Re: MIPS: ath79: regression after patch: sched/fair: do not expose some tracepoints to user if CONFIG_SCHEDSTATS is not set

[Yafang Shao <laoar.shao@gmail.com>]

On Mon, Jun 17, 2019 at 1:39 PM Oleksij Rempel <o.rempel@pengutronix.de> wrote:
>
> On Mon, Jun 17, 2019 at 10:04:46AM +0800, Yafang Shao wrote:
> > On Fri, Jun 14, 2019 at 4:34 PM Oleksij Rempel <o.rempel@pengutronix.de> wrote:
> > >
> > > On Fri, Jun 14, 2019 at 01:56:04PM +0800, Yafang Shao wrote:
> > > > On Fri, Jun 14, 2019 at 12:02 AM Oleksij Rempel <o.rempel@pengutronix.de> wrote:
> > > > >
> > > > > On Thu, Jun 13, 2019 at 04:12:57PM +0800, Yafang Shao wrote:
> > > > > > On Thu, Jun 13, 2019 at 4:08 PM Yafang Shao <laoar.shao@gmail.com> wrote:
> > > > > > >
> > > > > > > On Thu, Jun 13, 2019 at 2:30 PM Oleksij Rempel <o.rempel@pengutronix.de> wrote:
> > > > > > > >
> > > > > > > > Hi,
> > > > > > > >
> > > > > > > > After patch:
> > > > > > > > -----------------------------------------------------
> > > > > > > > commit 2a09b5de235a6b5f76193a2ed46546a2944f98bf
> > > > > > > > Author: Yafang Shao <laoar.shao@gmail.com>
> > > > > > > > Date:   Tue Mar 26 20:13:10 2019 +0800
> > > > > > > >     sched/fair: do not expose some tracepoints to user if CONFIG_SCHEDSTATS is not set
> > > > > > > >
> > > > > > > >     The tracepoints trace_sched_stat_{iowait, blocked, wait, sleep} should
> > > > > > > >     be not exposed to user if CONFIG_SCHEDSTATS is not set.
> > > > > > > >
> > > > > > > >     Link: http://lkml.kernel.org/r/1553602391-11926-3-git-send-email-laoar.shao@gmail.com
> > > > > > > >
> > > > > > > >     Acked-by: Peter Zijlstra (Intel) <peterz@infradead.org>
> > > > > > > >     Signed-off-by: Yafang Shao <laoar.shao@gmail.com>
> > > > > > > >     Signed-off-by: Steven Rostedt (VMware) <rostedt@goodmis.org>
> > > > > > > > -----------------------------------------------------
> > > > > > > >
> > > > > > > > i can't boot kernel on MIPS based system Atheros ar9331. The boot
> > > > > > > > process ends with following oops:
> > > > > > > > -----------------------------------------------------
> > > > > > > > [    0.000000] Built 1 zonelists, mobility grouping on.  Total pages: 16256                                                                                                [0/1518]
> > > > > > > > [    0.000000] Kernel command line:   ip=dhcp root=/dev/nfs nfsroot=192.168.23.4:/home/ore/nfsroot/dpt-module,v3,tcp
> > > > > > > > [    0.000000] Dentry cache hash table entries: 8192 (order: 3, 32768 bytes)
> > > > > > > > [    0.000000] Inode-cache hash table entries: 4096 (order: 2, 16384 bytes)
> > > > > > > > [    0.000000] Writing ErrCtl register=00000000
> > > > > > > > [    0.000000] Readback ErrCtl register=00000000
> > > > > > > > [    0.000000] Memory: 56376K/65536K available (5376K kernel code, 437K rwdata, 1068K rodata, 1432K init, 214K bss, 9160K reserved, 0K cma-reserved)
> > > > > > > > [    0.000000] random: get_random_u32 called from cache_random_seq_create+0x9c/0x170 with crng_init=0
> > > > > > > > [    0.000000] SLUB: HWalign=32, Order=0-3, MinObjects=0, CPUs=1, Nodes=1
> > > > > > > > [    0.000000] ftrace: allocating 18167 entries in 36 pages
> > > > > > > > [    0.000000] CPU 0 Unable to handle kernel paging request at virtual address 00000000, epc == 00000000, ra == 00000000
> > > > > > > > [    0.000000] Oops[#1]:
> > > > > > > > [    0.000000] CPU: 0 PID: 0 Comm: swapper Not tainted 5.2.0-rc4+ #308
> > > > > > > > [    0.000000] $ 0   : 00000000 8007e5fc 00000003 00000002
> > > > > > > > [    0.000000] $ 4   : 00000001 00000000 00000000 83c22470
> > > > > > > > [    0.000000] $ 8   : 00000000 00000000 00000000 00000000
> > > > > > > > [    0.000000] $12   : 806afe04 00000000 00000001 00000000
> > > > > > > > [    0.000000] $16   : 800adc0c 800adc00 00000000 00000000
> > > > > > > > [    0.000000] $20   : 00000000 0000048e 00000000 806c0000
> > > > > > > > [    0.000000] $24   : 00000003 800619e4
> > > > > > > > [    0.000000] $28   : 806ae000 806afe10 808b0000 00000000
> > > > > > > > [    0.000000] Hi    : 00000000
> > > > > > > > [    0.000000] Lo    : 00000000
> > > > > > > > [    0.000000] epc   : 00000000 0x0
> > > > > > > > [    0.000000] ra    : 00000000 0x0
> > > > > > > > [    0.000000] Status: 10000002 KERNEL EXL
> > > > > > > > [    0.000000] Cause : 10800008 (ExcCode 02)
> > > > > > > > [    0.000000] BadVA : 00000000
> > > > > > > > [    0.000000] PrId  : 00019374 (MIPS 24Kc)
> > > > > > > > [    0.000000] Modules linked in:
> > > > > > > > [    0.000000] Process swapper (pid: 0, threadinfo=(ptrval), task=(ptrval), tls=00000000)
> > > > > > > > [    0.000000] Stack : 00000002 8007e784 806ca284 00000000 800adc0c 00000000 80070eb0 8007108c
> > > > > > > > [    0.000000]         83c03c00 801ab7d8 83c03b80 00000000 00000000 83c03b80 00000000 801109cc
> > > > > > > > [    0.000000]         00000001 80880000 80880000 800c1318 00000000 ffffffff 806b0000 806b0000
> > > > > > > > [    0.000000]         83c22470 0000048e 8075eb00 807706dc 806b0000 80753158 806b0000 80755209
> > > > > > > > [    0.000000]         00000001 80880000 80880000 80726cc8 ffffffff 000046f7 00000024 00000003
> > > > > > > > [    0.000000]         ...
> > > > > > > > [    0.000000] Call Trace:
> > > > > > > > [    0.000000]
> > > > > > > > [    0.000000] [<8007e784>] r4k_flush_icache_range+0x10/0x18
> > > > > > > > [    0.000000] [<800adc0c>] preempt_count_sub+0x10/0xe4
> > > > > > > > [    0.000000] [<80070eb0>] ftrace_caller+0x0/0x54
> > > > > > > > [    0.000000] [<8007108c>] ftrace_make_nop+0x8c/0x94
> > > > > > > > [    0.000000] [<801ab7d8>] __get_free_pages+0x2c/0x60
> > > > > > > > [    0.000000] [<801109cc>] ftrace_process_locs+0x364/0x3ec
> > > > > > > > [    0.000000] [<800c1318>] vprintk_default+0x34/0x40
> > > > > > > > [    0.000000] [<80726cc8>] ftrace_init+0xb8/0x138
> > > > > > > > [    0.000000] [<8071ac98>] start_kernel+0x290/0x584
> > > > > > > > [    0.000000] [<8071a368>] unknown_bootoption+0x0/0x270
> > > > > > > > [    0.000000]
> > > > > > > > [    0.000000] Code: (Bad address in epc)
> > > > > > > > [    0.000000]
> > > > > > > > [    0.000000] ---[ end trace 64e6fa591c4a277c ]---
> > > > > > > > [    0.000000] Kernel panic - not syncing: Fatal exception
> > > > > > > > [    0.000000] Rebooting in
> > > > > > > > -----------------------------------------------------
> > > > > > > >
> > > > > > > > After reverting this patch I can normally boot the kernel.
> > > > > > > > The kernel is build with CONFIG_SCHEDSTATS not set.
> > > > > > > >
> > > > > > > > --
> > > > > > > > Pengutronix e.K.                           |                             |
> > > > > > > > Industrial Linux Solutions                 | http://www.pengutronix.de/  |
> > > > > > > > Peiner Str. 6-8, 31137 Hildesheim, Germany | Phone: +49-5121-206917-0    |
> > > > > > > > Amtsgericht Hildesheim, HRA 2686           | Fax:   +49-5121-206917-5555 |
> > > > > > >
> > > > > > > Hi Oleksij,
> > > > > > >
> > > > > > > This really makes me confused.
> > > > > > > If CONFIG_SCHEDSTATS is not set,  trace_sched_stat_{iowait, blocked,
> > > > > > > wait, sleep} will be optimized out by
> > > > > > > the compiler. That should not effect other codes.
> > > > > > >
> > > > > > > Would you pls. try bellow command on your build server?
> > > > > > > $ objdump -dr kernel/sched/fair.o | awk '/>:$/ { F=$2 } /sched_stat/ {
> > > > > > > print F " " $0 }'
> > > > > > >
> > > > > >
> > > > > > Plus, could you pls. show me the  difference on the objdump of these
> > > > > > two vmlinux (with and without my patch) ?
> > > > >
> > > > > Both files are in attachment.
> > > > >
> > > >
> > > > Hi Oleksij,
> > > >
> > > > Unfortunately I don't have a mips toolchain on hand.
> > > > It would be better if you could give me the  disassembled code of
> > > > these two vmlinuxs.
> > > > I will try to build a cross toolchain for MIPS 24Kc, that may take some time.
> > >
> > > disassembled code is in attachment.
> > >
> > > --
> >
> > + Steve,
> >
> > Hi Oleksij,
> >
> > I guess this issue is related with __mcount_loc, but I don't have a
> > clear idea how to fix it yet.
> > It would be better if you could share your .config file as well.
> >
> > Hi Steve,
> >
> > Do you have any suggestion on this issue?
> > My guess is when we define a tracepoint as nop, there's something
> > wrong in __mcount_loc,
> >  which will be used by MIPS and then this oops occurs.
>
> kernel config is in attachment.
>

Hi Oleksij,

What is the output of "ftrace: allocating %ld entries in %ld pages"
after revert this patch ?
I want to know whether this issue is related with the number of ftrace entries,
e.g, whehter more pages are allocated after this patch is reverted.

In this oops messages, the output is "ftrace: allocating 18167 entries
in 36 pages".

Thanks
Yafang