Re: [perf] perf probe fails sometimes on 4.9

From: Pintu Kumar
Date: Tue Jan 30 2018 - 08:50:42 EST

Next message: Michal Hocko: "[PATCH] net/netfilter/x_tables.c: make allocation less aggressive"
Previous message: Corey Minyard: "Re: [PATCH arm/aspeed/ast2500 v2] ipmi: add an Aspeed KCS IPMI BMC driver"
In reply to: Masami Hiramatsu: "Re: [perf] perf probe fails sometimes on 4.9"
Next in thread: Masami Hiramatsu: "Re: [perf] perf probe fails sometimes on 4.9"
Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]

On Tue, Jan 30, 2018 at 11:13 AM, Masami Hiramatsu <mhiramat@xxxxxxxxxx> wrote:
>
> On Mon, 29 Jan 2018 22:00:52 +0530
> Pintu Kumar <pintu.ping@xxxxxxxxx> wrote:
>
> > Dear Masami,
> >
> > Thank you so much for your reply.
> > Please find some of my answers inline.
> >
> >
> > On Mon, Jan 29, 2018 at 7:47 PM, Masami Hiramatsu <mhiramat@xxxxxxxxxx> wrote:
> > > On Mon, 29 Jan 2018 13:40:34 +0530
> > > Pintu Kumar <pintu.ping@xxxxxxxxx> wrote:
> > >
> > >> Hi All,
> > >>
> > >> 'perf probe' is failing sometimes on 4.9.20 with AMD-64.
> > >> # perf probe --add schedule
> > >> schedule is out of .text, skip it.
> > >> Error: Failed to add events.
> > >>
> > >> If any one have come across this problem please let me know the cause.
> > >
> > > Hi Pintu,
> > >
> > > Could you run it with --vv?
> > >
> > Ok, I will send verbose output by tomorrow.
> >
> > >>
> > >> Note: I don't have CONFIG_DEBUG_INFO enabled in kernel. Is this the problem?
> > >
> > > Without it, you can not probe source-level probe nor trace local variable.
> > >
> >
> > Currently I am facing problem in enabling DEBUG_INFO in our kernel 4.9.20
> > However, I will try to manually include "-g" option during compilation.
> >
> > >> However, I manually copied the vmlinux file to /boot/ directory, but
> > >> still it does not work.
> > >
> > > That doesn't work.
> > > CONFIG_DEBUG_INFO option enables gcc to compile kernel with extra debuginfo.
> > > Without that option, debuginfo is not generated with vmlinux.
> > >
> > >>
> > >> I checked upstream patches until 4.15 but could not find any clue.
> > >> Please let me know if there is any fixes available for this.
> > >
> > > Could you also ensure that you run perf by root user?
> > >
> >
> > Yes I am running with root user.
> >
> > My concern is sometimes it works but sometimes it fails.
>
> What I thought was that your kernel enables kptr_strict(but maybe not.)
> Can you also try to find "schedule" and "_etext" functions in
> /proc/kallsyms?
>

Yes these functions are present in kallsyms.

# cat /proc/kallsyms | grep -w schedule ; cat /proc/kallsyms | grep -w _etext
ffffffff829328e0 T schedule
ffffffff8293e751 T _etext

> > I still needs to figure out, in which condition it works and which
> > condition it fails.
>
> Yeah, it is important.
>
> > Usually I noticed that in fresh reboot case it works.
>
> So, after a while, it doesn't work again? If so, it sounds like a daemon
> process changes settings in background.
>

Fortunately or unfortunately, now it is passing every time.
Still not sure about the cause.
If it fails again, I will collect info with -v and report again.
Any specific info that I need to check when it fails ?

Thank You,
Pintu

> Thank you,
>
> >
> >
> > > Thank you,
> > >
> > >
> > >>
> > >>
> > >> Thank You!
> > >> Regards,
> > >> Pintu
> > >>
> > >>
> > >> On Thu, Jan 25, 2018 at 7:09 PM, Pintu Kumar <pintu.ping@xxxxxxxxx> wrote:
> > >> > Hi,
> > >> >
> > >> > ** Changed the subject now, since these issues are related to general
> > >> > perf commands.
> > >> >
> > >> > Following are the issues:
> > >> >
> > >> > 1) perf probe --add schedule - FAILED
> > >> > output:
> > >> > schedule is out of .text, skip it.
> > >> > Error: Failed to add events.
> > >> >
> > >> > what is the issue here?
> > >> > Sometimes it pass and sometimes it fails...
> > >> > Similar is the case of 'perf inject' as well.
> > >> >
> > >> > 2) perf test - 1 FAILURE
> > >> > 37.1: Test basic BPF filtering : FAILED!
> > >> > 37.2: Test BPF prologue generation : Skip
> > >> > 37.3: Test BPF relocation checker : Skip
> > >> >
> > >> > bpf: config program 'func=SyS_epoll_wait'
> > >> > symbol:SyS_epoll_wait file:(null) line:0 offset:0 return:0 lazy:(null)
> > >> > bpf: config 'func=SyS_epoll_wait' is ok
> > >> > Looking at the vmlinux_path (8 entries long)
> > >> > Using /boot/vmlinux for symbols
> > >> > Could not open debuginfo. Try to use symbols.
> > >> > SyS_epoll_wait is out of .text, skip it.
> > >> > bpf_probe: failed to convert perf probe eventsFailed to add events
> > >> > selected by BPF
> > >> > test child finished with -1
> > >> > ---- end ----
> > >> > Test BPF filter subtest 0: FAILED!
> > >> >
> > >> > Looks like both 1,2 are related.
> > >> > Since, CONFIG_DEBUG_INFO is not enabled, I manually copied the vmlinux
> > >> > to /boot/ folder.
> > >> >
> > >> > -------
> > >> > Some more info:
> > >> >
> > >> > Kernel build dir is set to /lib/modules/4.9.20-sc-amd-x86-64/build
> > >> > set env: KBUILD_DIR=/lib/modules/4.9.20-sc-amd-x86-64/build
> > >> > unset env: KBUILD_OPTS
> > >> > include option is set to -nostdinc -isystem
> > >> > /usr/lib/gcc/x86_64-linux-gnu/5/include -I./arch/x86/include
> > >> > -I./arch/x86/include/generated/uapi -I./arch/x86/include/generated
> > >> > -I./include -I./arch/x86/include/uapi -I./include/uapi
> > >> > -I./include/generated/uapi -include ./include/linux/kconfig.h
> > >> > set env: NR_CPUS=8
> > >> > set env: LINUX_VERSION_CODE=0x40914
> > >> > set env: CLANG_EXEC=/usr/bin/clang
> > >> > set env: CLANG_OPTIONS=-xc
> > >> > set env: KERNEL_INC_OPTIONS= -nostdinc -isystem
> > >> > /usr/lib/gcc/x86_64-linux-gnu/5/include -I./arch/x86/include
> > >> > -I./arch/x86/include/generated/uapi -I./arch/x86/include/generated
> > >> > -I./include -I./arch/x86/include/uapi -I./include/uapi
> > >> > -I./include/generated/uapi -include ./include/linux/kconfig.h
> > >> > set env: WORKING_DIR=/lib/modules/4.9.20-sc-amd-x86-64/build
> > >> >
> > >> >
> > >> > If you have any clue about these failure please help me.
> > >> >
> > >> >
> > >> > Thanks,
> > >> > Pintu
> > >> >
> > >> >
> > >> > On Wed, Jan 24, 2018 at 8:23 PM, Pintu Kumar <pintu.ping@xxxxxxxxx> wrote:
> > >> >> Hi,
> > >> >>
> > >> >> Thanks for your help.
> > >> >> Yes it was a sub version issue.
> > >> >>
> > >> >> Earlier, while building the kernel I left the SUBLEVEL as blank.
> > >> >>
> > >> >> make -j8 bindeb-pkg SUBLEVEL=20
> > >> >>
> > >> >> After passing the correct sublevel now the test is working.
> > >> >>
> > >> >> But still following are failing:
> > >> >>
> > >> >> 16: Try 'import perf' in python, checking link problems : FAILED!
> > >> >> 37.2: Test BPF prologue generation : FAILED!
> > >> >>
> > >> >>
> > >> >> This is the error I get:
> > >> >>
> > >> >> 16: Try 'import perf' in python, checking link problems :
> > >> >> --- start ---
> > >> >> test child forked, pid 7637
> > >> >> Traceback (most recent call last):
> > >> >> File "<stdin>", line 1, in <module>
> > >> >> ImportError: No module named perf
> > >> >> test child finished with -1
> > >> >> ---- end ----
> > >> >> Try 'import perf' in python, checking link problems: FAILED!
> > >> >>
> > >> >>
> > >> >> Looking at the vmlinux_path (8 entries long)
> > >> >> symsrc__init: cannot get elf header.
> > >> >> Failed to find the path for kernel: Invalid ELF file
> > >> >> bpf_probe: failed to convert perf probe eventsFailed to add events
> > >> >> selected by BPF
> > >> >> test child finished with -1
> > >> >> ---- end ----
> > >> >> Test BPF filter subtest 1: FAILED!
> > >> >>
> > >> >>
> > >> >>
> > >> >> Thanks,
> > >> >> Pintu
> > >> >>
> > >> >>
> > >> >> On Wed, Jan 24, 2018 at 6:39 AM, Wangnan (F) <wangnan0@xxxxxxxxxx> wrote:
> > >> >>>
> > >> >>>
> > >> >>> On 2018/1/23 20:37, Pintu Kumar wrote:
> > >> >>>>
> > >> >>>> Hi All,
> > >> >>>>
> > >> >>>> I am verifying all perf tests on Ubuntu-16 x86-64 platform using the
> > >> >>>> kernel version 4.9.20.
> > >> >>>> I have installed several others packages including: clang, llvm
> > >> >>>>
> > >> >>>> But, when I run 'perf test' I get some FAILURE.
> > >> >>>> Specially, 'perf test LLVM' is failing.
> > >> >>>>
> > >> >>>> Please check the below error logs:
> > >> >>>>
> > >> >>>> ----------------------------------------------------
> > >> >>>> # perf test LLVM
> > >> >>>> 35: Test LLVM searching and compiling :
> > >> >>>> 35.1: Basic BPF llvm compiling test : FAILED!
> > >> >>>> 35.2: Test kbuild searching : Skip
> > >> >>>> 35.3: Compile source for BPF prologue generation test : Skip
> > >> >>>> 35.4: Compile source for BPF relocation test : Skip
> > >> >>>> ----------------------------------------------------
> > >> >>>>
> > >> >>>> When I run with -v I get this:
> > >> >>>>
> > >> >>>> -----------------------------------------------------
> > >> >>>> # perf test -v LLVM
> > >> >>>> 35: Test LLVM searching and compiling :
> > >> >>>> 35.1: Basic BPF llvm compiling test :
> > >> >>>> --- start ---
> > >> >>>> test child forked, pid 3304
> > >> >>>> Unablt to get kernel version from uname '4.9-xxxx-amd-x86-64'
> > >> >>>> WARNING: unable to get correct kernel building directory.
> > >> >>>> Hint: Set correct kbuild directory using 'kbuild-dir' option in [llvm]
> > >> >>>> section of ~/.perfconfig or set it to "" to suppress kbuild
> > >> >>>> detection.
> > >> >>>>
> > >> >>>> Unablt to get kernel version from uname '4.9-xxxx-amd-x86-64'
> > >> >>>>
> > >> >>>>
> > >> >>>> int _version SEC("version") = LINUX_VERSION_CODE;
> > >> >>>> ' | $CLANG_EXEC -D__KERNEL__ -D__NR_CPUS__=$NR_CPUS
> > >> >>>> -DLINUX_VERSION_CODE=$LINUX_VERSION_CODE $CLANG_OPTIONS
> > >> >>>> $KERNEL_INC_OPTIONS -Wno-unused-value -Wno-pointer-sign
> > >> >>>> -working-directory $WORKING_DIR -c "$CLANG_SOURCE" -target bpf -O2 -o
> > >> >>>> -
> > >> >>>> libbpf: 2129190-4a0 doesn't provide kernel version
> > >> >>>> Failed to parse test case 'Basic BPF llvm compiling test'
> > >> >>>> test child finished with -1
> > >> >>>> ---- end ----
> > >> >>>> Test LLVM searching and compiling subtest 0: FAILED!
> > >> >>>> -----------------------------------------------------
> > >> >>>>
> > >> >>>>
> > >> >>>> Basic BPF LLVM compiling test is failing.
> > >> >>>> It reports that bpf could not able to read the kernel version from uname.
> > >> >>>> Is it because of this that 'perf test LLVM' is failing ?
> > >> >>>>
> > >> >>>> My uname says:
> > >> >>>> 4.9-xxxx-amd-x86-64
> > >> >>>
> > >> >>>
> > >> >>> 'perf test LLVM' requires a well formed uname string (%d.%d.%d). Please see
> > >> >>> tools/perf/fetch_kernel_version . If your distro uses another method to
> > >> >>> report
> > >> >>> kernel version number (from your uname string, sublevel can't be
> > >> >>> determined),
> > >> >>> please help us improve that function like what we have done for ubuntu
> > >> >>> (checking /proc/version_signature).
> > >> >>>
> > >> >>>
> > >> >>>
> > >> >>>> My kernel version is: 4.9.20 (from Makefile)
> > >> >>>>
> > >> >>>> So, I dont think there is any problem with kernel version.
> > >> >>>>
> > >> >>>> If any body have come across this problem please let me know how to
> > >> >>>> resolve this issue.
> > >> >>>>
> > >> >>>>
> > >> >>>>
> > >> >>>> Thank You!
> > >> >>>>
> > >> >>>> Regards,
> > >> >>>> Pintu
> > >> >>>
> > >> >>>
> > >> >>>
> > >
> > >
> > > --
> > > Masami Hiramatsu <mhiramat@xxxxxxxxxx>
>
>
> --
> Masami Hiramatsu <mhiramat@xxxxxxxxxx>

Next message: Michal Hocko: "[PATCH] net/netfilter/x_tables.c: make allocation less aggressive"
Previous message: Corey Minyard: "Re: [PATCH arm/aspeed/ast2500 v2] ipmi: add an Aspeed KCS IPMI BMC driver"
In reply to: Masami Hiramatsu: "Re: [perf] perf probe fails sometimes on 4.9"
Next in thread: Masami Hiramatsu: "Re: [perf] perf probe fails sometimes on 4.9"
Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]