date:20181113

Re: [PATCH v2 05/11] Doc: DT: thermal: new irq-mode for trip point

2018-11-13 Thread Lukasz Luba

Hi Krzysztof,

Thanks for the comments.

On 11/12/18 9:51 AM, Krzysztof Kozlowski wrote:
> On Wed, 7 Nov 2018 at 18:10, Lukasz Luba  wrote:
>>
> 
> Subject prefix:
> dt-bindings: thermal:
> 
>> Thermal trip point gets new flag in DT: irq-mode.
>> Trip point may have a new explicit flag which indicate
>> IRQ support when the temperature is met (so the thermal framework
>> deos not need to set polling for it).
>> It is useful for 'passive' cooling trip point,
>> which now will not register for polling the temperature.
> 
> You wrap lines in weird way making it more difficult to read.
> I already asked about this while reviewing v1. Please fix it.
> https://elixir.bootlin.com/linux/latest/source/Documentation/process/submitting-patches.rst#L633
>
I will fix it in v3.
>>
>> Update documentation about irq-mode for trip points.
>>
>> Cc: Zhang Rui 
>> Cc: Eduardo Valentin 
>> Cc: Daniel Lezcano 
>> Cc: Rob Herring 
>> Cc: Mark Rutland 
>> Cc: devicet...@vger.kernel.org
>> Signed-off-by: Lukasz Luba 
>> ---
>>   Documentation/devicetree/bindings/thermal/thermal.txt | 7 +++
>>   1 file changed, 7 insertions(+)
>>
>> diff --git a/Documentation/devicetree/bindings/thermal/thermal.txt 
>> b/Documentation/devicetree/bindings/thermal/thermal.txt
>> index ca14ba9..bee21e3 100644
>> --- a/Documentation/devicetree/bindings/thermal/thermal.txt
>> +++ b/Documentation/devicetree/bindings/thermal/thermal.txt
>> @@ -90,6 +90,10 @@ Required properties:
>>  "critical": Hardware not reliable.
>> Type: string
>>
>> +- irq-mode:A flag indicating that trip rises irq, so there is no
> 
> "rises IRQ" (it is an abbreviation).
ACK
> 
> Best regards,
> Krzysztof
> 


Regards,
Lukasz

Re: [PATCH v2 06/11] arm64: dts: exynos5433: add support for thermal trip irq-mode

2018-11-13 Thread Lukasz Luba

Hi Krzysztof,

On 11/12/18 10:00 AM, Krzysztof Kozlowski wrote:
> Thanks Lukasz for patches. I like your work!
> 
> I have few more comments which will probably apply for all the DTS commits.
> 
> On Wed, 7 Nov 2018 at 18:10, Lukasz Luba  wrote:
>>
>> This patch adds support for new flag which indicates
> 
> This patch does not add support. Support for flag was added in your
> first drivers/thermal patches in this set. This patch adds new flag.
> (and does something more, so read on)
Got it, will change the commit message.
> 
>> that trip point triggers IRQ when temperature is met.
>> Exynos5433 supports 8 trip point which will trigger IRQ.
> 
> /8 trip point/8 trip points/
> 
>> Above that number other trip points should be registered
>> without 'irq-mode' flag.
> 
> Please fix the line-wrap.
OK
> 
>>
>> Cc: Kukjin Kim 
>> Cc: Krzysztof Kozlowski 
>> Cc: devicet...@vger.kernel.org
>> Cc: linux-arm-ker...@lists.infradead.org
>> Cc: linux-samsung-...@vger.kernel.org
>> Signed-off-by: Lukasz Luba 
>> ---
>>   arch/arm64/boot/dts/exynos/exynos5433-tmu.dtsi | 105 
>> -
>>   1 file changed, 70 insertions(+), 35 deletions(-)
>>
>> diff --git a/arch/arm64/boot/dts/exynos/exynos5433-tmu.dtsi 
>> b/arch/arm64/boot/dts/exynos/exynos5433-tmu.dtsi
>> index fe3a0b1..c4330f6 100644
>> --- a/arch/arm64/boot/dts/exynos/exynos5433-tmu.dtsi
>> +++ b/arch/arm64/boot/dts/exynos/exynos5433-tmu.dtsi
>> @@ -17,37 +17,44 @@ thermal-zones {
>>  atlas0_alert_0: atlas0-alert-0 {
>>  temperature = <65000>;  /* millicelsius */
>>  hysteresis = <1000>;/* millicelsius */
>> -   type = "active";
>> +   type = "passive";
> 
> This change is not explained in commit msg.
> 
> Basically you are doing here two things (related to each other). You
> clearly define which trip points are IRQ-type and you correct the type
> of trip point. Active is incorrect. This second thing is missing in
> your commit msg and I believe it is main reason behind this patch. You
> should focus then on this reason - start from it. Subject could be
> like "Use proper passive type for thermal trip points".
> 
> Without this explanation I don't see the strong reason behind that
> patch. IOW, everything was working fine before... so why adding this
> new flag? Or maybe something was not fine... and then it is real
> reason why you are sending the patch. Usually commit message should
> answer to the most important question "WHY". Now, it explains
> "WHAT"... but I can see it from the source code. However from the code
> it is not easy to guess WHY.

OK, I will add the explanation 'why' similar to description from the
cover-letter to all the patches which add this flag in DT files.

> 
> Best regards,
> Krzysztof
> 
> 

Thank you for the review.

Regards,
Lukasz

Re: [PATCH v2 05/11] Doc: DT: thermal: new irq-mode for trip point

2018-11-13 Thread Lukasz Luba

Hi Rob,

On 11/12/18 8:09 PM, Rob Herring wrote:
> On Wed, Nov 07, 2018 at 06:09:47PM +0100, Lukasz Luba wrote:
>> Thermal trip point gets new flag in DT: irq-mode.
>> Trip point may have a new explicit flag which indicate
>> IRQ support when the temperature is met (so the thermal framework
>> deos not need to set polling for it).
>> It is useful for 'passive' cooling trip point,
>> which now will not register for polling the temperature.
>>
>> Update documentation about irq-mode for trip points.
> 
> This patch should come before you use it.
OK, I will re-order the patch set in v3.
> 
>>
>> Cc: Zhang Rui 
>> Cc: Eduardo Valentin 
>> Cc: Daniel Lezcano 
>> Cc: Rob Herring 
>> Cc: Mark Rutland 
>> Cc: devicet...@vger.kernel.org
>> Signed-off-by: Lukasz Luba 
>> ---
>>   Documentation/devicetree/bindings/thermal/thermal.txt | 7 +++
>>   1 file changed, 7 insertions(+)
>>
>> diff --git a/Documentation/devicetree/bindings/thermal/thermal.txt 
>> b/Documentation/devicetree/bindings/thermal/thermal.txt
>> index ca14ba9..bee21e3 100644
>> --- a/Documentation/devicetree/bindings/thermal/thermal.txt
>> +++ b/Documentation/devicetree/bindings/thermal/thermal.txt
>> @@ -90,6 +90,10 @@ Required properties:
>>  "critical": Hardware not reliable.
>> Type: string
>>   
>> +- irq-mode: A flag indicating that trip rises irq, so there is no
>> +  Type: boolneed of polling in thermal framework.
>> +  Size: one cell
> 
> Should be optional, right?
Yes, it is optional. I will mention about it in change v3.

Thank you for the review.

Regards,
Lukasz

Re: [PATCH V3] KSM: allow dedup all tasks memory

2018-11-13 Thread Oleksandr Natalenko


Hi.


ksm by default working only on memory that added by
madvise().

And only way get that work on other applications:
  * Use LD_PRELOAD and libraries
  * Patch kernel

Lets use kernel task list and add logic to import VMAs from tasks.

That behaviour controlled by new attributes:
  * mode:
I try mimic hugepages attribute, so mode have two states:
  * madvise  - old default behaviour
  * always [new] - allow ksm to get tasks vma and
   try working on that.
  * seeker_sleep_millisecs:
Add pauses between imports tasks VMA

For rate limiting proporses and tasklist locking time,
ksm seeker thread only import VMAs from one task per loop.

Some numbers from different not madvised workloads.
Formulas:
  Percentage ratio = (pages_sharing - pages_shared)/pages_unshared
  Memory saved = (pages_sharing - pages_shared)*4/1024 MiB
  Memory used = free -h

  * Name: My working laptop
Description: Many different chrome/electron apps + KDE
Ratio: 5%
Saved: ~100  MiB
Used:  ~2000 MiB

  * Name: K8s test VM
Description: Some small random running docker images
Ratio: 40%
Saved: ~160 MiB
Used:  ~920 MiB

  * Name: Ceph test VM
Description: Ceph Mon/OSD, some containers
Ratio: 20%
Saved: ~60 MiB
Used:  ~600 MiB

  * Name: BareMetal K8s backend server
Description: Different server apps in containers C, Java, GO & etc
Ratio: 72%
Saved: ~5800 MiB
Used:  ~35.7 GiB

  * Name: BareMetal K8s processing server
Description: Many instance of one CPU intensive application
Ratio: 55%
Saved: ~2600 MiB
Used:  ~28.0 GiB

  * Name: BareMetal Ceph node
Description: Only OSD storage daemons running
Raio: 2%
Saved: ~190 MiB
Used:  ~11.7 GiB


Out of curiosity, have you compared these results with UKSM [1]?

Thanks.

--
  Oleksandr Natalenko (post-factum)

[1] https://github.com/dolohow/uksm

Re: [PATCH v3 2/3] dt-bindings: i3c: Document Synopsys DesignWare I3C master bindings

2018-11-13 Thread vitor


Hi Boris,

On 12/11/18 09:17, Boris Brezillon wrote:

On Thu,  8 Nov 2018 17:14:10 +
Vitor soares  wrote:


Document Synopsys DesignWare I3C master module

Signed-off-by: Vitor soares 
---
Changes in v3:
- Remove dummy characters

Changes in v2:
- Address the changes in Documentation/devicetree/bindings/i3c/i3c.txt
- Add controller version on compatible string

  .../devicetree/bindings/i3c/snps,dw-i3c-master.txt | 42 ++
  1 file changed, 42 insertions(+)
  create mode 100644 
Documentation/devicetree/bindings/i3c/snps,dw-i3c-master.txt

diff --git a/Documentation/devicetree/bindings/i3c/snps,dw-i3c-master.txt 
b/Documentation/devicetree/bindings/i3c/snps,dw-i3c-master.txt
new file mode 100644
index 000..b930c09
--- /dev/null
+++ b/Documentation/devicetree/bindings/i3c/snps,dw-i3c-master.txt
@@ -0,0 +1,42 @@
+Bindings for Synopsys DesignWare I3C master block
+=
+
+Required properties:
+
+- compatible: shall be "snps,dw-i3c-master-1.00a"
+- clocks: shall reference the core_clk

Are you sure this IP only requires only one clock? If you're unsure,
you'd better have a clock-names prop.

Yes, the controller only has this clock dependency to configure the SCL.



+- interrupts: the interrupt line connected to this I3C master
+- reg: Offset and length of I3C master registers
+
+Mandatory properties defined by the generic binding (see
+Documentation/devicetree/bindings/i3c/i3c.txt for more details):
+
+- #address-cells: shall be set to 3
+- #size-cells: shall be set to 0
+
+Optional properties defined by the generic binding (see
+Documentation/devicetree/bindings/i3c/i3c.txt for more details):
+
+- i2c-scl-hz
+- i3c-scl-hz
+
+I3C device connected on the bus follow the generic description (see
+Documentation/devicetree/bindings/i3c/i3c.txt for more details).
+
+Example:
+
+   i3c-master@2000 {
+   compatible = "snps,dw-i3c-master-1.00a";
+   #address-cells = <3>;
+   #size-cells = <0>;
+   reg = <0x02000 0x1000>;
+   interrupts = <0>;
+   clocks = <&i3cclk>;
+
+   eeprom@57{
+   compatible = "atmel,24c01";
+   reg = < 0x57 0x0 0x10>;

   ^ remove this white space.


+   pagesize = <0x8>;
+   };
+   };
+


Best regards,

Vitor Soares

Re: [PATCH v3 2/3] dt-bindings: i3c: Document Synopsys DesignWare I3C master bindings

2018-11-13 Thread vitor


Hi Rob,

On 13/11/18 02:09, Rob Herring wrote:

On Thu, Nov 08, 2018 at 05:14:10PM +, Vitor soares wrote:

Document Synopsys DesignWare I3C master module

Signed-off-by: Vitor soares 

Please make your author and S-o-b emails match.


This is an alias and both email will work. Anyway I agree that should be 
the same that appear in the mailing list and I will change it.



---
Changes in v3:
- Remove dummy characters

Changes in v2:
- Address the changes in Documentation/devicetree/bindings/i3c/i3c.txt
- Add controller version on compatible string

  .../devicetree/bindings/i3c/snps,dw-i3c-master.txt | 42 ++
  1 file changed, 42 insertions(+)
  create mode 100644 
Documentation/devicetree/bindings/i3c/snps,dw-i3c-master.txt

Otherwise,

Reviewed-by: Rob Herring 


Thanks for review.


Best regards,

Vitor Soares

[PATCH] Documentation/features: Refresh the features list to v4.20-rc2

2018-11-13 Thread Ingo Molnar



* Palmer Dabbelt  wrote:

> I didn't even know this existed until David submitted a patch set that
> included updates to the documentation as a result of some features he
> added to RISC-V.  It looks like there may be a handful of other people
> who don't know this exists either, so I figured I'd just mail out a
> patch set containing all the updates split out as well as I can.
> 
> This smells like something that sholud be automatic, so if I'm jumping
> the gun here then feel free to drop this.  If nobody says anything then
> I guess I'll submit this as a separate PR to Linus from my personal
> tree, as it's not really a RISC-V thing but it seems like it's worth
> having docs that match the code where it's trivial -- I assume that's
> what this does, I didn't actually read anything but the diff because I
> never trust documentation to be up to date...
> 
> I feel compelled to say something like "maybe this should be part of
> checkpatch?", but I'm definately not looking to learn perl :)

I don't think it should be automated or part of checkpatch, but I 
(obviously) agree with the changes, except that I think it should be a 
single patch (combined patch attached below).

Thanks,

Ingo

Subject: Documentation/features: Refresh the features list to v4.20-rc2
From: Palmer Dabbelt 

Run Documentation/features/scripts/feature-refresh.sh to refresh the 
kernel features support matrix list:

 - The new 'csky' architecture was added
 - s390now supports KASAN
 - powerpc now supports stackprotector
 - xtensa  now supports sg-chain
 - arm64   now supports queued-spinlocks
 - parisc  now supports kprobes-events
 - RISC-V  now supports pte_special

[ mingo: combined the patches and the changelogs. ]

Signed-off-by: Palmer Dabbelt 
Signed-off-by: Ingo Molnar 
---

 Documentation/features/core/cBPF-JIT/arch-support.txt  | 1 +
 Documentation/features/core/eBPF-JIT/arch-support.txt  | 1 +
 Documentation/features/core/generic-idle-thread/arch-support.txt   | 1 +
 Documentation/features/core/jump-labels/arch-support.txt   | 1 +
 Documentation/features/core/tracehook/arch-support.txt | 1 +
 Documentation/features/debug/KASAN/arch-support.txt| 3 ++-
 Documentation/features/debug/gcov-profile-all/arch-support.txt | 1 +
 Documentation/features/debug/kgdb/arch-support.txt | 1 +
 Documentation/features/debug/kprobes-on-ftrace/arch-support.txt| 1 +
 Documentation/features/debug/kprobes/arch-support.txt  | 1 +
 Documentation/features/debug/kretprobes/arch-support.txt   | 1 +
 Documentation/features/debug/optprobes/arch-support.txt| 1 +
 Documentation/features/debug/stackprotector/arch-support.txt   | 3 ++-
 Documentation/features/debug/uprobes/arch-support.txt  | 1 +
 Documentation/features/debug/user-ret-profiler/arch-support.txt| 1 +
 Documentation/features/io/dma-contiguous/arch-support.txt  | 1 +
 Documentation/features/io/sg-chain/arch-support.txt| 3 ++-
 Documentation/features/locking/cmpxchg-local/arch-support.txt  | 1 +
 Documentation/features/locking/lockdep/arch-support.txt| 1 +
 Documentation/features/locking/queued-rwlocks/arch-support.txt | 1 +
 Documentation/features/locking/queued-spinlocks/arch-support.txt   | 3 ++-
 Documentation/features/locking/rwsem-optimized/arch-support.txt| 1 +
 Documentation/features/perf/kprobes-event/arch-support.txt | 3 ++-
 Documentation/features/perf/perf-regs/arch-support.txt | 1 +
 Documentation/features/perf/perf-stackdump/arch-support.txt| 1 +
 Documentation/features/sched/membarrier-sync-core/arch-support.txt | 1 +
 Documentation/features/sched/numa-balancing/arch-support.txt   | 1 +
 Documentation/features/seccomp/seccomp-filter/arch-support.txt | 1 +
 Documentation/features/time/arch-tick-broadcast/arch-support.txt   | 1 +
 Documentation/features/time/clockevents/arch-support.txt   | 1 +
 Documentation/features/time/context-tracking/arch-support.txt  | 1 +
 Documentation/features/time/irq-time-acct/arch-support.txt | 1 +
 Documentation/features/time/modern-timekeeping/arch-support.txt| 1 +
 Documentation/features/time/virt-cpuacct/arch-support.txt  | 1 +
 Documentation/features/vm/ELF-ASLR/arch-support.txt| 1 +
 Documentation/features/vm/PG_uncached/arch-support.txt | 1 +
 Documentation/features/vm/THP/arch-support.txt | 1 +
 Documentation/features/vm/TLB/arch-support.txt | 1 +
 Documentation/features/vm/huge-vmap/arch-support.txt   | 1 +
 Documentation/features/vm/ioremap_prot/arch-support.txt| 1 +
 Documentation/features/vm/numa-memblock/arch-support.txt   | 1 +
 Documentation/features/vm/pte_special/arch-support.txt | 3 ++-
 42 files changed, 48 insertions(+), 6 deletions(-)

diff --git a/Documentation/feat

Re: [PATCH V3] KSM: allow dedup all tasks memory

2018-11-13 Thread Timofey Titovets

вт, 13 нояб. 2018 г. в 04:49, Matthew Wilcox :
>
> On Tue, Nov 13, 2018 at 02:13:44AM +0300, Timofey Titovets wrote:
> > Some numbers from different not madvised workloads.
> > Formulas:
> >   Percentage ratio = (pages_sharing - pages_shared)/pages_unshared
> >   Memory saved = (pages_sharing - pages_shared)*4/1024 MiB
> >   Memory used = free -h
> >
> >   * Name: My working laptop
> > Description: Many different chrome/electron apps + KDE
> > Ratio: 5%
> > Saved: ~100  MiB
> > Used:  ~2000 MiB
>
> Your _laptop_ saves 100MB of RAM?  That's extraordinary.  Essentially
> that's like getting an extra 100MB of page cache for free.  Is there
> any observable slowdown?  I could even see there being a speedup (due
> to your working set being allowed to be 5% larger)
>
> I am now a big fan of this patch and shall try to give it the review
> that it deserves.

I'm not sure if this is sarcasm,
anyway i try do my best to get that working.

On any x86 desktop with mixed load (browser, docs, games & etc)
You will always see something like 40-200 MiB of deduped pages,
based on type of load of course.

I'm just don't try use that numbers as reason to get general KSM
deduplication in kernel.
Because in current generation with several gigabytes of memory,
several saved MiB not looks serious for most of people.

Thanks!

Re: [PATCH V3] KSM: allow dedup all tasks memory

2018-11-13 Thread Timofey Titovets

вт, 13 нояб. 2018 г. в 05:25, Pavel Tatashin :
>
> On 18-11-13 02:13:44, Timofey Titovets wrote:
> > From: Timofey Titovets 
> >
> > ksm by default working only on memory that added by
> > madvise().
> >
> > And only way get that work on other applications:
> >   * Use LD_PRELOAD and libraries
> >   * Patch kernel
> >
> > Lets use kernel task list and add logic to import VMAs from tasks.
> >
> > That behaviour controlled by new attributes:
> >   * mode:
> > I try mimic hugepages attribute, so mode have two states:
> >   * madvise  - old default behaviour
> >   * always [new] - allow ksm to get tasks vma and
> >try working on that.
> >   * seeker_sleep_millisecs:
> > Add pauses between imports tasks VMA
> >
> > For rate limiting proporses and tasklist locking time,
> > ksm seeker thread only import VMAs from one task per loop.
> >
> > Some numbers from different not madvised workloads.
> > Formulas:
> >   Percentage ratio = (pages_sharing - pages_shared)/pages_unshared
> >   Memory saved = (pages_sharing - pages_shared)*4/1024 MiB
> >   Memory used = free -h
> >
> >   * Name: My working laptop
> > Description: Many different chrome/electron apps + KDE
> > Ratio: 5%
> > Saved: ~100  MiB
> > Used:  ~2000 MiB
> >
> >   * Name: K8s test VM
> > Description: Some small random running docker images
> > Ratio: 40%
> > Saved: ~160 MiB
> > Used:  ~920 MiB
> >
> >   * Name: Ceph test VM
> > Description: Ceph Mon/OSD, some containers
> > Ratio: 20%
> > Saved: ~60 MiB
> > Used:  ~600 MiB
> >
> >   * Name: BareMetal K8s backend server
> > Description: Different server apps in containers C, Java, GO & etc
> > Ratio: 72%
> > Saved: ~5800 MiB
> > Used:  ~35.7 GiB
> >
> >   * Name: BareMetal K8s processing server
> > Description: Many instance of one CPU intensive application
> > Ratio: 55%
> > Saved: ~2600 MiB
> > Used:  ~28.0 GiB
> >
> >   * Name: BareMetal Ceph node
> > Description: Only OSD storage daemons running
> > Raio: 2%
> > Saved: ~190 MiB
> > Used:  ~11.7 GiB
> >
> > Changes:
> >   v1 -> v2:
> > * Rebase on v4.19.1 (must also apply on 4.20-rc2+)
> >   v2 -> v3:
> > * Reformat patch description
> > * Rename mode normal to madvise
> > * Add some memory numbers
> > * Fix checkpatch.pl warnings
> > * Separate ksm vma seeker to another kthread
> > * Fix: "BUG: scheduling while atomic: ksmd"
> >   by move seeker to another thread
> >
> > Signed-off-by: Timofey Titovets 
> > CC: Matthew Wilcox 
> > CC: linux...@kvack.org
> > CC: linux-doc@vger.kernel.org
> > ---
> >  Documentation/admin-guide/mm/ksm.rst |  15 ++
> >  mm/ksm.c | 215 +++
> >  2 files changed, 198 insertions(+), 32 deletions(-)
> >
> > diff --git a/Documentation/admin-guide/mm/ksm.rst 
> > b/Documentation/admin-guide/mm/ksm.rst
> > index 9303786632d1..7cffd47f9b38 100644
> > --- a/Documentation/admin-guide/mm/ksm.rst
> > +++ b/Documentation/admin-guide/mm/ksm.rst
> > @@ -116,6 +116,21 @@ run
> >  Default: 0 (must be changed to 1 to activate KSM, except if
> >  CONFIG_SYSFS is disabled)
> >
> > +mode
> > +* set always to allow ksm deduplicate memory of every process
> > +* set madvise to use only madvised memory
> > +
> > +Default: madvise (dedupulicate only madvised memory as in
> > +earlier releases)
> > +
> > +seeker_sleep_millisecs
> > +how many milliseconds ksmd task seeker should sleep try another
> > +task.
> > +e.g. ``echo 1000 > /sys/kernel/mm/ksm/seeker_sleep_millisecs``
> > +
> > +Default: 1000 (chosen for rate limit purposes)
> > +
> > +
> >  use_zero_pages
> >  specifies whether empty pages (i.e. allocated pages that only
> >  contain zeroes) should be treated specially.  When set to 1,
> > diff --git a/mm/ksm.c b/mm/ksm.c
> > index 5b0894b45ee5..1a03b28b6288 100644
> > --- a/mm/ksm.c
> > +++ b/mm/ksm.c
> > @@ -273,6 +273,9 @@ static unsigned int ksm_thread_pages_to_scan = 100;
> >  /* Milliseconds ksmd should sleep between batches */
> >  static unsigned int ksm_thread_sleep_millisecs = 20;
> >
> > +/* Milliseconds ksmd seeker should sleep between runs */
> > +static unsigned int ksm_thread_seeker_sleep_millisecs = 1000;
> > +
> >  /* Checksum of an empty (zeroed) page */
> >  static unsigned int zero_checksum __read_mostly;
> >
> > @@ -295,7 +298,12 @@ static int ksm_nr_node_ids = 1;
> >  static unsigned long ksm_run = KSM_RUN_STOP;
> >  static void wait_while_offlining(void);
> >
> > +#define KSM_MODE_MADVISE 0
> > +#define KSM_MODE_ALWAYS  1
> > +static unsigned long ksm_mode = KSM_MODE_MADVISE;
> > +
> >  static DECLARE_WAIT_QUEUE_HEAD(ksm_thread_wait);
> > +static DECLARE_WAIT_QUEUE_HEAD(ksm_seeker_thread_wait);
> >  static DEFINE_MUTEX(ksm_thread_mutex);
> >  static DEFINE_SPINLOCK(ksm_mmlist_lock);
> >
> > @@ -303,6 +3

[PATCH v4 0/8] clk: clkdev: managed clk lookup and provider registrations

2018-11-13 Thread Matti Vaittinen

Series add bd71837/bd71837 PMIC clock support + managed interfaces

Few clk drivers appear to be leaking clkdev lookup registrations at
driver remove. The patch series adds devm versions of lookup
registrations and cleans up few drivers. Driver clean-up patches have
not been tested as I lack the HW. All testing and comments if
driver/device removal is even possible for changed drivers is highly
appreciated. If removal is not possible I will gladly drop the patches
from series - although leaking lookups may serve as bad example for new
developers =)

Patch 8 adds support for clock gate in ROHM bd71837 and bd71847 PMICs.
This change is included in the series because it depends on new managed
interfaces introduced in this series.

bd718x7 driver and devm interfaces are tested on BeagleBoneBlack and
bd71837 break-out board. Clk area register interface of bd71847 is
identical to bd71837.

Changed drivers are:
clk-max77686, clk-st, clk-hi655x, rk808, clk-twl6040
and apcs-msm8916. New driver is clk-bd718x7

This series has been discussed for a while now. For those who want to
see whole discussion:

The bd71837 driver was originally proposed here
https://lore.kernel.org/lkml/d99c8762b0fbbcb18ec4f4d104191364c0ea798c.1528117485.git.matti.vaitti...@fi.rohmeurope.com/

clk portion was separated from that series and devm variants were
proposed here
https://lore.kernel.org/linux-clk/cover.1535630942.git.matti.vaitti...@fi.rohmeurope.com/

Cleanup to other drivers was initiated in this series while waiting for
MFD portions of bd718x7 to be applied. And now when MFD dependencies are in-tree
his version (4) combines bd718x7 driver back to this series.

Changelog (for this series) v4
- Add support for ROHM bd718x7 PMIC clock gate. Included in this patch
series because it depends on managed interfaces added in patch 1.

Changelog (for this series) v3
Address issues spotted by Krzysztof Kozlowski
- Drop patch 3 for clk-s3c2410-dclk as this device can never be removed
- Fix indentiation for clk-max77686
- Rest of the patches are unchanged.

Changelog (for this series) v2
Issue spotted by 0-Day test suite
- Add a stub function 'devm_of_clk_add_parent_hw_provider' for no OF config.
- patches 2-8 are unchanged.

This patch series is based on clk-next

---

Matti Vaittinen (8):
clk: clkdev/of_clk - add managed lookup and provider registrations
clk: clk-max77686: Clean clkdev lookup leak and use devm
clk: clk-st: avoid clkdev lookup leak at remove
clk: clk-hi655x: Free of_provider at remove
clk: rk808: use managed version of of_provider registration
clk: clk-twl6040: Free of_provider at remove
clk: apcs-msm8916: simplify probe cleanup by using devm
clk: bd718x7: Initial support for ROHM bd71837/bd71847 PMIC clock

86 matches

Mail list logo