The architecture specific information of the running processes could
be useful to the userland. Add support to examine process architecture
specific information externally.
Signed-off-by: Aubrey Li <[email protected]>
Cc: Peter Zijlstra <[email protected]>
Cc: Andi Kleen <[email protected]>
Cc: Tim Chen <[email protected]>
Cc: Dave Hansen <[email protected]>
Cc: Arjan van de Ven <[email protected]>
---
fs/proc/array.c | 5 +++++
1 file changed, 5 insertions(+)
diff --git a/fs/proc/array.c b/fs/proc/array.c
index 0ceb3b6b37e7..d8cb5b5fd7bb 100644
--- a/fs/proc/array.c
+++ b/fs/proc/array.c
@@ -392,6 +392,10 @@ static inline void task_core_dumping(struct seq_file *m, struct mm_struct *mm)
seq_putc(m, '\n');
}
+void __weak arch_proc_pid_status(struct seq_file *m, struct task_struct *task)
+{
+}
+
int proc_pid_status(struct seq_file *m, struct pid_namespace *ns,
struct pid *pid, struct task_struct *task)
{
@@ -414,6 +418,7 @@ int proc_pid_status(struct seq_file *m, struct pid_namespace *ns,
task_cpus_allowed(m, task);
cpuset_task_status_allowed(m, task);
task_context_switch_counts(m, task);
+ arch_proc_pid_status(m, task);
return 0;
}
--
2.17.1
Added AVX512_elapsed_ms in /proc/<pid>/status. Report it
in Documentation/filesystems/proc.txt
Signed-off-by: Aubrey Li <[email protected]>
Cc: Peter Zijlstra <[email protected]>
Cc: Andi Kleen <[email protected]>
Cc: Tim Chen <[email protected]>
Cc: Dave Hansen <[email protected]>
Cc: Arjan van de Ven <[email protected]>
---
Documentation/filesystems/proc.txt | 4 +++-
1 file changed, 3 insertions(+), 1 deletion(-)
diff --git a/Documentation/filesystems/proc.txt b/Documentation/filesystems/proc.txt
index 520f6a84cf50..8275ccd766e4 100644
--- a/Documentation/filesystems/proc.txt
+++ b/Documentation/filesystems/proc.txt
@@ -197,6 +197,7 @@ read the file /proc/PID/status:
Seccomp: 0
voluntary_ctxt_switches: 0
nonvoluntary_ctxt_switches: 1
+ AVX512_elapsed_ms: 1020
This shows you nearly the same information you would get if you viewed it with
the ps command. In fact, ps uses the proc file system to obtain its
@@ -214,7 +215,7 @@ asynchronous manner and the value may not be very precise. To see a precise
snapshot of a moment, you can see /proc/<pid>/smaps file and scan page table.
It's slow but very precise.
-Table 1-2: Contents of the status files (as of 4.8)
+Table 1-2: Contents of the status files (as of 5.1)
..............................................................................
Field Content
Name filename of the executable
@@ -275,6 +276,7 @@ Table 1-2: Contents of the status files (as of 4.8)
Mems_allowed_list Same as previous, but in "list format"
voluntary_ctxt_switches number of voluntary context switches
nonvoluntary_ctxt_switches number of non voluntary context switches
+ AVX512_elapsed_ms time elapsed since last AVX512 use in millisecond
..............................................................................
Table 1-3: Contents of the statm files (as of 2.6.8-rc3)
--
2.17.1
AVX-512 components use could cause core turbo frequency drop. So
it's useful to expose AVX-512 usage elapsed time as a heuristic hint
for the user space job scheduler to cluster the AVX-512 using tasks
together.
Example:
$ cat /proc/pid/status | grep AVX512_elapsed_ms
AVX512_elapsed_ms: 1020
The number '1020' denotes 1020 millisecond elapsed since last time
context switch the off-CPU task using AVX-512 components, thus the
task could cause core frequency drop.
Or:
$ cat /proc/pid/status | grep AVX512_elapsed_ms
AVX512_elapsed_ms: -1
The number '-1' indicates the task didn't use AVX-512 components
before thus unlikely has frequency drop issue.
User space tools may want to further check by:
$ perf stat --pid <pid> -e core_power.lvl2_turbo_license -- sleep 1
Performance counter stats for process id '3558':
3,251,565,961 core_power.lvl2_turbo_license
1.004031387 seconds time elapsed
Non-zero counter value confirms that the task causes frequency drop.
Signed-off-by: Aubrey Li <[email protected]>
Cc: Peter Zijlstra <[email protected]>
Cc: Andi Kleen <[email protected]>
Cc: Tim Chen <[email protected]>
Cc: Dave Hansen <[email protected]>
Cc: Arjan van de Ven <[email protected]>
---
arch/x86/include/asm/processor.h | 2 ++
arch/x86/kernel/fpu/xstate.c | 41 ++++++++++++++++++++++++++++++++
2 files changed, 43 insertions(+)
diff --git a/arch/x86/include/asm/processor.h b/arch/x86/include/asm/processor.h
index d53c54b842da..60ee932070fe 100644
--- a/arch/x86/include/asm/processor.h
+++ b/arch/x86/include/asm/processor.h
@@ -996,5 +996,7 @@ enum l1tf_mitigations {
};
extern enum l1tf_mitigations l1tf_mitigation;
+/* Add support for architecture specific output in /proc/pid/status */
+extern void arch_proc_pid_status(struct seq_file *m, struct task_struct *task);
#endif /* _ASM_X86_PROCESSOR_H */
diff --git a/arch/x86/kernel/fpu/xstate.c b/arch/x86/kernel/fpu/xstate.c
index 87a57b7642d3..dfd7cf847bc3 100644
--- a/arch/x86/kernel/fpu/xstate.c
+++ b/arch/x86/kernel/fpu/xstate.c
@@ -7,6 +7,7 @@
#include <linux/cpu.h>
#include <linux/mman.h>
#include <linux/pkeys.h>
+#include <linux/seq_file.h>
#include <asm/fpu/api.h>
#include <asm/fpu/internal.h>
@@ -1245,3 +1246,43 @@ int copy_user_to_xstate(struct xregs_state *xsave, const void __user *ubuf)
return 0;
}
+
+/*
+ * Report the amount of time elapsed in millisecond since last AVX512
+ * use in the task.
+ */
+void avx512_status(struct seq_file *m, struct task_struct *task)
+{
+ unsigned long timestamp = task->thread.fpu.avx512_timestamp;
+ long delta;
+
+ if (!timestamp) {
+ /*
+ * Report -1 if no AVX512 usage
+ */
+ delta = -1;
+ } else {
+ delta = (long)(jiffies - timestamp);
+ /*
+ * Cap to LONG_MAX if time difference > LONG_MAX
+ */
+ if (delta < 0)
+ delta = LONG_MAX;
+ delta = jiffies_to_msecs(delta);
+ }
+
+ seq_put_decimal_ll(m, "AVX512_elapsed_ms:\t", delta);
+ seq_putc(m, '\n');
+}
+
+/*
+ * Report architecture specific information
+ */
+void arch_proc_pid_status(struct seq_file *m, struct task_struct *task)
+{
+ /*
+ * Report AVX512 state if the processor and build option supported.
+ */
+ if (cpu_feature_enabled(X86_FEATURE_AVX512F))
+ avx512_status(m, task);
+}
--
2.17.1
On Tue, 12 Feb 2019, Aubrey Li wrote:
> The architecture specific information of the running processes could
> be useful to the userland. Add support to examine process architecture
> specific information externally.
>
> Signed-off-by: Aubrey Li <[email protected]>
> Cc: Peter Zijlstra <[email protected]>
> Cc: Andi Kleen <[email protected]>
> Cc: Tim Chen <[email protected]>
> Cc: Dave Hansen <[email protected]>
> Cc: Arjan van de Ven <[email protected]>
> ---
> fs/proc/array.c | 5 +++++
> 1 file changed, 5 insertions(+)
>
> diff --git a/fs/proc/array.c b/fs/proc/array.c
> index 0ceb3b6b37e7..d8cb5b5fd7bb 100644
> --- a/fs/proc/array.c
> +++ b/fs/proc/array.c
> @@ -392,6 +392,10 @@ static inline void task_core_dumping(struct seq_file *m, struct mm_struct *mm)
> seq_putc(m, '\n');
> }
>
> +void __weak arch_proc_pid_status(struct seq_file *m, struct task_struct *task)
> +{
> +}
This still lacks a prototype.
Thanks,
tglx
On Tue, 12 Feb 2019, Thomas Gleixner wrote:
> On Tue, 12 Feb 2019, Aubrey Li wrote:
>
> > The architecture specific information of the running processes could
> > be useful to the userland. Add support to examine process architecture
> > specific information externally.
> >
> > Signed-off-by: Aubrey Li <[email protected]>
> > Cc: Peter Zijlstra <[email protected]>
> > Cc: Andi Kleen <[email protected]>
> > Cc: Tim Chen <[email protected]>
> > Cc: Dave Hansen <[email protected]>
> > Cc: Arjan van de Ven <[email protected]>
> > ---
> > fs/proc/array.c | 5 +++++
> > 1 file changed, 5 insertions(+)
> >
> > diff --git a/fs/proc/array.c b/fs/proc/array.c
> > index 0ceb3b6b37e7..d8cb5b5fd7bb 100644
> > --- a/fs/proc/array.c
> > +++ b/fs/proc/array.c
> > @@ -392,6 +392,10 @@ static inline void task_core_dumping(struct seq_file *m, struct mm_struct *mm)
> > seq_putc(m, '\n');
> > }
> >
> > +void __weak arch_proc_pid_status(struct seq_file *m, struct task_struct *task)
> > +{
> > +}
>
> This still lacks a prototype.
Aside of that this doesn't apply to tip or Linus tree....
Thanks,
tglx
On Tue, 12 Feb 2019, Aubrey Li wrote:
> diff --git a/arch/x86/include/asm/processor.h b/arch/x86/include/asm/processor.h
> index d53c54b842da..60ee932070fe 100644
> --- a/arch/x86/include/asm/processor.h
> +++ b/arch/x86/include/asm/processor.h
> @@ -996,5 +996,7 @@ enum l1tf_mitigations {
> };
>
> extern enum l1tf_mitigations l1tf_mitigation;
> +/* Add support for architecture specific output in /proc/pid/status */
> +extern void arch_proc_pid_status(struct seq_file *m, struct task_struct *task);
Sigh. This is absolutely the wrong place. The weak function is declared and
used in fs/proc/... So the prototype wants to be in a header which is
included from there independent of x86...
Thanks,
tglx
On 2019/2/12 16:22, Thomas Gleixner wrote:
> On Tue, 12 Feb 2019, Aubrey Li wrote:
>> diff --git a/arch/x86/include/asm/processor.h b/arch/x86/include/asm/processor.h
>> index d53c54b842da..60ee932070fe 100644
>> --- a/arch/x86/include/asm/processor.h
>> +++ b/arch/x86/include/asm/processor.h
>> @@ -996,5 +996,7 @@ enum l1tf_mitigations {
>> };
>>
>> extern enum l1tf_mitigations l1tf_mitigation;
>> +/* Add support for architecture specific output in /proc/pid/status */
>> +extern void arch_proc_pid_status(struct seq_file *m, struct task_struct *task);
>
> Sigh. This is absolutely the wrong place. The weak function is declared and
> used in fs/proc/... So the prototype wants to be in a header which is
> included from there independent of x86...
Can the prototype be in the architecture header if they want to call the function?
Like the following? arch_report_meminfo() is used in fs/proc/... as well.
$ find . -name *.h | xargs grep arch_report_meminfo
./arch/s390/include/asm/pgtable.h:void arch_report_meminfo(struct seq_file *m);
./arch/x86/include/asm/pgtable_types.h:extern void arch_report_meminfo(struct seq_file *m);
./arch/parisc/include/asm/pgtable.h:extern void arch_report_meminfo(struct seq_file *m);
Thanks,
-Aubrey
On 2019/2/12 17:14, Li, Aubrey wrote:
> On 2019/2/12 16:22, Thomas Gleixner wrote:
>> On Tue, 12 Feb 2019, Aubrey Li wrote:
>>> diff --git a/arch/x86/include/asm/processor.h b/arch/x86/include/asm/processor.h
>>> index d53c54b842da..60ee932070fe 100644
>>> --- a/arch/x86/include/asm/processor.h
>>> +++ b/arch/x86/include/asm/processor.h
>>> @@ -996,5 +996,7 @@ enum l1tf_mitigations {
>>> };
>>>
>>> extern enum l1tf_mitigations l1tf_mitigation;
>>> +/* Add support for architecture specific output in /proc/pid/status */
>>> +extern void arch_proc_pid_status(struct seq_file *m, struct task_struct *task);
>>
>> Sigh. This is absolutely the wrong place. The weak function is declared and
>> used in fs/proc/... So the prototype wants to be in a header which is
>> included from there independent of x86...
>
> Can the prototype be in the architecture header if they want to call the function?
> Like the following? arch_report_meminfo() is used in fs/proc/... as well.
>
> $ find . -name *.h | xargs grep arch_report_meminfo
> ./arch/s390/include/asm/pgtable.h:void arch_report_meminfo(struct seq_file *m);
> ./arch/x86/include/asm/pgtable_types.h:extern void arch_report_meminfo(struct seq_file *m);
> ./arch/parisc/include/asm/pgtable.h:extern void arch_report_meminfo(struct seq_file *m);
>
Actually both way exist in the current kernel,the reason I chose to put the prototype
into architecture header file is that I found some architectures rename the function
name by a micro definition while others use prototype. See below:
$ find . -name *.h | xargs grep arch_irq_stat
./arch/arm64/include/asm/hardirq.h:#define arch_irq_stat_cpu smp_irq_stat_cpu
./arch/arm/include/asm/hardirq.h:#define arch_irq_stat_cpu smp_irq_stat_cpu
./arch/x86/include/asm/hardirq.h:extern u64 arch_irq_stat_cpu(unsigned int cpu);
This looks more flexible than it in the common header file.
Anyway, putting the prototype into the common header file like include/linux/proc_fs.h
is also acceptable to me if you persist, please just let me know, :)
Thanks,
-Aubrey
On Tue, 12 Feb 2019, Li, Aubrey wrote:
> On 2019/2/12 16:22, Thomas Gleixner wrote:
> > On Tue, 12 Feb 2019, Aubrey Li wrote:
> >> diff --git a/arch/x86/include/asm/processor.h b/arch/x86/include/asm/processor.h
> >> index d53c54b842da..60ee932070fe 100644
> >> --- a/arch/x86/include/asm/processor.h
> >> +++ b/arch/x86/include/asm/processor.h
> >> @@ -996,5 +996,7 @@ enum l1tf_mitigations {
> >> };
> >>
> >> extern enum l1tf_mitigations l1tf_mitigation;
> >> +/* Add support for architecture specific output in /proc/pid/status */
> >> +extern void arch_proc_pid_status(struct seq_file *m, struct task_struct *task);
> >
> > Sigh. This is absolutely the wrong place. The weak function is declared and
> > used in fs/proc/... So the prototype wants to be in a header which is
> > included from there independent of x86...
>
> Can the prototype be in the architecture header if they want to call the
> function?
Basic C programming course:
The prototype must be available before the declaration of the global
function.
fs/proc/array.c:404:13: warning: no previous prototype for ‘arch_proc_pid_status’ [-Wmissing-prototypes]
void __weak arch_proc_pid_status(struct seq_file *m, struct task_struct *task)
Oh well....
Thanks,
tglx
On Tue, 12 Feb 2019, Li, Aubrey wrote:
> $ find . -name *.h | xargs grep arch_irq_stat
> ./arch/arm64/include/asm/hardirq.h:#define arch_irq_stat_cpu smp_irq_stat_cpu
> ./arch/arm/include/asm/hardirq.h:#define arch_irq_stat_cpu smp_irq_stat_cpu
> ./arch/x86/include/asm/hardirq.h:extern u64 arch_irq_stat_cpu(unsigned int cpu);
>
> This looks more flexible than it in the common header file.
Looking more flexible does not make it more correct.
Thanks,
tglx
On 2019/2/12 19:19, Thomas Gleixner wrote:
> On Tue, 12 Feb 2019, Li, Aubrey wrote:
>> On 2019/2/12 16:22, Thomas Gleixner wrote:
>>> On Tue, 12 Feb 2019, Aubrey Li wrote:
>>>> diff --git a/arch/x86/include/asm/processor.h b/arch/x86/include/asm/processor.h
>>>> index d53c54b842da..60ee932070fe 100644
>>>> --- a/arch/x86/include/asm/processor.h
>>>> +++ b/arch/x86/include/asm/processor.h
>>>> @@ -996,5 +996,7 @@ enum l1tf_mitigations {
>>>> };
>>>>
>>>> extern enum l1tf_mitigations l1tf_mitigation;
>>>> +/* Add support for architecture specific output in /proc/pid/status */
>>>> +extern void arch_proc_pid_status(struct seq_file *m, struct task_struct *task);
>>>
>>> Sigh. This is absolutely the wrong place. The weak function is declared and
>>> used in fs/proc/... So the prototype wants to be in a header which is
>>> included from there independent of x86...
>>
>> Can the prototype be in the architecture header if they want to call the
>> function?
>
> Basic C programming course:
>
> The prototype must be available before the declaration of the global
> function.
>
> fs/proc/array.c:404:13: warning: no previous prototype for ‘arch_proc_pid_status’ [-Wmissing-prototypes]
> void __weak arch_proc_pid_status(struct seq_file *m, struct task_struct *task)
>
> Oh well....
Is this because patch 1/3 applied alone? If the whole patch set are applied,
the prototype is included in <asm/processor.h>, which is at the beginning of
array.c file, so it is available before the declaration.
Thanks,
-Aubrey
On 2019/2/12 19:27, Thomas Gleixner wrote:
> On Tue, 12 Feb 2019, Li, Aubrey wrote:
>> $ find . -name *.h | xargs grep arch_irq_stat
>> ./arch/arm64/include/asm/hardirq.h:#define arch_irq_stat_cpu smp_irq_stat_cpu
>> ./arch/arm/include/asm/hardirq.h:#define arch_irq_stat_cpu smp_irq_stat_cpu
>> ./arch/x86/include/asm/hardirq.h:extern u64 arch_irq_stat_cpu(unsigned int cpu);
>>
>> This looks more flexible than it in the common header file.
>
> Looking more flexible does not make it more correct.
Okay, will cook a new version to put it into the common header.
Thanks,
-Aubrey
On Tue, 12 Feb 2019, Li, Aubrey wrote:
> On 2019/2/12 19:19, Thomas Gleixner wrote:
> > On Tue, 12 Feb 2019, Li, Aubrey wrote:
> >> On 2019/2/12 16:22, Thomas Gleixner wrote:
> >>> On Tue, 12 Feb 2019, Aubrey Li wrote:
> >>>> diff --git a/arch/x86/include/asm/processor.h b/arch/x86/include/asm/processor.h
> >>>> index d53c54b842da..60ee932070fe 100644
> >>>> --- a/arch/x86/include/asm/processor.h
> >>>> +++ b/arch/x86/include/asm/processor.h
> >>>> @@ -996,5 +996,7 @@ enum l1tf_mitigations {
> >>>> };
> >>>>
> >>>> extern enum l1tf_mitigations l1tf_mitigation;
> >>>> +/* Add support for architecture specific output in /proc/pid/status */
> >>>> +extern void arch_proc_pid_status(struct seq_file *m, struct task_struct *task);
> >>>
> >>> Sigh. This is absolutely the wrong place. The weak function is declared and
> >>> used in fs/proc/... So the prototype wants to be in a header which is
> >>> included from there independent of x86...
> >>
> >> Can the prototype be in the architecture header if they want to call the
> >> function?
> >
> > Basic C programming course:
> >
> > The prototype must be available before the declaration of the global
> > function.
> >
> > fs/proc/array.c:404:13: warning: no previous prototype for ‘arch_proc_pid_status’ [-Wmissing-prototypes]
> > void __weak arch_proc_pid_status(struct seq_file *m, struct task_struct *task)
> >
> > Oh well....
>
> Is this because patch 1/3 applied alone? If the whole patch set are applied,
> the prototype is included in <asm/processor.h>, which is at the beginning of
> array.c file, so it is available before the declaration.
1) Each patch has to be correct stand alone
2) This file is compiled for every architecture the kernel supports and how
many of them are including arch/x86/include/asm/processor.h ?
There is a world outside x86 and it's rather large.
Thanks,
tglx
On 2019/2/12 19:55, Thomas Gleixner wrote:
> On Tue, 12 Feb 2019, Li, Aubrey wrote:
>> On 2019/2/12 19:19, Thomas Gleixner wrote:
>>> On Tue, 12 Feb 2019, Li, Aubrey wrote:
>>>> On 2019/2/12 16:22, Thomas Gleixner wrote:
>>>>> On Tue, 12 Feb 2019, Aubrey Li wrote:
>>>>>> diff --git a/arch/x86/include/asm/processor.h b/arch/x86/include/asm/processor.h
>>>>>> index d53c54b842da..60ee932070fe 100644
>>>>>> --- a/arch/x86/include/asm/processor.h
>>>>>> +++ b/arch/x86/include/asm/processor.h
>>>>>> @@ -996,5 +996,7 @@ enum l1tf_mitigations {
>>>>>> };
>>>>>>
>>>>>> extern enum l1tf_mitigations l1tf_mitigation;
>>>>>> +/* Add support for architecture specific output in /proc/pid/status */
>>>>>> +extern void arch_proc_pid_status(struct seq_file *m, struct task_struct *task);
>>>>>
>>>>> Sigh. This is absolutely the wrong place. The weak function is declared and
>>>>> used in fs/proc/... So the prototype wants to be in a header which is
>>>>> included from there independent of x86...
>>>>
>>>> Can the prototype be in the architecture header if they want to call the
>>>> function?
>>>
>>> Basic C programming course:
>>>
>>> The prototype must be available before the declaration of the global
>>> function.
>>>
>>> fs/proc/array.c:404:13: warning: no previous prototype for ‘arch_proc_pid_status’ [-Wmissing-prototypes]
>>> void __weak arch_proc_pid_status(struct seq_file *m, struct task_struct *task)
>>>
>>> Oh well....
>>
>> Is this because patch 1/3 applied alone? If the whole patch set are applied,
>> the prototype is included in <asm/processor.h>, which is at the beginning of
>> array.c file, so it is available before the declaration.
>
> 1) Each patch has to be correct stand alone
>
> 2) This file is compiled for every architecture the kernel supports and how
> many of them are including arch/x86/include/asm/processor.h ?
>
> There is a world outside x86 and it's rather large.
Got it, thanks!
>
> Thanks,
>
> tglx
>