Re: Recent aarch64 kprobes and uprobes patch systemtap testing

[Pratyush Anand <panand at redhat dot com>]

On 11/12/2015:03:59:53 PM, William Cohen wrote:
> On 12/10/2015 04:12 PM, David Long wrote:
> > On 12/10/2015 03:24 PM, William Cohen wrote:
> 
> >> The fslatency-nd and fsslower-nd tests need further investigation:
> >>
> >> PASS: ./systemtap.examples/lwtools/fslatency-nd build
> >> meta taglines 'test_installcheck: stap fslatency-nd.stp 1 1' tag 'test_installcheck' value 'stap fslatency-nd.stp 1 1'
> >> attempting command stap fslatency-nd.stp 1 1
> >> OUT ERROR: read fault [man error::fault] at 0x0000000000000034 (addr) near operator '@cast' at fslatency-nd.stp:66:15
> >> Tracing FS sync reads and writes... Output every 1 secs.
> >> WARNING: Number of errors: 1, skipped probes: 1
> >> WARNING: /root/systemtap_write/install/bin/staprun exited with status: 1
> >> Pass 5: run failed.  [man error::pass5]
> >> child process exited abnormally
> >> RC 1
> >> FAIL: ./systemtap.examples/lwtools/fslatency-nd run
> >>
> >> PASS: ./systemtap.examples/lwtools/fsslower-nd build
> >> meta taglines 'test_installcheck: stap fsslower-nd.stp -c "sleep 1"' tag 'test_installcheck' value 'stap fsslower-nd.stp -c "sleep 1"'
> >> attempting command stap fsslower-nd.stp -c "sleep 1"
> >> OUT ERROR: read fault [man error::fault] at 0x0000000000000034 (addr) near operator '@cast' at fsslower-nd.stp:68:15
> >> Tracing FS sync reads and writes slower than 10 ms... Hit Ctrl-C to end.
> >> TIME     PID    COMM             FUNC           SIZE     LAT(ms)
> >> WARNING: Number of errors: 1, skipped probes: 1
> >> WARNING: /root/systemtap_write/install/bin/staprun exited with status: 1
> >> Pass 5: run failed.  [man error::pass5]
> >> child process exited abnormally
> >> RC 1
> >> FAIL: ./systemtap.examples/lwtools/fsslower-nd run
> 
> > 
> > Cool. Wish I could make sense of systemtap error messages.
> > 
> >  At Will Deacon's suggested I tested probing the instruction in __copy_to_user that can cause a captured kernel exception when an application passes in a bad buffer address.  Unfortunately the result was a hang.  So copy_to/from user is going to have to be blacklisted for now, unless there turns out to be a simple fix. I'm worried there might be other places in the kernel where an otherwise probeable instruction might be expected to generate an exception.
> > 
> > -dl
> > 
> > 
> > 
> 
> Hi Dave and Pratyush,
> 
> I did some more experimentation with the fslatency-nd and fsslow-nd tests to see what is going on.  The problem seems to be related to the return probes.  I have a small reproducer attached which runs fine on x86_64 machine.  However on aarch64 it has the bogus read because some of the argument registers have changed value
> 
> # ../install/bin/stap ./aarch64_retkprobe_issue2.stp 
> ERROR: read fault [man error::fault] at 0x0000000000000034 (addr) near operator '@cast' at ./aarch64_retkprobe_issue2.stp:13:7
> pc : [<fffffe000021e37c>] lr : [<fffffe000021eb64>] pstate: 80000145
> sp : fffffe00bad7be30
> x29: fffffe00bad7be30 x28: fffffe00bad78000 
> x27: fffffe0000912000 x26: 000000000000003f 
> x25: 000000000000011d x24: 0000000000000015 
> x23: 0000000080000000 x22: 000003fff82b9760 
> x21: fffffe00bad7bec8 x20: 0000000000002004 
> x19: fffffe01b716e100 x18: 000003fff82b8160 
> x17: 000003ff849bf0a0 x16: fffffe000021f4a0 
> x15: 0000000000000004 x14: 000003fff82bb910 
> x13: 0000000000000001 x12: 000003ff7d75f200 
> x11: 00000000003d0f00 x10: 000003ff849b7af4 
> x9 : 0000000000000028 x8 : 0000000000000020 
> x7 : fffffe00bc5c3600 x6 : 0000000000000000 
> x5 : 0000000000000000 x4 : 0000000000000000 
> x3 : fffffe00bad7bec8 x2 : 0000000000002004 
> x1 : 000003fff82b9760 x0 : fffffe01b716e100 
> 
> pc : [<fffffe000021e37c>] lr : [<fffffe000009fbe0>] pstate: 60000145
> sp : fffffe00bad7be30
> x29: fffffe00bad7be30 x28: fffffe00bad78000 
> x27: fffffe0000912000 x26: 000000000000003f 
> x25: 000000000000011d x24: 0000000000000015 
> x23: 0000000080000000 x22: 000003fff82b9760 
> x21: fffffe00bad7bec8 x20: 0000000000002004 
> x19: fffffe01b716e100 x18: 000003fff82b8160 
> x17: 000003ff849bf0a0 x16: fffffe000021f4a0 
> x15: 0000000000000004 x14: 000003fff82bb910 
> x13: 0000000000000001 x12: 000003ff7d75f200 
> x11: 00000000003d0f00 x10: 000003ff849b7af4 
> x9 : 0000000000000028 x8 : 0000000000000020 
> x7 : fffffe00bc5c3600 x6 : 000003fff82b976c 
> x5 : 000003fff82b976c x4 : 0000000000000000 
> x3 : 0000000000000000 x2 : 0000000000000000 
> x1 : 0000000000000000 x0 : 000000000000000c 

Although I am not sure, but this is what it seems to me:

First argument (file) is in x0, and which is 0xC in case of kretprobe. But, can
x0 really be considered as 1st arg in case of kretprobe? 
I think, x0 should have return value of __vfs_read() in case of kretprobe. So,
0xC could be the number of bytes read.

With perf I see:

# perf probe -k vmlinux __vfs_read_exit=__vfs_read%return file
Semantic error :You can't specify local variable for kretprobe.

So, I am not sure what mechanism systemtap uses to get local variable in case of
kretprobe.

Moreover, on x86 I see that loop exits after the 1st print_regs() only. So it
means there was valid file->f_op->read() for the 1st file itself. If I comment
"kprobe.function("__vfs_read")", then there is no print at all. It means, we are
not hitting a case on x86 when callback was called for kretprobe and we had
nonzero 1st argument.

~Pratyush