Re: [PATCH] tracing: Fix complicated dependency of

From: Google
Date: Tue Dec 06 2022 - 07:56:38 EST

Next message: Abel Vesa: "[PATCH v5 0/5] clk: qcom: Add support for SM8550"
Previous message: Finn Behrens: "Re: [PATCH v2 13/28] rust: alloc: add `Vec::try_with_capacity{,_in}()` constructors"
In reply to: kernel test robot: "Re: [PATCH] tracing: Fix complicated dependency of"
Next in thread: David Howells: "Re: [PATCH] trace: Fix some checker warnings"
Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]

On Mon, 5 Dec 2022 23:10:19 +0800
kernel test robot <lkp@xxxxxxxxx> wrote:

> Hi Masami,
>
> Thank you for the patch! Yet something to improve:
>
> [auto build test ERROR on rostedt-trace/for-next]
> [also build test ERROR on linus/master]
> [cannot apply to rostedt-trace/for-next-urgent]
> [If your patch is applied to the wrong git tree, kindly drop us a note.
> And when submitting patch, we suggest to use '--base' as documented in
> https://git-scm.com/docs/git-format-patch#_base_tree_information]
>
> url: https://github.com/intel-lab-lkp/linux/commits/Masami-Hiramatsu/tracing-Fix-complicated-dependency-of/20221205-170923
> base: https://git.kernel.org/pub/scm/linux/kernel/git/rostedt/linux-trace.git for-next
> patch link: https://lore.kernel.org/r/20221205180617.9b9d3971cbe06ee536603523%40kernel.org
> patch subject: [PATCH] tracing: Fix complicated dependency of
> config: arm64-randconfig-r011-20221205
> compiler: aarch64-linux-gcc (GCC) 12.1.0
> reproduce (this is a W=1 build):
> wget https://raw.githubusercontent.com/intel/lkp-tests/master/sbin/make.cross -O ~/bin/make.cross
> chmod +x ~/bin/make.cross
> # https://github.com/intel-lab-lkp/linux/commit/32c12975c448f06f79a6797d4793d9e60b3040df
> git remote add linux-review https://github.com/intel-lab-lkp/linux
> git fetch --no-tags linux-review Masami-Hiramatsu/tracing-Fix-complicated-dependency-of/20221205-170923
> git checkout 32c12975c448f06f79a6797d4793d9e60b3040df
> # save the config file
> mkdir build_dir && cp config build_dir/.config
> COMPILER_INSTALL_PATH=$HOME/0day COMPILER=gcc-12.1.0 make.cross W=1 O=build_dir ARCH=arm64 SHELL=/bin/bash kernel/
>
> If you fix the issue, kindly add following tag where applicable
> | Reported-by: kernel test robot <lkp@xxxxxxxxx>
>
> All errors (new ones prefixed by >>):
>
> kernel/trace/trace.c: In function 'update_max_tr':
> kernel/trace/trace.c:1842:2: warning: label 'out_unlock' defined but not used [-Wunused-label]
> 1842 | out_unlock:
> | ^~~~~~~~~~

OK, this seems clearly an issue in this function with/without this patch.
But this patch kicks the issue. Why? since the original code is under
a complicated #ifdef hell.

As I pointed, CONFIG_OSNOISE_TRACER and CONFIG_HWLAT_TRACER will enable
a part of the CONFIG_TRACER_MAX_TRACE code. And the update_max_tr() is
NOT included that part.
The rest of the CONFIG_TRACER_MAX_TRACE code is enabled either one of
CONFIG_IRQSOFF_TRACER, CONFIG_PREEMPT_TRACER or CONFIG_SCHED_TRACER.
And those 3 kconfigs also selects CONFIG_TRACER_SNAPSHOT.

In the result, when CONFIG_OSNOISE_TRACER or CONFIG_HWLAT_TRACER is
enabled but all of CONFIG_IRQSOFF_TRACER, CONFIG_PREEMPT_TRACER and
CONFIG_SCHED_TRACER are disabled, the update_max_tr() is NOT compiled.
When one of CONFIG_IRQSOFF_TRACER, CONFIG_PREEMPT_TRACER or
CONFIG_SCHED_TRACER is enabled, update_max_tr() is compiled with
CONFIG_TRACER_SNAPSHOT=y.

For simplifying the dependency, I made the CONFIG_OSNOISE_TRACER and
CONFIG_HWLAT_TRACER selected CONFIG_TRACER_MAX_TRACE, but didn't select
CONFIG_TRACER_SNAPSHOT. Thus this error happens.

> kernel/trace/trace.c: In function 'trace_check_vprintf':
> kernel/trace/trace.c:3860:17: warning: function 'trace_check_vprintf' might be a candidate for 'gnu_printf' format attribute [-Wsuggest-attribute=format]
> 3860 | trace_seq_vprintf(&iter->seq, iter->fmt, ap);
> | ^~~~~~~~~~~~~~~~~
> kernel/trace/trace.c:3927:17: warning: function 'trace_check_vprintf' might be a candidate for 'gnu_printf' format attribute [-Wsuggest-attribute=format]
> 3927 | trace_seq_vprintf(&iter->seq, p, ap);
> | ^~~~~~~~~~~~~~~~~

These should be fixed by David's printf patch.

> kernel/trace/trace.c: In function 'tracing_set_tracer':
> >> kernel/trace/trace.c:6445:17: error: implicit declaration of function 'free_snapshot'; did you mean 'boot_snapshot'? [-Werror=implicit-function-declaration]
> 6445 | free_snapshot(tr);
> | ^~~~~~~~~~~~~
> | boot_snapshot
> cc1: some warnings being treated as errors

This is the same issue of the CONFIG_TRACER_MAX_TRACE and
CONFIG_TRACER_SNAPSHOT. I will fix it by adding #ifdef
CONFIG_TRACER_SNAPSHOT this time. But I'm not sure the
CONFIG_OSNOISE_TRACER and CONFIG_HWLAT_TRACER really don't need the
snapshot feature?

Thank you,

>
>
> vim +6445 kernel/trace/trace.c
>
> 09d23a1d8a82e8 Steven Rostedt (Red Hat 2015-02-03 6362)
> 9c5b9d3d65e485 Masami Hiramatsu 2020-01-11 6363 int tracing_set_tracer(struct trace_array *tr, const char *buf)
> 09d23a1d8a82e8 Steven Rostedt (Red Hat 2015-02-03 6364) {
> bc0c38d139ec7f Steven Rostedt 2008-05-12 6365 struct tracer *t;
> 12883efb670c28 Steven Rostedt (Red Hat 2013-03-05 6366) #ifdef CONFIG_TRACER_MAX_TRACE
> 34600f0e9c33c9 Steven Rostedt 2013-01-22 6367 bool had_max_tr;
> 12883efb670c28 Steven Rostedt (Red Hat 2013-03-05 6368) #endif
> d9e540762f5cdd Peter Zijlstra 2008-11-01 6369 int ret = 0;
> bc0c38d139ec7f Steven Rostedt 2008-05-12 6370
> 1027fcb206a0fb Steven Rostedt 2009-03-12 6371 mutex_lock(&trace_types_lock);
> 1027fcb206a0fb Steven Rostedt 2009-03-12 6372
> 73c5162aa362a5 Steven Rostedt 2009-03-11 6373 if (!ring_buffer_expanded) {
> 2b6080f28c7cc3 Steven Rostedt 2012-05-11 6374 ret = __tracing_resize_ring_buffer(tr, trace_buf_size,
> 438ced1720b584 Vaibhav Nagarnaik 2012-02-02 6375 RING_BUFFER_ALL_CPUS);
> 73c5162aa362a5 Steven Rostedt 2009-03-11 6376 if (ret < 0)
> 59f586db98919d Frederic Weisbecker 2009-03-15 6377 goto out;
> 73c5162aa362a5 Steven Rostedt 2009-03-11 6378 ret = 0;
> 73c5162aa362a5 Steven Rostedt 2009-03-11 6379 }
> 73c5162aa362a5 Steven Rostedt 2009-03-11 6380
> bc0c38d139ec7f Steven Rostedt 2008-05-12 6381 for (t = trace_types; t; t = t->next) {
> bc0c38d139ec7f Steven Rostedt 2008-05-12 6382 if (strcmp(t->name, buf) == 0)
> bc0c38d139ec7f Steven Rostedt 2008-05-12 6383 break;
> bc0c38d139ec7f Steven Rostedt 2008-05-12 6384 }
> c2931e05ec5965 Frederic Weisbecker 2008-10-04 6385 if (!t) {
> c2931e05ec5965 Frederic Weisbecker 2008-10-04 6386 ret = -EINVAL;
> c2931e05ec5965 Frederic Weisbecker 2008-10-04 6387 goto out;
> c2931e05ec5965 Frederic Weisbecker 2008-10-04 6388 }
> 2b6080f28c7cc3 Steven Rostedt 2012-05-11 6389 if (t == tr->current_trace)
> bc0c38d139ec7f Steven Rostedt 2008-05-12 6390 goto out;
> bc0c38d139ec7f Steven Rostedt 2008-05-12 6391
> a35873a0993b4d Tom Zanussi 2019-02-13 6392 #ifdef CONFIG_TRACER_SNAPSHOT
> a35873a0993b4d Tom Zanussi 2019-02-13 6393 if (t->use_max_tr) {
> c0a581d7126c0b Waiman Long 2022-09-22 6394 local_irq_disable();
> a35873a0993b4d Tom Zanussi 2019-02-13 6395 arch_spin_lock(&tr->max_lock);
> a35873a0993b4d Tom Zanussi 2019-02-13 6396 if (tr->cond_snapshot)
> a35873a0993b4d Tom Zanussi 2019-02-13 6397 ret = -EBUSY;
> a35873a0993b4d Tom Zanussi 2019-02-13 6398 arch_spin_unlock(&tr->max_lock);
> c0a581d7126c0b Waiman Long 2022-09-22 6399 local_irq_enable();
> a35873a0993b4d Tom Zanussi 2019-02-13 6400 if (ret)
> a35873a0993b4d Tom Zanussi 2019-02-13 6401 goto out;
> a35873a0993b4d Tom Zanussi 2019-02-13 6402 }
> a35873a0993b4d Tom Zanussi 2019-02-13 6403 #endif
> c7b3ae0bd2ca65 Ziqian SUN (Zamir 2017-09-11 6404) /* Some tracers won't work on kernel command line */
> c7b3ae0bd2ca65 Ziqian SUN (Zamir 2017-09-11 6405) if (system_state < SYSTEM_RUNNING && t->noboot) {
> c7b3ae0bd2ca65 Ziqian SUN (Zamir 2017-09-11 6406) pr_warn("Tracer '%s' is not allowed on command line, ignored\n",
> c7b3ae0bd2ca65 Ziqian SUN (Zamir 2017-09-11 6407) t->name);
> c7b3ae0bd2ca65 Ziqian SUN (Zamir 2017-09-11 6408) goto out;
> c7b3ae0bd2ca65 Ziqian SUN (Zamir 2017-09-11 6409) }
> c7b3ae0bd2ca65 Ziqian SUN (Zamir 2017-09-11 6410)
> 607e2ea167e56d Steven Rostedt (Red Hat 2013-11-06 6411) /* Some tracers are only allowed for the top level buffer */
> 607e2ea167e56d Steven Rostedt (Red Hat 2013-11-06 6412) if (!trace_ok_for_array(t, tr)) {
> 607e2ea167e56d Steven Rostedt (Red Hat 2013-11-06 6413) ret = -EINVAL;
> 607e2ea167e56d Steven Rostedt (Red Hat 2013-11-06 6414) goto out;
> 607e2ea167e56d Steven Rostedt (Red Hat 2013-11-06 6415) }
> 607e2ea167e56d Steven Rostedt (Red Hat 2013-11-06 6416)
> cf6ab6d9143b15 Steven Rostedt (Red Hat 2014-12-15 6417) /* If trace pipe files are being read, we can't change the tracer */
> 7ef282e05132d5 Steven Rostedt (VMware 2020-06-29 6418) if (tr->trace_ref) {
> cf6ab6d9143b15 Steven Rostedt (Red Hat 2014-12-15 6419) ret = -EBUSY;
> cf6ab6d9143b15 Steven Rostedt (Red Hat 2014-12-15 6420) goto out;
> cf6ab6d9143b15 Steven Rostedt (Red Hat 2014-12-15 6421) }
> cf6ab6d9143b15 Steven Rostedt (Red Hat 2014-12-15 6422)
> 9f029e83e968e5 Steven Rostedt 2008-11-12 6423 trace_branch_disable();
> 613f04a0f51e6e Steven Rostedt (Red Hat 2013-03-14 6424)
> 50512ab576e1ce Steven Rostedt (Red Hat 2014-01-14 6425) tr->current_trace->enabled--;
> 613f04a0f51e6e Steven Rostedt (Red Hat 2013-03-14 6426)
> 2b6080f28c7cc3 Steven Rostedt 2012-05-11 6427 if (tr->current_trace->reset)
> 2b6080f28c7cc3 Steven Rostedt 2012-05-11 6428 tr->current_trace->reset(tr);
> 613f04a0f51e6e Steven Rostedt (Red Hat 2013-03-14 6429)
> 7440172974e85b Paul E. McKenney 2018-11-06 6430 /* Current trace needs to be nop_trace before synchronize_rcu */
> 2b6080f28c7cc3 Steven Rostedt 2012-05-11 6431 tr->current_trace = &nop_trace;
> 34600f0e9c33c9 Steven Rostedt 2013-01-22 6432
> 45ad21ca5530ef Steven Rostedt (Red Hat 2013-03-05 6433) #ifdef CONFIG_TRACER_MAX_TRACE
> 45ad21ca5530ef Steven Rostedt (Red Hat 2013-03-05 6434) had_max_tr = tr->allocated_snapshot;
> 34600f0e9c33c9 Steven Rostedt 2013-01-22 6435
> 34600f0e9c33c9 Steven Rostedt 2013-01-22 6436 if (had_max_tr && !t->use_max_tr) {
> 34600f0e9c33c9 Steven Rostedt 2013-01-22 6437 /*
> 34600f0e9c33c9 Steven Rostedt 2013-01-22 6438 * We need to make sure that the update_max_tr sees that
> 34600f0e9c33c9 Steven Rostedt 2013-01-22 6439 * current_trace changed to nop_trace to keep it from
> 34600f0e9c33c9 Steven Rostedt 2013-01-22 6440 * swapping the buffers after we resize it.
> 34600f0e9c33c9 Steven Rostedt 2013-01-22 6441 * The update_max_tr is called from interrupts disabled
> 34600f0e9c33c9 Steven Rostedt 2013-01-22 6442 * so a synchronized_sched() is sufficient.
> 34600f0e9c33c9 Steven Rostedt 2013-01-22 6443 */
> 7440172974e85b Paul E. McKenney 2018-11-06 6444 synchronize_rcu();
> 3209cff4490bee Steven Rostedt (Red Hat 2013-03-12 @6445) free_snapshot(tr);
> ef710e100c1068 KOSAKI Motohiro 2010-07-01 6446 }
> 12883efb670c28 Steven Rostedt (Red Hat 2013-03-05 6447)
> 34600f0e9c33c9 Steven Rostedt 2013-01-22 6448 if (t->use_max_tr && !had_max_tr) {
> 2824f503324860 Steven Rostedt (VMware 2018-05-28 6449) ret = tracing_alloc_snapshot_instance(tr);
> ef710e100c1068 KOSAKI Motohiro 2010-07-01 6450 if (ret < 0)
> ef710e100c1068 KOSAKI Motohiro 2010-07-01 6451 goto out;
> ef710e100c1068 KOSAKI Motohiro 2010-07-01 6452 }
> 12883efb670c28 Steven Rostedt (Red Hat 2013-03-05 6453) #endif
> 577b785f55168d Steven Rostedt 2009-02-26 6454
> 1c80025a49855b Frederic Weisbecker 2008-11-16 6455 if (t->init) {
> b6f11df26fdc28 Arnaldo Carvalho de Melo 2009-02-05 6456 ret = tracer_init(t, tr);
> 1c80025a49855b Frederic Weisbecker 2008-11-16 6457 if (ret)
> 1c80025a49855b Frederic Weisbecker 2008-11-16 6458 goto out;
> 1c80025a49855b Frederic Weisbecker 2008-11-16 6459 }
> bc0c38d139ec7f Steven Rostedt 2008-05-12 6460
> 2b6080f28c7cc3 Steven Rostedt 2012-05-11 6461 tr->current_trace = t;
> 50512ab576e1ce Steven Rostedt (Red Hat 2014-01-14 6462) tr->current_trace->enabled++;
> 9f029e83e968e5 Steven Rostedt 2008-11-12 6463 trace_branch_enable(tr);
> bc0c38d139ec7f Steven Rostedt 2008-05-12 6464 out:
> bc0c38d139ec7f Steven Rostedt 2008-05-12 6465 mutex_unlock(&trace_types_lock);
> bc0c38d139ec7f Steven Rostedt 2008-05-12 6466
> d9e540762f5cdd Peter Zijlstra 2008-11-01 6467 return ret;
> d9e540762f5cdd Peter Zijlstra 2008-11-01 6468 }
> d9e540762f5cdd Peter Zijlstra 2008-11-01 6469
>
> --
> 0-DAY CI Kernel Test Service
> https://01.org/lkp

--
Masami Hiramatsu (Google) <mhiramat@xxxxxxxxxx>

Next message: Abel Vesa: "[PATCH v5 0/5] clk: qcom: Add support for SM8550"
Previous message: Finn Behrens: "Re: [PATCH v2 13/28] rust: alloc: add `Vec::try_with_capacity{,_in}()` constructors"
In reply to: kernel test robot: "Re: [PATCH] tracing: Fix complicated dependency of"
Next in thread: David Howells: "Re: [PATCH] trace: Fix some checker warnings"
Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]