On Wed, Sep 03, 2025 at 09:34:38AM +0800, Oliver Sang wrote:
> hi, Paul,
>
> On Fri, Aug 29, 2025 at 08:07:41AM -0700, Paul E. McKenney wrote:
> > On Thu, Aug 21, 2025 at 01:26:45PM +0800, kernel test robot wrote:
> > >
> > > hi, Paul,
> > >
> > > we also noticed there is similar commit in newer branch
> > > dev.2025.08.14a
> > > dev.2025.08.19a
> > > but we didn't finish any bisect for them so far.
> > >
> > > if the issue is already known and fixed in newer version, please just
> > > ignore
> > > this report. sorry if any inconvenience.
> > >
> > > below full report FYI.
> > >
> > >
> > > Hello,
> > >
> > > kernel test robot noticed
> > > "WARNING:at_arch/x86/mm/fault.c:#do_user_addr_fault" on:
> > >
> > > commit: 364ac25d46eea504eb90229d2a1f92e18c1a1eae ("tracing: Guard
> > > __DECLARE_TRACE() use of __DO_TRACE_CALL() with SRCU-fast")
> > > https://git.kernel.org/cgit/linux/kernel/git/paulmck/linux-rcu.git
> > > dev.2025.08.13a
> > >
> > > in testcase: boot
> > >
> > > config: i386-randconfig-004-20250819
> > > compiler: clang-20
> > > test machine: qemu-system-i386 -enable-kvm -cpu SandyBridge -smp 2 -m 4G
> > >
> > > (please refer to attached dmesg/kmsg for entire log/backtrace)
> >
> > Thank you for your testing efforts, and apologies for being slow!
> >
> > Could you please try the diagnostic patch at the end of this email?
> >
> > Thanx, Paul
>
> by applying the patch, the issue gone. but since you said this is a
> 'diagnostic
> patch', not sure if it's a real fix. anyway:
>
> Tested-by: kernel test robot <oliver.s...@intel.com>
Thank you very much! This tells me that something on the code path from
the tracepoint to the BPF program needs to have preemption disabled.
I will leave the diagnostic patch in my tree, and will be looking into
what the real fix should be.
Thanx, Paul
> [...]
>
> > > If you fix the issue in a separate patch/commit (i.e. not just a new
> > > version of
> > > the same patch/commit), kindly add following tags
> > > | Reported-by: kernel test robot <oliver.s...@intel.com>
> > > | Closes:
> > > https://lore.kernel.org/oe-lkp/202508211038.c93e8603-...@intel.com
>
> [...]
>
> > > The kernel config and materials to reproduce are available at:
> > > https://download.01.org/0day-ci/archive/20250821/202508211038.c93e8603-...@intel.com
> > > --
> > > 0-DAY CI Kernel Test Service
> > > https://github.com/intel/lkp-tests/wiki
> >
> >
> > commit 2d6142ce44dca77fb173bb96850634b169277214
> > Author: Paul E. McKenney <paul...@kernel.org>
> > Date: Thu Aug 28 12:56:42 2025 -0700
> >
> > EXP tracing: Diagnostic for __DECLARE_TRACE() use of SRCU-fast
> >
> > This patch is intended to test the theory that preemption needs to be
> > disabled in some portion of the tracing infrastructure extending from
> > the __DECLARE_TRACE() macro to the target BPF program.
> >
> > Signed-off-by: Paul E. McKenney <paul...@kernel.org>
> >
> > diff --git a/include/linux/tracepoint.h b/include/linux/tracepoint.h
> > index a22c1ab88560b8..c422e4c5ed51ed 100644
> > --- a/include/linux/tracepoint.h
> > +++ b/include/linux/tracepoint.h
> > @@ -273,7 +273,7 @@ static inline struct tracepoint
> > *tracepoint_ptr_deref(tracepoint_ptr_t *p)
> > static inline void __do_trace_##name(proto) \
> > { \
> > if (cond) { \
> > - guard(srcu_fast_notrace)(&tracepoint_srcu); \
> > + guard(preempt_notrace)(); \
> > __DO_TRACE_CALL(name, TP_ARGS(args)); \
> > } \
> > } \
