linux-arm-kernel.lists.infradead.org archive mirror
 help / color / mirror / Atom feed
From: ard.biesheuvel@linaro.org (Ard Biesheuvel)
To: linux-arm-kernel@lists.infradead.org
Subject: [PATCH 2/4] arm64: unwind: disregard frame.sp when validating frame pointer
Date: Mon, 24 Jul 2017 12:26:21 +0100	[thread overview]
Message-ID: <20170724112623.26109-3-ard.biesheuvel@linaro.org> (raw)
In-Reply-To: <20170724112623.26109-1-ard.biesheuvel@linaro.org>

Currently, when unwinding the call stack, we validate the frame pointer
of each frame against frame.sp, whose value is not clearly defined, and
which makes it more difficult to link stack frames together across
different stacks. It is far better to simply check whether the frame
pointer itself points into a valid stack.

Cc: Mark Rutland <mark.rutland@arm.com>
Cc: Catalin Marinas <catalin.marinas@arm.com>
Cc: James Morse <james.morse@arm.com>
Cc: Will Deacon <will.deacon@arm.com>
Signed-off-by: Ard Biesheuvel <ard.biesheuvel@linaro.org>
---
 arch/arm64/include/asm/irq.h   | 10 +++++++-
 arch/arm64/kernel/stacktrace.c | 24 ++++++--------------
 2 files changed, 16 insertions(+), 18 deletions(-)

diff --git a/arch/arm64/include/asm/irq.h b/arch/arm64/include/asm/irq.h
index c01a912fbb1e..57deb5abfc24 100644
--- a/arch/arm64/include/asm/irq.h
+++ b/arch/arm64/include/asm/irq.h
@@ -7,6 +7,7 @@
 #ifndef __ASSEMBLER__
 
 #include <linux/percpu.h>
+#include <linux/sched/task_stack.h>
 
 #include <asm-generic/irq.h>
 #include <asm/thread_info.h>
@@ -49,12 +50,19 @@ static inline int nr_legacy_irqs(void)
 
 static inline bool on_irq_stack(unsigned long sp)
 {
-	/* variable names the same as kernel/stacktrace.c */
 	unsigned long low = (unsigned long)raw_cpu_ptr(irq_stack);
 	unsigned long high = low + IRQ_STACK_START_SP;
 
 	return (low <= sp && sp <= high);
 }
 
+static inline bool on_task_stack(struct task_struct *tsk, unsigned long sp)
+{
+	unsigned long low = (unsigned long)task_stack_page(tsk);
+	unsigned long high = low + THREAD_SIZE;
+
+	return (low <= sp && sp < high);
+}
+
 #endif /* !__ASSEMBLER__ */
 #endif
diff --git a/arch/arm64/kernel/stacktrace.c b/arch/arm64/kernel/stacktrace.c
index 6ffb965be641..beaf51fb3088 100644
--- a/arch/arm64/kernel/stacktrace.c
+++ b/arch/arm64/kernel/stacktrace.c
@@ -42,9 +42,10 @@
  */
 int notrace unwind_frame(struct task_struct *tsk, struct stackframe *frame)
 {
-	unsigned long high, low;
 	unsigned long fp = frame->fp;
-	unsigned long irq_stack_ptr;
+
+	if (fp & 0xf)
+		return -EINVAL;
 
 	if (!tsk)
 		tsk = current;
@@ -53,19 +54,8 @@ int notrace unwind_frame(struct task_struct *tsk, struct stackframe *frame)
 	 * Switching between stacks is valid when tracing current and in
 	 * non-preemptible context.
 	 */
-	if (tsk == current && !preemptible())
-		irq_stack_ptr = IRQ_STACK_PTR();
-	else
-		irq_stack_ptr = 0;
-
-	low  = frame->sp;
-	/* irq stacks are not THREAD_SIZE aligned */
-	if (on_irq_stack(frame->sp))
-		high = irq_stack_ptr;
-	else
-		high = ALIGN(low, THREAD_SIZE) - 0x20;
-
-	if (fp < low || fp > high || fp & 0xf)
+	if (!(tsk == current && !preemptible() && on_irq_stack(fp)) &&
+	    !on_task_stack(tsk, fp))
 		return -EINVAL;
 
 	frame->sp = fp + 0x10;
@@ -94,9 +84,9 @@ int notrace unwind_frame(struct task_struct *tsk, struct stackframe *frame)
 	 * Check the frame->fp we read from the bottom of the irq_stack,
 	 * and the original task stack pointer are both in current->stack.
 	 */
-	if (frame->sp == irq_stack_ptr) {
+	if (frame->sp == IRQ_STACK_PTR()) {
 		struct pt_regs *irq_args;
-		unsigned long orig_sp = IRQ_STACK_TO_TASK_STACK(irq_stack_ptr);
+		unsigned long orig_sp = IRQ_STACK_TO_TASK_STACK(frame->sp);
 
 		if (object_is_on_stack((void *)orig_sp) &&
 		   object_is_on_stack((void *)frame->fp)) {
-- 
2.9.3

  parent reply	other threads:[~2017-07-24 11:26 UTC|newest]

Thread overview: 12+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2017-07-24 11:26 [PATCH 0/4] arm64: unwind: fix broken exception stack dump Ard Biesheuvel
2017-07-24 11:26 ` [PATCH 1/4] arm64: avoid percpu indirection for irq stack ops Ard Biesheuvel
2017-07-24 11:26 ` Ard Biesheuvel [this message]
2017-07-24 11:26 ` [PATCH 3/4] arm64: unwind: reference pt_regs via embedded stack frame Ard Biesheuvel
2017-07-24 17:54   ` Mark Rutland
2017-07-24 18:34     ` Ard Biesheuvel
2017-07-24 19:09       ` Ard Biesheuvel
2017-07-25  9:53         ` Mark Rutland
2017-07-25 10:07           ` Ard Biesheuvel
2017-07-25 13:13     ` Will Deacon
2017-07-25 13:58       ` Mark Rutland
2017-07-24 11:26 ` [PATCH 4/4] arm64: unwind: remove sp from struct stackframe Ard Biesheuvel

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20170724112623.26109-3-ard.biesheuvel@linaro.org \
    --to=ard.biesheuvel@linaro.org \
    --cc=linux-arm-kernel@lists.infradead.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).