When doing i2cdetect quick write mode, we would get transfer
error ENOMEM, and i2cdetect shows there's no device at the address.
Quoting from mt8183 datasheet, the number of transfers to be
transferred in one transaction should be set to bigger than 1,
so we should forbid zero-length transfer and
On Wed, Aug 21, 2019 at 06:17:12PM +0200, Marc Gonzalez wrote:
> On 21/08/2019 17:06, Mike Rapoport wrote:
>
> > Both pgtable_cache_init() and pgd_cache_init() are used to initialize kmem
> > cache for page table allocations on several architectures that do not use
> > PAGE_SIZE tables for one or
Greeting's
Today's linux-next kernel 5.3.0-rc5-next-20190820 failed to build on my
powerpc machine
Build errors:
drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c: In function amdgpu_exit:
drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c:1471:2: error: implicit
declaration of function mmu_notifier_synchronize
Hi
On 2019/08/17 2:34, Dmitry Torokhov wrote:
On Fri, Aug 16, 2019 at 05:30:33PM +0900, Jiada Wang wrote:
From: Nick Dyer
Signed-off-by: Nick Dyer
Acked-by: Benson Leung
Acked-by: Yufeng Shen
(cherry picked from ndyer/linux/for-upstream commit
ab95b5a30d2c098daaa9f88d9fcfae7eb516)
On Thu, 2019-08-22 at 07:06 +0200, Christophe Leroy wrote:
>
> Le 22/08/2019 à 02:27, Alastair D'Silva a écrit :
> > On Wed, 2019-08-21 at 22:27 +0200, Christophe Leroy wrote:
> > > Le 20/08/2019 à 06:36, Alastair D'Silva a écrit :
> > > > On Fri, 2019-08-16 at 15:52 +, Christophe Leroy
Greeting's
Today's linux-next kernel 5.3.0-rc5-next-20190820 failed to build on my
powerpc machine
Build errors:
drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c: In function amdgpu_exit:
drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c:1471:2: error: implicit
declaration of function mmu_notifier_synchronize
Hi Rob,
On 21/8/2019 9:35 PM, Rob Herring wrote:
On Wed, Aug 21, 2019 at 5:11 AM Ramuthevar,Vadivel MuruganX
wrote:
From: Ramuthevar Vadivel Murugan
Add a YAML schema to use the host controller driver with the
eMMC PHY on Intel's Lightning Mountain SoC.
Signed-off-by: Ramuthevar Vadivel
On Thu, 2019-08-22 at 06:27 +0200, h...@lst.de wrote:
> On Thu, Aug 22, 2019 at 04:01:24AM +, Atish Patra wrote:
> > The downside of this is that for every !cmask case in true SMP
> > (more
> > common probably) it will execute 2 extra cpumask instructions. As
> > tlbflush path is in
From: Tianyu Lan
Both Hyper-V tsc page and Hyper-V tsc MSR code use variable
hv_sched_clock_offset for their sched clock callback and so
define the variable regardless of CONFIG_HYPERV_TSCPAGE setting.
Signed-off-by: Tianyu Lan
---
This patch is based on the top of
On Wed, Aug 21, 2019 at 09:22:49PM -0700, Eric Biggers wrote:
> > > diff --git a/fs/fs_parser.c b/fs/fs_parser.c
> > > index 83b66c9e9a24..7498a44f18c0 100644
> > > --- a/fs/fs_parser.c
> > > +++ b/fs/fs_parser.c
> > > @@ -206,6 +206,9 @@ int fs_parse(struct fs_context *fc,
> > > case
Hi Dmitry
On 2019/08/17 2:18, Dmitry Torokhov wrote:
On Fri, Aug 16, 2019 at 05:28:53PM +0900, Jiada Wang wrote:
From: Nick Dyer
On some firmware variants, the size of the info block exceeds what can
be read in a single transfer.
Signed-off-by: Nick Dyer
(cherry picked from
From: Eran Ben Elisha
HV VHCA is a layer which provides PF to VF communication channel based on
HyperV PCI config channel. It implements Mellanox's Inter VHCA control
communication protocol. The protocol contains control block in order to
pass messages between the PF and VF drivers, and data
From: Eran Ben Elisha
Control agent is responsible over of the control block (ID 0). It should
update the PF via this block about every capability change. In addition,
upon block 0 invalidate, it should activate all other supported agents
with data requests from the PF.
Upon agent
This interface driver is a helper driver allows other drivers to
have a common interface with the Hyper-V PCI frontend driver.
Signed-off-by: Haiyang Zhang
Signed-off-by: Saeed Mahameed
---
MAINTAINERS | 1 +
drivers/pci/Kconfig | 1 +
From: Eran Ben Elisha
HV VHCA stats agent is responsible on running a preiodic rx/tx
packets/bytes stats update. Currently the supported format is version
MLX5_HV_VHCA_STATS_VERSION. Block ID 1 is dedicated for statistics data
transfer from the VF to the PF.
The reporter fetch the statistics
Le 22/08/2019 à 02:27, Alastair D'Silva a écrit :
On Wed, 2019-08-21 at 22:27 +0200, Christophe Leroy wrote:
Le 20/08/2019 à 06:36, Alastair D'Silva a écrit :
On Fri, 2019-08-16 at 15:52 +, Christophe Leroy wrote:
[...]
Thanks Christophe,
I'm trying a somewhat different approach
From: Dexuan Cui
Windows SR-IOV provides a backchannel mechanism in software for communication
between a VF driver and a PF driver. These "configuration blocks" are
similar in concept to PCI configuration space, but instead of doing reads and
writes in 32-bit chunks through a very slow path,
From: Eran Ben Elisha
Add wrapper functions for HyperV PCIe read / write /
block_invalidate_register operations. This will be used as an
infrastructure in the downstream patch for software communication.
This will be enabled by default if CONFIG_PCI_HYPERV_INTERFACE is set.
Signed-off-by:
This patch set adds paravirtual backchannel in software in pci_hyperv,
which is required by the mlx5e driver HV VHCA stats agent.
The stats agent is responsible on running a periodic rx/tx packets/bytes
stats update.
Dexuan Cui (1):
PCI: hv: Add a paravirtual backchannel in software
Eran Ben
@YueHaibing thanks for the patch.
On 2019-08-22 08:10, Viresh Kumar wrote:
On 21-08-19, 20:14, YueHaibing wrote:
drivers/cpufreq/qcom-cpufreq-hw.c: In function
qcom_cpufreq_hw_read_lut:
drivers/cpufreq/qcom-cpufreq-hw.c:89:38: warning:
variable prev_cc set but not used
On Wed, 21 Aug 2019, Michel Lespinasse wrote:
On Tue, Aug 13, 2019 at 03:46:18PM -0700, Davidlohr Bueso wrote:
o The border cases for overlapping differ -- interval trees are closed,
while memtype intervals are open. We need to maintain semantics such
that conflict detection and getting the
You already know the location of the source tree without this message.
Signed-off-by: Masahiro Yamada
---
Makefile | 1 -
1 file changed, 1 deletion(-)
diff --git a/Makefile b/Makefile
index 7e54a821b4b0..a77102e4ee90 100644
--- a/Makefile
+++ b/Makefile
@@ -1118,7 +1118,6 @@ PHONY += prepare
If you try out-of-tree build with an unclean source tree, Kbuild
suggests to run make mrproper. The path to the source tree may be
shown with a relative path, for example, "make O=foo" emits the
following:
.. is not clean, please run 'make mrproper'
in the '..' directory.
This is somewhat
Now prepare3 does nothing but depends on include/config/kernel.release
Signed-off-by: Masahiro Yamada
---
Makefile | 10 --
1 file changed, 4 insertions(+), 6 deletions(-)
diff --git a/Makefile b/Makefile
index ca6851f5ebc9..960df4d35b15 100644
--- a/Makefile
+++ b/Makefile
@@
With this commit, the error report is shown earlier, even before
running kconfig.
Signed-off-by: Masahiro Yamada
---
Makefile | 24 ++--
1 file changed, 10 insertions(+), 14 deletions(-)
diff --git a/Makefile b/Makefile
index 901fcb8fffbe..ca6851f5ebc9 100644
---
Move the outputmakefile target to the leftmost in the prerequisite list
so that this is checked first. There is no guarantee that Make runs the
prerequisites from left to right, but at least the released versions of
GNU Make work like that when the parallel build option is not given.
Of course,
Since commit 3a475b2166fd ("kbuild: Inform user to pass ARCH= for make
mrproper"), if you try out-of-tree build with an unclean source tree,
it suggests to run 'make ARCH= mrproper'.
This looks odd when you are not cross-compiling the kernel. Show the
'ARCH=' part only when ARCH= was given from
On Thu, Aug 22, 2019 at 04:01:24AM +, Atish Patra wrote:
> The downside of this is that for every !cmask case in true SMP (more
> common probably) it will execute 2 extra cpumask instructions. As
> tlbflush path is in performance critical path, I think we should favor
> more common case (SMP
On Thu, 2019-08-22 at 03:51 +0200, Christoph Hellwig wrote:
> On Wed, Aug 21, 2019 at 05:46:44PM -0700, Atish Patra wrote:
> > SBI calls are expensive. If cpumask is empty, there is no need to
> > trap via SBI as no remote tlb flushing is required.
> >
> > Signed-off-by: Atish Patra
> > ---
> >
[trimmed Cc list a bit]
On Thu, Aug 15, 2019 at 07:46:56PM -0700, Eric Biggers wrote:
> On Sat, Jul 20, 2019 at 07:29:49AM +0800, Yin Fengwei wrote:
> > syzbot reported general protection fault in kstrtouint:
> > https://lkml.org/lkml/2019/7/18/328
> >
> > From the log, if the mount option is
> -Original Message-
> From: linux-hyperv-ow...@vger.kernel.org ow...@vger.kernel.org> On Behalf Of David Miller
> Sent: Wednesday, August 21, 2019 9:09 PM
> To: Haiyang Zhang
> Cc: sas...@kernel.org; sae...@mellanox.com; l...@kernel.org;
> era...@mellanox.com;
From: Haiyang Zhang
Date: Wed, 21 Aug 2019 00:23:19 +
> This patch set adds paravirtual backchannel in software in pci_hyperv,
> which is required by the mlx5e driver HV VHCA stats agent.
>
> The stats agent is responsible on running a periodic rx/tx packets/bytes
> stats update.
These
On Thu, Aug 22, 2019 at 11:45:34AM +0800, Chester Lin wrote:
> adjust_lowmem_bounds() checks every memblocks in order to find the boundary
> between lowmem and highmem. However some memblocks could be marked as NOMAP
> so they are not used by kernel, which should be skipped while calculating
> the
From: Anders Roxell
Date: Tue, 20 Aug 2019 15:41:02 +0200
> When running xfrm_policy.sh we see the following
>
> # sysctl cannot stat /proc/sys/net/ipv4/conf/eth1/forwarding No such file or
> directory
> cannot: stat_/proc/sys/net/ipv4/conf/eth1/forwarding #
I don't understand how a
On Thu, 2019-08-22 at 03:50 +0200, Christoph Hellwig wrote:
> On Wed, Aug 21, 2019 at 05:46:43PM -0700, Atish Patra wrote:
> > + if (size <= PAGE_SIZE && size != -1)
> > + local_flush_tlb_page(start);
> > + else
> > + local_flush_tlb_all();
>
On Thu, 2019-08-22 at 03:46 +0200, Christoph Hellwig wrote:
> On Wed, Aug 21, 2019 at 05:46:42PM -0700, Atish Patra wrote:
> > In RISC-V, tlb flush happens via SBI which is expensive. If the
> > local
> > cpu is the only cpu in cpumask, there is no need to invoke a SBI
> > call.
> >
> > Just do a
Both relative path and absolute path have pros and cons. For example,
we can move the source and objtree around together by using the
relative path to the source tree.
Do not force the absolute path to the source tree. If you prefer the
absolute path, you can specify KBUILD_ABS_SRCTREE=1.
From: Hubert Feurstein
Date: Tue, 20 Aug 2019 10:48:29 +0200
> From: Hubert Feurstein
>
> Changelog:
> v3: mv88e6xxx_smi_indirect_write: forward ptp_sts only on the last write
> Copied Miroslav Lichvar because of PTP offset compensation patch
> v2: Added patch for PTP offset
On Wed, Aug 21, 2019 at 08:02:00PM -0700, Paul E. McKenney wrote:
> On Thu, Aug 22, 2019 at 10:53:43AM +0900, Christoph Hellwig wrote:
> > The kbuild reported a built failure due to a header loop when RCUTINY is
> > enabled with my pending riscv-nommu port. Switch rcutiny.h to only
> > include
adjust_lowmem_bounds() checks every memblocks in order to find the boundary
between lowmem and highmem. However some memblocks could be marked as NOMAP
so they are not used by kernel, which should be skipped while calculating
the boundary.
Signed-off-by: Chester Lin
---
arch/arm/mm/mmu.c | 3
Hi Thomas,
On 22/8/2019 12:47 AM, Andy Shevchenko wrote:
For DT we can actually avoid that completely. See below.
For ACPI not unfortunately as the stupid GSI mapping is hard coded.
The below works better for my case, so, if you are going with that
Tested-by: Andy Shevchenko
On Wed, Aug 21, 2019 at 4:51 PM Jason Gunthorpe wrote:
>
> On Wed, Aug 21, 2019 at 01:24:20PM -0300, Jason Gunthorpe wrote:
> > On Tue, Aug 20, 2019 at 07:58:22PM -0700, Dan Williams wrote:
> > > On Tue, Aug 20, 2019 at 6:27 AM Jason Gunthorpe wrote:
> > > >
> > > > On Mon, Aug 19, 2019 at
Hi
On 2019/08/22 2:54, Dmitry Torokhov wrote:
On Wed, Aug 21, 2019 at 10:26:31PM +0900, Jiada Wang wrote:
Hi Dmitry
On 2019/08/17 2:16, Dmitry Torokhov wrote:
On Fri, Aug 16, 2019 at 05:28:52PM +0900, Jiada Wang wrote:
From: Nick Dyer
The workaround of reading all messages until an
> > +What: /sys/kernel/debug/hyperv//fuzz_test_state
> > +Date: August 2019
> > +KernelVersion: 5.3
> > +Contact:Branden Bonaby
> > +Description:Fuzz testing status of a vmbus device, whether its in an ON
> > +state or a OFF state
>
> Document
Hi Joe,
On Tue, Aug 20, 2019 at 1:02 AM Joe Lawrence wrote:
>
> On 8/19/19 3:31 AM, Miroslav Benes wrote:
> > On Mon, 19 Aug 2019, Masahiro Yamada wrote:
> >
> >>
> >> I can review this series from the build system point of view,
> >> but I am not familiar enough with live-patching itself.
> >>
vfio_pci_enable() saves the device's initial configuration information
with the intent that it is restored in vfio_pci_disable(). However,
the commit referenced in Fixes: below replaced the call to
__pci_reset_function_locked(), which is not wrapped in a state save
and restore, with
From: Florian Fainelli
Date: Wed, 21 Aug 2019 17:07:46 -0700
> The SF2 binding does not specify that the CPU port should have
> properties mandatory for successfully instantiating a PHYLINK object. As
> such, there will be missing properties (including fixed-link) and when
> attempting to
On 8/21/19 6:38 PM, syzbot wrote:
> ==
> BUG: KMSAN: uninit-value in rtm_to_nh_config net/ipv4/nexthop.c:1317
> [inline]
> BUG: KMSAN: uninit-value in rtm_new_nexthop+0x447/0x98e0
> net/ipv4/nexthop.c:1474
I believed this is fixed in
On Wed, Aug 21, 2019 at 10:46 AM Nathan Huckleberry wrote:
>
> The stackframe setup when compiled with clang is different.
> Since the stack unwinder expects the gcc stackframe setup it
> fails to print backtraces. This patch adds support for the
> clang stackframe setup.
>
> Link:
If CONFIG_GPIOLIB is not, gpiochip_lock/unlock_as_irq will
conflict as this:
In file included from sound/soc/codecs/wm5100.c:18:0:
./include/linux/gpio.h:224:19: error: static declaration of
gpiochip_lock_as_irq follows non-static declaration
static inline int gpiochip_lock_as_irq(struct
On Thu, Aug 22, 2019 at 01:36:09AM +, Harry Zhang wrote:
> Tool function issues: Please validate args errors for '-p' and '--path',
> in or following validate_args_path().
>
> Comments of functionality:
> - it's confusing when fuzz_testing are all OFF, then user run ' python3
>
Thanks for reporting. I will send out fix patch.
-Original Message-
From: kbuild test robot
Sent: Thursday, August 22, 2019 10:25 AM
To: Tianyu Lan
Cc: kbuild-...@01.org; linux-kernel@vger.kernel.org; tipbu...@zytor.com; Thomas
Gleixner ; Michael Kelley
Subject: [tip:timers/core
On Mon, Aug 19, 2019 at 04:02:21PM +0200, Thomas Gleixner wrote:
> On Mon, 19 Aug 2019, Ming Lei wrote:
> > On Mon, Aug 19, 2019 at 03:13:58PM +0200, Thomas Gleixner wrote:
> > > On Mon, 19 Aug 2019, Ming Lei wrote:
> > >
> > > > Cc: Jon Derrick
> > > > Cc: Jens Axboe
> > > > Reported-by: Jon
On Mon, Aug 19, 2019 at 05:22:07AM -0700, syzbot wrote:
> Hello,
>
> syzbot found the following crash on:
>
> HEAD commit:5181b473 net: phy: realtek: add NBase-T PHY auto-detection
> git tree: net-next
> console output: https://syzkaller.appspot.com/x/log.txt?x=156b731c60
> kernel
From: Guo Ren
610 has vipt aliasing issue, so we need to finish the cache flush
apis mentioned in cachetlb.rst to avoid data corruption.
Here is the list of modified apis in the patch:
- flush_kernel_dcache_page (new add)
- flush_dcache_mmap_lock(new add)
-
On Thu, Aug 22, 2019 at 10:53:43AM +0900, Christoph Hellwig wrote:
> The kbuild reported a built failure due to a header loop when RCUTINY is
> enabled with my pending riscv-nommu port. Switch rcutiny.h to only
> include the minimal required header to get HZ instead.
>
> Signed-off-by: Christoph
Add the watchdog bindings for Freescale i.MX7ULP.
Signed-off-by: Anson Huang
---
No changes.
---
.../bindings/watchdog/fsl-imx7ulp-wdt.txt | 22 ++
1 file changed, 22 insertions(+)
create mode 100644
Documentation/devicetree/bindings/watchdog/fsl-imx7ulp-wdt.txt
The i.MX7ULP Watchdog Timer (WDOG) module is an independent timer
that is available for system use.
It provides a safety feature to ensure that software is executing
as planned and that the CPU is not stuck in an infinite loop or
executing unintended code. If the WDOG module is not serviced
Add wdog1 node to support watchdog driver.
Signed-off-by: Anson Huang
---
No changes.
---
arch/arm/boot/dts/imx7ulp.dtsi | 10 ++
1 file changed, 10 insertions(+)
diff --git a/arch/arm/boot/dts/imx7ulp.dtsi b/arch/arm/boot/dts/imx7ulp.dtsi
index 6859a3a..1fdb5a35 100644
---
Select CONFIG_IMX7ULP_WDT by default to support i.MX7ULP watchdog.
Signed-off-by: Anson Huang
---
No changes.
---
arch/arm/configs/imx_v6_v7_defconfig | 1 +
1 file changed, 1 insertion(+)
diff --git a/arch/arm/configs/imx_v6_v7_defconfig
b/arch/arm/configs/imx_v6_v7_defconfig
index
Hi, Guenter
> On Tue, Aug 20, 2019 at 10:07:56PM -0400, Anson Huang wrote:
> > The i.MX7ULP Watchdog Timer (WDOG) module is an independent timer
> that
> > is available for system use.
> > It provides a safety feature to ensure that software is executing as
> > planned and that the CPU is not
On 21-08-19, 16:16, Florian Fainelli wrote:
> Instead of printing the policy, which is incidentally a kernel pointer,
> so with limited interest, print the cpufreq driver name that failed to
> be suspend, which is more useful for debugging.
>
> Fixes: 2f0aea936360 ("cpufreq: suspend governors on
On 21-08-19, 20:14, YueHaibing wrote:
> drivers/cpufreq/qcom-cpufreq-hw.c: In function qcom_cpufreq_hw_read_lut:
> drivers/cpufreq/qcom-cpufreq-hw.c:89:38: warning:
> variable prev_cc set but not used [-Wunused-but-set-variable]
>
> It is not used since commit 3003e75a5045 ("cpufreq:
> qcom-hw:
ping,
On Thu, 15 Aug 2019 at 12:03, Wanpeng Li wrote:
>
> From: Wanpeng Li
>
> Even if for realtime CPUs, cache line bounces, frequency scaling, presence
> of higher-priority RT tasks, etc can still cause different response. These
> interferences should be considered and periodically revaluate
Hi Helen,
On Wed, Aug 21, 2019 at 06:46:15PM -0300, Helen Koike wrote:
> On 8/15/19 2:54 PM, Laurent Pinchart wrote:
> > On Wed, Aug 07, 2019 at 10:37:55AM -0300, Helen Koike wrote:
> >> On 8/7/19 10:05 AM, Sakari Ailus wrote:
> >>> On Tue, Jul 30, 2019 at 03:42:46PM -0300, Helen Koike wrote:
>
On Wed, Aug 21, 2019 at 10:43 AM Nathan Huckleberry wrote:
>
> On Tue, Aug 20, 2019 at 2:39 PM Nick Desaulniers
> wrote:
> >
> > On Tue, Aug 20, 2019 at 12:44 PM Nathan Huckleberry
> > wrote:
...snip...
> > > +tstr1, #0x10 @ 26 or 32-bit mode?
> > > +moveq mask, #0xfc03
Sorry to resend the email because of non-plain text issues.
On Wed, Aug 21, 2019 at 8:30 PM Adrian Hunter wrote:
>
> On 20/08/19 5:07 AM, Ben Chuang wrote:
> > From: Ben Chuang
> >
> > Add support for the GL9750 and GL9755 chipsets.
> >
> > The patches enable v4 mode and wait 5ms after set 1.8V
tree: https://kernel.googlesource.com/pub/scm/linux/kernel/git/tip/tip.git
timers/core
head: b74e1d61dbc614ff35ef3ad9267c61ed06b09051
commit: b74e1d61dbc614ff35ef3ad9267c61ed06b09051 [34/34] clocksource/hyperv:
Add Hyper-V specific sched clock function
config: i386-randconfig-g002-201933
On Thu, Aug 22, 2019 at 10:00 AM Keith Busch wrote:
>
> On Wed, Aug 21, 2019 at 7:34 PM Ming Lei wrote:
> > On Wed, Aug 21, 2019 at 04:27:00PM +, Long Li wrote:
> > > Here is the command to benchmark it:
> > >
> > > fio --bs=4k --ioengine=libaio --iodepth=128
> > >
Since only the root cfs_rq runnable_load_avg field is used any more,
we can move the field from struct sched_avg, which every sched_entity
has one of, directly into the struct cfs_rq, of which we have way fewer.
No functional changes.
Suggested-by: Dietmar Eggemann
Signed-off-by: Rik van Riel
The code in update_cfs_group / calc_group_shares has some logic to
quickly ramp up the load when a task has just started running in a
cgroup, in order to get sane values for the cgroup se->load.weight.
This code adds a similar hack to task_se_h_weight.
However, THIS CODE IS WRONG, since it does
Use an explicit "cfs_rq of parent sched_entity" helper in a few
strategic places, where cfs_rq_of(se) may no longer point at the
right runqueue once we flatten the hierarchical cgroup runqueues.
No functional change.
Signed-off-by: Rik van Riel
---
kernel/sched/fair.c | 17 +
1
LS1028A has a clock domain PXLCLK0 used for provide pixel clocks to Display
output interface. Add a YAML schema for this.
Signed-off-by: Wen He
Reviewed-by: Rob Herring
---
.../devicetree/bindings/clock/fsl,plldig.yaml | 43 +++
1 file changed, 43 insertions(+)
create mode
Add clock driver for QorIQ LS1028A Display output interfaces(LCD, DPHY),
as implemented in TSMC CLN28HPM PLL, this PLL supports the programmable
integer division and range of the display output pixel clock's 27-594MHz.
Signed-off-by: Wen He
---
change in v3:
- remove the OF dependency
Make it clear that update_curr only works on tasks any more.
There is no need for task_tick_fair to call it on every sched entity up
the hierarchy, so move the call out of entity_tick.
Signed-off-by: Rik van Riel `
Signed-off-by: Rik van Riel
---
kernel/sched/fair.c | 24
Add helper functions to make the flattened runqueue patch a little smaller.
The task_se_h_weight function is similar to task_se_h_load, but scales the
task weight by the group weight, without taking the task's duty cycle into
account.
The task_se_in_cgroup helper is functionally identical to
When a task wakes back up after having gone to sleep, place_entity
will limit the vruntime difference between min_vruntime and the
woken up task to half of sysctl_sched_latency.
The code in wakeup_preempt_entity calculates how much vruntime a
time slice for the woken up task represents, in
Reducing the overhead of the CPU controller is achieved by not walking
all the sched_entities every time a task is enqueued or dequeued.
One of the things being checked every single time is whether the cfs_rq
is on the rq->leaf_cfs_rq_list.
By only removing a cfs_rq from the list once it no
Now that enqueue_task_fair and dequeue_task_fair no longer iterate up
the hierarchy all the time, a method to lazily propagate sum_exec_runtime
up the hierarchy is necessary.
Once a tick, propagate the newly accumulated exec_runtime up the hierarchy,
and feed it into CFS bandwidth control.
The idea behind __sched_period makes sense, but the results do not always.
When a CPU has one high priority task and a large number of low priority
tasks, __sched_period will return a value larger than sysctl_sched_latency,
and the one high priority task may end up getting a timeslice all for
The way the time slice length is currently calculated, not only do high
priority tasks get longer time slices than low priority tasks, but due
to fixed point math, low priority tasks could end up with a zero length
time slice. This can lead to cache thrashing and other inefficiencies.
Cap the
The runnable_load magic is used to quickly propagate information about
runnable tasks up the hierarchy of runqueues. The runnable_load_avg is
mostly used for the load balancing code, which only examines the value at
the root cfs_rq.
Redefine the root cfs_rq runnable_load_avg to be the sum of
Remove some fields from /proc/sched_debug that are removed from
sched_entity in a subsequent patch, and add h_load, which comes in
very handy to debug CPU controller weight distribution.
Signed-off-by: Rik van Riel
Reviewed-by: Josef Bacik
---
kernel/sched/debug.c | 11 ++-
1 file
Flatten the hierarchical runqueues into just the per CPU rq.cfs runqueue.
Iteration of the sched_entity hierarchy is rate limited to once per jiffy
per sched_entity, which is a smaller change than it seems, because load
average adjustments were already rate limited to once per jiffy before this
Sometimes the hierarchical load of a sched_entity needs to be calculated.
Rename task_h_load to task_se_h_load, and directly pass a sched_entity to
that function.
Move the function declaration up above where it will be used later.
No functional changes.
Signed-off-by: Rik van Riel
Reviewed-by:
The current implementation of the CPU controller uses hierarchical
runqueues, where on wakeup a task is enqueued on its group's runqueue,
the group is enqueued on the runqueue of the group above it, etc.
This increases a fairly large amount of overhead for workloads that
do a lot of wakeups a
Refactor enqueue_entity, dequeue_entity, and update_load_avg, in order
to split out the things we still want to happen at every level in the
cgroup hierarchy with a flat runqueue from the things we only need to
happen once.
No functional changes.
Signed-off-by: Rik van Riel
---
On Wed, Aug 21, 2019 at 7:34 PM Ming Lei wrote:
> On Wed, Aug 21, 2019 at 04:27:00PM +, Long Li wrote:
> > Here is the command to benchmark it:
> >
> > fio --bs=4k --ioengine=libaio --iodepth=128
> >
The kbuild reported a built failure due to a header loop when RCUTINY is
enabled with my pending riscv-nommu port. Switch rcutiny.h to only
include the minimal required header to get HZ instead.
Signed-off-by: Christoph Hellwig
---
include/linux/rcutiny.h | 2 +-
1 file changed, 1
> On Aug 21, 2019, at 9:31 PM, Baoquan He wrote:
>
> On 08/21/19 at 05:12pm, Qian Cai wrote:
Does disabling CONFIG_RANDOMIZE_BASE help? Maybe that workaround has
regressed. Effectively we need to find what is causing the kernel to
sometimes be placed in the middle of a custom
On Wed, Aug 21, 2019 at 05:46:44PM -0700, Atish Patra wrote:
> SBI calls are expensive. If cpumask is empty, there is no need to
> trap via SBI as no remote tlb flushing is required.
>
> Signed-off-by: Atish Patra
> ---
> arch/riscv/mm/tlbflush.c | 3 +++
> 1 file changed, 3 insertions(+)
>
>
On Wed, Aug 21, 2019 at 05:46:43PM -0700, Atish Patra wrote:
> + if (size <= PAGE_SIZE && size != -1)
> + local_flush_tlb_page(start);
> + else
> + local_flush_tlb_all();
As Andreas pointed out (unsigned long)-1 is actually larger
On Wed, Aug 21, 2019 at 05:46:42PM -0700, Atish Patra wrote:
> In RISC-V, tlb flush happens via SBI which is expensive. If the local
> cpu is the only cpu in cpumask, there is no need to invoke a SBI call.
>
> Just do a local flush and return.
>
> Signed-off-by: Atish Patra
> ---
>
On 2019/8/21 20:38, James Clark wrote:
> Hi,
>
> I also had a look at this and had a question about the --spe option.
> It seems that whatever options I give it, the output is the same:
>
> perf report
> And
> perf report --spe=t
>
> Both give the same result:
>
> # Samples:
Tool function issues: Please validate args errors for '-p' and '--path', in
or following validate_args_path().
Comments of functionality:
- it's confusing when fuzz_testing are all OFF, then user run ' python3
/home/lisa/vmbus_testing -p
On Wed, Aug 21, 2019 at 04:27:00PM +, Long Li wrote:
> >>>Subject: Re: [PATCH 0/3] fix interrupt swamp in NVMe
> >>>
> >>>On Wed, Aug 21, 2019 at 07:47:44AM +, Long Li wrote:
> >>>Subject: Re: [PATCH 0/3] fix interrupt swamp in NVMe
> >>>
> >>>On 20/08/2019 09:25, Ming Lei
On 08/21/19 at 05:12pm, Qian Cai wrote:
> > > Does disabling CONFIG_RANDOMIZE_BASE help? Maybe that workaround has
> > > regressed. Effectively we need to find what is causing the kernel to
> > > sometimes be placed in the middle of a custom reserved memmap= range.
> >
> > Yes, disabling KASLR
From: Marco Hartman Sent: Wednesday, August 21, 2019 7:44 PM
> IEEE 802.3ae clause 45 defines a modified MDIO protocol that uses a two
> staged access model in order to increase the address space.
>
> This patch adds support for C45 MDIO read and write accesses, which are
> used whenever the
On Wed, 2019-08-21 at 20:52 +0300, Andrey Ryabinin wrote:
>
> On 8/20/19 8:37 AM, Walter Wu wrote:
> > On Tue, 2019-08-06 at 13:43 +0800, Walter Wu wrote:
> >> This patch adds memory corruption identification at bug report for
> >> software tag-based mode, the report show whether it is
On Thu, Aug 22, 2019 at 03:02:04AM +0200, Frederic Weisbecker wrote:
> > which repeats every time I fetch. I can't of anythign particular on
> > my side that would cause this.
>
> Yeah I had to run "git remote prune tip" and fetch again.
>
> Apparently there was an old remote branch
1 - 100 of 998 matches
Mail list logo