Re: [PATCH bpf-next v9 7/8] bpf: lsm: Add selftests for BPF_PROG_TYPE_LSM

From: KP Singh <kpsingh@chromium.org>
To: Alexei Starovoitov <alexei.starovoitov@gmail.com>
Cc: KP Singh <kpsingh@chromium.org>, bpf <bpf@vger.kernel.org>,
	Andrii Nakryiko <andriin@fb.com>,
	Brendan Jackman <jackmanb@google.com>,
	Florent Revest <revest@google.com>,
	Thomas Garnier <thgarnie@google.com>,
	Daniel Borkmann <daniel@iogearbox.net>,
	Kees Cook <keescook@chromium.org>, Paul Turner <pjt@google.com>,
	Jann Horn <jannh@google.com>,
	Florent Revest <revest@chromium.org>,
	Brendan Jackman <jackmanb@chromium.org>
Subject: Re: [PATCH bpf-next v9 7/8] bpf: lsm: Add selftests for BPF_PROG_TYPE_LSM
Date: Thu, 2 Apr 2020 06:03:57 +0200	[thread overview]
Message-ID: <20200402040357.GA217889@google.com> (raw)
In-Reply-To: <CAADnVQKP3mOTUkkzjWM6Qii+v-dCDwV9Ms_-4ptsbdwyDW1MCA@mail.gmail.com>

On 01-Apr 17:09, Alexei Starovoitov wrote:
> On Sat, Mar 28, 2020 at 5:44 PM KP Singh <kpsingh@chromium.org> wrote:
> > +int BPF_PROG(test_int_hook, struct vm_area_struct *vma,
> > +            unsigned long reqprot, unsigned long prot, int ret)
> > +{
> > +       if (ret != 0)
> > +               return ret;
> > +
> > +       __u32 pid = bpf_get_current_pid_tgid() >> 32;
> > +       int is_heap = 0;
> > +
> > +       is_heap = (vma->vm_start >= vma->vm_mm->start_brk &&
> > +                  vma->vm_end <= vma->vm_mm->brk);
> 
> This test fails for me.

Trying this from bpf/master:

  b9258a2cece4 ("slcan: Don't transmit uninitialized stack data in padding")

also from bpf-next/master:

 1a323ea5356e ("x86: get rid of 'errret' argument to __get_user_xyz() macross")

and I am unable to reproduce the failure (the output when using bpf/master):

 ./test_progs -t test_lsm
#70 test_lsm:OK
Summary: 1/0 PASSED, 0 SKIPPED, 0 FAILED

cat /sys/kernel/debug/tracing/trace
# tracer: nop
#
# entries-in-buffer/entries-written: 10/10   #P:4
#
#                              _-----=> irqs-off
#                             / _----=> need-resched
#                            | / _---=> hardirq/softirq
#                            || / _--=> preempt-depth
#                            ||| /     delay
#           TASK-PID   CPU#  ||||    TIMESTAMP  FUNCTION
#              | |       |   ||||       |         |
            true-322   [001] ...2   187.127231: 0: start 7fc7ffccc000 556a623be000
            true-322   [001] ...2   187.127238: 0: end 7fc7ffe8c000 556a623be000
            true-322   [001] ...2   187.128233: 0: start 7fc7ffe82000 556a623be000
            true-322   [001] ...2   187.128237: 0: end 7fc7ffe88000 556a623be000
            true-322   [001] ...2   187.128306: 0: start 556a604f7000 556a623be000
            true-322   [001] ...2   187.128309: 0: end 556a604f9000 556a623be000
            true-322   [001] ...2   187.128372: 0: start 7fc7ffebf000 556a623be000
            true-322   [001] ...2   187.128375: 0: end 7fc7ffec1000 556a623be000
      test_progs-321   [000] ...2   187.129952: 0: start 55dc6e8df000 55dc6e8df000
      test_progs-321   [000] ...2   187.129955: 0: end 55dc6e906000 55dc6e906000

The full run also works for me:

./test_progs

[...]

#70 test_lsm:OK
#71 test_overhead:OK
#72 tp_attach_query:OK

My config is:

  https://gist.githubusercontent.com/sinkap/cb24b955e1b6e6c1dc736054a774fb41/raw/

and the kernel commandline is (on a QEMU VM):

cat /proc/cmdline
console=ttyS0,115200 root=/dev/sda rw nokaslr

Could you share your config and cmdline?

Also, I am wondering if this happens just in the BPF program or also
in the kernel as the other variable I can think of is the compiled
bpf program itself which might be reading a different value thinking
it's vm->vma_start, possible something to do with BTF / CO RE due to a
compiler bug:

Here's the version of clang I am using:

  clang version 10.0.0 (https://github.com/llvm/llvm-project.git 2026d7b80a1a5534b5e263683c85aa95e7593b98)

- KP

> I've added:
>         bpf_printk("start %llx %llx\n", vma->vm_start, vma->vm_mm->start_brk);
>         bpf_printk("end %llx %llx\n", vma->vm_end, vma->vm_mm->brk);
> and see
> cat /sys/kernel/debug/tracing/trace_pipe
>             true-2285  [001] ...2   858.717432: 0: start 7f66470a2000 607000
>             true-2285  [001] ...2   858.717440: 0: end 7f6647443000 607000
>             true-2285  [001] ...2   858.717658: 0: start 7f6647439000 607000
>             true-2285  [001] ...2   858.717659: 0: end 7f664743f000 607000
>             true-2285  [001] ...2   858.717691: 0: start 605000 607000
>             true-2285  [001] ...2   858.717692: 0: end 607000 607000
>             true-2285  [001] ...2   858.717700: 0: start 7f6647666000 607000
>             true-2285  [001] ...2   858.717701: 0: end 7f6647668000 607000
>       test_progs-2283  [000] ...2   858.718030: 0: start 523000 39b9000
>       test_progs-2283  [000] ...2   858.718033: 0: end 39e0000 39e0000
> 
> 523000 is not >= 39b9000.
> 523000 is higher than vm_mm->end_data, but lower than vm_mm->start_brk.
> No idea why this addr is passed into security_file_mprotect().
> The address user space is passing to mprotect() is 0x39c0000 which is correct.
> Could you please help debug?