This is where we describe the different new and generic options used by
the ST BCH driver.
Cc: devicet...@vger.kernel.org
Signed-off-by: Lee Jones
---
Documentation/devicetree/bindings/mtd/stm-nand.txt | 123 +
1 file changed, 123 insertions(+)
create mode 100644
First introduction of the driver. Includes the basic device struct
(some functionality isn't utilised as of yet) and supplies some of the
important resources required for basic running of the Controller.
Signed-off-by: Lee Jones
---
drivers/mtd/nand/Kconfig| 7 ++
Provide device information and flash layout for the NAND Micron
MT29F8G08ABABAWP chip found on the STM B2020 Development Board.
Signed-off-by: Lee Jones
---
arch/arm/boot/dts/stih41x-b2020.dtsi | 39
1 file changed, 39 insertions(+)
diff --git
Provide register and bit definitions used by STM's BCH Controller driver.
We place these into a shared location as they will be used by other STM
NAND Controllers as they appear.
Signed-off-by: Lee Jones
---
drivers/mtd/nand/stm_nand_regs.h | 302 +++
1 file
Obtain IRQ number and request IRQ resource via the usual methods. We're
also registering an IRQ handler to inform us of any completed tasks.
Notice that we're starting to make use of the device struct that we
defined before. In keeping with the subject of the patch, we're also
adding the related
The EMISS is a device which, amongst other things, controls various
flash memory modes. We make use of it here merely to flip the
HAMMING_NOT_BCH bit dependant on whether we wish to operate in Hamming
or BCH modes. The STM BCH driver makes good use of the helper introduced
here.
Signed-off-by:
Add support for clocks when, and only when, they are supplied. It is
not yet compulsory to provide the BCH and EMI clocks, as Common Clk isn't
supported Mainline yet. Until an implementation lands upstream all clocks
located on STM boards default to always-on.
Signed-off-by: Lee Jones
---
Provide some more in-depth structures which will be used heavily within
the driver. We also add a convenience function, used to set the default
values.
Signed-off-by: Lee Jones
---
drivers/mtd/nand/stm_nand_bch.c | 67 +
1 file changed, 67 insertions(+)
Populate the standard PM call-backs for; suspend, resume and restore.
Signed-off-by: Lee Jones
---
drivers/mtd/nand/stm_nand_bch.c | 39 +++
1 file changed, 39 insertions(+)
diff --git a/drivers/mtd/nand/stm_nand_bch.c b/drivers/mtd/nand/stm_nand_bch.c
index
This call resets the ECC stats, boot-mode controller and the AFM
(Advanced Flex Mode) controller, then enables Advanced Flex Mode,
configures the DMA plugs and finally enables the NANDi controller
IRQs. All required initialisation for clean running of the driver.
Signed-off-by: Lee Jones
---
Fetch platform specific data from Device Tree. Any functions which
are useful to other STM NAND Controllers have been separated into a
separate file so they can be easily referenced by them as they
appear.
Signed-off-by: Lee Jones
---
drivers/mtd/nand/Kconfig| 7 ++
Here we automatically select the strongest ECC scheme compatible with
the size of the OOB.
Signed-off-by: Lee Jones
---
drivers/mtd/nand/stm_nand_bch.c | 43 +
1 file changed, 43 insertions(+)
diff --git a/drivers/mtd/nand/stm_nand_bch.c
A few of the working variables can either be taken or derived from
existing knowledge about the connected chip. Rather than attempt to
provide each specification manually, here we make assumptions based
on information already obtained.
Signed-off-by: Lee Jones
---
Signed-off-by: Lee Jones
---
drivers/mtd/nand/stm_nand_bch.c | 9 +
1 file changed, 9 insertions(+)
diff --git a/drivers/mtd/nand/stm_nand_bch.c b/drivers/mtd/nand/stm_nand_bch.c
index fdb06517..a1c25d0 100644
--- a/drivers/mtd/nand/stm_nand_bch.c
+++ b/drivers/mtd/nand/stm_nand_bch.c
On Thu, May 01, 2014 at 11:33:56AM +0900, Jungseok Lee wrote:
> This patch adds virtual address space size and a level of translation
> tables to kernel configuration. It facilicates introduction of
> different MMU options, such as 4KB + 4 levels, 16KB + 4 levels and
> 64KB + 3 levels, easily.
>
On Thu, May 01, 2014 at 11:34:05AM +0900, Jungseok Lee wrote:
> This patch adds memory layout and translation lookup information
> about 48-bit address space with 4K pages. The description is based
> on 4 levels of translation tables.
>
> Cc: Catalin Marinas
> Cc: Steve Capper
> Signed-off-by:
Tune BCH programs according to device found and ECC mode.
Signed-off-by: Lee Jones
---
drivers/mtd/nand/stm_nand_bch.c | 121
1 file changed, 121 insertions(+)
diff --git a/drivers/mtd/nand/stm_nand_bch.c b/drivers/mtd/nand/stm_nand_bch.c
index
Including one for programming the BCH sequence, taking care not to
program two registers marked 'reserved' in the NANDi specification.
Failing to take this precaution results in an imprecise data exception
on ARM platforms. We also add an operation which waits on a completion
timeout (released by
Adding a function which simply erases a single flash block.
Signed-off-by: Lee Jones
---
drivers/mtd/nand/stm_nand_bch.c | 27 +++
1 file changed, 27 insertions(+)
diff --git a/drivers/mtd/nand/stm_nand_bch.c b/drivers/mtd/nand/stm_nand_bch.c
index 40767b8..6d7ab79
Initialise working buffers, accomodating DMA alignment constraints.
Signed-off-by: Lee Jones
---
drivers/mtd/nand/stm_nand_bch.c | 45 +
1 file changed, 45 insertions(+)
diff --git a/drivers/mtd/nand/stm_nand_bch.c b/drivers/mtd/nand/stm_nand_bch.c
index
Fixing some checkpatch warnings(remove global initialization, move __initdata,
coalesce formats ...)
Cc: Ingo Molnar
Cc: Andrew Morton
Signed-off-by: Fabian Frederick
---
init/main.c | 25 -
1 file changed, 12 insertions(+), 13 deletions(-)
diff --git a/init/main.c
Detect an erased page, tolerating and correcting up to a specified number
of bits at '0'. Downgrade uncorrectable ECC error for an erased page,
tolerating 'sectors_per_page' bits at '0'.
Signed-off-by: Lee Jones
---
drivers/mtd/nand/stm_nand_bch.c | 27 +++
1 file
The STM NAND BCH Controller driver has extended the IBBT header with
some extra convenience properties. This function populates both those
and the standard header specified by 'struct nand_ibbt_header'.
Signed-off-by: Lee Jones
---
drivers/mtd/nand/stm_nand_bch.c | 17 +
1 file
Use DMA to write a single page of data.
Signed-off-by: Lee Jones
---
drivers/mtd/nand/stm_nand_bch.c | 76 +
1 file changed, 76 insertions(+)
diff --git a/drivers/mtd/nand/stm_nand_bch.c b/drivers/mtd/nand/stm_nand_bch.c
index 72c9734..e875446 100644
---
Use DMA to read a single page of data.
Signed-off-by: Lee Jones
---
drivers/mtd/nand/stm_nand_bch.c | 96 +
1 file changed, 96 insertions(+)
diff --git a/drivers/mtd/nand/stm_nand_bch.c b/drivers/mtd/nand/stm_nand_bch.c
index 2ff72c9..72c9734 100644
Bad Block Markers (BBMs) keep track of unusable/unsuitable memory
blocks which can prevent unnecessary data loss. These helpers aid in
reading and writing to/from the Bad Block Tables (BBTs) where the BBMs
are stored.
Signed-off-by: Lee Jones
---
drivers/mtd/nand/stm_nand_bch.c | 29
Scan block for IBBT signature, either "Bbt0" or "1tbB" from
pre-specified data block.
Signed-off-by: Lee Jones
---
drivers/mtd/nand/stm_nand_bch.c | 91 +
1 file changed, 91 insertions(+)
diff --git a/drivers/mtd/nand/stm_nand_bch.c
Search for suitable block, erase and write table data to flash. If any
errors occur on selected block, mark it as bad and select another one.
Keep trying until both the Primary and Mirror BBTs are a) synced and
b) written successfully to flash.
Signed-off-by: Lee Jones
---
On Tue, Apr 29, 2014 at 05:24:32PM +0200, Jiri Kosina wrote:
> On Tue, 29 Apr 2014, Steven Rostedt wrote:
>
> > > According to 38.4 of [1], when SMM mode is entered while the CPU is
> > > handling NMI, the end result might be that upon exit from SMM, NMIs will
> > > be re-enabled and latched
Including; a check for Ready/NotBusy (RBn) (i.e. wait for current
operation to finish and a method to issue Flex commands to the
controller.
Signed-off-by: Lee Jones
---
drivers/mtd/nand/stm_nand_bch.c | 36
1 file changed, 36 insertions(+)
diff --git
Write BBT contents to the first page of a specified block, then
update the IBBT header and write it to last page of the same block.
Signed-off-by: Lee Jones
---
drivers/mtd/nand/stm_nand_bch.c | 29 +
1 file changed, 29 insertions(+)
diff --git
If at start-up time we can't find our own pre-built memory-resident BBT
(e.g. at first run) this routine will search for factory set BBMs and
duplicate them into our own BBT for local consumption.
Signed-off-by: Lee Jones
---
drivers/mtd/nand/stm_nand_bch.c | 27 +++
1
If a BBT already exists in flash, it's this function's task to locate
it and load it into the driver for local consumption.
Signed-off-by: Lee Jones
---
drivers/mtd/nand/stm_nand_bch.c | 87 +
1 file changed, 87 insertions(+)
diff --git
If load of existing BBT fails (first run), scan device for factory-set
BBMs (Bad-Block Markers) and create a new flash-resident.
Signed-off-by: Lee Jones
---
drivers/mtd/nand/stm_nand_bch.c | 26 ++
1 file changed, 26 insertions(+)
diff --git
Once the correct READ/WRITE(SEQIN) commands and address locations have
been sent to the Controller, these calls are able to read/write the
requested information from the FLEX_DATA register.
Signed-off-by: Lee Jones
---
drivers/mtd/nand/stm_nand_bch.c | 35 +++
1
read or write in 4 Byte chunks at a time until there is less than
8 Bytes remaining, at which point split down into reading the
reset a single Byte at a time.
Signed-off-by: Lee Jones
---
drivers/mtd/nand/stm_nand_bch.c | 72 +
1 file changed, 72
Inform the user of any known bad blocks during initialisation.
Conversely, if there aren't any known bad blocks, let the user know
the good news.
Signed-off-by: Lee Jones
---
drivers/mtd/nand/stm_nand_bch.c | 24
1 file changed, 24 insertions(+)
diff --git
Use the local BBM helpers to check for and mark bad blocks.
Signed-off-by: Lee Jones
---
drivers/mtd/nand/stm_nand_bch.c | 49 +
1 file changed, 49 insertions(+)
diff --git a/drivers/mtd/nand/stm_nand_bch.c b/drivers/mtd/nand/stm_nand_bch.c
index
Carry out the normal checks, then attempt to erase a given block.
Signed-off-by: Lee Jones
---
drivers/mtd/nand/stm_nand_bch.c | 11 +++
1 file changed, 11 insertions(+)
diff --git a/drivers/mtd/nand/stm_nand_bch.c b/drivers/mtd/nand/stm_nand_bch.c
index e742655..11735ee 100644
---
Including a dummy function for selecting chips which isn't supported by
our driver, the wait function as requested by the NAND sub system, a read
Byte function also requested by NAND and finally a command function which
provides a partial implementation of MTD/NAND Interface, based on
Hamming-FLEX
If none is provided by the platform, the default ECC mode will be used.
Signed-off-by: Lee Jones
---
drivers/mtd/nand/stm_nand_bch.c | 21 +
1 file changed, 21 insertions(+)
diff --git a/drivers/mtd/nand/stm_nand_bch.c b/drivers/mtd/nand/stm_nand_bch.c
index
Issue the core parse partitions and register as MTD device call.
Signed-off-by: Lee Jones
---
drivers/mtd/nand/stm_nand_bch.c | 5 -
1 file changed, 4 insertions(+), 1 deletion(-)
diff --git a/drivers/mtd/nand/stm_nand_bch.c b/drivers/mtd/nand/stm_nand_bch.c
index 6aaa118..0f96f75 100644
The OOB is filled with ECC data which is handled by the hardware
controller. It's a mistake for the framework to attempt to read or
write to the OOB area. If that happens, we'll BUG() out.
Signed-off-by: Lee Jones
---
drivers/mtd/nand/stm_nand_bch.c | 17 +
1 file changed, 17
Signed-off-by: Lee Jones
---
drivers/mtd/nand/stm_nand_bch.c | 4
1 file changed, 4 insertions(+)
diff --git a/drivers/mtd/nand/stm_nand_bch.c b/drivers/mtd/nand/stm_nand_bch.c
index bec7e23..e7a4bed 100644
--- a/drivers/mtd/nand/stm_nand_bch.c
+++ b/drivers/mtd/nand/stm_nand_bch.c
@@
Display a BUG() message and return appropriate/expected value.
Signed-off-by: Lee Jones
---
drivers/mtd/nand/stm_nand_bch.c | 39 +++
1 file changed, 39 insertions(+)
diff --git a/drivers/mtd/nand/stm_nand_bch.c b/drivers/mtd/nand/stm_nand_bch.c
index
Search the specified (first and, second or last) page in a block for
the BBT. Support Hamming or Advanced Flex Mode (AFM) BBMs too.
Signed-off-by: Lee Jones
---
drivers/mtd/nand/stm_nand_bch.c | 62 +
1 file changed, 62 insertions(+)
diff --git
Use the core nand_scan_ident() routine to locate connected NAND chips.
Signed-off-by: Lee Jones
---
drivers/mtd/nand/stm_nand_bch.c | 7 +++
1 file changed, 7 insertions(+)
diff --git a/drivers/mtd/nand/stm_nand_bch.c b/drivers/mtd/nand/stm_nand_bch.c
index 8a63c4a..2670cec 100644
---
This patch adds a new structure, 'nand_timing_spec', to capture the A/C
timing characteristics of NAND devices.
Unfortunately, there is no universally accepted standard for defining
NAND timing parameters. Different datasheets list different sets of
parameters. The ONFI standard gets close, but
This patch adds support for configuring the NAND Controller timing
registers according to 'nand_timing_spec' data. Since the stm_nand_bch
driver falls back to Hamming FLEX mode for certain operations, it is
necessary to configure the timing registers of both the Hamming
Controller and the BCH
Some chip characteristics have known incompatibilities with the function
of this driver. Here we check for this characteristics and refuse to run
if they are present.
Signed-off-by: Lee Jones
---
drivers/mtd/nand/stm_nand_bch.c | 39 ++-
1 file changed, 38
Signed-off-by: Lee Jones
---
drivers/mtd/nand/stm_nand_bch.c | 36
1 file changed, 36 insertions(+)
diff --git a/drivers/mtd/nand/stm_nand_bch.c b/drivers/mtd/nand/stm_nand_bch.c
index 19d090c..0c6aad2 100644
--- a/drivers/mtd/nand/stm_nand_bch.c
+++
--
Hello ,
My name is Lisa from North Africa . I am a woman seeking for true
friendship and partner, please if you don't mind I will like you
to respond to my mail so I can send you my pictures and tell you more
about myself,
Lisa.
--
To unsubscribe from this list: send the line "unsubscribe
Two points about latest version (dentry_kill-2):
- Doing anything with dentry->d_parent in case of DCACHE_DENTRY_KILLED looks
seriously wrong. Parent has been dealt with, at that point, by the other
caller, no?
- "bool foo = flag & FLAG" looks suspicious. Is this guaranteed not to
On Thu, 1 May 2014 10:04:30 +0200 Peter Zijlstra wrote:
> On Thu, May 01, 2014 at 12:41:43PM +1000, NeilBrown wrote:
> > diff --git a/include/linux/wait.h b/include/linux/wait.h
> > index 438dc6044587..162cbcde9dae 100644
> > --- a/include/linux/wait.h
> > +++ b/include/linux/wait.h
> > @@ -25,6
From: Alexey Charkov [mailto:alch...@gmail.com]
...
> This patch removes the cast altogether, and instead stores an actual
> pointer to u8 in match->data. All instances of 'revision' are also
> unified to u8 instead of an assortment of different integer types,
> in line with the definition of
On Thu, May 01, 2014 at 07:26:05PM +1000, NeilBrown wrote:
> On Thu, 1 May 2014 09:42:57 +0200 Peter Zijlstra wrote:
>
> > On Thu, May 01, 2014 at 12:37:38PM +1000, NeilBrown wrote:
> > > +static inline int
> > > +wait_on_bit(void *word, int bit, unsigned mode)
> > > +{
> > > + if
Higher priority does not provide exclusive privilege
of one fair task over the other. In this case priority
boosting looks excess.
On RT patch with enabled PREEMPT_RT_FULL I see a lot of
rt_mutex_setprio() actions like
120 -> 118
118 -> 120
They harm RT tasks.
RT patch has lazy
On Thu, 2014-03-06 at 14:52 +1100, Alistair Popple wrote:
> This patch adds support for the IBM Akebono board.
This patch just landed in linux-next (ie, in next-20140501).
> Signed-off-by: Alistair Popple
>[...]
> --- a/arch/powerpc/platforms/44x/Kconfig
> +++ b/arch/powerp
On Thu, 1 May 2014 09:42:57 +0200 Peter Zijlstra wrote:
> On Thu, May 01, 2014 at 12:37:38PM +1000, NeilBrown wrote:
> > +static inline int
> > +wait_on_bit(void *word, int bit, unsigned mode)
> > +{
> > + if (!test_bit(bit, word))
> > + return 0;
> > + return
On Wed, Apr 30, 2014 at 03:03:53PM -0500, Felipe Balbi wrote:
> On Thu, May 01, 2014 at 03:16:04AM -0400, Zhuang Jin Can wrote:
> > endpoint.maxburst may be 0 if a gadget doesn't call config_ep_by_speed()
> > to update it from the companion descriptor.
> > And endpoint.maxburst - 1 returns 1b
On Thu, May 01, 2014 at 09:04:19AM +0200, Andrzej Hajda wrote:
> 2. You replace calls of component_add and component_del with calls
> to interface_tracker_ifup(dev, INTERFACE_TRACKER_TYPE_COMPONENT,
> _component_ops),
> or interface_tracker_ifdown.
> Thats all for components.
How does the
On Thu, May 01, 2014 at 10:38:11AM +0200, Christian Engelmayer wrote:
> Fix a potential leak in the error path of rtw_set_key(). In case the requested
> algorithm is not supported by the driver, the function returns without
> enqueuing or freeing the already allocated command and parameter
On 2014/05/01 09:23 AM, Raghavendra Ganiga wrote:
> This is a patch to fix coding style
> warnings found by checkpatch.pl tool
>
> Signed-off-by: Raghavendra Chandra Ganiga
> ---
> drivers/staging/comedi/comedi_fops.c |2 ++
> 1 file changed, 2 insertions(+)
>
> diff --git
Hi Peter,
On Wed, Apr 30, 2014 at 10:22:19PM +0100, H. Peter Anvin wrote:
> On 04/23/2014 09:52 AM, Will Deacon wrote:
> > diff --git a/include/asm-generic/word-at-a-time.h
> > b/include/asm-generic/word-at-a-time.h
> > index d3909effd725..d96deb443f18 100644
> > ---
On 2014/04/30 06:34 PM, Raghavendra Ganiga wrote:
> This is a patch to fix coding style
> warnings found by the checkpatch.pl tool
>
> Signed-off-by: Raghavendra Chandra Ganiga
> ---
> drivers/staging/comedi/comedi_fops.c |6 --
> 1 file changed, 4 insertions(+), 2 deletions(-)
>
>
[ CC linux-sh and Magnus Damm (shmobile co-maintainer) ]
On Wed, Apr 30, 2014 at 11:26:37PM -0700, Brian Norris wrote:
> These defconfigs contain the CONFIG_M25P80 symbol, which is now
> dependent on the MTD_SPI_NOR symbol. Add CONFIG_MTD_SPI_NOR to satisfy
> the new dependency.
>
> At the same
If a zone cannot be used for a dirty page then it gets marked "full"
which is cached in the zlc and later potentially skipped by allocation
requests that have nothing to do with dirty zones.
Signed-off-by: Mel Gorman
Acked-by: Johannes Weiner
---
mm/page_alloc.c | 2 +-
1 file changed, 1
If cpusets are not in use then we still check a global variable on every
page allocation. Use jump labels to avoid the overhead.
Signed-off-by: Mel Gorman
---
include/linux/cpuset.h | 31 +++
kernel/cpuset.c| 8 ++--
mm/page_alloc.c| 3 ++-
3
There is no need to calculate zone_idx(preferred_zone) multiple times
or use the pgdat to figure it out.
Signed-off-by: Mel Gorman
---
mm/page_alloc.c | 55 ---
1 file changed, 32 insertions(+), 23 deletions(-)
diff --git a/mm/page_alloc.c
I was investigating a performance bug that looked like dd to tmpfs
had regressed. The bulk of the problem turned out to be a difference
in Kconfig but it got me looking at the unnecessary overhead in tmpfs,
mark_page_accessed and parts of the allocator. This series is the result.
The patches
A node/zone index is used to check if pages are compatible for merging
but this happens unconditionally even if the buddy page is not free. Defer
the calculation as long as possible. Ideally we would check the zone boundary
but nodes can overlap.
Signed-off-by: Mel Gorman
Acked-by: Johannes
Currently it's calculated once per zone in the zonelist.
Signed-off-by: Mel Gorman
Acked-by: Johannes Weiner
---
mm/page_alloc.c | 5 +++--
1 file changed, 3 insertions(+), 2 deletions(-)
diff --git a/mm/page_alloc.c b/mm/page_alloc.c
index 8971953..2e576fd 100644
--- a/mm/page_alloc.c
+++
The zlc is used on NUMA machines to quickly skip over zones that are full.
However it is always updated, even for the first zone scanned when the
zlc might not even be active. As it's a write to a bitmap that potentially
bounces cache line it's deceptively expensive and most machines will not
The test_bit operations in get/set pageblock flags are expensive. This patch
reads the bitmap on a word basis and use shifts and masks to isolate the bits
of interest. Similarly masks are used to set a local copy of the bitmap and then
use cmpxchg to update the bitmap if there have been no other
In the free path we calculate page_to_pfn multiple times. Reduce that.
Signed-off-by: Mel Gorman
---
include/linux/mmzone.h | 9 +++--
include/linux/pageblock-flags.h | 35 +++
mm/page_alloc.c | 32 ++--
3
shmem_getpage_gfp uses an atomic operation to set the SwapBacked field
before it's even added to the LRU or visible. This is unnecessary as what
could it possible race against? Use an unlocked variant.
Signed-off-by: Mel Gorman
Acked-by: Johannes Weiner
---
include/linux/page-flags.h | 1 +
ALLOC_NO_WATERMARK is set in a few cases. Always by kswapd, always for
__GFP_MEMALLOC, sometimes for swap-over-nfs, tasks etc. Each of these cases
are relatively rare events but the ALLOC_NO_WATERMARK check is an unlikely
branch in the fast path. This patch moves the check out of the fast path
get_pageblock_migratetype() is called during free with IRQs disabled. This
is unnecessary and disables IRQs for longer than necessary.
Signed-off-by: Mel Gorman
---
mm/page_alloc.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/mm/page_alloc.c b/mm/page_alloc.c
index
There should be no references to it any more and a parallel mark should
not be reordered against us. Use non-locked varient to clear page active.
Signed-off-by: Mel Gorman
---
mm/swap.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/mm/swap.c b/mm/swap.c
index
cold is a bool, make it one. Make the likely case the "if" part of the
block instead of the else as according to the optimisation manual this
is preferred.
Signed-off-by: Mel Gorman
---
arch/tile/mm/homecache.c | 2 +-
fs/fuse/dev.c| 2 +-
include/linux/gfp.h | 4 ++--
X86 prefers the use of unsigned types for iterators and there is a
tendency to mix whether a signed or unsigned type if used for page
order. This converts a number of sites in mm/page_alloc.c to use
unsigned int for order where possible.
Signed-off-by: Mel Gorman
---
include/linux/mmzone.h | 8
From: Nick Piggin
This patch introduces a new page flag for 64-bit capable machines,
PG_waiters, to signal there are processes waiting on PG_lock and uses it to
avoid memory barriers and waitqueue hash lookup in the unlock_page fastpath.
This adds a few branches to the fast path but avoids
When adding pages to the LRU we clear the active bit unconditionally. As the
page could be reachable from other paths we cannot use unlocked operations
without risk of corruption such as a parallel mark_page_accessed. This
patch test if is necessary to clear the atomic flag before using an atomic
The most obvious example of the problem being tackled is that
aops->write_begin may allocate a new page and make it visible only to
have mark_page_accessed called almost immediately after. Once the page is
visible the atomic operations are necessary which is noticable overhead when
writing to an
Hi Balbi,
On Wed, Apr 30, 2014 at 02:58:29PM -0500, Felipe Balbi wrote:
> On Thu, May 01, 2014 at 02:36:08AM -0400, Zhuang Jin Can wrote:
> > At least we should giveback the current request to the
> > gadget. Otherwise, the gadget will be stuck without knowing
> > anything.
> >
> > It was
Hi Greg,
I noticed that v3.14 broke my development system which relies on nfsroot
for embedded targets, but soon found Bruce's fix (which has now been in
Linus' tree for a while):
082f31a2169b ("nfsd: revert v2 half of "nfsd: don't return high
mode bits")
The fix
Fix a potential leak in the error path of rtw_set_key(). In case the requested
algorithm is not supported by the driver, the function returns without
enqueuing or freeing the already allocated command and parameter structs. Use
a centralized exit path and make sure that all memory is freed
Hi,
I occasionally build the odd program against raw perf and use perf.h for
this.
Now I find that no longer works because of:
52502bf201a85 ("perf tests: Add framework for automated perf_event_attr tests")
So revert the hunk touching on perf.h.
Also, it seems to be for snake shit only, and
This is a patch to fix coding style
warnings found by checkpatch.pl tool
Signed-off-by: Raghavendra Chandra Ganiga
---
drivers/staging/comedi/comedi_fops.c |2 ++
1 file changed, 2 insertions(+)
diff --git a/drivers/staging/comedi/comedi_fops.c
b/drivers/staging/comedi/comedi_fops.c
index
If kthread_run on spi_init_queue() fails, spi_destroy_queue can lead to
hang timeout.
Before 786235eeba0e1e85e5cbbb9f97d1087ad03dfa21 kthread:
make kthread_create() killable" it was very unlikely that kthread_run
would fail, but after that patch a kill to modprobe can lead to this
kind of error.
On Thu, May 01, 2014 at 12:41:43PM +1000, NeilBrown wrote:
> diff --git a/include/linux/wait.h b/include/linux/wait.h
> index 438dc6044587..162cbcde9dae 100644
> --- a/include/linux/wait.h
> +++ b/include/linux/wait.h
> @@ -25,6 +25,7 @@ struct wait_bit_key {
> void
On Wed, 30 Apr 2014 23:54:37 +0200, Geert Uytterhoeven
wrote:
> Hi Grant,
>
> On Tue, Apr 29, 2014 at 3:16 PM, Grant Likely
> wrote:
> > On Fri, 25 Apr 2014 16:44:58 -0700, Kevin Hilman wrote:
> >> Geert Uytterhoeven writes:
> >>
> >> > When adding a device from DT, check if its clocks are
On Wed, Apr 30, 2014 at 6:07 PM, Greg Kroah-Hartman
wrote:
> On Wed, Apr 30, 2014 at 02:33:21PM -0600, Bjorn Helgaas wrote:
>> dma_declare_coherent_memory() takes two addresses for a region of memory: a
>> "bus_addr" and a "device_addr". I think the intent is that "bus_addr" is
>> the physical
Ah. Fantastic. That's all great stuff. I'm on holiday today but I'll
send a new later in the week.
regards,
dan carpenter
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majord...@vger.kernel.org
More majordomo info at
On Thu, May 01, 2014 at 12:37:38PM +1000, NeilBrown wrote:
> +static inline int
> +wait_on_bit(void *word, int bit, unsigned mode)
> +{
> + if (!test_bit(bit, word))
> + return 0;
> + return out_of_line_wait_on_bit(word, bit,
> +bit_wait,
> +
Hi all,
This tree still fails (more than usual) the powerpc allyesconfig build.
Changes since 20140430:
The powerpc tree still had its build failure.
The pm tree gained a conflict against the mips tree.
The mfd-lj tree lost its build failure.
The staging tree gained a conflict against the
2014-05-01 6:19 GMT+09:00, Andrew Morton :
> On Tue, 29 Apr 2014 11:53:10 +0900 Daeseok Youn
> wrote:
>
>> "dev" cannot be NULL because it is already checked before
>> calling dma_pool_create().
>>
>> Signed-off-by: Daeseok Youn
>> ---
>> If dev can be NULL, it has NULL deferencing when
On Wed, Apr 30, 2014 at 10:30:21PM +0200, Arnd Bergmann wrote:
> On Wednesday 30 April 2014 22:21:30 Andrew Lunn wrote:
> > > Yes, we *can* move it in kirkwood.dtsi, but IMHO we *should* keep it in
> > > the board files. I see the "chosen" node and its properties as _the_
> > > board node.
> > >
On 05/01/2014 04:11 PM, Ben Skeggs wrote:
On Fri, Apr 25, 2014 at 5:19 PM, Alexandre Courbot wrote:
Changes since v2:
- Enabled software class
- Removed unneeded changes to nouveau_accel_init()
- Replaced use of architecture-private pfn_to_dma() and dma_to_pfn() with
the portable
entry(cmd->ll_list) may belong to new request once end_cmd()
returns, so fix the bug with the patch.
Without the change, it is easy to observe oops when
doing null_blk(timer) test.
Signed-off-by: Ming Lei
---
drivers/block/null_blk.c |2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
401 - 500 of 1092 matches
Mail list logo
