On Sun, Sep 2, 2012 at 10:46 AM, Yinghai Lu wrote:
> This one intend to fix bugs:
> when efi booting have too many memmap entries, will need to double memblock
> memory array or reserved array.
Okay, why do we need to do that?
> +RESERVE_BRK(early_pgt_alloc, 65536);
What is this needed for?
>
At 09/01/2012 05:06 AM, Andrew Morton Wrote:
> On Tue, 28 Aug 2012 18:00:15 +0800
> we...@cn.fujitsu.com wrote:
>
>> From: Yasuaki Ishimatsu
>>
>> When (hot)adding memory into system, /sys/firmware/memmap/X/{end, start,
>> type}
>> sysfs files are created. But there is no code to remove these
On Sat, Sep 01, 2012 at 07:13:38PM -0400, Christoph Hellwig wrote:
> I'd suspect it's something with the actual radix tree code, Ccing
> linux-mm in case they know more.
I don't think it has anything to do with the radix tree code
> On Mon, Aug 27, 2012 at 11:00:10AM -0400, Justin Piszcz
On Sun, Sep 2, 2012 at 10:46 AM, Yinghai Lu wrote:
> +static void __init with_all_ram_ranges(
> + void (*work_fn)(unsigned long, unsigned long, void *),
> + void *data)
> +static void __init size_work_fn(unsigned long start, unsigned long end, void
>
On Sun, Sep 2, 2012 at 10:46 AM, Yinghai Lu wrote:
> From: Jacob Shin
>
> Current logic finds enough space for direct mapping page tables from 0
> to end. Instead, we only need to find enough space to cover mr[0].start
> to mr[nr_range].end -- the range that is actually being mapped by
>
On Sun, Sep 2, 2012 at 10:46 AM, Yinghai Lu wrote:
> From: Jacob Shin
>
> Currently direct mappings are created for [ 0 to max_low_pfn< and [ 4GB to max_pfn< backed by actual DRAM. This is fine for holes under 4GB which are covered
> by fixed and variable range MTRRs to be UC. However, we run
On Sun, Sep 2, 2012 at 10:46 AM, Yinghai Lu wrote:
> From: Jacob Shin
>
> Update code that previously assumed pfns [ 0 - max_low_pfn_mapped ) and
> [ 4GB - max_pfn_mapped ) were always direct mapped, to now look up
> pfn_mapped ranges instead.
What problem does this fix? How did you find about
On Sun, Sep 2, 2012 at 10:46 AM, Yinghai Lu wrote:
> From: Jacob Shin
>
> There could be cases where user supplied memmap=exactmap memory
> mappings do not mark the region where the kernel .text .data and
> .bss reside as E820_RAM, as reported here:
>
> https://lkml.org/lkml/2012/8/14/86
>
>
On Sun, Sep 2, 2012 at 10:46 AM, Yinghai Lu wrote:
> So later could make it call split_mem_range...
>
> Signed-off-by: Yinghai Lu
The commit title is utterly confusing. And it has a trailing dot (".").
As for the actual change:
Reviewed-by: Pekka Enberg
--
To unsubscribe from this list: send
On 08/28/2012 05:17 PM, Robert Richter wrote:
On 27.08.12 09:32:13, wei.y...@windriver.com wrote:
From: Wei Yang
Upon enabling the call-graph functionality of oprofile, A few minutes
later the following calltrace will always occur.
BUG: unable to handle kernel paging request at 656d6153
This
On Sun, Sep 2, 2012 at 10:46 AM, Yinghai Lu wrote:
> it should take physical address range that will need to be mapped.
> and find_early_table_space should take range that pgt buff should be in.
> Separate those two to reduce confusion.
>
> Signed-off-by: Yinghai Lu
Reviewed-by: Pekka Enberg
On Sun, Sep 2, 2012 at 10:46 AM, Yinghai Lu wrote:
> Should not do that in every calling of init_memory_mapping.
> Actually in early time, only need do once.
>
> Also move down early_memtest.
>
> Signed-off-by: Yinghai Lu
The changelog is too terse for my liking. I think it could use some
more
On Sun, Sep 2, 2012 at 10:46 AM, Yinghai Lu wrote:
> So we could put page table high again for 64bit.
>
> Signed-off-by: Yinghai Lu
The changelog for this is too terse for me to actually understand why
this is needed.
> ---
> arch/x86/mm/init.c |2 +-
> 1 files changed, 1 insertions(+), 1
On Sun, Sep 2, 2012 at 10:46 AM, Yinghai Lu wrote:
> from setup.c to mm/init.c
>
> So could update all related calling together later
>
> Signed-off-by: Yinghai Lu
Reviewed-by: Pekka Enberg
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to
On Sun, Sep 2, 2012 at 10:46 AM, Yinghai Lu wrote:
> from init_memory_mapping, so make init_memory_mapping readable.
>
> Suggested-by: Ingo Molnar
> Signed-off-by: Yinghai Lu
Reviewed-by: Pekka Enberg
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a
On Sun, Sep 2, 2012 at 10:46 AM, Yinghai Lu wrote:
> detect if need to use 1G or 2M and store them in page_size_mask.
>
> Only probe them one time.
>
> Suggested-by: Ingo Molnar
> Signed-off-by: Yinghai Lu
Reviewed-by: Pekka Enberg
--
To unsubscribe from this list: send the line "unsubscribe
The vast majority of systems either use initramfs or mount a root
filesystem directly from the kernel. Distros have defaulted to
initramfs for years. Only highly specialized systems would use an
actual filesystem-image initrd at this point, and such systems don't
rely on defconfig anyway. Drop
CONFIG_CRC_T10DIF explicitly states that it exists only for use by
out-of-tree modules; anything in-kernel that needs it selects it. Thus,
compile it out by default.
Signed-off-by: Josh Triplett
---
arch/x86/configs/i386_defconfig |1 -
arch/x86/configs/x86_64_defconfig |1 -
2 files
The current x86 and x86-64 defconfigs do not enable ext4, which most
current distributions default to. Switch the defconfigs to ext4, so
they will boot on current systems without additional configuration.
Signed-off-by: Josh Triplett
---
arch/x86/configs/i386_defconfig |7 +++
On Mon, Sep 3, 2012 at 4:45 AM, Namhyung Kim wrote:
> On Mon, 3 Sep 2012 03:04:32 +0300, Irina Tirdea wrote:
>> From: Irina Tirdea
>>
>> When NDEBUG is defined, the assert macro will be expanded to nothing.
>> Some assert calls used in perf are also including some functionality
>> (e.g. system
The x86 defconfigs have become somewhat out of date compared to the
current result of "make savedefconfig". Update them to the current output, as
a prelude to further defconfig changes, to avoid unrelated noise in
those further changes.
Signed-off-by: Josh Triplett
---
The x86 defconfigs include exactly one module: test_nx.ko, a
special-purpose module which just exists to do evil things like
executing code off the stack to see if the kernel has enabled NX
support. Anyone who actually uses that module can easily enable it
themselves, but the vast majority of
After repeatedly going through the cycle of building a "make defconfig" kernel,
trying to boot it, getting a kernel panic, turning on ext4, and rebuilding, I
figured I'd actually get the defconfigs fixed to work on modern systems with
ext4 root filesystems. Patch 2 of this patch series does
Whoops, need VIRTIO_RING too
>From 79dc9ae40e40cefd6079e4197cac858a1d59e1d8 Mon Sep 17 00:00:00 2001
From: Kent Overstreet
Date: Sun, 2 Sep 2012 21:44:37 -0700
Subject: [PATCH] virtio-blk: Fix kconfig option
CONFIG_VIRTIO isn't exposed, everything else is supposed to select it
instead.
HID_LOGITECH_DJ uses "default m", which enables it in default kernel
builds. Since this module just enables extra, non-critical
functionality for one particular piece of hardware (specifically,
differentiating multiple wireless keyboards and mice as separate input
devices rather than treating
CONFIG_VIRTIO isn't exposed, everything else is supposed to select it
instead.
Signed-off-by: Kent Overstreet
---
drivers/block/Kconfig |3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)
diff --git a/drivers/block/Kconfig b/drivers/block/Kconfig
index a796407..d4e1d12 100644
---
On Mon, Sep 03, 2012 at 11:56:59AM +0900, Nobuhiro Iwamatsu wrote:
> On Fri, Aug 31, 2012 at 10:10 PM, Al Viro wrote:
> > Smells like broken TIF_NOTIFY_RESUME hookup on sh... Arrrgh. Looks like
> > while it is in the relevant masks *and* checked in do_notify_resume() both
> > on 32bit and 64bit
From: Michael Wang
In real world of x86, during an interrupt, if current thread need to
be reschedule, we will do it after invoke do_IRQ.
And in linsched, while handle the softirq, it may cause the
reschedule ipi on other cpu, so we need to do schedule for them at
that time, otherwise we will
From: Michael Wang
tick_nohz_irq_exit() will make sure the tick timer reprogram correctly
after cpu enter idle.
With out this check, after the interrupt, tick timer will be enabled
even cpu is still in idle, this will cause inaccuracy.
Signed-off-by: Michael Wang
---
From: Michael Wang
In real world, tick_nohz_idle_enter() will be invoked by idle thread
when the cpu change from active to idle, and will only be invoked
again after tick_nohz_idle_exit() was invoked by idle thread when
cpu is going to recover, invoke it multiple times in one idle may
cause
From: Michael Wang
process_all_softirqs() will handle softirq for all the cpu even
it's not the right timing for them, this will cause inaccuracy.
This patch stop invoke process_all_softirqs(), so softirq will
only be handled after timer interrupt arrived.
Signed-off-by: Michael Wang
---
From: Michael Wang
This patch set fix several issues in linsched, help it to
behave more close to the real world, in order to make the
test results more accurately.
Signed-off-by: Michael Wang
---
b/arch/linsched/kernel/irq.c |4
b/tools/linsched/hrtimer.c |1 -
On Sun, Sep 02, 2012 at 10:19:06PM +0800, Wei Yongjun wrote:
> From: Wei Yongjun
>
> buf has been allocated in this function and should be
> freed before leaving from the error handling case.
>
> spatch with a semantic match is used to found this problem.
> (http://coccinelle.lip6.fr/)
>
>
The use of if (!head) BUG(); can be replaced with single line
BUG_ON(!head).
Signed-off-by: Prasad Joshi
---
fs/proc/proc_sysctl.c |3 +--
1 files changed, 1 insertions(+), 2 deletions(-)
diff --git a/fs/proc/proc_sysctl.c b/fs/proc/proc_sysctl.c
index dfafeb2..63bdfa0 100644
---
On Sun, Sep 02, 2012 at 06:59:37PM -0700, H. Peter Anvin wrote:
> Well, appending is an action, not really a property of the variable that
> sticks around, no?
We could do this for append, but what happens if an attribute with
similar variables appears and doesn't neatly map to an O_ option? We
Hi hpa,
> Well, appending is an action, not really a property of the variable
> that sticks around, no?
True, but they're still all defined as the same thing in the UEFI spec.
If you're looking to define which attributes to pass, you now need to
know the extra information that you pass most of
On Mon, Sep 03, 2012 at 10:31:03AM +0800, Dong Aisheng wrote:
> >
> > I think of_node_put should be moved out from here and put into
> > syscon_node_to_regmap and syscon_regmap_lookup_by_compatible.
> >
> I guess no, if you want to move of_node_put into
> syscon_regmap_lookup_by_phandle,
> then
On Fri, Aug 31, 2012 at 10:02:48AM +0800, Shawn Guo wrote:
> On Wed, Aug 29, 2012 at 06:56:53PM +0800, Dong Aisheng wrote:
> > From: Dong Aisheng
> >
> > Include headfile for easy using.
> >
> > Acked-by: Stephen Warren
> > Signed-off-by: Dong Aisheng
> > ---
> > arch/arm/boot/dts/imx6q.dtsi
SNDRV_MAIN_OBJECT_FILE hasn't done anything since the pre-git days, and
the only remaining reference occurs as a #define in sound/last.c. Drop
that last mention of it.
Signed-off-by: Josh Triplett
---
sound/last.c |1 -
1 file changed, 1 deletion(-)
diff --git a/sound/last.c
From: Namhyung Kim
Now we can support color using pango markup with this change.
Acked-by: Pekka Enberg
Signed-off-by: Namhyung Kim
---
tools/perf/ui/gtk/browser.c | 101 +---
tools/perf/ui/gtk/gtk.h | 1 +
tools/perf/ui/gtk/setup.c | 1 +
3
From: Namhyung Kim
Current hist print functions are messy because it has to consider many
of command line options and the code doing that is scattered around to
places. So when someone wants to add an option to manipulate the hist
output it'd very easy to miss to update all of them in sync. And
From: Namhyung Kim
Signed-off-by: Namhyung Kim
---
tools/perf/ui/hist.c | 27 +++
tools/perf/util/hist.c | 33 -
2 files changed, 27 insertions(+), 33 deletions(-)
diff --git a/tools/perf/ui/hist.c b/tools/perf/ui/hist.c
index
From: Namhyung Kim
Override hpp->color functions for TUI. Because line coloring is done
outside of the function, it just sets the percent value and pass it.
Signed-off-by: Namhyung Kim
---
tools/perf/ui/browsers/hists.c | 94 --
From: Namhyung Kim
When a field separator is given, the output format doesn't need to be
fancy like aligning to column length, coloring the percent value and
so on. And since there's a slight difference to normal format, fix it
not to break backward compatibility.
Signed-off-by: Namhyung Kim
Hi,
This is a cleanup and refactoring patchset for the hist printing code
by adding perf_hpp__format functions and perf_hpp. I believe it makes
the code easy to maintain and to add new features like upcoming group
viewing and callchain accumulation.
Any comments are welcome, thanks.
Namhyung
On Fri, Aug 31, 2012 at 10:10 PM, Al Viro wrote:
> On Fri, Aug 31, 2012 at 04:32:45PM +0900, Nobuhiro Iwamatsu wrote:
>> Hi, Al.
>>
>> I can not get output of command line on SH in latest linux kernel.
>> I bisected, I confirmed that this problem occurred by following commit.
>>
>> -
>>
On Fri, Aug 31, 2012 at 09:39:26AM +0800, Shawn Guo wrote:
> On Wed, Aug 29, 2012 at 06:56:58PM +0800, Dong Aisheng wrote:
> > -config MFD_ANATOP
> > - bool "Support for Freescale i.MX on-chip ANATOP controller"
> > - depends on SOC_IMX6Q
> > - help
> > - Select this option to enable
On Fri, Aug 31, 2012 at 09:26:29AM +0800, Shawn Guo wrote:
> On Wed, Aug 29, 2012 at 06:56:52PM +0800, Dong Aisheng wrote:
> > +config MFD_SYSCON
> > +bool "System Controller Register R/W Based on Regmap"
>
> If the driver only compiles and works for device tree probe, we should
> have
Commit 56449f437add737a1e5e1cb7e00f63ac8ead1938, in April 2009, made
trace_clock available unconditionally, since CONFIG_X86_DS used it too.
Commit faa4602e47690fb11221e00f9b9697c8dc0d4b19, in March 2010, removed
CONFIG_X86_DS, and now only CONFIG_RING_BUFFER (split out from
CONFIG_TRACING for
On Fri, Aug 31, 2012 at 11:27:01AM +0200, Philipp Zabel wrote:
> This driver requests and remaps a memory region as configured in the
> device tree. It serves memory from this region via the genalloc API.
>
> Other drivers can retrieve the genalloc pool from a phandle pointing
> to this drivers'
On 08/22/2012 10:40 AM, Michael Wang wrote:
> From: Michael Wang
>
> Fengguang Wu has reported the bug:
>
> [0.043953] BUG: scheduling while atomic: swapper/0/1/0x1002
> [0.044017] no locks held by swapper/0/1.
> [0.044692] Pid: 1, comm: swapper/0 Not tainted
From: Robin Dong
When trying to modify flashcache to request based (current it's bio based), we
need
to make request from bios by ourselves, but dm_io() will submit these bios
directly,
so we propose to modify the dm_io() to return bios instead of submiting it.
This could also improve the
From: Robin Dong
We are now trying to modify flashcache(https://github.com/facebook/flashcache)
to make it request based so that
we can let cfq io-controller control the bandwidth between different
io cgroups.
A search in the dm directory tells me that only multipath is a request
based dm
On Fri, Aug 31, 2012 at 11:27:00AM +0200, Philipp Zabel wrote:
> This patch keeps all created pools in a global list and adds two
> functions that allow to retrieve the gen_pool pointer from a known
> physical address and from a device tree node.
>
> Signed-off-by: Philipp Zabel
> ---
>
Well, appending is an action, not really a property of the variable that sticks
around, no?
Jeremy Kerr wrote:
>hi hpa,
>
>> Wouldn't that be better handled by O_APPEND?
>
>Possibly, but this then means that there are now two "interfaces" that
>specify the variable attributes.
>
>[Also, in
I do not understand the point of introducing those imx patches, 1 ~ 4
and 7, 8. They are all unnecessary churns to me. IMO, 4 patches are
enough.
* genalloc: add a global pool list, allow to find pools by phys address
* misc: Generic on-chip SRAM allocation driver
* ARM i.MX: remove
On Mon, 3 Sep 2012 03:04:32 +0300, Irina Tirdea wrote:
> From: Irina Tirdea
>
> When NDEBUG is defined, the assert macro will be expanded to nothing.
> Some assert calls used in perf are also including some functionality
> (e.g. system calls), not only validity checks. Therefore, if NDEBUG is
>
Driver for the Microchip MRF24J40 802.15.4 WPAN module.
Signed-off-by: Alan Ott
---
drivers/net/ieee802154/Kconfig| 11 +
drivers/net/ieee802154/Makefile | 1 +
drivers/net/ieee802154/mrf24j40.c | 767 ++
3 files changed, 779 insertions(+)
create
This is a driver for the Microchip MRF24J40 802.15.4 transceiver.
This is the same as v6 except that it now puts the driver in
drivers/net/ieee802154 instead of drivers/ieee802154, since
drivers/ieee802154 was recently moved to drivers/net/ieee802154 in
0739d643b8dda866d1011bcf
Full history and
hi hpa,
> Wouldn't that be better handled by O_APPEND?
Possibly, but this then means that there are now two "interfaces" that
specify the variable attributes.
[Also, in that case we should support the same mechanism through open();
llseek(0, SEEK_END) then, right?]
In general, I think the
On Fri, Aug 31, 2012 at 07:13:48PM -0700, Tejun Heo wrote:
> Hello, Vivek.
>
> On Thu, Aug 30, 2012 at 06:07:45PM -0400, Vivek Goyal wrote:
> > Here is one quick and dirty proof of concept patch. It checks for stack
> > depth and if remaining space is less than 20% of stack size, then it
> >
On Fri, Aug 31, 2012 at 11:01:59AM -0400, Vivek Goyal wrote:
> On Thu, Aug 30, 2012 at 06:43:59PM -0700, Kent Overstreet wrote:
> > On Thu, Aug 30, 2012 at 06:07:45PM -0400, Vivek Goyal wrote:
> > > On Wed, Aug 29, 2012 at 10:13:45AM -0700, Kent Overstreet wrote:
> > >
> > > [..]
> > > > >
At 09/01/2012 04:55 AM, Andrew Morton Wrote:
> On Tue, 28 Aug 2012 18:00:11 +0800
> we...@cn.fujitsu.com wrote:
>
>> +int remove_memory(int nid, u64 start, u64 size)
>> +{
>> +int ret = -EBUSY;
>> +lock_memory_hotplug();
>> +/*
>> + * The memory might become online by other task,
On Mon, Sep 03, 2012 at 10:49:27AM +1000, Dave Chinner wrote:
> Given that we are working around stack depth issues in the
> filesystems already in several places, and now it seems like there's
> a reason to work around it in the block layers as well, shouldn't we
> simply increase the default
On Wed, 29 Aug 2012 01:22:16 +0300, Irina Tirdea wrote:
> mempcpy is not supported by bionic in Android
> and will lead to compilation errors.
>
> Replacing mempcpy with memcpy so it will work in Android.
Unfortunately I've added another instance of the mempcpy in a
libtraceevent code. :-/
Hi,
On Wed, 29 Aug 2012 01:04:25 +0300, Irina Tirdea wrote:
> perf defines __used to for marking unused variables.
> The variable __used is defined to __attribute__((__unused__)), which
> contradicts the kernel definition to __attribute__((__used__))
> for new gcc versions. On Android), __used is
On Thu, Aug 30, 2012 at 06:07:45PM -0400, Vivek Goyal wrote:
> On Wed, Aug 29, 2012 at 10:13:45AM -0700, Kent Overstreet wrote:
>
> [..]
> > > Performance aside, punting submission to per device worker in case of deep
> > > stack usage sounds cleaner solution to me.
> >
> > Agreed, but
The option allows you to remove TTY and compile without errors. This
saves space on systems that won't support TTY interfaces anyway.
bloat-o-meter output is below.
The bulk of this patch consists of Kconfig changes adding "depends on
TTY" to various serial devices and similar drivers that
On Wed, Aug 29, 2012 at 02:32:04PM +0900, Jun'ichi Nomura wrote:
> On 08/29/12 11:59, Dave Chinner wrote:
> > On Mon, Aug 27, 2012 at 06:05:06PM -0400, Naoya Horiguchi wrote:
> >> And yes, I understand it's ideal, but many applications choose not to
> >> do that for performance reason.
> >> So I
Deleted the no longer valid example of which x86 CPUs lack a hardware
IOMMU, and moved the "If unsure..." statement to a new line to follow
the style of surrounding options.
Signed-off-by: Joe Millenbach
Reviewed-by: Josh Triplett
---
v2: Incorporated feedback from Konrad Rzeszutek Wilk that no
On Mon, Sep 03, 2012 at 02:03:31AM +0200, Davidlohr Bueso wrote:
> Calling this function no longer makes sense as /proc/acpi/event
> is being removed.
>
> Signed-off-by: Davidlohr Bueso
> :
> drivers/platform/x86/fujitsu-laptop.c |4
> :
> diff --git
On Mon, Sep 03, 2012 at 12:38:31AM +0800, Fengguang Wu wrote:
> On Sun, Sep 02, 2012 at 03:41:34PM +0200, Bjørn Mork wrote:
> > Fengguang Wu writes:
> >
> > > After the __devinit* removal series, I can still get kernel panic in
> > > show_uevent(). So there are more sources of bug..
> > >
> > >
From: Irina Tirdea
When NDEBUG is defined, the assert macro will be expanded to nothing.
Some assert calls used in perf are also including some functionality
(e.g. system calls), not only validity checks. Therefore, if NDEBUG is
defined, these functionality will be removed along with the assert.
Calling this function no longer makes sense as /proc/acpi/event
is being removed.
Signed-off-by: Davidlohr Bueso
---
drivers/platform/x86/asus-laptop.c |1 -
drivers/platform/x86/eeepc-laptop.c |1 -
drivers/platform/x86/fujitsu-laptop.c |4
Calling this function no longer makes sense as /proc/acpi/event
is being removed.
Signed-off-by: Davidlohr Bueso
---
drivers/pci/hotplug/acpiphp_ibm.c |1 -
1 files changed, 0 insertions(+), 1 deletions(-)
diff --git a/drivers/pci/hotplug/acpiphp_ibm.c
b/drivers/pci/hotplug/acpiphp_ibm.c
Calling this function no longer makes sense as /proc/acpi/event
is being removed.
Signed-off-by: Davidlohr Bueso
---
drivers/char/sonypi.c |5 -
1 files changed, 0 insertions(+), 5 deletions(-)
diff --git a/drivers/char/sonypi.c b/drivers/char/sonypi.c
index f877805..4543473 100644
---
The long time deprecated procfs interface for ACPI power devices has
been scheduled for removal since linux 2.6.39.
Signed-off-by: Davidlohr Bueso
---
Documentation/feature-removal-schedule.txt | 11 -
drivers/acpi/Kconfig | 17 --
drivers/acpi/Makefile
The /proc/acpi/event interface has been replaced by events through the
input layer and netlink, and scheduled for removal over four years ago.
Signed-off-by: Davidlohr Bueso
---
Documentation/feature-removal-schedule.txt |8 --
drivers/acpi/Kconfig | 18 -
Hi,
This patchset is a first attempt to remove some of the deprecated procfs
ACPI interfaces - in the final overall idea to remove /proc/acpi entirely.
Based on the feature removal file, the CONFIG_ACPI_PROCFS_POWER and
CONFIG_ACPI_PROC_EVENT
options are dropped.
patch 1: removes
On Fri, 2012-08-31 at 14:37 -0300, Marcelo Tosatti wrote:
> On Fri, Aug 31, 2012 at 06:10:48PM +0200, Davidlohr Bueso wrote:
> > For processors that support VPIDs we should invalidate the page table entry
> > specified by the lineal address. For this purpose add support for individual
> > address
Since I had a rare occasion to physically access the machine, I did
the following experiment: connect another machine to the serial
console, run
while true ; do date ; cat /proc/slabinfo ; echo '***' ; sleep 3 ; done
and generate lots of IPv6 traffic through the box (as I mentioned, for
some
It is main portion of pci_rescan_bus().
Separate it out and need to use it for pci root bus hot add later.
Signed-off-by: Yinghai Lu
---
drivers/pci/setup-bus.c | 32 ++--
include/linux/pci.h |1 +
2 files changed, 19 insertions(+), 14 deletions(-)
diff
We have pci_assign_unassigned_bus_resources() in as global function now.
So could move back pci_rescan_bus to probe.c where it should be.
Signed-off-by: Yinghai Lu
---
drivers/pci/probe.c | 21 +
drivers/pci/setup-bus.c | 22 --
2 files changed,
We want to create rescan in sys only for pci bridge instead of all pci dev.
We could use attribute_groups/is_visible method to do that.
Now pci dev does not use device type yet. So add pci_dev_type to take
attr_groups with it.
Add pci_dev_bridge_attrs_are_visible() to control attr_bridge_group
From: Jiang Liu
Function pci_get_dev_by_id() will hold a reference count on the pci device
returned, so pci_dev_present() should release the corresponding reference
count to avoid memory leakage.
Signed-off-by: Jiang Liu
---
drivers/pci/search.c | 10 +-
1 files changed, 5
From: Jiang Liu
The function pci_create_root_bus() allocates the pci bus structure,
registers the bus device and creates the legacy files for a pci root
bus, but returns without setting the is_added flag. The is_added flag
for a pci root bus will be set by function pci_scan_child_bus().
If a pci
From: Jiang Liu
If pci_stop_and_remove_bus() is called to remove a pci root bus,
the host_bridge structure may have already been freed after returning
from pci_remove_bus(). To avoid that, hold an extra reference count
to the root bus before calling pci_remove_bus(), so we can safely
access the
Add missing hot_remove support for root device.
How to use it?
Find out root bus number to acpi root name mapping from dmesg or /sys
echo "\_SB.PCIB 3" > /proc/acpi/sci/notify
to remove root bus
Signed-off-by: Yinghai Lu
Cc: Len Brown
Cc: linux-a...@vger.kernel.org
---
From: Jiang Liu
When hot-plugging PCI root bridge, acpi_pci_drivers' add()/remove()
methods should be invoked to notify registered drivers.
-v2: Move add calling to acpi_pci_root_start by Yinghai
Signed-off-by: Jiang Liu
Signed-off-by: Yinghai Lu
---
drivers/acpi/pci_root.c | 11
After we get hot-added ioapic registered.
pci_enable_bridges will try to enable ioapic irq for pci bridge.
So need to move it down.
Or We can move out pcibios_enable_irq() out of pci_enable_device()
and call pcibios_enable_irq in pci_bus_add_devices ?
also will need to move ...
During stop pci drivers, it still need to access ioapic and iommu.
So need to make sure those drivers need to be stop at first.
Also change the acpi_pci_drivers remove calling sequence to reverse.
Signed-off-by: Yinghai Lu
---
drivers/acpi/pci_root.c | 11 +++
1 files changed, 7
This part is to make sure ioapic and dmar handling sequence.
aka need to start driver for ioapic and dmar before regular pci device drivers.
and need to stop dmar and ioapic driver after pci device drivers.
user acpi_pci_driver to start and stop ioapic and dmar driver.
could get from
Pre-installed will be handled later, need to skip the one only for hot-added
roots.
Otherwise will get anonying failing message about can not reserve, irq ...
Signed-off-by: Yinghai Lu
---
drivers/acpi/pci_root.c|3 ++-
drivers/acpi/pci_root_hp.c |1 +
include/acpi/acpi_bus.h|
From: Jiang Liu
There are two global lists inf file drivers/acpi/pci_root.c.
One is for registered acpi_pci_drivers, the other is for
enumerated ACPI PCI root bridge objects. These two global
lists may change dynamically when registering/deregistering
acpi_pci_drivers or adding/removing ACPI PCI
otherwise irq_desc for pci bridge with hot-added ioapic can not be
on local node.
Signed-off-by: Yinghai Lu
---
drivers/pci/probe.c |1 +
1 files changed, 1 insertions(+), 0 deletions(-)
diff --git a/drivers/pci/probe.c b/drivers/pci/probe.c
index 59345ac..dd84224 100644
---
From: Jiang Liu
Use normal list for struct acpi_pci_driver to simplify code.
Signed-off-by: Jiang Liu
Signed-off-by: Yinghai Lu
---
drivers/acpi/pci_root.c | 16 +++-
include/linux/acpi.h|2 +-
2 files changed, 4 insertions(+), 14 deletions(-)
diff --git
Remove local defined acpiphp_hp_work.
Signed-off-by: Yinghai Lu
Cc: Len Brown
Cc: linux-a...@vger.kernel.org
---
drivers/pci/hotplug/acpiphp_glue.c | 42 +--
1 files changed, 6 insertions(+), 36 deletions(-)
diff --git a/drivers/pci/hotplug/acpiphp_glue.c
It will call new added pci_stop_and_remove_bus() to stop/remove pci root bus.
Also checking if that pci_root_bus get removed already in bus remove in /sys
Signed-off-by: Yinghai Lu
Cc: Len Brown
Cc: linux-a...@vger.kernel.org
---
drivers/acpi/pci_root.c | 14 ++
1 files changed,
It causes confusion.
We may only need acpi hp for pci host bridge.
Split host bridge hot-add support to pci_root_hp, and keep acpiphp simple.
Also remove not used res_lock in the struct.
-v2: put back pci_root_hp change in one patch
-v3: add pcibios_resource_survey_bus() calling
-v4: remove
No external user anymore.
Signed-off-by: Yinghai Lu
Cc: Len Brown
Cc: linux-a...@vger.kernel.org
---
drivers/acpi/osl.c |3 +--
include/acpi/acpiosxf.h |2 --
2 files changed, 1 insertions(+), 4 deletions(-)
diff --git a/drivers/acpi/osl.c b/drivers/acpi/osl.c
index
1 - 100 of 500 matches
Mail list logo
