic-v4.c new file mode 100644 index 000000000000..2370e6d9e603 --- /dev/null +++ b/drivers/irqchip/irq-gic-v4.c @@ -0,0 +1,225 @@ +/* + * Copyright (C) 2016,2017 ARM Limited, All Rights Reserved. + * Author: Marc Zyngier <marc.zyngier@arm.com> + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program. If not, see <http://www.gnu.org/licenses/>. + */ + +#include <linux/interrupt.h> +#include <linux/irq.h> +#include <linux/irqdomain.h> +#include <linux/msi.h> +#include <linux/sched.h> + +#include <linux/irqchip/arm-gic-v4.h> + +/* + * WARNING: The blurb below assumes that you understand the + * intricacies of GICv3, GICv4, and how a guest's view of a GICv3 gets + * translated into GICv4 commands. So it effectively targets at most + * two individuals. You know who you are. + * + * The core GICv4 code is designed to *avoid* exposing too much of the + * core GIC code (that would in turn leak into the hypervisor code), + * and instead provide a hypervisor agnostic interface to the HW (of + * course, the astute reader will quickly realize that hypervisor + * agnostic actually means KVM-specific - what were you thinking?). + * + * In order to achieve a modicum of isolation, we try to hide most of + * the GICv4 "stuff" behind normal irqchip operations: + * + * - Any guest-visible VLPI is backed by a Linux interrupt (and a + * physical LPI which gets unmapped when the guest maps the + * VLPI). This allows the same DevID/EventID pair to be either + * mapped to the LPI (host) or the VLPI (guest). Note that this is + * exclusive, and you cannot have both. + * + * - Enabling/disabling a VLPI is done by issuing mask/unmask calls. + * + * - Guest INT/CLEAR commands are implemented through + * irq_set_irqchip_state(). + * + * - The *bizarre* stuff (mapping/unmapping an interrupt to a VLPI, or + * issuing an INV after changing a priority) gets shoved into the + * irq_set_vcpu_affinity() method. While this is quite horrible + * (let's face it, this is the irqchip version of an ioctl), it + * confines the crap to a single location. And map/unmap really is + * about setting the affinity of a VLPI to a vcpu, so only INV is + * majorly out of place. So there. + * + * A number of commands are simply not provided by this interface, as + * they do not make direct sense. For example, MAPD is purely local to + * the virtual ITS (because it references a virtual device, and the + * physical ITS is still very much in charge of the physical + * device). Same goes for things like MAPC (the physical ITS deals + * with the actual vPE affinity, and not the braindead concept of + * collection). SYNC is not provided either, as each and every command + * is followed by a VSYNC. This could be relaxed in the future, should + * this be seen as a bottleneck (yes, this means *never*). + * + * But handling VLPIs is only one side of the job of the GICv4 + * code. The other (darker) side is to take care of the doorbell + * interrupts which are delivered when a VLPI targeting a non-running + * vcpu is being made pending. + * + * The choice made here is that each vcpu (VPE in old northern GICv4 + * dialect) gets a single doorbell LPI, no matter how many interrupts + * are targeting it. This has a nice property, which is that the + * interrupt becomes a handle for the VPE, and that the hypervisor + * code can manipulate it through the normal interrupt API: + * + * - VMs (or rather the VM abstraction that matters to the GIC) + * contain an irq domain where each interrupt maps to a VPE. In + * turn, this domain sits on top of the normal LPI allocator, and a + * specially crafted irq_chip implementation. + * + * - mask/unmask do what is expected on the doorbell interrupt. + * + * - irq_set_affinity is used to move a VPE from one redistributor to + * another. + * + * - irq_set_vcpu_affinity once again gets hijacked for the purpose of + * creating a new sub-API, namely scheduling/descheduling a VPE + * (which involves programming GICR_V{PROP,PEND}BASER) and + * performing INVALL operations. + */ + +static struct irq_domain *gic_domain; +static const struct irq_domain_ops *vpe_domain_ops; + +int its_alloc_vcpu_irqs(struct its_vm *vm) +{ + int vpe_base_irq, i; + + vm->fwnode = irq_domain_alloc_named_id_fwnode("GICv4-vpe", + task_pid_nr(current)); + if (!vm->fwnode) + goto err; + + vm->domain = irq_domain_create_hierarchy(gic_domain, 0, vm->nr_vpes, + vm->fwnode, vpe_domain_ops, + vm); + if (!vm->domain) + goto err; + + for (i = 0; i < vm->nr_vpes; i++) { + vm->vpes[i]->its_vm = vm; + vm->vpes[i]->idai = true; + } + + vpe_base_irq = __irq_domain_alloc_irqs(vm->domain, -1, vm->nr_vpes, + NUMA_NO_NODE, vm, + false, NULL); + if (vpe_base_irq <= 0) + goto err; + + for (i = 0; i < vm->nr_vpes; i++) + vm->vpes[i]->irq = vpe_base_irq + i; + + return 0; + +err: + if (vm->domain) + irq_domain_remove(vm->domain); + if (vm->fwnode) + irq_domain_free_fwnode(vm->fwnode); + + return -ENOMEM; +} + +void its_free_vcpu_irqs(struct its_vm *vm) +{ + irq_domain_free_irqs(vm->vpes[0]->irq, vm->nr_vpes); + irq_domain_remove(vm->domain); + irq_domain_free_fwnode(vm->fwnode); +} + +static int its_send_vpe_cmd(struct its_vpe *vpe, struct its_cmd_info *info) +{ + return irq_set_vcpu_affinity(vpe->irq, info); +} + +int its_schedule_vpe(struct its_vpe *vpe, bool on) +{ + struct its_cmd_info info; + + WARN_ON(preemptible()); + + info.cmd_type = on ? SCHEDULE_VPE : DESCHEDULE_VPE; + + return its_send_vpe_cmd(vpe, &info); +} + +int its_invall_vpe(struct its_vpe *vpe) +{ + struct its_cmd_info info = { + .cmd_type = INVALL_VPE, + }; + + return its_send_vpe_cmd(vpe, &info); +} + +int its_map_vlpi(int irq, struct its_vlpi_map *map) +{ + struct its_cmd_info info = { + .cmd_type = MAP_VLPI, + .map = map, + }; + + /* + * The host will never see that interrupt firing again, so it + * is vital that we don't do any lazy masking. + */ + irq_set_status_flags(irq, IRQ_DISABLE_UNLAZY); + + return irq_set_vcpu_affinity(irq, &info); +} + +int its_get_vlpi(int irq, struct its_vlpi_map *map) +{ + struct its_cmd_info info = { + .cmd_type = GET_VLPI, + .map = map, + }; + + return irq_set_vcpu_affinity(irq, &info); +} + +int its_unmap_vlpi(int irq) +{ + irq_clear_status_flags(irq, IRQ_DISABLE_UNLAZY); + return irq_set_vcpu_affinity(irq, NULL); +} + +int its_prop_update_vlpi(int irq, u8 config, bool inv) +{ + struct its_cmd_info info = { + .cmd_type = inv ? PROP_UPDATE_AND_INV_VLPI : PROP_UPDATE_VLPI, + .config = config, + }; + + return irq_set_vcpu_affinity(irq, &info); +} + +int its_init_v4(struct irq_domain *domain, const struct irq_domain_ops *ops) +{ + if (domain) { + pr_info("ITS: Enabling GICv4 support\n"); + gic_domain = domain; + vpe_domain_ops = ops; + return 0; + } + + pr_err("ITS: No GICv4 VPE domain allocated\n"); + return -ENODEV; +} diff --git a/drivers/irqchip/irq-gic.c b/drivers/irqchip/irq-gic.c index d3e7c43718b8..651d726e8b12 100644 --- a/drivers/irqchip/irq-gic.c +++ b/drivers/irqchip/irq-gic.c @@ -344,6 +344,8 @@ static int gic_set_affinity(struct irq_data *d, const struct cpumask *mask_val, writel_relaxed(val | bit, reg); gic_unlock_irqrestore(flags); + irq_data_update_effective_affinity(d, cpumask_of(cpu)); + return IRQ_SET_MASK_OK_DONE; } #endif @@ -413,7 +415,7 @@ static void gic_handle_cascade_irq(struct irq_desc *desc) chained_irq_exit(chip, desc); } -static struct irq_chip gic_chip = { +static const struct irq_chip gic_chip = { .irq_mask = gic_mask_irq, .irq_unmask = gic_unmask_irq, .irq_eoi = gic_eoi_irq, @@ -969,6 +971,7 @@ static int gic_irq_domain_map(struct irq_domain *d, unsigned int irq, irq_domain_set_info(d, irq, hw, &gic->chip, d->host_data, handle_fasteoi_irq, NULL, NULL); irq_set_probe(irq); + irqd_set_single_target(irq_desc_get_irq_data(irq_to_desc(irq))); } return 0; } diff --git a/drivers/irqchip/irq-hip04.c b/drivers/irqchip/irq-hip04.c index c1b4ee955dbe..5b4fd2f4e5f8 100644 --- a/drivers/irqchip/irq-hip04.c +++ b/drivers/irqchip/irq-hip04.c @@ -165,6 +165,8 @@ static int hip04_irq_set_affinity(struct irq_data *d, writel_relaxed(val | bit, reg); raw_spin_unlock(&irq_controller_lock); + irq_data_update_effective_affinity(d, cpumask_of(cpu)); + return IRQ_SET_MASK_OK; } #endif @@ -312,6 +314,7 @@ static int hip04_irq_domain_map(struct irq_domain *d, unsigned int irq, irq_set_chip_and_handler(irq, &hip04_irq_chip, handle_fasteoi_irq); irq_set_probe(irq); + irqd_set_single_target(irq_desc_get_irq_data(irq_to_desc(irq))); } irq_set_chip_data(irq, d->host_data); return 0; diff --git a/drivers/irqchip/irq-imx-gpcv2.c b/drivers/irqchip/irq-imx-gpcv2.c index bb36f572e322..675eda5ff2b8 100644 --- a/drivers/irqchip/irq-imx-gpcv2.c +++ b/drivers/irqchip/irq-imx-gpcv2.c @@ -214,13 +214,13 @@ static int __init imx_gpcv2_irqchip_init(struct device_node *node, int i; if (!parent) { - pr_err("%s: no parent, giving up\n", node->full_name); + pr_err("%pOF: no parent, giving up\n", node); return -ENODEV; } parent_domain = irq_find_host(parent); if (!parent_domain) { - pr_err("%s: unable to get parent domain\n", node->full_name); + pr_err("%pOF: unable to get parent domain\n", node); return -ENXIO; } diff --git a/drivers/irqchip/irq-lpc32xx.c b/drivers/irqchip/irq-lpc32xx.c index 1034aeb2e98a..a48357d369b5 100644 --- a/drivers/irqchip/irq-lpc32xx.c +++ b/drivers/irqchip/irq-lpc32xx.c @@ -191,7 +191,7 @@ static int __init lpc32xx_of_ic_init(struct device_node *node, irqc->base = of_iomap(node, 0); if (!irqc->base) { - pr_err("%s: unable to map registers\n", node->full_name); + pr_err("%pOF: unable to map registers\n", node); kfree(irqc); return -EINVAL; } diff --git a/drivers/irqchip/irq-ls-scfg-msi.c b/drivers/irqchip/irq-ls-scfg-msi.c index 02cca74cab94..119f4ef0d421 100644 --- a/drivers/irqchip/irq-ls-scfg-msi.c +++ b/drivers/irqchip/irq-ls-scfg-msi.c @@ -17,13 +17,32 @@ #include <linux/irq.h> #include <linux/irqchip/chained_irq.h> #include <linux/irqdomain.h> +#include <linux/of_irq.h> #include <linux/of_pci.h> #include <linux/of_platform.h> #include <linux/spinlock.h> -#define MSI_MAX_IRQS 32 -#define MSI_IBS_SHIFT 3 -#define MSIR 4 +#define MSI_IRQS_PER_MSIR 32 +#define MSI_MSIR_OFFSET 4 + +#define MSI_LS1043V1_1_IRQS_PER_MSIR 8 +#define MSI_LS1043V1_1_MSIR_OFFSET 0x10 + +struct ls_scfg_msi_cfg { + u32 ibs_shift; /* Shift of interrupt bit select */ + u32 msir_irqs; /* The irq number per MSIR */ + u32 msir_base; /* The base address of MSIR */ +}; + +struct ls_scfg_msir { + struct ls_scfg_msi *msi_data; + unsigned int index; + unsigned int gic_irq; + unsigned int bit_start; + unsigned int bit_end; + unsigned int srs; /* Shared interrupt register select */ + void __iomem *reg; +}; struct ls_scfg_msi { spinlock_t lock; @@ -32,8 +51,11 @@ struct ls_scfg_msi { struct irq_domain *msi_domain; void __iomem *regs; phys_addr_t msiir_addr; - int irq; - DECLARE_BITMAP(used, MSI_MAX_IRQS); + struct ls_scfg_msi_cfg *cfg; + u32 msir_num; + struct ls_scfg_msir *msir; + u32 irqs_num; + unsigned long *used; }; static struct irq_chip ls_scfg_msi_irq_chip = { @@ -49,19 +71,56 @@ static struct msi_domain_info ls_scfg_msi_domain_info = { .chip = &ls_scfg_msi_irq_chip, }; +static int msi_affinity_flag = 1; + +static int __init early_parse_ls_scfg_msi(char *p) +{ + if (p && strncmp(p, "no-affinity", 11) == 0) + msi_affinity_flag = 0; + else + msi_affinity_flag = 1; + + return 0; +} +early_param("lsmsi", early_parse_ls_scfg_msi); + static void ls_scfg_msi_compose_msg(struct irq_data *data, struct msi_msg *msg) { struct ls_scfg_msi *msi_data = irq_data_get_irq_chip_data(data); msg->address_hi = upper_32_bits(msi_data->msiir_addr); msg->address_lo = lower_32_bits(msi_data->msiir_addr); - msg->data = data->hwirq << MSI_IBS_SHIFT; + msg->data = data->hwirq; + + if (msi_affinity_flag) + msg->data |= cpumask_first(data->common->affinity); } static int ls_scfg_msi_set_affinity(struct irq_data *irq_data, const struct cpumask *mask, bool force) { - return -EINVAL; + struct ls_scfg_msi *msi_data = irq_data_get_irq_chip_data(irq_data); + u32 cpu; + + if (!msi_affinity_flag) + return -EINVAL; + + if (!force) + cpu = cpumask_any_and(mask, cpu_online_mask); + else + cpu = cpumask_first(mask); + + if (cpu >= msi_data->msir_num) + return -EINVAL; + + if (msi_data->msir[cpu].gic_irq <= 0) { + pr_warn("cannot bind the irq to cpu%d\n", cpu); + return -EINVAL; + } + + cpumask_copy(irq_data->common->affinity, mask); + + return IRQ_SET_MASK_OK; } static struct irq_chip ls_scfg_msi_parent_chip = { @@ -81,8 +140,8 @@ static int ls_scfg_msi_domain_irq_alloc(struct irq_domain *domain, WARN_ON(nr_irqs != 1); spin_lock(&msi_data->lock); - pos = find_first_zero_bit(msi_data->used, MSI_MAX_IRQS); - if (pos < MSI_MAX_IRQS) + pos = find_first_zero_bit(msi_data->used, msi_data->irqs_num); + if (pos < msi_data->irqs_num) __set_bit(pos, msi_data->used); else err = -ENOSPC; @@ -106,7 +165,7 @@ static void ls_scfg_msi_domain_irq_free(struct irq_domain *domain, int pos; pos = d->hwirq; - if (pos < 0 || pos >= MSI_MAX_IRQS) { + if (pos < 0 || pos >= msi_data->irqs_num) { pr_err("failed to teardown msi. Invalid hwirq %d\n", pos); return; } @@ -123,15 +182,22 @@ static const struct irq_domain_ops ls_scfg_msi_domain_ops = { static void ls_scfg_msi_irq_handler(struct irq_desc *desc) { - struct ls_scfg_msi *msi_data = irq_desc_get_handler_data(desc); + struct ls_scfg_msir *msir = irq_desc_get_handler_data(desc); + struct ls_scfg_msi *msi_data = msir->msi_data; unsigned long val; - int pos, virq; + int pos, size, virq, hwirq; chained_irq_enter(irq_desc_get_chip(desc), desc); - val = ioread32be(msi_data->regs + MSIR); - for_each_set_bit(pos, &val, MSI_MAX_IRQS) { - virq = irq_find_mapping(msi_data->parent, (31 - pos)); + val = ioread32be(msir->reg); + + pos = msir->bit_start; + size = msir->bit_end + 1; + + for_each_set_bit_from(pos, &val, size) { + hwirq = ((msir->bit_end - pos) << msi_data->cfg->ibs_shift) | + msir->srs; + virq = irq_find_mapping(msi_data->parent, hwirq); if (virq) generic_handle_irq(virq); } @@ -143,7 +209,7 @@ static int ls_scfg_msi_domains_init(struct ls_scfg_msi *msi_data) { /* Initialize MSI domain parent */ msi_data->parent = irq_domain_add_linear(NULL, - MSI_MAX_IRQS, + msi_data->irqs_num, &ls_scfg_msi_domain_ops, msi_data); if (!msi_data->parent) { @@ -164,16 +230,117 @@ static int ls_scfg_msi_domains_init(struct ls_scfg_msi *msi_data) return 0; } +static int ls_scfg_msi_setup_hwirq(struct ls_scfg_msi *msi_data, int index) +{ + struct ls_scfg_msir *msir; + int virq, i, hwirq; + + virq = platform_get_irq(msi_data->pdev, index); + if (virq <= 0) + return -ENODEV; + + msir = &msi_data->msir[index]; + msir->index = index; + msir->msi_data = msi_data; + msir->gic_irq = virq; + msir->reg = msi_data->regs + msi_data->cfg->msir_base + 4 * index; + + if (msi_data->cfg->msir_irqs == MSI_LS1043V1_1_IRQS_PER_MSIR) { + msir->bit_start = 32 - ((msir->index + 1) * + MSI_LS1043V1_1_IRQS_PER_MSIR); + msir->bit_end = msir->bit_start + + MSI_LS1043V1_1_IRQS_PER_MSIR - 1; + } else { + msir->bit_start = 0; + msir->bit_end = msi_data->cfg->msir_irqs - 1; + } + + irq_set_chained_handler_and_data(msir->gic_irq, + ls_scfg_msi_irq_handler, + msir); + + if (msi_affinity_flag) { + /* Associate MSIR interrupt to the cpu */ + irq_set_affinity(msir->gic_irq, get_cpu_mask(index)); + msir->srs = 0; /* This value is determined by the CPU */ + } else + msir->srs = index; + + /* Release the hwirqs corresponding to this MSIR */ + if (!msi_affinity_flag || msir->index == 0) { + for (i = 0; i < msi_data->cfg->msir_irqs; i++) { + hwirq = i << msi_data->cfg->ibs_shift | msir->index; + bitmap_clear(msi_data->used, hwirq, 1); + } + } + + return 0; +} + +static int ls_scfg_msi_teardown_hwirq(struct ls_scfg_msir *msir) +{ + struct ls_scfg_msi *msi_data = msir->msi_data; + int i, hwirq; + + if (msir->gic_irq > 0) + irq_set_chained_handler_and_data(msir->gic_irq, NULL, NULL); + + for (i = 0; i < msi_data->cfg->msir_irqs; i++) { + hwirq = i << msi_data->cfg->ibs_shift | msir->index; + bitmap_set(msi_data->used, hwirq, 1); + } + + return 0; +} + +static struct ls_scfg_msi_cfg ls1021_msi_cfg = { + .ibs_shift = 3, + .msir_irqs = MSI_IRQS_PER_MSIR, + .msir_base = MSI_MSIR_OFFSET, +}; + +static struct ls_scfg_msi_cfg ls1046_msi_cfg = { + .ibs_shift = 2, + .msir_irqs = MSI_IRQS_PER_MSIR, + .msir_base = MSI_MSIR_OFFSET, +}; + +static struct ls_scfg_msi_cfg ls1043_v1_1_msi_cfg = { + .ibs_shift = 2, + .msir_irqs = MSI_LS1043V1_1_IRQS_PER_MSIR, + .msir_base = MSI_LS1043V1_1_MSIR_OFFSET, +}; + +static const struct of_device_id ls_scfg_msi_id[] = { + /* The following two misspelled compatibles are obsolete */ + { .compatible = "fsl,1s1021a-msi", .data = &ls1021_msi_cfg}, + { .compatible = "fsl,1s1043a-msi", .data = &ls1021_msi_cfg}, + + { .compatible = "fsl,ls1021a-msi", .data = &ls1021_msi_cfg }, + { .compatible = "fsl,ls1043a-msi", .data = &ls1021_msi_cfg }, + { .compatible = "fsl,ls1043a-v1.1-msi", .data = &ls1043_v1_1_msi_cfg }, + { .compatible = "fsl,ls1046a-msi", .data = &ls1046_msi_cfg }, + {}, +}; +MODULE_DEVICE_TABLE(of, ls_scfg_msi_id); + static int ls_scfg_msi_probe(struct platform_device *pdev) { + const struct of_device_id *match; struct ls_scfg_msi *msi_data; struct resource *res; - int ret; + int i, ret; + + match = of_match_device(ls_scfg_msi_id, &pdev->dev); + if (!match) + return -ENODEV; msi_data = devm_kzalloc(&pdev->dev, sizeof(*msi_data), GFP_KERNEL); if (!msi_data) return -ENOMEM; + msi_data->cfg = (struct ls_scfg_msi_cfg *) match->data; + res = platform_get_resource(pdev, IORESOURCE_MEM, 0); msi_data->regs = devm_ioremap_resource(&pdev->dev, res); if (IS_ERR(msi_data->regs)) { @@ -182,23 +349,48 @@ static int ls_scfg_msi_probe(struct platform_device *pdev) } msi_data->msiir_addr = res->start; - msi_data->irq = platform_get_irq(pdev, 0); - if (msi_data->irq <= 0) { - dev_err(&pdev->dev, "failed to get MSI irq\n"); - return -ENODEV; - } - msi_data->pdev = pdev; spin_lock_init(&msi_data->lock); + msi_data->irqs_num = MSI_IRQS_PER_MSIR * + (1 << msi_data->cfg->ibs_shift); + msi_data->used = devm_kcalloc(&pdev->dev, + BITS_TO_LONGS(msi_data->irqs_num), + sizeof(*msi_data->used), + GFP_KERNEL); + if (!msi_data->used) + return -ENOMEM; + /* + * Reserve all the hwirqs + * The available hwirqs will be released in ls1_msi_setup_hwirq() + */ + bitmap_set(msi_data->used, 0, msi_data->irqs_num); + + msi_data->msir_num = of_irq_count(pdev->dev.of_node); + + if (msi_affinity_flag) { + u32 cpu_num; + + cpu_num = num_possible_cpus(); + if (msi_data->msir_num >= cpu_num) + msi_data->msir_num = cpu_num; + else + msi_affinity_flag = 0; + } + + msi_data->msir = devm_kcalloc(&pdev->dev, msi_data->msir_num, + sizeof(*msi_data->msir), + GFP_KERNEL); + if (!msi_data->msir) + return -ENOMEM; + + for (i = 0; i < msi_data->msir_num; i++) + ls_scfg_msi_setup_hwirq(msi_data, i); + ret = ls_scfg_msi_domains_init(msi_data); if (ret) return ret; - irq_set_chained_handler_and_data(msi_data->irq, - ls_scfg_msi_irq_handler, - msi_data); - platform_set_drvdata(pdev, msi_data); return 0; @@ -207,8 +399,10 @@ static int ls_scfg_msi_probe(struct platform_device *pdev) static int ls_scfg_msi_remove(struct platform_device *pdev) { struct ls_scfg_msi *msi_data = platform_get_drvdata(pdev); + int i; - irq_set_chained_handler_and_data(msi_data->irq, NULL, NULL); + for (i = 0; i < msi_data->msir_num; i++) + ls_scfg_msi_teardown_hwirq(&msi_data->msir[i]); irq_domain_remove(msi_data->msi_domain); irq_domain_remove(msi_data->parent); @@ -218,12 +412,6 @@ static int ls_scfg_msi_remove(struct platform_device *pdev) return 0; } -static const struct of_device_id ls_scfg_msi_id[] = { - { .compatible = "fsl,1s1021a-msi", }, - { .compatible = "fsl,1s1043a-msi", }, - {}, -}; - static struct platform_driver ls_scfg_msi_driver = { .driver = { .name = "ls-scfg-msi", diff --git a/drivers/irqchip/irq-metag-ext.c b/drivers/irqchip/irq-metag-ext.c index 0cdd923d1535..be7216bfb8dd 100644 --- a/drivers/irqchip/irq-metag-ext.c +++ b/drivers/irqchip/irq-metag-ext.c @@ -518,6 +518,8 @@ static int meta_intc_set_affinity(struct irq_data *data, metag_out32(TBI_TRIG_VEC(TBID_SIGNUM_TR2(thread)), vec_addr); + irq_data_update_effective_affinity(data, cpumask_of(cpu)); + return 0; } #else @@ -578,6 +580,8 @@ static int meta_intc_map(struct irq_domain *d, unsigned int irq, else irq_set_chip_and_handler(irq, &meta_intc_edge_chip, handle_edge_irq); + + irqd_set_single_target(irq_desc_get_irq_data(irq_to_desc(irq))); return 0; } diff --git a/drivers/irqchip/irq-mips-gic.c b/drivers/irqchip/irq-mips-gic.c index 48ee1bad473f..b3a60da088db 100644 --- a/drivers/irqchip/irq-mips-gic.c +++ b/drivers/irqchip/irq-mips-gic.c @@ -445,24 +445,27 @@ static int gic_set_affinity(struct irq_data *d, const struct cpumask *cpumask, unsigned int irq = GIC_HWIRQ_TO_SHARED(d->hwirq); cpumask_t tmp = CPU_MASK_NONE; unsigned long flags; - int i; + int i, cpu; cpumask_and(&tmp, cpumask, cpu_online_mask); if (cpumask_empty(&tmp)) return -EINVAL; + cpu = cpumask_first(&tmp); + /* Assumption : cpumask refers to a single CPU */ spin_lock_irqsave(&gic_lock, flags); /* Re-route this IRQ */ - gic_map_to_vpe(irq, mips_cm_vp_id(cpumask_first(&tmp))); + gic_map_to_vpe(irq, mips_cm_vp_id(cpu)); /* Update the pcpu_masks */ for (i = 0; i < min(gic_vpes, NR_CPUS); i++) clear_bit(irq, pcpu_masks[i].pcpu_mask); - set_bit(irq, pcpu_masks[cpumask_first(&tmp)].pcpu_mask); + set_bit(irq, pcpu_masks[cpu].pcpu_mask); cpumask_copy(irq_data_get_affinity_mask(d), cpumask); + irq_data_update_effective_affinity(d, cpumask_of(cpu)); spin_unlock_irqrestore(&gic_lock, flags); return IRQ_SET_MASK_OK_NOCOPY; @@ -716,6 +719,7 @@ static int gic_irq_domain_map(struct irq_domain *d, unsigned int virq, if (err) return err; + irqd_set_single_target(irq_desc_get_irq_data(irq_to_desc(virq))); return gic_shared_irq_domain_map(d, virq, hwirq, 0); } diff --git a/drivers/irqchip/irq-mmp.c b/drivers/irqchip/irq-mmp.c index 013fc9659a84..25f32e1d7764 100644 --- a/drivers/irqchip/irq-mmp.c +++ b/drivers/irqchip/irq-mmp.c @@ -181,13 +181,13 @@ const struct irq_domain_ops mmp_irq_domain_ops = { .xlate = mmp_irq_domain_xlate, }; -static struct mmp_intc_conf mmp_conf = { +static const struct mmp_intc_conf mmp_conf = { .conf_enable = 0x51, .conf_disable = 0x0, .conf_mask = 0x7f, }; -static struct mmp_intc_conf mmp2_conf = { +static const struct mmp_intc_conf mmp2_conf = { .conf_enable = 0x20, .conf_disable = 0x0, .conf_mask = 0x7f, diff --git a/drivers/irqchip/irq-mtk-sysirq.c b/drivers/irqchip/irq-mtk-sysirq.c index eeac512ec5a8..90aaf190157f 100644 --- a/drivers/irqchip/irq-mtk-sysirq.c +++ b/drivers/irqchip/irq-mtk-sysirq.c @@ -178,8 +178,7 @@ static int __init mtk_sysirq_of_init(struct device_node *node, chip_data->intpol_words[i] = size / 4; chip_data->intpol_bases[i] = of_iomap(node, i); if (ret || !chip_data->intpol_bases[i]) { - pr_err("%s: couldn't map region %d\n", - node->full_name, i); + pr_err("%pOF: couldn't map region %d\n", node, i); ret = -ENODEV; goto out_free_intpol; } diff --git a/drivers/irqchip/irq-mxs.c b/drivers/irqchip/irq-mxs.c index 05fa9f7af53c..e8b31f52e071 100644 --- a/drivers/irqchip/irq-mxs.c +++ b/drivers/irqchip/irq-mxs.c @@ -179,7 +179,7 @@ static void __init icoll_add_domain(struct device_node *np, &icoll_irq_domain_ops, NULL); if (!icoll_domain) - panic("%s: unable to create irq domain", np->full_name); + panic("%pOF: unable to create irq domain", np); } static void __iomem * __init icoll_init_iobase(struct device_node *np) @@ -188,7 +188,7 @@ static void __iomem * __init icoll_init_iobase(struct device_node *np) icoll_base = of_io_request_and_map(np, 0, np->name); if (IS_ERR(icoll_base)) - panic("%s: unable to map resource", np->full_name); + panic("%pOF: unable to map resource", np); return icoll_base; } diff --git a/drivers/irqchip/irq-stm32-exti.c b/drivers/irqchip/irq-stm32-exti.c index 491568c95aa5..45363ff8d06f 100644 --- a/drivers/irqchip/irq-stm32-exti.c +++ b/drivers/irqchip/irq-stm32-exti.c @@ -140,7 +140,7 @@ static int __init stm32_exti_init(struct device_node *node, base = of_iomap(node, 0); if (!base) { - pr_err("%s: Unable to map registers\n", node->full_name); + pr_err("%pOF: Unable to map registers\n", node); return -ENOMEM; } @@ -149,7 +149,7 @@ static int __init stm32_exti_init(struct device_node *node, nr_exti = fls(readl_relaxed(base + EXTI_RTSR)); writel_relaxed(0, base + EXTI_RTSR); - pr_info("%s: %d External IRQs detected\n", node->full_name, nr_exti); + pr_info("%pOF: %d External IRQs detected\n", node, nr_exti); domain = irq_domain_add_linear(node, nr_exti, &irq_exti_domain_ops, NULL); @@ -163,8 +163,8 @@ static int __init stm32_exti_init(struct device_node *node, ret = irq_alloc_domain_generic_chips(domain, nr_exti, 1, "exti", handle_edge_irq, clr, 0, 0); if (ret) { - pr_err("%s: Could not allocate generic interrupt chip.\n", - node->full_name); + pr_err("%pOF: Could not allocate generic interrupt chip.\n", + node); goto out_free_domain; } diff --git a/drivers/irqchip/irq-sun4i.c b/drivers/irqchip/irq-sun4i.c index 376b28074e0d..e3e5b9132b75 100644 --- a/drivers/irqchip/irq-sun4i.c +++ b/drivers/irqchip/irq-sun4i.c @@ -97,8 +97,8 @@ static int __init sun4i_of_init(struct device_node *node, { sun4i_irq_base = of_iomap(node, 0); if (!sun4i_irq_base) - panic("%s: unable to map IC registers\n", - node->full_name); + panic("%pOF: unable to map IC registers\n", + node); /* Disable all interrupts */ writel(0, sun4i_irq_base + SUN4I_IRQ_ENABLE_REG(0)); @@ -124,7 +124,7 @@ static int __init sun4i_of_init(struct device_node *node, sun4i_irq_domain = irq_domain_add_linear(node, 3 * 32, &sun4i_irq_ops, NULL); if (!sun4i_irq_domain) - panic("%s: unable to create IRQ domain\n", node->full_name); + panic("%pOF: unable to create IRQ domain\n", node); set_handle_irq(sun4i_handle_irq); diff --git a/drivers/irqchip/irq-tegra.c b/drivers/irqchip/irq-tegra.c index 3973a14bb15b..0abc0cd1c32e 100644 --- a/drivers/irqchip/irq-tegra.c +++ b/drivers/irqchip/irq-tegra.c @@ -291,13 +291,13 @@ static int __init tegra_ictlr_init(struct device_node *node, int err; if (!parent) { - pr_err("%s: no parent, giving up\n", node->full_name); + pr_err("%pOF: no parent, giving up\n", node); return -ENODEV; } parent_domain = irq_find_host(parent); if (!parent_domain) { - pr_err("%s: unable to obtain parent domain\n", node->full_name); + pr_err("%pOF: unable to obtain parent domain\n", node); return -ENXIO; } @@ -329,29 +329,29 @@ static int __init tegra_ictlr_init(struct device_node *node, } if (!num_ictlrs) { - pr_err("%s: no valid regions, giving up\n", node->full_name); + pr_err("%pOF: no valid regions, giving up\n", node); err = -ENOMEM; goto out_free; } WARN(num_ictlrs != soc->num_ictlrs, - "%s: Found %u interrupt controllers in DT; expected %u.\n", - node->full_name, num_ictlrs, soc->num_ictlrs); + "%pOF: Found %u interrupt controllers in DT; expected %u.\n", + node, num_ictlrs, soc->num_ictlrs); domain = irq_domain_add_hierarchy(parent_domain, 0, num_ictlrs * 32, node, &tegra_ictlr_domain_ops, lic); if (!domain) { - pr_err("%s: failed to allocated domain\n", node->full_name); + pr_err("%pOF: failed to allocated domain\n", node); err = -ENOMEM; goto out_unmap; } tegra_ictlr_syscore_init(); - pr_info("%s: %d interrupts forwarded to %s\n", - node->full_name, num_ictlrs * 32, parent->full_name); + pr_info("%pOF: %d interrupts forwarded to %pOF\n", + node, num_ictlrs * 32, parent); return 0; diff --git a/drivers/irqchip/irq-uniphier-aidet.c b/drivers/irqchip/irq-uniphier-aidet.c new file mode 100644 index 000000000000..7ba7f253470e --- /dev/null +++ b/drivers/irqchip/irq-uniphier-aidet.c @@ -0,0 +1,261 @@ +/* + * Driver for UniPhier AIDET (ARM Interrupt Detector) + * + * Copyright (C) 2017 Socionext Inc. + * Author: Masahiro Yamada <yamada.masahiro@socionext.com> + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + */ + +#include <linux/bitops.h> +#include <linux/init.h> +#include <linux/irq.h> +#include <linux/irqdomain.h> +#include <linux/kernel.h> +#include <linux/of.h> +#include <linux/of_device.h> +#include <linux/of_irq.h> +#include <linux/platform_device.h> +#include <linux/spinlock.h> + +#define UNIPHIER_AIDET_NR_IRQS 256 + +#define UNIPHIER_AIDET_DETCONF 0x04 /* inverter register base */ + +struct uniphier_aidet_priv { + struct irq_domain *domain; + void __iomem *reg_base; + spinlock_t lock; + u32 saved_vals[UNIPHIER_AIDET_NR_IRQS / 32]; +}; + +static void uniphier_aidet_reg_update(struct uniphier_aidet_priv *priv, + unsigned int reg, u32 mask, u32 val) +{ + unsigned long flags; + u32 tmp; + + spin_lock_irqsave(&priv->lock, flags); + tmp = readl_relaxed(priv->reg_base + reg); + tmp &= ~mask; + tmp |= mask & val; + writel_relaxed(tmp, priv->reg_base + reg); + spin_unlock_irqrestore(&priv->lock, flags); +} + +static void uniphier_aidet_detconf_update(struct uniphier_aidet_priv *priv, + unsigned long index, unsigned int val) +{ + unsigned int reg; + u32 mask; + + reg = UNIPHIER_AIDET_DETCONF + index / 32 * 4; + mask = BIT(index % 32); + + uniphier_aidet_reg_update(priv, reg, mask, val ? mask : 0); +} + +static int uniphier_aidet_irq_set_type(struct irq_data *data, unsigned int type) +{ + struct uniphier_aidet_priv *priv = data->chip_data; + unsigned int val; + + /* enable inverter for active low triggers */ + switch (type) { + case IRQ_TYPE_EDGE_RISING: + case IRQ_TYPE_LEVEL_HIGH: + val = 0; + break; + case IRQ_TYPE_EDGE_FALLING: + val = 1; + type = IRQ_TYPE_EDGE_RISING; + break; + case IRQ_TYPE_LEVEL_LOW: + val = 1; + type = IRQ_TYPE_LEVEL_HIGH; + break; + default: + return -EINVAL; + } + + uniphier_aidet_detconf_update(priv, data->hwirq, val); + + return irq_chip_set_type_parent(data, type); +} + +static struct irq_chip uniphier_aidet_irq_chip = { + .name = "AIDET", + .irq_mask = irq_chip_mask_parent, + .irq_unmask = irq_chip_unmask_parent, + .irq_eoi = irq_chip_eoi_parent, + .irq_set_affinity = irq_chip_set_affinity_parent, + .irq_set_type = uniphier_aidet_irq_set_type, +}; + +static int uniphier_aidet_domain_translate(struct irq_domain *domain, + struct irq_fwspec *fwspec, + unsigned long *out_hwirq, + unsigned int *out_type) +{ + if (WARN_ON(fwspec->param_count < 2)) + return -EINVAL; + + *out_hwirq = fwspec->param[0]; + *out_type = fwspec->param[1] & IRQ_TYPE_SENSE_MASK; + + return 0; +} + +static int uniphier_aidet_domain_alloc(struct irq_domain *domain, + unsigned int virq, unsigned int nr_irqs, + void *arg) +{ + struct irq_fwspec parent_fwspec; + irq_hw_number_t hwirq; + unsigned int type; + int ret; + + if (nr_irqs != 1) + return -EINVAL; + + ret = uniphier_aidet_domain_translate(domain, arg, &hwirq, &type); + if (ret) + return ret; + + switch (type) { + case IRQ_TYPE_EDGE_RISING: + case IRQ_TYPE_LEVEL_HIGH: + break; + case IRQ_TYPE_EDGE_FALLING: + type = IRQ_TYPE_EDGE_RISING; + break; + case IRQ_TYPE_LEVEL_LOW: + type = IRQ_TYPE_LEVEL_HIGH; + break; + default: + return -EINVAL; + } + + if (hwirq >= UNIPHIER_AIDET_NR_IRQS) + return -ENXIO; + + ret = irq_domain_set_hwirq_and_chip(domain, virq, hwirq, + &uniphier_aidet_irq_chip, + domain->host_data); + if (ret) + return ret; + + /* parent is GIC */ + parent_fwspec.fwnode = domain->parent->fwnode; + parent_fwspec.param_count = 3; + parent_fwspec.param[0] = 0; /* SPI */ + parent_fwspec.param[1] = hwirq; + parent_fwspec.param[2] = type; + + return irq_domain_alloc_irqs_parent(domain, virq, 1, &parent_fwspec); +} + +static const struct irq_domain_ops uniphier_aidet_domain_ops = { + .alloc = uniphier_aidet_domain_alloc, + .free = irq_domain_free_irqs_common, + .translate = uniphier_aidet_domain_translate, +}; + +static int uniphier_aidet_probe(struct platform_device *pdev) +{ + struct device *dev = &pdev->dev; + struct device_node *parent_np; + struct irq_domain *parent_domain; + struct uniphier_aidet_priv *priv; + struct resource *res; + + parent_np = of_irq_find_parent(dev->of_node); + if (!parent_np) + return -ENXIO; + + parent_domain = irq_find_host(parent_np); + of_node_put(parent_np); + if (!parent_domain) + return -EPROBE_DEFER; + + priv = devm_kzalloc(dev, sizeof(*priv), GFP_KERNEL); + if (!priv) + return -ENOMEM; + + res = platform_get_resource(pdev, IORESOURCE_MEM, 0); + priv->reg_base = devm_ioremap_resource(dev, res); + if (IS_ERR(priv->reg_base)) + return PTR_ERR(priv->reg_base); + + spin_lock_init(&priv->lock); + + priv->domain = irq_domain_create_hierarchy( + parent_domain, 0, + UNIPHIER_AIDET_NR_IRQS, + of_node_to_fwnode(dev->of_node), + &uniphier_aidet_domain_ops, priv); + if (!priv->domain) + return -ENOMEM; + + platform_set_drvdata(pdev, priv); + + return 0; +} + +static int __maybe_unused uniphier_aidet_suspend(struct device *dev) +{ + struct uniphier_aidet_priv *priv = dev_get_drvdata(dev); + int i; + + for (i = 0; i < ARRAY_SIZE(priv->saved_vals); i++) + priv->saved_vals[i] = readl_relaxed( + priv->reg_base + UNIPHIER_AIDET_DETCONF + i * 4); + + return 0; +} + +static int __maybe_unused uniphier_aidet_resume(struct device *dev) +{ + struct uniphier_aidet_priv *priv = dev_get_drvdata(dev); + int i; + + for (i = 0; i < ARRAY_SIZE(priv->saved_vals); i++) + writel_relaxed(priv->saved_vals[i], + priv->reg_base + UNIPHIER_AIDET_DETCONF + i * 4); + + return 0; +} + +static const struct dev_pm_ops uniphier_aidet_pm_ops = { + SET_NOIRQ_SYSTEM_SLEEP_PM_OPS(uniphier_aidet_suspend, + uniphier_aidet_resume) +}; + +static const struct of_device_id uniphier_aidet_match[] = { + { .compatible = "socionext,uniphier-ld4-aidet" }, + { .compatible = "socionext,uniphier-pro4-aidet" }, + { .compatible = "socionext,uniphier-sld8-aidet" }, + { .compatible = "socionext,uniphier-pro5-aidet" }, + { .compatible = "socionext,uniphier-pxs2-aidet" }, + { .compatible = "socionext,uniphier-ld11-aidet" }, + { .compatible = "socionext,uniphier-ld20-aidet" }, + { .compatible = "socionext,uniphier-pxs3-aidet" }, + { /* sentinel */ } +}; + +static struct platform_driver uniphier_aidet_driver = { + .probe = uniphier_aidet_probe, + .driver = { + .name = "uniphier-aidet", + .of_match_table = uniphier_aidet_match, + .pm = &uniphier_aidet_pm_ops, + }, +}; +builtin_platform_driver(uniphier_aidet_driver); diff --git a/drivers/irqchip/irq-xilinx-intc.c b/drivers/irqchip/irq-xilinx-intc.c index 3db7ab1c9741..e3043ded8973 100644 --- a/drivers/irqchip/irq-xilinx-intc.c +++ b/drivers/irqchip/irq-xilinx-intc.c @@ -186,8 +186,8 @@ static int __init xilinx_intc_of_init(struct device_node *intc, if (irqc->intr_mask >> nr_irq) pr_warn("irq-xilinx: mismatch in kind-of-intr param\n"); - pr_info("irq-xilinx: %s: num_irq=%d, edge=0x%x\n", - intc->full_name, nr_irq, irqc->intr_mask); + pr_info("irq-xilinx: %pOF: num_irq=%d, edge=0x%x\n", + intc, nr_irq, irqc->intr_mask); /* diff --git a/drivers/irqchip/irq-xtensa-mx.c b/drivers/irqchip/irq-xtensa-mx.c index 72a391e01011..a15a9510c904 100644 --- a/drivers/irqchip/irq-xtensa-mx.c +++ b/drivers/irqchip/irq-xtensa-mx.c @@ -32,6 +32,7 @@ static int xtensa_mx_irq_map(struct irq_domain *d, unsigned int irq, irq_set_status_flags(irq, IRQ_LEVEL); return 0; } + irqd_set_single_target(irq_desc_get_irq_data(irq_to_desc(irq))); return xtensa_irq_map(d, irq, hw); } @@ -121,9 +122,12 @@ static int xtensa_mx_irq_retrigger(struct irq_data *d) static int xtensa_mx_irq_set_affinity(struct irq_data *d, const struct cpumask *dest, bool force) { - unsigned mask = 1u << cpumask_any_and(dest, cpu_online_mask); + int cpu = cpumask_any_and(dest, cpu_online_mask); + unsigned mask = 1u << cpu; set_er(mask, MIROUT(d->hwirq - HW_IRQ_MX_BASE)); + irq_data_update_effective_affinity(d, cpumask_of(cpu)); + return 0; } |