Re: [PATCH v4 2/2] arm64: Expand the stack trace feature to support IRQ stack

From: Jungseok Lee
Date: Wed Oct 14 2015 - 08:55:42 EST


On Oct 14, 2015, at 9:24 PM, Jungseok Lee wrote:
> On Oct 14, 2015, at 4:13 PM, AKASHI Takahiro wrote:
>> On 10/09/2015 11:24 PM, James Morse wrote:
>>> Hi Jungseok,
>>>
>>> On 07/10/15 16:28, Jungseok Lee wrote:
>>>> Currently, a call trace drops a process stack walk when a separate IRQ
>>>> stack is used. It makes a call trace information much less useful when
>>>> a system gets paniked in interrupt context.
>>>
>>> panicked
>>>
>>>> This patch addresses the issue with the following schemes:
>>>>
>>>> - Store aborted stack frame data
>>>> - Decide whether another stack walk is needed or not via current sp
>>>> - Loosen the frame pointer upper bound condition
>>>
>>> It may be worth merging this patch with its predecessor - anyone trying to
>>> bisect a problem could land between these two patches, and spend time
>>> debugging the truncated call traces.
>>>
>>>
>>>> diff --git a/arch/arm64/include/asm/irq.h b/arch/arm64/include/asm/irq.h
>>>> index 6ea82e8..e5904a1 100644
>>>> --- a/arch/arm64/include/asm/irq.h
>>>> +++ b/arch/arm64/include/asm/irq.h
>>>> @@ -2,13 +2,25 @@
>>>> #define __ASM_IRQ_H
>>>>
>>>> #include <linux/irqchip/arm-gic-acpi.h>
>>>> +#include <asm/stacktrace.h>
>>>>
>>>> #include <asm-generic/irq.h>
>>>>
>>>> struct irq_stack {
>>>> void *stack;
>>>> + struct stackframe frame;
>>>> };
>>>>
>>>> +DECLARE_PER_CPU(struct irq_stack, irq_stacks);
>>>
>>> Good idea, storing this in the per-cpu data makes it immune to stack
>>> corruption.
>>
>> Is this the only reason that you have a dummy stack frame in per-cpu data?
>> By placing this frame in an interrupt stack, I think, we will be able to eliminate
>> changes in dump_stace(). and
>>
>>>
>>>> diff --git a/arch/arm64/kernel/stacktrace.c b/arch/arm64/kernel/stacktrace.c
>>>> index 407991b..5124649 100644
>>>> --- a/arch/arm64/kernel/stacktrace.c
>>>> +++ b/arch/arm64/kernel/stacktrace.c
>>>> @@ -43,7 +43,27 @@ int notrace unwind_frame(struct stackframe *frame)
>>>> low = frame->sp;
>>>> high = ALIGN(low, THREAD_SIZE);
>>>>
>>>> - if (fp < low || fp > high - 0x18 || fp & 0xf)
>>>> + /*
>>>> + * A frame pointer would reach an upper bound if a prologue of the
>>>> + * first function of call trace looks as follows:
>>>> + *
>>>> + * stp x29, x30, [sp,#-16]!
>>>> + * mov x29, sp
>>>> + *
>>>> + * Thus, the upper bound is (top of stack - 0x20) with consideration
>>>
>>> The terms 'top' and 'bottom' of the stack are confusing, your 'top' appears
>>> to be the highest address, which is used first, making it the bottom of the
>>> stack.
>>>
>>> I would try to use the terms low/est and high/est, in keeping with the
>>> variable names in use here.
>>>
>>>
>>>> + * of a 16-byte empty space in THREAD_START_SP.
>>>> + *
>>>> + * The value, 0x20, however, does not cover all cases as interrupts
>>>> + * are handled using a separate stack. That is, a call trace can start
>>>> + * from elx_irq exception vectors. The symbols could not be promoted
>>>> + * to candidates for a stack trace under the restriction, 0x20.
>>>> + *
>>>> + * The scenario is handled without complexity as 1) considering
>>>> + * (bottom of stack + THREAD_START_SP) as a dummy frame pointer, the
>>>> + * content of which is 0, and 2) allowing the case, which changes
>>>> + * the value to 0x10 from 0x20.
>>>
>>> Where has 0x20 come from? The old value was 0x18.
>>>
>>> My understanding is the highest part of the stack looks like this:
>>> high [ off-stack ]
>>> high - 0x08 [ left free by THREAD_START_SP ]
>>> high - 0x10 [ left free by THREAD_START_SP ]
>>> high - 0x18 [#1 x30 ]
>>> high - 0x20 [#1 x29 ]
>>>
>>> So the condition 'fp > high - 0x18' prevents returning either 'left free'
>>> address, or off-stack-value as a frame. Changing it to 'fp > high - 0x10'
>>> allows the first half of that reserved area to be a valid stack frame.
>>>
>>> This change is breaking perf using incantations [0] and [1]:
>>>
>>> Before, with just patch 1/2:
>>> ---__do_softirq
>>> |
>>> |--92.95%-- __handle_domain_irq
>>> | __irqentry_text_start
>>> | el1_irq
>>> |
>>>
>>> After, with both patches:
>>> ---__do_softirq
>>> |
>>> |--83.83%-- __handle_domain_irq
>>> | __irqentry_text_start
>>> | el1_irq
>>> | |
>>> | |--99.39%-- 0x400008040d00000c
>>> | --0.61%-- [...]
>>> |
>>
>> This also shows that walk_stackframe() doesn't walk through a process stack.
>> Now I'm trying the following hack on top of Jungseok's patch.
>> (It doesn't traverse from an irq stack to an process stack yet. I need modify
>> unwind_frame().)
>
> I've got a difference between perf and dump_backtrace() as reviewing perf call
> chain operation. Perf relies on walk_stackframe(), but dump_backtrace() does not.
> That is, a symbol is printed out *before* unwind_frame() call in case of perf.
> By contrast, dump_backtrace() records a symbol *after* unwind_frame(). I think
> perf behavior is correct since frame.pc is retrieved from a valid stack frame.
>
> So, the following diff is a prerequisite. It looks reasonable to remove dump_mem()
> call since frame.sp is calculated incorrectly now. If accepted, dump_backtrace()
> could utilize walk_stackframe(), which simplifies the code.
>
> ----8<----
> diff --git a/arch/arm64/kernel/traps.c b/arch/arm64/kernel/traps.c
> index f93aae5..e18be43 100644
> --- a/arch/arm64/kernel/traps.c
> +++ b/arch/arm64/kernel/traps.c
> @@ -103,12 +103,15 @@ static void dump_mem(const char *lvl, const char *str, unsigned long bottom,
> set_fs(fs);
> }
>
> -static void dump_backtrace_entry(unsigned long where, unsigned long stack)
> +static void dump_backtrace_entry(unsigned long where)
> {
> + /*
> + * PC has a physical address when MMU is disabled.
> + */
> + if (!kernel_text_address(where))
> + where = (unsigned long)phys_to_virt(where);
> +
> print_ip_sym(where);
> - if (in_exception_text(where))
> - dump_mem("", "Exception stack", stack,
> - stack + sizeof(struct pt_regs), false);
> }
>
> static void dump_instr(const char *lvl, struct pt_regs *regs)
> @@ -172,12 +175,17 @@ static void dump_backtrace(struct pt_regs *regs, struct task_struct *tsk)
> pr_emerg("Call trace:\n");
> while (1) {
> unsigned long where = frame.pc;
> + unsigned long stack;
> int ret;
>
> + dump_backtrace_entry(where);
> ret = unwind_frame(&frame);
> if (ret < 0)
> break;
> - dump_backtrace_entry(where, frame.sp);
> + stack = frame.sp;
> + if (in_exception_text(where))
> + dump_mem("", "Exception stack", stack,
> + stack + sizeof(struct pt_regs), false);
> }
> }
> ----8<----
>
>> Thanks,
>> -Takahiro AKASHI
>> ----8<----
>> diff --git a/arch/arm64/kernel/entry.S b/arch/arm64/kernel/entry.S
>> index 650cc05..5fbd1ea 100644
>> --- a/arch/arm64/kernel/entry.S
>> +++ b/arch/arm64/kernel/entry.S
>> @@ -185,14 +185,12 @@ alternative_endif
>> mov x23, sp
>> and x23, x23, #~(THREAD_SIZE - 1)
>> cmp x20, x23 // check irq re-enterance
>> + mov x19, sp
>> beq 1f
>> - str x29, [x19, #IRQ_FRAME_FP]
>> - str x21, [x19, #IRQ_FRAME_SP]
>> - str x22, [x19, #IRQ_FRAME_PC]
>> - mov x29, x24
>> -1: mov x19, sp
>> - csel x23, x19, x24, eq // x24 = top of irq stack
>> - mov sp, x23
>> + mov sp, x24 // x24 = top of irq stack
>> + stp x29, x22, [sp, #-32]!
>> + mov x29, sp
>> +1:
>> .endm
>>
>> /*
>
> Is it possible to decide which stack is used without aborted SP information?

We could know which stack is used via current SP, but how could we decide
a variable 'low' in unwind_frame() when walking a process stack?

Sorry for confusion.

Best Regards
Jungseok Lee--
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/