* Re: [PATCH v5 08/13] ACPI: add perf low power callback
[not found] ` <20220126233454.3362047-9-eranian@google.com>
@ 2022-01-27 4:58 ` kernel test robot
2022-01-27 7:41 ` kernel test robot
2022-01-27 13:06 ` Peter Zijlstra
2 siblings, 0 replies; 6+ messages in thread
From: kernel test robot @ 2022-01-27 4:58 UTC (permalink / raw)
To: kbuild-all
[-- Attachment #1: Type: text/plain, Size: 4535 bytes --]
Hi Stephane,
Thank you for the patch! Yet something to improve:
[auto build test ERROR on next-20220125]
[cannot apply to tip/x86/core rafael-pm/linux-next tip/perf/core v5.17-rc1 v5.16 v5.16-rc8 v5.17-rc1]
[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]
url: https://github.com/0day-ci/linux/commits/Stephane-Eranian/perf-core-add-perf_clear_branch_entry_bitfields-helper/20220127-083516
base: d25ee88530253138d0b20d43511ca5acbda4e9f7
config: x86_64-allyesconfig (https://download.01.org/0day-ci/archive/20220127/202201271224.g4ALolMJ-lkp(a)intel.com/config)
compiler: gcc-9 (Debian 9.3.0-22) 9.3.0
reproduce (this is a W=1 build):
# https://github.com/0day-ci/linux/commit/fe7943af61c65b96cff36b8388246531402e7f26
git remote add linux-review https://github.com/0day-ci/linux
git fetch --no-tags linux-review Stephane-Eranian/perf-core-add-perf_clear_branch_entry_bitfields-helper/20220127-083516
git checkout fe7943af61c65b96cff36b8388246531402e7f26
# save the config file to linux build tree
mkdir build_dir
make W=1 O=build_dir ARCH=x86_64 SHELL=/bin/bash drivers/
If you fix the issue, kindly add following tag as appropriate
Reported-by: kernel test robot <lkp@intel.com>
All errors (new ones prefixed by >>):
drivers/acpi/acpi_pad.c: In function 'power_saving_thread':
>> drivers/acpi/acpi_pad.c:168:4: error: implicit declaration of function 'perf_lopwr_cb' [-Werror=implicit-function-declaration]
168 | perf_lopwr_cb(true);
| ^~~~~~~~~~~~~
cc1: some warnings being treated as errors
vim +/perf_lopwr_cb +168 drivers/acpi/acpi_pad.c
135
136 static unsigned int idle_pct = 5; /* percentage */
137 static unsigned int round_robin_time = 1; /* second */
138 static int power_saving_thread(void *data)
139 {
140 int do_sleep;
141 unsigned int tsk_index = (unsigned long)data;
142 u64 last_jiffies = 0;
143
144 sched_set_fifo_low(current);
145
146 while (!kthread_should_stop()) {
147 unsigned long expire_time;
148
149 /* round robin to cpus */
150 expire_time = last_jiffies + round_robin_time * HZ;
151 if (time_before(expire_time, jiffies)) {
152 last_jiffies = jiffies;
153 round_robin_cpu(tsk_index);
154 }
155
156 do_sleep = 0;
157
158 expire_time = jiffies + HZ * (100 - idle_pct) / 100;
159
160 while (!need_resched()) {
161 if (tsc_detected_unstable && !tsc_marked_unstable) {
162 /* TSC could halt in idle, so notify users */
163 mark_tsc_unstable("TSC halts in idle");
164 tsc_marked_unstable = 1;
165 }
166 local_irq_disable();
167
> 168 perf_lopwr_cb(true);
169
170 tick_broadcast_enable();
171 tick_broadcast_enter();
172 stop_critical_timings();
173
174 mwait_idle_with_hints(power_saving_mwait_eax, 1);
175
176 start_critical_timings();
177 tick_broadcast_exit();
178
179 perf_lopwr_cb(false);
180
181 local_irq_enable();
182
183 if (time_before(expire_time, jiffies)) {
184 do_sleep = 1;
185 break;
186 }
187 }
188
189 /*
190 * current sched_rt has threshold for rt task running time.
191 * When a rt task uses 95% CPU time, the rt thread will be
192 * scheduled out for 5% CPU time to not starve other tasks. But
193 * the mechanism only works when all CPUs have RT task running,
194 * as if one CPU hasn't RT task, RT task from other CPUs will
195 * borrow CPU time from this CPU and cause RT task use > 95%
196 * CPU time. To make 'avoid starvation' work, takes a nap here.
197 */
198 if (unlikely(do_sleep))
199 schedule_timeout_killable(HZ * idle_pct / 100);
200
201 /* If an external event has set the need_resched flag, then
202 * we need to deal with it, or this loop will continue to
203 * spin without calling __mwait().
204 */
205 if (unlikely(need_resched()))
206 schedule();
207 }
208
209 exit_round_robin(tsk_index);
210 return 0;
211 }
212
---
0-DAY CI Kernel Test Service, Intel Corporation
https://lists.01.org/hyperkitty/list/kbuild-all(a)lists.01.org
^ permalink raw reply [flat|nested] 6+ messages in thread* Re: [PATCH v5 08/13] ACPI: add perf low power callback
[not found] ` <20220126233454.3362047-9-eranian@google.com>
@ 2022-01-27 7:41 ` kernel test robot
2022-01-27 7:41 ` kernel test robot
2022-01-27 13:06 ` Peter Zijlstra
2 siblings, 0 replies; 6+ messages in thread
From: kernel test robot @ 2022-01-27 7:41 UTC (permalink / raw)
To: Stephane Eranian; +Cc: llvm, kbuild-all
Hi Stephane,
Thank you for the patch! Yet something to improve:
[auto build test ERROR on next-20220125]
[cannot apply to tip/x86/core rafael-pm/linux-next tip/perf/core v5.17-rc1 v5.16 v5.16-rc8 v5.17-rc1]
[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]
url: https://github.com/0day-ci/linux/commits/Stephane-Eranian/perf-core-add-perf_clear_branch_entry_bitfields-helper/20220127-083516
base: d25ee88530253138d0b20d43511ca5acbda4e9f7
config: i386-randconfig-a011-20220124 (https://download.01.org/0day-ci/archive/20220127/202201271501.7zuL3YEw-lkp@intel.com/config)
compiler: clang version 14.0.0 (https://github.com/llvm/llvm-project f400a6012c668dfaa73462caf067ceb074e66c47)
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/0day-ci/linux/commit/fe7943af61c65b96cff36b8388246531402e7f26
git remote add linux-review https://github.com/0day-ci/linux
git fetch --no-tags linux-review Stephane-Eranian/perf-core-add-perf_clear_branch_entry_bitfields-helper/20220127-083516
git checkout fe7943af61c65b96cff36b8388246531402e7f26
# save the config file to linux build tree
mkdir build_dir
COMPILER_INSTALL_PATH=$HOME/0day COMPILER=clang make.cross W=1 O=build_dir ARCH=i386 SHELL=/bin/bash drivers/acpi/
If you fix the issue, kindly add following tag as appropriate
Reported-by: kernel test robot <lkp@intel.com>
All errors (new ones prefixed by >>):
>> drivers/acpi/acpi_pad.c:168:4: error: implicit declaration of function 'perf_lopwr_cb' [-Werror,-Wimplicit-function-declaration]
perf_lopwr_cb(true);
^
1 error generated.
vim +/perf_lopwr_cb +168 drivers/acpi/acpi_pad.c
135
136 static unsigned int idle_pct = 5; /* percentage */
137 static unsigned int round_robin_time = 1; /* second */
138 static int power_saving_thread(void *data)
139 {
140 int do_sleep;
141 unsigned int tsk_index = (unsigned long)data;
142 u64 last_jiffies = 0;
143
144 sched_set_fifo_low(current);
145
146 while (!kthread_should_stop()) {
147 unsigned long expire_time;
148
149 /* round robin to cpus */
150 expire_time = last_jiffies + round_robin_time * HZ;
151 if (time_before(expire_time, jiffies)) {
152 last_jiffies = jiffies;
153 round_robin_cpu(tsk_index);
154 }
155
156 do_sleep = 0;
157
158 expire_time = jiffies + HZ * (100 - idle_pct) / 100;
159
160 while (!need_resched()) {
161 if (tsc_detected_unstable && !tsc_marked_unstable) {
162 /* TSC could halt in idle, so notify users */
163 mark_tsc_unstable("TSC halts in idle");
164 tsc_marked_unstable = 1;
165 }
166 local_irq_disable();
167
> 168 perf_lopwr_cb(true);
169
170 tick_broadcast_enable();
171 tick_broadcast_enter();
172 stop_critical_timings();
173
174 mwait_idle_with_hints(power_saving_mwait_eax, 1);
175
176 start_critical_timings();
177 tick_broadcast_exit();
178
179 perf_lopwr_cb(false);
180
181 local_irq_enable();
182
183 if (time_before(expire_time, jiffies)) {
184 do_sleep = 1;
185 break;
186 }
187 }
188
189 /*
190 * current sched_rt has threshold for rt task running time.
191 * When a rt task uses 95% CPU time, the rt thread will be
192 * scheduled out for 5% CPU time to not starve other tasks. But
193 * the mechanism only works when all CPUs have RT task running,
194 * as if one CPU hasn't RT task, RT task from other CPUs will
195 * borrow CPU time from this CPU and cause RT task use > 95%
196 * CPU time. To make 'avoid starvation' work, takes a nap here.
197 */
198 if (unlikely(do_sleep))
199 schedule_timeout_killable(HZ * idle_pct / 100);
200
201 /* If an external event has set the need_resched flag, then
202 * we need to deal with it, or this loop will continue to
203 * spin without calling __mwait().
204 */
205 if (unlikely(need_resched()))
206 schedule();
207 }
208
209 exit_round_robin(tsk_index);
210 return 0;
211 }
212
---
0-DAY CI Kernel Test Service, Intel Corporation
https://lists.01.org/hyperkitty/list/kbuild-all@lists.01.org
^ permalink raw reply [flat|nested] 6+ messages in thread* Re: [PATCH v5 08/13] ACPI: add perf low power callback
@ 2022-01-27 7:41 ` kernel test robot
0 siblings, 0 replies; 6+ messages in thread
From: kernel test robot @ 2022-01-27 7:41 UTC (permalink / raw)
To: kbuild-all
[-- Attachment #1: Type: text/plain, Size: 4743 bytes --]
Hi Stephane,
Thank you for the patch! Yet something to improve:
[auto build test ERROR on next-20220125]
[cannot apply to tip/x86/core rafael-pm/linux-next tip/perf/core v5.17-rc1 v5.16 v5.16-rc8 v5.17-rc1]
[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]
url: https://github.com/0day-ci/linux/commits/Stephane-Eranian/perf-core-add-perf_clear_branch_entry_bitfields-helper/20220127-083516
base: d25ee88530253138d0b20d43511ca5acbda4e9f7
config: i386-randconfig-a011-20220124 (https://download.01.org/0day-ci/archive/20220127/202201271501.7zuL3YEw-lkp(a)intel.com/config)
compiler: clang version 14.0.0 (https://github.com/llvm/llvm-project f400a6012c668dfaa73462caf067ceb074e66c47)
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/0day-ci/linux/commit/fe7943af61c65b96cff36b8388246531402e7f26
git remote add linux-review https://github.com/0day-ci/linux
git fetch --no-tags linux-review Stephane-Eranian/perf-core-add-perf_clear_branch_entry_bitfields-helper/20220127-083516
git checkout fe7943af61c65b96cff36b8388246531402e7f26
# save the config file to linux build tree
mkdir build_dir
COMPILER_INSTALL_PATH=$HOME/0day COMPILER=clang make.cross W=1 O=build_dir ARCH=i386 SHELL=/bin/bash drivers/acpi/
If you fix the issue, kindly add following tag as appropriate
Reported-by: kernel test robot <lkp@intel.com>
All errors (new ones prefixed by >>):
>> drivers/acpi/acpi_pad.c:168:4: error: implicit declaration of function 'perf_lopwr_cb' [-Werror,-Wimplicit-function-declaration]
perf_lopwr_cb(true);
^
1 error generated.
vim +/perf_lopwr_cb +168 drivers/acpi/acpi_pad.c
135
136 static unsigned int idle_pct = 5; /* percentage */
137 static unsigned int round_robin_time = 1; /* second */
138 static int power_saving_thread(void *data)
139 {
140 int do_sleep;
141 unsigned int tsk_index = (unsigned long)data;
142 u64 last_jiffies = 0;
143
144 sched_set_fifo_low(current);
145
146 while (!kthread_should_stop()) {
147 unsigned long expire_time;
148
149 /* round robin to cpus */
150 expire_time = last_jiffies + round_robin_time * HZ;
151 if (time_before(expire_time, jiffies)) {
152 last_jiffies = jiffies;
153 round_robin_cpu(tsk_index);
154 }
155
156 do_sleep = 0;
157
158 expire_time = jiffies + HZ * (100 - idle_pct) / 100;
159
160 while (!need_resched()) {
161 if (tsc_detected_unstable && !tsc_marked_unstable) {
162 /* TSC could halt in idle, so notify users */
163 mark_tsc_unstable("TSC halts in idle");
164 tsc_marked_unstable = 1;
165 }
166 local_irq_disable();
167
> 168 perf_lopwr_cb(true);
169
170 tick_broadcast_enable();
171 tick_broadcast_enter();
172 stop_critical_timings();
173
174 mwait_idle_with_hints(power_saving_mwait_eax, 1);
175
176 start_critical_timings();
177 tick_broadcast_exit();
178
179 perf_lopwr_cb(false);
180
181 local_irq_enable();
182
183 if (time_before(expire_time, jiffies)) {
184 do_sleep = 1;
185 break;
186 }
187 }
188
189 /*
190 * current sched_rt has threshold for rt task running time.
191 * When a rt task uses 95% CPU time, the rt thread will be
192 * scheduled out for 5% CPU time to not starve other tasks. But
193 * the mechanism only works when all CPUs have RT task running,
194 * as if one CPU hasn't RT task, RT task from other CPUs will
195 * borrow CPU time from this CPU and cause RT task use > 95%
196 * CPU time. To make 'avoid starvation' work, takes a nap here.
197 */
198 if (unlikely(do_sleep))
199 schedule_timeout_killable(HZ * idle_pct / 100);
200
201 /* If an external event has set the need_resched flag, then
202 * we need to deal with it, or this loop will continue to
203 * spin without calling __mwait().
204 */
205 if (unlikely(need_resched()))
206 schedule();
207 }
208
209 exit_round_robin(tsk_index);
210 return 0;
211 }
212
---
0-DAY CI Kernel Test Service, Intel Corporation
https://lists.01.org/hyperkitty/list/kbuild-all(a)lists.01.org
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: [PATCH v5 08/13] ACPI: add perf low power callback
[not found] ` <20220126233454.3362047-9-eranian@google.com>
2022-01-27 4:58 ` [PATCH v5 08/13] ACPI: add perf low power callback kernel test robot
2022-01-27 7:41 ` kernel test robot
@ 2022-01-27 13:06 ` Peter Zijlstra
2022-02-01 7:33 ` Stephane Eranian
2 siblings, 1 reply; 6+ messages in thread
From: Peter Zijlstra @ 2022-01-27 13:06 UTC (permalink / raw)
To: Stephane Eranian
Cc: linux-kernel, kim.phillips, acme, jolsa, songliubraving,
Rafael J. Wysocki, linux-acpi
On Wed, Jan 26, 2022 at 03:34:49PM -0800, Stephane Eranian wrote:
> This patch add an optional callback needed by some PMU features, e.g., AMD
"This patch" is a documented fail.
> BRS, to give a chance to the perf_events code to change its state before
> going to low power and after coming back.
>
> The callback is void when the PERF_NEEDS_LOPWR_CB flag is not set.
> This flag must be set in arch specific perf_event.h header whenever needed.
> When not set, there is no impact of the ACPI code.
>
> Signed-off-by: Stephane Eranian <eranian@google.com>
> ---
> drivers/acpi/acpi_pad.c | 6 ++++++
> drivers/acpi/processor_idle.c | 5 +++++
> include/linux/perf_event.h | 6 ++++++
> 3 files changed, 17 insertions(+)
>
> diff --git a/drivers/acpi/acpi_pad.c b/drivers/acpi/acpi_pad.c
> index f45979aa2d64..a306a07a60b5 100644
> --- a/drivers/acpi/acpi_pad.c
> +++ b/drivers/acpi/acpi_pad.c
> @@ -164,6 +164,9 @@ static int power_saving_thread(void *data)
> tsc_marked_unstable = 1;
> }
> local_irq_disable();
> +
> + perf_lopwr_cb(true);
> +
> tick_broadcast_enable();
> tick_broadcast_enter();
> stop_critical_timings();
> @@ -172,6 +175,9 @@ static int power_saving_thread(void *data)
>
> start_critical_timings();
> tick_broadcast_exit();
> +
> + perf_lopwr_cb(false);
> +
> local_irq_enable();
>
> if (time_before(expire_time, jiffies)) {
You forgot to Cc the maintainers of this stinking pile of poo. Can we
please delete it instead?
^ permalink raw reply [flat|nested] 6+ messages in thread* Re: [PATCH v5 08/13] ACPI: add perf low power callback
2022-01-27 13:06 ` Peter Zijlstra
@ 2022-02-01 7:33 ` Stephane Eranian
2022-02-01 10:20 ` Peter Zijlstra
0 siblings, 1 reply; 6+ messages in thread
From: Stephane Eranian @ 2022-02-01 7:33 UTC (permalink / raw)
To: Peter Zijlstra
Cc: linux-kernel, kim.phillips, acme, jolsa, songliubraving,
Rafael J. Wysocki, linux-acpi
On Thu, Jan 27, 2022 at 5:06 AM Peter Zijlstra <peterz@infradead.org> wrote:
>
> On Wed, Jan 26, 2022 at 03:34:49PM -0800, Stephane Eranian wrote:
> > This patch add an optional callback needed by some PMU features, e.g., AMD
>
> "This patch" is a documented fail.
>
I understand this is not ideal. But the BRS feature must be disabled
before going low-power.
Do you have a better proposal? Been trying to find the simplest way to
support this with no
impact when not needed.
> > BRS, to give a chance to the perf_events code to change its state before
> > going to low power and after coming back.
> >
> > The callback is void when the PERF_NEEDS_LOPWR_CB flag is not set.
> > This flag must be set in arch specific perf_event.h header whenever needed.
> > When not set, there is no impact of the ACPI code.
> >
> > Signed-off-by: Stephane Eranian <eranian@google.com>
> > ---
> > drivers/acpi/acpi_pad.c | 6 ++++++
> > drivers/acpi/processor_idle.c | 5 +++++
> > include/linux/perf_event.h | 6 ++++++
> > 3 files changed, 17 insertions(+)
> >
> > diff --git a/drivers/acpi/acpi_pad.c b/drivers/acpi/acpi_pad.c
> > index f45979aa2d64..a306a07a60b5 100644
> > --- a/drivers/acpi/acpi_pad.c
> > +++ b/drivers/acpi/acpi_pad.c
> > @@ -164,6 +164,9 @@ static int power_saving_thread(void *data)
> > tsc_marked_unstable = 1;
> > }
> > local_irq_disable();
> > +
> > + perf_lopwr_cb(true);
> > +
> > tick_broadcast_enable();
> > tick_broadcast_enter();
> > stop_critical_timings();
> > @@ -172,6 +175,9 @@ static int power_saving_thread(void *data)
> >
> > start_critical_timings();
> > tick_broadcast_exit();
> > +
> > + perf_lopwr_cb(false);
> > +
> > local_irq_enable();
> >
> > if (time_before(expire_time, jiffies)) {
>
> You forgot to Cc the maintainers of this stinking pile of poo. Can we
> please delete it instead?
^ permalink raw reply [flat|nested] 6+ messages in thread* Re: [PATCH v5 08/13] ACPI: add perf low power callback
2022-02-01 7:33 ` Stephane Eranian
@ 2022-02-01 10:20 ` Peter Zijlstra
0 siblings, 0 replies; 6+ messages in thread
From: Peter Zijlstra @ 2022-02-01 10:20 UTC (permalink / raw)
To: Stephane Eranian
Cc: linux-kernel, kim.phillips, acme, jolsa, songliubraving,
Rafael J. Wysocki, linux-acpi
On Mon, Jan 31, 2022 at 11:33:10PM -0800, Stephane Eranian wrote:
> On Thu, Jan 27, 2022 at 5:06 AM Peter Zijlstra <peterz@infradead.org> wrote:
> >
> > On Wed, Jan 26, 2022 at 03:34:49PM -0800, Stephane Eranian wrote:
> > > This patch add an optional callback needed by some PMU features, e.g., AMD
> >
> > "This patch" is a documented fail.
> >
> I understand this is not ideal.
git grep -i "this patch" Documentation/process/submitting-patches.rst
"Describe your changes in imperative mood, e.g. "make xyzzy do frotz"
instead of "[This patch] makes xyzzy do frotz" or "[I] changed xyzzy
to do frotz", as if you are giving orders to the codebase to change
its behaviour."
IOW, don't use "This patch" to start a Changelog (lot's of your patches
here do so).
^ permalink raw reply [flat|nested] 6+ messages in thread