On Thu, Dec 27, 2012 at 9:59 AM, Kent Overstreet wrote:
> @@ -1585,18 +1278,27 @@ static int io_submit_one(struct kioctx *ctx, struct
> iocb __user *user_iocb,
> * don't see ctx->dead set here, io_destroy() waits for our IO to
> * finish.
> */
> - if (ctx->dead) {
On Thu, Dec 27, 2012 at 9:59 AM, Kent Overstreet wrote:
> - /*
> -* This is so that aio_complete knows it doesn't need to
> -* pull the iocb off the run list (We can't just call
> -* INIT_LIST_HEAD because we don't want a kick_iocb to
> -* queue this on the ru
On Thu, Dec 27, 2012 at 11:42 PM, Zlatko Calusic
wrote:
> On 21.12.2012 12:51, Hillf Danton wrote:
>>
>> On Thu, Dec 20, 2012 at 7:25 AM, Zlatko Calusic
>> wrote:
>>>
>>> static unsigned long balance_pgdat(pg_data_t *pgdat, int order,
>>>
On Fri, Dec 28, 2012 at 12:08 AM, Alex Xu wrote:
> On 25/12/12 07:05 AM, Hillf Danton wrote:
>> On Tue, Dec 25, 2012 at 12:38 PM, Zhouping Liu wrote:
>>> Hello all,
>>>
>>> I found the below kernel bug using latest mainline(637704cbc95),
>>> my hardware has 2 numa nodes, and it's easy to reproduc
On Thu, Dec 20, 2012 at 11:44:04AM +0200, Hiroshi Doyu wrote:
> Initial support for Tegra 114 SoC. This is expected to be included in
> the board DTS files, Tegra 114 SoC based evaluation board family.
>
> Signed-off-by: Hiroshi Doyu
Hi,
I think it could be a good idea to start documenting the
On Fri, 2012-12-28 at 21:19 -0800, H. Peter Anvin wrote:
> Again, we could hack a simulator and try it.
Yeah, shouldn't be too hard to wedge into qemu/ovmf.
N�r��yb�X��ǧv�^�){.n�+{zX����ܨ}���Ơz�&j:+v���zZ+��+zf���h���~i���z� �w���?�&�)ߢf��^jǫy�m��@A�a���
0��h���i
On Fri, Dec 28, 2012 at 20:05 -0800, Eric W. Biederman wrote:
> Vasily Kulikov writes:
>
> > Currently there is completely no limiting in number of user namespaces
> > created by unprivileged users. One can freely create thousands of
> > user_ns'es and exhaust kernel memory without even bumping
Again, we could hack a simulator and try it.
Matthew Garrett wrote:
>On Fri, 2012-12-28 at 16:42 -0800, H. Peter Anvin wrote:
>> On 12/28/2012 03:39 PM, Matthew Garrett wrote:
>> > On Fri, 2012-12-28 at 15:32 -0800, H. Peter Anvin wrote:
>> >> No, that is explicitly the ordering that is excludab
On Fri, Dec 28, 2012 at 08:05:32PM -0800, Eric W. Biederman wrote:
> Yes. Gcc can't turn a tail call into a jump in even the most basic
> cases apparently.
What. The. Fuck?
You have introduced unlimited recursion on kernel stack. OK, it's
unpleasant, but it can happen to anybody. But then y
Hi Linus,
I'll leave it up to you to decide if you want to pull this or not --
it can wait to 3.9 as well.
This branch came in from Tony at the start of the merge window, and I
wanted to give it a little time in -next (but it got more than expected
due to travel).
It's a mix of lowish priority f
On Fri, 2012-12-28 at 16:42 -0800, H. Peter Anvin wrote:
> On 12/28/2012 03:39 PM, Matthew Garrett wrote:
> > On Fri, 2012-12-28 at 15:32 -0800, H. Peter Anvin wrote:
> >> No, that is explicitly the ordering that is excludable.
> >
> > Windows uses the UEFI time functions. The fact that they fail
Hi Linus,
The following changes since commit a49f0d1ea3ec94fc7cf33a7c36a16343b74bd565:
Linux 3.8-rc1 (2012-12-21 17:19:00 -0800)
are available in the git repository at:
git://git.kernel.org/pub/scm/linux/kernel/git/arm/arm-soc.git tags/fixes
for you to fetch changes up to 2e376799b2f6af2e8
From: Roland Dreier
print_prefix() passes a NULL buf to print_time() to get the length of
the time prefix; when printk times are enabled, the current code just
returns the constant 15, which matches the format "[%5lu.%06lu] " used
to print the time value. However, this is obviously incorrect whe
Vasily Kulikov writes:
> Currently there is completely no limiting in number of user namespaces
> created by unprivileged users. One can freely create thousands of
> user_ns'es and exhaust kernel memory without even bumping in
> RLIMIT_NPROC or similar.
First for a proper sense of scale it will
This patch is brought to you by the letter 'H'.
Commit 20b279 breaks compatiblity with older perf binaries when run with
precise modifier (:p or :pp) by requiring the exclude_guest attribute to be
set. Older binaries default exclude_guest to 0 (ie., wanting guest-based
samples) unless host only pr
Since commit 2aacdff entitled "cpufreq: Move common part from governors
to separate file", whenever the drivers that depend on this new file
(cpufreq_ondemand or cpufreq_conservative) are built as modules, a new
module named cpufreq_governor is created because the Makefile includes
cpufreq_governor
It looks pointless to pass max_vol to s5m8767_convert_voltage_to_sel().
Compare selected voltage to desc->max is enough to ensure selected voltage
is in supported range.
Signed-off-by: Axel Lin
---
drivers/regulator/s5m8767.c | 33 ++---
1 file changed, 10 insertion
On 12/28/2012 05:00 PM, joeyli wrote:
> 於 五，2012-12-28 於 17:43 +，Matthew Garrett 提到：
>> On Sat, 2012-12-29 at 00:26 +0800, Lee, Chun-Yi wrote:
>>> UEFI time services, GetTime(), SetTime(), GetWakeupTime(), SetWakeupTime()
>>> are also
>>> supported by other non-IA64 architecutre with UEFI BIOS
於 五，2012-12-28 於 17:43 +，Matthew Garrett 提到：
> On Sat, 2012-12-29 at 00:26 +0800, Lee, Chun-Yi wrote:
> > UEFI time services, GetTime(), SetTime(), GetWakeupTime(), SetWakeupTime()
> > are also
> > supported by other non-IA64 architecutre with UEFI BIOS, e.g. x86.
> >
> > This patch changed R
On 12/28/2012 06:33 PM, Fabio Baltieri wrote:
On Fri, Dec 28, 2012 at 05:45:54PM -0600, Larry Finger wrote:
I wonder if that's avoidable? The intention is not to create an additional
module, clearly.
It appears not to be possible. I don't know enough about to kmake to
understand why it is for
On 12/26/2012 07:27 AM, Shawn Guo wrote:
> On Tue, Nov 27, 2012 at 04:29:11PM -0700, Stephen Warren wrote:
>> From: Grant Likely
>>
>> The current rules have the .dtb files build in a different directory
>> from the .dts files. The only reason for this is that it was what
>> PowerPC has done histo
On 12/28/2012 03:39 PM, Matthew Garrett wrote:
> On Fri, 2012-12-28 at 15:32 -0800, H. Peter Anvin wrote:
>> No, that is explicitly the ordering that is excludable.
>
> Windows uses the UEFI time functions. The fact that they fail for us on
> a bunch of systems is just another symptom of our gener
From: Stephen Warren
if_changed (used by the *.dts->*.dtc rule) rebuilds files if they aren't
contained in $(targets). (make V=2 indicates this). Add $(dtb-y) to
$(targets) to prevent *.dtb from always being rebuilt. Note
This fixes a regression introduced by the .dtb rule rework in da4cbc6
"arm
From: Stephen Warren
if_changed (used by the *.dts->*.dtc rule) rebuilds files if they aren't
contained in $(targets). (make V=2 indicates this). Add $(dtb-y) to
$(targets) to prevent *.dtb from always being rebuilt.
This fixes a regression introduced by the .dtb rule rework in 499cd82
"ARM: dt:
On Thu, 2012-12-27 at 14:31 -0500, Rik van Riel wrote:
> to use a bigger/smaller one.
>
> I guess we want a larger value.
>
> With your hashed lock approach, we can get away with
> larger values - they will not penalize other locks
> the same way a single value per cpu might have.
Then, we absol
Em Fri, 28 Dec 2012 20:23:34 +0100
Geert Uytterhoeven escreveu:
> dma_common_mmap() and dma_common_get_sgtable() are defined in
> drivers/base/dma-mapping.c, and always compiled if CONFIG_HAS_DMA=y.
>
> However, their forward declarations and the inline functions defined on top
> of them (dma_mm
On Fri, Dec 28, 2012 at 05:45:54PM -0600, Larry Finger wrote:
> >I wonder if that's avoidable? The intention is not to create an additional
> >module, clearly.
>
> It appears not to be possible. I don't know enough about to kmake to
> understand why it is forcing a new module. Perhaps some expert
Since commit 2aacdff entitled "cpufreq: Move common part from governors to separate file", whenever the drivers that
depend on this new file (cpufreq_ondemand or cpufreq_conservative) are built as modules, a new module named
cpufreq_governor is created. It seems that kmake is smart enough to crea
On Fri, Dec 28, 2012 at 10:18:11AM -0800, David Decotigny wrote:
> Thank you, Josh,
>
> A few comments below, and the revised version shortly.
Responses below.
> On Thu, Dec 27, 2012 at 8:04 PM, Josh Triplett wrote:
> > On Thu, Dec 27, 2012 at 11:24:34AM -0800, David Decotigny wrote:
> >> In so
On 12/28/2012 05:01 PM, Rafael J. Wysocki wrote:
On Friday, December 28, 2012 04:17:24 PM Larry Finger wrote:
Since commit 2aacdff entitled "cpufreq: Move common part from governors to
separate file", whenever the drivers that depend on this new file
(cpufreq_ondemand or cpufreq_conservative) ar
On 12/28/2012 12:49 PM, Matthew Garrett wrote:
> On Fri, 2012-12-28 at 12:40 -0800, H. Peter Anvin wrote:
>
>>> I suspect that what we *should* do looks like:
>>>
>>> 1. If ACPI exports a Time and Alarm Device (ACPI000E) the use it;
>>> 2. If ACPI exports an PC/AT device (PNP0B00/1/2) then use it(
On Fri, 2012-12-28 at 15:32 -0800, H. Peter Anvin wrote:
> No, that is explicitly the ordering that is excludable.
Windows uses the UEFI time functions. The fact that they fail for us on
a bunch of systems is just another symptom of our general failure to
accurately mimic Windows when making runti
From: Larry Finger
Date: Thu, 27 Dec 2012 21:25:41 -0600
> With 3.8-rc1, the first call of pci_map_single() that is not checked
> with a corresponding pci_dma_mapping_error() call results in a warning
> with a splat as follows:
>
> WARNING: at lib/dma-debug.c:933 check_unmap+0x480/0x950()
> Hard
On Sat, 29 Dec 2012, Andreas Schwab wrote:
> Signed-off-by: Andreas Schwab
> ---
> scripts/tags.sh | 50 +-
> 1 file changed, 25 insertions(+), 25 deletions(-)
>
> diff --git a/scripts/tags.sh b/scripts/tags.sh
> index 08f06c0..4c53b7d 100755
> --
Hierarchical stats for cfq-iosched will need __blkg_prfill_rwstat().
Export it.
Signed-off-by: Tejun Heo
Reported-by: Fengguang Wu
---
Fengguang's build test discovered that cfq now uses
__blkg_prfill_rwstat() which isn't exported leading to build failures
when cfq is built as a module. Export
Signed-off-by: Andreas Schwab
---
scripts/tags.sh | 50 +-
1 file changed, 25 insertions(+), 25 deletions(-)
diff --git a/scripts/tags.sh b/scripts/tags.sh
index 08f06c0..4c53b7d 100755
--- a/scripts/tags.sh
+++ b/scripts/tags.sh
@@ -201,34 +201,34
On Friday, December 28, 2012 04:17:24 PM Larry Finger wrote:
> Since commit 2aacdff entitled "cpufreq: Move common part from governors to
> separate file", whenever the drivers that depend on this new file
> (cpufreq_ondemand or cpufreq_conservative) are built as modules, a new module
> named cp
On Fri, 28 Dec 2012, Guennadi Liakhovetski wrote:
> On Fri, 28 Dec 2012, Borislav Petkov wrote:
>
> > On Thu, Dec 27, 2012 at 11:36:13PM -0800, Eric W. Biederman wrote:
> > > git-ls-files | xargs fgrep 'struct f2fs_inode'
> > >
> > > That returns instantly and tells me where to look. If you can
On Fri, 28 Dec 2012, Borislav Petkov wrote:
> On Thu, Dec 27, 2012 at 11:36:13PM -0800, Eric W. Biederman wrote:
> > git-ls-files | xargs fgrep 'struct f2fs_inode'
> >
> > That returns instantly and tells me where to look. If you can do an
> > instant brute force search what is the point of an i
On Fri, Dec 28, 2012 at 5:01 PM, Linus Torvalds
wrote:
> On Fri, Dec 28, 2012 at 9:07 AM, wrote:
>> From: Christoffer Dall
>>
>> Unfortunately with !CONFIG_PAGEFLAGS_EXTENDED, (!PageHead) is false, and
>> (PageHead) is true, for tail pages. This breaks cache cleaning on some
>> ARM systems, an
Since commit 2aacdff entitled "cpufreq: Move common part from governors to
separate file", whenever the drivers that depend on this new file
(cpufreq_ondemand or cpufreq_conservative) are built as modules, a new module
named cpufreq_governor is created. It seems that kmake is smart enough to cre
On Tue, Aug 14, 2012 at 5:17 PM, Fengguang Wu wrote:
> Rationals from Eric:
>
> So I just looked a little deeper and it appears architectures that do
> not support atomic64_t are broken.
>
> The generic atomic64 support came in 2009 to support the perf subsystem
> with the expectation that all arc
On Fri, Dec 28, 2012 at 2:07 PM, Borislav Petkov wrote:
> On Thu, Dec 27, 2012 at 11:36:13PM -0800, Eric W. Biederman wrote:
>> git-ls-files | xargs fgrep 'struct f2fs_inode'
>>
>> That returns instantly and tells me where to look. If you can do an
>> instant brute force search what is the point
On Fri, Dec 28, 2012 at 9:07 AM, wrote:
> From: Christoffer Dall
>
> Unfortunately with !CONFIG_PAGEFLAGS_EXTENDED, (!PageHead) is false, and
> (PageHead) is true, for tail pages. This breaks cache cleaning on some
> ARM systems, and may cause other bugs.
So this already got committed earlier
On Fri, 2012-12-28 at 13:44 -0800, David Decotigny wrote:
> Thanks,
>
> Ok for the cpu_rmap_put helper. Will do this in v3 of this patch.
>
> Your comments suggest more refactoring, which might be better in the
> form of 1 or 2 additional patches that:
> - rename alloc_cpu_rmap & co according to
On Wed, Dec 26, 2012 at 01:04:02PM +0900, Jingoo Han wrote:
> On Saturday, December 22, 2012 12:06 PM, Kukjin Kim wrote
> > Tejun Heo wrote:
> > >
> > > There's no need to test whether a (delayed) work item in pending
> > > before queueing, flushing or cancelling it. Most uses are unnecessary
> >
On Fri, Dec 28, 2012 at 09:40:56PM +0100, Geert Uytterhoeven wrote:
> Subject: [PATCH] mips: Export min_low_pfn if CONFIG_FLATMEM
>
> If CONFIG_FLATMEM=y on MIPS, pfn_valid() (and thus virt_addr_valid()) uses
> min_low_pfn. However, min_low_pfn is not exported by the generic bootmem
> code.
>
>
Thanks,
Ok for the cpu_rmap_put helper. Will do this in v3 of this patch.
Your comments suggest more refactoring, which might be better in the
form of 1 or 2 additional patches that:
- rename alloc_cpu_rmap & co according to new conventions (cpu_rmap_*)
- remove the cpu_rmap sub-API altogether,
On Fri, Dec 21, 2012 at 08:21:25PM -0800, Greg Kroah-Hartman wrote:
> On Fri, Dec 21, 2012 at 05:57:11PM -0800, Tejun Heo wrote:
> > There's no need to test whether a (delayed) work item in pending
> > before queueing, flushing or cancelling it. Most uses are unnecessary
> > and quite a few of the
There's no need to test whether a (delayed) work item in pending
before queueing, flushing or cancelling it. Most uses are unnecessary
and quite a few of them are buggy.
Remove unnecessary pending tests from x86/mce. Only compile tested.
v2: Local var work removed from mce_schedule_work() as su
On Sat, Dec 22, 2012 at 09:22:13PM +0100, Johannes Berg wrote:
> On Fri, 2012-12-21 at 17:57 -0800, Tejun Heo wrote:
> > There's no need to test whether a (delayed) work item in pending
> > before queueing, flushing or cancelling it. Most uses are unnecessary
> > and quite a few of them are buggy.
On Sat, Dec 22, 2012 at 04:14:29PM +0200, Luciano Coelho wrote:
> On Fri, 2012-12-21 at 17:56 -0800, Tejun Heo wrote:
> > There's no need to test whether a (delayed) work item in pending
> > before queueing, flushing or cancelling it. Most uses are unnecessary
> > and quite a few of them are buggy
On Sat, Dec 22, 2012 at 09:55:04PM -0200, Henrique de Moraes Holschuh wrote:
> On Fri, 21 Dec 2012, Tejun Heo wrote:
> > There's no need to test whether a (delayed) work item in pending
> > before queueing, flushing or cancelling it. Most uses are unnecessary
> > and quite a few of them are buggy.
On Sat, Dec 22, 2012 at 02:29:57PM -0800, Bing Zhao wrote:
> Hi Tejun,
>
> Thanks for the patch.
>
> > Drop work_pending() test from mwifiex_sdio_card_reset(). As
> > work_pending() becomes %false before sdio_card_reset_worker() starts
> > executing, it doesn't really protect anything. reset_ho
On Sat, Dec 22, 2012 at 12:01:11AM -0800, David Miller wrote:
> From: Tejun Heo
> Date: Fri, 21 Dec 2012 17:56:53 -0800
>
> > There's no need to test whether a (delayed) work item in pending
> > before queueing, flushing or cancelling it. Most uses are unnecessary
> > and quite a few of them are
On Fri, 2012-12-28 at 16:09 +, Al Viro wrote:
> On Thu, Dec 27, 2012 at 11:36:13PM -0800, Eric W. Biederman wrote:
> > But then I am probably peculiar keeping an index of the source code in
> > my head. When I need to look for something and I don't know where to
> > find it I do.
> >
> > git-
>From a0a4bddd2779a51b6529afa113c5671ebcc21b14 Mon Sep 17 00:00:00 2001
From: Tejun Heo
Date: Fri, 28 Dec 2012 13:18:28 -0800
5edee61ede ("cgroup: cgroup_subsys->fork() should be called after the
task is added to css_set") removed cgroup_fork_callbacks() but forgot
to remove its dummy version for
From: Rafael J. Wysocki
As the kernel Bugzilla report #42696 indicates, it generally is not
sufficient to use _ADR to get an ACPI device node corresponding to
the given PCI device, because there may be multiple objects with
matching _ADR in the ACPI namespace (this probably is against the
spec, b
On Mon, Dec 24, 2012 at 10:25:00PM -0700, Stephen Warren wrote:
> On 12/21/2012 11:50 PM, Terje Bergström wrote:
> > On 21.12.2012 16:36, Thierry Reding wrote:
> >> On Fri, Dec 21, 2012 at 01:39:21PM +0200, Terje Bergstrom wrote:
> >>> +static struct platform_driver tegra_drm_platform_driver = {
>
On Fri, 2012-12-28 at 11:03 -0800, David Decotigny wrote:
> In some cases, free_irq_cpu_rmap() is called while holding a lock
> (eg. rtnl). This can lead to deadlocks, because it invokes
> flush_scheduled_work() which ends up waiting for whole system
> workqueue to flush, but some pending works mig
On Fri, 2012-12-28 at 12:40 -0800, H. Peter Anvin wrote:
> > I suspect that what we *should* do looks like:
> >
> > 1. If ACPI exports a Time and Alarm Device (ACPI000E) the use it;
> > 2. If ACPI exports an PC/AT device (PNP0B00/1/2) then use it(*);
> > 3. If we have an EFI RTC use it;
> > 4. Pro
On Fri, Dec 28, 2012 at 11:49:53AM -0800, Eric W. Biederman wrote:
> Al Viro writes:
>
> > On Thu, Dec 27, 2012 at 11:36:13PM -0800, Eric W. Biederman wrote:
> >> But then I am probably peculiar keeping an index of the source code in
> >> my head. When I need to look for something and I don't kn
Add two blkcg_policy methods, ->online_pd_fn() and ->offline_pd_fn(),
which are invoked as the policy_data gets activated and deactivated
while holding both blkcg and q locks.
Also, add blkcg_gq->online bool, which is set and cleared as the
blkcg_gq gets activated and deactivated. This flag also
Add pd->plid so that the policy a pd belongs to can be identified
easily. This will be used to implement hierarchical blkg_[rw]stats.
Signed-off-by: Tejun Heo
---
block/blk-cgroup.c | 2 ++
block/blk-cgroup.h | 3 ++-
2 files changed, 4 insertions(+), 1 deletion(-)
diff --git a/block/blk-cgrou
Instead of holding blkcg->lock while walking ->blkg_list and executing
prfill(), RCU walk ->blkg_list and hold the blkg's queue lock while
executing prfill(). This makes prfill() implementations easier as
stats are mostly protected by queue lock.
This will be used to implement hierarchical stats.
* Rename out_* labels to err_*.
* Do ERR_PTR() conversion once in the error return path.
This patch is cosmetic and to prepare for the hierarchy support.
Signed-off-by: Tejun Heo
Acked-by: Vivek Goyal
---
block/blk-cgroup.c | 15 +++
1 file changed, 7 insertions(+), 8 deletions(-)
From: Vivek Goyal
Use of local varibale "n" seems to be unnecessary. Remove it. This brings
it inline with function __cfq_group_st_add(), which is also doing the
similar operation of adding a group to a rb tree.
No functionality change here.
Signed-off-by: Vivek Goyal
Acked-by: Jeff Moyer
Sig
On 12/28/2012 11:17 AM, Matthew Garrett wrote:
On Fri, 2012-12-28 at 11:07 -0800, H. Peter Anvin wrote:
We do have such machines, which is why this change has been reverted twice
already. I believe we should stick to the priority scheme I proposed a few
weeks ago.
I seem to have missed that
Fix:
ERROR: "parport_pc_probe_port" [drivers/staging/sb105x/sb105x.ko] undefined!
by making the code calling it conditional. Also remove the external reference to
parport_pc_probe_port() from the sb105x source file, as it doesn't belong there.
Signed-off-by: Guenter Roeck
---
drivers/staging/sb
If CONFIG_FLATMEM=y on MIPS, pfn_valid() (and thus virt_addr_valid()) uses
min_low_pfn. However, min_low_pfn is not exported by the generic bootmem
code.
As of commit e52a29326462badd9ceec90a9eb2ac2a8550e02e ("aoe: avoid races
between device destruction and discovery"), aoeblk_open() uses
virt_add
From: Vivek Goyal
Currently CFQ has three IO classes, RT, BE and IDLE. At many a places we
are calling workloads belonging to these classes as "prio". This gets
very confusing as one starts to associate it with ioprio.
So this patch just does bunch of renaming so that reading code becomes
easier
From: Vivek Goyal
choose_service_tree() selects/sets both wl_class and wl_type. Rename it to
choose_wl_class_and_type() to make it very clear.
cfq_choose_wl() only selects and sets wl_type. It is easy to confuse
it with choose_st(). So rename it to cfq_choose_wl_type() to make
it clear what doe
Currently a child blkg (blkcg_gq) can be created even if its parent
doesn't exist. ie. Given a blkg, it's not guaranteed that its
ancestors will exist. This makes it difficult to implement proper
hierarchy support for blkcg policies.
Always create blkgs recursively and make a child blkg hold a r
Reorganize such that
* __blkg_lookup() takes bool param @update_hint to determine whether
to update hint.
* __blkg_lookup_create() no longer performs lookup before trying to
create. Renamed to blkg_create().
* blkg_lookup_create() now performs lookup and then invokes
blkg_create() if look
From: Vivek Goyal
Some more renaming. Again making the code uniform w.r.t use of
wl_class/class to represent IO class (RT, BE, IDLE) and using
wl_type/type to represent subclass (SYNC, SYNC-IDLE, ASYNC).
At places this patch shortens the string "workload" to "wl".
Renamed "saved_workload" to "sa
Currently, cfqg charges are scaled directly according to cfqg->weight.
Regardless of the number of active cfqgs or the amount of active
weights, a given weight value always scales charge the same way. This
works fine as long as all cfqgs are treated equally regardless of
their positions in the hie
Rename blkg_rwstat_sum() to blkg_rwstat_total(). sum will be used for
summing up stats from multiple blkgs.
Signed-off-by: Tejun Heo
---
block/blk-cgroup.h | 4 ++--
block/cfq-iosched.c | 4 ++--
2 files changed, 4 insertions(+), 4 deletions(-)
diff --git a/block/blk-cgroup.h b/block/blk-cgro
Implement blkg_[rw]stat_recursive_sum() and blkg_[rw]stat_merge().
The former two collect the [rw]stats designated by the target policy
data and offset from the pd's subtree. The latter two add one
[rw]stat to another.
Note that the recursive sum functions require the queue lock to be
held on ent
Separate out cfqg_stats_reset() which takes struct cfqg_stats * from
cfq_pd_reset_stats() and move the latter to where other pd methods are
defined. cfqg_stats_reset() will be used to implement hierarchical
stats.
Signed-off-by: Tejun Heo
---
block/cfq-iosched.c | 13 +
1 file chang
To prepare for blkcg hierarchy support, add cfqg->nr_active and
->children_weight. cfqg->nr_active counts the number of active cfqgs
at the cfqg's level and ->children_weight is sum of weights of those
cfqgs. The level covers itself (cfqg->leaf_weight) and immediate
children.
The two values are
cfq_group_slice() calculates slice by taking a fraction of
cfq_target_latency according to the ratio of cfqg->weight against
service_tree->total_weight. This currently works only because all
cfqgs are treated to be at the same level.
To prepare for proper hierarchy support, convert cfq_group_slic
cfq blkcg is about to grow proper hierarchy handling, where a child
blkg's weight would nest inside the parent's. This makes tasks in a
blkg to compete against both tasks in the sibling blkgs and the tasks
of child blkgs.
We're gonna use the existing weight as the group weight which decides
the b
To support hierarchical stats, it's necessary to remember stats from
dead children. Add cfqg->dead_stats and make a dying cfqg transfer
its stats to the parent's dead-stats.
The transfer happens form ->pd_offline_fn() and it is possible that
there are some residual IOs completing afterwards. Cur
Unfortunately, at this point, there's no way to make the existing
statistics hierarchical without creating nasty surprises for the
existing users. Just create recursive counterpart of the existing
stats.
Signed-off-by: Tejun Heo
---
block/cfq-iosched.c | 105
RCU free request_queue so that blkcg_gq->q can be dereferenced under
RCU lock. This will be used to implement hierarchical stats.
Signed-off-by: Tejun Heo
---
block/blk-sysfs.c | 9 -
include/linux/blkdev.h | 2 ++
2 files changed, 10 insertions(+), 1 deletion(-)
diff --git a/bloc
From: Vivek Goyal
Currently we attach a character "S" or "A" to the cfqq, to represent
whether queues is sync or async. Add one more character "N" to represent
whether it is sync-noidle queue or sync queue. So now three different
type of queues will look as follows.
cfq1234S --> sync queus
cfq
With the previous two patches, all cfqg scheduling decisions are based
on vfraction and ready for hierarchy support. The only thing which
keeps the behavior flat is cfqg_flat_parent() which makes vfraction
calculation consider all non-root cfqgs children of the root cfqg.
Replace it with cfqg_par
From: Vivek Goyal
At quite a few places we use the keyword "service_tree". At some places,
especially local variables, I have abbreviated it to "st".
Also at couple of places moved binary operator "+" from beginning of line
to end of previous line, as per Tejun's feedback.
v2:
Reverted most of
blkg_alloc() was mistakenly checking blkcg_policy_enabled() twice.
The latter test should have been on whether pol->pd_init_fn() exists.
This doesn't cause actual problems because both blkcg policies
implement pol->pd_init_fn(). Fix it.
Signed-off-by: Tejun Heo
Acked-by: Vivek Goyal
---
block/
Hello,
This is the second iteration to implement blkcg hierarchy support in
cfq-iosched. Changes from the first task[L] are
* Vivek's cfq cleanup patches are included in the series for
convenience.
* Divide by zero bug when !CONFIG_CFQ_GROUP_IOSCHED reported by
Fengguang fixed.
* Updated t
commit da3d4c5fa56236dd924d77ffc4f982356816b93b ("get rid of pt_regs
argument of do_execve()") removed the parameter, but forgot to update
hpux_execve():
arch/parisc/hpux/fs.c: In function 'hpux_execve':
arch/parisc/hpux/fs.c:47:6: error: too many arguments to function 'do_execve'
Signed-off-by:
3.2-stable review patch. If anyone has any objections, please let me know.
--
From: Seth Heasley
commit 8bc039a1e15a72da8426b84293723fb7181f0b5e upstream.
This patch adds the HD Audio DeviceIDs for the Intel Lynx Point PCH.
Signed-off-by: Seth Heasley
Signed-off-by: Takashi
3.2-stable review patch. If anyone has any objections, please let me know.
--
From: Petr Matousek
commit 6d1068b3a98519247d8ba4ec85cd40ac136dbdf9 upstream.
On hosts without the XSAVE support unprivileged local user can trigger
oops similar to the one below by setting X86_CR4_O
3.2-stable review patch. If anyone has any objections, please let me know.
--
From: Wu Fengguang
commit 6edc59e602b36cd3c95a426ef6e8cad0344af8c7 upstream.
[the order sorted by tiwai]
Signed-off-by: Wu Fengguang
Signed-off-by: Takashi Iwai
Signed-off-by: Ben Hutchings
---
This is the start of the stable review cycle for the 3.2.36 release.
There are 173 patches in this series, which will be posted as responses
to this one. If anyone has any issues with these being applied, please
let me know.
Responses should be made by Wed Jan 2 20:00:00 GMT 2013.
Anything recei
3.2-stable review patch. If anyone has any objections, please let me know.
--
From: Oleg Nesterov
This patch is only for pre-v3.3 stable trees which backported
b40a7959 "freezer: exec should clear PF_NOFREEZE along with PF_KTHREAD".
v3.3+ doesn't need this fix.
b40a7959 is the
On Fri, Dec 28, 2012 at 6:42 AM, JoonSoo Kim wrote:
>
> I have a different idea.
> How about removing fallback allocation in bootmem.c completely?
> I don't know why it is there exactly.
> But, warning for 'slab_is_available()' is there for a long time.
> So, most people who misuse fallback alloca
3.2-stable review patch. If anyone has any objections, please let me know.
--
From: Alan Cox
commit 879dca019dc43a1622edca3e7dde644b14b5acc5 upstream.
We handle NOTIFY_THROTTLING so don't then fall through to unsupported event.
Signed-off-by: Alan Cox
Signed-off-by: Rafael J
3.2-stable review patch. If anyone has any objections, please let me know.
--
From: David Milburn
commit b03e66a6be91f8389fcd902ab6c1563db1c9c06b upstream.
If kdump is triggered with pending IO, controller may not respond causing
kdump to fail.
http://marc.info/?l=linux-ide&m
On Fri, 2012-12-28 at 20:27 +0100, Geert Uytterhoeven wrote:
> For removed files, get_maintainer.pl doesn't find any maintainers (besides
> the default linux-kernel@vger.kernel.org), as it only looks at the "+++"
> lines, which are "/dev/null" for removals.
> Fix this by extending the parsing to th
1 - 100 of 378 matches
Mail list logo
