linux.kernel - 26 new messages in 21 topics - digest

linux.kernel
http://groups.google.com/group/linux.kernel?hl=en

linux.kernel@googlegroups.com

Today's topics:

* cgroup: clean up cgroup_subsys names and initialization - 2 messages, 1
author
http://groups.google.com/group/linux.kernel/t/3a3c8584cdd64136?hl=en
* Bluetooth: remove direct compilation of 6lowpan_iphc.c - 1 messages, 1
author
http://groups.google.com/group/linux.kernel/t/b90801d1da5dd141?hl=en
* Networking - 1 messages, 1 author
http://groups.google.com/group/linux.kernel/t/1f2a0629c49b068b?hl=en
* mmc: sdhci: fix possible scheduling while atomic - 2 messages, 2 authors
http://groups.google.com/group/linux.kernel/t/5220ab2fee9bc8b4?hl=en
* ACPI, PCI: reuse ACPI hotplug framework to support PCI host bridge hotplug -
2 messages, 2 authors
http://groups.google.com/group/linux.kernel/t/6f6826cf9783dcfc?hl=en
* rcu: Eliminate softirq processing from rcutree - 2 messages, 1 author
http://groups.google.com/group/linux.kernel/t/183beaeeca586438?hl=en
* numa,sched: do statistics calculation using local variables only - 1
messages, 1 author
http://groups.google.com/group/linux.kernel/t/34ca8de3d5507b71?hl=en
* ACPI / idle: Move idle_boot_override out of the arch directory - 2 messages,
1 author
http://groups.google.com/group/linux.kernel/t/34ac1de26bb70a98?hl=en
* clk: Export more clk-provider functions - 1 messages, 1 author
http://groups.google.com/group/linux.kernel/t/80a8144f15d64067?hl=en
* [v3] driver/memory:Move Freescale IFC driver to a common driver - 1 messages,
1 author
http://groups.google.com/group/linux.kernel/t/376e2fddb7708f4c?hl=en
* ARM64 / ACPI: Introduce arm_core.c and its related head file - 1 messages, 1
author
http://groups.google.com/group/linux.kernel/t/5f4cccbbd3b92300?hl=en
* [PATCH] tracing: Fix buggered tee(2) on tracing_pipe - 1 messages, 1 author
http://groups.google.com/group/linux.kernel/t/bab4ee0ab1c9cdcb?hl=en
* usbcore: fix BABBLE failed enumeration of legacy USB2 devices on USB3 bus -
1 messages, 1 author
http://groups.google.com/group/linux.kernel/t/b9228cf2e58abcd5?hl=en
* net: rfkill: gpio: add device tree support - 1 messages, 1 author
http://groups.google.com/group/linux.kernel/t/502551497c79a009?hl=en
* gpio: mcp23s08: fix casting caused build warning - 1 messages, 1 author
http://groups.google.com/group/linux.kernel/t/63baaf004b3dd79d?hl=en
* USB: at91: fix the number of endpoint parameter - 1 messages, 1 author
http://groups.google.com/group/linux.kernel/t/6f2322f68527c2bd?hl=en
* net: core: orphan frags before queuing to slow qdisc - 1 messages, 1 author
http://groups.google.com/group/linux.kernel/t/3cd38982d779582a?hl=en
* math_state_restore and kernel_fpu_end disable interrupts? - 1 messages, 1
author
http://groups.google.com/group/linux.kernel/t/4f2afb452e630694?hl=en
* [v5] driver/memory:Move Freescale IFC driver to a common driver - 1 messages,
1 author
http://groups.google.com/group/linux.kernel/t/c5ca3c8645103d14?hl=en
* [v5] powerpc/config: Enable memory driver - 1 messages, 1 author
http://groups.google.com/group/linux.kernel/t/ec1203ad6008baf7?hl=en
* ext4: add cross rename support - 1 messages, 1 author
http://groups.google.com/group/linux.kernel/t/2b03f04905ff68d3?hl=en

==============================================================================
TOPIC: cgroup: clean up cgroup_subsys names and initialization
http://groups.google.com/group/linux.kernel/t/3a3c8584cdd64136?hl=en
==============================================================================

== 1 of 2 ==
Date: Fri, Jan 17 2014 7:20 pm
From: David Miller


From: Tejun Heo <tj@kernel.org>
Date: Fri, 17 Jan 2014 13:11:54 -0500

> Signed-off-by: Tejun Heo <tj@kernel.org>

Acked-by: David S. Miller <davem@davemloft.net>
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/




== 2 of 2 ==
Date: Fri, Jan 17 2014 7:20 pm
From: David Miller


From: Tejun Heo <tj@kernel.org>
Date: Fri, 17 Jan 2014 13:11:52 -0500

> net_cls and net_prio are the only cgroups which are allowed to be
> built as modules. The savings from allowing the two controllers to be
> built as modules are tiny especially given that cgroup module support
> itself adds quite a bit of complexity.
>
> The following are the sizes of vmlinux with both built as module and
> both built as part of the kernel image with cgroup module support
> removed.
>
> text data bss dec
> 20292207 2411496 10784768 33488471
> 20293421 2412568 10784768 33490757
>
> The total difference is 2286 bytes. Given that none of other
> controllers has much chance of being made a module and that we're
> unlikely to add new modular controllers, the added complexity is
> simply not justifiable.
>
> As a first step to drop cgroup module support, this patch changes the
> two config options to bool from tristate and drops module related code
> from the two controllers.
>
> Signed-off-by: Tejun Heo <tj@kernel.org>

Acked-by: David S. Miller <davem@davemloft.net>
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/





==============================================================================
TOPIC: Bluetooth: remove direct compilation of 6lowpan_iphc.c
http://groups.google.com/group/linux.kernel/t/b90801d1da5dd141?hl=en
==============================================================================

== 1 of 1 ==
Date: Fri, Jan 17 2014 7:20 pm
From: David Miller


From: Stephen Warren <swarren@wwwdotorg.org>
Date: Fri, 17 Jan 2014 12:29:24 -0700

> From: Stephen Warren <swarren@nvidia.com>
>
> It's now built as a separate utility module, and enabling BT selects
> that module in Kconfig. This fixes:
...
> (this change probably simply wasn't "git add"d to a53d34c3465b)
>
> Fixes: a53d34c3465b ("net: move 6lowpan compression code to separate module")
> Fixes: 18722c247023 ("Bluetooth: Enable 6LoWPAN support for BT LE devices")
> Signed-off-by: Stephen Warren <swarren@nvidia.com>

Applied to net-next, thanks a lot.
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/





==============================================================================
TOPIC: Networking
http://groups.google.com/group/linux.kernel/t/1f2a0629c49b068b?hl=en
==============================================================================

== 1 of 1 ==
Date: Fri, Jan 17 2014 7:30 pm
From: David Miller



1) The value choosen for the new SO_MAX_PACING_RATE socket option on
parisc was very poorly choosen, let's fix it while we still can.
From Eric Dumazet.

2) Our generic reciprocal divide was found to handle some edge cases
incorrectly, part of this is encoded into the BPF as deep as the
JIT engines themselves. Just use a real divide throughout for now.
From Eric Dumazet.

3) Because the initial lookup is lockless, the TCP metrics engine
can end up creating two entries for the same lookup key. Fix this
by doing a second lookup under the lock before we actually create
the new entry. From Christoph Paasch.

4) Fix scatter-gather list init in usbnet driver, from Bj�rn Mork.

5) Fix unintended 32-bit truncation in cxgb4 driver's bit shifting.
From Dan Carpenter.

6) Netlink socket dumping uses the wrong socket state for timewait
sockets. Fix from Neal Cardwell.

7) Fix netlink memory leak in ieee802154_add_iface(), from Christian
Engelmayer.

8) Multicast forwarding in ipv4 can overflow the per-rule reference
counts, causing all multicast traffic to cease. Fix from
Hannes Frederic Sowa.

9) via-rhine needs to stop all TX queues when it resets the device,
from Richard Weinberger.

10) Fix RDS per-cpu accesses broken by the this_cpu_* conversions.
From Gerald Schaefer.

Please pull, thanks a lot!

The following changes since commit 228fdc083b017eaf90e578fa86fb1ecfd5ffae87:

Merge git://git.kernel.org/pub/scm/linux/kernel/git/davem/net (2014-01-11 06:37:11 +0700)

are available in the git repository at:


git://git.kernel.org/pub/scm/linux/kernel/git/davem/net.git master

for you to fetch changes up to 3af57f78c38131b7a66e2b01e06fdacae01992a3:

s390/bpf,jit: fix 32 bit divisions, use unsigned divide instructions (2014-01-17 18:54:49 -0800)

----------------------------------------------------------------
Bj�rn Mork (1):
net: usbnet: fix SG initialisation

Christian Engelmayer (1):
ieee802154: Fix memory leak in ieee802154_add_iface()

Christoph Paasch (1):
tcp: metrics: Avoid duplicate entries with the same destination-IP

Dan Carpenter (1):
cxgb4: silence shift wrapping static checker warning

David S. Miller (1):
Merge tag 'batman-adv-fix-for-davem' of git://git.open-mesh.org/linux-merge

Eric Dumazet (2):
bpf: do not use reciprocal divide
parisc: fix SO_MAX_PACING_RATE typo

Gerald Schaefer (1):
net: rds: fix per-cpu helper usage

Hannes Frederic Sowa (2):
net: avoid reference counter overflows on fib_rules in multicast forwarding
ipv6: simplify detection of first operational link-local address on interface

Heiko Carstens (1):
s390/bpf,jit: fix 32 bit divisions, use unsigned divide instructions

Ivan Vecera (1):
be2net: add dma_mapping_error() check for dma_map_page()

Jitendra Kalsaria (1):
qlge: Fix vlan netdev features.

Marek Lindner (1):
batman-adv: fix batman-adv header overhead calculation

Michael S. Tsirkin (1):
MAINTAINERS: add virtio-dev ML for virtio

Mika Westerberg (1):
e1000e: Fix compilation warning when !CONFIG_PM_SLEEP

Neal Cardwell (1):
inet_diag: fix inet_diag_dump_icsk() to use correct state for timewait sockets

Peter Korsgaard (1):
dm9601: add USB IDs for new dm96xx variants

Richard Weinberger (1):
net,via-rhine: Fix tx_timeout handling

Yuval Mintz (1):
bnx2x: Don't release PCI bars on shutdown

MAINTAINERS | 3 +++
arch/arm/net/bpf_jit_32.c | 6 +++---
arch/parisc/include/uapi/asm/socket.h | 2 +-
arch/powerpc/net/bpf_jit_comp.c | 7 ++++---
arch/s390/net/bpf_jit_comp.c | 29 +++++++++++++++++-----------
arch/sparc/net/bpf_jit_comp.c | 17 ++++++++++++++---
arch/x86/net/bpf_jit_comp.c | 14 ++++++++++----
drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c | 29 ++++++++++++++--------------
drivers/net/ethernet/chelsio/cxgb4/l2t.c | 2 +-
drivers/net/ethernet/emulex/benet/be_main.c | 11 +++++++++--
drivers/net/ethernet/intel/e1000e/netdev.c | 8 ++------
drivers/net/ethernet/qlogic/qlge/qlge_main.c | 2 ++
drivers/net/ethernet/via/via-rhine.c | 1 +
drivers/net/usb/dm9601.c | 12 ++++++++++++
drivers/net/usb/usbnet.c | 2 +-
include/net/if_inet6.h | 1 -
net/batman-adv/main.c | 2 +-
net/core/filter.c | 30 ++---------------------------
net/ieee802154/nl-phy.c | 6 ++++--
net/ipv4/inet_diag.c | 5 ++++-
net/ipv4/ipmr.c | 7 +++++--
net/ipv4/tcp_metrics.c | 51 +++++++++++++++++++++++++++++++-------------------
net/ipv6/addrconf.c | 38 +++++++++++++++++--------------------
net/ipv6/ip6mr.c | 7 +++++--
net/rds/ib_recv.c | 7 +++----
25 files changed, 169 insertions(+), 130 deletions(-)
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/





==============================================================================
TOPIC: mmc: sdhci: fix possible scheduling while atomic
http://groups.google.com/group/linux.kernel/t/5220ab2fee9bc8b4?hl=en
==============================================================================

== 1 of 2 ==
Date: Fri, Jan 17 2014 7:30 pm
From: Andrew Bresticker


On Fri, Jan 17, 2014 at 3:40 PM, Chris Ball <chris@printf.net> wrote:
> Hi, adding Aisheng,
>
> On Fri, Jan 17 2014, Andrew Bresticker wrote:
>> On Fri, Jan 17, 2014 at 3:11 PM, John Tobias <john.tobias.ph@gmail.com> wrote:
>>> There's an existing patch for that...
>>> http://www.spinics.net/lists/arm-kernel/msg296596.html
>>
>> Ah, I see. Looks like it has yet to be picked up...
>
> The patches aren't quite identical -- Andrew's leaves the
> disable_irq() call in and Aisheng's removes it. Which should I take?

Since the disable_irq() is now redundant, I suppose Aisheng's is more correct.

Thanks,
Andrew

>
> Thanks,
>
> - Chris.
> --
> Chris Ball <chris@printf.net> <http://printf.net/>
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/




== 2 of 2 ==
Date: Fri, Jan 17 2014 7:50 pm
From: Chris Ball


Hi,

On Sat, Jan 18 2014, Andrew Bresticker wrote:
>>>> There's an existing patch for that...
>>>> http://www.spinics.net/lists/arm-kernel/msg296596.html
>>>
>>> Ah, I see. Looks like it has yet to be picked up...
>>
>> The patches aren't quite identical -- Andrew's leaves the
>> disable_irq() call in and Aisheng's removes it. Which should I take?
>
> Since the disable_irq() is now redundant, I suppose Aisheng's is more correct

Thanks, pushed Aisheng's version to mmc-next for 3.14.

- Chris.
--
Chris Ball <chris@printf.net> <http://printf.net/>
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/





==============================================================================
TOPIC: ACPI, PCI: reuse ACPI hotplug framework to support PCI host bridge
hotplug
http://groups.google.com/group/linux.kernel/t/6f6826cf9783dcfc?hl=en
==============================================================================

== 1 of 2 ==
Date: Fri, Jan 17 2014 7:30 pm
From: Yinghai Lu


On Fri, Jan 17, 2014 at 6:48 PM, Jiang Liu <jiang.liu@linux.intel.com> wrote:
> Reuse ACPI hotplug framework to support PCI host bridge hotplug, this
> makes PCI host bridge hotplug implementation simpler and more clear.
>
> It also fixes a bug in support of PCI host bridge hot-addition.
> Currently pci_root driver fails to install notification handler for
> PCI host bridge absent at boot time because acpi_is_root_bridge()
> returns false if no ACPI device created for handle. So PCI host
> bridge hot-addition event will just be ignored by system.
>
> Signed-off-by: Jiang Liu <jiang.liu@linux.intel.com>

is the same as

commit 3338db0057ed9f554050bd06863731c515d79672
Author: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
Date: Fri Nov 22 21:55:20 2013 +0100

ACPI / hotplug: Make ACPI PCI root hotplug use common hotplug code

in rafael tree pm/linux-next for 3.14?

> ---
> drivers/acpi/internal.h | 1 -
> drivers/acpi/pci_root.c | 117 +++++++----------------------------------------
> drivers/acpi/scan.c | 2 -
> 3 files changed, 17 insertions(+), 103 deletions(-)
>
> diff --git a/drivers/acpi/internal.h b/drivers/acpi/internal.h
> index a29739c..03efa56 100644
> --- a/drivers/acpi/internal.h
> +++ b/drivers/acpi/internal.h
> @@ -28,7 +28,6 @@ int init_acpi_device_notify(void);
> int acpi_scan_init(void);
> void acpi_pci_root_init(void);
> void acpi_pci_link_init(void);
> -void acpi_pci_root_hp_init(void);
> void acpi_processor_init(void);
> void acpi_platform_init(void);
> int acpi_sysfs_init(void);
> diff --git a/drivers/acpi/pci_root.c b/drivers/acpi/pci_root.c
> index 20360e4..6f6e6c1 100644
> --- a/drivers/acpi/pci_root.c
> +++ b/drivers/acpi/pci_root.c
> @@ -50,6 +50,7 @@ ACPI_MODULE_NAME("pci_root");
> static int acpi_pci_root_add(struct acpi_device *device,
> const struct acpi_device_id *not_used);
> static void acpi_pci_root_remove(struct acpi_device *device);
> +static int handle_hotplug_event_root(acpi_handle handle, u32 type, void *ctx);
>
> #define ACPI_PCIE_REQ_SUPPORT (OSC_PCI_EXT_CONFIG_SUPPORT \
> | OSC_PCI_ASPM_SUPPORT \
> @@ -61,12 +62,14 @@ static const struct acpi_device_id root_device_ids[] = {
> {"", 0},
> };
>
> +
> static struct acpi_scan_handler pci_root_handler = {
> .ids = root_device_ids,
> + .prepare = handle_hotplug_event_root,
> .attach = acpi_pci_root_add,
> .detach = acpi_pci_root_remove,
> .hotplug = {
> - .ignore = true,
> + .enabled = true,
> },
> };
>
> @@ -627,113 +630,27 @@ void __init acpi_pci_root_init(void)
>
> if (!acpi_pci_disabled) {
> pci_acpi_crs_quirks();
> - acpi_scan_add_handler(&pci_root_handler);
> - }
> -}
> -/* Support root bridge hotplug */
> -
> -static void handle_root_bridge_insertion(acpi_handle handle)
> -{
> - struct acpi_device *device;
> -
> - if (!acpi_bus_get_device(handle, &device)) {
> - dev_printk(KERN_DEBUG, &device->dev,
> - "acpi device already exists; ignoring notify\n");
> - return;
> + acpi_scan_add_handler_with_hotplug(&pci_root_handler,
> + "pci_hostbridge");
> }
> -
> - if (acpi_bus_scan(handle))
> - acpi_handle_err(handle, "cannot add bridge to acpi list\n");
> }
>
> -static void hotplug_event_root(void *data, u32 type)
> +static int handle_hotplug_event_root(acpi_handle handle, u32 type, void *ctx)
> {
> - acpi_handle handle = data;
> + int ret = NOTIFY_OK;
> struct acpi_pci_root *root;
>
> - acpi_scan_lock_acquire();
> -
> - root = acpi_pci_find_root(handle);
> -
> - switch (type) {
> - case ACPI_NOTIFY_BUS_CHECK:
> - /* bus enumerate */
> - acpi_handle_printk(KERN_DEBUG, handle,
> - "Bus check notify on %s\n", __func__);
> - if (root)
> + if (type == ACPI_NOTIFY_BUS_CHECK) {
> + acpi_scan_lock_acquire();
> + root = acpi_pci_find_root(handle);
> + if (root) {
> + acpi_handle_printk(KERN_DEBUG, handle,
> + "Bus check notify on %s\n", __func__);
> acpiphp_check_host_bridge(handle);
> - else
> - handle_root_bridge_insertion(handle);
> -
> - break;
> -
> - case ACPI_NOTIFY_DEVICE_CHECK:
> - /* device check */
> - acpi_handle_printk(KERN_DEBUG, handle,
> - "Device check notify on %s\n", __func__);
> - if (!root)
> - handle_root_bridge_insertion(handle);
> - break;
> -
> - case ACPI_NOTIFY_EJECT_REQUEST:
> - /* request device eject */
> - acpi_handle_printk(KERN_DEBUG, handle,
> - "Device eject notify on %s\n", __func__);
> - if (!root)
> - break;
> -
> - get_device(&root->device->dev);
> -
> + ret = NOTIFY_STOP;
> + }
> acpi_scan_lock_release();
> -
> - acpi_bus_device_eject(root->device, ACPI_NOTIFY_EJECT_REQUEST);
> - return;
> - default:
> - acpi_handle_warn(handle,
> - "notify_handler: unknown event type 0x%x\n",
> - type);
> - break;
> }
>
> - acpi_scan_lock_release();
> -}
> -
> -static void handle_hotplug_event_root(acpi_handle handle, u32 type,
> - void *context)
> -{
> - acpi_hotplug_execute(hotplug_event_root, handle, type);
> -}
> -
> -static acpi_status __init
> -find_root_bridges(acpi_handle handle, u32 lvl, void *context, void **rv)
> -{
> - acpi_status status;
> - int *count = (int *)context;
> -
> - if (!acpi_is_root_bridge(handle))
> - return AE_OK;
> -
> - (*count)++;
> -
> - status = acpi_install_notify_handler(handle, ACPI_SYSTEM_NOTIFY,
> - handle_hotplug_event_root, NULL);
> - if (ACPI_FAILURE(status))
> - acpi_handle_printk(KERN_DEBUG, handle,
> - "notify handler is not installed, exit status: %u\n",
> - (unsigned int)status);
> - else
> - acpi_handle_printk(KERN_DEBUG, handle,
> - "notify handler is installed\n");
> -
> - return AE_OK;
> -}
> -
> -void __init acpi_pci_root_hp_init(void)
> -{
> - int num = 0;
> -
> - acpi_walk_namespace(ACPI_TYPE_DEVICE, ACPI_ROOT_OBJECT,
> - ACPI_UINT32_MAX, find_root_bridges, NULL, &num, NULL);
> -
> - printk(KERN_DEBUG "Found %d acpi root devices\n", num);
> + return ret;
> }
> diff --git a/drivers/acpi/scan.c b/drivers/acpi/scan.c
> index 6b0f419..d83e0ff 100644
> --- a/drivers/acpi/scan.c
> +++ b/drivers/acpi/scan.c
> @@ -2057,8 +2057,6 @@ int __init acpi_scan_init(void)
>
> acpi_update_all_gpes();
>
> - acpi_pci_root_hp_init();
> -
> out:
> mutex_unlock(&acpi_scan_lock);
> return result;
> --
> 1.7.10.4
>
> --
> To unsubscribe from this list: send the line "unsubscribe linux-acpi" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at http://vger.kernel.org/majordomo-info.html
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/




== 2 of 2 ==
Date: Fri, Jan 17 2014 7:50 pm
From: Jiang Liu


Hi yinghai,
Sorry for the noise. I didn't noticed Rafael's work,
so I generated this patchset when encountered this issue
during testing PCI host bridge hotplug. It should achieve
the same goal.
Thanks!
Gerry

On 2014/1/18 11:23, Yinghai Lu wrote:
> ACPI / hotplug: Make ACPI PCI root hotplug use common hotplug code
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/





==============================================================================
TOPIC: rcu: Eliminate softirq processing from rcutree
http://groups.google.com/group/linux.kernel/t/183beaeeca586438?hl=en
==============================================================================

== 1 of 2 ==
Date: Fri, Jan 17 2014 7:30 pm
From: Mike Galbraith


On Fri, 2014-01-17 at 18:14 +0100, Sebastian Andrzej Siewior wrote:
> * Mike Galbraith | 2013-12-25 18:37:37 [+0100]:
>
> >On Tue, 2013-12-24 at 23:55 -0800, Paul E. McKenney wrote:
> >> On Wed, Dec 25, 2013 at 04:07:34AM +0100, Mike Galbraith wrote:
> >
> >Having sufficiently recovered from turkey overdose to be able to slither
> >upstairs (bump bump bump) to check on the box, commenting..
> >
> ># timers-do-not-raise-softirq-unconditionally.patch
> ># rtmutex-use-a-trylock-for-waiter-lock-in-trylock.patch
> >
> >..those two out does seem to have stabilized the thing.
>
> timers-do-not-raise-softirq-unconditionally.patch is on its way out.
>
> rtmutex-use-a-trylock-for-waiter-lock-in-trylock.patch confues me.
> Didn't you report once that your box deadlocks without this patch? Now
> your 64way box on the other hand does not work with it?

If 'do not raise' is applied, 'use a trylock' won't save you. If 'do
not raise' is not applied, _and_ you wisely do not try to turn on very
expensive nohz_full, things work fine without 'use a trylock'.

-Mike

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/




== 2 of 2 ==
Date: Fri, Jan 17 2014 7:40 pm
From: Mike Galbraith


On Fri, 2014-01-17 at 18:23 +0100, Sebastian Andrzej Siewior wrote:

> So I had rtmutex-take-the-waiter-lock-with-irqs-off.patch in my queue
> which took the waiter lock with irqs off. This should be the same thing
> you try do here.

(yeah, these are just whacked mole body bags;)

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/





==============================================================================
TOPIC: numa,sched: do statistics calculation using local variables only
http://groups.google.com/group/linux.kernel/t/34ca8de3d5507b71?hl=en
==============================================================================

== 1 of 1 ==
Date: Fri, Jan 17 2014 7:40 pm
From: Rik van Riel


On 01/17/2014 04:12 PM, riel@redhat.com wrote:
> From: Rik van Riel <riel@redhat.com>
>
> The current code in task_numa_placement calculates the difference
> between the old and the new value, but also temporarily stores half
> of the old value in the per-process variables.
>
> The NUMA balancing code looks at those per-process variables, and
> having other tasks temporarily see halved statistics could lead to
> unwanted numa migrations. This can be avoided by doing all the math
> in local variables.
>
> This change also simplifies the code a little.

I am seeing what looks like a performance improvement
with this patch, so it is not just a theoretical bug.
The improvement is small, as is to be expected with
such a small race, but with two 32-warehouse specjbb
instances on a 4-node, 10core/20thread per node system,
I see the following change in performance, and reduced
numa page migrations.

Without the patch:
run 1: throughput 367660 367660, migrated 3112982
run 2: throughput 353821 355612, migrated 2881317
run 3: throughput 355027 355027, migrated 3358105
run 4: throughput 354366 354366, migrated 3466687
run 5: throughput 356186 356186, migrated 3152194
run 6: throughput 361431 361431, migrated 3336219
run 7: throughput 354704 354704, migrated 3345418
run 8: throughput 363770 363770, migrated 3642925
run 9: throughput 363380 363380, migrated 3192836
run 10: throughput 358440 358440, migrated 3354028
avg: througphut 358968, migrated 3284271

With the patch:
run 1: throughput 360580 360580, migrated 3169872
run 2: throughput 361303 361303, migrated 3220280
run 3: throughput 367692 367692, migrated 3096093
run 4: throughput 362320 362320, migrated 2981762
run 5: throughput 364201 364201, migrated 3089107
run 6: throughput 364561 364561, migrated 2892364
run 7: throughput 360771 360771, migrated 3086638
run 8: throughput 361530 361530, migrated 2933256
run 9: throughput 365841 365841, migrated 3356944
run 10: throughput 359188 359188, migrated 3394545
avg: througphut 362798, migrated 3122086


--
All rights reversed
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/





==============================================================================
TOPIC: ACPI / idle: Move idle_boot_override out of the arch directory
http://groups.google.com/group/linux.kernel/t/34ac1de26bb70a98?hl=en
==============================================================================

== 1 of 2 ==
Date: Fri, Jan 17 2014 7:50 pm
From: Hanjun Guo


On 2014-1-17 20:06, Sudeep Holla wrote:
> On 17/01/14 02:03, Hanjun Guo wrote:
>> Move idle_boot_override out of the arch directory to be a single enum
>> including both platforms values, this will make it rather easier to
>> avoid ifdefs around which definitions are for which processor in
>> generally used ACPI code.
>>
>> IDLE_FORCE_MWAIT for IA64 is not used anywhere, so romove it.
>>
>> No functional change in this patch.
>>
>> Suggested-by: Alan <gnomes@lxorguk.ukuu.org.uk>
>> Signed-off-by: Hanjun Guo <hanjun.guo@linaro.org>
>> ---
>> arch/ia64/include/asm/processor.h | 3 ---
>> arch/powerpc/include/asm/processor.h | 1 -
>> arch/x86/include/asm/processor.h | 3 ---
>> arch/x86/kernel/process.c | 1 +
>> include/linux/cpu.h | 8 ++++++++
>> 5 files changed, 9 insertions(+), 7 deletions(-)
>>
>> diff --git a/arch/ia64/include/asm/processor.h b/arch/ia64/include/asm/processor.h
>> index 5a84b3a..ccd63a0 100644
>> --- a/arch/ia64/include/asm/processor.h
>> +++ b/arch/ia64/include/asm/processor.h
>> @@ -698,9 +698,6 @@ prefetchw (const void *x)
>>
>> extern unsigned long boot_option_idle_override;
>>
>> -enum idle_boot_override {IDLE_NO_OVERRIDE=0, IDLE_HALT, IDLE_FORCE_MWAIT,
>> - IDLE_NOMWAIT, IDLE_POLL};
>> -
>> void default_idle(void);
>>
>> #define ia64_platform_is(x) (strcmp(x, ia64_platform_name) == 0)
>> diff --git a/arch/powerpc/include/asm/processor.h b/arch/powerpc/include/asm/processor.h
>> index fc14a38..06689c0 100644
>> --- a/arch/powerpc/include/asm/processor.h
>> +++ b/arch/powerpc/include/asm/processor.h
>> @@ -440,7 +440,6 @@ static inline unsigned long get_clean_sp(unsigned long sp, int is_32)
>>