summaryrefslogtreecommitdiffstats
path: root/arch/mips/sibyte/sb1250/irq.c
diff options
context:
space:
mode:
Diffstat (limited to 'arch/mips/sibyte/sb1250/irq.c')
-rw-r--r--arch/mips/sibyte/sb1250/irq.c318
1 files changed, 318 insertions, 0 deletions
diff --git a/arch/mips/sibyte/sb1250/irq.c b/arch/mips/sibyte/sb1250/irq.c
new file mode 100644
index 000000000..2f08ad267
--- /dev/null
+++ b/arch/mips/sibyte/sb1250/irq.c
@@ -0,0 +1,318 @@
+// SPDX-License-Identifier: GPL-2.0-or-later
+/*
+ * Copyright (C) 2000, 2001, 2002, 2003 Broadcom Corporation
+ */
+#include <linux/kernel.h>
+#include <linux/init.h>
+#include <linux/linkage.h>
+#include <linux/interrupt.h>
+#include <linux/spinlock.h>
+#include <linux/smp.h>
+#include <linux/mm.h>
+#include <linux/kernel_stat.h>
+
+#include <asm/errno.h>
+#include <asm/signal.h>
+#include <asm/time.h>
+#include <asm/io.h>
+
+#include <asm/sibyte/sb1250_regs.h>
+#include <asm/sibyte/sb1250_int.h>
+#include <asm/sibyte/sb1250_uart.h>
+#include <asm/sibyte/sb1250_scd.h>
+#include <asm/sibyte/sb1250.h>
+
+/*
+ * These are the routines that handle all the low level interrupt stuff.
+ * Actions handled here are: initialization of the interrupt map, requesting of
+ * interrupt lines by handlers, dispatching if interrupts to handlers, probing
+ * for interrupt lines
+ */
+
+#ifdef CONFIG_SIBYTE_HAS_LDT
+extern unsigned long ldt_eoi_space;
+#endif
+
+/* Store the CPU id (not the logical number) */
+int sb1250_irq_owner[SB1250_NR_IRQS];
+
+static DEFINE_RAW_SPINLOCK(sb1250_imr_lock);
+
+void sb1250_mask_irq(int cpu, int irq)
+{
+ unsigned long flags;
+ u64 cur_ints;
+
+ raw_spin_lock_irqsave(&sb1250_imr_lock, flags);
+ cur_ints = ____raw_readq(IOADDR(A_IMR_MAPPER(cpu) +
+ R_IMR_INTERRUPT_MASK));
+ cur_ints |= (((u64) 1) << irq);
+ ____raw_writeq(cur_ints, IOADDR(A_IMR_MAPPER(cpu) +
+ R_IMR_INTERRUPT_MASK));
+ raw_spin_unlock_irqrestore(&sb1250_imr_lock, flags);
+}
+
+void sb1250_unmask_irq(int cpu, int irq)
+{
+ unsigned long flags;
+ u64 cur_ints;
+
+ raw_spin_lock_irqsave(&sb1250_imr_lock, flags);
+ cur_ints = ____raw_readq(IOADDR(A_IMR_MAPPER(cpu) +
+ R_IMR_INTERRUPT_MASK));
+ cur_ints &= ~(((u64) 1) << irq);
+ ____raw_writeq(cur_ints, IOADDR(A_IMR_MAPPER(cpu) +
+ R_IMR_INTERRUPT_MASK));
+ raw_spin_unlock_irqrestore(&sb1250_imr_lock, flags);
+}
+
+#ifdef CONFIG_SMP
+static int sb1250_set_affinity(struct irq_data *d, const struct cpumask *mask,
+ bool force)
+{
+ int i = 0, old_cpu, cpu, int_on;
+ unsigned int irq = d->irq;
+ u64 cur_ints;
+ unsigned long flags;
+
+ i = cpumask_first_and(mask, cpu_online_mask);
+
+ /* Convert logical CPU to physical CPU */
+ cpu = cpu_logical_map(i);
+
+ /* Protect against other affinity changers and IMR manipulation */
+ raw_spin_lock_irqsave(&sb1250_imr_lock, flags);
+
+ /* Swizzle each CPU's IMR (but leave the IP selection alone) */
+ old_cpu = sb1250_irq_owner[irq];
+ cur_ints = ____raw_readq(IOADDR(A_IMR_MAPPER(old_cpu) +
+ R_IMR_INTERRUPT_MASK));
+ int_on = !(cur_ints & (((u64) 1) << irq));
+ if (int_on) {
+ /* If it was on, mask it */
+ cur_ints |= (((u64) 1) << irq);
+ ____raw_writeq(cur_ints, IOADDR(A_IMR_MAPPER(old_cpu) +
+ R_IMR_INTERRUPT_MASK));
+ }
+ sb1250_irq_owner[irq] = cpu;
+ if (int_on) {
+ /* unmask for the new CPU */
+ cur_ints = ____raw_readq(IOADDR(A_IMR_MAPPER(cpu) +
+ R_IMR_INTERRUPT_MASK));
+ cur_ints &= ~(((u64) 1) << irq);
+ ____raw_writeq(cur_ints, IOADDR(A_IMR_MAPPER(cpu) +
+ R_IMR_INTERRUPT_MASK));
+ }
+ raw_spin_unlock_irqrestore(&sb1250_imr_lock, flags);
+
+ return 0;
+}
+#endif
+
+static void disable_sb1250_irq(struct irq_data *d)
+{
+ unsigned int irq = d->irq;
+
+ sb1250_mask_irq(sb1250_irq_owner[irq], irq);
+}
+
+static void enable_sb1250_irq(struct irq_data *d)
+{
+ unsigned int irq = d->irq;
+
+ sb1250_unmask_irq(sb1250_irq_owner[irq], irq);
+}
+
+
+static void ack_sb1250_irq(struct irq_data *d)
+{
+ unsigned int irq = d->irq;
+#ifdef CONFIG_SIBYTE_HAS_LDT
+ u64 pending;
+
+ /*
+ * If the interrupt was an HT interrupt, now is the time to
+ * clear it. NOTE: we assume the HT bridge was set up to
+ * deliver the interrupts to all CPUs (which makes affinity
+ * changing easier for us)
+ */
+ pending = __raw_readq(IOADDR(A_IMR_REGISTER(sb1250_irq_owner[irq],
+ R_IMR_LDT_INTERRUPT)));
+ pending &= ((u64)1 << (irq));
+ if (pending) {
+ int i;
+ for (i=0; i<NR_CPUS; i++) {
+ int cpu;
+#ifdef CONFIG_SMP
+ cpu = cpu_logical_map(i);
+#else
+ cpu = i;
+#endif
+ /*
+ * Clear for all CPUs so an affinity switch
+ * doesn't find an old status
+ */
+ __raw_writeq(pending,
+ IOADDR(A_IMR_REGISTER(cpu,
+ R_IMR_LDT_INTERRUPT_CLR)));
+ }
+
+ /*
+ * Generate EOI. For Pass 1 parts, EOI is a nop. For
+ * Pass 2, the LDT world may be edge-triggered, but
+ * this EOI shouldn't hurt. If they are
+ * level-sensitive, the EOI is required.
+ */
+ *(uint32_t *)(ldt_eoi_space+(irq<<16)+(7<<2)) = 0;
+ }
+#endif
+ sb1250_mask_irq(sb1250_irq_owner[irq], irq);
+}
+
+static struct irq_chip sb1250_irq_type = {
+ .name = "SB1250-IMR",
+ .irq_mask_ack = ack_sb1250_irq,
+ .irq_unmask = enable_sb1250_irq,
+ .irq_mask = disable_sb1250_irq,
+#ifdef CONFIG_SMP
+ .irq_set_affinity = sb1250_set_affinity
+#endif
+};
+
+void __init init_sb1250_irqs(void)
+{
+ int i;
+
+ for (i = 0; i < SB1250_NR_IRQS; i++) {
+ irq_set_chip_and_handler(i, &sb1250_irq_type,
+ handle_level_irq);
+ sb1250_irq_owner[i] = 0;
+ }
+}
+
+
+/*
+ * arch_init_irq is called early in the boot sequence from init/main.c via
+ * init_IRQ. It is responsible for setting up the interrupt mapper and
+ * installing the handler that will be responsible for dispatching interrupts
+ * to the "right" place.
+ */
+/*
+ * For now, map all interrupts to IP[2]. We could save
+ * some cycles by parceling out system interrupts to different
+ * IP lines, but keep it simple for bringup. We'll also direct
+ * all interrupts to a single CPU; we should probably route
+ * PCI and LDT to one cpu and everything else to the other
+ * to balance the load a bit.
+ *
+ * On the second cpu, everything is set to IP5, which is
+ * ignored, EXCEPT the mailbox interrupt. That one is
+ * set to IP[2] so it is handled. This is needed so we
+ * can do cross-cpu function calls, as required by SMP
+ */
+
+#define IMR_IP2_VAL K_INT_MAP_I0
+#define IMR_IP3_VAL K_INT_MAP_I1
+#define IMR_IP4_VAL K_INT_MAP_I2
+#define IMR_IP5_VAL K_INT_MAP_I3
+#define IMR_IP6_VAL K_INT_MAP_I4
+
+void __init arch_init_irq(void)
+{
+
+ unsigned int i;
+ u64 tmp;
+ unsigned int imask = STATUSF_IP4 | STATUSF_IP3 | STATUSF_IP2 |
+ STATUSF_IP1 | STATUSF_IP0;
+
+ /* Default everything to IP2 */
+ for (i = 0; i < SB1250_NR_IRQS; i++) { /* was I0 */
+ __raw_writeq(IMR_IP2_VAL,
+ IOADDR(A_IMR_REGISTER(0,
+ R_IMR_INTERRUPT_MAP_BASE) +
+ (i << 3)));
+ __raw_writeq(IMR_IP2_VAL,
+ IOADDR(A_IMR_REGISTER(1,
+ R_IMR_INTERRUPT_MAP_BASE) +
+ (i << 3)));
+ }
+
+ init_sb1250_irqs();
+
+ /*
+ * Map the high 16 bits of the mailbox registers to IP[3], for
+ * inter-cpu messages
+ */
+ /* Was I1 */
+ __raw_writeq(IMR_IP3_VAL,
+ IOADDR(A_IMR_REGISTER(0, R_IMR_INTERRUPT_MAP_BASE) +
+ (K_INT_MBOX_0 << 3)));
+ __raw_writeq(IMR_IP3_VAL,
+ IOADDR(A_IMR_REGISTER(1, R_IMR_INTERRUPT_MAP_BASE) +
+ (K_INT_MBOX_0 << 3)));
+
+ /* Clear the mailboxes. The firmware may leave them dirty */
+ __raw_writeq(0xffffffffffffffffULL,
+ IOADDR(A_IMR_REGISTER(0, R_IMR_MAILBOX_CLR_CPU)));
+ __raw_writeq(0xffffffffffffffffULL,
+ IOADDR(A_IMR_REGISTER(1, R_IMR_MAILBOX_CLR_CPU)));
+
+ /* Mask everything except the mailbox registers for both cpus */
+ tmp = ~((u64) 0) ^ (((u64) 1) << K_INT_MBOX_0);
+ __raw_writeq(tmp, IOADDR(A_IMR_REGISTER(0, R_IMR_INTERRUPT_MASK)));
+ __raw_writeq(tmp, IOADDR(A_IMR_REGISTER(1, R_IMR_INTERRUPT_MASK)));
+
+ /* Enable necessary IPs, disable the rest */
+ change_c0_status(ST0_IM, imask);
+}
+
+extern void sb1250_mailbox_interrupt(void);
+
+static inline void dispatch_ip2(void)
+{
+ unsigned int cpu = smp_processor_id();
+ unsigned long long mask;
+
+ /*
+ * Default...we've hit an IP[2] interrupt, which means we've got to
+ * check the 1250 interrupt registers to figure out what to do. Need
+ * to detect which CPU we're on, now that smp_affinity is supported.
+ */
+ mask = __raw_readq(IOADDR(A_IMR_REGISTER(cpu,
+ R_IMR_INTERRUPT_STATUS_BASE)));
+ if (mask)
+ do_IRQ(fls64(mask) - 1);
+}
+
+asmlinkage void plat_irq_dispatch(void)
+{
+ unsigned int cpu = smp_processor_id();
+ unsigned int pending;
+
+ /*
+ * What a pain. We have to be really careful saving the upper 32 bits
+ * of any * register across function calls if we don't want them
+ * trashed--since were running in -o32, the calling routing never saves
+ * the full 64 bits of a register across a function call. Being the
+ * interrupt handler, we're guaranteed that interrupts are disabled
+ * during this code so we don't have to worry about random interrupts
+ * blasting the high 32 bits.
+ */
+
+ pending = read_c0_cause() & read_c0_status() & ST0_IM;
+
+ if (pending & CAUSEF_IP7) /* CPU performance counter interrupt */
+ do_IRQ(MIPS_CPU_IRQ_BASE + 7);
+ else if (pending & CAUSEF_IP4)
+ do_IRQ(K_INT_TIMER_0 + cpu); /* sb1250_timer_interrupt() */
+
+#ifdef CONFIG_SMP
+ else if (pending & CAUSEF_IP3)
+ sb1250_mailbox_interrupt();
+#endif
+
+ else if (pending & CAUSEF_IP2)
+ dispatch_ip2();
+ else
+ spurious_interrupt();
+}