Thanks so much. :)
regards,
dan carpenter
___
devel mailing list
de...@linuxdriverproject.org
http://driverdev.linuxdriverproject.org/mailman/listinfo/driverdev-devel
On Wed, Dec 4, 2013 at 2:02 PM, Greg KH wrote:
> On Wed, Dec 04, 2013 at 01:55:34PM -0800, Colin Cross wrote:
>> On Wed, Dec 4, 2013 at 1:43 PM, Greg KH wrote:
>> > On Wed, Dec 04, 2013 at 12:46:42PM -0800, Colin Cross wrote:
>> >> On Wed, Dec 4, 2013 at 10:35 AM, Greg KH
>> >> wrote:
>> >>
>>
On Wed, Dec 4, 2013 at 10:53 PM, Mark Einon wrote:
> On Wed, Dec 04, 2013 at 09:32:27PM +0800, ZHAO Gang wrote:
>>
>> By re-examine the code I found the calculation is not correct. The
>> real impact is too big to apply this patch, so this patch and
>> following patches should be dropped.
>
> This
This is the fourth patch of a series.(4)
Replaced spaces in margin w/ 1 tab for lines:
181-185, 187-188
Signed-off-by: Gary Alan Rookard
On branch staging-next
---
drivers/staging/bcm/DDRInit.c | 14 +++---
1 file changed, 7 insertions(+), 7 deletions(-)
diff --git a/drivers/staging/b
On Thu, Dec 5, 2013 at 7:23 AM, Greg Kroah-Hartman
wrote:
> On Wed, Dec 04, 2013 at 03:24:14PM +0800, ZHAO Gang wrote:
>> As TODO file suggested, drop packet instead of return NETDEV_TX_BUSY
>> when tx failed.
>
> Really? That's ok to do? Seems like you are changing the logic of the
> function a
This is the fifth and final patch for replace spaces w/tab.(5)
Replaced spaces in margin w/ 1 tab for lines:
193-197, 199-200, 202-205, 207, 209-242, 244, 246
Signed-off-by: Gary Alan Rookard
On branch staging-next
---
drivers/staging/bcm/DDRInit.c | 96 +---
This is the third patch of a series.(3)
Replaced spaces in margin w/ 1 tab for lines:
121-125, 128-134, 136, 138-171, 173, 175
Signed-off-by: Gary Alan Rookard
On branch staging-next
---
drivers/staging/bcm/DDRInit.c | 98 +--
1 file changed, 49 insertio
This is the third patch of a series.(3)
Replaced spaces in margin w/ 1 tab for lines:
121-125, 128-134, 136, 138-171, 173, 175
Signed-off-by: Gary Alan Rookard
On branch staging-next
---
drivers/staging/bcm/DDRInit.c | 98 +--
1 file changed, 49 insertio
This is the third patch of a series.(3)
Replaced spaces in margin w/ 1 tab for lines:
121-125, 128-134, 136, 138-171, 173, 175
Signed-off-by: Gary Alan Rookard
On branch staging-next
---
drivers/staging/bcm/DDRInit.c | 98 +--
1 file changed, 49 insertio
> None of this (the patch series or the original code) is mine. My
Sorry mistraced the attribution sequence
> question was more of a general one on designing ioctls, as well as
> concerns with changing the existing 32-bit api.
I think in general my advice would be:
If its already been screwed
This is the second patch of a series.(2)
Replaced spaces in margin w/ 1 tab for lines:
69-78, 80-114, 116
Signed-off-by: Gary Alan Rookard
On branch staging-next
---
drivers/staging/bcm/DDRInit.c | 92 +--
1 file changed, 46 insertions(+), 46 deletions(-
- Original Message -
> From: "Martyn Welch"
> Sent: Thursday, November 7, 2013 7:00:35 AM
>
> On 05/11/13 20:58, Aaron Sierra wrote:
> > - Original Message -
> >> From: "Martyn Welch"
> >> Subject: Re: [PATCH 1/3] vme_user: Ensure driver compiles after VME
> >> bridges
> >>
> >>
This is the first patch of a series.
Replaced spaces in margin w/ 1 tab for lines:
11-15, 17-23, 25-58, 60, 62, 64
On branch staging-next
---
drivers/staging/bcm/DDRInit.c | 98 +--
1 file changed, 49 insertions(+), 49 deletions(-)
diff --git a/drivers/s
On Wed, Dec 4, 2013 at 4:02 PM, Greg KH wrote:
> On Wed, Dec 04, 2013 at 02:22:13PM -0800, Colin Cross wrote:
>> On Wed, Dec 4, 2013 at 2:02 PM, Greg KH wrote:
>> > On Wed, Dec 04, 2013 at 01:55:34PM -0800, Colin Cross wrote:
>> >> On Wed, Dec 4, 2013 at 1:43 PM, Greg KH
>> >> wrote:
>> >> > On
On Wed, Dec 04, 2013 at 02:22:13PM -0800, Colin Cross wrote:
> On Wed, Dec 4, 2013 at 2:02 PM, Greg KH wrote:
> > On Wed, Dec 04, 2013 at 01:55:34PM -0800, Colin Cross wrote:
> >> On Wed, Dec 4, 2013 at 1:43 PM, Greg KH wrote:
> >> > On Wed, Dec 04, 2013 at 12:46:42PM -0800, Colin Cross wrote:
>
On Wed, 4 Dec 2013 10:35:54 -0800
Greg KH wrote:
> On Wed, Dec 04, 2013 at 06:09:41PM +, Serban Constantinescu wrote:
> > +#define size_helper(x) ({ \
> > + size_t __size; \
> > + if (!is_com
On Wed, Dec 04, 2013 at 06:09:34PM +, Serban Constantinescu wrote:
> This patch adds binder_copy_to_user() to be used for copying binder
> commands to user address space. This way we can abstract away the
> copy_to_user() calls and add separate handling for the compat layer.
>
> Signed-off-by:
On Wed, Dec 04, 2013 at 03:24:11PM +0800, ZHAO Gang wrote:
> 1. As TODO list suggested, do this sort of things to reduce split lines:
> struct fbr_lookup *fbr;
> fbr = rx_local->fbr[id];
> Then replace all the instances of "rx_local->fbr[id]" with fbr.
>
> 2. Some code style chan
On Wed, Dec 04, 2013 at 11:46:48AM +, Ian Abbott wrote:
> On 2013-12-03 19:07, H Hartley Sweeten wrote:
> > The comedi subsystem only requires the drivers to support interrupts if one
> > or more of the subdevices support async commands. Since this is optional:
> >
> > 1) don't fail the attach
On Wed, Dec 04, 2013 at 03:24:14PM +0800, ZHAO Gang wrote:
> As TODO file suggested, drop packet instead of return NETDEV_TX_BUSY
> when tx failed.
Really? That's ok to do? Seems like you are changing the logic of the
function a lot here, how does the code let userspace know packets were
dropped
On Wed, Dec 4, 2013 at 3:21 PM, One Thousand Gnomes
wrote:
> On Wed, 4 Dec 2013 10:35:54 -0800
> Greg KH wrote:
>
>> On Wed, Dec 04, 2013 at 06:09:41PM +, Serban Constantinescu wrote:
>> > +#define size_helper(x) ({ \
>> > + size_t __size;
On Wed, Dec 4, 2013 at 2:02 PM, Greg KH wrote:
> On Wed, Dec 04, 2013 at 01:55:34PM -0800, Colin Cross wrote:
>> On Wed, Dec 4, 2013 at 1:43 PM, Greg KH wrote:
>> > On Wed, Dec 04, 2013 at 12:46:42PM -0800, Colin Cross wrote:
>> >> On Wed, Dec 4, 2013 at 10:35 AM, Greg KH
>> >> wrote:
>> >>
>>
On Wed, Dec 04, 2013 at 01:55:34PM -0800, Colin Cross wrote:
> On Wed, Dec 4, 2013 at 1:43 PM, Greg KH wrote:
> > On Wed, Dec 04, 2013 at 12:46:42PM -0800, Colin Cross wrote:
> >> On Wed, Dec 4, 2013 at 10:35 AM, Greg KH
> >> wrote:
> >>
> >>
> >> > And finally, is this all really needed? Why
On Wed, Dec 4, 2013 at 1:43 PM, Greg KH wrote:
> On Wed, Dec 04, 2013 at 12:46:42PM -0800, Colin Cross wrote:
>> On Wed, Dec 4, 2013 at 10:35 AM, Greg KH wrote:
>>
>>
>> > And finally, is this all really needed? Why not just fix the structures
>> > to be "correct", and then fix userspace to use
On Wed, Dec 04, 2013 at 12:46:42PM -0800, Colin Cross wrote:
> On Wed, Dec 4, 2013 at 10:35 AM, Greg KH wrote:
>
>
> > And finally, is this all really needed? Why not just fix the structures
> > to be "correct", and then fix userspace to use the correct structures as
> > well, thereby not needi
On Wed, Dec 4, 2013 at 10:35 AM, Greg KH wrote:
> And finally, is this all really needed? Why not just fix the structures
> to be "correct", and then fix userspace to use the correct structures as
> well, thereby not needing a compat layer at all?
Some of the binder ioctls take userspace point
On Wed, Dec 04, 2013 at 06:09:41PM +, Serban Constantinescu wrote:
> +#define size_helper(x) ({\
> + size_t __size; \
> + if (!is_compat_task())
The "space before a non-naked semicolon" test has
unwanted output when used in "for ( ;; )" loops.
Make the test work only on end-of-line statement
termination semicolons.
Signed-off-by: Joe Perches
---
On Wed, 2013-12-04 at 11:21 +0300, Dan Carpenter wrote:
> You and I generally agree on style
This patch moves some of the logic for binder_thread_write() into
subfunctions. This way we can share more code with the binder compat
layer.
Signed-off-by: Serban Constantinescu
---
drivers/staging/android/binder.c | 403 +-
1 file changed, 220 insertions(+)
This patch adds all the needed compat structures to binder.h. All the
structures defined in this patch mirror the structure and size of 32bit
ones.
Signed-off-by: Serban Constantinescu
---
drivers/staging/android/binder.h | 109 ++
1 file changed, 109 inserti
Hi all,
The patches attached add support for 32bit userspace running on 64bit
kernels. This is the last series of patches needed for basic 32bit Android
bring-up on 64bit kernels.
The series is split into refactoring the binder driver and the addition
of the compat layer. Please note that the hel
This patch adds support for 32bit userspace running on 64bit kernels.
All the changes done in this patch have been tested on 32bit and 64bit
systems.
Signed-off-by: Serban Constantinescu
---
drivers/staging/android/binder.c | 355 +-
1 file changed, 353 inser
This patch modifies the functions that need to be passed the explicit
command to use a boolean flag. This way we can reuse the code for 64bit
compat commands.
Signed-off-by: Serban Constantinescu
---
drivers/staging/android/binder.c | 35 ++-
1 file changed, 18
This patch adds a deref_helper() macro that will be used to dereference
the binder offsets on 64bit systems where the offset is either a 32bit
or a 64bit value, depending on the userpace used (32bit /64bit)
This patch is a temporary patch that will be extended with 32bit compat
handling.
Signed-o
This patch adds binder_copy_to_user() to be used for copying binder
commands to user address space. This way we can abstract away the
copy_to_user() calls and add separate handling for the compat layer.
Signed-off-by: Serban Constantinescu
---
drivers/staging/android/binder.c | 39
This patch adds copy_flat_binder_object macro() that will help
dereference struct flat_binder_object on 64bit systems where the
structure differs between 32bit and 64bit userspace.
This patch is a temporary patch that will be extended with 32bit compat
handling.
Signed-off-by: Serban Constantines
This patch adds align_helper() macro that will be used for enforcing
the desired alignment on 64bit systems where the alignment will differ
depending on the userspace used (32bit /64bit).
This patch is a temporary patch that will be extended with 32bit compat
handling.
Signed-off-by: Serban Const
This patch adds size_helper() macro that will be used for indexing into
different buffers on 64bit systems where the size of particular
structures will differ depending on the userspace used (32bit /64bit).
This patch is a temporary patch that will be extended with 32bit compat
handling.
Signed-o
On Wed, Dec 04, 2013 at 09:32:27PM +0800, ZHAO Gang wrote:
>
> By re-examine the code I found the calculation is not correct. The
> real impact is too big to apply this patch, so this patch and
> following patches should be dropped.
This will mess up the patch numbering for your set (e.g. only 7
On 2013/12/04 01:05 PM, Dan Carpenter wrote:
> On Wed, Dec 04, 2013 at 11:10:05AM +, Ian Abbott wrote:
>> On 2013-12-03 19:07, H Hartley Sweeten wrote:
>>> Do the request_irq() before setting up the subdevices. This removes an
>>> indent level and makes the code a bit cleaner.
>>>
>>> Also, rem
On Wed, Dec 4, 2013 at 8:01 PM, ZHAO Gang wrote:
> On Wed, Dec 4, 2013 at 5:05 PM, Mark Einon wrote:
>> On Wed, Dec 04, 2013 at 03:24:18PM +0800, ZHAO Gang wrote:
>>> The original code allocate rx dma memory in several dma_alloc_coherent
>>> calls,
>>> which causes some problems:
>>> 1. since dm
On Wed, Dec 04, 2013 at 11:10:05AM +, Ian Abbott wrote:
> On 2013-12-03 19:07, H Hartley Sweeten wrote:
> >Do the request_irq() before setting up the subdevices. This removes an
> >indent level and makes the code a bit cleaner.
> >
> >Also, remove the dev_warn() noise about the irq.
> >
> >Sign
On Wed, Dec 04, 2013 at 04:53:09AM -0500, Gary Rookard wrote:
> Fixed non_standard kernel style.
>
You're doing too many things at once, it needs to be broken into
a patch series which do one type of change at a time.
The changelog is useless.
> Signed off by: Gary Alan Rookard
> ---
> driver
On Wed, 2013-12-04 at 10:38 +0300, Dan Carpenter wrote:
> On Wed, Dec 04, 2013 at 06:35:15AM +0200, Aldo Iljazi wrote:
> > Samuel Thibault wrote:
> >
> > > Err, I'd rather make it really visible that the for loop doesn't have
> > > its first statement?
> >
> > Wouldn't it be better if you add a
On Wed, Dec 4, 2013 at 5:05 PM, Mark Einon wrote:
> On Wed, Dec 04, 2013 at 03:24:18PM +0800, ZHAO Gang wrote:
>> The original code allocate rx dma memory in several dma_alloc_coherent calls,
>> which causes some problems:
>> 1. since dma_alloc_coherent allocate at least one page memory, it wastes
On 2013-12-03 19:07, H Hartley Sweeten wrote:
The comedi subsystem only requires the drivers to support interrupts if one
or more of the subdevices support async commands. Since this is optional:
1) don't fail the attach if the irq is not available
2) only hookup the async command support if the
On 2013-12-03 19:07, H Hartley Sweeten wrote:
The dev->irq passed to request_irq() will always be 0 when the auto_attach
function is called. The pcidev->irq should be used instead to get the correct
irq number.
It looks like this bug was introduced in the 3.7 kernel. I'll check it
for applica
On 2013-12-03 19:07, H Hartley Sweeten wrote:
Do the request_irq() before setting up the subdevices. Only hook up
the command support of the irq was sucessfully requested.
Signed-off-by: H Hartley Sweeten
Cc: Ian Abbott
Cc: Greg Kroah-Hartman
---
drivers/staging/comedi/drivers/ni_pcidio.c |
On 2013-12-03 19:07, H Hartley Sweeten wrote:
Clean up the irq request in the attach of this driver and remove the
dev_{level} noise.
Signed-off-by: H Hartley Sweeten
Cc: Ian Abbott
Cc: Greg Kroah-Hartman
---
drivers/staging/comedi/drivers/das1800.c | 64 ++--
1
On 2013-12-03 19:07, H Hartley Sweeten wrote:
Do the request_irq() before setting up the subdevices. This removes an
indent level and makes the code a bit cleaner.
Also, remove the dev_warn() noise about the irq.
Signed-off-by: H Hartley Sweeten
Cc: Ian Abbott
Cc: Greg Kroah-Hartman
---
dr
On 2013-12-03 19:07, H Hartley Sweeten wrote:
The irq is only needed to support async commands. Tidy up the
code that does the request_irq() and remove the noise.
Only hookup the subdevice command support if the irq is available.
Signed-off-by: H Hartley Sweeten
Cc: Ian Abbott
Cc: Greg Kroah-
On Wed, Dec 04, 2013 at 03:24:18PM +0800, ZHAO Gang wrote:
> The original code allocate rx dma memory in several dma_alloc_coherent calls,
> which causes some problems:
> 1. since dma_alloc_coherent allocate at least one page memory, it wastes some
>memory when allocation size is smaller than o
You and I generally agree on style preferences... I think the warning
should be limited to grep " ;$".
I did a grep on the kernel for ' ;' and found 8000 results. 6000 of
them are caught by my semicolon before the newline rule. The remaining
2000 are assembly, macros, and crappy for loops.
reg
53 matches
Mail list logo
