diff options
Diffstat (limited to 'arch/powerpc/platforms/powernv/pci-ioda.c')
-rw-r--r-- | arch/powerpc/platforms/powernv/pci-ioda.c | 312 |
1 files changed, 119 insertions, 193 deletions
diff --git a/arch/powerpc/platforms/powernv/pci-ioda.c b/arch/powerpc/platforms/powernv/pci-ioda.c index a6c92c78c9b2..dd807446801e 100644 --- a/arch/powerpc/platforms/powernv/pci-ioda.c +++ b/arch/powerpc/platforms/powernv/pci-ioda.c @@ -17,11 +17,10 @@ #include <linux/delay.h> #include <linux/string.h> #include <linux/init.h> -#include <linux/bootmem.h> +#include <linux/memblock.h> #include <linux/irq.h> #include <linux/io.h> #include <linux/msi.h> -#include <linux/memblock.h> #include <linux/iommu.h> #include <linux/rculist.h> #include <linux/sizes.h> @@ -46,17 +45,14 @@ #include "powernv.h" #include "pci.h" +#include "../../../../drivers/pci/pci.h" #define PNV_IODA1_M64_NUM 16 /* Number of M64 BARs */ #define PNV_IODA1_M64_SEGS 8 /* Segments per M64 BAR */ #define PNV_IODA1_DMA32_SEGSIZE 0x10000000 -#define POWERNV_IOMMU_DEFAULT_LEVELS 1 -#define POWERNV_IOMMU_MAX_LEVELS 5 - static const char * const pnv_phb_names[] = { "IODA1", "IODA2", "NPU_NVLINK", "NPU_OCAPI" }; -static void pnv_pci_ioda2_table_free_pages(struct iommu_table *tbl); void pe_level_printk(const struct pnv_ioda_pe *pe, const char *level, const char *fmt, ...) @@ -1976,9 +1972,10 @@ static void pnv_pci_p7ioc_tce_invalidate(struct iommu_table *tbl, mb(); /* Ensure above stores are visible */ while (start <= end) { if (rm) - __raw_rm_writeq(cpu_to_be64(start), invalidate); + __raw_rm_writeq_be(start, invalidate); else - __raw_writeq(cpu_to_be64(start), invalidate); + __raw_writeq_be(start, invalidate); + start += inc; } @@ -2006,7 +2003,7 @@ static int pnv_ioda1_tce_build(struct iommu_table *tbl, long index, static int pnv_ioda1_tce_xchg(struct iommu_table *tbl, long index, unsigned long *hpa, enum dma_data_direction *direction) { - long ret = pnv_tce_xchg(tbl, index, hpa, direction); + long ret = pnv_tce_xchg(tbl, index, hpa, direction, true); if (!ret) pnv_pci_p7ioc_tce_invalidate(tbl, index, 1, false); @@ -2017,7 +2014,7 @@ static int pnv_ioda1_tce_xchg(struct iommu_table *tbl, long index, static int pnv_ioda1_tce_xchg_rm(struct iommu_table *tbl, long index, unsigned long *hpa, enum dma_data_direction *direction) { - long ret = pnv_tce_xchg(tbl, index, hpa, direction); + long ret = pnv_tce_xchg(tbl, index, hpa, direction, false); if (!ret) pnv_pci_p7ioc_tce_invalidate(tbl, index, 1, true); @@ -2039,6 +2036,7 @@ static struct iommu_table_ops pnv_ioda1_iommu_ops = { #ifdef CONFIG_IOMMU_API .exchange = pnv_ioda1_tce_xchg, .exchange_rm = pnv_ioda1_tce_xchg_rm, + .useraddrptr = pnv_tce_useraddrptr, #endif .clear = pnv_ioda1_tce_free, .get = pnv_tce_get, @@ -2055,9 +2053,9 @@ static void pnv_pci_phb3_tce_invalidate_entire(struct pnv_phb *phb, bool rm) mb(); /* Ensure previous TCE table stores are visible */ if (rm) - __raw_rm_writeq(cpu_to_be64(val), invalidate); + __raw_rm_writeq_be(val, invalidate); else - __raw_writeq(cpu_to_be64(val), invalidate); + __raw_writeq_be(val, invalidate); } static inline void pnv_pci_phb3_tce_invalidate_pe(struct pnv_ioda_pe *pe) @@ -2067,7 +2065,7 @@ static inline void pnv_pci_phb3_tce_invalidate_pe(struct pnv_ioda_pe *pe) unsigned long val = PHB3_TCE_KILL_INVAL_PE | (pe->pe_number & 0xFF); mb(); /* Ensure above stores are visible */ - __raw_writeq(cpu_to_be64(val), invalidate); + __raw_writeq_be(val, invalidate); } static void pnv_pci_phb3_tce_invalidate(struct pnv_ioda_pe *pe, bool rm, @@ -2090,9 +2088,9 @@ static void pnv_pci_phb3_tce_invalidate(struct pnv_ioda_pe *pe, bool rm, while (start <= end) { if (rm) - __raw_rm_writeq(cpu_to_be64(start), invalidate); + __raw_rm_writeq_be(start, invalidate); else - __raw_writeq(cpu_to_be64(start), invalidate); + __raw_writeq_be(start, invalidate); start += inc; } } @@ -2170,7 +2168,7 @@ static int pnv_ioda2_tce_build(struct iommu_table *tbl, long index, static int pnv_ioda2_tce_xchg(struct iommu_table *tbl, long index, unsigned long *hpa, enum dma_data_direction *direction) { - long ret = pnv_tce_xchg(tbl, index, hpa, direction); + long ret = pnv_tce_xchg(tbl, index, hpa, direction, true); if (!ret) pnv_pci_ioda2_tce_invalidate(tbl, index, 1, false); @@ -2181,7 +2179,7 @@ static int pnv_ioda2_tce_xchg(struct iommu_table *tbl, long index, static int pnv_ioda2_tce_xchg_rm(struct iommu_table *tbl, long index, unsigned long *hpa, enum dma_data_direction *direction) { - long ret = pnv_tce_xchg(tbl, index, hpa, direction); + long ret = pnv_tce_xchg(tbl, index, hpa, direction, false); if (!ret) pnv_pci_ioda2_tce_invalidate(tbl, index, 1, true); @@ -2198,20 +2196,16 @@ static void pnv_ioda2_tce_free(struct iommu_table *tbl, long index, pnv_pci_ioda2_tce_invalidate(tbl, index, npages, false); } -static void pnv_ioda2_table_free(struct iommu_table *tbl) -{ - pnv_pci_ioda2_table_free_pages(tbl); -} - static struct iommu_table_ops pnv_ioda2_iommu_ops = { .set = pnv_ioda2_tce_build, #ifdef CONFIG_IOMMU_API .exchange = pnv_ioda2_tce_xchg, .exchange_rm = pnv_ioda2_tce_xchg_rm, + .useraddrptr = pnv_tce_useraddrptr, #endif .clear = pnv_ioda2_tce_free, .get = pnv_tce_get, - .free = pnv_ioda2_table_free, + .free = pnv_pci_ioda2_table_free_pages, }; static int pnv_pci_ioda_dev_dma_weight(struct pci_dev *dev, void *data) @@ -2461,13 +2455,9 @@ void pnv_pci_ioda2_set_bypass(struct pnv_ioda_pe *pe, bool enable) pe->tce_bypass_enabled = enable; } -static long pnv_pci_ioda2_table_alloc_pages(int nid, __u64 bus_offset, - __u32 page_shift, __u64 window_size, __u32 levels, - struct iommu_table *tbl); - static long pnv_pci_ioda2_create_table(struct iommu_table_group *table_group, int num, __u32 page_shift, __u64 window_size, __u32 levels, - struct iommu_table **ptbl) + bool alloc_userspace_copy, struct iommu_table **ptbl) { struct pnv_ioda_pe *pe = container_of(table_group, struct pnv_ioda_pe, table_group); @@ -2484,7 +2474,7 @@ static long pnv_pci_ioda2_create_table(struct iommu_table_group *table_group, ret = pnv_pci_ioda2_table_alloc_pages(nid, bus_offset, page_shift, window_size, - levels, tbl); + levels, alloc_userspace_copy, tbl); if (ret) { iommu_tce_table_put(tbl); return ret; @@ -2517,7 +2507,7 @@ static long pnv_pci_ioda2_setup_default_config(struct pnv_ioda_pe *pe) rc = pnv_pci_ioda2_create_table(&pe->table_group, 0, IOMMU_PAGE_SHIFT_4K, window_size, - POWERNV_IOMMU_DEFAULT_LEVELS, &tbl); + POWERNV_IOMMU_DEFAULT_LEVELS, false, &tbl); if (rc) { pe_err(pe, "Failed to create 32-bit TCE table, err %ld", rc); @@ -2604,7 +2594,16 @@ static unsigned long pnv_pci_ioda2_get_table_size(__u32 page_shift, tce_table_size, direct_table_size); } - return bytes; + return bytes + bytes; /* one for HW table, one for userspace copy */ +} + +static long pnv_pci_ioda2_create_table_userspace( + struct iommu_table_group *table_group, + int num, __u32 page_shift, __u64 window_size, __u32 levels, + struct iommu_table **ptbl) +{ + return pnv_pci_ioda2_create_table(table_group, + num, page_shift, window_size, levels, true, ptbl); } static void pnv_ioda2_take_ownership(struct iommu_table_group *table_group) @@ -2633,7 +2632,7 @@ static void pnv_ioda2_release_ownership(struct iommu_table_group *table_group) static struct iommu_table_group_ops pnv_pci_ioda2_ops = { .get_table_size = pnv_pci_ioda2_get_table_size, - .create_table = pnv_pci_ioda2_create_table, + .create_table = pnv_pci_ioda2_create_table_userspace, .set_window = pnv_pci_ioda2_set_window, .unset_window = pnv_pci_ioda2_unset_window, .take_ownership = pnv_ioda2_take_ownership, @@ -2681,14 +2680,23 @@ static struct pnv_ioda_pe *gpe_table_group_to_npe( static long pnv_pci_ioda2_npu_set_window(struct iommu_table_group *table_group, int num, struct iommu_table *tbl) { + struct pnv_ioda_pe *npe = gpe_table_group_to_npe(table_group); + int num2 = (num == 0) ? 1 : 0; long ret = pnv_pci_ioda2_set_window(table_group, num, tbl); if (ret) return ret; - ret = pnv_npu_set_window(gpe_table_group_to_npe(table_group), num, tbl); - if (ret) + if (table_group->tables[num2]) + pnv_npu_unset_window(npe, num2); + + ret = pnv_npu_set_window(npe, num, tbl); + if (ret) { pnv_pci_ioda2_unset_window(table_group, num); + if (table_group->tables[num2]) + pnv_npu_set_window(npe, num2, + table_group->tables[num2]); + } return ret; } @@ -2697,12 +2705,24 @@ static long pnv_pci_ioda2_npu_unset_window( struct iommu_table_group *table_group, int num) { + struct pnv_ioda_pe *npe = gpe_table_group_to_npe(table_group); + int num2 = (num == 0) ? 1 : 0; long ret = pnv_pci_ioda2_unset_window(table_group, num); if (ret) return ret; - return pnv_npu_unset_window(gpe_table_group_to_npe(table_group), num); + if (!npe->table_group.tables[num]) + return 0; + + ret = pnv_npu_unset_window(npe, num); + if (ret) + return ret; + + if (table_group->tables[num2]) + ret = pnv_npu_set_window(npe, num2, table_group->tables[num2]); + + return ret; } static void pnv_ioda2_npu_take_ownership(struct iommu_table_group *table_group) @@ -2717,7 +2737,7 @@ static void pnv_ioda2_npu_take_ownership(struct iommu_table_group *table_group) static struct iommu_table_group_ops pnv_pci_ioda2_npu_ops = { .get_table_size = pnv_pci_ioda2_get_table_size, - .create_table = pnv_pci_ioda2_create_table, + .create_table = pnv_pci_ioda2_create_table_userspace, .set_window = pnv_pci_ioda2_npu_set_window, .unset_window = pnv_pci_ioda2_npu_unset_window, .take_ownership = pnv_ioda2_npu_take_ownership, @@ -2751,142 +2771,32 @@ static void pnv_pci_ioda_setup_iommu_api(void) static void pnv_pci_ioda_setup_iommu_api(void) { }; #endif -static __be64 *pnv_pci_ioda2_table_do_alloc_pages(int nid, unsigned shift, - unsigned levels, unsigned long limit, - unsigned long *current_offset, unsigned long *total_allocated) +static unsigned long pnv_ioda_parse_tce_sizes(struct pnv_phb *phb) { - struct page *tce_mem = NULL; - __be64 *addr, *tmp; - unsigned order = max_t(unsigned, shift, PAGE_SHIFT) - PAGE_SHIFT; - unsigned long allocated = 1UL << (order + PAGE_SHIFT); - unsigned entries = 1UL << (shift - 3); - long i; - - tce_mem = alloc_pages_node(nid, GFP_KERNEL, order); - if (!tce_mem) { - pr_err("Failed to allocate a TCE memory, order=%d\n", order); - return NULL; - } - addr = page_address(tce_mem); - memset(addr, 0, allocated); - *total_allocated += allocated; - - --levels; - if (!levels) { - *current_offset += allocated; - return addr; - } - - for (i = 0; i < entries; ++i) { - tmp = pnv_pci_ioda2_table_do_alloc_pages(nid, shift, - levels, limit, current_offset, total_allocated); - if (!tmp) - break; - - addr[i] = cpu_to_be64(__pa(tmp) | - TCE_PCI_READ | TCE_PCI_WRITE); - - if (*current_offset >= limit) - break; - } - - return addr; -} - -static void pnv_pci_ioda2_table_do_free_pages(__be64 *addr, - unsigned long size, unsigned level); - -static long pnv_pci_ioda2_table_alloc_pages(int nid, __u64 bus_offset, - __u32 page_shift, __u64 window_size, __u32 levels, - struct iommu_table *tbl) -{ - void *addr; - unsigned long offset = 0, level_shift, total_allocated = 0; - const unsigned window_shift = ilog2(window_size); - unsigned entries_shift = window_shift - page_shift; - unsigned table_shift = max_t(unsigned, entries_shift + 3, PAGE_SHIFT); - const unsigned long tce_table_size = 1UL << table_shift; - - if (!levels || (levels > POWERNV_IOMMU_MAX_LEVELS)) - return -EINVAL; - - if (!is_power_of_2(window_size)) - return -EINVAL; - - /* Adjust direct table size from window_size and levels */ - entries_shift = (entries_shift + levels - 1) / levels; - level_shift = entries_shift + 3; - level_shift = max_t(unsigned, level_shift, PAGE_SHIFT); - - if ((level_shift - 3) * levels + page_shift >= 60) - return -EINVAL; - - /* Allocate TCE table */ - addr = pnv_pci_ioda2_table_do_alloc_pages(nid, level_shift, - levels, tce_table_size, &offset, &total_allocated); - - /* addr==NULL means that the first level allocation failed */ - if (!addr) - return -ENOMEM; + struct pci_controller *hose = phb->hose; + struct device_node *dn = hose->dn; + unsigned long mask = 0; + int i, rc, count; + u32 val; - /* - * First level was allocated but some lower level failed as - * we did not allocate as much as we wanted, - * release partially allocated table. - */ - if (offset < tce_table_size) { - pnv_pci_ioda2_table_do_free_pages(addr, - 1ULL << (level_shift - 3), levels - 1); - return -ENOMEM; + count = of_property_count_u32_elems(dn, "ibm,supported-tce-sizes"); + if (count <= 0) { + mask = SZ_4K | SZ_64K; + /* Add 16M for POWER8 by default */ + if (cpu_has_feature(CPU_FTR_ARCH_207S) && + !cpu_has_feature(CPU_FTR_ARCH_300)) + mask |= SZ_16M | SZ_256M; + return mask; } - /* Setup linux iommu table */ - pnv_pci_setup_iommu_table(tbl, addr, tce_table_size, bus_offset, - page_shift); - tbl->it_level_size = 1ULL << (level_shift - 3); - tbl->it_indirect_levels = levels - 1; - tbl->it_allocated_size = total_allocated; - - pr_devel("Created TCE table: ws=%08llx ts=%lx @%08llx\n", - window_size, tce_table_size, bus_offset); - - return 0; -} - -static void pnv_pci_ioda2_table_do_free_pages(__be64 *addr, - unsigned long size, unsigned level) -{ - const unsigned long addr_ul = (unsigned long) addr & - ~(TCE_PCI_READ | TCE_PCI_WRITE); - - if (level) { - long i; - u64 *tmp = (u64 *) addr_ul; - - for (i = 0; i < size; ++i) { - unsigned long hpa = be64_to_cpu(tmp[i]); - - if (!(hpa & (TCE_PCI_READ | TCE_PCI_WRITE))) - continue; - - pnv_pci_ioda2_table_do_free_pages(__va(hpa), size, - level - 1); - } + for (i = 0; i < count; i++) { + rc = of_property_read_u32_index(dn, "ibm,supported-tce-sizes", + i, &val); + if (rc == 0) + mask |= 1ULL << val; } - free_pages(addr_ul, get_order(size << 3)); -} - -static void pnv_pci_ioda2_table_free_pages(struct iommu_table *tbl) -{ - const unsigned long size = tbl->it_indirect_levels ? - tbl->it_level_size : tbl->it_size; - - if (!tbl->it_size) - return; - - pnv_pci_ioda2_table_do_free_pages((__be64 *)tbl->it_base, size, - tbl->it_indirect_levels); + return mask; } static void pnv_pci_ioda2_setup_dma_pe(struct pnv_phb *phb, @@ -2913,7 +2823,7 @@ static void pnv_pci_ioda2_setup_dma_pe(struct pnv_phb *phb, pe->table_group.max_dynamic_windows_supported = IOMMU_TABLE_GROUP_MAX_TABLES; pe->table_group.max_levels = POWERNV_IOMMU_MAX_LEVELS; - pe->table_group.pgsizes = SZ_4K | SZ_64K | SZ_16M; + pe->table_group.pgsizes = pnv_ioda_parse_tce_sizes(phb); #ifdef CONFIG_IOMMU_API pe->table_group.ops = &pnv_pci_ioda2_ops; #endif @@ -3088,7 +2998,7 @@ static void pnv_pci_ioda_fixup_iov_resources(struct pci_dev *pdev) struct pci_dn *pdn; int mul, total_vfs; - if (!pdev->is_physfn || pdev->is_added) + if (!pdev->is_physfn || pci_dev_is_added(pdev)) return; pdn = pci_get_pdn(pdev); @@ -3317,12 +3227,49 @@ static void pnv_pci_ioda_create_dbgfs(void) #endif /* CONFIG_DEBUG_FS */ } +static void pnv_pci_enable_bridge(struct pci_bus *bus) +{ + struct pci_dev *dev = bus->self; + struct pci_bus *child; + + /* Empty bus ? bail */ + if (list_empty(&bus->devices)) + return; + + /* + * If there's a bridge associated with that bus enable it. This works + * around races in the generic code if the enabling is done during + * parallel probing. This can be removed once those races have been + * fixed. + */ + if (dev) { + int rc = pci_enable_device(dev); + if (rc) + pci_err(dev, "Error enabling bridge (%d)\n", rc); + pci_set_master(dev); + } + + /* Perform the same to child busses */ + list_for_each_entry(child, &bus->children, node) + pnv_pci_enable_bridge(child); +} + +static void pnv_pci_enable_bridges(void) +{ + struct pci_controller *hose; + + list_for_each_entry(hose, &hose_list, list_node) + pnv_pci_enable_bridge(hose->bus); +} + static void pnv_pci_ioda_fixup(void) { pnv_pci_ioda_setup_PEs(); pnv_pci_ioda_setup_iommu_api(); pnv_pci_ioda_create_dbgfs(); + pnv_pci_enable_bridges(); + #ifdef CONFIG_EEH pnv_eeh_post_init(); #endif @@ -3525,7 +3472,7 @@ static resource_size_t pnv_pci_iov_resource_alignment(struct pci_dev *pdev, /* Prevent enabling devices for which we couldn't properly * assign a PE */ -bool pnv_pci_enable_device_hook(struct pci_dev *dev) +static bool pnv_pci_enable_device_hook(struct pci_dev *dev) { struct pci_controller *hose = pci_bus_to_host(dev->bus); struct pnv_phb *phb = hose->private_data; @@ -3621,7 +3568,6 @@ static void pnv_pci_ioda2_release_pe_dma(struct pnv_ioda_pe *pe) WARN_ON(pe->table_group.group); } - pnv_pci_ioda2_table_free_pages(tbl); iommu_tce_table_put(tbl); } @@ -3794,26 +3740,6 @@ static const struct pci_controller_ops pnv_npu_ocapi_ioda_controller_ops = { .shutdown = pnv_pci_ioda_shutdown, }; -#ifdef CONFIG_CXL_BASE -const struct pci_controller_ops pnv_cxl_cx4_ioda_controller_ops = { - .dma_dev_setup = pnv_pci_dma_dev_setup, - .dma_bus_setup = pnv_pci_dma_bus_setup, -#ifdef CONFIG_PCI_MSI - .setup_msi_irqs = pnv_cxl_cx4_setup_msi_irqs, - .teardown_msi_irqs = pnv_cxl_cx4_teardown_msi_irqs, -#endif - .enable_device_hook = pnv_cxl_enable_device_hook, - .disable_device = pnv_cxl_disable_device, - .release_device = pnv_pci_release_device, - .window_alignment = pnv_pci_window_alignment, - .setup_bridge = pnv_pci_setup_bridge, - .reset_secondary_bus = pnv_pci_reset_secondary_bus, - .dma_set_mask = pnv_pci_ioda_dma_set_mask, - .dma_get_required_mask = pnv_pci_ioda_dma_get_required_mask, - .shutdown = pnv_pci_ioda_shutdown, -}; -#endif - static void __init pnv_pci_init_ioda_phb(struct device_node *np, u64 hub_id, int ioda_type) { @@ -3843,7 +3769,7 @@ static void __init pnv_pci_init_ioda_phb(struct device_node *np, phb_id = be64_to_cpup(prop64); pr_debug(" PHB-ID : 0x%016llx\n", phb_id); - phb = memblock_virt_alloc(sizeof(struct pnv_phb), 0); + phb = memblock_alloc(sizeof(*phb), SMP_CACHE_BYTES); /* Allocate PCI controller */ phb->hose = hose = pcibios_alloc_controller(np); @@ -3889,7 +3815,7 @@ static void __init pnv_pci_init_ioda_phb(struct device_node *np, else phb->diag_data_size = PNV_PCI_DIAG_BUF_SIZE; - phb->diag_data = memblock_virt_alloc(phb->diag_data_size, 0); + phb->diag_data = memblock_alloc(phb->diag_data_size, SMP_CACHE_BYTES); /* Parse 32-bit and IO ranges (if any) */ pci_process_bridge_OF_ranges(hose, np, !hose->global_number); @@ -3948,7 +3874,7 @@ static void __init pnv_pci_init_ioda_phb(struct device_node *np, } pemap_off = size; size += phb->ioda.total_pe_num * sizeof(struct pnv_ioda_pe); - aux = memblock_virt_alloc(size, 0); + aux = memblock_alloc(size, SMP_CACHE_BYTES); phb->ioda.pe_alloc = aux; phb->ioda.m64_segmap = aux + m64map_off; phb->ioda.m32_segmap = aux + m32map_off; |