Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]

Groups > linux.kernel > #1484976

[PATCH] x86/dumpstack: remove NULL task pointer convention

Show all headers | View raw

On Fri, Sep 16, 2016 at 02:57:54PM +0200, Ingo Molnar wrote:
> 
> * Josh Poimboeuf <jpoimboe@redhat.com> wrote:
> 
> > On Fri, Sep 16, 2016 at 06:55:57AM -0500, Josh Poimboeuf wrote:
> > > On Fri, Sep 16, 2016 at 02:17:46AM -0700, tip-bot for Andy Lutomirski wrote:
> > > > Commit-ID:  1959a60182f48879635812a03a99c02231ea8677
> > > > Gitweb:     http://git.kernel.org/tip/1959a60182f48879635812a03a99c02231ea8677
> > > > Author:     Andy Lutomirski <luto@kernel.org>
> > > > AuthorDate: Thu, 15 Sep 2016 22:45:45 -0700
> > > > Committer:  Ingo Molnar <mingo@kernel.org>
> > > > CommitDate: Fri, 16 Sep 2016 09:18:53 +0200
> > > > 
> > > > x86/dumpstack: Pin the target stack when dumping it
> > > > 
> > > > Specifically, pin the stack in save_stack_trace_tsk() and
> > > > show_trace_log_lvl().
> > > > 
> > > > This will prevent a crash if the target task dies before or while
> > > > dumping its stack once we start freeing task stacks early.
> > > 
> > > This causes a hang:
> > 
> > The problem is that show_stack_log_lvl() can be called with a NULL
> > task_struct pointer to indicate 'current'.
> 
> Could you please send a quick fix that addresses the hang by turning the NULL into 
> 'current' or so?
> 
> > No idea why that convention exists -- IMO we should just require the
> > caller to pass 'current' directly.
> 
> For hysterical raisins I believe. It appears just a single caller passes 
> non-current. Feel free to change this.

I've verified that this fixes it.  It can be applied before Andy's patch
(or after if you don't want to rebase).

---

From: Josh Poimboeuf <jpoimboe@redhat.com>
Subject: [PATCH] x86/dumpstack: remove NULL task pointer convention

show_stack_log_lvl() and friends allow a NULL pointer for the
task_struct to indicate the current task.  This creates confusion and
can cause sneaky bugs.

Instead require the caller to pass 'current' directly.

This only changes the internal workings of the dumpstack code.  The
dump_trace() and show_stack() interfaces still allow a NULL task
pointer.  Those interfaces should also probably be fixed as well.

Signed-off-by: Josh Poimboeuf <jpoimboe@redhat.com>
---
 arch/x86/include/asm/stacktrace.h | 4 ++--
 arch/x86/kernel/dumpstack.c       | 4 +++-
 arch/x86/kernel/dumpstack_32.c    | 2 +-
 arch/x86/kernel/dumpstack_64.c    | 2 +-
 4 files changed, 7 insertions(+), 5 deletions(-)

diff --git a/arch/x86/include/asm/stacktrace.h b/arch/x86/include/asm/stacktrace.h
index 780a83e..ed2be1b 100644
--- a/arch/x86/include/asm/stacktrace.h
+++ b/arch/x86/include/asm/stacktrace.h
@@ -94,7 +94,7 @@ get_frame_pointer(struct task_struct *task, struct pt_regs *regs)
 	if (regs)
 		return (unsigned long *)regs->bp;
 
-	if (!task || task == current)
+	if (task == current)
 		return __builtin_frame_address(0);
 
 	return (unsigned long *)((struct inactive_task_frame *)task->thread.sp)->bp;
@@ -113,7 +113,7 @@ get_stack_pointer(struct task_struct *task, struct pt_regs *regs)
 	if (regs)
 		return (unsigned long *)kernel_stack_pointer(regs);
 
-	if (!task || task == current)
+	if (task == current)
 		return __builtin_frame_address(0);
 
 	return (unsigned long *)task->thread.sp;
diff --git a/arch/x86/kernel/dumpstack.c b/arch/x86/kernel/dumpstack.c
index aa208e5..e0648f7 100644
--- a/arch/x86/kernel/dumpstack.c
+++ b/arch/x86/kernel/dumpstack.c
@@ -175,11 +175,13 @@ void show_stack(struct task_struct *task, unsigned long *sp)
 {
 	unsigned long bp = 0;
 
+	task = task ? : current;
+
 	/*
 	 * Stack frames below this one aren't interesting.  Don't show them
 	 * if we're printing for %current.
 	 */
-	if (!sp && (!task || task == current)) {
+	if (!sp && task == current) {
 		sp = get_stack_pointer(current, NULL);
 		bp = (unsigned long)get_frame_pointer(current, NULL);
 	}
diff --git a/arch/x86/kernel/dumpstack_32.c b/arch/x86/kernel/dumpstack_32.c
index 2d65cfa..ded285c 100644
--- a/arch/x86/kernel/dumpstack_32.c
+++ b/arch/x86/kernel/dumpstack_32.c
@@ -200,7 +200,7 @@ void show_regs(struct pt_regs *regs)
 		u8 *ip;
 
 		pr_emerg("Stack:\n");
-		show_stack_log_lvl(NULL, regs, NULL, 0, KERN_EMERG);
+		show_stack_log_lvl(current, regs, NULL, 0, KERN_EMERG);
 
 		pr_emerg("Code:");
 
diff --git a/arch/x86/kernel/dumpstack_64.c b/arch/x86/kernel/dumpstack_64.c
index 8cb6004..4fac236 100644
--- a/arch/x86/kernel/dumpstack_64.c
+++ b/arch/x86/kernel/dumpstack_64.c
@@ -273,7 +273,7 @@ void show_regs(struct pt_regs *regs)
 		u8 *ip;
 
 		printk(KERN_DEFAULT "Stack:\n");
-		show_stack_log_lvl(NULL, regs, NULL, 0, KERN_DEFAULT);
+		show_stack_log_lvl(current, regs, NULL, 0, KERN_DEFAULT);
 
 		printk(KERN_DEFAULT "Code: ");
 
-- 
2.7.4

Back to linux.kernel | Previous | Next — Previous in thread | Next in thread | Find similar | Unroll thread

Thread

[PATCH v2 0/8] thread_info cleanups and stack caching Andy Lutomirski <luto@kernel.org> - 2016-09-16 07:50 +0200
  [PATCH v2 1/8] x86/entry/64: Fix a minor comment rebase error Andy Lutomirski <luto@kernel.org> - 2016-09-16 07:50 +0200
    [tip:x86/asm] x86/entry/64: Fix a minor comment rebase error tip-bot for Andy Lutomirski <tipbot@zytor.com> - 2016-09-16 11:20 +0200
  [PATCH v2 4/8] x86/dumpstack: Pin the target stack when dumping it Andy Lutomirski <luto@kernel.org> - 2016-09-16 07:50 +0200
    [tip:x86/asm] x86/dumpstack: Pin the target stack when dumping it tip-bot for Andy Lutomirski <tipbot@zytor.com> - 2016-09-16 11:20 +0200
      Re: [tip:x86/asm] x86/dumpstack: Pin the target stack when dumping it Josh Poimboeuf <jpoimboe@redhat.com> - 2016-09-16 14:00 +0200
        Re: [tip:x86/asm] x86/dumpstack: Pin the target stack when dumping it Josh Poimboeuf <jpoimboe@redhat.com> - 2016-09-16 14:30 +0200
          Re: [tip:x86/asm] x86/dumpstack: Pin the target stack when dumping it Ingo Molnar <mingo@kernel.org> - 2016-09-16 15:00 +0200
            [PATCH] x86/dumpstack: remove NULL task pointer convention Josh Poimboeuf <jpoimboe@redhat.com> - 2016-09-16 15:10 +0200
  [PATCH v2 6/8] lib/syscall: Pin the task stack in collect_syscall() Andy Lutomirski <luto@kernel.org> - 2016-09-16 07:50 +0200
    [tip:x86/asm] lib/syscall: Pin the task stack in collect_syscall() tip-bot for Andy Lutomirski <tipbot@zytor.com> - 2016-09-16 11:20 +0200
  [PATCH v2 8/8] fork: Cache two thread stacks per cpu if CONFIG_VMAP_STACK is set Andy Lutomirski <luto@kernel.org> - 2016-09-16 07:50 +0200
    [tip:x86/asm] fork: Optimize task creation by caching two thread  stacks per CPU if CONFIG_VMAP_STACK=y tip-bot for Andy Lutomirski <tipbot@zytor.com> - 2016-09-16 11:30 +0200
  [PATCH v2 3/8] kthread: to_live_kthread() needs try_get_task_stack() Andy Lutomirski <luto@kernel.org> - 2016-09-16 07:50 +0200
    [tip:x86/asm] kthread: Pin the stack via  try_get_task_stack()/put_task_stack() in to_live_kthread() function tip-bot for Oleg Nesterov <tipbot@zytor.com> - 2016-09-16 11:20 +0200

csiph-web