static int setup_kdump_elfcorehdr(void)
{
size_t elfcorebuf_sz;
char *elfcorebuf;
if (!OLDMEM_BASE || is_kdump_kernel())
return -EINVAL;
s390_elf_corehdr_create(&elfcorebuf, &elfcorebuf_sz);
elfcorehdr_addr = (unsigned long long) elfcorebuf;
elfcorehdr_size = elfcorebuf_sz;
return 0;
}
/* IPoIB mlx5 netdev profile */
static void mlx5i_build_nic_params(struct mlx5_core_dev *mdev,
struct mlx5e_params *params)
{
/* Override RQ params as IPoIB supports only LINKED LIST RQ for now */
mlx5e_set_rq_type_params(mdev, params, MLX5_WQ_TYPE_LINKED_LIST);
/* RQ size in ipoib by default is 512 */
params->log_rq_size = is_kdump_kernel() ?
MLX5E_PARAMS_MINIMUM_LOG_RQ_SIZE :
MLX5I_PARAMS_DEFAULT_LOG_RQ_SIZE;
params->lro_en = false;
}
static int iommu_load_old_irte(struct intel_iommu *iommu)
{
struct irte *old_ir_table;
phys_addr_t irt_phys;
unsigned int i;
size_t size;
u64 irta;
if (!is_kdump_kernel()) {
pr_warn("IRQ remapping was enabled on %s but we are not in kdump mode\n",
iommu->name);
clear_ir_pre_enabled(iommu);
iommu_disable_irq_remapping(iommu);
return -EINVAL;
}
/* Check whether the old ir-table has the same size as ours */
irta = dmar_readq(iommu->reg + DMAR_IRTA_REG);
if ((irta & INTR_REMAP_TABLE_REG_SIZE_MASK)
!= INTR_REMAP_TABLE_REG_SIZE)
return -EINVAL;
irt_phys = irta & VTD_PAGE_MASK;
size = INTR_REMAP_TABLE_ENTRIES*sizeof(struct irte);
/* Map the old IR table */
old_ir_table = memremap(irt_phys, size, MEMREMAP_WB);
if (!old_ir_table)
return -ENOMEM;
/* Copy data over */
memcpy(iommu->ir_table->base, old_ir_table, size);
__iommu_flush_cache(iommu, iommu->ir_table->base, size);
/*
* Now check the table for used entries and mark those as
* allocated in the bitmap
*/
for (i = 0; i < INTR_REMAP_TABLE_ENTRIES; i++) {
if (iommu->ir_table->base[i].present)
bitmap_set(iommu->ir_table->bitmap, i, 1);
}
memunmap(old_ir_table);
return 0;
}
/* IPoIB mlx5 netdev profile */
static void mlx5i_build_nic_params(struct mlx5_core_dev *mdev,
struct mlx5e_params *params)
{
/* Override RQ params as IPoIB supports only LINKED LIST RQ for now */
MLX5E_SET_PFLAG(params, MLX5E_PFLAG_RX_STRIDING_RQ, false);
mlx5e_set_rq_type(mdev, params);
mlx5e_init_rq_type_params(mdev, params);
/* RQ size in ipoib by default is 512 */
params->log_rq_mtu_frames = is_kdump_kernel() ?
MLX5E_PARAMS_MINIMUM_LOG_RQ_SIZE :
MLX5I_PARAMS_DEFAULT_LOG_RQ_SIZE;
params->lro_en = false;
params->hard_mtu = MLX5_IB_GRH_BYTES + MLX5_IPOIB_HARD_LEN;
}
static void __init smp_get_save_area(unsigned int cpu, unsigned int phy_cpu)
{
if (ipl_info.type != IPL_TYPE_FCP_DUMP && !OLDMEM_BASE)
return;
if (is_kdump_kernel())
return;
if (cpu >= NR_CPUS) {
pr_warning("CPU %i exceeds the maximum %i and is excluded from "
"the dump\n", cpu, NR_CPUS - 1);
return;
}
zfcpdump_save_areas[cpu] = kmalloc(sizeof(struct save_area), GFP_KERNEL);
while (raw_sigp(phy_cpu, sigp_stop_and_store_status) == sigp_busy)
cpu_relax();
memcpy_real(zfcpdump_save_areas[cpu],
(void *)(unsigned long) store_prefix() + SAVE_AREA_BASE,
sizeof(struct save_area));
}
static void iommu_table_setparms(struct pci_controller *phb,
struct device_node *dn,
struct iommu_table *tbl)
{
struct device_node *node;
const unsigned long *basep;
const u32 *sizep;
node = phb->dn;
basep = of_get_property(node, "linux,tce-base", NULL);
sizep = of_get_property(node, "linux,tce-size", NULL);
if (basep == NULL || sizep == NULL) {
printk(KERN_ERR "PCI_DMA: iommu_table_setparms: %s has "
"missing tce entries !\n", dn->full_name);
return;
}
tbl->it_base = (unsigned long)__va(*basep);
if (!is_kdump_kernel())
memset((void *)tbl->it_base, 0, *sizep);
tbl->it_busno = phb->bus->number;
/* Units of tce entries */
tbl->it_offset = phb->dma_window_base_cur >> IOMMU_PAGE_SHIFT;
/* Test if we are going over 2GB of DMA space */
if (phb->dma_window_base_cur + phb->dma_window_size > 0x80000000ul) {
udbg_printf("PCI_DMA: Unexpected number of IOAs under this PHB.\n");
panic("PCI_DMA: Unexpected number of IOAs under this PHB.\n");
}
phb->dma_window_base_cur += phb->dma_window_size;
/* Set the tce table size - measured in entries */
tbl->it_size = phb->dma_window_size >> IOMMU_PAGE_SHIFT;
tbl->it_index = 0;
tbl->it_blocksize = 16;
tbl->it_type = TCE_PCI;
}
static int intel_setup_irq_remapping(struct intel_iommu *iommu)
{
struct ir_table *ir_table;
struct fwnode_handle *fn;
unsigned long *bitmap;
struct page *pages;
if (iommu->ir_table)
return 0;
ir_table = kzalloc(sizeof(struct ir_table), GFP_KERNEL);
if (!ir_table)
return -ENOMEM;
pages = alloc_pages_node(iommu->node, GFP_KERNEL | __GFP_ZERO,
INTR_REMAP_PAGE_ORDER);
if (!pages) {
pr_err("IR%d: failed to allocate pages of order %d\n",
iommu->seq_id, INTR_REMAP_PAGE_ORDER);
goto out_free_table;
}
bitmap = kcalloc(BITS_TO_LONGS(INTR_REMAP_TABLE_ENTRIES),
sizeof(long), GFP_ATOMIC);
if (bitmap == NULL) {
pr_err("IR%d: failed to allocate bitmap\n", iommu->seq_id);
goto out_free_pages;
}
fn = irq_domain_alloc_named_id_fwnode("INTEL-IR", iommu->seq_id);
if (!fn)
goto out_free_bitmap;
iommu->ir_domain =
irq_domain_create_hierarchy(arch_get_ir_parent_domain(),
0, INTR_REMAP_TABLE_ENTRIES,
fn, &intel_ir_domain_ops,
iommu);
irq_domain_free_fwnode(fn);
if (!iommu->ir_domain) {
pr_err("IR%d: failed to allocate irqdomain\n", iommu->seq_id);
goto out_free_bitmap;
}
iommu->ir_msi_domain =
arch_create_remap_msi_irq_domain(iommu->ir_domain,
"INTEL-IR-MSI",
iommu->seq_id);
ir_table->base = page_address(pages);
ir_table->bitmap = bitmap;
iommu->ir_table = ir_table;
/*
* If the queued invalidation is already initialized,
* shouldn't disable it.
*/
if (!iommu->qi) {
/*
* Clear previous faults.
*/
dmar_fault(-1, iommu);
dmar_disable_qi(iommu);
if (dmar_enable_qi(iommu)) {
pr_err("Failed to enable queued invalidation\n");
goto out_free_bitmap;
}
}
init_ir_status(iommu);
if (ir_pre_enabled(iommu)) {
if (!is_kdump_kernel()) {
pr_warn("IRQ remapping was enabled on %s but we are not in kdump mode\n",
iommu->name);
clear_ir_pre_enabled(iommu);
iommu_disable_irq_remapping(iommu);
} else if (iommu_load_old_irte(iommu))
pr_err("Failed to copy IR table for %s from previous kernel\n",
iommu->name);
else
pr_info("Copied IR table for %s from previous kernel\n",
iommu->name);
}
iommu_set_irq_remapping(iommu, eim_mode);
return 0;
out_free_bitmap:
kfree(bitmap);
out_free_pages:
__free_pages(pages, INTR_REMAP_PAGE_ORDER);
out_free_table:
kfree(ir_table);
iommu->ir_table = NULL;
return -ENOMEM;
}
static void __init
setup_memory(void)
{
unsigned long bootmap_size;
unsigned long start_pfn, end_pfn;
int i;
/*
* partially used pages are not usable - thus
* we are rounding upwards:
*/
start_pfn = PFN_UP(__pa(&_end));
end_pfn = max_pfn = PFN_DOWN(memory_end);
#ifdef CONFIG_BLK_DEV_INITRD
/*
* Move the initrd in case the bitmap of the bootmem allocater
* would overwrite it.
*/
if (INITRD_START && INITRD_SIZE) {
unsigned long bmap_size;
unsigned long start;
bmap_size = bootmem_bootmap_pages(end_pfn - start_pfn + 1);
bmap_size = PFN_PHYS(bmap_size);
if (PFN_PHYS(start_pfn) + bmap_size > INITRD_START) {
start = PFN_PHYS(start_pfn) + bmap_size + PAGE_SIZE;
#ifdef CONFIG_CRASH_DUMP
if (OLDMEM_BASE) {
/* Move initrd behind kdump oldmem */
if (start + INITRD_SIZE > OLDMEM_BASE &&
start < OLDMEM_BASE + OLDMEM_SIZE)
start = OLDMEM_BASE + OLDMEM_SIZE;
}
#endif
if (start + INITRD_SIZE > memory_end) {
pr_err("initrd extends beyond end of "
"memory (0x%08lx > 0x%08lx) "
"disabling initrd\n",
start + INITRD_SIZE, memory_end);
INITRD_START = INITRD_SIZE = 0;
} else {
pr_info("Moving initrd (0x%08lx -> "
"0x%08lx, size: %ld)\n",
INITRD_START, start, INITRD_SIZE);
memmove((void *) start, (void *) INITRD_START,
INITRD_SIZE);
INITRD_START = start;
}
}
}
#endif
/*
* Initialize the boot-time allocator
*/
bootmap_size = init_bootmem(start_pfn, end_pfn);
/*
* Register RAM areas with the bootmem allocator.
*/
for (i = 0; i < MEMORY_CHUNKS && memory_chunk[i].size > 0; i++) {
unsigned long start_chunk, end_chunk, pfn;
if (memory_chunk[i].type != CHUNK_READ_WRITE &&
memory_chunk[i].type != CHUNK_CRASHK)
continue;
start_chunk = PFN_DOWN(memory_chunk[i].addr);
end_chunk = start_chunk + PFN_DOWN(memory_chunk[i].size);
end_chunk = min(end_chunk, end_pfn);
if (start_chunk >= end_chunk)
continue;
memblock_add_node(PFN_PHYS(start_chunk),
PFN_PHYS(end_chunk - start_chunk), 0);
pfn = max(start_chunk, start_pfn);
for (; pfn < end_chunk; pfn++)
page_set_storage_key(PFN_PHYS(pfn),
PAGE_DEFAULT_KEY, 0);
}
psw_set_key(PAGE_DEFAULT_KEY);
free_bootmem_with_active_regions(0, max_pfn);
/*
* Reserve memory used for lowcore/command line/kernel image.
*/
reserve_bootmem(0, (unsigned long)_ehead, BOOTMEM_DEFAULT);
reserve_bootmem((unsigned long)_stext,
PFN_PHYS(start_pfn) - (unsigned long)_stext,
BOOTMEM_DEFAULT);
/*
* Reserve the bootmem bitmap itself as well. We do this in two
* steps (first step was init_bootmem()) because this catches
* the (very unlikely) case of us accidentally initializing the
* bootmem allocator with an invalid RAM area.
*/
reserve_bootmem(start_pfn << PAGE_SHIFT, bootmap_size,
BOOTMEM_DEFAULT);
#ifdef CONFIG_CRASH_DUMP
if (crashk_res.start)
reserve_bootmem(crashk_res.start,
crashk_res.end - crashk_res.start + 1,
BOOTMEM_DEFAULT);
if (is_kdump_kernel())
reserve_bootmem(elfcorehdr_addr - OLDMEM_BASE,
PAGE_ALIGN(elfcorehdr_size), BOOTMEM_DEFAULT);
#endif
#ifdef CONFIG_BLK_DEV_INITRD
if (INITRD_START && INITRD_SIZE) {
if (INITRD_START + INITRD_SIZE <= memory_end) {
reserve_bootmem(INITRD_START, INITRD_SIZE,
BOOTMEM_DEFAULT);
initrd_start = INITRD_START;
initrd_end = initrd_start + INITRD_SIZE;
} else {
pr_err("initrd extends beyond end of "
"memory (0x%08lx > 0x%08lx) "
"disabling initrd\n",
initrd_start + INITRD_SIZE, memory_end);
initrd_start = initrd_end = 0;
}
}
#endif
}
static void acpi_bus_osc_support(void)
{
u32 capbuf[2];
struct acpi_osc_context context = {
.uuid_str = sb_uuid_str,
.rev = 1,
.cap.length = 8,
.cap.pointer = capbuf,
};
acpi_handle handle;
capbuf[OSC_QUERY_TYPE] = OSC_QUERY_ENABLE;
capbuf[OSC_SUPPORT_TYPE] = OSC_SB_PR3_SUPPORT; /* _PR3 is in use */
#if defined(CONFIG_ACPI_PROCESSOR_AGGREGATOR) ||\
defined(CONFIG_ACPI_PROCESSOR_AGGREGATOR_MODULE)
capbuf[OSC_SUPPORT_TYPE] |= OSC_SB_PAD_SUPPORT;
#endif
#if defined(CONFIG_ACPI_PROCESSOR) || defined(CONFIG_ACPI_PROCESSOR_MODULE)
capbuf[OSC_SUPPORT_TYPE] |= OSC_SB_PPC_OST_SUPPORT;
#endif
if (ACPI_FAILURE(acpi_get_handle(NULL, "\\_SB", &handle)))
return;
if (is_uv_system() && is_kdump_kernel()) {
/*
* There is no need to parse the OS Capabilities table
* in the crash kernel. And it should not be done, as
* that parsing includes destructive writes to io ports to
* initialize UV system controller interrupts.
*/
return;
}
if (ACPI_SUCCESS(acpi_run_osc(handle, &context)))
kfree(context.ret.pointer);
/* do we need to check the returned cap? Sounds no */
}
/* --------------------------------------------------------------------------
Event Management
-------------------------------------------------------------------------- */
#ifdef CONFIG_ACPI_PROC_EVENT
static DEFINE_SPINLOCK(acpi_bus_event_lock);
LIST_HEAD(acpi_bus_event_list);
DECLARE_WAIT_QUEUE_HEAD(acpi_bus_event_queue);
extern int event_is_open;
int acpi_bus_generate_proc_event4(const char *device_class, const char *bus_id, u8 type, int data)
{
struct acpi_bus_event *event;
unsigned long flags = 0;
/* drop event on the floor if no one's listening */
if (!event_is_open)
return 0;
event = kmalloc(sizeof(struct acpi_bus_event), GFP_ATOMIC);
if (!event)
return -ENOMEM;
strcpy(event->device_class, device_class);
strcpy(event->bus_id, bus_id);
event->type = type;
event->data = data;
spin_lock_irqsave(&acpi_bus_event_lock, flags);
list_add_tail(&event->node, &acpi_bus_event_list);
spin_unlock_irqrestore(&acpi_bus_event_lock, flags);
wake_up_interruptible(&acpi_bus_event_queue);
return 0;
}
EXPORT_SYMBOL_GPL(acpi_bus_generate_proc_event4);
int acpi_bus_generate_proc_event(struct acpi_device *device, u8 type, int data)
{
if (!device)
return -EINVAL;
return acpi_bus_generate_proc_event4(device->pnp.device_class,
device->pnp.bus_id, type, data);
}
EXPORT_SYMBOL(acpi_bus_generate_proc_event);
int acpi_bus_receive_event(struct acpi_bus_event *event)
{
unsigned long flags = 0;
struct acpi_bus_event *entry = NULL;
DECLARE_WAITQUEUE(wait, current);
if (!event)
return -EINVAL;
if (list_empty(&acpi_bus_event_list)) {
set_current_state(TASK_INTERRUPTIBLE);
add_wait_queue(&acpi_bus_event_queue, &wait);
if (list_empty(&acpi_bus_event_list))
schedule();
remove_wait_queue(&acpi_bus_event_queue, &wait);
set_current_state(TASK_RUNNING);
if (signal_pending(current))
return -ERESTARTSYS;
}
spin_lock_irqsave(&acpi_bus_event_lock, flags);
if (!list_empty(&acpi_bus_event_list)) {
entry = list_entry(acpi_bus_event_list.next,
struct acpi_bus_event, node);
list_del(&entry->node);
}
static void __init smp_detect_cpus(void)
{
unsigned int cpu, c_cpus, s_cpus;
struct sclp_cpu_info *info;
u16 boot_cpu_addr, cpu_addr;
c_cpus = 1;
s_cpus = 0;
boot_cpu_addr = __cpu_logical_map[0];
info = kmalloc(sizeof(*info), GFP_KERNEL);
if (!info)
panic("smp_detect_cpus failed to allocate memory\n");
#ifdef CONFIG_CRASH_DUMP
if (OLDMEM_BASE && !is_kdump_kernel()) {
struct save_area *save_area;
save_area = kmalloc(sizeof(*save_area), GFP_KERNEL);
if (!save_area)
panic("could not allocate memory for save area\n");
copy_oldmem_page(1, (void *) save_area, sizeof(*save_area),
0x200, 0);
zfcpdump_save_areas[0] = save_area;
}
#endif
/* Use sigp detection algorithm if sclp doesn't work. */
if (sclp_get_cpu_info(info)) {
smp_use_sigp_detection = 1;
for (cpu = 0; cpu <= MAX_CPU_ADDRESS; cpu++) {
if (cpu == boot_cpu_addr)
continue;
if (!raw_cpu_stopped(cpu))
continue;
smp_get_save_area(c_cpus, cpu);
c_cpus++;
}
goto out;
}
if (info->has_cpu_type) {
for (cpu = 0; cpu < info->combined; cpu++) {
if (info->cpu[cpu].address == boot_cpu_addr) {
smp_cpu_type = info->cpu[cpu].type;
break;
}
}
}
for (cpu = 0; cpu < info->combined; cpu++) {
if (info->has_cpu_type && info->cpu[cpu].type != smp_cpu_type)
continue;
cpu_addr = info->cpu[cpu].address;
if (cpu_addr == boot_cpu_addr)
continue;
if (!raw_cpu_stopped(cpu_addr)) {
s_cpus++;
continue;
}
smp_get_save_area(c_cpus, cpu_addr);
c_cpus++;
}
out:
kfree(info);
pr_info("%d configured CPUs, %d standby CPUs\n", c_cpus, s_cpus);
get_online_cpus();
__smp_rescan_cpus();
put_online_cpus();
}
