Add the rt linux 4.1.3-rt3 as base
[kvmfornfv.git] / kernel / arch / powerpc / kernel / kprobes.c
diff --git a/kernel/arch/powerpc/kernel/kprobes.c b/kernel/arch/powerpc/kernel/kprobes.c
new file mode 100644 (file)
index 0000000..7c053f2
--- /dev/null
@@ -0,0 +1,559 @@
+/*
+ *  Kernel Probes (KProbes)
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * 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, write to the Free Software
+ * Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA.
+ *
+ * Copyright (C) IBM Corporation, 2002, 2004
+ *
+ * 2002-Oct    Created by Vamsi Krishna S <vamsi_krishna@in.ibm.com> Kernel
+ *             Probes initial implementation ( includes contributions from
+ *             Rusty Russell).
+ * 2004-July   Suparna Bhattacharya <suparna@in.ibm.com> added jumper probes
+ *             interface to access function arguments.
+ * 2004-Nov    Ananth N Mavinakayanahalli <ananth@in.ibm.com> kprobes port
+ *             for PPC64
+ */
+
+#include <linux/kprobes.h>
+#include <linux/ptrace.h>
+#include <linux/preempt.h>
+#include <linux/module.h>
+#include <linux/kdebug.h>
+#include <linux/slab.h>
+#include <asm/code-patching.h>
+#include <asm/cacheflush.h>
+#include <asm/sstep.h>
+#include <asm/uaccess.h>
+
+DEFINE_PER_CPU(struct kprobe *, current_kprobe) = NULL;
+DEFINE_PER_CPU(struct kprobe_ctlblk, kprobe_ctlblk);
+
+struct kretprobe_blackpoint kretprobe_blacklist[] = {{NULL, NULL}};
+
+int __kprobes arch_prepare_kprobe(struct kprobe *p)
+{
+       int ret = 0;
+       kprobe_opcode_t insn = *p->addr;
+
+       if ((unsigned long)p->addr & 0x03) {
+               printk("Attempt to register kprobe at an unaligned address\n");
+               ret = -EINVAL;
+       } else if (IS_MTMSRD(insn) || IS_RFID(insn) || IS_RFI(insn)) {
+               printk("Cannot register a kprobe on rfi/rfid or mtmsr[d]\n");
+               ret = -EINVAL;
+       }
+
+       /* insn must be on a special executable page on ppc64.  This is
+        * not explicitly required on ppc32 (right now), but it doesn't hurt */
+       if (!ret) {
+               p->ainsn.insn = get_insn_slot();
+               if (!p->ainsn.insn)
+                       ret = -ENOMEM;
+       }
+
+       if (!ret) {
+               memcpy(p->ainsn.insn, p->addr,
+                               MAX_INSN_SIZE * sizeof(kprobe_opcode_t));
+               p->opcode = *p->addr;
+               flush_icache_range((unsigned long)p->ainsn.insn,
+                       (unsigned long)p->ainsn.insn + sizeof(kprobe_opcode_t));
+       }
+
+       p->ainsn.boostable = 0;
+       return ret;
+}
+
+void __kprobes arch_arm_kprobe(struct kprobe *p)
+{
+       *p->addr = BREAKPOINT_INSTRUCTION;
+       flush_icache_range((unsigned long) p->addr,
+                          (unsigned long) p->addr + sizeof(kprobe_opcode_t));
+}
+
+void __kprobes arch_disarm_kprobe(struct kprobe *p)
+{
+       *p->addr = p->opcode;
+       flush_icache_range((unsigned long) p->addr,
+                          (unsigned long) p->addr + sizeof(kprobe_opcode_t));
+}
+
+void __kprobes arch_remove_kprobe(struct kprobe *p)
+{
+       if (p->ainsn.insn) {
+               free_insn_slot(p->ainsn.insn, 0);
+               p->ainsn.insn = NULL;
+       }
+}
+
+static void __kprobes prepare_singlestep(struct kprobe *p, struct pt_regs *regs)
+{
+       enable_single_step(regs);
+
+       /*
+        * On powerpc we should single step on the original
+        * instruction even if the probed insn is a trap
+        * variant as values in regs could play a part in
+        * if the trap is taken or not
+        */
+       regs->nip = (unsigned long)p->ainsn.insn;
+}
+
+static void __kprobes save_previous_kprobe(struct kprobe_ctlblk *kcb)
+{
+       kcb->prev_kprobe.kp = kprobe_running();
+       kcb->prev_kprobe.status = kcb->kprobe_status;
+       kcb->prev_kprobe.saved_msr = kcb->kprobe_saved_msr;
+}
+
+static void __kprobes restore_previous_kprobe(struct kprobe_ctlblk *kcb)
+{
+       __this_cpu_write(current_kprobe, kcb->prev_kprobe.kp);
+       kcb->kprobe_status = kcb->prev_kprobe.status;
+       kcb->kprobe_saved_msr = kcb->prev_kprobe.saved_msr;
+}
+
+static void __kprobes set_current_kprobe(struct kprobe *p, struct pt_regs *regs,
+                               struct kprobe_ctlblk *kcb)
+{
+       __this_cpu_write(current_kprobe, p);
+       kcb->kprobe_saved_msr = regs->msr;
+}
+
+void __kprobes arch_prepare_kretprobe(struct kretprobe_instance *ri,
+                                     struct pt_regs *regs)
+{
+       ri->ret_addr = (kprobe_opcode_t *)regs->link;
+
+       /* Replace the return addr with trampoline addr */
+       regs->link = (unsigned long)kretprobe_trampoline;
+}
+
+static int __kprobes kprobe_handler(struct pt_regs *regs)
+{
+       struct kprobe *p;
+       int ret = 0;
+       unsigned int *addr = (unsigned int *)regs->nip;
+       struct kprobe_ctlblk *kcb;
+
+       /*
+        * We don't want to be preempted for the entire
+        * duration of kprobe processing
+        */
+       preempt_disable();
+       kcb = get_kprobe_ctlblk();
+
+       /* Check we're not actually recursing */
+       if (kprobe_running()) {
+               p = get_kprobe(addr);
+               if (p) {
+                       kprobe_opcode_t insn = *p->ainsn.insn;
+                       if (kcb->kprobe_status == KPROBE_HIT_SS &&
+                                       is_trap(insn)) {
+                               /* Turn off 'trace' bits */
+                               regs->msr &= ~MSR_SINGLESTEP;
+                               regs->msr |= kcb->kprobe_saved_msr;
+                               goto no_kprobe;
+                       }
+                       /* We have reentered the kprobe_handler(), since
+                        * another probe was hit while within the handler.
+                        * We here save the original kprobes variables and
+                        * just single step on the instruction of the new probe
+                        * without calling any user handlers.
+                        */
+                       save_previous_kprobe(kcb);
+                       set_current_kprobe(p, regs, kcb);
+                       kcb->kprobe_saved_msr = regs->msr;
+                       kprobes_inc_nmissed_count(p);
+                       prepare_singlestep(p, regs);
+                       kcb->kprobe_status = KPROBE_REENTER;
+                       return 1;
+               } else {
+                       if (*addr != BREAKPOINT_INSTRUCTION) {
+                               /* If trap variant, then it belongs not to us */
+                               kprobe_opcode_t cur_insn = *addr;
+                               if (is_trap(cur_insn))
+                                       goto no_kprobe;
+                               /* The breakpoint instruction was removed by
+                                * another cpu right after we hit, no further
+                                * handling of this interrupt is appropriate
+                                */
+                               ret = 1;
+                               goto no_kprobe;
+                       }
+                       p = __this_cpu_read(current_kprobe);
+                       if (p->break_handler && p->break_handler(p, regs)) {
+                               goto ss_probe;
+                       }
+               }
+               goto no_kprobe;
+       }
+
+       p = get_kprobe(addr);
+       if (!p) {
+               if (*addr != BREAKPOINT_INSTRUCTION) {
+                       /*
+                        * PowerPC has multiple variants of the "trap"
+                        * instruction. If the current instruction is a
+                        * trap variant, it could belong to someone else
+                        */
+                       kprobe_opcode_t cur_insn = *addr;
+                       if (is_trap(cur_insn))
+                               goto no_kprobe;
+                       /*
+                        * The breakpoint instruction was removed right
+                        * after we hit it.  Another cpu has removed
+                        * either a probepoint or a debugger breakpoint
+                        * at this address.  In either case, no further
+                        * handling of this interrupt is appropriate.
+                        */
+                       ret = 1;
+               }
+               /* Not one of ours: let kernel handle it */
+               goto no_kprobe;
+       }
+
+       kcb->kprobe_status = KPROBE_HIT_ACTIVE;
+       set_current_kprobe(p, regs, kcb);
+       if (p->pre_handler && p->pre_handler(p, regs))
+               /* handler has already set things up, so skip ss setup */
+               return 1;
+
+ss_probe:
+       if (p->ainsn.boostable >= 0) {
+               unsigned int insn = *p->ainsn.insn;
+
+               /* regs->nip is also adjusted if emulate_step returns 1 */
+               ret = emulate_step(regs, insn);
+               if (ret > 0) {
+                       /*
+                        * Once this instruction has been boosted
+                        * successfully, set the boostable flag
+                        */
+                       if (unlikely(p->ainsn.boostable == 0))
+                               p->ainsn.boostable = 1;
+
+                       if (p->post_handler)
+                               p->post_handler(p, regs, 0);
+
+                       kcb->kprobe_status = KPROBE_HIT_SSDONE;
+                       reset_current_kprobe();
+                       preempt_enable_no_resched();
+                       return 1;
+               } else if (ret < 0) {
+                       /*
+                        * We don't allow kprobes on mtmsr(d)/rfi(d), etc.
+                        * So, we should never get here... but, its still
+                        * good to catch them, just in case...
+                        */
+                       printk("Can't step on instruction %x\n", insn);
+                       BUG();
+               } else if (ret == 0)
+                       /* This instruction can't be boosted */
+                       p->ainsn.boostable = -1;
+       }
+       prepare_singlestep(p, regs);
+       kcb->kprobe_status = KPROBE_HIT_SS;
+       return 1;
+
+no_kprobe:
+       preempt_enable_no_resched();
+       return ret;
+}
+
+/*
+ * Function return probe trampoline:
+ *     - init_kprobes() establishes a probepoint here
+ *     - When the probed function returns, this probe
+ *             causes the handlers to fire
+ */
+static void __used kretprobe_trampoline_holder(void)
+{
+       asm volatile(".global kretprobe_trampoline\n"
+                       "kretprobe_trampoline:\n"
+                       "nop\n");
+}
+
+/*
+ * Called when the probe at kretprobe trampoline is hit
+ */
+static int __kprobes trampoline_probe_handler(struct kprobe *p,
+                                               struct pt_regs *regs)
+{
+       struct kretprobe_instance *ri = NULL;
+       struct hlist_head *head, empty_rp;
+       struct hlist_node *tmp;
+       unsigned long flags, orig_ret_address = 0;
+       unsigned long trampoline_address =(unsigned long)&kretprobe_trampoline;
+
+       INIT_HLIST_HEAD(&empty_rp);
+       kretprobe_hash_lock(current, &head, &flags);
+
+       /*
+        * It is possible to have multiple instances associated with a given
+        * task either because an multiple functions in the call path
+        * have a return probe installed on them, and/or more than one return
+        * return probe was registered for a target function.
+        *
+        * We can handle this because:
+        *     - instances are always inserted at the head of the list
+        *     - when multiple return probes are registered for the same
+        *       function, the first instance's ret_addr will point to the
+        *       real return address, and all the rest will point to
+        *       kretprobe_trampoline
+        */
+       hlist_for_each_entry_safe(ri, tmp, head, hlist) {
+               if (ri->task != current)
+                       /* another task is sharing our hash bucket */
+                       continue;
+
+               if (ri->rp && ri->rp->handler)
+                       ri->rp->handler(ri, regs);
+
+               orig_ret_address = (unsigned long)ri->ret_addr;
+               recycle_rp_inst(ri, &empty_rp);
+
+               if (orig_ret_address != trampoline_address)
+                       /*
+                        * This is the real return address. Any other
+                        * instances associated with this task are for
+                        * other calls deeper on the call stack
+                        */
+                       break;
+       }
+
+       kretprobe_assert(ri, orig_ret_address, trampoline_address);
+       regs->nip = orig_ret_address;
+
+       reset_current_kprobe();
+       kretprobe_hash_unlock(current, &flags);
+       preempt_enable_no_resched();
+
+       hlist_for_each_entry_safe(ri, tmp, &empty_rp, hlist) {
+               hlist_del(&ri->hlist);
+               kfree(ri);
+       }
+       /*
+        * By returning a non-zero value, we are telling
+        * kprobe_handler() that we don't want the post_handler
+        * to run (and have re-enabled preemption)
+        */
+       return 1;
+}
+
+/*
+ * Called after single-stepping.  p->addr is the address of the
+ * instruction whose first byte has been replaced by the "breakpoint"
+ * instruction.  To avoid the SMP problems that can occur when we
+ * temporarily put back the original opcode to single-step, we
+ * single-stepped a copy of the instruction.  The address of this
+ * copy is p->ainsn.insn.
+ */
+static int __kprobes post_kprobe_handler(struct pt_regs *regs)
+{
+       struct kprobe *cur = kprobe_running();
+       struct kprobe_ctlblk *kcb = get_kprobe_ctlblk();
+
+       if (!cur)
+               return 0;
+
+       /* make sure we got here for instruction we have a kprobe on */
+       if (((unsigned long)cur->ainsn.insn + 4) != regs->nip)
+               return 0;
+
+       if ((kcb->kprobe_status != KPROBE_REENTER) && cur->post_handler) {
+               kcb->kprobe_status = KPROBE_HIT_SSDONE;
+               cur->post_handler(cur, regs, 0);
+       }
+
+       /* Adjust nip to after the single-stepped instruction */
+       regs->nip = (unsigned long)cur->addr + 4;
+       regs->msr |= kcb->kprobe_saved_msr;
+
+       /*Restore back the original saved kprobes variables and continue. */
+       if (kcb->kprobe_status == KPROBE_REENTER) {
+               restore_previous_kprobe(kcb);
+               goto out;
+       }
+       reset_current_kprobe();
+out:
+       preempt_enable_no_resched();
+
+       /*
+        * if somebody else is singlestepping across a probe point, msr
+        * will have DE/SE set, in which case, continue the remaining processing
+        * of do_debug, as if this is not a probe hit.
+        */
+       if (regs->msr & MSR_SINGLESTEP)
+               return 0;
+
+       return 1;
+}
+
+int __kprobes kprobe_fault_handler(struct pt_regs *regs, int trapnr)
+{
+       struct kprobe *cur = kprobe_running();
+       struct kprobe_ctlblk *kcb = get_kprobe_ctlblk();
+       const struct exception_table_entry *entry;
+
+       switch(kcb->kprobe_status) {
+       case KPROBE_HIT_SS:
+       case KPROBE_REENTER:
+               /*
+                * We are here because the instruction being single
+                * stepped caused a page fault. We reset the current
+                * kprobe and the nip points back to the probe address
+                * and allow the page fault handler to continue as a
+                * normal page fault.
+                */
+               regs->nip = (unsigned long)cur->addr;
+               regs->msr &= ~MSR_SINGLESTEP; /* Turn off 'trace' bits */
+               regs->msr |= kcb->kprobe_saved_msr;
+               if (kcb->kprobe_status == KPROBE_REENTER)
+                       restore_previous_kprobe(kcb);
+               else
+                       reset_current_kprobe();
+               preempt_enable_no_resched();
+               break;
+       case KPROBE_HIT_ACTIVE:
+       case KPROBE_HIT_SSDONE:
+               /*
+                * We increment the nmissed count for accounting,
+                * we can also use npre/npostfault count for accounting
+                * these specific fault cases.
+                */
+               kprobes_inc_nmissed_count(cur);
+
+               /*
+                * We come here because instructions in the pre/post
+                * handler caused the page_fault, this could happen
+                * if handler tries to access user space by
+                * copy_from_user(), get_user() etc. Let the
+                * user-specified handler try to fix it first.
+                */
+               if (cur->fault_handler && cur->fault_handler(cur, regs, trapnr))
+                       return 1;
+
+               /*
+                * In case the user-specified fault handler returned
+                * zero, try to fix up.
+                */
+               if ((entry = search_exception_tables(regs->nip)) != NULL) {
+                       regs->nip = entry->fixup;
+                       return 1;
+               }
+
+               /*
+                * fixup_exception() could not handle it,
+                * Let do_page_fault() fix it.
+                */
+               break;
+       default:
+               break;
+       }
+       return 0;
+}
+
+/*
+ * Wrapper routine to for handling exceptions.
+ */
+int __kprobes kprobe_exceptions_notify(struct notifier_block *self,
+                                      unsigned long val, void *data)
+{
+       struct die_args *args = (struct die_args *)data;
+       int ret = NOTIFY_DONE;
+
+       if (args->regs && user_mode(args->regs))
+               return ret;
+
+       switch (val) {
+       case DIE_BPT:
+               if (kprobe_handler(args->regs))
+                       ret = NOTIFY_STOP;
+               break;
+       case DIE_SSTEP:
+               if (post_kprobe_handler(args->regs))
+                       ret = NOTIFY_STOP;
+               break;
+       default:
+               break;
+       }
+       return ret;
+}
+
+unsigned long arch_deref_entry_point(void *entry)
+{
+       return ppc_global_function_entry(entry);
+}
+
+int __kprobes setjmp_pre_handler(struct kprobe *p, struct pt_regs *regs)
+{
+       struct jprobe *jp = container_of(p, struct jprobe, kp);
+       struct kprobe_ctlblk *kcb = get_kprobe_ctlblk();
+
+       memcpy(&kcb->jprobe_saved_regs, regs, sizeof(struct pt_regs));
+
+       /* setup return addr to the jprobe handler routine */
+       regs->nip = arch_deref_entry_point(jp->entry);
+#ifdef CONFIG_PPC64
+#if defined(_CALL_ELF) && _CALL_ELF == 2
+       regs->gpr[12] = (unsigned long)jp->entry;
+#else
+       regs->gpr[2] = (unsigned long)(((func_descr_t *)jp->entry)->toc);
+#endif
+#endif
+
+       return 1;
+}
+
+void __used __kprobes jprobe_return(void)
+{
+       asm volatile("trap" ::: "memory");
+}
+
+static void __used __kprobes jprobe_return_end(void)
+{
+};
+
+int __kprobes longjmp_break_handler(struct kprobe *p, struct pt_regs *regs)
+{
+       struct kprobe_ctlblk *kcb = get_kprobe_ctlblk();
+
+       /*
+        * FIXME - we should ideally be validating that we got here 'cos
+        * of the "trap" in jprobe_return() above, before restoring the
+        * saved regs...
+        */
+       memcpy(regs, &kcb->jprobe_saved_regs, sizeof(struct pt_regs));
+       preempt_enable_no_resched();
+       return 1;
+}
+
+static struct kprobe trampoline_p = {
+       .addr = (kprobe_opcode_t *) &kretprobe_trampoline,
+       .pre_handler = trampoline_probe_handler
+};
+
+int __init arch_init_kprobes(void)
+{
+       return register_kprobe(&trampoline_p);
+}
+
+int __kprobes arch_trampoline_kprobe(struct kprobe *p)
+{
+       if (p->addr == (kprobe_opcode_t *)&kretprobe_trampoline)
+               return 1;
+
+       return 0;
+}