On 08/09/2012 10:53 AM, Lee Jones wrote:
> of_irq_find_parent is a handy function to use outside the confines of
> the Open Firmware subsystem. One such use-case is when the IRQ Domain
> wishes to find an IRQ domain for a given device node. Currently it can
> not take any notice of the 'interrupt-p
On Thu, 9 Aug 2012, Will Deacon wrote:
> I think we could actually fix this entirely in mutex-xchg.h by doing
> something in fastpath_lock similar to what we do for trylock:
>
>
> diff --git a/include/asm-generic/mutex-xchg.h
> b/include/asm-generic/mutex-xchg.h
> index 580a6d3..c082e99 100644
On 08/09/2012 09:45 AM, Mark Brown wrote:
> If the regulator doesn't supply a way of reading back the voltage but does
> provide a list_voltage() operation then use that with a selector of zero
> to read the voltage.
Ah, this makes sense. The only issue is:
static int add_regulator_attributes(str
On Thu, Aug 9, 2012 at 10:24 PM, Sarah Sharp
wrote:
> On Thu, Aug 09, 2012 at 05:31:51PM +0800, Keng-Yu Lin wrote:
>> With a previous patch to enable the EHCI/XHCI port switching, it switches
>> all the available ports.
>>
>> The assumption is not correct because the BIOS may expect some ports
>>
On Thu, Aug 9, 2012 at 12:05 PM, Eric Paris wrote:
> Paul, are you looking into this? This is a bandaide, not a fix
Yep, I mentioned this a few times in the other thread. The problem is
there is not going to be an easy fix for the labeling so I'd rather we
see this patch, or something like
On Thu, Aug 9, 2012 at 9:50 AM, Dave Martin wrote:
> On Thu, Aug 09, 2012 at 09:32:59AM -0500, Matt Sealey wrote:
>> Matt Sealey
>> Product Development Analyst, Genesi USA, Inc.
>>
>>
>> On Thu, Aug 9, 2012 at 5:24 AM, Dave Martin wrote:
>> > On Wed, Aug 08, 2012 at 12:32:39PM -0500, Matt Sealey
On Thu, Aug 9, 2012 at 11:36 AM, Eric Dumazet wrote:
> On Thu, 2012-08-09 at 11:07 -0400, Paul Moore wrote:
>
>> Is is possible to do the call to security_sk_alloc() in the ip_init()
>> function
>> or does the per-cpu nature of the socket make this a pain?
>>
>
> Its a pain, if we want NUMA affin
From: James Ralston
This patch adds the AHCI-mode SATA Device IDs for the Intel Lynx Point-LP PCH
Signed-off-by: James Ralston
---
drivers/ata/ahci.c |8
1 files changed, 8 insertions(+), 0 deletions(-)
diff --git a/drivers/ata/ahci.c b/drivers/ata/ahci.c
index ebaf67e..c590259 1
Hi Linus,
Please pull my for-linus-3.6 branch:
git://git.kernel.org/pub/scm/linux/kernel/git/mason/linux-btrfs.git
for-linus-3.6
It fixes a merging error in rc1. The calls to mnt_want_write should
have been removed.
Alexander Block (1):
Btrfs: remove mnt_want_write call in btrfs_mksubvo
The platform attempts to register platform device 'snd_soc_u8500'
which doesn't actually exist. Here we change the reference to the
correct one 'snd_soc_mop500'.
Signed-off-by: Lee Jones
---
arch/arm/mach-ux500/board-mop500-msp.c | 10 +-
1 file changed, 5 insertions(+), 5 deletions(-)
Previous attempts to add platform probing of the Audio related devices
only call from non-DT initialisation functions. This patch extends that
functionality to the Device Tree related ones too.
Signed-off-by: Lee Jones
---
arch/arm/mach-ux500/board-mop500.c |4
1 file changed, 4 inserti
On Thu, Aug 9, 2012 at 11:36 AM, Eric Dumazet wrote:
> On Thu, 2012-08-09 at 11:07 -0400, Paul Moore wrote:
>
>> Is is possible to do the call to security_sk_alloc() in the ip_init()
>> function
>> or does the per-cpu nature of the socket make this a pain?
>>
>
> Its a pain, if we want NUMA affin
I'm announcing the release of the 3.5.1 kernel.
All users of the 3.5 kernel series must upgrade.
The updated 3.5.y git tree can be found at:
git://git.kernel.org/pub/scm/linux/kernel/git/stable/linux-stable.git
linux-3.5.y
and can be browsed at the normal kernel.org git web browser:
Hi Ashley,
> diff --git a/drivers/char/tpm/Makefile b/drivers/char/tpm/Makefile
> index 547509d..b53da57 100644
> --- a/drivers/char/tpm/Makefile
> +++ b/drivers/char/tpm/Makefile
> @@ -2,9 +2,15 @@
> # Makefile for the kernel tpm device drivers.
> #
> obj-$(CONFIG_TCG_TPM) += tpm.o
> +obj-$(CO
All AB8500 devices are now registered via MFD core, so Device Tree
capability is no longer required for probing. Here we pull the DT
match table to ensure we're no longer probed during Device Tree
start-up.
CC: Alessandro Zummo
CC: rtc-li...@googlegroups.com
Signed-off-by: Lee Jones
---
drivers
I'm announcing the release of the 3.4.8 kernel.
All users of the 3.4 kernel series must upgrade.
The updated 3.4.y git tree can be found at:
git://git.kernel.org/pub/scm/linux/kernel/git/stable/linux-stable.git
linux-3.4.y
and can be browsed at the normal kernel.org git web browser:
This was left over during a recent clean-up which removed Device Tree
helper structs. There is no longer a requirement for it, so we can just
remove it.
Signed-off-by: Lee Jones
---
arch/arm/mach-ux500/board-mop500.c |5 -
1 file changed, 5 deletions(-)
diff --git a/arch/arm/mach-ux500/
The compatible string found in the Power-On-Key's MFD cell matches
the device name provided in the driver. Keeping this naming convention
seems like a good idea, so we're changing the one found in the DTS
file to match.
Signed-off-by: Lee Jones
---
arch/arm/boot/dts/dbx5x0.dtsi |2 +-
1 file
irq_find_host() currently ignores the 'interrupt-parent' property
even if it's specified in the Device Tree. Meaning that a node can
match to a domain in its hierarchy even if it doesn't belong to it.
By searching for the parent first using of_irq_find_parent() we
insist that the 'interrupt-parent'
diff --git a/Documentation/stable_kernel_rules.txt
b/Documentation/stable_kernel_rules.txt
index e1f856b..22bf11b 100644
--- a/Documentation/stable_kernel_rules.txt
+++ b/Documentation/stable_kernel_rules.txt
@@ -1,4 +1,4 @@
-Everything you ever wanted to know about Linux 2.6 -stable releases.
+Ev
I'm announcing the release of the 3.0.40 kernel.
All users of the 3.0 kernel series must upgrade.
The updated 3.0.y git tree can be found at:
git://git.kernel.org/pub/scm/linux/kernel/git/stable/linux-stable.git
linux-3.0.y
and can be browsed at the normal kernel.org git web browser:
We're just about to provide the DB8500-PRCMU with its own IRQ domain,
so that its subordinate drivers can use it as an interrupt controller.
It's obligatory for all IRQ controllers to reference themselves as
such from its own node in Device Tree. This patch does just that.
Signed-off-by: Lee Jones
It's understood that the AB8500 should be subordinate to the DB8500;
however, the AB8500 uses the GIC as it's interrupt controller. If
we do not specify which IRQ controller to use the default is to use
the next encountered IRQ controller as we climb the tree. This would
be the DB8500. This patch e
Before the AB8500 had its own IRQ domain, the IRQ handler would take
the fired local IRQ (hwirq) and add it to the irq_base to convert it
to an IRQ number which Linux would understand (virq). However, the
IRQ base is not always used anymore since we can make use of Linear
domains. It's better to us
There was a plan to place ab8500_irq_get_virq() calls in each AB8500
child device prior to requesting an IRQ, but as we're no longer using
Device Tree to collect our IRQ numbers, it's actually better to allow
the core to do this during device registration time. So the IRQ number
we pull from its re
Without this patch the default behaviour is to climb the Device
Tree and use the first encountered interrupt controller. This
does not take into account if a device node has specified to use
a particular IRQ controller using the interrupt-parent property.
This patch ensures that property is adhered
of_irq_find_parent is a handy function to use outside the confines of
the Open Firmware subsystem. One such use-case is when the IRQ Domain
wishes to find an IRQ domain for a given device node. Currently it can
not take any notice of the 'interrupt-parent' property. Instead it
just uses the first I
The PRCMU has its own USB, Thermal, GPIO, Modem, HSI and RTC drivers,
amongst other things. This patch allows those subordinate devices to
use it as an interrupt controller as and when they are DT enabled.
CC: Samuel Ortiz
Signed-off-by: Lee Jones
---
drivers/mfd/db8500-prcmu.c | 54 +++
Here we apply some fixes to the way the AB8500 handles IRQs and
provide the DB8500 with an IRQ domain. We also handle hwirq to
virq conversion slightly differently when registering MFD devices.
The other changes pertain to the way IRQs are requested and dealt
with by AB8500 child devices.
arch/ar
In the initial submission of the MSP driver msp1 and msp3's associated
pinctrl mechanism was passed back to platform code using a plat_init()
call-back routine, but it has no place in platform code. The MSP driver
should set this up for the appropriate ports. Instead we use a use_pinctrl
identifier
Here we ensure that the MOP500 audio driver will be probed during a
Device Tree boot. We also parse the sound node to link together the
codec, dma and the CPU-side Digital Audio Interface.
CC: alsa-de...@alsa-project.org
Signed-off-by: Lee Jones
---
sound/soc/ux500/mop500.c | 40 ++
The PCM is a pseudo-device. It doesn't have any of it's own registers
or hardware. It rather acts as a layer of abstraction for DMA
transfers. Hence, instead of classifying it as a device in its own
right, we call the initialisation from the MSP driver.
CC: alsa-de...@alsa-project.org
Signed-off-b
On Thu, Aug 09, 2012 at 05:40:46PM +0200, Borislav Petkov wrote:
> From: Borislav Petkov
>
> One day I was playing with disabling the caches over CR0.CD and I wanted
> to check the current setting of the CD bit but there wasn't a way to do
> so from userspace because we #GP when accessing CR0.
>
We continue to allow the AB8500 CODEC to be registered via the AB8500
Multi Functional Device API, only this time we extract its configuration
from the Device Tree binary.
CC: alsa-de...@alsa-project.org
Signed-off-by: Lee Jones
---
include/linux/mfd/abx500/ab8500-codec.h |6 ++-
sound/soc/c
This patch contains a couple of general MSP clean-ups pertaining to
layout changes and changing functions to be void instead of int instead
of regardlessly returning '0'.
Signed-off-by: Lee Jones
---
arch/arm/mach-ux500/board-mop500-msp.c | 24 +++-
arch/arm/mach-ux500/boar
Ensure correct probing and pass though important configuration
options to the AB8500 CODEC driver when DT is enabled
Signed-off-by: Lee Jones
---
arch/arm/boot/dts/db8500.dtsi |6 ++
1 file changed, 6 insertions(+)
diff --git a/arch/arm/boot/dts/db8500.dtsi b/arch/arm/boot/dts/db8500.dt
We've done this before and it worked well last time. Here we're
duplicating a complex registration function to ease the process
of enabling it for Device Tree. As there are quite a few steps
taken during the registration process, it makes sense to break
them up into more manageable chunks. This pat
In this patch we prevent MOP500 driver registration from platform
code and rely solely on Device Tree to do the probing for us.
Signed-off-by: Lee Jones
---
arch/arm/mach-ux500/board-mop500-msp.c |3 ---
1 file changed, 3 deletions(-)
diff --git a/arch/arm/mach-ux500/board-mop500-msp.c
b/a
We now initialise the PCM driver through the MSP DAI, so there is
no need to register it though platform code anymore. This patch
strips out all PCM platform registration.
Signed-off-by: Lee Jones
---
arch/arm/mach-ux500/board-mop500-msp.c |6 --
1 file changed, 6 deletions(-)
diff --gi
Hi everyone,
This first pull is the bulk of our changes for the next rc. It is
against the 3.5 kernel so people testing the new features have a stable
point to work against. This was tested against Linus' current tree as
well.
The second pull is just one fix against 3.6-rc1 (in another email).
Here we add the required documentation for the new Device Tree
bindings pertaining to the MOP500 Audio Machine driver.
CC: alsa-de...@alsa-project.org
Signed-off-by: Lee Jones
---
.../devicetree/bindings/sound/ux500-mop500.txt | 39
1 file changed, 39 insertions(+)
cr
Here we add the required documentation for the new Device Tree
bindings pertaining to the MSP CPU-side DAI Driver.
CC: alsa-de...@alsa-project.org
Signed-off-by: Lee Jones
---
.../devicetree/bindings/sound/ux500-msp.txt| 43
1 file changed, 43 insertions(+)
create
If codec->control_data is not populated SoC Core assumes we want to
use regmap, which fails catastrophically, as we don't have one:
Unable to handle kernel NULL pointer dereference at virtual address 0080
pgd = c0004000
[0080] *pgd=
Internal error: Oops: 17 [#1] PREEMPT SMP ARM
Mod
This patch removes platform device registration of all 4 MSP
devices. It also takes care of all redundant infrastructure now
that each of the ux500 audio components have been Device Tree
enabled.
Signed-off-by: Lee Jones
---
arch/arm/mach-ux500/board-mop500-msp.c | 14 --
arch/arm/
Signed-off-by: Lee Jones
---
arch/arm/boot/dts/db8500.dtsi | 633
arch/arm/boot/dts/dbx5x0.dtsi | 633
arch/arm/boot/dts/snowball.dts |2 +-
3 files changed, 634 insertions(+), 634 deletions(-)
delete mode
This patch inserts all known MSP devices into the DBx5x0 Device Tree
disabled and enables the two important ones for the Snowball low-cost
development board.
Signed-off-by: Lee Jones
---
arch/arm/boot/dts/dbx5x0.dtsi | 33 +
arch/arm/boot/dts/snowball.dts |
This is where we link together all of the SoC sound components for
a given platform. This all encompassing sound node is only found
in the very lowest hierarchical DTS file, since the component mix may
well change from board variant to board variant.
Signed-off-by: Lee Jones
---
arch/arm/boot/dt
Here we document the AB8500 Audio CODEC in the existing MFD AB8500 document.
CC: alsa-de...@alsa-project.org
Signed-off-by: Lee Jones
---
Documentation/devicetree/bindings/mfd/ab8500.txt | 15 +++
1 file changed, 15 insertions(+)
diff --git a/Documentation/devicetree/bindings/mfd/
The 'msp' board file does more than just register MSP devices. It
also registers some other components necessary to get audio working
on ux500 based platforms; such as the PCM and Machine Drivers. For
that reason we're changing the filename to be more encompassing -
'audio'.
Signed-off-by: Lee Jon
It isn't currently possible to pass all platform specific configuration
though Device Tree. Thinks like device names used in the clock
infrastructure, call-backs and DMA information have to be passed in via
AUXDATA structures and the MSP is no exception. Here we're passing DMA
settings.
Signed-off
Register both parts of the MSP driver from Device Tree so that they
are probed when Device Tree is enabled. Also, as there is platform
data involved, we ensure that there is allocated memory to place the
configuration into and that the correct information is extracted from
the DT binary.
CC: alsa-
Another fine instalment of Audio fixes and Device Tree enablement
surrounding the mop500 sound driver. Hopefully containing all of
the recommendations suggested by Mark Brown.
Documentation/devicetree/bindings/mfd/ab8500.txt | 15 +
.../devicetree/bindings/sound/ux500-mop500.txt | 39 +
YAMA et al rely on on i_uid/i_gid to be populated in order to perform
their checks. While these really cannot be guarenteed as the underlying
filesystem may not even have the concept, they are expected to be filled
when possible. To quote Al Viro:
"Ideally, yes, we'd want to have ->i_uid use
Hi, everyone
I wish to alloc physically contiguous memory in kernel mode.
I used the __get_free_pages(gfp_mask, order) function,
but it fails occasionary.
I guess it is caused by memory fragmentation.
My code is kernel driver module which cannot use
bootmem allocation and its platform is x86_64.
M
If the regulator doesn't supply a way of reading back the voltage but does
provide a list_voltage() operation then use that with a selector of zero
to read the voltage.
Reported-by: Stephen Warren
Signed-off-by: Mark Brown
---
drivers/regulator/core.c |2 ++
1 file changed, 2 insertions(+)
On Thu, Aug 09, 2012 at 09:22:55AM -0600, Stephen Warren wrote:
> I don't think there's a *get*_voltage_linear. All the existing
> get_voltage "standard" ops require reading a register, hence why I added
> this new standard op. I guess I'll look at converting all the existing
> users.
Hrm, right
Hi,
> We should offline and remove memory when removing the memory device.
> The memory device can be removed by 2 ways:
> 1. send eject request by SCI
> 2. echo 1 >/sys/bus/pci/devices/PNP0C80:XX/eject
>
[snip]
> +
> +static int acpi_memory_disable_device(struct acpi_memory_device *mem_device)
From: Borislav Petkov
One day I was playing with disabling the caches over CR0.CD and I wanted
to check the current setting of the CD bit but there wasn't a way to do
so from userspace because we #GP when accessing CR0.
And then I thought, well, sysrq must have this. And it does:
$ echo p > /pr
On Wed, 8 Aug 2012, Dave Martin wrote:
> On Mon, Aug 06, 2012 at 03:27:05PM +0100, Stefano Stabellini wrote:
> > Use r12 to pass the hypercall number to the hypervisor.
> >
> > We need a register to pass the hypercall number because we might not
> > know it at compile time and HVC only takes an im
On 08/09/2012 06:13 PM, Paolo Bonzini wrote:
> Il 05/07/2012 12:29, Jason Wang ha scritto:
>> Sometimes, virtio device need to configure irq affiniry hint to maximize the
>> performance. Instead of just exposing the irq of a virtqueue, this patch
>> introduce an API to set the affinity for a virtqu
On Thu, 2012-08-09 at 11:07 -0400, Paul Moore wrote:
> Is is possible to do the call to security_sk_alloc() in the ip_init()
> function
> or does the per-cpu nature of the socket make this a pain?
>
Its a pain, if we want NUMA affinity.
Here, each cpu should get memory from its closest node.
From: Andi Kleen
Add a description of the JIT interface in the perf symbol resolution
code. I reverse engineered the format from the source.
Cc: Pekka Enberg
Signed-off-by: Andi Kleen
---
tools/perf/Documentation/jit-interface.txt | 16
1 files changed, 16 insertions(+), 0
>>> On 09.08.12 at 17:03, "Kirill A. Shutemov"
>>> wrote:
> ENTRY(clear_page_nocache)
> CFI_STARTPROC
> - xorl %eax,%eax
> - movl $4096/64,%ecx
> + push %rdi
> + call kernel_fpu_begin
> + pop%rdi
You use CFI annotations elsewhere, so why don't you use
pushq
On Thu, 2012-08-09 at 08:22 -0700, Joe Perches wrote:
> Yes, I sent signed-off patches to their original RFC submission
> (back last year?). To me, this single large patch is like a
> snapshot of a git tree and not all contributors to that git
> tree should be noted as signers of the entire thing
> While on 64-bit this is fine, I fail to see how you avoid using the
> SSE2 instruction on non-SSE2 systems.
You're right, this needs a fallback path for 32bit non sse
(and fixing the ABI)
-Andi
--
a...@linux.intel.com -- Speaking for myself only
--
To unsubscribe from this list: send the line
On 08/09/2012 12:27 AM, Bjorn Helgaas wrote:
> On Wed, Aug 8, 2012 at 8:44 AM, Jiang Liu wrote:
>> On 08/08/2012 07:38 AM, Toshi Kani wrote:
>
>>> It is nice to see redundant ACPI namespace walks removed from the ACPI
>>> drivers. But why do you need to add a new enumerator to create the
>>> acp
On 08/09/2012 08:03 AM, Kirill A. Shutemov wrote:
From: Andi Kleen
Add a cache avoiding version of clear_page. Straight forward integer variant
of the existing 64bit clear_page, for both 32bit and 64bit.
Also add the necessary glue for highmem including a layer that non cache
coherent architec
On 08/09/2012 04:59 AM, Mark Brown wrote:
> On Wed, Aug 08, 2012 at 03:18:46PM -0600, Stephen Warren wrote:
>
>> Fixed regulators always output desc->min_uV. Add a helper get_voltage
>> op to save duplicating this code in drivers.
>
> Just use get_voltage_linear() (or convert the existing users).
On Thu, 2012-08-09 at 10:54 -0400, Steven Rostedt wrote:
> On Thu, Aug 09, 2012 at 06:49:25AM +, Ren, Cloud wrote:
> > Luis has ever asked joe for adding as a Signed-off-by.
[]
> I'm curious, did Luis ask on list or off?
Off vger lists with some qca list or exploder added.
> The Signed-off ta
>>> On 09.08.12 at 17:03, "Kirill A. Shutemov"
>>> wrote:
> From: Andi Kleen
>
> Add a cache avoiding version of clear_page. Straight forward integer variant
> of the existing 64bit clear_page, for both 32bit and 64bit.
While on 64-bit this is fine, I fail to see how you avoid using the
SSE2 i
Fix two checkpatch.pl warnings about printk issues by using
pr_info(...) instead of printk(KERN_INFO, ...).
Signed-off-by: Güngör Erseymen
---
drivers/staging/comedi/drivers/ssv_dnp.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/drivers/staging/comedi/drivers/ssv_dnp.c
On 07/20/2012 05:50 AM, Kirill A. Shutemov wrote:
From: "Kirill A. Shutemov"
Clearing a 2MB huge page will typically blow away several levels of CPU
caches. To avoid this only cache clear the 4K area around the fault
address and use a cache avoiding clears for the rest of the 2MB area.
It wou
On Thu, Aug 09, 2012 at 11:08:03AM +0100, Mark Brown wrote:
> On Wed, Aug 08, 2012 at 06:13:12PM -0300, Herton Ronaldo Krzesinski wrote:
>
> Always delete irrelevant text from your quotes.
>
> > Doesn't build on 3.4:
>
> > linux-stable/sound/soc/soc-dapm.c: In function 'soc_dapm_shutdown_codec':
On Thu, Aug 09, 2012 at 02:53:53AM +0100, Ben Hutchings wrote:
> On Tue, 2012-08-07 at 15:34 -0700, Greg Kroah-Hartman wrote:
> > From: Greg KH
> >
> > 3.4-stable review patch. If anyone has any objections, please let me know.
> >
> > --
> >
> > From: Antonio Quartulli
> >
>
On Thu, Aug 09, 2012 at 04:13:28AM +0100, Ben Hutchings wrote:
> On Tue, 2012-08-07 at 15:25 -0700, Greg Kroah-Hartman wrote:
> > From: Greg KH
> >
> > 3.5-stable review patch. If anyone has any objections, please let me know.
> >
> > --
> >
> > From: Antonio Quartulli
> >
>
On 08/09/2012 04:09 AM, Mark Brown wrote:
> On Wed, Aug 08, 2012 at 03:18:45PM -0600, Stephen Warren wrote:
>
>> + rdev->desc->enable_mask |
>> + rdev->desc->en_dis_set_mask |
>> + rdev->desc->en_dis_clr_mask,
>
Il 30/07/2012 08:27, Paolo Bonzini ha scritto:
>> > Did you set the affinity manually in your experiments, or perhaps
>> > there
>> > is a difference between scsi and networking... (interrupt mitigation?)
>> >
>> > You need to run irqbalancer in guest to make it actually work. Do y
Il 05/07/2012 12:29, Jason Wang ha scritto:
> Sometimes, virtio device need to configure irq affiniry hint to maximize the
> performance. Instead of just exposing the irq of a virtqueue, this patch
> introduce an API to set the affinity for a virtqueue.
>
> The api is best-effort, the affinity hin
On Thu, Aug 09, 2012 at 11:48:36AM -0300, Rafael Aquini wrote:
> Sure!
> what do you think of:
>
> +/* putback_lru_page() counterpart for a ballooned page */
> +void putback_balloon_page(struct page *page)
> +{
> + lock_page(page);
> + if (!WARN_ON(!movable_balloon_page(page))) {
> +
On Thursday, August 09, 2012 04:50:33 PM Eric Dumazet wrote:
> From: Eric Dumazet
>
> commit be9f4a44e7d41cee (ipv4: tcp: remove per net tcp_sock) added a
> selinux regression, reported and bisected by John Stultz
>
> selinux_ip_postroute_compat() expect to find a valid sk->sk_security
> pointer
From: Andi Kleen
With multiple threads vector stores are more efficient, so use them.
This will cause the page clear to run non preemptable and add some
overhead. However on 32bit it was already non preempable (due to
kmap_atomic) and there is an preemption opportunity every 4K unit.
On a NPB (N
On Thu, Aug 9, 2012 at 10:27 AM, Eric Dumazet wrote:
> On Thu, 2012-08-09 at 09:30 -0400, Paul Moore wrote:
>
>> In the case of a TCP syn-recv and timewait ACK things are a little less
>> clear.
>> Eric (Dumazet), it looks like we have a socket in tcp_v4_reqsk_send_ack() and
>> tcp_v4_timewait_ac
From: Andi Kleen
Use the fault address, not the rounded down hpage address for NUMA
policy purposes. In some circumstances this can give more exact
NUMA policy.
Signed-off-by: Andi Kleen
Signed-off-by: Kirill A. Shutemov
---
mm/huge_memory.c |8
1 files changed, 4 insertions(+),
From: Andi Kleen
Signed-off-by: Andi Kleen
Signed-off-by: Kirill A. Shutemov
---
mm/huge_memory.c |9 +
1 files changed, 5 insertions(+), 4 deletions(-)
diff --git a/mm/huge_memory.c b/mm/huge_memory.c
index 70737ec..ecd93f8 100644
--- a/mm/huge_memory.c
+++ b/mm/huge_memory.c
@@
From: Andi Kleen
Add a cache avoiding version of clear_page. Straight forward integer variant
of the existing 64bit clear_page, for both 32bit and 64bit.
Also add the necessary glue for highmem including a layer that non cache
coherent architectures that use the virtual address for flushing can
From: Andi Kleen
hugetlb does not necessarily pass in an aligned address, so the
low level address computation is wrong.
This will fix architectures that actually use the address for flushing
the cleared address (very few, like xtensa/sparc/...?)
Signed-off-by: Andi Kleen
Signed-off-by: Kirill
From: Andi Kleen
Clearing a 2MB huge page will typically blow away several levels
of CPU caches. To avoid this only cache clear the 4K area
around the fault address and use a cache avoiding clears
for the rest of the 2MB area.
Signed-off-by: Andi Kleen
Signed-off-by: Kirill A. Shutemov
---
mm
From: "Kirill A. Shutemov"
Clearing a 2MB huge page will typically blow away several levels of CPU
caches. To avoid this only cache clear the 4K area around the fault
address and use a cache avoiding clears for the rest of the 2MB area.
This patchset implements cache avoiding version of clear_p
"DebBarma, Tarun Kanti" writes:
> On Wed, Aug 8, 2012 at 10:40 PM, Kevin Hilman wrote:
>> Tarun Kanti DebBarma writes:
>>
>>> Add *remove* callback so that necessary cleanup operations are
>>> performed when device is unregistered. The device is deleted
>>> from the list and associated clock ha
On Thu, Aug 09, 2012 at 06:49:25AM +, Ren, Cloud wrote:
>
> Please calm down. I will follow rule of sending email from now on. Sorry for
> causing
> your trouble. Luis has ever asked joe for adding as a Signed-off-by. At that
> time joe agreed
> with it. If joe prefers his name not be listed
Hi Greg,
On 08/09/2012 04:23 PM, Greg KH wrote:
> On Thu, Aug 09, 2012 at 11:36:21AM +0200, Tomasz Stanislawski wrote:
>> This patch adds reference counting on a module that exported dma-buf and
>> implements its operations. This prevents the module from being unloaded while
>> DMABUF file is in u
On Thu, Aug 09, 2012 at 08:36:12AM -0600, Jim Schutt wrote:
> Hi Mel,
>
> On 08/09/2012 07:49 AM, Mel Gorman wrote:
> >Changelog since V2
> >o Capture !MIGRATE_MOVABLE pages where possible
> >o Document the treatment of MIGRATE_MOVABLE pages while capturing
> >o Expand changelogs
> >
> >Changelog
From: Eric Dumazet
commit be9f4a44e7d41cee (ipv4: tcp: remove per net tcp_sock) added a
selinux regression, reported and bisected by John Stultz
selinux_ip_postroute_compat() expect to find a valid sk->sk_security
pointer, but this field is NULL for unicast_sock
Fix this by adding a new 'kernel
Hi Nicolas,
Thanks for the replies.
On Thu, Aug 09, 2012 at 06:12:15AM +0100, Nicolas Pitre wrote:
> On Tue, 7 Aug 2012, Will Deacon wrote:
> > diff --git a/kernel/mutex.c b/kernel/mutex.c
> > index a307cc9..27b7887 100644
> > --- a/kernel/mutex.c
> > +++ b/kernel/mutex.c
> > @@ -170,7 +170,7 @@
On Thu, Aug 09, 2012 at 09:32:59AM -0500, Matt Sealey wrote:
> Matt Sealey
> Product Development Analyst, Genesi USA, Inc.
>
>
> On Thu, Aug 9, 2012 at 5:24 AM, Dave Martin wrote:
> > On Wed, Aug 08, 2012 at 12:32:39PM -0500, Matt Sealey wrote:
> >
> > [...]
> >
> >> I'm going to do a trapse th
On Thu, Aug 09, 2012 at 10:00:19AM +0100, Mel Gorman wrote:
> On Wed, Aug 08, 2012 at 07:53:19PM -0300, Rafael Aquini wrote:
> > Memory fragmentation introduced by ballooning might reduce significantly
> > the number of 2MB contiguous memory blocks that can be used within a guest,
> > thus imposing
On Thu, Aug 09, 2012 at 08:55:26PM +0800, Fengguang Wu wrote:
> Hi Guenter,
>
> This commit triggered an oops which can be fixed by the attached diff.
> Should it be folded into the original one (preferable for me), or be
> resent as a standalone patch?
>
I folded it into the original commit.
Th
> > This patch adds the checking code for the two register so that only the
> > switchable ports are altered.
And since when has trusting BIOS provided data beena good idea ?
Absent a lot of precise examples of where it causes actual failures I
don't believe we should change our behaviour here.
On Thu, Aug 09, 2012 at 04:34:42PM +0200, Takashi Iwai wrote:
> At Thu, 9 Aug 2012 16:25:36 +0200,
> Thierry Reding wrote:
> >
> > On Thu, Aug 09, 2012 at 04:10:55PM +0200, Takashi Iwai wrote:
> > > For processing the firmware handling properly for built-in kernels,
> > > implement an asynchronous
Hi Mel,
On 08/09/2012 07:49 AM, Mel Gorman wrote:
Changelog since V2
o Capture !MIGRATE_MOVABLE pages where possible
o Document the treatment of MIGRATE_MOVABLE pages while capturing
o Expand changelogs
Changelog since V1
o Dropped kswapd related patch, basically a no-op and regresses if fixed
301 - 400 of 651 matches
Mail list logo
