Skip to contentSkip to conversation

x86/ftrace: use uaccess in atomic context

PATCH Applied Frédéric Weisbecker 3 participants 3 messages Sep 24, 2008

Subsystems (2)
lore
Share & tools
lore
#0 lore raw

PATCH 1/1 x86/ftrace: use uaccess in atomic context

Hello,

With last -tip I get this bug:

[   49.439988] in_atomic():0, irqs_disabled():1
[   49.440118] INFO: lockdep is turned off.
[   49.440118] Pid: 2814, comm: modprobe Tainted: G        W 2.6.27-rc7 #4
[   49.440118]  [<c01215e1>] __might_sleep+0xe1/0x120
[   49.440118]  [<c01148ea>] ftrace_modify_code+0x2a/0xd0
[   49.440118]  [<c01148a2>] ? ftrace_test_p6nop+0x0/0xa
[   49.440118]  [<c016e80e>] __ftrace_update_code+0xfe/0x2f0
[   49.440118]  [<c01148a2>] ? ftrace_test_p6nop+0x0/0xa
[   49.440118]  [<c016f190>] ftrace_convert_nops+0x50/0x80
[   49.440118]  [<c016f1d6>] ftrace_init_module+0x16/0x20
[   49.440118]  [<c015498b>] load_module+0x185b/0x1d30
[   49.440118]  [<c01767a0>] ? find_get_page+0x0/0xf0
[   49.440118]  [<c02463c0>] ? sprintf+0x0/0x30
[   49.440118]  [<c034e012>] ? mutex_lock_interruptible_nested+0x1f2/0x350
[   49.440118]  [<c0154eb3>] sys_init_module+0x53/0x1b0
[   49.440118]  [<c0352340>] ? do_page_fault+0x0/0x740
[   49.440118]  [<c0104012>] syscall_call+0x7/0xb
[   49.440118]  =======================

It is because ftrace_modify_code() calls copy_to_user and copy_from_user.
These functions have been inserted after guessing that there couldn't be any race condition but copy_[to/from]_user might sleep and __ftrace_update_code is called with local_irq_saved.

These function have been inserted since this commit:
d5e92e8978fd2574e415dc2792c5eb592978243d:
"ftrace: x86 use copy from user function"

Following is a patch which correct it.
---
Adapt uaccess functions's use to atomic context in ftrace_modify_code() since we are in IRQ-off context: __ftrace_update_code() calls local_irq_save().

---

arch/x86/kernel/ftrace.c +2 -2
index 082d996..fd667f1 100644
@@ -71,13 +71,13 @@ ftrace_modify_code(unsigned long ip, unsigned char *old_code,
 	 * No real locking needed, this code is run through
 	 * kstop_machine, or before SMP starts.
 	 */
-	if (__copy_from_user(replaced, (char __user *)ip, MCOUNT_INSN_SIZE))
+	if (__copy_from_user_inatomic(replaced, (char __user *)ip, MCOUNT_INSN_SIZE))
 		return 1;
 
 	if (memcmp(replaced, old_code, MCOUNT_INSN_SIZE) != 0)
 		return 2;
 
-	WARN_ON_ONCE(__copy_to_user((char __user *)ip, new_code,
+	WARN_ON_ONCE(__copy_to_user_inatomic((char __user *)ip, new_code,
 				    MCOUNT_INSN_SIZE));
 
 	sync_core();
maintainer Acked-by on v1 1/1
#1 lore raw
Quoted 33 lines · Fr?d?ric Weisbecker

Hello,

With last -tip I get this bug:

[   49.439988] in_atomic():0, irqs_disabled():1
[   49.440118] INFO: lockdep is turned off.
[   49.440118] Pid: 2814, comm: modprobe Tainted: G        W 2.6.27-rc7 #4
[   49.440118]  [<c01215e1>] __might_sleep+0xe1/0x120
[   49.440118]  [<c01148ea>] ftrace_modify_code+0x2a/0xd0
[   49.440118]  [<c01148a2>] ? ftrace_test_p6nop+0x0/0xa
[   49.440118]  [<c016e80e>] __ftrace_update_code+0xfe/0x2f0
[   49.440118]  [<c01148a2>] ? ftrace_test_p6nop+0x0/0xa
[   49.440118]  [<c016f190>] ftrace_convert_nops+0x50/0x80
[   49.440118]  [<c016f1d6>] ftrace_init_module+0x16/0x20
[   49.440118]  [<c015498b>] load_module+0x185b/0x1d30
[   49.440118]  [<c01767a0>] ? find_get_page+0x0/0xf0
[   49.440118]  [<c02463c0>] ? sprintf+0x0/0x30
[   49.440118]  [<c034e012>] ? mutex_lock_interruptible_nested+0x1f2/0x350
[   49.440118]  [<c0154eb3>] sys_init_module+0x53/0x1b0
[   49.440118]  [<c0352340>] ? do_page_fault+0x0/0x740
[   49.440118]  [<c0104012>] syscall_call+0x7/0xb
[   49.440118]  =======================

It is because ftrace_modify_code() calls copy_to_user and copy_from_user.
These functions have been inserted after guessing that there couldn't be any race condition but copy_[to/from]_user might sleep and __ftrace_update_code is called with local_irq_saved.

These function have been inserted since this commit:
d5e92e8978fd2574e415dc2792c5eb592978243d:
"ftrace: x86 use copy from user function"

Quoted 8 lines

Following is a patch which correct it.
---
Adapt uaccess functions's use to atomic context in ftrace_modify_code() since we are in IRQ-off context: __ftrace_update_code() calls local_irq_save().

---

I had this exact patch sitting it my quilt queue. I thought I sent it out, but it must have been lost in the noise.

Thanks,

-- Steve

Applied
#2 lore raw
In reply to Frédéric Weisbecker
Quoted 29 lines · Frédéric Weisbecker

Hello,

With last -tip I get this bug:

[   49.439988] in_atomic():0, irqs_disabled():1
[   49.440118] INFO: lockdep is turned off.
[   49.440118] Pid: 2814, comm: modprobe Tainted: G        W 2.6.27-rc7 #4
[   49.440118]  [<c01215e1>] __might_sleep+0xe1/0x120
[   49.440118]  [<c01148ea>] ftrace_modify_code+0x2a/0xd0
[   49.440118]  [<c01148a2>] ? ftrace_test_p6nop+0x0/0xa
[   49.440118]  [<c016e80e>] __ftrace_update_code+0xfe/0x2f0
[   49.440118]  [<c01148a2>] ? ftrace_test_p6nop+0x0/0xa
[   49.440118]  [<c016f190>] ftrace_convert_nops+0x50/0x80
[   49.440118]  [<c016f1d6>] ftrace_init_module+0x16/0x20
[   49.440118]  [<c015498b>] load_module+0x185b/0x1d30
[   49.440118]  [<c01767a0>] ? find_get_page+0x0/0xf0
[   49.440118]  [<c02463c0>] ? sprintf+0x0/0x30
[   49.440118]  [<c034e012>] ? mutex_lock_interruptible_nested+0x1f2/0x350
[   49.440118]  [<c0154eb3>] sys_init_module+0x53/0x1b0
[   49.440118]  [<c0352340>] ? do_page_fault+0x0/0x740
[   49.440118]  [<c0104012>] syscall_call+0x7/0xb
[   49.440118]  =======================

It is because ftrace_modify_code() calls copy_to_user and copy_from_user.
These functions have been inserted after guessing that there couldn't be any race condition but copy_[to/from]_user might sleep and __ftrace_update_code is called with local_irq_saved.

Quoted 5 lines

These function have been inserted since this commit:
d5e92e8978fd2574e415dc2792c5eb592978243d:
"ftrace: x86 use copy from user function"

Following is a patch which correct it.

applied to tip/tracing/ftrace, thanks Frédéric!

Ingo