[PATCH 11/16] LTTng : Linux Trace Toolkit Next Generation 0.5.95, kernel 2.6.17
From: Mathieu Desnoyers
Date: Wed Aug 30 2006 - 19:20:45 EST
11- LTTng architecture dependant instrumentation : MIPS
patch-2.6.17-lttng-0.5.95-instrumentation-mips.diff
OpenPGP public key: http://krystal.dyndns.org:8080/key/compudj.gpg
Key fingerprint: 8CD5 52C3 8E3C 4140 715F BA06 3F25 A8FE 3BAE 9A68
--- a/arch/mips/kernel/irq.c
+++ b/arch/mips/kernel/irq.c
@@ -22,6 +22,7 @@ #include <linux/random.h>
#include <linux/sched.h>
#include <linux/seq_file.h>
#include <linux/kallsyms.h>
+#include <linux/ltt/ltt-facility-kernel.h>
#include <asm/atomic.h>
#include <asm/system.h>
@@ -59,8 +60,12 @@ asmlinkage unsigned int do_IRQ(unsigned
irq_enter();
__DO_IRQ_SMTC_HOOK();
+ trace_kernel_irq_entry(irq, !(user_mode(regs)));
+
__do_IRQ(irq, regs);
+ trace_kernel_irq_exit();
+
irq_exit();
return 1;
diff --git a/arch/mips/kernel/process.c b/arch/mips/kernel/process.c
index 199a06e..58bddec 100644
--- a/arch/mips/kernel/process.c
+++ b/arch/mips/kernel/process.c
@@ -26,6 +26,7 @@ #include <linux/a.out.h>
#include <linux/init.h>
#include <linux/completion.h>
#include <linux/kallsyms.h>
+#include <linux/ltt/ltt-facility-process.h>
#include <asm/abi.h>
#include <asm/bootinfo.h>
@@ -256,6 +257,7 @@ ATTRIB_NORET void kernel_thread_helper(v
long kernel_thread(int (*fn)(void *), void *arg, unsigned long flags)
{
struct pt_regs regs;
+ long pid;
memset(®s, 0, sizeof(regs));
@@ -271,7 +273,13 @@ #else
#endif
/* Ok, create the new process.. */
- return do_fork(flags | CLONE_VM | CLONE_UNTRACED, 0, ®s, 0, NULL, NULL);
+ pid = do_fork(flags | CLONE_VM | CLONE_UNTRACED,
+ 0, ®s, 0, NULL, NULL);
+#ifdef CONFIG_LTT
+ if(pid >= 0)
+ trace_process_kernel_thread(pid, fn);
+#endif //CONFIG_LTT
+ return pid;
}
static struct mips_frame_info {
diff --git a/arch/mips/kernel/scall32-o32.S b/arch/mips/kernel/scall32-o32.S
index a0ac0e5..e32d298 100644
--- a/arch/mips/kernel/traps.c
+++ b/arch/mips/kernel/traps.c
@@ -11,6 +11,8 @@
* Copyright (C) 2000, 01 MIPS Technologies, Inc.
* Copyright (C) 2002, 2003, 2004, 2005 Maciej W. Rozycki
*/
+#include <linux/ltt/ltt-facility-kernel.h>
+#include <asm/ltt/ltt-facility-kernel_arch_mips.h>
#include <linux/config.h>
#include <linux/init.h>
#include <linux/mm.h>
@@ -19,6 +21,7 @@ #include <linux/sched.h>
#include <linux/smp.h>
#include <linux/smp_lock.h>
#include <linux/spinlock.h>
+#include <linux/unistd.h>
#include <linux/kallsyms.h>
#include <linux/bootmem.h>
@@ -256,7 +259,7 @@ void show_regs(struct pt_regs *regs)
printk("Cause : %08x\n", cause);
- cause = (cause & CAUSEF_EXCCODE) >> CAUSEB_EXCCODE;
+ cause = CAUSE_EXCCODE(cause);
if (1 <= cause && cause <= 5)
printk("BadVA : %0*lx\n", field, regs->cp0_badvaddr);
@@ -570,6 +573,7 @@ asmlinkage void do_ov(struct pt_regs *re
*/
asmlinkage void do_fpe(struct pt_regs *regs, unsigned long fcr31)
{
+ trace_kernel_trap_entry(CAUSE_EXCCODE(regs->cp0_cause), (void*)regs->cp0_epc);
if (fcr31 & FPU_CSR_UNI_X) {
int sig;
@@ -727,16 +731,22 @@ asmlinkage void do_cpu(struct pt_regs *r
die_if_kernel("do_cpu invoked from kernel context!", regs);
+ trace_kernel_trap_entry(CAUSE_EXCCODE(regs->cp0_cause), (void*)regs->cp0_epc);
+
cpid = (regs->cp0_cause >> CAUSEB_CE) & 3;
switch (cpid) {
case 0:
if (!cpu_has_llsc)
- if (!simulate_llsc(regs))
+ if (!simulate_llsc(regs)) {
+ trace_kernel_trap_exit();
return;
+ }
- if (!simulate_rdhwr(regs))
+ if (!simulate_rdhwr(regs)) {
+ trace_kernel_trap_exit();
return;
+ }
break;
@@ -789,7 +799,7 @@ #ifdef CONFIG_MIPS_MT_FPAFF
}
#endif /* CONFIG_MIPS_MT_FPAFF */
}
-
+ trace_kernel_trap_exit();
return;
case 2:
@@ -799,6 +809,7 @@ #endif /* CONFIG_MIPS_MT_FPAFF */
}
force_sig(SIGILL, current);
+ trace_kernel_trap_exit();
}
asmlinkage void do_mdmx(struct pt_regs *regs)
@@ -1512,3 +1523,90 @@ #endif
flush_icache_range(ebase, ebase + 0x400);
flush_tlb_handlers();
}
+
+
+
+#ifdef CONFIG_LTT
+asmlinkage void trace_real_syscall_entry(struct pt_regs *regs)
+{
+ trace_kernel_arch_syscall_entry(
+ (enum lttng_syscall_name)regs->regs[2],
+ (void*)regs->cp0_epc);
+}
+
+asmlinkage void trace_real_syscall_exit(void)
+{
+ trace_kernel_arch_syscall_exit();
+}
+
+#endif //CONFIG_LTT
+
+
+#if 0
+asmlinkage void trace_real_syscall_entry(struct pt_regs * regs)
+{
+ unsigned long addr;
+ int depth = 0;
+ unsigned long end_code;
+ unsigned long lower_bound;
+ int seek_depth;
+ unsigned long *stack;
+ unsigned long start_code;
+ unsigned long *start_stack;
+ ltt_syscall_entry trace_syscall_event;
+ unsigned long upper_bound;
+ int use_bounds;
+ int use_depth;
+
+ /* syscall_id will be negative for SVR4, IRIX5, BSD43, and POSIX
+ * syscalls -- these are not supported at this point by LTT
+ */
+ trace_syscall_event.syscall_id = (uint8_t) (regs->regs[2] - __NR_Linux);
+
+ trace_syscall_event.address = regs->cp0_epc;
+
+ if (!user_mode(regs))
+ goto trace_syscall_end;
+
+ if (ltt_get_trace_config(&use_depth, &use_bounds, &seek_depth,
+ (void*)&lower_bound, (void*)&upper_bound) < 0)
+ goto trace_syscall_end;
+
+ /* Heuristic that might work:
+ * (BUT DOESN'T WORK for any of the cases I tested...) zzz
+ * Search through stack until a value is found that is within the
+ * range start_code .. end_code. (This is looking for a return
+ * pointer to where a shared library was called from.) If a stack
+ * variable contains a valid code address then an incorrect
+ * result will be generated.
+ */
+ if ((use_depth == 1) || (use_bounds == 1)) {
+ stack = (unsigned long*) regs->regs[29];
+ end_code = current->mm->end_code;
+ start_code = current->mm->start_code;
+ start_stack = (unsigned long *)current->mm->start_stack;
+
+ while ((stack <= start_stack) && (!__get_user(addr, stack))) {
+ if ((addr > start_code) && (addr < end_code)) {
+ if (((use_depth == 1) && (depth == seek_depth)) ||
+ ((use_bounds == 1) && (addr > lower_bound) && (addr < upper_bound))) {
+ trace_syscall_event.address = addr;
+ goto trace_syscall_end;
+ } else {
+ depth++;
+ }
+ }
+ stack++;
+ }
+ }
+
+trace_syscall_end:
+ ltt_log_event(LTT_EV_SYSCALL_ENTRY, &trace_syscall_event);
+}
+
+asmlinkage void trace_real_syscall_exit(void)
+{
+ ltt_log_event(LTT_EV_SYSCALL_EXIT, NULL);
+}
+
+#endif /* 0 */
diff --git a/arch/mips/kernel/unaligned.c b/arch/mips/kernel/unaligned.c
index 5b5a373..857d793 100644
--- a/arch/mips/kernel/unaligned.c
+++ b/arch/mips/kernel/unaligned.c
@@ -72,6 +72,7 @@
* A store crossing a page boundary might be executed only partially.
* Undo the partial store in this case.
*/
+#include <linux/ltt/ltt-facility-kernel.h>
#include <linux/config.h>
#include <linux/mm.h>
#include <linux/module.h>
@@ -497,14 +498,18 @@ asmlinkage void do_ade(struct pt_regs *r
unsigned int __user *pc;
mm_segment_t seg;
+ trace_kernel_trap_entry(CAUSE_EXCCODE(regs->cp0_cause), (void*)regs->cp0_epc);
+
/*
* Address errors may be deliberately induced by the FPU emulator to
* retake control of the CPU after executing the instruction in the
* delay slot of an emulated branch.
*/
/* Terminate if exception was recognized as a delay slot return */
- if (do_dsemulret(regs))
+ if (do_dsemulret(regs)) {
+ trace_kernel_trap_exit();
return;
+ }
/* Otherwise handle as normal */
@@ -538,6 +543,8 @@ asmlinkage void do_ade(struct pt_regs *r
}
set_fs(seg);
+ trace_kernel_trap_exit();
+
return;
sigbus:
@@ -547,4 +554,6 @@ sigbus:
/*
* XXX On return from the signal handler we should advance the epc
*/
+
+ trace_kernel_trap_exit();
}
diff --git a/arch/mips/mm/fault.c b/arch/mips/mm/fault.c
index e3a6172..24f4333 100644
--- a/arch/mips/kernel/syscall.c
+++ b/arch/mips/kernel/syscall.c
@@ -7,6 +7,7 @@
* Copyright (C) 1999, 2000 Silicon Graphics, Inc.
* Copyright (C) 2001 MIPS Technologies, Inc.
*/
+#include <linux/ltt/ltt-facility-ipc.h>
#include <linux/config.h>
#include <linux/a.out.h>
#include <linux/capability.h>
@@ -308,6 +309,8 @@ asmlinkage int sys_ipc (uint call, int f
version = call >> 16; /* hack for backward compatibility */
call &= 0xffff;
+
+ trace_ipc_call(call, first);
switch (call) {
case SEMOP:
diff --git a/arch/mips/kernel/time.c b/arch/mips/kernel/time.c
index 13ff4da..f3f20bb 100644
--- a/arch/mips/kernel/scall32-o32.S
+++ b/arch/mips/kernel/scall32-o32.S
@@ -54,6 +54,17 @@ #endif
bgez t3, stackargs
stack_done:
+#ifdef CONFIG_LTT
+ sw t2, PT_R1(sp)
+ move a0, sp
+ jal trace_real_syscall_entry
+ lw t2, PT_R1(sp)
+
+ lw a0, PT_R4(sp) # Restore argument registers
+ lw a1, PT_R5(sp)
+ lw a2, PT_R6(sp)
+ lw a3, PT_R7(sp)
+#endif // CONFIG_LTT
lw t0, TI_FLAGS($28) # syscall tracing enabled?
li t1, _TIF_SYSCALL_TRACE | _TIF_SYSCALL_AUDIT
and t0, t1
@@ -71,6 +82,10 @@ stack_done:
# restarting
1: sw v0, PT_R2(sp) # result
+#ifdef CONFIG_LTT
+ jal trace_real_syscall_exit
+#endif // CONFIG_LTT
+
o32_syscall_exit:
local_irq_disable # make sure need_resched and
# signals dont change between
@@ -112,6 +127,10 @@ syscall_trace_entry:
# restarting
1: sw v0, PT_R2(sp) # result
+#ifdef CONFIG_LTT
+ jal trace_real_syscall_exit
+#endif //CONFIG_LTT)
+
j syscall_exit
/* ------------------------------------------------------------------------ */
diff --git a/arch/mips/kernel/scall64-64.S b/arch/mips/kernel/scall64-64.S
index 9ba7508..10c2a7e 100644
--- a/arch/mips/kernel/scall64-64.S
+++ b/arch/mips/kernel/scall64-64.S
@@ -52,6 +52,18 @@ #endif
# syscall routine
sd a3, PT_R26(sp) # save a3 for syscall restarting
+#ifdef CONFIG_LTT
+ sd t2, PT_R1(sp)
+ move a0, sp
+ jal trace_real_syscall_entry
+ ld t2, PT_R1(sp)
+ ld a0, PT_R4(sp) # Restore argument registers
+ ld a1, PT_R5(sp)
+ ld a2, PT_R6(sp)
+ ld a3, PT_R7(sp)
+ ld a4, PT_R8(sp)
+ ld a5, PT_R9(sp)
+#endif
li t1, _TIF_SYSCALL_TRACE | _TIF_SYSCALL_AUDIT
LONG_L t0, TI_FLAGS($28) # syscall tracing enabled?
@@ -69,6 +81,9 @@ #endif
sd v0, PT_R0(sp) # set flag for syscall
# restarting
1: sd v0, PT_R2(sp) # result
+#ifdef CONFIG_LTT
+ jal trace_real_syscall_exit
+#endif
n64_syscall_exit:
local_irq_disable # make sure need_resched and
@@ -111,6 +126,9 @@ syscall_trace_entry:
dnegu v0 # error
sd v0, PT_R0(sp) # set flag for syscall restarting
1: sd v0, PT_R2(sp) # result
+#ifdef CONFIG_LTT
+ jal trace_real_syscall_exit
+#endif
j syscall_exit
diff --git a/arch/mips/kernel/scall64-n32.S b/arch/mips/kernel/scall64-n32.S
index 942aca2..41a82ab 100644
--- a/arch/mips/kernel/scall64-n32.S
+++ b/arch/mips/kernel/scall64-n32.S
@@ -52,6 +52,18 @@ #endif
sd a3, PT_R26(sp) # save a3 for syscall restarting
+#ifdef CONFIG_LTT
+ sd t2, PT_R1(sp)
+ move a0, sp
+ jal trace_real_syscall_entry
+ ld t2, PT_R1(sp)
+ ld a0, PT_R4(sp) # Restore argument registers
+ ld a1, PT_R5(sp)
+ ld a2, PT_R6(sp)
+ ld a3, PT_R7(sp)
+ ld a4, PT_R8(sp)
+ ld a5, PT_R9(sp)
+#endif
li t1, _TIF_SYSCALL_TRACE | _TIF_SYSCALL_AUDIT
LONG_L t0, TI_FLAGS($28) # syscall tracing enabled?
and t0, t1, t0
@@ -67,6 +79,9 @@ #endif
dnegu v0 # error
sd v0, PT_R0(sp) # set flag for syscall restarting
1: sd v0, PT_R2(sp) # result
+#ifdef CONFIG_LTT
+ jal trace_real_syscall_exit
+#endif
local_irq_disable # make sure need_resched and
# signals dont change between
@@ -108,6 +123,9 @@ n32_syscall_trace_entry:
dnegu v0 # error
sd v0, PT_R0(sp) # set flag for syscall restarting
1: sd v0, PT_R2(sp) # result
+#ifdef CONFIG_LTT
+ jal trace_real_syscall_exit
+#endif
j syscall_exit
diff --git a/arch/mips/kernel/scall64-o32.S b/arch/mips/kernel/scall64-o32.S
index 8efb23a..2dc7b72 100644
--- a/arch/mips/kernel/scall64-o32.S
+++ b/arch/mips/kernel/scall64-o32.S
@@ -80,6 +80,25 @@ #endif
PTR 4b, bad_stack
.previous
+#ifdef CONFIG_LTT
+ sd a4, PT_R8(sp) # Save argument registers
+ sd a5, PT_R9(sp)
+ sd a6, PT_R10(sp)
+ sd a7, PT_R11(sp) # For indirect syscalls
+ sd t2, PT_R1(sp)
+ move a0, sp
+ jal trace_real_syscall_entry
+ ld t2, PT_R1(sp)
+ ld a0, PT_R4(sp) # Restore argument registers
+ ld a1, PT_R5(sp)
+ ld a2, PT_R6(sp)
+ ld a3, PT_R7(sp)
+ ld a4, PT_R8(sp)
+ ld a5, PT_R9(sp)
+ ld a6, PT_R10(sp)
+ ld a7, PT_R11(sp) # For indirect syscalls
+#endif
+
li t1, _TIF_SYSCALL_TRACE | _TIF_SYSCALL_AUDIT
LONG_L t0, TI_FLAGS($28) # syscall tracing enabled?
and t0, t1, t0
@@ -95,6 +114,9 @@ #endif
dnegu v0 # error
sd v0, PT_R0(sp) # flag for syscall restarting
1: sd v0, PT_R2(sp) # result
+#ifdef CONFIG_LTT
+ jal trace_real_syscall_exit
+#endif
o32_syscall_exit:
local_irq_disable # make need_resched and
@@ -144,6 +166,9 @@ trace_a_syscall:
dnegu v0 # error
sd v0, PT_R0(sp) # set flag for syscall restarting
1: sd v0, PT_R2(sp) # result
+#ifdef CONFIG_LTT
+ jal trace_real_syscall_exit
+#endif
j syscall_exit
diff --git a/arch/mips/kernel/syscall.c b/arch/mips/kernel/syscall.c
index 5e8a18a..476f7ff 100644
--- a/arch/mips/mm/fault.c
+++ b/arch/mips/mm/fault.c
@@ -5,6 +5,7 @@
*
* Copyright (C) 1995 - 2000 by Ralf Baechle
*/
+#include <linux/ltt/ltt-facility-kernel.h>
#include <linux/signal.h>
#include <linux/sched.h>
#include <linux/interrupt.h>
@@ -61,6 +62,8 @@ #endif
if (unlikely(address >= VMALLOC_START && address <= VMALLOC_END))
goto vmalloc_fault;
+ trace_kernel_trap_entry(CAUSE_EXCCODE(regs->cp0_cause), (void*)regs->cp0_epc);
+
/*
* If we're in an interrupt or have no user
* context, we must not take the fault..
@@ -115,6 +118,7 @@ survive:
}
up_read(&mm->mmap_sem);
+ trace_kernel_trap_exit();
return;
/*
@@ -143,6 +147,7 @@ #endif
/* info.si_code has been set above */
info.si_addr = (void __user *) address;
force_sig_info(SIGSEGV, &info, tsk);
+ trace_kernel_trap_exit();
return;
}
@@ -208,6 +213,7 @@ #endif
info.si_addr = (void __user *) address;
force_sig_info(SIGBUS, &info, tsk);
+ trace_kernel_trap_exit();
return;
vmalloc_fault:
{
@@ -245,6 +251,9 @@ vmalloc_fault:
pte_k = pte_offset_kernel(pmd_k, address);
if (!pte_present(*pte_k))
goto no_context;
+ trace_kernel_trap_entry(CAUSE_EXCCODE(regs->cp0_cause), (void*)regs->cp0_epc);
+ trace_kernel_trap_exit();
return;
}
+ trace_kernel_trap_exit();
}
diff --git a/arch/powerpc/Kconfig b/arch/powerpc/Kconfig
index 6729c98..1c770b3 100644