linux-kernel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Masami Hiramatsu <mhiramat@kernel.org>
To: Steven Rostedt <rostedt@goodmis.org>, Ingo Molnar <mingo@kernel.org>
Cc: X86 ML <x86@kernel.org>, Masami Hiramatsu <mhiramat@kernel.org>,
	Daniel Xu <dxu@dxuuu.xyz>,
	linux-kernel@vger.kernel.org, bpf@vger.kernel.org,
	kuba@kernel.org, mingo@redhat.com, ast@kernel.org,
	tglx@linutronix.de, kernel-team@fb.com, yhs@fb.com,
	Josh Poimboeuf <jpoimboe@redhat.com>
Subject: [PATCH -tip v2 09/10] x86/unwind/orc: Fixup kretprobe trampoline entry
Date: Fri, 12 Mar 2021 15:43:23 +0900	[thread overview]
Message-ID: <161553140351.1038734.4282308401106671492.stgit@devnote2> (raw)
In-Reply-To: <161553130371.1038734.7661319550287837734.stgit@devnote2>

Since the kretprobe replaces the function return address with
the kretprobe_trampoline on the stack, the ORC unwinder can not
continue the stack unwinding at that point.

To fix this issue, correct state->ip as like as function-graph
tracer in the unwind_next_frame().

Signed-off-by: Masami Hiramatsu <mhiramat@kernel.org>
---
  Changes in v2:
   - Remove kretprobe wrapper functions from unwind_orc.c
   - Do not fixup state->ip when unwinding with regs because
     kretprobe fixup instruction pointer before calling handler.
---
 arch/x86/include/asm/unwind.h |    4 ++++
 arch/x86/kernel/unwind_orc.c  |   16 ++++++++++++++++
 2 files changed, 20 insertions(+)

diff --git a/arch/x86/include/asm/unwind.h b/arch/x86/include/asm/unwind.h
index 70fc159ebe69..ab5e45b848d5 100644
--- a/arch/x86/include/asm/unwind.h
+++ b/arch/x86/include/asm/unwind.h
@@ -4,6 +4,7 @@
 
 #include <linux/sched.h>
 #include <linux/ftrace.h>
+#include <linux/llist.h>
 #include <asm/ptrace.h>
 #include <asm/stacktrace.h>
 
@@ -20,6 +21,9 @@ struct unwind_state {
 	bool signal, full_regs;
 	unsigned long sp, bp, ip;
 	struct pt_regs *regs, *prev_regs;
+#if defined(CONFIG_KRETPROBES)
+	struct llist_node *kr_iter;
+#endif
 #elif defined(CONFIG_UNWINDER_FRAME_POINTER)
 	bool got_irq;
 	unsigned long *bp, *orig_sp, ip;
diff --git a/arch/x86/kernel/unwind_orc.c b/arch/x86/kernel/unwind_orc.c
index 2a1d47f47eee..1d1b9388a1b1 100644
--- a/arch/x86/kernel/unwind_orc.c
+++ b/arch/x86/kernel/unwind_orc.c
@@ -2,6 +2,7 @@
 #include <linux/objtool.h>
 #include <linux/module.h>
 #include <linux/sort.h>
+#include <linux/kprobes.h>
 #include <asm/ptrace.h>
 #include <asm/stacktrace.h>
 #include <asm/unwind.h>
@@ -536,6 +537,21 @@ bool unwind_next_frame(struct unwind_state *state)
 
 		state->ip = ftrace_graph_ret_addr(state->task, &state->graph_idx,
 						  state->ip, (void *)ip_p);
+		/*
+		 * When the stack unwinder is called from the kretprobe handler
+		 * or the interrupt handler which occurs in the kretprobe
+		 * trampoline code, %sp is shown on the stack instead of the
+		 * return address because kretprobe_trampoline() does
+		 * "push %sp" at first.
+		 * And also the unwinder may find the kretprobe_trampoline
+		 * instead of the real return address on stack.
+		 * In those cases, find the correct return address from
+		 * task->kretprobe_instances list.
+		 */
+		if (state->ip == sp ||
+		    is_kretprobe_trampoline(state->ip))
+			state->ip = kretprobe_find_ret_addr(state->task,
+							    &state->kr_iter);
 
 		state->sp = sp;
 		state->regs = NULL;


  parent reply	other threads:[~2021-03-12  6:44 UTC|newest]

Thread overview: 20+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-03-12  6:41 [PATCH -tip v2 00/10] kprobes: Fix stacktrace with kretprobes Masami Hiramatsu
2021-03-12  6:41 ` [PATCH -tip v2 01/10] ia64: kprobes: Fix to pass correct trampoline address to the handler Masami Hiramatsu
2021-03-12  6:42 ` [PATCH -tip v2 02/10] kprobes: treewide: Replace arch_deref_entry_point() with dereference_function_descriptor() Masami Hiramatsu
2021-05-24  9:26   ` Naveen N. Rao
2021-05-24 23:41     ` Masami Hiramatsu
2021-03-12  6:42 ` [PATCH -tip v2 03/10] kprobes: treewide: Remove trampoline_address from kretprobe_trampoline_handler() Masami Hiramatsu
2021-03-12  6:42 ` [PATCH -tip v2 04/10] kprobes: stacktrace: Recover the address changed by kretprobe Masami Hiramatsu
2021-03-17  0:27   ` Masami Hiramatsu
2021-03-12  6:42 ` [PATCH -tip v2 05/10] x86/kprobes: Add UNWIND_HINT_FUNC on kretprobe_trampoline code Masami Hiramatsu
2021-03-12  6:42 ` [PATCH -tip v2 06/10] ARC: Add instruction_pointer_set() API Masami Hiramatsu
2021-03-12  6:43 ` [PATCH -tip v2 07/10] ia64: " Masami Hiramatsu
2021-03-12  6:43 ` [PATCH -tip v2 08/10] kprobes: Setup instruction pointer in __kretprobe_trampoline_handler Masami Hiramatsu
2021-03-12  6:43 ` Masami Hiramatsu [this message]
2021-03-12  6:43 ` [PATCH -tip v2 10/10] tracing: Remove kretprobe unknown indicator from stacktrace Masami Hiramatsu
2021-03-12 18:56 ` [PATCH -tip v2 00/10] kprobes: Fix stacktrace with kretprobes Daniel Xu
2021-03-16  2:30 ` Josh Poimboeuf
2021-03-16  6:30   ` Masami Hiramatsu
2021-05-17 21:06     ` Andrii Nakryiko
2021-05-23 14:22       ` Masami Hiramatsu
2021-05-24 17:49         ` Andrii Nakryiko

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=161553140351.1038734.4282308401106671492.stgit@devnote2 \
    --to=mhiramat@kernel.org \
    --cc=ast@kernel.org \
    --cc=bpf@vger.kernel.org \
    --cc=dxu@dxuuu.xyz \
    --cc=jpoimboe@redhat.com \
    --cc=kernel-team@fb.com \
    --cc=kuba@kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=mingo@kernel.org \
    --cc=mingo@redhat.com \
    --cc=rostedt@goodmis.org \
    --cc=tglx@linutronix.de \
    --cc=x86@kernel.org \
    --cc=yhs@fb.com \
    /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).