Re: [PATCH 2/2 v2] sched: use load_avg for selecting idlest group

On Fri, 02 Dec, at 04:20:54PM, Vincent Guittot wrote: > > Matt, > > Have you been able to get some results for the patchset ? Sorry, I messed up the test config and the tests never ran :( I've redone everything and fast-tracked the results on the SUSE grid now. Should have results by morning.

Re: [PATCH 2/2 v2] sched: use load_avg for selecting idlest group

On Fri, 02 Dec, at 04:20:54PM, Vincent Guittot wrote: > > Matt, > > Have you been able to get some results for the patchset ? Sorry, I messed up the test config and the tests never ran :( I've redone everything and fast-tracked the results on the SUSE grid now. Should have results by morning.

[PATCH V2] x86/intel_rdt: Implement show_options() for resctrlfs

Implement show_options() callback for intel resource control filesystem to expose the active mount options in /proc/ This applies to tip tree x86/cache branch. Cc: Fenghua Yu Cc: Thomas Gleixner Signed-off-by: Shaohua Li ---

[PATCH V2] x86/intel_rdt: Implement show_options() for resctrlfs

Implement show_options() callback for intel resource control filesystem to expose the active mount options in /proc/ This applies to tip tree x86/cache branch. Cc: Fenghua Yu Cc: Thomas Gleixner Signed-off-by: Shaohua Li --- arch/x86/kernel/cpu/intel_rdt_rdtgroup.c | 12 ++-- 1 file

Re: [PATCH v2] intelrdt: resctrl: recommend locking for resctrlfs

On Fri, Dec 02, 2016 at 12:20:29PM +0100, Thomas Gleixner wrote: > On Thu, 1 Dec 2016, Marcelo Tosatti wrote: > > > > There is a locking problem between different applications > > reading/writing to resctrlfs directory at the same time (read the patch > > below for details). > > > > Suggest a

Re: [PATCH v2] intelrdt: resctrl: recommend locking for resctrlfs

On Fri, Dec 02, 2016 at 12:20:29PM +0100, Thomas Gleixner wrote: > On Thu, 1 Dec 2016, Marcelo Tosatti wrote: > > > > There is a locking problem between different applications > > reading/writing to resctrlfs directory at the same time (read the patch > > below for details). > > > > Suggest a

[PATCH v2] rpmsg: qcom_smd: Add support for "label" property

Add support for the "label" property, used to give the edge a name other than the one of the DT node. This allows the implementor to provide consistently named edges when using the rpmsg character device. Signed-off-by: Bjorn Andersson --- Changes since v1: - Moved

[PATCH v2] rpmsg: qcom_smd: Add support for "label" property

Add support for the "label" property, used to give the edge a name other than the one of the DT node. This allows the implementor to provide consistently named edges when using the rpmsg character device. Signed-off-by: Bjorn Andersson --- Changes since v1: - Moved sysfs attribute to the

[PATCH v2 3/4] rpmsg: Driver for user space endpoint interface

This driver allows rpmsg instances to expose access to rpmsg endpoints to user space processes. It provides a control interface, allowing userspace to export endpoints and an endpoint interface for each exposed endpoint. The implementation is based on prior art by Texas Instrument, Google,

[PATCH v2 3/4] rpmsg: Driver for user space endpoint interface

This driver allows rpmsg instances to expose access to rpmsg endpoints to user space processes. It provides a control interface, allowing userspace to export endpoints and an endpoint interface for each exposed endpoint. The implementation is based on prior art by Texas Instrument, Google,

[PATCH v2 4/4] rpmsg: smd: Register rpmsg user space interface for edges

Create and register a rpmsg device for use with the rpmsg user space interface, allowing user space to access SMD channels. Also provide the "rpmsg_name" device attribute to expose the edge name in sysfs, allowing the user to write udev rules for specific rpmsg devices and their children.

[PATCH v2 4/4] rpmsg: smd: Register rpmsg user space interface for edges

Create and register a rpmsg device for use with the rpmsg user space interface, allowing user space to access SMD channels. Also provide the "rpmsg_name" device attribute to expose the edge name in sysfs, allowing the user to write udev rules for specific rpmsg devices and their children.

[PATCH v2 1/4] rpmsg: Introduce "poll" to endpoint ops

This allows rpmsg backends to implement polling of the outgoing buffer, which provides poll support to user space when using the rpmsg character device. Signed-off-by: Bjorn Andersson --- Changes since v1: - New patch drivers/rpmsg/rpmsg_core.c | 20

[PATCH v2 2/4] rpmsg: qcom_smd: Implement endpoint "poll"

Add support for polling the status of the write buffer so that user space can use rpmsg character devices in non-blocking mode. Signed-off-by: Bjorn Andersson --- Changes since v1: - New patch drivers/rpmsg/qcom_smd.c | 16 1 file changed, 16

[PATCH v2 1/4] rpmsg: Introduce "poll" to endpoint ops

This allows rpmsg backends to implement polling of the outgoing buffer, which provides poll support to user space when using the rpmsg character device. Signed-off-by: Bjorn Andersson --- Changes since v1: - New patch drivers/rpmsg/rpmsg_core.c | 20

[PATCH v2 2/4] rpmsg: qcom_smd: Implement endpoint "poll"

Add support for polling the status of the write buffer so that user space can use rpmsg character devices in non-blocking mode. Signed-off-by: Bjorn Andersson --- Changes since v1: - New patch drivers/rpmsg/qcom_smd.c | 16 1 file changed, 16 insertions(+) diff --git

[PATCH net-next] tools: hv: Enable network manager for bonding scripts on RHEL

From: Haiyang Zhang We found network manager is necessary on RHEL to make the synthetic NIC, VF NIC bonding operations handled automatically. So, enabling network manager here. Signed-off-by: Haiyang Zhang Reviewed-by: K. Y. Srinivasan

[PATCH net-next] tools: hv: Enable network manager for bonding scripts on RHEL

From: Haiyang Zhang We found network manager is necessary on RHEL to make the synthetic NIC, VF NIC bonding operations handled automatically. So, enabling network manager here. Signed-off-by: Haiyang Zhang Reviewed-by: K. Y. Srinivasan --- tools/hv/bondvf.sh |4 ++-- 1 files changed, 2

Re: [PATCH] audit: remove the audit freelist

On Thu, Dec 1, 2016 at 7:09 PM, Florian Westphal wrote: > Paul Moore wrote: >> On Wed, Nov 30, 2016 at 8:44 PM, Florian Westphal wrote: >> > Paul Moore wrote: >> >> On Tue, Nov 15, 2016 at 8:16 AM, Florian Westphal

Re: [PATCH] audit: remove the audit freelist

On Thu, Dec 1, 2016 at 7:09 PM, Florian Westphal wrote: > Paul Moore wrote: >> On Wed, Nov 30, 2016 at 8:44 PM, Florian Westphal wrote: >> > Paul Moore wrote: >> >> On Tue, Nov 15, 2016 at 8:16 AM, Florian Westphal wrote: >> >> > allows better debugging as freeing audit buffers now always

Re: [PATCH v4] x86/suspend: fix false positive KASAN warning on suspend/resume

On Fri, Dec 02, 2016 at 10:09:03PM +0100, Pavel Machek wrote: > On Fri 2016-12-02 11:42:21, Josh Poimboeuf wrote: > > Resuming from a suspend operation is showing a KASAN false positive > > warning: > > > > > > Reported-by: Scott Bauer > > Signed-off-by: Josh Poimboeuf

Re: [PATCH v4] x86/suspend: fix false positive KASAN warning on suspend/resume

On Fri, Dec 02, 2016 at 10:09:03PM +0100, Pavel Machek wrote: > On Fri 2016-12-02 11:42:21, Josh Poimboeuf wrote: > > Resuming from a suspend operation is showing a KASAN false positive > > warning: > > > > > > Reported-by: Scott Bauer > > Signed-off-by: Josh Poimboeuf > > Acked-by: Pavel

Re: [GIT PULL 1/3] ARM: exynos: Soc/mach for v4.10

On Thursday, December 1, 2016 8:34:04 PM CET Krzysztof Kozlowski wrote: > On Thu, Nov 24, 2016 at 08:08:27AM +0200, Krzysztof Kozlowski wrote: > > Hi, > > > > This contains previous dts branch because SCU node in dts is needed > > prior to removing it from mach code. > > > > Below you will find

Re: [GIT PULL 1/3] ARM: exynos: Soc/mach for v4.10

On Thursday, December 1, 2016 8:34:04 PM CET Krzysztof Kozlowski wrote: > On Thu, Nov 24, 2016 at 08:08:27AM +0200, Krzysztof Kozlowski wrote: > > Hi, > > > > This contains previous dts branch because SCU node in dts is needed > > prior to removing it from mach code. > > > > Below you will find

[PATCH] adm80211: add checks for dma mapping errors

The driver does not check if mapping dma memory succeed. The patch adds the checks and failure handling. Found by Linux Driver Verification project (linuxtesting.org). Signed-off-by: Alexey Khoroshilov --- drivers/net/wireless/admtek/adm8211.c | 24

[PATCH] adm80211: add checks for dma mapping errors

The driver does not check if mapping dma memory succeed. The patch adds the checks and failure handling. Found by Linux Driver Verification project (linuxtesting.org). Signed-off-by: Alexey Khoroshilov --- drivers/net/wireless/admtek/adm8211.c | 24 ++-- 1 file changed, 22

RE: [PATCH 02/15] hyperv: Add a function to detect hv_device

> -Original Message- > From: Greg KH [mailto:gre...@linuxfoundation.org] > Sent: Friday, December 2, 2016 8:03 AM > To: KY Srinivasan > Cc: o...@aepfle.de; jasow...@redhat.com; Haiyang Zhang > ; linux-kernel@vger.kernel.org; >

RE: [PATCH 02/15] hyperv: Add a function to detect hv_device

> -Original Message- > From: Greg KH [mailto:gre...@linuxfoundation.org] > Sent: Friday, December 2, 2016 8:03 AM > To: KY Srinivasan > Cc: o...@aepfle.de; jasow...@redhat.com; Haiyang Zhang > ; linux-kernel@vger.kernel.org; > a...@canonical.com; de...@linuxdriverproject.org; >

[PATCH v2 0/3] Add Broadcom SPU Crypto Driver

Change in v2: - select CRYPTO_DES in Kconfig The Broadcom SPU crypto driver provides access to SPU hardware for symmetric crypto offload. The driver supports ablkcipher, ahash, and aead operations. The driver supports several Broadcom SoCs with different revisions of the SPU hardware. The driver

[PATCH v2 0/3] Add Broadcom SPU Crypto Driver

Change in v2: - select CRYPTO_DES in Kconfig The Broadcom SPU crypto driver provides access to SPU hardware for symmetric crypto offload. The driver supports ablkcipher, ahash, and aead operations. The driver supports several Broadcom SoCs with different revisions of the SPU hardware. The driver

[PATCH v2 3/3] crypto: brcm: Add Broadcom SPU driver DT entry.

Add Northstar2 device tree entry for Broadcom Secure Processing Unit (SPU) crypto driver. Signed-off-by: Steve Lin Signed-off-by: Rob Rice --- arch/arm64/boot/dts/broadcom/ns2.dtsi | 12 1 file changed, 12 insertions(+) diff --git

[PATCH v2 1/3] crypto: brcm: DT documentation for Broadcom SPU driver

Device tree documentation for Broadcom Secure Processing Unit (SPU) crypto driver. Signed-off-by: Steve Lin Signed-off-by: Rob Rice --- .../devicetree/bindings/crypto/brcm,spu-crypto.txt | 25 ++ 1 file changed, 25

[PATCH v2 3/3] crypto: brcm: Add Broadcom SPU driver DT entry.

Add Northstar2 device tree entry for Broadcom Secure Processing Unit (SPU) crypto driver. Signed-off-by: Steve Lin Signed-off-by: Rob Rice --- arch/arm64/boot/dts/broadcom/ns2.dtsi | 12 1 file changed, 12 insertions(+) diff --git a/arch/arm64/boot/dts/broadcom/ns2.dtsi

[PATCH v2 1/3] crypto: brcm: DT documentation for Broadcom SPU driver

Device tree documentation for Broadcom Secure Processing Unit (SPU) crypto driver. Signed-off-by: Steve Lin Signed-off-by: Rob Rice --- .../devicetree/bindings/crypto/brcm,spu-crypto.txt | 25 ++ 1 file changed, 25 insertions(+) create mode 100644

Re: [PATCH v2] lkdtm: Add tests for LIST_POISON and ZERO_SIZE_PTR

On Thu, Dec 1, 2016 at 8:22 PM, Michael Ellerman wrote: > This adds two tests, to check that a read or write to LIST_POISON1 and > ZERO_SIZE_PTR are blocked. > > The default values for both (256 and 16) typically fall in the range > of valid user space addresses. However in

Re: [PATCH v2] lkdtm: Add tests for LIST_POISON and ZERO_SIZE_PTR

On Thu, Dec 1, 2016 at 8:22 PM, Michael Ellerman wrote: > This adds two tests, to check that a read or write to LIST_POISON1 and > ZERO_SIZE_PTR are blocked. > > The default values for both (256 and 16) typically fall in the range > of valid user space addresses. However in general mmap_min_addr

Re: [PATCH] tty: serial: fsl_lpuart: fix del_timer_sync() vs timer routine deadlock

>> Problem found via lockdep: >> >> - lpuart_set_termios() calls del_timer_sync(>lpuart_timer) while >> holding sport->port.lock >> >> - sport->lpuart_timer routine is lpuart_timer_func() that calls >> lpuart_copy_rx_to_tty() that acquires same lock. >> >> To fix, move Rx DMA stopping out of

Re: [PATCH] tty: serial: fsl_lpuart: fix del_timer_sync() vs timer routine deadlock

>> Problem found via lockdep: >> >> - lpuart_set_termios() calls del_timer_sync(>lpuart_timer) while >> holding sport->port.lock >> >> - sport->lpuart_timer routine is lpuart_timer_func() that calls >> lpuart_copy_rx_to_tty() that acquires same lock. >> >> To fix, move Rx DMA stopping out of

[PATCH V2 03/13] perf/x86: output sampling overhead

From: Kan Liang On x86, NMI handler is the most important part which brings overhead for sampling. Adding a pmu specific overhead type PERF_PMU_SAMPLE_OVERHEAD for it. For other architectures which may don't have NMI, the overhead type can be reused. Signed-off-by: Kan

[PATCH V2 13/13] perf report: warn on high overhead

From: Kan Liang Warning the user if the overhead is too high. The overhead evaluation metrics is as below. - Kerenl profiling cost evaluation MAX(cpu0_overhead, cpu1_overhead,...) / perf monotonic wall time If the rate is > 10%, a warning is printed. - perf record

[PATCH V2 12/13] perf tools: record user space profiling cost

From: Kan Liang Record the cpu time and elapsed time of perf record. Show them in perf report. # To display the perf.data header info, please use --header/--header-only options. # # # Elapsed time: 1199332708(ns) # Perf record cpu time: 197780038(ns) # #

[PATCH V2 03/13] perf/x86: output sampling overhead

From: Kan Liang On x86, NMI handler is the most important part which brings overhead for sampling. Adding a pmu specific overhead type PERF_PMU_SAMPLE_OVERHEAD for it. For other architectures which may don't have NMI, the overhead type can be reused. Signed-off-by: Kan Liang ---

[PATCH V2 13/13] perf report: warn on high overhead

From: Kan Liang Warning the user if the overhead is too high. The overhead evaluation metrics is as below. - Kerenl profiling cost evaluation MAX(cpu0_overhead, cpu1_overhead,...) / perf monotonic wall time If the rate is > 10%, a warning is printed. - perf record profiling cost evaluation

[PATCH V2 12/13] perf tools: record user space profiling cost

From: Kan Liang Record the cpu time and elapsed time of perf record. Show them in perf report. # To display the perf.data header info, please use --header/--header-only options. # # # Elapsed time: 1199332708(ns) # Perf record cpu time: 197780038(ns) # # CPU SAMSAM

[PATCH V2 00/13] export perf overheads information

From: Kan Liang Profiling brings additional overhead. High overhead may impacts the behavior of the profiling object, impacts the accuracy of the profiling result, and even hang the system. Currently, perf has dynamic interrupt throttle mechanism to lower the sample rate and

[PATCH] fpga: fix sparse warnings in fpga-mgr and fpga-bridge

Fix up these sparse warnings: drivers/fpga/fpga-mgr.c:189:21: warning: symbol '__fpga_mgr_get' was not declared. Should it be static? drivers/fpga/fpga-bridge.c:30:12: warning: symbol 'bridge_list_lock' was not declared. Should it be static? Signed-off-by: Dinh Nguyen ---

[GIT PULL] ARM: SoC fixes for v4.9

The following changes since commit 9c763584b7c8911106bb77af7e648bef09af9d80: Linux 4.9-rc6 (2016-11-20 13:52:19 -0800) are available in the git repository at: git://git.kernel.org/pub/scm/linux/kernel/git/arm/arm-soc.git tags/fixes-for-linus for you to fetch changes up to

[PATCH V2 00/13] export perf overheads information

From: Kan Liang Profiling brings additional overhead. High overhead may impacts the behavior of the profiling object, impacts the accuracy of the profiling result, and even hang the system. Currently, perf has dynamic interrupt throttle mechanism to lower the sample rate and overhead. But it has

[PATCH] fpga: fix sparse warnings in fpga-mgr and fpga-bridge

Fix up these sparse warnings: drivers/fpga/fpga-mgr.c:189:21: warning: symbol '__fpga_mgr_get' was not declared. Should it be static? drivers/fpga/fpga-bridge.c:30:12: warning: symbol 'bridge_list_lock' was not declared. Should it be static? Signed-off-by: Dinh Nguyen ---

[GIT PULL] ARM: SoC fixes for v4.9

The following changes since commit 9c763584b7c8911106bb77af7e648bef09af9d80: Linux 4.9-rc6 (2016-11-20 13:52:19 -0800) are available in the git repository at: git://git.kernel.org/pub/scm/linux/kernel/git/arm/arm-soc.git tags/fixes-for-linus for you to fetch changes up to

[PATCH V2 04/13] perf/core: output multiplexing overhead

From: Kan Liang Multiplexing overhead is one of the key overhead when the number of events is more than available counters. The multiplexing overhead PERF_CORE_MUX_OVERHEAD is a common overhead type. Signed-off-by: Kan Liang ---

[PATCH V2 02/13] perf/core: output overhead when sched out from context

From: Kan Liang Outputing every overhead when it happens is very costly. The accumulated time is more meaningful. So the overhead information should be outputted at the very end. The overhead information is outputted when task is scheduling out or the event is going to be

[PATCH V2 04/13] perf/core: output multiplexing overhead

From: Kan Liang Multiplexing overhead is one of the key overhead when the number of events is more than available counters. The multiplexing overhead PERF_CORE_MUX_OVERHEAD is a common overhead type. Signed-off-by: Kan Liang --- include/linux/perf_event.h | 2 ++

[PATCH V2 02/13] perf/core: output overhead when sched out from context

From: Kan Liang Outputing every overhead when it happens is very costly. The accumulated time is more meaningful. So the overhead information should be outputted at the very end. The overhead information is outputted when task is scheduling out or the event is going to be disabled. The arch

[PATCH V2 07/13] perf tools: handle PERF_RECORD_OVERHEAD record type

From: Kan Liang The infrastructure to handle PERF_RECORD_OVERHEAD record type. A new perf report option is introduced as a knob to show the profiling overhead information. The option also forces tty output. The overhead information is auxillary information, so it should be

[char-misc-next RESEND] mei: bus: enable non-blocking RX

From: Alexander Usyskin Enable non-blocking receive for drivers on mei bus, this allows checking for data availability by mei client drivers. This is most effective for fixed address clients, that lacks flow control. This function adds new API function

[PATCH V2 05/13] perf/core: output side-band events overhead

From: Kan Liang Iterating all events which need to receive side-band events also bring some overhead. The side-band events overhead PERF_CORE_SB_OVERHEAD is a common overhead type. Signed-off-by: Kan Liang --- include/linux/perf_event.h | 2 ++

[PATCH V2 10/13] perf tools: add time related functions

From: Kan Liang Make get_nsecs visible to collect monotonic wall time. Introduce get_vnsecs to collect CPU time. Signed-off-by: Kan Liang --- tools/perf/builtin-record.c | 9 + tools/perf/builtin-sched.c | 2 +- tools/perf/builtin.h|

[PATCH V2 08/13] perf tools: show kernel overhead

From: Kan Liang The kernel overhead include sample overhead, multiplexing overhead and side-band events overhead. All the information is printed in the head of output file if the show profiling cost option is applied. The result is sorted by CPU. If there is no available CPU

[PATCH V2 07/13] perf tools: handle PERF_RECORD_OVERHEAD record type

From: Kan Liang The infrastructure to handle PERF_RECORD_OVERHEAD record type. A new perf report option is introduced as a knob to show the profiling overhead information. The option also forces tty output. The overhead information is auxillary information, so it should be possible to access it

[char-misc-next RESEND] mei: bus: enable non-blocking RX

From: Alexander Usyskin Enable non-blocking receive for drivers on mei bus, this allows checking for data availability by mei client drivers. This is most effective for fixed address clients, that lacks flow control. This function adds new API function mei_cldev_recv_nonblock(), it retuns

[PATCH V2 05/13] perf/core: output side-band events overhead

From: Kan Liang Iterating all events which need to receive side-band events also bring some overhead. The side-band events overhead PERF_CORE_SB_OVERHEAD is a common overhead type. Signed-off-by: Kan Liang --- include/linux/perf_event.h | 2 ++ include/uapi/linux/perf_event.h | 1 +

[PATCH V2 10/13] perf tools: add time related functions

From: Kan Liang Make get_nsecs visible to collect monotonic wall time. Introduce get_vnsecs to collect CPU time. Signed-off-by: Kan Liang --- tools/perf/builtin-record.c | 9 + tools/perf/builtin-sched.c | 2 +- tools/perf/builtin.h| 2 ++ 3 files changed, 12 insertions(+), 1

[PATCH V2 08/13] perf tools: show kernel overhead

From: Kan Liang The kernel overhead include sample overhead, multiplexing overhead and side-band events overhead. All the information is printed in the head of output file if the show profiling cost option is applied. The result is sorted by CPU. If there is no available CPU information in

[PATCH V2 06/13] perf tools: option to disable overhead collection

From: Kan Liang Perf collects overhead information by default. A new option --no-profile-cost-info is introduced to disable the information collection. Signed-off-by: Kan Liang --- tools/include/uapi/linux/perf_event.h| 42

[PATCH V2 06/13] perf tools: option to disable overhead collection

From: Kan Liang Perf collects overhead information by default. A new option --no-profile-cost-info is introduced to disable the information collection. Signed-off-by: Kan Liang --- tools/include/uapi/linux/perf_event.h| 42 +++-

[PATCH V2 11/13] perf tools: introduce PERF_RECORD_USER_OVERHEAD

From: Kan Liang It is used to track user space perf tool profiling time cost. Signed-off-by: Kan Liang --- tools/perf/util/event.c | 1 + tools/perf/util/event.h | 5 + tools/perf/util/session.c | 4 3 files changed, 10 insertions(+)

[PATCH V2 11/13] perf tools: introduce PERF_RECORD_USER_OVERHEAD

From: Kan Liang It is used to track user space perf tool profiling time cost. Signed-off-by: Kan Liang --- tools/perf/util/event.c | 1 + tools/perf/util/event.h | 5 + tools/perf/util/session.c | 4 3 files changed, 10 insertions(+) diff --git a/tools/perf/util/event.c

[PATCH V2 09/13] perf script: show kernel overhead

From: Kan Liang Shows kernel overhead in perf script. The output is as below: perf script --show-profiling-cost-events perf 29001 79989.093958: 1 cycles: 81064ca6 native_write_msr (/lib/ sleep 29001 79989.094282: 7661 cycles:

[PATCH V2 01/13] perf/core: Introduce PERF_RECORD_OVERHEAD

From: Kan Liang A new perf record is introduced to export perf overhead information to userspace. So the user can measure the overhead of sampling directly. If the user doesn't want to use this feature, it can be switched off by configuring the user space tool. To output

[PATCH V2 09/13] perf script: show kernel overhead

From: Kan Liang Shows kernel overhead in perf script. The output is as below: perf script --show-profiling-cost-events perf 29001 79989.093958: 1 cycles: 81064ca6 native_write_msr (/lib/ sleep 29001 79989.094282: 7661 cycles: 810dc433

[PATCH V2 01/13] perf/core: Introduce PERF_RECORD_OVERHEAD

From: Kan Liang A new perf record is introduced to export perf overhead information to userspace. So the user can measure the overhead of sampling directly. If the user doesn't want to use this feature, it can be switched off by configuring the user space tool. To output the overhead

[PATCH v2 3/3] debugobjects: Reduce contention on the global pool_lock

On a large SMP system with many CPUs, the global pool_lock may become a performance bottleneck as all the CPUs that need to allocate or free debug objects have to take the lock. That can sometimes cause soft lockups like: NMI watchdog: BUG: soft lockup - CPU#35 stuck for 22s! [rcuos/1:21] ...

[PATCH v2 3/3] debugobjects: Reduce contention on the global pool_lock

On a large SMP system with many CPUs, the global pool_lock may become a performance bottleneck as all the CPUs that need to allocate or free debug objects have to take the lock. That can sometimes cause soft lockups like: NMI watchdog: BUG: soft lockup - CPU#35 stuck for 22s! [rcuos/1:21] ...

[PATCH v2 0/3] debugobjects: Reduce global pool_lock contention

v1->v2: - Move patch 2 in front of patch 1. - Fix merge conflict with linux-next. This patchset aims to reduce contention of the global pool_lock while improving performance at the same time. It is done to resolve the following soft lockup problem with a debug kernel in some of the large SMP

[PATCH v2 0/3] debugobjects: Reduce global pool_lock contention

v1->v2: - Move patch 2 in front of patch 1. - Fix merge conflict with linux-next. This patchset aims to reduce contention of the global pool_lock while improving performance at the same time. It is done to resolve the following soft lockup problem with a debug kernel in some of the large SMP

[PATCH v2 2/3] debugobjects: Scale thresholds with # of CPUs

On a large SMP systems with hundreds of CPUs, the current thresholds for allocating and freeing debug objects (256 and 1024 respectively) may not work well. This can cause a lot of needless calls to kmem_aloc() and kmem_free() on those systems. To alleviate this thrashing problem, the object

[PATCH v2 2/3] debugobjects: Scale thresholds with # of CPUs

On a large SMP systems with hundreds of CPUs, the current thresholds for allocating and freeing debug objects (256 and 1024 respectively) may not work well. This can cause a lot of needless calls to kmem_aloc() and kmem_free() on those systems. To alleviate this thrashing problem, the object

[PATCH v2 1/3] debugobjects: Track number of kmem_cache_alloc/kmem_cache_free done

New debugfs stat counters are added to track the numbers of kmem_cache_alloc() and kmem_cache_free() function calls to get a sense of how the internal debug objects cache management is performing. Signed-off-by: Waiman Long --- lib/debugobjects.c | 10 ++ 1 file

[PATCH v2 1/3] debugobjects: Track number of kmem_cache_alloc/kmem_cache_free done

New debugfs stat counters are added to track the numbers of kmem_cache_alloc() and kmem_cache_free() function calls to get a sense of how the internal debug objects cache management is performing. Signed-off-by: Waiman Long --- lib/debugobjects.c | 10 ++ 1 file changed, 10

Re: [PATCH v2 5/6] x86/xen: Add a Xen-specific sync_core() implementation

On Fri, Dec 2, 2016 at 12:41 PM, Andy Lutomirski wrote: > > Because, if so, we should maybe serialize whenever we migrate a > process to a different CPU. The intel docs are bad on this issue. Technically what we do could fall under the "cross-modifying code" case, where one

Re: [PATCH v2 5/6] x86/xen: Add a Xen-specific sync_core() implementation

On Fri, Dec 2, 2016 at 12:41 PM, Andy Lutomirski wrote: > > Because, if so, we should maybe serialize whenever we migrate a > process to a different CPU. The intel docs are bad on this issue. Technically what we do could fall under the "cross-modifying code" case, where one CPU does the write,

Re: [PATCH v4] x86/suspend: fix false positive KASAN warning on suspend/resume

On Fri 2016-12-02 11:42:21, Josh Poimboeuf wrote: > Resuming from a suspend operation is showing a KASAN false positive > warning: > > > Reported-by: Scott Bauer > Signed-off-by: Josh Poimboeuf Acked-by: Pavel Machek > diff --git

Re: [PATCH v4] x86/suspend: fix false positive KASAN warning on suspend/resume

On Fri 2016-12-02 11:42:21, Josh Poimboeuf wrote: > Resuming from a suspend operation is showing a KASAN false positive > warning: > > > Reported-by: Scott Bauer > Signed-off-by: Josh Poimboeuf Acked-by: Pavel Machek > diff --git a/mm/kasan/kasan.c b/mm/kasan/kasan.c > index

Re: Unkillable processes due to PTRACE_TRACEME again

On Fri 2016-12-02 19:48:40, Dmitry Vyukov wrote: > Hello, > > There was an issue discussed a year ago which leads to > unkillalble/unwaitable zombie processes due to PTRACE_TRACEME: > https://groups.google.com/d/msg/syzkaller/uGzwvhlCXAw/E-cfY2ejAgAJ > and I though it has been fixed by

Re: Unkillable processes due to PTRACE_TRACEME again

On Fri 2016-12-02 19:48:40, Dmitry Vyukov wrote: > Hello, > > There was an issue discussed a year ago which leads to > unkillalble/unwaitable zombie processes due to PTRACE_TRACEME: > https://groups.google.com/d/msg/syzkaller/uGzwvhlCXAw/E-cfY2ejAgAJ > and I though it has been fixed by

Re: [PATCH 2/8 linux-next] libnvdimm, namespace: add verbosity to persistence warning

On Fri, Dec 2, 2016 at 11:20 AM, Fabian Frederick wrote: > Signed-off-by: Fabian Frederick > --- > drivers/nvdimm/pmem.c | 2 +- > 1 file changed, 1 insertion(+), 1 deletion(-) > > diff --git a/drivers/nvdimm/pmem.c b/drivers/nvdimm/pmem.c > index

Re: [PATCH 2/8 linux-next] libnvdimm, namespace: add verbosity to persistence warning

On Fri, Dec 2, 2016 at 11:20 AM, Fabian Frederick wrote: > Signed-off-by: Fabian Frederick > --- > drivers/nvdimm/pmem.c | 2 +- > 1 file changed, 1 insertion(+), 1 deletion(-) > > diff --git a/drivers/nvdimm/pmem.c b/drivers/nvdimm/pmem.c > index 2461843..822a712 100644 > ---

Re: [PATCH 3/8 linux-next] libnvdimm, namespace: move persistence warning

On Fri, Dec 2, 2016 at 11:20 AM, Fabian Frederick wrote: > There's no need to warn on persistence yet as some validity tests > are on the way. > > Signed-off-by: Fabian Frederick > --- > drivers/nvdimm/pmem.c | 6 +++--- > 1 file changed, 3 insertions(+), 3

Re: [PATCH 3/8 linux-next] libnvdimm, namespace: move persistence warning

On Fri, Dec 2, 2016 at 11:20 AM, Fabian Frederick wrote: > There's no need to warn on persistence yet as some validity tests > are on the way. > > Signed-off-by: Fabian Frederick > --- > drivers/nvdimm/pmem.c | 6 +++--- > 1 file changed, 3 insertions(+), 3 deletions(-) > > diff --git

Re: [PATCH v2 5/6] x86/xen: Add a Xen-specific sync_core() implementation

On Fri, Dec 2, 2016 at 11:35 AM, Linus Torvalds wrote: > On Fri, Dec 2, 2016 at 11:30 AM, Andy Lutomirski wrote: >> >> How's this? > > Looks ok. I do think that > >> I suppose it could be an unconditional IRET-to-self, but that's a good >> deal

Re: [PATCH v2 5/6] x86/xen: Add a Xen-specific sync_core() implementation

On Fri, Dec 2, 2016 at 11:35 AM, Linus Torvalds wrote: > On Fri, Dec 2, 2016 at 11:30 AM, Andy Lutomirski wrote: >> >> How's this? > > Looks ok. I do think that > >> I suppose it could be an unconditional IRET-to-self, but that's a good >> deal slower and not a whole lot simpler. Although if we

Re: [PATCH] net: wireless: realtek: constify rate_control_ops structures

On 12/02/2016 03:50 AM, Bhumika Goyal wrote: The structures rate_control_ops are only passed as an argument to the functions ieee80211_rate_control_{register/unregister}. This argument is of type const, so rate_control_ops having this property can also be declared as const. Done using

Re: [PATCH] net: wireless: realtek: constify rate_control_ops structures

On 12/02/2016 03:50 AM, Bhumika Goyal wrote: The structures rate_control_ops are only passed as an argument to the functions ieee80211_rate_control_{register/unregister}. This argument is of type const, so rate_control_ops having this property can also be declared as const. Done using

[PATCH v3 07/13] net: ethernet: ti: cpts: clean up event list if event pool is empty

From: WingMan Kwok When a CPTS user does not exit gracefully by disabling cpts timestamping and leaving a joined multicast group, the system continues to receive and timestamps the ptp packets which eventually occupy all the event list entries. When this happns, the added code

[PATCH v3 05/13] net: ethernet: ti: cpts: fix registration order

The ptp clock registered before spinlock, which is protecting it, and before timecounter and cyclecounter initialization in cpts_register(). So, ensure that ptp clock is registered the last, after everything else is done. Acked-by: Richard Cochran Signed-off-by:

[PATCH RESEND] tsl2550: fix lux1_input error in low light

ADC channel 0 photodiode detects both infrared + visible light, but ADC channel 1 just detects infrared. However, the latter is a bit more sensitive in that range so complete darkness or low light causes a error condition in which the chan0 - chan1 is negative that results in a -EAGAIN. This

[PATCH v3 07/13] net: ethernet: ti: cpts: clean up event list if event pool is empty

From: WingMan Kwok When a CPTS user does not exit gracefully by disabling cpts timestamping and leaving a joined multicast group, the system continues to receive and timestamps the ptp packets which eventually occupy all the event list entries. When this happns, the added code tries to remove

[PATCH v3 05/13] net: ethernet: ti: cpts: fix registration order

The ptp clock registered before spinlock, which is protecting it, and before timecounter and cyclecounter initialization in cpts_register(). So, ensure that ptp clock is registered the last, after everything else is done. Acked-by: Richard Cochran Signed-off-by: Grygorii Strashko ---

[PATCH RESEND] tsl2550: fix lux1_input error in low light

ADC channel 0 photodiode detects both infrared + visible light, but ADC channel 1 just detects infrared. However, the latter is a bit more sensitive in that range so complete darkness or low light causes a error condition in which the chan0 - chan1 is negative that results in a -EAGAIN. This

<    1   2   3   4   5   6   7   8   9   10   >