1 files changed, 337 insertions, 23 deletions

diff --git a/kernel/kprobes.c b/kernel/kprobes.c
index 037142b72a49..90c0e82b650c 100644
--- a/kernel/kprobes.c
+++ b/kernel/kprobes.c
@@ -27,12 +27,16 @@
  *              interface to access function arguments.
  * 2004-Sep     Prasanna S Panchamukhi <prasanna@in.ibm.com> Changed Kprobes
  *              exceptions notifier to be first on the priority list.
+ * 2005-May     Hien Nguyen <hien@us.ibm.com>, Jim Keniston
+ *              <jkenisto@us.ibm.com> and Prasanna S Panchamukhi
+ *              <prasanna@in.ibm.com> added function-return probes.
  */
 #include <linux/kprobes.h>
 #include <linux/spinlock.h>
 #include <linux/hash.h>
 #include <linux/init.h>
 #include <linux/module.h>
+#include <linux/moduleloader.h>
 #include <asm/cacheflush.h>
 #include <asm/errno.h>
 #include <asm/kdebug.h>
@@ -41,11 +45,112 @@
 #define KPROBE_TABLE_SIZE (1 << KPROBE_HASH_BITS)
 
 static struct hlist_head kprobe_table[KPROBE_TABLE_SIZE];
+static struct hlist_head kretprobe_inst_table[KPROBE_TABLE_SIZE];
 
 unsigned int kprobe_cpu = NR_CPUS;
 static DEFINE_SPINLOCK(kprobe_lock);
 static struct kprobe *curr_kprobe;
 
+/*
+ * kprobe->ainsn.insn points to the copy of the instruction to be
+ * single-stepped. x86_64, POWER4 and above have no-exec support and
+ * stepping on the instruction on a vmalloced/kmalloced/data page
+ * is a recipe for disaster
+ */
+#define INSNS_PER_PAGE  (PAGE_SIZE/(MAX_INSN_SIZE * sizeof(kprobe_opcode_t)))
+
+struct kprobe_insn_page {
+        struct hlist_node hlist;
+        kprobe_opcode_t *insns;         /* Page of instruction slots */
+        char slot_used[INSNS_PER_PAGE];
+        int nused;
+};
+
+static struct hlist_head kprobe_insn_pages;
+
+/**
+ * get_insn_slot() - Find a slot on an executable page for an instruction.
+ * We allocate an executable page if there's no room on existing ones.
+ */
+kprobe_opcode_t *get_insn_slot(void)
+{
+        struct kprobe_insn_page *kip;
+        struct hlist_node *pos;
+
+        hlist_for_each(pos, &kprobe_insn_pages) {
+                kip = hlist_entry(pos, struct kprobe_insn_page, hlist);
+                if (kip->nused < INSNS_PER_PAGE) {
+                        int i;
+                        for (i = 0; i < INSNS_PER_PAGE; i++) {
+                                if (!kip->slot_used[i]) {
+                                        kip->slot_used[i] = 1;
+                                        kip->nused++;
+                                        return kip->insns + (i * MAX_INSN_SIZE);
+                                }
+                        }
+                        /* Surprise!  No unused slots.  Fix kip->nused. */
+                        kip->nused = INSNS_PER_PAGE;
+                }
+        }
+
+        /* All out of space.  Need to allocate a new page. Use slot 0.*/
+        kip = kmalloc(sizeof(struct kprobe_insn_page), GFP_KERNEL);
+        if (!kip) {
+                return NULL;
+        }
+
+        /*
+         * Use module_alloc so this page is within +/- 2GB of where the
+         * kernel image and loaded module images reside. This is required
+         * so x86_64 can correctly handle the %rip-relative fixups.
+         */
+        kip->insns = module_alloc(PAGE_SIZE);
+        if (!kip->insns) {
+                kfree(kip);
+                return NULL;
+        }
+        INIT_HLIST_NODE(&kip->hlist);
+        hlist_add_head(&kip->hlist, &kprobe_insn_pages);
+        memset(kip->slot_used, 0, INSNS_PER_PAGE);
+        kip->slot_used[0] = 1;
+        kip->nused = 1;
+        return kip->insns;
+}
+
+void free_insn_slot(kprobe_opcode_t *slot)
+{
+        struct kprobe_insn_page *kip;
+        struct hlist_node *pos;
+
+        hlist_for_each(pos, &kprobe_insn_pages) {
+                kip = hlist_entry(pos, struct kprobe_insn_page, hlist);
+                if (kip->insns <= slot &&
+                    slot < kip->insns + (INSNS_PER_PAGE * MAX_INSN_SIZE)) {
+                        int i = (slot - kip->insns) / MAX_INSN_SIZE;
+                        kip->slot_used[i] = 0;
+                        kip->nused--;
+                        if (kip->nused == 0) {
+                                /*
+                                 * Page is no longer in use.  Free it unless
+                                 * it's the last one.  We keep the last one
+                                 * so as not to have to set it up again the
+                                 * next time somebody inserts a probe.
+                                 */
+                                hlist_del(&kip->hlist);
+                                if (hlist_empty(&kprobe_insn_pages)) {
+                                        INIT_HLIST_NODE(&kip->hlist);
+                                        hlist_add_head(&kip->hlist,
+                                                &kprobe_insn_pages);
+                                } else {
+                                        module_free(NULL, kip->insns);
+                                        kfree(kip);
+                                }
+                        }
+                        return;
+                }
+        }
+}
+
 /* Locks kprobe: irqs must be disabled */
 void lock_kprobes(void)
 {
@@ -78,22 +183,23 @@ struct kprobe *get_kprobe(void *addr)
  * Aggregate handlers for multiple kprobes support - these handlers
  * take care of invoking the individual kprobe handlers on p->list
  */
-int aggr_pre_handler(struct kprobe *p, struct pt_regs *regs)
+static int aggr_pre_handler(struct kprobe *p, struct pt_regs *regs)
 {
         struct kprobe *kp;
 
         list_for_each_entry(kp, &p->list, list) {
                 if (kp->pre_handler) {
                         curr_kprobe = kp;
-                        kp->pre_handler(kp, regs);
-                        curr_kprobe = NULL;
+                        if (kp->pre_handler(kp, regs))
+                                return 1;
                 }
+                curr_kprobe = NULL;
         }
         return 0;
 }
 
-void aggr_post_handler(struct kprobe *p, struct pt_regs *regs,
-                unsigned long flags)
+static void aggr_post_handler(struct kprobe *p, struct pt_regs *regs,
+                              unsigned long flags)
 {
         struct kprobe *kp;
 
@@ -107,7 +213,8 @@ void aggr_post_handler(struct kprobe *p, struct pt_regs *regs,
         return;
 }
 
-int aggr_fault_handler(struct kprobe *p, struct pt_regs *regs, int trapnr)
+static int aggr_fault_handler(struct kprobe *p, struct pt_regs *regs,
+                              int trapnr)
 {
         /*
          * if we faulted "during" the execution of a user specified
@@ -120,19 +227,159 @@ int aggr_fault_handler(struct kprobe *p, struct pt_regs *regs, int trapnr)
         return 0;
 }
 
+static int aggr_break_handler(struct kprobe *p, struct pt_regs *regs)
+{
+        struct kprobe *kp = curr_kprobe;
+        if (curr_kprobe && kp->break_handler) {
+                if (kp->break_handler(kp, regs)) {
+                        curr_kprobe = NULL;
+                        return 1;
+                }
+        }
+        curr_kprobe = NULL;
+        return 0;
+}
+
+struct kretprobe_instance *get_free_rp_inst(struct kretprobe *rp)
+{
+        struct hlist_node *node;
+        struct kretprobe_instance *ri;
+        hlist_for_each_entry(ri, node, &rp->free_instances, uflist)
+                return ri;
+        return NULL;
+}
+
+static struct kretprobe_instance *get_used_rp_inst(struct kretprobe *rp)
+{
+        struct hlist_node *node;
+        struct kretprobe_instance *ri;
+        hlist_for_each_entry(ri, node, &rp->used_instances, uflist)
+                return ri;
+        return NULL;
+}
+
+void add_rp_inst(struct kretprobe_instance *ri)
+{
+        /*
+         * Remove rp inst off the free list -
+         * Add it back when probed function returns
+         */
+        hlist_del(&ri->uflist);
+
+        /* Add rp inst onto table */
+        INIT_HLIST_NODE(&ri->hlist);
+        hlist_add_head(&ri->hlist,
+                        &kretprobe_inst_table[hash_ptr(ri->task, KPROBE_HASH_BITS)]);
+
+        /* Also add this rp inst to the used list. */
+        INIT_HLIST_NODE(&ri->uflist);
+        hlist_add_head(&ri->uflist, &ri->rp->used_instances);
+}
+
+void recycle_rp_inst(struct kretprobe_instance *ri)
+{
+        /* remove rp inst off the rprobe_inst_table */
+        hlist_del(&ri->hlist);
+        if (ri->rp) {
+                /* remove rp inst off the used list */
+                hlist_del(&ri->uflist);
+                /* put rp inst back onto the free list */
+                INIT_HLIST_NODE(&ri->uflist);
+                hlist_add_head(&ri->uflist, &ri->rp->free_instances);
+        } else
+                /* Unregistering */
+                kfree(ri);
+}
+
+struct hlist_head * kretprobe_inst_table_head(struct task_struct *tsk)
+{
+        return &kretprobe_inst_table[hash_ptr(tsk, KPROBE_HASH_BITS)];
+}
+
+/*
+ * This function is called from exit_thread or flush_thread when task tk's
+ * stack is being recycled so that we can recycle any function-return probe
+ * instances associated with this task. These left over instances represent
+ * probed functions that have been called but will never return.
+ */
+void kprobe_flush_task(struct task_struct *tk)
+{
+        struct kretprobe_instance *ri;
+        struct hlist_head *head;
+        struct hlist_node *node, *tmp;
+        unsigned long flags = 0;
+
+        spin_lock_irqsave(&kprobe_lock, flags);
+        head = kretprobe_inst_table_head(current);
+        hlist_for_each_entry_safe(ri, node, tmp, head, hlist) {
+                if (ri->task == tk)
+                        recycle_rp_inst(ri);
+        }
+        spin_unlock_irqrestore(&kprobe_lock, flags);
+}
+
+/*
+ * This kprobe pre_handler is registered with every kretprobe. When probe
+ * hits it will set up the return probe.
+ */
+static int pre_handler_kretprobe(struct kprobe *p, struct pt_regs *regs)
+{
+        struct kretprobe *rp = container_of(p, struct kretprobe, kp);
+
+        /*TODO: consider to only swap the RA after the last pre_handler fired */
+        arch_prepare_kretprobe(rp, regs);
+        return 0;
+}
+
+static inline void free_rp_inst(struct kretprobe *rp)
+{
+        struct kretprobe_instance *ri;
+        while ((ri = get_free_rp_inst(rp)) != NULL) {
+                hlist_del(&ri->uflist);
+                kfree(ri);
+        }
+}
+
+/*
+ * Keep all fields in the kprobe consistent
+ */
+static inline void copy_kprobe(struct kprobe *old_p, struct kprobe *p)
+{
+        memcpy(&p->opcode, &old_p->opcode, sizeof(kprobe_opcode_t));
+        memcpy(&p->ainsn, &old_p->ainsn, sizeof(struct arch_specific_insn));
+}
+
+/*
+* Add the new probe to old_p->list. Fail if this is the
+* second jprobe at the address - two jprobes can't coexist
+*/
+static int add_new_kprobe(struct kprobe *old_p, struct kprobe *p)
+{
+        struct kprobe *kp;
+
+        if (p->break_handler) {
+                list_for_each_entry(kp, &old_p->list, list) {
+                        if (kp->break_handler)
+                                return -EEXIST;
+                }
+                list_add_tail(&p->list, &old_p->list);
+        } else
+                list_add(&p->list, &old_p->list);
+        return 0;
+}
+
 /*
  * Fill in the required fields of the "manager kprobe". Replace the
  * earlier kprobe in the hlist with the manager kprobe
  */
 static inline void add_aggr_kprobe(struct kprobe *ap, struct kprobe *p)
 {
+        copy_kprobe(p, ap);
         ap->addr = p->addr;
-        ap->opcode = p->opcode;
-        memcpy(&ap->ainsn, &p->ainsn, sizeof(struct arch_specific_insn));
-
         ap->pre_handler = aggr_pre_handler;
         ap->post_handler = aggr_post_handler;
         ap->fault_handler = aggr_fault_handler;
+        ap->break_handler = aggr_break_handler;
 
         INIT_LIST_HEAD(&ap->list);
         list_add(&p->list, &ap->list);
@@ -153,16 +400,16 @@ static int register_aggr_kprobe(struct kprobe *old_p, struct kprobe *p)
         int ret = 0;
         struct kprobe *ap;
 
-        if (old_p->break_handler || p->break_handler) {
-                ret = -EEXIST;  /* kprobe and jprobe can't (yet) coexist */
-        } else if (old_p->pre_handler == aggr_pre_handler) {
-                list_add(&p->list, &old_p->list);
+        if (old_p->pre_handler == aggr_pre_handler) {
+                copy_kprobe(old_p, p);
+                ret = add_new_kprobe(old_p, p);
         } else {
                 ap = kcalloc(1, sizeof(struct kprobe), GFP_ATOMIC);
                 if (!ap)
                         return -ENOMEM;
                 add_aggr_kprobe(ap, old_p);
-                list_add(&p->list, &ap->list);
+                copy_kprobe(ap, p);
+                ret = add_new_kprobe(ap, p);
         }
         return ret;
 }
@@ -170,10 +417,8 @@ static int register_aggr_kprobe(struct kprobe *old_p, struct kprobe *p)
 /* kprobe removal house-keeping routines */
 static inline void cleanup_kprobe(struct kprobe *p, unsigned long flags)
 {
-        *p->addr = p->opcode;
+        arch_disarm_kprobe(p);
         hlist_del(&p->hlist);
-        flush_icache_range((unsigned long) p->addr,
-                   (unsigned long) p->addr + sizeof(kprobe_opcode_t));
         spin_unlock_irqrestore(&kprobe_lock, flags);
         arch_remove_kprobe(p);
 }
@@ -200,6 +445,7 @@ int register_kprobe(struct kprobe *p)
         }
         spin_lock_irqsave(&kprobe_lock, flags);
         old_p = get_kprobe(p->addr);
+        p->nmissed = 0;
         if (old_p) {
                 ret = register_aggr_kprobe(old_p, p);
                 goto out;
@@ -210,10 +456,8 @@ int register_kprobe(struct kprobe *p)
         hlist_add_head(&p->hlist,
                        &kprobe_table[hash_ptr(p->addr, KPROBE_HASH_BITS)]);
 
-        p->opcode = *p->addr;
-        *p->addr = BREAKPOINT_INSTRUCTION;
-        flush_icache_range((unsigned long) p->addr,
-                           (unsigned long) p->addr + sizeof(kprobe_opcode_t));
+        arch_arm_kprobe(p);
+
 out:
         spin_unlock_irqrestore(&kprobe_lock, flags);
 rm_kprobe:
@@ -257,16 +501,83 @@ void unregister_jprobe(struct jprobe *jp)
         unregister_kprobe(&jp->kp);
 }
 
+#ifdef ARCH_SUPPORTS_KRETPROBES
+
+int register_kretprobe(struct kretprobe *rp)
+{
+        int ret = 0;
+        struct kretprobe_instance *inst;
+        int i;
+
+        rp->kp.pre_handler = pre_handler_kretprobe;
+
+        /* Pre-allocate memory for max kretprobe instances */
+        if (rp->maxactive <= 0) {
+#ifdef CONFIG_PREEMPT
+                rp->maxactive = max(10, 2 * NR_CPUS);
+#else
+                rp->maxactive = NR_CPUS;
+#endif
+        }
+        INIT_HLIST_HEAD(&rp->used_instances);
+        INIT_HLIST_HEAD(&rp->free_instances);
+        for (i = 0; i < rp->maxactive; i++) {
+                inst = kmalloc(sizeof(struct kretprobe_instance), GFP_KERNEL);
+                if (inst == NULL) {
+                        free_rp_inst(rp);
+                        return -ENOMEM;
+                }
+                INIT_HLIST_NODE(&inst->uflist);
+                hlist_add_head(&inst->uflist, &rp->free_instances);
+        }
+
+        rp->nmissed = 0;
+        /* Establish function entry probe point */
+        if ((ret = register_kprobe(&rp->kp)) != 0)
+                free_rp_inst(rp);
+        return ret;
+}
+
+#else /* ARCH_SUPPORTS_KRETPROBES */
+
+int register_kretprobe(struct kretprobe *rp)
+{
+        return -ENOSYS;
+}
+
+#endif /* ARCH_SUPPORTS_KRETPROBES */
+
+void unregister_kretprobe(struct kretprobe *rp)
+{
+        unsigned long flags;
+        struct kretprobe_instance *ri;
+
+        unregister_kprobe(&rp->kp);
+        /* No race here */
+        spin_lock_irqsave(&kprobe_lock, flags);
+        free_rp_inst(rp);
+        while ((ri = get_used_rp_inst(rp)) != NULL) {
+                ri->rp = NULL;
+                hlist_del(&ri->uflist);
+        }
+        spin_unlock_irqrestore(&kprobe_lock, flags);
+}
+
 static int __init init_kprobes(void)
 {
         int i, err = 0;
 
         /* FIXME allocate the probe table, currently defined statically */
         /* initialize all list heads */
-        for (i = 0; i < KPROBE_TABLE_SIZE; i++)
+        for (i = 0; i < KPROBE_TABLE_SIZE; i++) {
                 INIT_HLIST_HEAD(&kprobe_table[i]);
+                INIT_HLIST_HEAD(&kretprobe_inst_table[i]);
+        }
+
+        err = arch_init();
+        if (!err)
+                err = register_die_notifier(&kprobe_exceptions_nb);
 
-        err = register_die_notifier(&kprobe_exceptions_nb);
         return err;
 }
 
@@ -277,3 +588,6 @@ EXPORT_SYMBOL_GPL(unregister_kprobe);
 EXPORT_SYMBOL_GPL(register_jprobe);
 EXPORT_SYMBOL_GPL(unregister_jprobe);
 EXPORT_SYMBOL_GPL(jprobe_return);
+EXPORT_SYMBOL_GPL(register_kretprobe);
+EXPORT_SYMBOL_GPL(unregister_kretprobe);
+