On (10/10/19 20:23), Joe Perches wrote:
> On Fri, 2019-10-11 at 10:52 +0900, Sergey Senozhatsky wrote:
> > Hi Joe,
>
> Hi Sergey.
Hi Joe,
For some reason your reply triggered gmail spam filter; took
me a while to notice and "recover" it from spam folder.
[..]
> > Both LINUX_VERSION_CODE and KER
Dear Friend,
I came across your e-mail contact prior a private search whilst in
need of your partnership for investment assistance in your country. I
am opportune to use this medium to exhibit my legal intentions towards
investing to your country under your management. I am fully convinced
that yo
On Sun, Oct 20, 2019 at 3:33 AM wrote:
> On Sat, 2019-10-19 at 11:31 +0300, Dan Carpenter wrote:
> > On Fri, Oct 18, 2019 at 06:42:13PM -0400, Ayman Bagabas wrote:
> > > Check if battery thresholds are within 0 and 100.
> > Don't forget to add your Signed-off-by: though.
Please, resend all three
Any review comments for this?
Or it should be merged?
Thanks.
On Thu, Aug 22, 2019 at 02:08:14PM +0200, Martin Kepplinger wrote:
> Am 19.08.2019 12:08 schrieb Dixit Parmar:
> > From: Dixit Parmar
> >
> > For Sitronix st1633 multi-touch controller driver the co-ordinates
> > reported
> > for mu
On the basis of the additional information (thanks), there might be a
more specific path to investigate: It is that the PCI code does not
enumerate the second cardbus bridge PCI function in the more recent 4.19
kernel compared to the anvient (and working) 2.6 kernel.
Namely, only one CardBus bridg
On Mon, 2019-10-14 at 16:51:28 UTC, Christophe Leroy wrote:
> Make sure starting addr is aligned to segment boundary so that when
> incrementing the segment, the starting address of the new segment is
> below the end address. Otherwise the last segment might get missed.
>
> Fixes: a68c31fc01ef ("
There is no functional changs, just some cleanup and renaming to increase
readability.
Patch 1 is newly added from v2.
Patcd 2 and 3 is seperated from Patch 4.
Xiaoyao Li (4):
KVM: VMX: Write VPID to vmcs when creating vcpu
KVM: VMX: Remove vmx->hv_deadline_tsc initialization from
vmx_vcp
Move the code that writes vmx->vpid to vmcs from vmx_vcpu_reset() to
vmx_vcpu_setup(), because vmx->vpid is allocated when creating vcpu and
never changed. So we don't need to update the vmcs.vpid when resetting
vcpu.
Signed-off-by: Xiaoyao Li
---
arch/x86/kvm/vmx/vmx.c | 6 +++---
1 file change
Rename {vmx,nested_vmx}_vcpu_setup() to match what they really do.
Signed-off-by: Xiaoyao Li
---
Changes in v3:
- Move vmcs unrelated changes into 2 seperate patches.
- refine the function name.
---
arch/x86/kvm/vmx/nested.c | 2 +-
arch/x86/kvm/vmx/nested.h | 2 +-
arch/x86/kvm/vmx/vmx.c
Move the initialization of vmx->guest_msrs[] from vmx_vcpu_setup() to
vmx_create_vcpu(), and put it right after its allocation.
This also is the preperation for next patch.
Signed-off-by: Xiaoyao Li
---
arch/x86/kvm/vmx/vmx.c | 34 --
1 file changed, 16 insertion
... It can be removed here because the same code is called later in
vmx_vcpu_reset() as the flow:
kvm_arch_vcpu_setup()
-> kvm_vcpu_reset()
-> vmx_vcpu_reset()
Signed-off-by: Xiaoyao Li
---
arch/x86/kvm/vmx/vmx.c | 1 -
1 file changed, 1 deletion(-)
diff --git a/arch/x8
On Fri, Oct 18, 2019 at 06:40:24PM -0700, Andrew Morton wrote:
> On Fri, 18 Oct 2019 15:09:59 +0100 Mel Gorman
> wrote:
>
> > > > Cc: sta...@vger.kernel.org # v4.15+
> > >
> > > Hmm, are you sure about 4.15? Doesn't this go all the way down to
> > > deferred initialization? I do not see any rec
Subject: DNS Records for Teo En Ming Corporation (Office 365 Business Premium)
(Using Linux command dig)
Good day from Singapore,
Office 365 Business Premium User: CEO at teo-en-ming-corp.com
Linux command dig:
teo-en-ming@debian:~$ dig @8.8.8.8 teo-en-ming-corp.com ANY
; <<>> DiG 9.11.5-P4-5
Linus,
please pull the latest x86-urgent-for-linus git tree from:
git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip.git
x86-urgent-for-linus
up to: 228d120051a2: x86/boot/acpi: Move get_cmdline_acpi_rsdp() under #ifdef
guard
A small set of x86 fixes:
- Prevent a NULL pointer derefer
Linus,
please pull the latest irq-urgent-for-linus git tree from:
git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip.git
irq-urgent-for-linus
up to: c9b59181c2b0: Merge tag 'irqchip-fixes-5.4-1' of
git://git.kernel.org/pub/scm/linux/kernel/git/maz/arm-platforms into irq/urgent
A small
Linus,
please pull the latest timers-urgent-for-linus git tree from:
git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip.git
timers-urgent-for-linus
up to: ff229eee3d89: hrtimer: Annotate lockless access to timer->base
A single commit annotating the lockcless access to timer->base with
R
Linus,
please pull the latest core-urgent-for-linus git tree from:
git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip.git
core-urgent-for-linus
up to: b1fc58333575: stop_machine: Avoid potential race behaviour
A single fix, amending stop machine with WRITE/READ_ONCE() to address the
fal
The pull request you sent on Sun, 20 Oct 2019 10:13:56 -:
> git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip.git
> core-urgent-for-linus
has been merged into torvalds/linux.git:
https://git.kernel.org/torvalds/c/589f1222e043ab1f5cbce558cfaed0c0fa76e7bb
Thank you!
--
Deet-doot-dot, I
The pull request you sent on Sun, 20 Oct 2019 10:13:56 -:
> git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip.git x86-urgent-for-linus
has been merged into torvalds/linux.git:
https://git.kernel.org/torvalds/c/4fe34d61a3a9d9ed954cbc90713506a7598d6b36
Thank you!
--
Deet-doot-dot, I am a
The pull request you sent on Sun, 20 Oct 2019 10:13:56 -:
> git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip.git
> timers-urgent-for-linus
has been merged into torvalds/linux.git:
https://git.kernel.org/torvalds/c/188768f3c0727d4cb95eaad6c67799449aebcc03
Thank you!
--
Deet-doot-dot,
The pull request you sent on Sun, 20 Oct 2019 10:13:56 -:
> git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip.git irq-urgent-for-linus
has been merged into torvalds/linux.git:
https://git.kernel.org/torvalds/c/81c4bc31c4cd6a1e45805604aefd25eee1e72ade
Thank you!
--
Deet-doot-dot, I am a
On Fri, 20 Sep 2019, Andy Lutomirski wrote:
> On Thu, Sep 19, 2019 at 8:09 AM Thomas Gleixner wrote:
> > +#ifndef _TIF_AUDIT
> > +# define _TIF_AUDIT(0)
> > +#endif
>
> I'm wondering if these should be __TIF (double-underscore) or
> MAYBE_TIF_ or something to avoid errors wher
On Sat, Oct 19, 2019 at 11:14 PM Alexander Sverdlin
wrote:
> On Sat, 19 Oct 2019 22:44:18 +0200
> Arnd Bergmann wrote:
> > Ah, that makes sense. so all interrupt numbers need to
> > be shifted by a fixed number (e.g. 1) like we did for
> > other platforms (see attachment).
>
> Yes, the below patc
Hi Dan,
On Mon, Sep 30, 2019 at 09:31:37AM -0500, Dan Murphy wrote:
> Introduce the bq25150 and bq25155 supply chargers.
>
> Signed-off-by: Dan Murphy
> ---
> drivers/power/supply/Kconfig | 8 +
> drivers/power/supply/Makefile | 1 +
> drivers/power/supply/bq2515x_charger
Dear Friend,
I am Ms Lisa Hugh work with the department of Audit and accounting manager here
in the Bank(B.O.A).
Please i need your assistance for the transferring of thIs fund to your bank
account for both of us benefit for life time investment, amount (US$4.5M
DOLLARS).
I have every inqu
Hi,
On Mon, Sep 30, 2019 at 09:31:35AM -0500, Dan Murphy wrote:
> Add HEALTH_WARM, HEALTH_COOL and HEALTH_HOT to the health enum.
You used OVERHEAT instead of HOT in the implementation,
which makes sense to me, so please drop HOT here. Also
this needs to be documented in the ABI documentation:
D
Hi Dan,
On Mon, Sep 30, 2019 at 09:31:36AM -0500, Dan Murphy wrote:
> Add the bindings for the bq25150 and bq25155 500mA
> charging IC from Texas Instruments.
>
> Datasheet:
> http://www.ti.com/lit/ds/symlink/bq25150.pdf
> http://www.ti.com/lit/ds/symlink/bq25155.pdf
>
> Signed-off-by: Dan Murp
The patch documents and updates the memory barriers in ipc/sem.c:
- Add smp_store_release() to wake_up_sem_queue_prepare() and
document why it is needed.
- Read q->status using READ_ONCE+smp_acquire__after_ctrl_dep().
as the pair for the barrier inside wake_up_sem_queue_prepare().
- Add comme
Patch from Davidlohr, I just added this change log.
pipelined_send() and pipelined_receive() are identical, so merge them.
Signed-off-by: Davidlohr Bueso
Signed-off-by: Manfred Spraul
---
ipc/mqueue.c | 31 ++-
1 file changed, 18 insertions(+), 13 deletions(-)
diff
Transfer findings from ipc/mqueue.c:
- A control barrier was missing for the lockless receive case
So in theory, not yet initialized data may have been copied
to user space - obviously only for architectures where
control barriers are not NOP.
- use smp_store_release(). In theory, the refoun
Hi,
Updated series, based on input from Davidlohr and Peter Zijlstra:
- I've dropped the documentation update for wake_q_add, as what it
states is normal: When you call a function and pass a parameter
to a structure, you as caller are responsible to ensure that the
parameter is valid, and
When adding the _{acquire|release|relaxed}() variants of some atomic
operations, it was forgotten to update Documentation/memory_barrier.txt:
smp_mb__{before,after}_atomic() is now intended for all RMW operations
that do not imply a memory barrier.
1)
smp_mb__before_atomic();
atom
Update and document memory barriers for mqueue.c:
- ewp->state is read without any locks, thus READ_ONCE is required.
- add smp_aquire__after_ctrl_dep() after the READ_ONCE, we need
acquire semantics if the value is STATE_READY.
- use wake_q_add_safe()
- document why __set_current_state() may
Hi,
Here are some more notes.
On Mon, Sep 30, 2019 at 09:31:37AM -0500, Dan Murphy wrote:
> Introduce the bq25150 and bq25155 supply chargers.
>
> Signed-off-by: Dan Murphy
> ---
[...]
> +/* initial field values, converted to register values */
> +struct bq2515x_init_data {
> + u8 ichg;
The following commit has been merged into the perf/urgent branch of tip:
Commit-ID: 13301c6b16a6d809b331bb88e40ab9ce38238b8b
Gitweb:
https://git.kernel.org/tip/13301c6b16a6d809b331bb88e40ab9ce38238b8b
Author:Jiri Olsa
AuthorDate:Sun, 20 Oct 2019 00:07:26 +02:00
Committer:
Hi,
On Tue, Oct 15, 2019 at 09:43:41PM +0530, madhuparnabhowmi...@gmail.com wrote:
> From: Madhuparna Bhowmik
>
> Fixed Code indentation error caused due to using spaces
> instead of tabs.
> The error reported by checkpatch.pl is:
> ERROR: code indent should use tabs where possible
> The warnin
Hi,
On Tue, Oct 15, 2019 at 04:54:14PM +0100, Ben Dooks (Codethink) wrote:
> The writel() takes standard integers, not __le32 so
> fix the following sparse warnings by removing the
> cpu_to_le32() calls.
>
> drivers/power/reset/at91-reset.c:134:9: warning: cast from restricted __le32
> drivers/po
Hi,
On Wed, Oct 16, 2019 at 10:33:12AM +0200, Linus Walleij wrote:
> On Fri, Oct 4, 2019 at 5:07 PM Krzysztof Kozlowski wrote:
>
> > It is logical to cleanup in probe's error path in reverse order to
> > previous actions. It also makes easier to add additional goto labels
> > within this error
--
Good afternooni' m Michael A. Grum Do you received my previous email
let me know please :Thanks:
On 19-10-19, 12:58, Bjorn Andersson wrote:
> Most Qualcomm platforms contain a pseudo random number generator
> hardware block. Enable the driver for this block and also enable the
> interface for exposing this to userspace.
Reviewed-by: Vinod Koul
Tested-by: Vinod Koul
>
> Signed-off-by: Bjor
On 02-10-19, 21:13, Bjorn Andersson wrote:
> Add a node describing the watchdog found in the application subsystem.
Reviewed-by: Vinod Koul
>
> Signed-off-by: Bjorn Andersson
> ---
> arch/arm64/boot/dts/qcom/sdm845.dtsi | 6 ++
> 1 file changed, 6 insertions(+)
>
> diff --git a/arch/arm6
From: Ondrej Jirman
Enable Allwinner's USB 3.0 phy and the host controller. Orange Pi 3
board has GL3510 USB 3.0 4-port hub connected to the SoC's USB 3.0
port. All four ports are exposed via USB3-A connectors. VBUS is
always on, since it's powered directly from DCIN (VCC-5V) and
not switchable.
From: Ondrej Jirman
This series implements USB 3 support for Xunlong Orange Pi 3 board.
This is a re-hash of the Icenowy's earlier USB3 work[1] without code
that caused controversy previously. Orange Pi 3 board doesn't need vbus
supply to be dynamically enabled, so that code is not needed to sup
From: Ondrej Jirman
The new Allwinner H6 SoC contains a USB3 PHY that is wired to the
external USB3 pins of the SoC.
Add a device tree binding for the PHY.
Signed-off-by: Ondrej Jirman
---
.../phy/allwinner,sun50i-h6-usb3-phy.yaml | 47 +++
1 file changed, 47 insertions(+)
From: Icenowy Zheng
Allwinner H6 SoC features USB3 functionality, with a DWC3 controller and
a custom PHY.
Add device tree nodes for them.
Signed-off-by: Ondrej Jirman
Signed-off-by: Icenowy Zheng
Reviewed-by: Chen-Yu Tsai
---
arch/arm64/boot/dts/allwinner/sun50i-h6.dtsi | 32 ++
From: Icenowy Zheng
Allwinner H6 SoC contains a USB3 PHY (with USB2 DP/DM lines also
controlled).
Add a driver for it.
The register operations in this driver is mainly extracted from the BSP
USB3 driver.
Signed-off-by: Ondrej Jirman
Signed-off-by: Icenowy Zheng
Reviewed-by: Chen-Yu Tsai
---
On 03-10-19, 17:09, Green Wan wrote:
> diff --git a/drivers/dma/Kconfig b/drivers/dma/Kconfig
> index 7af874b69ffb..03dc82094857 100644
> --- a/drivers/dma/Kconfig
> +++ b/drivers/dma/Kconfig
> @@ -661,6 +661,8 @@ source "drivers/dma/qcom/Kconfig"
>
> source "drivers/dma/dw/Kconfig"
>
> +sour
On 15-10-19, 20:18, Radhey Shyam Pandey wrote:
> This patchset adds callback result, descriptor residue
> calculation and some regression fixes.
Applied, thanks
--
~Vinod
> … But after this release the the return statement
> tries to access the label field of the rule which results in
> use-after-free. Before releaseing the rule, copy errNo and return it
> after releasing rule.
Please avoid a duplicate word and a typo in this change description.
…
> +++ b/securit
On 19-10-09 09:58:14, Peng Fan wrote:
> From: Peng Fan
>
> According Architecture definition guide, SYS_PLL1 is fixed at
> 800MHz, SYS_PLL2 is fixed at 1000MHz, so let's use imx_clk_fixed
> to register the clocks and drop code that could change the rate.
>
> Signed-off-by: Peng Fan
For the ent
On 19-10-16 10:14:23, Anson Huang wrote:
> usdhc's clock rate is different according to different devices
> connected, so clock rate assignment should be placed in board
> DT according to different devices connected on each usdhc port.
>
> Signed-off-by: Anson Huang
For the entire patchset:
Rev
This device is sold as 'ThinkPad USB-C Dock Gen 2 (40AS)'.
Chipset is RTL8153 and works with r8152.
Without this, the generic cdc_ether grabs the device, and the device jam
connected networks up when the machine suspends.
Signed-off-by: Kazutoshi Noguchi
---
drivers/net/usb/cdc_ether.c | 7 +
Use the correct macro when adding the MFD devices instead of using
directly '-1' value.
Signed-off-by: Fabien Parent
Reviewed-by: Matthias Brugger
---
drivers/mfd/mt6397-core.c | 12 ++--
1 file changed, 6 insertions(+), 6 deletions(-)
diff --git a/drivers/mfd/mt6397-core.c b/drivers/m
On Sun, Oct 20, 2019 at 09:43:04PM +0800, Hillf Danton wrote:
> First on the page side, page->prio that is used to mirror the prio
> of page owner tasks is added, and a couple of helpers for setting,
> copying and comparing page->prio to help to add pages to lru.
Um, no. struct page is 64 bytes a
Hi Stephen,
On Tue, Sep 17, 2019 at 01:38:53PM -0700, Stephen Boyd wrote:
> Quoting Manivannan Sadhasivam (2019-09-16 09:14:41)
> > The new implementation for determining parent map uses multiple ways
> > to pass parent info. The order in which it gets processed depends on
> > the first available
On Tue, Sep 17, 2019 at 01:39:56PM -0700, Stephen Boyd wrote:
> Quoting Manivannan Sadhasivam (2019-09-16 09:14:40)
> > The clk_init_data struct needs to be initialized to zero for the new
> > parent_map implementation to work correctly. Otherwise, the member which
> > is available first will get p
On Tue, Sep 17, 2019 at 01:40:42PM -0700, Stephen Boyd wrote:
> Quoting Manivannan Sadhasivam (2019-09-16 09:14:42)
> > This function has been delcared but not defined anywhere. Hence, this
> > commit adds definition for it.
> >
> > Signed-off-by: Manivannan Sadhasivam
> > ---
>
> Can you add a
PM8916 supports an earpiece as another (small) speaker.
The earpiece is routed through RX MIX1 similarly to
the headphones, except that RDAC2 MUX is set to RX1.
Signed-off-by: Stephan Gerhold
---
This patch is based on code from the downstream driver
(msm8x16-wcd.c, [1]) with some cleanup.
This
According to the PM8916 Hardware Register Description,
CDC_D_CDC_CONN_HPHR_DAC_CTL has only a single bit (RX_SEL)
to switch between RX1 (0) and RX2 (1). It is not possible to
disable it entirely to achieve the "ZERO" state.
However, at the moment the "RDAC2 MUX" mixer defines three possible
values
Add 27 MHz oscillator clock node.
Signed-off-by: Andreas Färber
---
v3: New (from previously blocking clk patch series)
arch/arm64/boot/dts/realtek/rtd129x.dtsi | 7 +++
1 file changed, 7 insertions(+)
diff --git a/arch/arm64/boot/dts/realtek/rtd129x.dtsi
b/arch/arm64/boot/dts/realtek/r
Add the watchdog node to the RTD129x Device Tree.
Acked-by: Rob Herring
Acked-by: Guenter Roeck
[AF: Moved from RTD1295 to new RTD129x]
Signed-off-by: Andreas Färber
---
v2 -> v3:
* rtd129x.dtsi was factored out of rtd1295.dtsi, add it there
v1 -> v2: Unchanged
arch/arm64/boot/dts/realt
Hi Stephen,
On Tue, Sep 17, 2019 at 09:47:44PM -0700, Stephen Boyd wrote:
> Quoting Manivannan Sadhasivam (2019-09-16 09:14:46)
> > diff --git a/drivers/clk/Kconfig b/drivers/clk/Kconfig
> > index 801fa1cd0321..e70c64e43ff9 100644
> > --- a/drivers/clk/Kconfig
> > +++ b/drivers/clk/Kconfig
> > @@
On PREEMPT_RT_FULL while running netperf, a corruption
of the skb queue causes an oops.
This appears to be caused by a race condition here
__skb_queue_tail(&trans->queue, skb);
tasklet_schedule(&trans->tasklet);
Where the queue is changed before the tasklet is locked by
tasklet_sch
From: Ira Weiny
xfs_ioctl_setattr_dax_invalidate() currently checks if the DAX flag is
changing as a quick check.
But the implementation mixes the physical (XFS_DIFLAG2_DAX) and
effective (S_DAX) DAX flags.
Remove the use of the effective flag when determining if a change of the
physical flag i
From: Ira Weiny
xfs_inode_supports_dax() should reflect if the inode can support DAX not
that it is enabled for DAX. Leave that to other helper functions.
Change the caller of xfs_inode_supports_dax() to call
xfs_inode_use_dax() which reflects new logic to override the effective
DAX flag with e
From: Ira Weiny
Rather than open coding xfs_inode_supports_dax() in
xfs_ioctl_setattr_dax_invalidate() export xfs_inode_supports_dax() and
call it in preparation for swapping dax flags.
This also means updating xfs_inode_supports_dax() to return true for a
directory.
Signed-off-by: Ira Weiny
-
From: Ira Weiny
In order for users to determine if a file is currently operating in DAX
mode (effective DAX). Define a statx attribute value and set that
attribute if the effective DAX flag is set.
To go along with this we propose the following addition to the statx man
page:
STATX_ATTR_DAX
From: Ira Weiny
At LSF/MM'19 [1] [2] we discussed applications that overestimate memory
consumption due to their inability to detect whether the kernel will
instantiate page cache for a file, and cases where a global dax enable via a
mount option is too coarse.
The following patch series enables
From: Ira Weiny
Switching between DAX and non-DAX on a file is racy with respect to data
operations. However, if no data is involved the flag is safe to switch.
Allow toggling the physical flag if a file is empty. The file length
check is not racy with respect to other operations as it is perf
From: Andi Kleen
The kernel perf subsystem has to IPI to the target CPU for many
operations. On systems with many CPUs and when managing many events the
overhead can be dominated by lots of IPIs.
An alternative is to set up CPU affinity in the perf tool, then set up
all the events for that CPU,
From: Andi Kleen
In some weak fallback cases close can be called a lot with -1. Check
for this case and avoid calling close then.
This is mainly to shut up valgrind which complains about this case.
Signed-off-by: Andi Kleen
---
tools/perf/lib/evsel.c | 3 ++-
tools/perf/util/evsel.c | 3 ++-
From: Andi Kleen
Restructure the event opening in perf stat to cycle through
the events by CPU after setting affinity to that CPU.
This eliminates IPI overhead in the perf API.
We have to loop through the CPU in the outter builtin-stat
code instead of leaving that to low level functions.
It has
From: Andi Kleen
Closing a perf fd can also trigger an IPI to the target CPU.
Use the same affinity technique as we use for reading/enabling events
to closing to optimize the CPU transitions.
Before on a large test case with 94 CPUs:
% time seconds usecs/call callserrors syscall
--
From: Andi Kleen
pmu.c does a lot of redundant /sys accesses while parsing aliases
and probing for PMUs. On large systems with a lot of PMUs this
can get expensive (>2s):
% time seconds usecs/call callserrors syscall
-- --- --- - - ---
From: Andi Kleen
Restructure event reading to use affinity to minimize the number
of IPIs needed.
Before on a large test case with 94 CPUs:
% time seconds usecs/call callserrors syscall
-- --- --- - -
3.160.106079
From: Andi Kleen
Restructure event enabling/disabling to use affinity, which
minimizes the number of IPIs needed.
Before on a large test case with 94 CPUs:
% time seconds usecs/call callserrors syscall
-- --- --- - -
54.651.
From: Andi Kleen
In some cases when perf_event_open fails, it may do some closes to clean
up. In special cases these closes can fail too, which overwrites the
errno of the perf_event_open, which is then incorrectly reported.
Save/restore errno around closes.
Signed-off-by: Andi Kleen
---
tool
From: Andi Kleen
Add some common code that is needed to iterate over all events
in CPU order. Used in followon patches
Signed-off-by: Andi Kleen
---
tools/perf/util/evlist.c | 33 +
tools/perf/util/evlist.h | 4
tools/perf/util/evsel.h | 1 +
3 files cha
This patch kit optimizes perf stat for a large number of events
on systems with many CPUs and PMUs.
Some profiling shows that the most overhead is doing IPIs to
all the target CPUs. We can optimize this by using sched_setaffinity
to set the affinity to a target CPU once and then doing
the perf op
Hi Linus,
Please pull some Kbuild fixes. Thanks.
The following changes since commit 4f5cafb5cb8471e54afdc9054d973535614f7675:
Linux 5.4-rc3 (2019-10-13 16:37:36 -0700)
are available in the Git repository at:
git://git.kernel.org/pub/scm/linux/kernel/git/masahiroy/linux-kbuild.git
tags/kb
"config" regex is too loose, giving a few false positives
CONFIG_and drivers/pci/controller/Kconfig /^config and
MMIO accesses.$/;" r
CONFIG_option arch/x86/Kconfig/^config option.$/;"r
CONFIG_option init/Kconfig/^config optio
The pull request you sent on Mon, 21 Oct 2019 01:17:57 +0900:
> git://git.kernel.org/pub/scm/linux/kernel/git/masahiroy/linux-kbuild.git
> tags/kbuild-fixes-v5.4-2
has been merged into torvalds/linux.git:
https://git.kernel.org/torvalds/c/e2ab4ef83fe279e6f8e06ffa490144d082b95493
Thank you!
--
On 10/18/19 1:33 AM, Xingyu Chen wrote:
The watchdog controller on the Meson-A/C series SoCs is moved to secure
world, watchdog operation needs to be done in secure EL3 mode via ATF,
Non-secure world can call SMC instruction to trap to AFT for watchdog
operation.
Signed-off-by: Xingyu Chen
---
Check if battery thresholds are within 0 and 100.
Fixes: 355a070b09ab ("platform/x86: huawei-wmi: Add battery charging
thresholds")
Signed-off-by: Ayman Bagabas
---
drivers/platform/x86/huawei-wmi.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/drivers/platform/x86/huawei-
No need to check for battery name, we already check if the WMI function is
available in huawei_wmi_battery_setup.
Fixes: 355a070b09ab ("platform/x86: huawei-wmi: Add battery charging
thresholds")
Signed-off-by: Ayman Bagabas
---
drivers/platform/x86/huawei-wmi.c | 4
1 file changed, 4 dele
battery_lock mutex is never used and not needed.
Fixes: 355a070b09ab ("platform/x86: huawei-wmi: Add battery charging
thresholds")
Signed-off-by: Ayman Bagabas
---
drivers/platform/x86/huawei-wmi.c | 2 --
1 file changed, 2 deletions(-)
diff --git a/drivers/platform/x86/huawei-wmi.c
b/drivers
Greetings
My name is Barrister Hans Erich.
I have a client who is interested to invest in your country, she is a well
known politician in her country and deserve a lucrative investment partnership
with you outside her country without any delay Please can you manage such
investment please Kin
From: Anatol Belski
Signed-off-by: Anatol Belski
---
include/linux/byteorder/generic.h | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/include/linux/byteorder/generic.h
b/include/linux/byteorder/generic.h
index 4b13e0a3e15b..c9a4c96c9943 100644
--- a/include/linux/byteo
(euid, egid) pair is snapshotted correctly from task under RCU,
but writeback to inode can be done in any order.
Fix by doing writeback under inode->i_lock where necessary
(/proc/* , /proc/*/fd/* , /proc/*/map_files/* revalidate).
Reported-by: syzbot+e392f8008a294fdf8...@syzkaller.appspotmail.com
On Sat, 19 Oct 2019 16:03:48 +0800, Huazhong Tan wrote:
> This patchset includes some cleanups and optimizations for the HNS3
> ethernet driver.
>
> [patch 1/8] removes unused and unnecessary structures.
>
> [patch 2/8] uses a ETH_ALEN u8 array to replace two mac_addr_*
> field in struct hclge_ma
From: Andi Kleen
pmu.c does a lot of redundant /sys accesses while parsing aliases
and probing for PMUs. On large systems with a lot of PMUs this
can get expensive (>2s):
% time seconds usecs/call callserrors syscall
-- --- --- - - ---
From: Andi Kleen
The kernel perf subsystem has to IPI to the target CPU for many
operations. On systems with many CPUs and when managing many events the
overhead can be dominated by lots of IPIs.
An alternative is to set up CPU affinity in the perf tool, then set up
all the events for that CPU,
From: Andi Kleen
Restructure event enabling/disabling to use affinity, which
minimizes the number of IPIs needed.
Before on a large test case with 94 CPUs:
% time seconds usecs/call callserrors syscall
-- --- --- - -
54.651.
From: Andi Kleen
In some cases when perf_event_open fails, it may do some closes to clean
up. In special cases these closes can fail too, which overwrites the
errno of the perf_event_open, which is then incorrectly reported.
Save/restore errno around closes.
Signed-off-by: Andi Kleen
---
tool
From: Andi Kleen
In some weak fallback cases close can be called a lot with -1. Check
for this case and avoid calling close then.
This is mainly to shut up valgrind which complains about this case.
Signed-off-by: Andi Kleen
---
tools/perf/lib/evsel.c | 3 ++-
tools/perf/util/evsel.c | 3 ++-
From: Andi Kleen
Restructure event reading to use affinity to minimize the number
of IPIs needed.
Before on a large test case with 94 CPUs:
% time seconds usecs/call callserrors syscall
-- --- --- - -
3.160.106079
From: Andi Kleen
Add some common code that is needed to iterate over all events
in CPU order. Used in followon patches
Signed-off-by: Andi Kleen
---
tools/perf/util/evlist.c | 33 +
tools/perf/util/evlist.h | 4
tools/perf/util/evsel.h | 1 +
3 files cha
[The earlier v1 version had a lot of conflicts against some
recent libperf changes in tip/perf/core. Resolve that and
also fix some minor issues.]
This patch kit optimizes perf stat for a large number of events
on systems with many CPUs and PMUs.
Some profiling shows that the most overhead is do
From: Andi Kleen
Restructure the event opening in perf stat to cycle through
the events by CPU after setting affinity to that CPU.
This eliminates IPI overhead in the perf API.
We have to loop through the CPU in the outter builtin-stat
code instead of leaving that to low level functions.
It has
From: Andi Kleen
Closing a perf fd can also trigger an IPI to the target CPU.
Use the same affinity technique as we use for reading/enabling events
to closing to optimize the CPU transitions.
Before on a large test case with 94 CPUs:
% time seconds usecs/call callserrors syscall
--
1 - 100 of 335 matches
Mail list logo
