linux/arch/s390/kernel/irq.c
Heiko Carstens 6472a2dcc4 s390/irq,idle: simplify idle check
Use the per-cpu CIF_ENABLED_WAIT flag to decide if an interrupt
occurred while a cpu was idle, instead of checking two conditions
within the old psw.

Also move clearing of the CIF_ENABLED_WAIT bit to the early interrupt
handler, which in turn makes arch_vcpu_is_preempted() also a bit more
precise, since the flag is now cleared before interrupt handlers have
been called.

Reviewed-by: Sven Schnelle <svens@linux.ibm.com>
Signed-off-by: Heiko Carstens <hca@linux.ibm.com>
2023-02-15 11:07:01 +01:00

403 lines
11 KiB
C

// SPDX-License-Identifier: GPL-2.0
/*
* Copyright IBM Corp. 2004, 2011
* Author(s): Martin Schwidefsky <schwidefsky@de.ibm.com>,
* Holger Smolinski <Holger.Smolinski@de.ibm.com>,
* Thomas Spatzier <tspat@de.ibm.com>,
*
* This file contains interrupt related functions.
*/
#include <linux/kernel_stat.h>
#include <linux/interrupt.h>
#include <linux/seq_file.h>
#include <linux/proc_fs.h>
#include <linux/profile.h>
#include <linux/export.h>
#include <linux/kernel.h>
#include <linux/ftrace.h>
#include <linux/errno.h>
#include <linux/slab.h>
#include <linux/init.h>
#include <linux/cpu.h>
#include <linux/irq.h>
#include <linux/entry-common.h>
#include <asm/irq_regs.h>
#include <asm/cputime.h>
#include <asm/lowcore.h>
#include <asm/irq.h>
#include <asm/hw_irq.h>
#include <asm/stacktrace.h>
#include <asm/softirq_stack.h>
#include "entry.h"
DEFINE_PER_CPU_SHARED_ALIGNED(struct irq_stat, irq_stat);
EXPORT_PER_CPU_SYMBOL_GPL(irq_stat);
struct irq_class {
int irq;
char *name;
char *desc;
};
/*
* The list of "main" irq classes on s390. This is the list of interrupts
* that appear both in /proc/stat ("intr" line) and /proc/interrupts.
* Historically only external and I/O interrupts have been part of /proc/stat.
* We can't add the split external and I/O sub classes since the first field
* in the "intr" line in /proc/stat is supposed to be the sum of all other
* fields.
* Since the external and I/O interrupt fields are already sums we would end
* up with having a sum which accounts each interrupt twice.
*/
static const struct irq_class irqclass_main_desc[NR_IRQS_BASE] = {
{.irq = EXT_INTERRUPT, .name = "EXT"},
{.irq = IO_INTERRUPT, .name = "I/O"},
{.irq = THIN_INTERRUPT, .name = "AIO"},
};
/*
* The list of split external and I/O interrupts that appear only in
* /proc/interrupts.
* In addition this list contains non external / I/O events like NMIs.
*/
static const struct irq_class irqclass_sub_desc[] = {
{.irq = IRQEXT_CLK, .name = "CLK", .desc = "[EXT] Clock Comparator"},
{.irq = IRQEXT_EXC, .name = "EXC", .desc = "[EXT] External Call"},
{.irq = IRQEXT_EMS, .name = "EMS", .desc = "[EXT] Emergency Signal"},
{.irq = IRQEXT_TMR, .name = "TMR", .desc = "[EXT] CPU Timer"},
{.irq = IRQEXT_TLA, .name = "TAL", .desc = "[EXT] Timing Alert"},
{.irq = IRQEXT_PFL, .name = "PFL", .desc = "[EXT] Pseudo Page Fault"},
{.irq = IRQEXT_DSD, .name = "DSD", .desc = "[EXT] DASD Diag"},
{.irq = IRQEXT_VRT, .name = "VRT", .desc = "[EXT] Virtio"},
{.irq = IRQEXT_SCP, .name = "SCP", .desc = "[EXT] Service Call"},
{.irq = IRQEXT_IUC, .name = "IUC", .desc = "[EXT] IUCV"},
{.irq = IRQEXT_CMS, .name = "CMS", .desc = "[EXT] CPU-Measurement: Sampling"},
{.irq = IRQEXT_CMC, .name = "CMC", .desc = "[EXT] CPU-Measurement: Counter"},
{.irq = IRQEXT_FTP, .name = "FTP", .desc = "[EXT] HMC FTP Service"},
{.irq = IRQIO_CIO, .name = "CIO", .desc = "[I/O] Common I/O Layer Interrupt"},
{.irq = IRQIO_DAS, .name = "DAS", .desc = "[I/O] DASD"},
{.irq = IRQIO_C15, .name = "C15", .desc = "[I/O] 3215"},
{.irq = IRQIO_C70, .name = "C70", .desc = "[I/O] 3270"},
{.irq = IRQIO_TAP, .name = "TAP", .desc = "[I/O] Tape"},
{.irq = IRQIO_VMR, .name = "VMR", .desc = "[I/O] Unit Record Devices"},
{.irq = IRQIO_LCS, .name = "LCS", .desc = "[I/O] LCS"},
{.irq = IRQIO_CTC, .name = "CTC", .desc = "[I/O] CTC"},
{.irq = IRQIO_ADM, .name = "ADM", .desc = "[I/O] EADM Subchannel"},
{.irq = IRQIO_CSC, .name = "CSC", .desc = "[I/O] CHSC Subchannel"},
{.irq = IRQIO_VIR, .name = "VIR", .desc = "[I/O] Virtual I/O Devices"},
{.irq = IRQIO_QAI, .name = "QAI", .desc = "[AIO] QDIO Adapter Interrupt"},
{.irq = IRQIO_APB, .name = "APB", .desc = "[AIO] AP Bus"},
{.irq = IRQIO_PCF, .name = "PCF", .desc = "[AIO] PCI Floating Interrupt"},
{.irq = IRQIO_PCD, .name = "PCD", .desc = "[AIO] PCI Directed Interrupt"},
{.irq = IRQIO_MSI, .name = "MSI", .desc = "[AIO] MSI Interrupt"},
{.irq = IRQIO_VAI, .name = "VAI", .desc = "[AIO] Virtual I/O Devices AI"},
{.irq = IRQIO_GAL, .name = "GAL", .desc = "[AIO] GIB Alert"},
{.irq = NMI_NMI, .name = "NMI", .desc = "[NMI] Machine Check"},
{.irq = CPU_RST, .name = "RST", .desc = "[CPU] CPU Restart"},
};
static void do_IRQ(struct pt_regs *regs, int irq)
{
if (tod_after_eq(S390_lowcore.int_clock,
S390_lowcore.clock_comparator))
/* Serve timer interrupts first. */
clock_comparator_work();
generic_handle_irq(irq);
}
static int on_async_stack(void)
{
unsigned long frame = current_frame_address();
return ((S390_lowcore.async_stack ^ frame) & ~(THREAD_SIZE - 1)) == 0;
}
static void do_irq_async(struct pt_regs *regs, int irq)
{
if (on_async_stack()) {
do_IRQ(regs, irq);
} else {
call_on_stack(2, S390_lowcore.async_stack, void, do_IRQ,
struct pt_regs *, regs, int, irq);
}
}
static int irq_pending(struct pt_regs *regs)
{
int cc;
asm volatile("tpi 0\n"
"ipm %0" : "=d" (cc) : : "cc");
return cc >> 28;
}
void noinstr do_io_irq(struct pt_regs *regs)
{
irqentry_state_t state = irqentry_enter(regs);
struct pt_regs *old_regs = set_irq_regs(regs);
bool from_idle;
irq_enter_rcu();
if (user_mode(regs)) {
update_timer_sys();
if (static_branch_likely(&cpu_has_bear))
current->thread.last_break = regs->last_break;
}
from_idle = test_and_clear_cpu_flag(CIF_ENABLED_WAIT);
if (from_idle)
account_idle_time_irq();
do {
regs->tpi_info = S390_lowcore.tpi_info;
if (S390_lowcore.tpi_info.adapter_IO)
do_irq_async(regs, THIN_INTERRUPT);
else
do_irq_async(regs, IO_INTERRUPT);
} while (MACHINE_IS_LPAR && irq_pending(regs));
irq_exit_rcu();
set_irq_regs(old_regs);
irqentry_exit(regs, state);
if (from_idle)
regs->psw.mask &= ~(PSW_MASK_EXT | PSW_MASK_IO | PSW_MASK_WAIT);
}
void noinstr do_ext_irq(struct pt_regs *regs)
{
irqentry_state_t state = irqentry_enter(regs);
struct pt_regs *old_regs = set_irq_regs(regs);
bool from_idle;
irq_enter_rcu();
if (user_mode(regs)) {
update_timer_sys();
if (static_branch_likely(&cpu_has_bear))
current->thread.last_break = regs->last_break;
}
regs->int_code = S390_lowcore.ext_int_code_addr;
regs->int_parm = S390_lowcore.ext_params;
regs->int_parm_long = S390_lowcore.ext_params2;
from_idle = test_and_clear_cpu_flag(CIF_ENABLED_WAIT);
if (from_idle)
account_idle_time_irq();
do_irq_async(regs, EXT_INTERRUPT);
irq_exit_rcu();
set_irq_regs(old_regs);
irqentry_exit(regs, state);
if (from_idle)
regs->psw.mask &= ~(PSW_MASK_EXT | PSW_MASK_IO | PSW_MASK_WAIT);
}
static void show_msi_interrupt(struct seq_file *p, int irq)
{
struct irq_desc *desc;
unsigned long flags;
int cpu;
rcu_read_lock();
desc = irq_to_desc(irq);
if (!desc)
goto out;
raw_spin_lock_irqsave(&desc->lock, flags);
seq_printf(p, "%3d: ", irq);
for_each_online_cpu(cpu)
seq_printf(p, "%10u ", irq_desc_kstat_cpu(desc, cpu));
if (desc->irq_data.chip)
seq_printf(p, " %8s", desc->irq_data.chip->name);
if (desc->action)
seq_printf(p, " %s", desc->action->name);
seq_putc(p, '\n');
raw_spin_unlock_irqrestore(&desc->lock, flags);
out:
rcu_read_unlock();
}
/*
* show_interrupts is needed by /proc/interrupts.
*/
int show_interrupts(struct seq_file *p, void *v)
{
int index = *(loff_t *) v;
int cpu, irq;
cpus_read_lock();
if (index == 0) {
seq_puts(p, " ");
for_each_online_cpu(cpu)
seq_printf(p, "CPU%-8d", cpu);
seq_putc(p, '\n');
}
if (index < NR_IRQS_BASE) {
seq_printf(p, "%s: ", irqclass_main_desc[index].name);
irq = irqclass_main_desc[index].irq;
for_each_online_cpu(cpu)
seq_printf(p, "%10u ", kstat_irqs_cpu(irq, cpu));
seq_putc(p, '\n');
goto out;
}
if (index < nr_irqs) {
show_msi_interrupt(p, index);
goto out;
}
for (index = 0; index < NR_ARCH_IRQS; index++) {
seq_printf(p, "%s: ", irqclass_sub_desc[index].name);
irq = irqclass_sub_desc[index].irq;
for_each_online_cpu(cpu)
seq_printf(p, "%10u ",
per_cpu(irq_stat, cpu).irqs[irq]);
if (irqclass_sub_desc[index].desc)
seq_printf(p, " %s", irqclass_sub_desc[index].desc);
seq_putc(p, '\n');
}
out:
cpus_read_unlock();
return 0;
}
unsigned int arch_dynirq_lower_bound(unsigned int from)
{
return from < NR_IRQS_BASE ? NR_IRQS_BASE : from;
}
/*
* ext_int_hash[index] is the list head for all external interrupts that hash
* to this index.
*/
static struct hlist_head ext_int_hash[32] ____cacheline_aligned;
struct ext_int_info {
ext_int_handler_t handler;
struct hlist_node entry;
struct rcu_head rcu;
u16 code;
};
/* ext_int_hash_lock protects the handler lists for external interrupts */
static DEFINE_SPINLOCK(ext_int_hash_lock);
static inline int ext_hash(u16 code)
{
BUILD_BUG_ON(!is_power_of_2(ARRAY_SIZE(ext_int_hash)));
return (code + (code >> 9)) & (ARRAY_SIZE(ext_int_hash) - 1);
}
int register_external_irq(u16 code, ext_int_handler_t handler)
{
struct ext_int_info *p;
unsigned long flags;
int index;
p = kmalloc(sizeof(*p), GFP_ATOMIC);
if (!p)
return -ENOMEM;
p->code = code;
p->handler = handler;
index = ext_hash(code);
spin_lock_irqsave(&ext_int_hash_lock, flags);
hlist_add_head_rcu(&p->entry, &ext_int_hash[index]);
spin_unlock_irqrestore(&ext_int_hash_lock, flags);
return 0;
}
EXPORT_SYMBOL(register_external_irq);
int unregister_external_irq(u16 code, ext_int_handler_t handler)
{
struct ext_int_info *p;
unsigned long flags;
int index = ext_hash(code);
spin_lock_irqsave(&ext_int_hash_lock, flags);
hlist_for_each_entry_rcu(p, &ext_int_hash[index], entry) {
if (p->code == code && p->handler == handler) {
hlist_del_rcu(&p->entry);
kfree_rcu(p, rcu);
}
}
spin_unlock_irqrestore(&ext_int_hash_lock, flags);
return 0;
}
EXPORT_SYMBOL(unregister_external_irq);
static irqreturn_t do_ext_interrupt(int irq, void *dummy)
{
struct pt_regs *regs = get_irq_regs();
struct ext_code ext_code;
struct ext_int_info *p;
int index;
ext_code.int_code = regs->int_code;
if (ext_code.code != EXT_IRQ_CLK_COMP)
set_cpu_flag(CIF_NOHZ_DELAY);
index = ext_hash(ext_code.code);
rcu_read_lock();
hlist_for_each_entry_rcu(p, &ext_int_hash[index], entry) {
if (unlikely(p->code != ext_code.code))
continue;
p->handler(ext_code, regs->int_parm, regs->int_parm_long);
}
rcu_read_unlock();
return IRQ_HANDLED;
}
static void __init init_ext_interrupts(void)
{
int idx;
for (idx = 0; idx < ARRAY_SIZE(ext_int_hash); idx++)
INIT_HLIST_HEAD(&ext_int_hash[idx]);
irq_set_chip_and_handler(EXT_INTERRUPT,
&dummy_irq_chip, handle_percpu_irq);
if (request_irq(EXT_INTERRUPT, do_ext_interrupt, 0, "EXT", NULL))
panic("Failed to register EXT interrupt\n");
}
void __init init_IRQ(void)
{
BUILD_BUG_ON(ARRAY_SIZE(irqclass_sub_desc) != NR_ARCH_IRQS);
init_cio_interrupts();
init_airq_interrupts();
init_ext_interrupts();
}
static DEFINE_SPINLOCK(irq_subclass_lock);
static unsigned char irq_subclass_refcount[64];
void irq_subclass_register(enum irq_subclass subclass)
{
spin_lock(&irq_subclass_lock);
if (!irq_subclass_refcount[subclass])
ctl_set_bit(0, subclass);
irq_subclass_refcount[subclass]++;
spin_unlock(&irq_subclass_lock);
}
EXPORT_SYMBOL(irq_subclass_register);
void irq_subclass_unregister(enum irq_subclass subclass)
{
spin_lock(&irq_subclass_lock);
irq_subclass_refcount[subclass]--;
if (!irq_subclass_refcount[subclass])
ctl_clear_bit(0, subclass);
spin_unlock(&irq_subclass_lock);
}
EXPORT_SYMBOL(irq_subclass_unregister);