Re: [PATCH 3/4] LoongArch: Add stacktrace support

[Qing Zhang]

On 2022/8/2 上午9:30, Huacai Chen wrote:
> Hi, Qing,
>
> On Tue, Aug 2, 2022 at 9:20 AM zhangqing <zhangqing@xxxxxxxxxxx> wrote:
> > On 2022/8/1 下午11:30, Huacai Chen wrote:
> > > Hi, Qing,
> > >
> > > On Mon, Aug 1, 2022 at 8:17 PM Qing Zhang <zhangqing@xxxxxxxxxxx> wrote:
> > > > Use common arch_stack_walk infrastructure to avoid duplicated code and
> > > > avoid taking care of the stack storage and filtering.
> > > > Add sra (means __schedule return address) and scfa (means __schedule call
> > > > frame address) to thread_info and store it in switch_to().
> > > >
> > > > Now we can print the process stack by cat /proc/*/stack and can better
> > > > support ftrace.
> > > >
> > > > Signed-off-by: Qing Zhang <zhangqing@xxxxxxxxxxx>
> > > > ---
> > > >    arch/loongarch/Kconfig                 |  5 ++++
> > > >    arch/loongarch/include/asm/processor.h |  9 +++++++
> > > >    arch/loongarch/include/asm/switch_to.h | 14 ++++++----
> > > >    arch/loongarch/include/asm/uaccess.h   |  4 +--
> > > >    arch/loongarch/kernel/Makefile         |  1 +
> > > >    arch/loongarch/kernel/asm-offsets.c    |  2 ++
> > > >    arch/loongarch/kernel/process.c        |  3 +++
> > > >    arch/loongarch/kernel/stacktrace.c     | 37 ++++++++++++++++++++++++++
> > > >    arch/loongarch/kernel/switch.S         |  2 ++
> > > >    9 files changed, 70 insertions(+), 7 deletions(-)
> > > >    create mode 100644 arch/loongarch/kernel/stacktrace.c
> > > >
> > > > diff --git a/arch/loongarch/include/asm/uaccess.h b/arch/loongarch/include/asm/uaccess.h
> > > > index 2b44edc604a2..a8ae2af4025a 100644
> > > > --- a/arch/loongarch/include/asm/uaccess.h
> > > > +++ b/arch/loongarch/include/asm/uaccess.h
> > > > @@ -229,13 +229,13 @@ extern unsigned long __copy_user(void *to, const void *from, __kernel_size_t n);
> > > >    static inline unsigned long __must_check
> > > >    raw_copy_from_user(void *to, const void __user *from, unsigned long n)
> > > >    {
> > > > -       return __copy_user(to, from, n);
> > > > +       return __copy_user(to, (__force const void *)from, n);
> > > >    }
> > > >
> > > >    static inline unsigned long __must_check
> > > >    raw_copy_to_user(void __user *to, const void *from, unsigned long n)
> > > >    {
> > > > -       return __copy_user(to, from, n);
> > > > +       return __copy_user((__force void *)to, from, n);
> > > Why this? Does it have something to do with stacktrace?
> > >
> > > Huacai
> >
> > Hi, huacai
> >
> > This is kernel test robot report sparse warnings:
> > I reproduced locally and found that other architectures calling
> > __copy_user also use __force conversion, Is this modification appropriate?
> >
> > kernel/trace/trace_events_user.c: note: in included file (through
> > include/linux/uaccess.h, include/linux/sched/task.h,
> > include/linux/sched/signal.h, ...):
> >      arch/loongarch/include/asm/uaccess.h:232:32: sparse: sparse:
> > incorrect type in argument 2 (different address spaces) @@     expected
> > void const *from @@     got void const [noderef] __user *from @@
> >      arch/loongarch/include/asm/uaccess.h:232:32: sparse:     expected
> > void const *from
> >      arch/loongarch/include/asm/uaccess.h:232:32: sparse:     got void
> > const [noderef] __user *from
> The modification might be correct, but should it be in this patch?
>
> Huacai
I will send this patch separately,  it has nothing to do with the series 
patch.

arch/loongarch/kernel/ptrace.c: note: in included file (through 
include/linux/uaccess.h, include/linux/sched/task.h, 
include/linux/sched/signal.h, include/linux/ptrace.h, include/linux/audit.h)

-Qing
> > Thanks,
> > -Qing