[PATCH] kmemdup: some users
[safe/jmp/linux-2.6] / arch / i386 / kernel / kprobes.c
index 137bf61..afe6505 100644 (file)
  *             <prasanna@in.ibm.com> added function-return probes.
  */
 
-#include <linux/config.h>
 #include <linux/kprobes.h>
 #include <linux/ptrace.h>
 #include <linux/preempt.h>
 #include <asm/cacheflush.h>
 #include <asm/kdebug.h>
 #include <asm/desc.h>
+#include <asm/uaccess.h>
 
 void jprobe_return_end(void);
 
@@ -42,7 +42,7 @@ DEFINE_PER_CPU(struct kprobe *, current_kprobe) = NULL;
 DEFINE_PER_CPU(struct kprobe_ctlblk, kprobe_ctlblk);
 
 /* insert a jmp code */
-static inline void set_jmp_op(void *from, void *to)
+static __always_inline void set_jmp_op(void *from, void *to)
 {
        struct __arch_jmp_op {
                char op;
@@ -56,38 +56,89 @@ static inline void set_jmp_op(void *from, void *to)
 /*
  * returns non-zero if opcodes can be boosted.
  */
-static inline int can_boost(kprobe_opcode_t opcode)
+static __always_inline int can_boost(kprobe_opcode_t *opcodes)
 {
-       switch (opcode & 0xf0 ) {
+#define W(row,b0,b1,b2,b3,b4,b5,b6,b7,b8,b9,ba,bb,bc,bd,be,bf)               \
+       (((b0##UL << 0x0)|(b1##UL << 0x1)|(b2##UL << 0x2)|(b3##UL << 0x3) |   \
+         (b4##UL << 0x4)|(b5##UL << 0x5)|(b6##UL << 0x6)|(b7##UL << 0x7) |   \
+         (b8##UL << 0x8)|(b9##UL << 0x9)|(ba##UL << 0xa)|(bb##UL << 0xb) |   \
+         (bc##UL << 0xc)|(bd##UL << 0xd)|(be##UL << 0xe)|(bf##UL << 0xf))    \
+        << (row % 32))
+       /*
+        * Undefined/reserved opcodes, conditional jump, Opcode Extension
+        * Groups, and some special opcodes can not be boost.
+        */
+       static const unsigned long twobyte_is_boostable[256 / 32] = {
+               /*      0 1 2 3 4 5 6 7 8 9 a b c d e f         */
+               /*      -------------------------------         */
+               W(0x00, 0,0,1,1,0,0,1,0,1,1,0,0,0,0,0,0)| /* 00 */
+               W(0x10, 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0), /* 10 */
+               W(0x20, 1,1,1,1,0,0,0,0,0,0,0,0,0,0,0,0)| /* 20 */
+               W(0x30, 0,1,0,0,0,0,0,0,0,0,0,0,0,0,0,0), /* 30 */
+               W(0x40, 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1)| /* 40 */
+               W(0x50, 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0), /* 50 */
+               W(0x60, 1,1,1,1,1,1,1,1,1,1,1,1,0,0,1,1)| /* 60 */
+               W(0x70, 0,0,0,0,1,1,1,1,0,0,0,0,0,0,1,1), /* 70 */
+               W(0x80, 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0)| /* 80 */
+               W(0x90, 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1), /* 90 */
+               W(0xa0, 1,1,0,1,1,1,0,0,1,1,0,1,1,1,0,1)| /* a0 */
+               W(0xb0, 1,1,1,1,1,1,1,1,0,0,0,1,1,1,1,1), /* b0 */
+               W(0xc0, 1,1,0,0,0,0,0,0,1,1,1,1,1,1,1,1)| /* c0 */
+               W(0xd0, 0,1,1,1,0,1,0,0,1,1,0,1,1,1,0,1), /* d0 */
+               W(0xe0, 0,1,1,0,0,1,0,0,1,1,0,1,1,1,0,1)| /* e0 */
+               W(0xf0, 0,1,1,1,0,1,0,0,1,1,1,0,1,1,1,0)  /* f0 */
+               /*      -------------------------------         */
+               /*      0 1 2 3 4 5 6 7 8 9 a b c d e f         */
+       };
+#undef W
+       kprobe_opcode_t opcode;
+       kprobe_opcode_t *orig_opcodes = opcodes;
+retry:
+       if (opcodes - orig_opcodes > MAX_INSN_SIZE - 1)
+               return 0;
+       opcode = *(opcodes++);
+
+       /* 2nd-byte opcode */
+       if (opcode == 0x0f) {
+               if (opcodes - orig_opcodes > MAX_INSN_SIZE - 1)
+                       return 0;
+               return test_bit(*opcodes, twobyte_is_boostable);
+       }
+
+       switch (opcode & 0xf0) {
+       case 0x60:
+               if (0x63 < opcode && opcode < 0x67)
+                       goto retry; /* prefixes */
+               /* can't boost Address-size override and bound */
+               return (opcode != 0x62 && opcode != 0x67);
        case 0x70:
                return 0; /* can't boost conditional jump */
-       case 0x90:
-               /* can't boost call and pushf */
-               return opcode != 0x9a && opcode != 0x9c;
        case 0xc0:
-               /* can't boost undefined opcodes and soft-interruptions */
-               return (0xc1 < opcode && opcode < 0xc6) ||
-                       (0xc7 < opcode && opcode < 0xcc) || opcode == 0xcf;
+               /* can't boost software-interruptions */
+               return (0xc1 < opcode && opcode < 0xcc) || opcode == 0xcf;
        case 0xd0:
                /* can boost AA* and XLAT */
                return (opcode == 0xd4 || opcode == 0xd5 || opcode == 0xd7);
        case 0xe0:
-               /* can boost in/out and (may be) jmps */
-               return (0xe3 < opcode && opcode != 0xe8);
+               /* can boost in/out and absolute jmps */
+               return ((opcode & 0x04) || opcode == 0xea);
        case 0xf0:
+               if ((opcode & 0x0c) == 0 && opcode != 0xf1)
+                       goto retry; /* lock/rep(ne) prefix */
                /* clear and set flags can be boost */
                return (opcode == 0xf5 || (0xf7 < opcode && opcode < 0xfe));
        default:
-               /* currently, can't boost 2 bytes opcodes */
-               return opcode != 0x0f;
+               if (opcode == 0x26 || opcode == 0x36 || opcode == 0x3e)
+                       goto retry; /* prefixes */
+               /* can't boost CS override and call */
+               return (opcode != 0x2e && opcode != 0x9a);
        }
 }
 
-
 /*
  * returns non-zero if opcode modifies the interrupt flag.
  */
-static inline int is_IF_modifier(kprobe_opcode_t opcode)
+static int __kprobes is_IF_modifier(kprobe_opcode_t opcode)
 {
        switch (opcode) {
        case 0xfa:              /* cli */
@@ -108,7 +159,7 @@ int __kprobes arch_prepare_kprobe(struct kprobe *p)
 
        memcpy(p->ainsn.insn, p->addr, MAX_INSN_SIZE * sizeof(kprobe_opcode_t));
        p->opcode = *p->addr;
-       if (can_boost(p->opcode)) {
+       if (can_boost(p->addr)) {
                p->ainsn.boostable = 0;
        } else {
                p->ainsn.boostable = -1;
@@ -137,7 +188,7 @@ void __kprobes arch_remove_kprobe(struct kprobe *p)
        mutex_unlock(&kprobe_mutex);
 }
 
-static inline void save_previous_kprobe(struct kprobe_ctlblk *kcb)
+static void __kprobes save_previous_kprobe(struct kprobe_ctlblk *kcb)
 {
        kcb->prev_kprobe.kp = kprobe_running();
        kcb->prev_kprobe.status = kcb->kprobe_status;
@@ -145,7 +196,7 @@ static inline void save_previous_kprobe(struct kprobe_ctlblk *kcb)
        kcb->prev_kprobe.saved_eflags = kcb->kprobe_saved_eflags;
 }
 
-static inline void restore_previous_kprobe(struct kprobe_ctlblk *kcb)
+static void __kprobes restore_previous_kprobe(struct kprobe_ctlblk *kcb)
 {
        __get_cpu_var(current_kprobe) = kcb->prev_kprobe.kp;
        kcb->kprobe_status = kcb->prev_kprobe.status;
@@ -153,7 +204,7 @@ static inline void restore_previous_kprobe(struct kprobe_ctlblk *kcb)
        kcb->kprobe_saved_eflags = kcb->prev_kprobe.saved_eflags;
 }
 
-static inline void set_current_kprobe(struct kprobe *p, struct pt_regs *regs,
+static void __kprobes set_current_kprobe(struct kprobe *p, struct pt_regs *regs,
                                struct kprobe_ctlblk *kcb)
 {
        __get_cpu_var(current_kprobe) = p;
@@ -163,7 +214,7 @@ static inline void set_current_kprobe(struct kprobe *p, struct pt_regs *regs,
                kcb->kprobe_saved_eflags &= ~IF_MASK;
 }
 
-static inline void prepare_singlestep(struct kprobe *p, struct pt_regs *regs)
+static void __kprobes prepare_singlestep(struct kprobe *p, struct pt_regs *regs)
 {
        regs->eflags |= TF_MASK;
        regs->eflags &= ~IF_MASK;
@@ -203,12 +254,10 @@ static int __kprobes kprobe_handler(struct pt_regs *regs)
 {
        struct kprobe *p;
        int ret = 0;
-       kprobe_opcode_t *addr = NULL;
-       unsigned long *lp;
+       kprobe_opcode_t *addr;
        struct kprobe_ctlblk *kcb;
-#ifdef CONFIG_PREEMPT
-       unsigned pre_preempt_count = preempt_count();
-#endif /* CONFIG_PREEMPT */
+
+       addr = (kprobe_opcode_t *)(regs->eip - sizeof(kprobe_opcode_t));
 
        /*
         * We don't want to be preempted for the entire
@@ -217,17 +266,6 @@ static int __kprobes kprobe_handler(struct pt_regs *regs)
        preempt_disable();
        kcb = get_kprobe_ctlblk();
 
-       /* Check if the application is using LDT entry for its code segment and
-        * calculate the address by reading the base address from the LDT entry.
-        */
-       if ((regs->xcs & 4) && (current->mm)) {
-               lp = (unsigned long *) ((unsigned long)((regs->xcs >> 3) * 8)
-                                       + (char *) current->mm->context.ldt);
-               addr = (kprobe_opcode_t *) (get_desc_base(lp) + regs->eip -
-                                               sizeof(kprobe_opcode_t));
-       } else {
-               addr = (kprobe_opcode_t *)(regs->eip - sizeof(kprobe_opcode_t));
-       }
        /* Check we're not actually recursing */
        if (kprobe_running()) {
                p = get_kprobe(addr);
@@ -251,10 +289,6 @@ static int __kprobes kprobe_handler(struct pt_regs *regs)
                        kcb->kprobe_status = KPROBE_REENTER;
                        return 1;
                } else {
-                       if (regs->eflags & VM_MASK) {
-                       /* We are in virtual-8086 mode. Return 0 */
-                               goto no_kprobe;
-                       }
                        if (*addr != BREAKPOINT_INSTRUCTION) {
                        /* The breakpoint instruction was removed by
                         * another cpu right after we hit, no further
@@ -274,11 +308,6 @@ static int __kprobes kprobe_handler(struct pt_regs *regs)
 
        p = get_kprobe(addr);
        if (!p) {
-               if (regs->eflags & VM_MASK) {
-                       /* We are in virtual-8086 mode. Return 0 */
-                       goto no_kprobe;
-               }
-
                if (*addr != BREAKPOINT_INSTRUCTION) {
                        /*
                         * The breakpoint instruction was removed right
@@ -303,22 +332,16 @@ static int __kprobes kprobe_handler(struct pt_regs *regs)
                /* handler has already set things up, so skip ss setup */
                return 1;
 
-       if (p->ainsn.boostable == 1 &&
-#ifdef CONFIG_PREEMPT
-           !(pre_preempt_count) && /*
-                                      * This enables booster when the direct
-                                      * execution path aren't preempted.
-                                      */
-#endif /* CONFIG_PREEMPT */
-           !p->post_handler && !p->break_handler ) {
+ss_probe:
+#ifndef CONFIG_PREEMPT
+       if (p->ainsn.boostable == 1 && !p->post_handler){
                /* Boost up -- we can execute copied instructions directly */
                reset_current_kprobe();
                regs->eip = (unsigned long)p->ainsn.insn;
                preempt_enable_no_resched();
                return 1;
        }
-
-ss_probe:
+#endif
        prepare_singlestep(p, regs);
        kcb->kprobe_status = KPROBE_HIT_SS;
        return 1;
@@ -333,17 +356,44 @@ no_kprobe:
  * here. When a retprobed function returns, this probe is hit and
  * trampoline_probe_handler() runs, calling the kretprobe's handler.
  */
- void kretprobe_trampoline_holder(void)
+ void __kprobes kretprobe_trampoline_holder(void)
  {
-       asm volatile (  ".global kretprobe_trampoline\n"
+       asm volatile ( ".global kretprobe_trampoline\n"
                        "kretprobe_trampoline: \n"
-                       "nop\n");
- }
+                       "       pushf\n"
+                       /* skip cs, eip, orig_eax, es, ds */
+                       "       subl $20, %esp\n"
+                       "       pushl %eax\n"
+                       "       pushl %ebp\n"
+                       "       pushl %edi\n"
+                       "       pushl %esi\n"
+                       "       pushl %edx\n"
+                       "       pushl %ecx\n"
+                       "       pushl %ebx\n"
+                       "       movl %esp, %eax\n"
+                       "       call trampoline_handler\n"
+                       /* move eflags to cs */
+                       "       movl 48(%esp), %edx\n"
+                       "       movl %edx, 44(%esp)\n"
+                       /* save true return address on eflags */
+                       "       movl %eax, 48(%esp)\n"
+                       "       popl %ebx\n"
+                       "       popl %ecx\n"
+                       "       popl %edx\n"
+                       "       popl %esi\n"
+                       "       popl %edi\n"
+                       "       popl %ebp\n"
+                       "       popl %eax\n"
+                       /* skip eip, orig_eax, es, ds */
+                       "       addl $16, %esp\n"
+                       "       popf\n"
+                       "       ret\n");
+}
 
 /*
- * Called when we hit the probe point at kretprobe_trampoline
+ * Called from kretprobe_trampoline
  */
-int __kprobes trampoline_probe_handler(struct kprobe *p, struct pt_regs *regs)
+fastcall void *__kprobes trampoline_handler(struct pt_regs *regs)
 {
         struct kretprobe_instance *ri = NULL;
         struct hlist_head *head;
@@ -372,8 +422,11 @@ int __kprobes trampoline_probe_handler(struct kprobe *p, struct pt_regs *regs)
                        /* another task is sharing our hash bucket */
                         continue;
 
-               if (ri->rp && ri->rp->handler)
+               if (ri->rp && ri->rp->handler){
+                       __get_cpu_var(current_kprobe) = &ri->rp->kp;
                        ri->rp->handler(ri, regs);
+                       __get_cpu_var(current_kprobe) = NULL;
+               }
 
                orig_ret_address = (unsigned long)ri->ret_addr;
                recycle_rp_inst(ri);
@@ -388,18 +441,10 @@ int __kprobes trampoline_probe_handler(struct kprobe *p, struct pt_regs *regs)
        }
 
        BUG_ON(!orig_ret_address || (orig_ret_address == trampoline_address));
-       regs->eip = orig_ret_address;
 
-       reset_current_kprobe();
        spin_unlock_irqrestore(&kretprobe_lock, flags);
-       preempt_enable_no_resched();
 
-       /*
-        * 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;
+       return (void*)orig_ret_address;
 }
 
 /*
@@ -439,10 +484,11 @@ static void __kprobes resume_execution(struct kprobe *p,
                *tos &= ~(TF_MASK | IF_MASK);
                *tos |= kcb->kprobe_old_eflags;
                break;
-       case 0xc3:              /* ret/lret */
-       case 0xcb:
-       case 0xc2:
+       case 0xc2:              /* iret/ret/lret */
+       case 0xc3:
        case 0xca:
+       case 0xcb:
+       case 0xcf:
        case 0xea:              /* jmp absolute -- eip is correct */
                /* eip is already adjusted, no more changes required */
                p->ainsn.boostable = 1;
@@ -450,10 +496,13 @@ static void __kprobes resume_execution(struct kprobe *p,
        case 0xe8:              /* call relative - Fix return addr */
                *tos = orig_eip + (*tos - copy_eip);
                break;
+       case 0x9a:              /* call absolute -- same as call absolute, indirect */
+               *tos = orig_eip + (*tos - copy_eip);
+               goto no_change;
        case 0xff:
                if ((p->ainsn.insn[1] & 0x30) == 0x10) {
-                       /* call absolute, indirect */
                        /*
+                        * call absolute, indirect
                         * Fix return addr; eip is correct.
                         * But this is not boostable
                         */
@@ -494,7 +543,7 @@ no_change:
  * Interrupts are disabled on entry as trap1 is an interrupt gate and they
  * remain disabled thoroughout this function.
  */
-static inline int post_kprobe_handler(struct pt_regs *regs)
+static int __kprobes post_kprobe_handler(struct pt_regs *regs)
 {
        struct kprobe *cur = kprobe_running();
        struct kprobe_ctlblk *kcb = get_kprobe_ctlblk();
@@ -530,20 +579,62 @@ out:
        return 1;
 }
 
-static inline int kprobe_fault_handler(struct pt_regs *regs, int trapnr)
+static int __kprobes kprobe_fault_handler(struct pt_regs *regs, int trapnr)
 {
        struct kprobe *cur = kprobe_running();
        struct kprobe_ctlblk *kcb = get_kprobe_ctlblk();
 
-       if (cur->fault_handler && cur->fault_handler(cur, regs, trapnr))
-               return 1;
-
-       if (kcb->kprobe_status & KPROBE_HIT_SS) {
-               resume_execution(cur, regs, kcb);
+       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 eip points back to the probe address
+                * and allow the page fault handler to continue as a
+                * normal page fault.
+                */
+               regs->eip = (unsigned long)cur->addr;
                regs->eflags |= kcb->kprobe_old_eflags;
-
-               reset_current_kprobe();
+               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 accouting
+                * 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 (fixup_exception(regs))
+                       return 1;
+
+               /*
+                * fixup_exception() could not handle it,
+                * Let do_page_fault() fix it.
+                */
+               break;
+       default:
+               break;
        }
        return 0;
 }
@@ -557,6 +648,9 @@ int __kprobes kprobe_exceptions_notify(struct notifier_block *self,
        struct die_args *args = (struct die_args *)data;
        int ret = NOTIFY_DONE;
 
+       if (args->regs && user_mode_vm(args->regs))
+               return ret;
+
        switch (val) {
        case DIE_INT3:
                if (kprobe_handler(args->regs))
@@ -646,12 +740,7 @@ int __kprobes longjmp_break_handler(struct kprobe *p, struct pt_regs *regs)
        return 0;
 }
 
-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);
+       return 0;
 }