[PATCH] tile: improve stack backtrace

From: Chris Metcalf
Date: Fri May 08 2015 - 11:25:39 EST

This commit fixes a number of issues with the tile backtrace code.

- Don't try to identify userspace shared object or executable paths
if we are doing a backtrace from an interrupt; it's not legal,
and also unlikely to be interesting. Likewise, don't try to do
it for other address spaces, since d_path() assumes it is being
called in "current" context.

- Move "in_backtrace" from thread_struct to thread_info.
This way we can access it even if our stack thread_info has been
clobbered, which makes backtracing more robust.

- Avoid using "current" directly when testing for is_sigreturn().
Since "current" may be corrupt, we're better off using kbt->task
explicitly to look up the vdso_base for the current task.
Conveniently, this simplifies the internal APIs (we only need
one is_sigreturn() function now).

- Avoid bogus "Odd fault" warning when pc/sp/ex1 are all zero,
as is true for kernel threads above the last frame.

- Hook into Tejun Heo's dump_stack() framework in lib/dump_stack.c.

- Write last entry in save_stack_trace() as ULONG_MAX, not zero,
since ftrace (at least) relies on finding that marker.

- Implement save_stack_trace_regs() and save_strack_trace_user(),
and set CONFIG_USER_STACKTRACE_SUPPORT.

Signed-off-by: Chris Metcalf <cmetcalf@xxxxxxxxxx>
---
arch/tile/Kconfig | 1 +
arch/tile/include/asm/processor.h | 2 -
arch/tile/include/asm/stack.h | 13 ++--
arch/tile/include/asm/thread_info.h | 1 +
arch/tile/kernel/entry.S | 7 --
arch/tile/kernel/process.c | 36 +++++++----
arch/tile/kernel/stack.c | 125 +++++++++++++++++++++---------------
arch/tile/kernel/traps.c | 3 +
arch/tile/lib/exports.c | 1 -
9 files changed, 107 insertions(+), 82 deletions(-)

diff --git a/arch/tile/Kconfig b/arch/tile/Kconfig
index 8fac40f56ec7..bcc7d66976f1 100644
--- a/arch/tile/Kconfig
+++ b/arch/tile/Kconfig
@@ -24,6 +24,7 @@ config TILE
select MODULES_USE_ELF_RELA
select HAVE_ARCH_TRACEHOOK
select HAVE_SYSCALL_TRACEPOINTS
+ select USER_STACKTRACE_SUPPORT
select ARCH_HAS_ATOMIC64_DEC_IF_POSITIVE
select HAVE_DEBUG_STACKOVERFLOW
select ARCH_WANT_FRAME_POINTERS
diff --git a/arch/tile/include/asm/processor.h b/arch/tile/include/asm/processor.h
index dd4f9f17e30a..139dfdee0134 100644
--- a/arch/tile/include/asm/processor.h
+++ b/arch/tile/include/asm/processor.h
@@ -111,8 +111,6 @@ struct thread_struct {
unsigned long long interrupt_mask;
/* User interrupt-control 0 state */
unsigned long intctrl_0;
- /* Is this task currently doing a backtrace? */
- bool in_backtrace;
/* Any other miscellaneous processor state bits */
unsigned long proc_status;
#if !CHIP_HAS_FIXED_INTVEC_BASE()
diff --git a/arch/tile/include/asm/stack.h b/arch/tile/include/asm/stack.h
index 0e9d382a2d45..c3cb42615a9f 100644
--- a/arch/tile/include/asm/stack.h
+++ b/arch/tile/include/asm/stack.h
@@ -58,17 +58,14 @@ extern int KBacktraceIterator_end(struct KBacktraceIterator *kbt);
/* Advance to the next frame. */
extern void KBacktraceIterator_next(struct KBacktraceIterator *kbt);

+/* Dump just the contents of the pt_regs structure. */
+extern void tile_show_regs(struct pt_regs *);
+
/*
* Dump stack given complete register info. Use only from the
* architecture-specific code; show_stack()
- * and dump_stack() (in entry.S) are architecture-independent entry points.
+ * and dump_stack() are architecture-independent entry points.
*/
-extern void tile_show_stack(struct KBacktraceIterator *, int headers);
-
-/* Dump stack of current process, with registers to seed the backtrace. */
-extern void dump_stack_regs(struct pt_regs *);
-
-/* Helper method for assembly dump_stack(). */
-extern void _dump_stack(int dummy, ulong pc, ulong lr, ulong sp, ulong r52);
+extern void tile_show_stack(struct KBacktraceIterator *);

#endif /* _ASM_TILE_STACK_H */
diff --git a/arch/tile/include/asm/thread_info.h b/arch/tile/include/asm/thread_info.h
index f804c39a5e4d..dc1fb28d9636 100644
--- a/arch/tile/include/asm/thread_info.h
+++ b/arch/tile/include/asm/thread_info.h
@@ -42,6 +42,7 @@ struct thread_info {
unsigned long unalign_jit_tmp[4]; /* temp r0..r3 storage */
void __user *unalign_jit_base; /* unalign fixup JIT base */
#endif
+ bool in_backtrace; /* currently doing backtrace? */
};

/*
diff --git a/arch/tile/kernel/entry.S b/arch/tile/kernel/entry.S
index 3d9175992a20..670a3569450f 100644
--- a/arch/tile/kernel/entry.S
+++ b/arch/tile/kernel/entry.S
@@ -27,13 +27,6 @@ STD_ENTRY(current_text_addr)
{ move r0, lr; jrp lr }
STD_ENDPROC(current_text_addr)

-STD_ENTRY(dump_stack)
- { move r2, lr; lnk r1 }
- { move r4, r52; addli r1, r1, dump_stack - . }
- { move r3, sp; j _dump_stack }
- jrp lr /* keep backtracer happy */
- STD_ENDPROC(dump_stack)
-
STD_ENTRY(KBacktraceIterator_init_current)
{ move r2, lr; lnk r1 }
{ move r4, r52; addli r1, r1, KBacktraceIterator_init_current - . }
diff --git a/arch/tile/kernel/process.c b/arch/tile/kernel/process.c
index 96ea75e04582..a45213781ad0 100644
--- a/arch/tile/kernel/process.c
+++ b/arch/tile/kernel/process.c
@@ -546,33 +546,43 @@ void exit_thread(void)
#endif
}

-void show_regs(struct pt_regs *regs)
+void tile_show_regs(struct pt_regs *regs)
{
- struct task_struct *tsk = validate_current();
int i;
-
- if (tsk != &corrupt_current)
- show_regs_print_info(KERN_ERR);
#ifdef __tilegx__
for (i = 0; i < 17; i++)
- pr_err(" r%-2d: " REGFMT " r%-2d: " REGFMT " r%-2d: " REGFMT "\n",
+ pr_err(" r%-2d: "REGFMT" r%-2d: "REGFMT" r%-2d: "REGFMT"\n",
i, regs->regs[i], i+18, regs->regs[i+18],
i+36, regs->regs[i+36]);
- pr_err(" r17: " REGFMT " r35: " REGFMT " tp : " REGFMT "\n",
+ pr_err(" r17: "REGFMT" r35: "REGFMT" tp : "REGFMT"\n",
regs->regs[17], regs->regs[35], regs->tp);
- pr_err(" sp : " REGFMT " lr : " REGFMT "\n", regs->sp, regs->lr);
+ pr_err(" sp : "REGFMT" lr : "REGFMT"\n", regs->sp, regs->lr);
#else
for (i = 0; i < 13; i++)
- pr_err(" r%-2d: " REGFMT " r%-2d: " REGFMT " r%-2d: " REGFMT " r%-2d: " REGFMT "\n",
+ pr_err(" r%-2d: "REGFMT" r%-2d: "REGFMT
+ " r%-2d: "REGFMT" r%-2d: "REGFMT"\n",
i, regs->regs[i], i+14, regs->regs[i+14],
i+27, regs->regs[i+27], i+40, regs->regs[i+40]);
- pr_err(" r13: " REGFMT " tp : " REGFMT " sp : " REGFMT " lr : " REGFMT "\n",
+ pr_err(" r13: "REGFMT" tp : "REGFMT" sp : "REGFMT" lr : "REGFMT"\n",
regs->regs[13], regs->tp, regs->sp, regs->lr);
#endif
- pr_err(" pc : " REGFMT " ex1: %ld faultnum: %ld\n",
- regs->pc, regs->ex1, regs->faultnum);
+ pr_err(" pc : "REGFMT" ex1: %ld faultnum: %ld flags:%s%s%s%s\n",
+ regs->pc, regs->ex1, regs->faultnum,
+ is_compat_task() ? " compat" : "",
+ (regs->flags & PT_FLAGS_DISABLE_IRQ) ? " noirq" : "",
+ !(regs->flags & PT_FLAGS_CALLER_SAVES) ? " nocallersave" : "",
+ (regs->flags & PT_FLAGS_RESTORE_REGS) ? " restoreregs" : "");
+}
+
+void show_regs(struct pt_regs *regs)
+{
+ struct KBacktraceIterator kbt;
+
+ show_regs_print_info(KERN_DEFAULT);
+ tile_show_regs(regs);

- dump_stack_regs(regs);
+ KBacktraceIterator_init(&kbt, NULL, regs);
+ tile_show_stack(&kbt);
}

/* To ensure stack dump on tiles occurs one by one. */
diff --git a/arch/tile/kernel/stack.c b/arch/tile/kernel/stack.c
index c42dce50acd8..35d34635e4f1 100644
--- a/arch/tile/kernel/stack.c
+++ b/arch/tile/kernel/stack.c
@@ -23,6 +23,7 @@
#include <linux/mmzone.h>
#include <linux/dcache.h>
#include <linux/fs.h>
+#include <linux/hardirq.h>
#include <linux/string.h>
#include <asm/backtrace.h>
#include <asm/page.h>
@@ -109,7 +110,7 @@ static struct pt_regs *valid_fault_handler(struct KBacktraceIterator* kbt)
if (kbt->verbose)
pr_err(" <%s while in user mode>\n", fault);
} else {
- if (kbt->verbose)
+ if (kbt->verbose && (p->pc != 0 || p->sp != 0 || p->ex1 != 0))
pr_err(" (odd fault: pc %#lx, sp %#lx, ex1 %#lx?)\n",
p->pc, p->sp, p->ex1);
return NULL;
@@ -119,10 +120,12 @@ static struct pt_regs *valid_fault_handler(struct KBacktraceIterator* kbt)
return p;
}

-/* Is the pc pointing to a sigreturn trampoline? */
-static int is_sigreturn(unsigned long pc)
+/* Is the iterator pointing to a sigreturn trampoline? */
+static int is_sigreturn(struct KBacktraceIterator *kbt)
{
- return current->mm && (pc == VDSO_SYM(&__vdso_rt_sigreturn));
+ return kbt->task->mm &&
+ (kbt->it.pc == ((ulong)kbt->task->mm->context.vdso_base +
+ (ulong)&__vdso_rt_sigreturn));
}

/* Return a pt_regs pointer for a valid signal handler frame */
@@ -131,7 +134,7 @@ static struct pt_regs *valid_sigframe(struct KBacktraceIterator* kbt,
{
BacktraceIterator *b = &kbt->it;

- if (is_sigreturn(b->pc) && b->sp < PAGE_OFFSET &&
+ if (is_sigreturn(kbt) && b->sp < PAGE_OFFSET &&
b->sp % sizeof(long) == 0) {
int retval;
pagefault_disable();
@@ -151,11 +154,6 @@ static struct pt_regs *valid_sigframe(struct KBacktraceIterator* kbt,
return NULL;
}

-static int KBacktraceIterator_is_sigreturn(struct KBacktraceIterator *kbt)
-{
- return is_sigreturn(kbt->it.pc);
-}
-
static int KBacktraceIterator_restart(struct KBacktraceIterator *kbt)
{
struct pt_regs *p;
@@ -178,7 +176,7 @@ static int KBacktraceIterator_next_item_inclusive(
{
for (;;) {
do {
- if (!KBacktraceIterator_is_sigreturn(kbt))
+ if (!is_sigreturn(kbt))
return KBT_ONGOING;
} while (backtrace_next(&kbt->it));

@@ -357,51 +355,50 @@ static void describe_addr(struct KBacktraceIterator *kbt,
*/
static bool start_backtrace(void)
{
- if (current->thread.in_backtrace) {
+ if (current_thread_info()->in_backtrace) {
pr_err("Backtrace requested while in backtrace!\n");
return false;
}
- current->thread.in_backtrace = true;
+ current_thread_info()->in_backtrace = true;
return true;
}

static void end_backtrace(void)
{
- current->thread.in_backtrace = false;
+ current_thread_info()->in_backtrace = false;
}

/*
* This method wraps the backtracer's more generic support.
* It is only invoked from the architecture-specific code; show_stack()
- * and dump_stack() (in entry.S) are architecture-independent entry points.
+ * and dump_stack() are architecture-independent entry points.
*/
-void tile_show_stack(struct KBacktraceIterator *kbt, int headers)
+void tile_show_stack(struct KBacktraceIterator *kbt)
{
int i;
int have_mmap_sem = 0;

if (!start_backtrace())
return;
- if (headers) {
- /*
- * Add a blank line since if we are called from panic(),
- * then bust_spinlocks() spit out a space in front of us
- * and it will mess up our KERN_ERR.
- */
- pr_err("Starting stack dump of tid %d, pid %d (%s) on cpu %d at cycle %lld\n",
- kbt->task->pid, kbt->task->tgid, kbt->task->comm,
- raw_smp_processor_id(), get_cycles());
- }
kbt->verbose = 1;
i = 0;
for (; !KBacktraceIterator_end(kbt); KBacktraceIterator_next(kbt)) {
char namebuf[KSYM_NAME_LEN+100];
unsigned long address = kbt->it.pc;

- /* Try to acquire the mmap_sem as we pass into userspace. */
- if (address < PAGE_OFFSET && !have_mmap_sem && kbt->task->mm)
+ /*
+ * Try to acquire the mmap_sem as we pass into userspace.
+ * If we're in an interrupt context, don't even try, since
+ * it's not safe to call e.g. d_path() from an interrupt,
+ * since it uses spin locks without disabling interrupts.
+ * Note we test "kbt->task == current", not "kbt->is_current",
+ * since we're checking that "current" will work in d_path().
+ */
+ if (kbt->task == current && address < PAGE_OFFSET &&
+ !have_mmap_sem && kbt->task->mm && !in_interrupt()) {
have_mmap_sem =
down_read_trylock(&kbt->task->mm->mmap_sem);
+ }

describe_addr(kbt, address, have_mmap_sem,
namebuf, sizeof(namebuf));
@@ -416,24 +413,12 @@ void tile_show_stack(struct KBacktraceIterator *kbt, int headers)
}
if (kbt->end == KBT_LOOP)
pr_err("Stack dump stopped; next frame identical to this one\n");
- if (headers)
- pr_err("Stack dump complete\n");
if (have_mmap_sem)
up_read(&kbt->task->mm->mmap_sem);
end_backtrace();
}
EXPORT_SYMBOL(tile_show_stack);

-
-/* This is called from show_regs() and _dump_stack() */
-void dump_stack_regs(struct pt_regs *regs)
-{
- struct KBacktraceIterator kbt;
- KBacktraceIterator_init(&kbt, NULL, regs);
- tile_show_stack(&kbt, 1);
-}
-EXPORT_SYMBOL(dump_stack_regs);
-
static struct pt_regs *regs_to_pt_regs(struct pt_regs *regs,
ulong pc, ulong lr, ulong sp, ulong r52)
{
@@ -445,11 +430,15 @@ static struct pt_regs *regs_to_pt_regs(struct pt_regs *regs,
return regs;
}

-/* This is called from dump_stack() and just converts to pt_regs */
+/* Deprecated function currently only used by kernel_double_fault(). */
void _dump_stack(int dummy, ulong pc, ulong lr, ulong sp, ulong r52)
{
+ struct KBacktraceIterator kbt;
struct pt_regs regs;
- dump_stack_regs(regs_to_pt_regs(&regs, pc, lr, sp, r52));
+
+ regs_to_pt_regs(&regs, pc, lr, sp, r52);
+ KBacktraceIterator_init(&kbt, NULL, &regs);
+ tile_show_stack(&kbt);
}

/* This is called from KBacktraceIterator_init_current() */
@@ -461,22 +450,30 @@ void _KBacktraceIterator_init_current(struct KBacktraceIterator *kbt, ulong pc,
regs_to_pt_regs(&regs, pc, lr, sp, r52));
}

-/* This is called only from kernel/sched/core.c, with esp == NULL */
+/*
+ * Called from sched_show_task() with task != NULL, or dump_stack()
+ * with task == NULL. The esp argument is always NULL.
+ */
void show_stack(struct task_struct *task, unsigned long *esp)
{
struct KBacktraceIterator kbt;
- if (task == NULL || task == current)
+ if (task == NULL || task == current) {
KBacktraceIterator_init_current(&kbt);
- else
+ KBacktraceIterator_next(&kbt); /* don't show first frame */
+ } else {
KBacktraceIterator_init(&kbt, task, NULL);
- tile_show_stack(&kbt, 0);
+ }
+ tile_show_stack(&kbt);
}

#ifdef CONFIG_STACKTRACE

/* Support generic Linux stack API too */

-void save_stack_trace_tsk(struct task_struct *task, struct stack_trace *trace)
+static void save_stack_trace_common(struct task_struct *task,
+ struct pt_regs *regs,
+ bool user,
+ struct stack_trace *trace)
{
struct KBacktraceIterator kbt;
int skip = trace->skip;
@@ -484,31 +481,57 @@ void save_stack_trace_tsk(struct task_struct *task, struct stack_trace *trace)

if (!start_backtrace())
goto done;
- if (task == NULL || task == current)
+ if (regs != NULL) {
+ KBacktraceIterator_init(&kbt, NULL, regs);
+ } else if (task == NULL || task == current) {
KBacktraceIterator_init_current(&kbt);
- else
+ skip++; /* don't show KBacktraceIterator_init_current */
+ } else {
KBacktraceIterator_init(&kbt, task, NULL);
+ }
for (; !KBacktraceIterator_end(&kbt); KBacktraceIterator_next(&kbt)) {
if (skip) {
--skip;
continue;
}
- if (i >= trace->max_entries || kbt.it.pc < PAGE_OFFSET)
+ if (i >= trace->max_entries ||
+ (!user && kbt.it.pc < PAGE_OFFSET))
break;
trace->entries[i++] = kbt.it.pc;
}
end_backtrace();
done:
+ if (i < trace->max_entries)
+ trace->entries[i++] = ULONG_MAX;
trace->nr_entries = i;
}
+
+void save_stack_trace_tsk(struct task_struct *task, struct stack_trace *trace)
+{
+ save_stack_trace_common(task, NULL, false, trace);
+}
EXPORT_SYMBOL(save_stack_trace_tsk);

void save_stack_trace(struct stack_trace *trace)
{
- save_stack_trace_tsk(NULL, trace);
+ save_stack_trace_common(NULL, NULL, false, trace);
}
EXPORT_SYMBOL_GPL(save_stack_trace);

+void save_stack_trace_regs(struct pt_regs *regs, struct stack_trace *trace)
+{
+ save_stack_trace_common(NULL, regs, false, trace);
+}
+
+void save_stack_trace_user(struct stack_trace *trace)
+{
+ /* Trace user stack if we are not a kernel thread. */
+ if (current->mm)
+ save_stack_trace_common(NULL, task_pt_regs(current),
+ true, trace);
+ else if (trace->nr_entries < trace->max_entries)
+ trace->entries[trace->nr_entries++] = ULONG_MAX;
+}
#endif

/* In entry.S */
diff --git a/arch/tile/kernel/traps.c b/arch/tile/kernel/traps.c
index 855f7316f1ee..0011a9ff0525 100644
--- a/arch/tile/kernel/traps.c
+++ b/arch/tile/kernel/traps.c
@@ -407,6 +407,9 @@ void do_nmi(struct pt_regs *regs, int fault_num, unsigned long reason)
}
}

+/* Deprecated function currently only used here. */
+extern void _dump_stack(int dummy, ulong pc, ulong lr, ulong sp, ulong r52);
+
void kernel_double_fault(int dummy, ulong pc, ulong lr, ulong sp, ulong r52)
{
_dump_stack(dummy, pc, lr, sp, r52);
diff --git a/arch/tile/lib/exports.c b/arch/tile/lib/exports.c
index 16326f288177..9d171ca4302c 100644
--- a/arch/tile/lib/exports.c
+++ b/arch/tile/lib/exports.c
@@ -26,7 +26,6 @@ EXPORT_SYMBOL(finv_user_asm);
#include <linux/kernel.h>
#include <asm/processor.h>
EXPORT_SYMBOL(current_text_addr);
-EXPORT_SYMBOL(dump_stack);

/* arch/tile/kernel/head.S */
EXPORT_SYMBOL(empty_zero_page);
--
2.1.2

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/