microblaze: Support ptrace syscall tracing.

Signed-off-by: Edgar E. Iglesias <edgar.iglesias@gmail.com>
Signed-off-by: Michal Simek <monstr@monstr.eu>

2 files changed, 122 insertions, 12 deletions

diff --git a/arch/microblaze/kernel/entry.S b/arch/microblaze/kernel/entry.S
index c7353e7..acc1f05 100644
--- a/arch/microblaze/kernel/entry.S
+++ b/arch/microblaze/kernel/entry.S
@@ -308,38 +308,69 @@ C_ENTRY(_user_exception):
 	swi	r12, r1, PTO+PT_R0;
 	tovirt(r1,r1)
 
-	la	r15, r0, ret_from_trap-8
 /* where the trap should return need -8 to adjust for rtsd r15, 8*/
 /* Jump to the appropriate function for the system call number in r12
  * (r12 is not preserved), or return an error if r12 is not valid. The LP
  * register should point to the location where
  * the called function should return.  [note that MAKE_SYS_CALL uses label 1] */
-	/* See if the system call number is valid.  */
+
+	# Step into virtual mode.
+	set_vms;
+	addik	r11, r0, 3f
+	rtid	r11, 0
+	nop
+3:
+	add	r11, r0, CURRENT_TASK	 /* Get current task ptr into r11 */
+	lwi	r11, r11, TS_THREAD_INFO /* get thread info */
+	lwi	r11, r11, TI_FLAGS	 /* get flags in thread info */
+	andi	r11, r11, _TIF_WORK_SYSCALL_MASK
+	beqi	r11, 4f
+
+	addik	r3, r0, -ENOSYS
+	swi	r3, r1, PTO + PT_R3
+	brlid	r15, do_syscall_trace_enter
+	addik	r5, r1, PTO + PT_R0
+
+	# do_syscall_trace_enter returns the new syscall nr.
+	addk	r12, r0, r3
+	lwi	r5, r1, PTO+PT_R5;
+	lwi	r6, r1, PTO+PT_R6;
+	lwi	r7, r1, PTO+PT_R7;
+	lwi	r8, r1, PTO+PT_R8;
+	lwi	r9, r1, PTO+PT_R9;
+	lwi	r10, r1, PTO+PT_R10;
+4:
+/* Jump to the appropriate function for the system call number in r12
+ * (r12 is not preserved), or return an error if r12 is not valid.
+ * The LP register should point to the location where the called function
+ * should return.  [note that MAKE_SYS_CALL uses label 1] */
+	/* See if the system call number is valid */
 	addi	r11, r12, -__NR_syscalls;
-	bgei	r11,1f;
+	bgei	r11,5f;
 	/* Figure out which function to use for this system call.  */
 	/* Note Microblaze barrel shift is optional, so don't rely on it */
 	add	r12, r12, r12;			/* convert num -> ptr */
 	add	r12, r12, r12;
 
 	/* Trac syscalls and stored them to r0_ram */
-	lwi	r3, r12, 0x400 + TOPHYS(r0_ram)
+	lwi	r3, r12, 0x400 + r0_ram
 	addi	r3, r3, 1
-	swi	r3, r12, 0x400 + TOPHYS(r0_ram)
+	swi	r3, r12, 0x400 + r0_ram
+
+	# Find and jump into the syscall handler.
+	lwi	r12, r12, sys_call_table
+	/* where the trap should return need -8 to adjust for rtsd r15, 8 */
+	la	r15, r0, ret_from_trap-8
+	bra	r12
 
-	lwi	r12, r12, TOPHYS(sys_call_table); /* Function ptr */
-	/* Make the system call.  to r12*/
-	set_vms;
-	rtid	r12, 0;
-	nop;
 	/* The syscall number is invalid, return an error.  */
-1:	VM_ON;	/* RETURN() expects virtual mode*/
+5:
 	addi	r3, r0, -ENOSYS;
 	rtsd	r15,8;		/* looks like a normal subroutine return */
 	or 	r0, r0, r0
 
 
-/* Entry point used to return from a syscall/trap.  */
+/* Entry point used to return from a syscall/trap */
 /* We re-enable BIP bit before state restore */
 C_ENTRY(ret_from_trap):
 	set_bip;			/*  Ints masked for state restore*/
@@ -349,6 +380,23 @@ C_ENTRY(ret_from_trap):
 
 	/* We're returning to user mode, so check for various conditions that
 	 * trigger rescheduling. */
+	# FIXME: Restructure all these flag checks.
+	add	r11, r0, CURRENT_TASK;	/* Get current task ptr into r11 */
+	lwi	r11, r11, TS_THREAD_INFO;	/* get thread info */
+	lwi	r11, r11, TI_FLAGS;		/* get flags in thread info */
+	andi	r11, r11, _TIF_WORK_SYSCALL_MASK
+	beqi	r11, 1f
+
+	swi	r3, r1, PTO + PT_R3
+	swi	r4, r1, PTO + PT_R4
+	brlid	r15, do_syscall_trace_leave
+	addik	r5, r1, PTO + PT_R0
+	lwi	r3, r1, PTO + PT_R3
+	lwi	r4, r1, PTO + PT_R4
+1:
+
+	/* We're returning to user mode, so check for various conditions that
+	 * trigger rescheduling. */
 	/* Get current task ptr into r11 */
 	add	r11, r0, CURRENT_TASK;	/* Get current task ptr into r11 */
 	lwi	r11, r11, TS_THREAD_INFO;	/* get thread info */
diff --git a/arch/microblaze/kernel/ptrace.c b/arch/microblaze/kernel/ptrace.c
index 53ff39a..4b3ac32 100644
--- a/arch/microblaze/kernel/ptrace.c
+++ b/arch/microblaze/kernel/ptrace.c
@@ -29,6 +29,10 @@
 #include <linux/sched.h>
 #include <linux/ptrace.h>
 #include <linux/signal.h>
+#include <linux/elf.h>
+#include <linux/audit.h>
+#include <linux/seccomp.h>
+#include <linux/tracehook.h>
 
 #include <linux/errno.h>
 #include <asm/processor.h>
@@ -174,6 +178,64 @@ long arch_ptrace(struct task_struct *child, long request, long addr, long data)
 	return rval;
 }
 
+asmlinkage long do_syscall_trace_enter(struct pt_regs *regs)
+{
+	long ret = 0;
+
+	secure_computing(regs->r12);
+
+	if (test_thread_flag(TIF_SYSCALL_TRACE) &&
+	    tracehook_report_syscall_entry(regs))
+		/*
+		 * Tracing decided this syscall should not happen.
+		 * We'll return a bogus call number to get an ENOSYS
+		 * error, but leave the original number in regs->regs[0].
+		 */
+		ret = -1L;
+
+	if (unlikely(current->audit_context))
+		audit_syscall_entry(EM_XILINX_MICROBLAZE, regs->r12,
+				    regs->r5, regs->r6,
+				    regs->r7, regs->r8);
+
+	return ret ?: regs->r12;
+}
+
+asmlinkage void do_syscall_trace_leave(struct pt_regs *regs)
+{
+	int step;
+
+	if (unlikely(current->audit_context))
+		audit_syscall_exit(AUDITSC_RESULT(regs->r3), regs->r3);
+
+	step = test_thread_flag(TIF_SINGLESTEP);
+	if (step || test_thread_flag(TIF_SYSCALL_TRACE))
+		tracehook_report_syscall_exit(regs, step);
+}
+
+#if 0
+static asmlinkage void syscall_trace(void)
+{
+	if (!test_thread_flag(TIF_SYSCALL_TRACE))
+		return;
+	if (!(current->ptrace & PT_PTRACED))
+		return;
+	/* The 0x80 provides a way for the tracing parent to distinguish
+	 between a syscall stop and SIGTRAP delivery */
+	ptrace_notify(SIGTRAP | ((current->ptrace & PT_TRACESYSGOOD)
+				? 0x80 : 0));
+	/*
+	 * this isn't the same as continuing with a signal, but it will do
+	 * for normal use. strace only continues with a signal if the
+	 * stopping signal is not SIGTRAP. -brl
+	 */
+	if (current->exit_code) {
+		send_sig(current->exit_code, current, 1);
+		current->exit_code = 0;
+	}
+}
+#endif
+
 void ptrace_disable(struct task_struct *child)
 {
 	/* nothing to do */