On Mon, Nov 19, 2012 at 03:39:12PM +0800, Qing Xu wrote:
> From: Qing Xu
JFYI, I added the following text:
This patch adds support for initializing platform data from the
device-tree description.
Plus added include.
> Signed-off-by: Qing Xu
> ---
Applied, thank you!
--
To unsubscribe
On Mon, Nov 19, 2012 at 7:57 AM, Liu, Chuansheng
wrote:
> Rechecked these codes, the trace event runtime_pm_status is added newly, this
> is different with vanilla
> Linux.
Not sure I'm following - can you point out which tree are you working with ?
> So I still think that calling
From: Qing Xu
Signed-off-by: Qing Xu
---
drivers/power/max8925_power.c | 50 -
1 files changed, 49 insertions(+), 1 deletions(-)
diff --git a/drivers/power/max8925_power.c b/drivers/power/max8925_power.c
index daa333b..2d77107 100644
---
On 11/05/2012 09:08 AM, Rusty Russell wrote:
Jason Wang writes:
+struct virtnet_info {
+ u16 num_queue_pairs;/* # of RX/TX vq pairs */
+ u16 total_queue_pairs;
+
+ struct send_queue *sq;
+ struct receive_queue *rq;
+ struct virtqueue *cvq;
+
+
Check the waitqueue task list to be non empty before entering the critical
section. This prevents locking the spin lock needlessly in case the queue
was empty, and therefor also prevent scheduling overhead on a PREEMPT_RT
system.
Signed-off-by: Ivo Sieben
---
a second repost of this patch v2:
On 11/18/2012 04:30 AM, Luming Yu wrote:
> I'd be glad to do anything to push this tool into upstream. Please
> let me know your thoughts. Thanks /l
I'm also happy to help test. I'll take a look over the TG holiday at
your latest version.
Jon.
--
To unsubscribe from this list: send the
On 11/19/2012 03:06 PM, Anton Vorontsov wrote:
On Mon, Nov 19, 2012 at 02:52:24PM +0800, Qing Xu wrote:
From: Qing Xu
Signed-off-by: Qing Xu
---
drivers/power/max8925_power.c | 61 +---
1 files changed, 56 insertions(+), 5 deletions(-)
diff --git
Hi,
I'm Ryo Onodera. I created this very tiny patch to enable /proc/diskstats for
/dev/ramX.
I'm contacting Nick Piggin as the maintainer of ramdisk ram block.
This is my first patch to the Linux kernel. Sorry if this is not the expected
way of submitting patches.
Regards
Ryo Onodera
On Mon,
Hi Mathieu,
On Sun, Nov 11, 2012 at 01:24:48PM -0700, mathieu.poir...@linaro.org wrote:
> From: "Mathieu J. Poirier"
>
> This patch adds keyreset functionality to the sysrq driver. It
> allows certain button/key combinations to be used in order to
> trigger device resets.
>
> The first time
There are two reasons to remove the "chip" parameter in nand_get_device():
[1] The nand_release_device() does not have the "chip" parameter.
[2] We can get the nand_chip by the mtd->priv field.
This patch removes the "chip" parameter in nand_get_device().
Signed-off-by: Huang Shijie
---
The nand_get_device() does not select the chip, but nand_release_device()
does de-select the chip. It is really strange.
With the current code, nand_sync() will de-select the chip, even if the chip
has never been selected.
To make the balance of select/de-select chip, it's better to remove the
On Fri, Nov 16, 2012 at 06:54:43PM +0200, Johan Meiring wrote:
> This commit sorts out several coding style issues, including the use of C99
> // comments, spaces at the start of lines, spaces before tabs, incorrect
> brace placement, and macros with complex values that have not been wrapped in
Signed-off-by: Ryo Onodera
---
drivers/block/brd.c |7 +++
1 file changed, 7 insertions(+)
diff --git a/drivers/block/brd.c b/drivers/block/brd.c
index 531ceb3..181b2d7 100644
--- a/drivers/block/brd.c
+++ b/drivers/block/brd.c
@@ -332,6 +332,7 @@ static void brd_make_request(struct
On 11/19/2012 02:45 PM, Thierry Reding wrote:
> * PGP Signed by an unknown key
>
> On Mon, Nov 19, 2012 at 02:18:51PM +0800, Mark Zhang wrote:
>> On 11/17/2012 12:29 AM, Stephen Warren wrote:
>>> On 11/15/2012 09:58 PM, Mark Zhang wrote:
This patch is based on Thierry's drm patch for
Tejun Heo writes:
> On Sat, Nov 17, 2012 at 12:45:48PM +0100, David Herrmann wrote:
>> Sysfs doesn't allow two devices with the same name, but we register a
>> sysfs entry for each cuse device without checking for name collisions.
>> This extends the registration to first check whether the name
On Fri, Nov 16, 2012 at 06:52:39PM +0200, Johan Meiring wrote:
> COMP_ID_FW_AP,
> - 2, /*
> Variant /
^^
You fixed a bug here. :)
On Mon, Nov 19, 2012 at 02:52:24PM +0800, Qing Xu wrote:
> From: Qing Xu
>
> Signed-off-by: Qing Xu
> ---
> drivers/power/max8925_power.c | 61 +---
> 1 files changed, 56 insertions(+), 5 deletions(-)
>
> diff --git a/drivers/power/max8925_power.c
On Sun, Nov 18, 2012 at 09:28:12PM -0800, Josh Triplett wrote:
> Nothing outside of kernel/resource.c references reallocate_resource, so
> mark it static. This eliminates warnings from GCC
> (-Wmissing-prototypes) and Sparse (-Wdecl).
>
> kernel/resource.c:476:5: warning: no previous prototype
On 2012-11-19 00:42, David Rientjes wrote:
On Wed, 14 Nov 2012, Nithin Nayak Sujir wrote:
On 11/14/2012 07:30 PM, David Rientjes wrote:
> On Wed, 14 Nov 2012, Nithin Nayak Sujir wrote:
>
> > This was fixed by
> >
> > commit de0a41484c47d783dd4d442914815076aa2caac2
> > Author: Paul Gortmaker
>
+-- On Sun, 18 Nov 2012, Kees Cook wrote --+
| This is the second problem. I view this as less critical because it's only
| 64 instead of 4, but it certainly should be solved as well.
I don't mean to be rude, but the patch I had sent solves both of these
problems with much less performance
On 11/19/2012 08:24 AM, Anton Vorontsov wrote:
On Tue, Nov 06, 2012 at 03:42:44PM +0800, Qing Xu wrote:
From: Qing Xu
Signed-off-by: Qing Xu
---
W/ this patch I'm getting this warning:
CC drivers/power/max8925_power.o
drivers/power/max8925_power.c: In function
From: Qing Xu
Signed-off-by: Qing Xu
---
drivers/power/max8925_power.c | 61 +---
1 files changed, 56 insertions(+), 5 deletions(-)
diff --git a/drivers/power/max8925_power.c b/drivers/power/max8925_power.c
index daa333b..aca3f68 100644
---
On Mon, Nov 19, 2012 at 12:51 AM, Dave Chinner wrote:
> On Sun, Nov 18, 2012 at 04:29:22PM +0100, Torsten Kaiser wrote:
>> On Sun, Nov 18, 2012 at 11:24 AM, Torsten Kaiser
>> wrote:
>> > On Tue, Oct 30, 2012 at 9:37 PM, Torsten Kaiser
>> > wrote:
>> >> I will keep LOCKDEP enabled on that
Hi there!
Looks like it is bug in scsi.h with missing definitions in 3.2.34.
I compared scsi.h with one in 3.2.29 kernel and there are lots of
definitions missing in 3.2.34:
18014 bytes in old file and only 7037 in new one.
CC block/scsi_ioctl.o
block/scsi_ioctl.c:1:0: warning: SSE
On Mon, Nov 19, 2012 at 02:18:51PM +0800, Mark Zhang wrote:
> On 11/17/2012 12:29 AM, Stephen Warren wrote:
> > On 11/15/2012 09:58 PM, Mark Zhang wrote:
> >> This patch is based on Thierry's drm patch for Tegra20:
> >> - [PATCH v2 0/6] Device tree updates for host1x support
> >> - [PATCH v3 0/2]
Fix the below warning
drivers/spi/spi-omap2-mcspi.c:336:34: warning: symbol 'tx' shadows an earlier
one
drivers/spi/spi-omap2-mcspi.c:327:12: originally declared here
So delete the u8 tx as it is assigned and not used(resigned afterwards).
Signed-off-by: Shubhrajyoti D
---
On 19 November 2012 12:05, Rajanikanth HV wrote:
> On 19 November 2012 12:00, Viresh Kumar wrote:
>> Firstly you tried square braces [ ], I am not sure if that is allowed.
>> Can you point me to the specification?
> http://www.devicetree.org/Device_Tree_Usage
> "
> a-byte-data-property = [0x01
On 19 November 2012 12:00, Viresh Kumar wrote:
> Firstly you tried square braces [ ], I am not sure if that is allowed.
> Can you point me to the specification?
http://www.devicetree.org/Device_Tree_Usage
"
a-byte-data-property = [0x01 0x23 0x34 0x56];
"
>
> And simply passing 0x50, 0x60 etc..
On 19 November 2012 11:54, Rajanikanth HV wrote:
>> data1 = /bits/ 8 <0x50 0x60 0x70>;
> as per spec, format for data byte defines will be:
> data1 = [ 0x50 0x60 0x70 ];
> however, i see a parse error from device tree compiler when i tried.
Firstly you tried square braces [ ], I am not sure if
> > +static int ps_get_cur_chrage_cntl_limit(struct thermal_cooling_device
> > +*tcd,
>
> Typo, "charge".
>
> > + unsigned long *state)
> > +{
> > + struct power_supply *psy;
> > + union power_supply_propval val;
> > + int ret;
> > +
> > +
On 19 November 2012 09:24, Viresh Kumar wrote:
> On 12 November 2012 09:03, Viresh Kumar wrote:
>> On 12 November 2012 01:12, Rob Herring wrote:
>>> I don't think the size is stored in the dtb. It is only in the dts. You
>>> need to define the size in the binding definitions and use '/bits/'
On 11/05/2012 07:46 AM, Rusty Russell wrote:
Jason Wang writes:
This patch implement the {set|get}_channels method of ethool to allow user to
change the number of queues dymaically when the device is running. This would
let the user to tune the device for specific applications.
...
+
On 11/17/2012 12:29 AM, Stephen Warren wrote:
> On 11/15/2012 09:58 PM, Mark Zhang wrote:
>> This patch is based on Thierry's drm patch for Tegra20:
>> - [PATCH v2 0/6] Device tree updates for host1x support
>> - [PATCH v3 0/2] NVIDIA Tegra DRM driver
>>
>> It adds the support for NVIDIA Tegra30.
On 11/05/2012 07:16 AM, Rusty Russell wrote:
Jason Wang writes:
This addes multiqueue support to virtio_net driver. There's two mode supported:
single queue pair mode and multiple queue pairs mode. An obvious
difference compared with a physical mq card is that virtio-net reserve
first two
On 11/16/2012 05:12 PM, Paul Mackerras wrote:
> On Fri, Nov 16, 2012 at 02:29:04PM +0530, Anshuman Khandual wrote:
>> Signed-off-by: Anshuman Khandual
>
> That's not a sufficient description of why you are making this
> change. In particular, what is the motivation for and impact of using
>
Rob,
On Sat, Nov 17, 2012 at 6:11 AM, Rob Landley wrote:
> On 11/16/2012 01:01:42 AM, Michael Kerrisk wrote:
>>
>> After migrating most of the information in
>> Documentation/filesystems/proc.txt to the proc(5) man page,
>> Jim Paris pointed out to me that the description of SwapFree
>> in the
> -Original Message-
> From: Ohad Ben-Cohen [mailto:o...@wizery.com]
> Sent: Saturday, November 17, 2012 3:33 AM
> To: Liu, Chuansheng
> Cc: Chris Ball; linux-...@vger.kernel.org; linux-kernel@vger.kernel.org
> Subject: Re: [PATCH] mmc,sdio: Fix the panic due to devname NULL when
>
[In addition to the patch mails sent as replies to this email, I've made this
series available for pulling as a signed tag from:
git://git.kernel.org/pub/scm/linux/kernel/git/josh/linux
tags/Wmissing-prototypes-allnoconfig
for you to fetch changes up to
Nothing outside of kernel/trace/trace.c references tracing_dentry_percpu.
Signed-off-by: Josh Triplett
---
kernel/trace/trace.c |2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/kernel/trace/trace.c b/kernel/trace/trace.c
index 31e4f55..032f9ee 100644
---
Nothing outside of fs/sysfs/file.c references this function, so mark it static.
Signed-off-by: Josh Triplett
---
fs/sysfs/file.c |4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/fs/sysfs/file.c b/fs/sysfs/file.c
index 00012e3..602f56d 100644
--- a/fs/sysfs/file.c
+++
Nothing outside of drivers/tty/tty_io.c references these functions, so
mark them static.
Signed-off-by: Josh Triplett
---
drivers/tty/tty_io.c |4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/drivers/tty/tty_io.c b/drivers/tty/tty_io.c
index 2ea176b..023d1fd 100644
---
Both rcutiny and rcutree define a helper funtion
rcu_is_cpu_rrupt_from_idle, each used exactly once, later in the same
file. Declare these helper functions static.
Signed-off-by: Josh Triplett
---
kernel/rcutiny.c |2 +-
kernel/rcutree.c |2 +-
2 files changed, 2 insertions(+), 2
On 2012/11/14 4:27, Tejun Heo wrote:
>>From bbf2566f9f4fc79ff3320e83cafb69533efc9ea0 Mon Sep 17 00:00:00 2001
> From: Tejun Heo
> Date: Tue, 13 Nov 2012 12:21:50 -0800
>
> Currently CGRP_CPUSET_CLONE_CHILDREN triggers ->post_clone(). Now
> that clone_children is cpuset specific, there's no
On 2012/11/14 4:27, Tejun Heo wrote:
>>From 6e405c1ae4018d813e8ed9e0bd463d6976aebfa8 Mon Sep 17 00:00:00 2001
> From: Tejun Heo
> Date: Tue, 13 Nov 2012 12:21:50 -0800
>
> clone_children is only meaningful for cpuset and will stay that way.
> Rename the flag to reflect that and update
C files should include the header files that prototype their functions.
This keeps the types in sync, and eliminates warnings from GCC
(-Wmissing-prototypes) and Sparse (-Wdecl).
Signed-off-by: Josh Triplett
---
kernel/time/timekeeping.c |1 +
1 file changed, 1 insertion(+)
diff --git
Nothing outside of arch/x86/kernel/hpet.c references this function, so
mark it static.
Signed-off-by: Josh Triplett
---
arch/x86/kernel/hpet.c |2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/arch/x86/kernel/hpet.c b/arch/x86/kernel/hpet.c
index 1460a5d..2dd92ae 100644
---
C files should include the header files that prototype their functions.
This keeps the types in sync, and eliminates warnings from GCC
(-Wmissing-prototypes) and Sparse (-Wdecl).
lib/iommu-helper.c:9:5: warning: no previous prototype for
‘iommu_is_span_boundary’ [-Wmissing-prototypes]
C files should include the header files that prototype their functions.
This keeps the types in sync, and eliminates warnings from GCC
(-Wmissing-prototypes) and Sparse (-Wdecl).
Signed-off-by: Josh Triplett
---
drivers/tty/vt/vt_ioctl.c |1 +
1 file changed, 1 insertion(+)
diff --git
Nothing outside of block/blk-lib.c references __blkdev_issue_zeroout, so
mark it static.
Signed-off-by: Josh Triplett
---
block/blk-lib.c |2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/block/blk-lib.c b/block/blk-lib.c
index 9373b58..b19377f 100644
--- a/block/blk-lib.c
On 11/16/12 04:26, Dave, Tushar N wrote:
>> Would you please help to fine the offset of max payload size in eeprom?
>> I'd like to have a try to modify it by ethtool.
>
> It is defined using bit 8 of word 0x1A.
> Bit value 0 = 128B , bit value 1 = 256B
Hi Tushar,
I checked one of my server
arch/x86/kernel/apic/io_apic.c defines acpi_get_override_irq
unconditionally, but linux/acpi.h only prototypes it if CONFIG_ACPI=y,
and nothing calls it unless CONFIG_ACPI=y. Make the definition itself
conditional on CONFIG_ACPI. This saves space when CONFIG_ACPI=n, and
eliminates warnings from
Nothing outside of kernel/resource.c references reallocate_resource, so
mark it static. This eliminates warnings from GCC
(-Wmissing-prototypes) and Sparse (-Wdecl).
kernel/resource.c:476:5: warning: no previous prototype for
‘reallocate_resource’ [-Wmissing-prototypes]
Signed-off-by: Josh
C files should include the header files that prototype their functions.
This keeps the types in sync, and eliminates warnings from GCC
(-Wmissing-prototypes) and Sparse (-Wdecl).
arch/x86/kernel/apic/apic_flat_64.c:46:6: warning: no previous prototype for
‘flat_init_apic_ldr’
arch/x86/vdso/vclock_gettime.c and arch/x86/vdso/vgetcpu.c define
several functions prefixed by __vdso_*, used in the alias definitions
for the actual vdso symbols calls. Add prototypes of these functions
right before their definitions, to satisfy gcc (-Wmissing-prototypes)
and Sparse (-Wdecl).
kernel/sched/core.c defines a function sched_set_stop_task, only
referenced from kernel/stop_machine.c via an extern declaration; no
header file defines a prototype for sched_set_stop_task. Add a
prototype right before its definition, to satisfy GCC
(-Wmissing-prototypes) and Sparse (-Wdecl).
kernel/sched/fair.c defines stub versions of free_fair_sched_group,
alloc_fair_sched_group, and unregister_fair_sched_group, for use when
CONFIG_FAIR_GROUP_SCHED=n. However, sched.h only declares the
prototypes of those functions when CONFIG_CGROUP_SCHED=y, since the
scheduler core only calls
kernel/sched/fair.c defines a stub version of init_cfs_bandwidth when
CONFIG_CFS_BANDWIDTH=n. A few other places in the scheduler call
init_cfs_bandwidth, but only when CONFIG_FAIR_GROUP_SCHED=y, and
kernel/sched/sched.h only prototypes init_cfs_bandwidth when
CONFIG_FAIR_GROUP_SCHED=y. Wrap the
kernel/sched/rt.c defines stub versions of free_rt_sched_group and
alloc_rt_sched_group, for use when CONFIG_RT_GROUP_SCHED=n. However,
sched.h only declares the prototypes of those functions when
CONFIG_CGROUP_SCHED=y, since the scheduler core only calls those
functions when
include/linux/syscalls.h has prototypes for most syscalls, but other
prototypes appear in the corresponding asm/syscalls.h instead. Include
the latter in the former, giving include/linux/syscalls.h a more
complete set of syscall prototypes. This eliminates warnings from GCC
include/linux/syscalls.h contains prototypes for most syscall functions;
however, it lacked the prototypes for sys_rt_sigaction and
sys_rt_sigsuspend. Add them. This eliminates warnings from GCC
(-Wmissing-prototypes) and Sparse (-Wdecl).
kernel/signal.c:3183:2: warning: no previous prototype
arch/x86/kernel/irq_work.c defines a function smp_irq_work_interrupt,
only called from assembly. Add a prototype right before its definition,
to satisfy GCC (-Wmissing-prototypes) and Sparse (-Wdecl).
arch/x86/kernel/irq_work.c:12:6: warning: no previous prototype for
‘smp_irq_work_interrupt’
arch/x86/kernel/cpu/perf_event.c defines a function
arch_perf_update_userpage, overriding a __weak stub for that function
defined in kernel/events/core.c. Nothing currently prototypes that
function. Add a prototype right before the definition, to satisfy gcc
(-Wmissing-prototypes) and Sparse
Nothing outside of arch/x86/kernel/x86_init.c references iommu_init_noop
or iommu_shutdown_noop, so mark them static. This eliminates warnings
from GCC (-Wmissing-prototypes) and Sparse (-Wdecl).
arch/x86/kernel/x86_init.c:29:115: warning: no previous prototype for
‘iommu_init_noop’
C files should include the header files that prototype their functions.
This keeps the types in sync, and eliminates warnings from GCC
(-Wmissing-prototypes) and Sparse (-Wdecl).
arch/x86/kernel/time.c:69:116: warning: no previous prototype for
‘setup_default_timer_irq’ [-Wmissing-prototypes]
arch/x86/kernel/irqinit.c defines an __init function init_IRQ, called
from init/main.c. No header file prototypes init_IRQ, because
init/main.c tends to directly define prototypes for init functions it
calls, rather than including appropriate headers. So, add a prototype
of init_IRQ to
C files should include the header files that prototype their functions.
This keeps the types in sync, and eliminates warnings from GCC
(-Wmissing-prototypes) and Sparse (-Wdecl).
arch/x86/kernel/process.c:221:6: warning: no previous prototype for
‘__switch_to_xtra’ [-Wmissing-prototypes]
Commit 7856f6cce4a8cda8c1f94b99605c07d16b8d8dec renamed
mce_threshold_interrupt to smp_threshold_interrupt, but did not update
the prototype in arch/x86/include/asm/traps.h to match. Update the
prototype to the new name.
This also eliminates warnings from GCC (-Wmissing-prototypes) and Sparse
mm/page_alloc.c defines is_pageblock_removable_nolock unconditionally,
but include/linux/memory_hotplug.h only prototypes it when
CONFIG_MEMORY_HOTPLUG=y and CONFIG_MEMORY_HOTREMOVE=y. Add the
corresponding conditions around the definition, too.
This also eliminates warnings from GCC
mm/page_alloc.c defines a fallback version of __early_pfn_to_nid for
architectures without CONFIG_HAVE_ARCH_EARLY_PFN_TO_NID. Nothing
outside of mm/page_alloc.c calls that function, so mark it static. This
eliminates warnings from GCC (-Wmissing-prototypes) and Sparse (-Wdecl).
init/calibrate.c defines an empty stub of calibrate_delay_is_known with
__attribute__((weak)), for use on architectures without an alternate
implementation. Define a prototype of that function, to satisfy gcc
(-Wmissing-prototypes) and Sparse (-Wdecl).
init/calibrate.c:257:150: warning: no
kernel/events/hw_breakpoint.c defines an empty stub of
arch_unregister_hw_breakpoint with __weak, for use on architectures
without an alternate implementation. Define a prototype of that
function, to satisfy gcc (-Wmissing-prototypes) and Sparse (-Wdecl).
kernel/events/hw_breakpoint.c:247:28:
kernel/panic.c defines an empty stub of panic_smp_self_stop with __weak,
for use on architectures without an alternate implementation. Define a
prototype of that function, to satisfy gcc (-Wmissing-prototypes) and
Sparse (-Wdecl).
kernel/panic.c:55:28: warning: no previous prototype for
kernel/events/core.c defines an empty stub of arch_perf_update_userpage
with __weak, for use on architectures without an alternate
implementation. Define a prototype of that function, to satisfy gcc
(-Wmissing-prototypes) and Sparse (-Wdecl).
arch/x86/kernel/cpu/perf_event.c:1692:6: warning: no
Nothing outside of block/blk-core.c references blk_drain_queue, so mark
it static. This eliminates warnings from GCC (-Wmissing-prototypes) and
Sparse (-Wdecl).
block/blk-core.c:360:6: warning: no previous prototype for ‘blk_drain_queue’
[-Wmissing-prototypes]
Signed-off-by: Josh Triplett
---
kernel/fork.c defines an empty stub of arch_release_thread_info with
__weak, for use on architectures without an alternate implementation.
Define a prototype of that function, to satisfy gcc
(-Wmissing-prototypes) and Sparse (-Wdecl).
kernel/fork.c:137:28: warning: no previous prototype for
kernel/fork.c defines an __init function fork_init, called from
init/main.c. No header file prototypes fork_init, because init/main.c
tends to directly define prototypes for init functions it calls, rather
than including appropriate headers. So, add a prototype of fork_init to
kernel/fork.c
arch/x86/kernel/asm-offsets.c defines an intentionally unused function
common, just to slurp out the generated assembly and obtain the values
for some constants. Add a prototype of this function, to satisfy GCC
(-Wmissing-prototypes) and Sparse (-Wdecl).
arch/x86/kernel/asm-offsets.c:30:6:
kernel/bounds.c defines an intentionally unused function foo, just to
slurp out the generated assembly and obtain the values for some
constants. Add a prototype of this function, to satisfy GCC
(-Wmissing-prototypes) and Sparse (-Wdecl).
kernel/bounds.c:14:6: warning: no previous prototype for
C files should include the header files that prototype their functions.
This keeps the types in sync, and eliminates warnings from GCC
(-Wmissing-prototypes) and Sparse (-Wdecl).
kernel/resource.c:365:27: warning: no previous prototype for ‘page_is_ram’
[-Wmissing-prototypes]
Signed-off-by:
kernel/events/core.c defines a no-op stub of perf_cgroup_switch when
CONFIG_CGROUP_PERF=n. However, nothing actually calls
perf_cgroup_switch outside of an #ifdef CONFIG_CGROUP_PERF. So, drop
the stub.
This also eliminates a warning from gcc (-Wmissing-prototypes) and from
Sparse (-Wdecl).
Commit e43473b7f223ec866f7db273697e76c337c390f9 in September 2010
("blkio: Core implementation of throttle policy") re-added the
previously removed function kblockd_schedule_delayed_work, and added a
single call from throtl_schedule_delayed_work in block/blk-throttle.c.
Commit
include/linux/huge_mm.h declares is_vma_temporary_stack inside an ifdef
CONFIG_TRANSPARENT_HUGEPAGE; however, mm/rmap.c defines the function
unconditionally. Move the function outside of the ifdef.
This eliminates a warning from gcc (-Wmissing-prototypes) and from
Sparse (-Wdecl).
Nothing outside of mm/memory.c references copy_pte_range.
linux/huge_mm.h prototypes it, but nothing uses that prototype. Commit
71e3aac0724ffe8918992d76acfe3aad7d8724a5 in January 2011 explicitly made
copy_pte_range non-static, but no commit ever introduced a caller for
copy_pte_range outside of
include/linux/huge_mm.h declares handle_pte_fault inside an ifdef
CONFIG_TRANSPARENT_HUGEPAGE; however, mm/memory.c defines the function
unconditionally. Move the function outside of the ifdef.
This eliminates a warning from gcc (-Wmissing-prototypes) and from
Sparse (-Wdecl).
mm/internal.h declares vma_address inside an ifdef
CONFIG_TRANSPARENT_HUGEPAGE; however, mm/rmap.c defines the function
unconditionally. Move the function outside of the ifdef.
This eliminates a warning from gcc (-Wmissing-prototypes) and from
Sparse (-Wdecl).
mm/rmap.c:527:1: warning: no
Commit 3ee48b6af49cf534ca2f481ecc484b156a41451d added the
set_iounmap_nonlazy function for use on x86, but actually defined it on
all architectures. Only x86 prototypes this function (in asm/io.h), and
only x86 uses it (in arch/x86/kernel/crash_dump_64.c), so avoid defining
it on non-x86.
Nothing outside of arch/x86/kernel/cpu/common.c references
cpu_detect_tlb, so mark it static.
Signed-off-by: Josh Triplett
---
arch/x86/kernel/cpu/common.c |2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/arch/x86/kernel/cpu/common.c b/arch/x86/kernel/cpu/common.c
index
Nothing outside of kernel/events/core.c references perf_cgroup_switch,
perf_init_event, or perf_event_init_context; mark these functions
static.
Signed-off-by: Josh Triplett
---
kernel/events/core.c |6 +++---
1 file changed, 3 insertions(+), 3 deletions(-)
diff --git
Both mm/bootmem.c and mm/nobootmem.c define an internal function
___alloc_bootmem_node. Nothing outside of those source files references
that function, so declare it static in both cases.
Signed-off-by: Josh Triplett
---
mm/bootmem.c |8 +---
mm/nobootmem.c |8 +---
2 files
Commit f93a20541134fa767e8dc4eb32e956d30b9f6b92, in April 2010, added
the ability for hardware breakpoints to have variable "weight" in
allocations, and added a __weak function hw_breakpoint_weight that
architectures could override to set the weight of a breakpoint. No
architecture has ever done
Nothing outside of arch/x86/kernel/cpu/perf_event_intel_uncore.c
references the functions nhmex_mbox_alter_er, nhmex_rbox_alter_er,
uncore_alloc_box, or uncore_pmu_event_init; mark these funtions static.
Signed-off-by: Josh Triplett
---
arch/x86/kernel/cpu/perf_event_intel_uncore.c |8
Nothing outside of drivers/base/core.c references this function.
Signed-off-by: Josh Triplett
---
drivers/base/core.c |2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/drivers/base/core.c b/drivers/base/core.c
index abea76c..720441d 100644
--- a/drivers/base/core.c
+++
Several architectures define arch_irq_work_raise, and kernel/irq_work.c
defines a __weak version for architectures that do not. However,
nothing ever prototypes this function, leading to warnings from GCC and
Sparse like this:
kernel/irq_work.c:52:28: warning: no previous prototype for
Nothing outside of fs/locks.c references these functions, so mark them
static.
Signed-off-by: Josh Triplett
---
fs/locks.c |4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/fs/locks.c b/fs/locks.c
index a94e331..01051de 100644
--- a/fs/locks.c
+++ b/fs/locks.c
@@
Several source files reference swiotlb_late_init_with_default_size with
an extern prototype; declare it in linux/swiotlb.h instead.
This also ensures that lib/swiotlb.c has a matching prototype for
swiotlb_late_init_with_default_size, keeping the types in sync, and
eliminating warnings from GCC
default_init_apic_ldr only exists for use in the 32-bit-only
arch/x86/kernel/apic/probe_32.c, and asm/apic.h prototypes it in an
ifdef CONFIG_X86_32 block, but arch/x86/kernel/apic/apic.c defines the
function unconditionally. Wrap the definition in an ifdef CONFIG_X86_32
block.
Signed-off-by:
Nothing outside of arch/x86/kernel/apic/apic.c references enable_IR.
Signed-off-by: Josh Triplett
---
arch/x86/kernel/apic/apic.c |2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/arch/x86/kernel/apic/apic.c b/arch/x86/kernel/apic/apic.c
index b17416e..5a92aeb 100644
---
Nothing outside of drivers/tty/pty.c references pty_resize.
Signed-off-by: Josh Triplett
---
drivers/tty/pty.c |2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/drivers/tty/pty.c b/drivers/tty/pty.c
index a82b399..2fd1ae7 100644
--- a/drivers/tty/pty.c
+++
Nothing outside of kernel/sched/fair.c references scale_rt_power, so
mark it static.
Signed-off-by: Josh Triplett
---
kernel/sched/fair.c |2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c
index 6b800a1..7e29e80 100644
---
On 11/18/12 5:06 PM, "NeilBrown" wrote:
>
>Hi Dan,
> could you comment on this please? Would it make sense to arrange for
>errors
> to propagate up? Or should we arrange to do a software-fallback in the
>dma
> engine is a problem? What sort of things can cause error here anyway?
1 - 100 of 658 matches
Mail list logo
