RE: [PATCH] scripts/gdb: document lx_current is only supported by x86

[Song Bao Hua (Barry Song)]

> -----Original Message-----
> From: Kieran Bingham [mailto:kieran.bingham@xxxxxxxxxxxxxxxx]
> Sent: Tuesday, February 23, 2021 12:06 AM
> To: Song Bao Hua (Barry Song) <song.bao.hua@xxxxxxxxxxxxx>; corbet@xxxxxxx;
> linux-doc@xxxxxxxxxxxxxxx; jan.kiszka@xxxxxxxxxxx
> Cc: linux-kernel@xxxxxxxxxxxxxxx; linuxarm@xxxxxxxxxxxxx
> Subject: Re: [PATCH] scripts/gdb: document lx_current is only supported by x86
> 
> Hi Barry
> 
> On 21/02/2021 21:35, Barry Song wrote:
> > lx_current depends on the per_cpu current_task which exists on x86 only:
> >
> > arch$ git grep current_task | grep -i per_cpu
> > x86/include/asm/current.h:DECLARE_PER_CPU(struct task_struct *,
> current_task);
> > x86/kernel/cpu/common.c:DEFINE_PER_CPU(struct task_struct *, current_task)
> ____cacheline_aligned =
> > x86/kernel/cpu/common.c:EXPORT_PER_CPU_SYMBOL(current_task);
> > x86/kernel/cpu/common.c:DEFINE_PER_CPU(struct task_struct *, current_task)
> = &init_task;
> > x86/kernel/cpu/common.c:EXPORT_PER_CPU_SYMBOL(current_task);
> > x86/kernel/smpboot.c:	per_cpu(current_task, cpu) = idle;
> >
> > On other architectures, lx_current() will lead to a python exception:
> > (gdb) p $lx_current().pid
> > Python Exception <class 'gdb.error'> No symbol "current_task" in current
> context.:
> > Error occurred in Python: No symbol "current_task" in current context.
> >
> > To avoid more people struggling and wasting time in other architectures,
> > document it.
> >
> > Cc: Jan Kiszka <jan.kiszka@xxxxxxxxxxx>
> > Signed-off-by: Barry Song <song.bao.hua@xxxxxxxxxxxxx>
> > ---
> >  Documentation/dev-tools/gdb-kernel-debugging.rst |  2 +-
> >  scripts/gdb/linux/cpus.py                        | 10 ++++++++--
> >  2 files changed, 9 insertions(+), 3 deletions(-)
> >
> > diff --git a/Documentation/dev-tools/gdb-kernel-debugging.rst
> b/Documentation/dev-tools/gdb-kernel-debugging.rst
> > index 4756f6b3a04e..1586901b683c 100644
> > --- a/Documentation/dev-tools/gdb-kernel-debugging.rst
> > +++ b/Documentation/dev-tools/gdb-kernel-debugging.rst
> > @@ -114,7 +114,7 @@ Examples of using the Linux-provided gdb helpers
> >      [     0.000000] BIOS-e820: [mem 0x000000000009fc00-0x000000000009ffff]
> reserved
> >      ....
> >
> > -- Examine fields of the current task struct::
> > +- Examine fields of the current task struct(supported by x86 only)::
> >
> >      (gdb) p $lx_current().pid
> >      $1 = 4998
> > diff --git a/scripts/gdb/linux/cpus.py b/scripts/gdb/linux/cpus.py
> > index 008e62f3190d..f382762509d3 100644
> > --- a/scripts/gdb/linux/cpus.py
> > +++ b/scripts/gdb/linux/cpus.py
> > @@ -156,6 +156,13 @@ Note that VAR has to be quoted as string."""
> >
> >  PerCpu()
> >
> > +def get_current_task(cpu):
> > +    if utils.is_target_arch("x86"):
> > +         var_ptr = gdb.parse_and_eval("&current_task")
> > +         return per_cpu(var_ptr, cpu).dereference()
> > +    else:
> > +        raise gdb.GdbError("Sorry, obtaining the current task is not yet "
> > +                           "supported with this arch")
> 
> I've wondered in the past how we should handle the architecture specific
> layers.
> 
> Perhaps we need to have an interface of functionality to implement on
> each architecture so that we can create a per-arch set of helpers.
> 
> or break it up into arch specific subdirs at least...
> 
> 
> >  class LxCurrentFunc(gdb.Function):
> >      """Return current task.
> > @@ -167,8 +174,7 @@ number. If CPU is omitted, the CPU of the current context
> is used."""
> >          super(LxCurrentFunc, self).__init__("lx_current")
> >
> >      def invoke(self, cpu=-1):
> > -        var_ptr = gdb.parse_and_eval("&current_task")
> > -        return per_cpu(var_ptr, cpu).dereference()
> > +        return get_current_task(cpu)
> >
> 
> And then perhaps we simply shouldn't even expose commands which can not
> be supported on those architectures?

I feel it is better to tell users this function is not supported on its arch
than simply hiding the function.

If we hide it, users still have many chances to try it as they have got
information of lx_current from google or somewhere.
They will try, if it turns out the lx_current is not in the list and an
error like  "invalid data type for function to be called", they will
probably suspect their gdb/python environment is not set up correctly,
and continue to waste time in checking their environment. 
Finally they figure out this function is not supported by its arch so it is
not exposed. But they have wasted a couple of hours before knowing that.

It seems it is more friendly to directly tell users this is not supported
on its arch explicitly and clearly than reporting a "invalid data type
for function to be called.

> 
> Is it easy to disable this command if it's not supportable on the
> architecture?
> 

TBH, I'm not a python expert. I don't know how to do that in an elegant
way :-)  on the other hand, it seems lx_current isn’t a command like
lx-dmesg. Lx_current is actually similar with lx_per_cpu, we use gdb's
print(p) command to show its content.

> Presumably you are working on non-x86, have you investigated adding this
> support for your architecture (arm/arm64?)?

Yes. I've thought about it. But It would be quite trivial to bring up
this function on arm64.

arch/arm64/include/asm/current.h:
static __always_inline struct task_struct *get_current(void)
{
	unsigned long sp_el0;

	asm ("mrs %0, sp_el0" : "=r" (sp_el0));

	return (struct task_struct *)sp_el0;
}

We have to read a special register named sp_el0 and convert it to
task_struct while we are running in kernel mode. In gdb I can do
it by:
(gdb)p/x $SP_EL0
$20 = 0xffffffc011492400
(gdb)p ((struct task_struct *0xffffffc011492400))->pid
$21 = 0

What is more complex is that if we are running in user mode(EL0), this
register doesn't describe current task any more. so we have to
differentiate the modes of processor and make sure it only returns
current task while we are running in EL1(processor's kernel mode).

> 
> The fact you have run the command implies it would be useful for you ?
> 

Yes. I think it is a common requirement to get current task. lx_current
convenience function can help everyone. Since there is a document saying
this command exists, everyone using scripts/gdb would like to try it
I guess.

The simplest way would be adding current_task per_cpu variable for other
arch, but I believe hardly arch maintainers will accept it as its only
benefit is bringing up the lx_current. So 99.9% no maintainer wants it.

Thus, for the time being, I moved to just stop people from wasting time
like what I had done with a couple of hours.

> 
> >  LxCurrentFunc()
> >
> 
> 
> --
> Regards
> --
> Kieran

Thanks
Barry