On Thu, Apr 23, 2015 at 02:32:26PM -0600, Stephen Warren wrote:
> On 04/23/2015 08:06 AM, Johan Hovold wrote:
> > Make sure only to copy any actual data rather than the whole buffer,
> > when releasing the temporary buffer used for unaligned non-isochronous
> > transfers.
> >
> > Compile-tested onl
On 23 April 2015 at 21:59, Greg KH wrote:
> On Thu, Apr 23, 2015 at 09:50:14PM +0100, Chris Bainbridge wrote:
>> Hardware: IvyBridge Macbook (Panther Point PCH), 2 external USB3 ports
>> Verified in kernels: 4.0.0+ (1fc1499), Debian 3.16.4
>>
>> All USB3 devices appear as USB2 high-speed (480mbps)
Hi Steve,
> If you are looking for a good C websockets library, gpl etc try:
> https://libwebsockets.org/trac/libwebsockets
> I used it in a project and was very pleased with how it worked and
> the support where it is hosted.
I checked libwebsockets ago but I misunderstood it doesn't support
c
Fixed several style issues with: comments, function perenthesis,
indentation, and conditional braces
Signed-off-by: Jason Eastman
---
drivers/usb/image/mdc800.c | 769 +++-
1 file changed, 336 insertions(+), 433 deletions(-)
diff --git a/drivers/usb/imag
And this one too :(
Please, consider only the "X/2" patches.
BR, David
On Thu, Apr 23, 2015 at 03:59:26PM -0700, David Cohen wrote:
> From: Dan Carpenter
>
> commit 73a30bfc0d526db899033165db6f95c427e70505 upstream.
>
> These were cut and paste from the ->disconnect function.
>
> Fixes commit
Hi,
Please ignore this patch. It was sent by mistake :(
My apologies.
Br, David
On Thu, Apr 23, 2015 at 04:00:52PM -0700, David Cohen wrote:
> From: Dan Carpenter
>
> commit 73a30bfc0d526db899033165db6f95c427e70505 upstream.
>
> These were cut and paste from the ->disconnect function.
>
> Fi
From: Dan Carpenter
commit 73a30bfc0d526db899033165db6f95c427e70505 upstream.
These were cut and paste from the ->disconnect function.
Fixes commit 30d577b9bcc4 ('usb: dwc3: gadget: call gadget driver's
->suspend/->resume')
Cc: # 3.14
Signed-off-by: Dan Carpenter
Signed-off-by: Felipe Balbi
From: Dan Carpenter
commit 73a30bfc0d526db899033165db6f95c427e70505 upstream.
These were cut and paste from the ->disconnect function.
Fixes commit 30d577b9bcc4 ('usb: dwc3: gadget: call gadget driver's
->suspend/->resume')
Cc: # 3.14
Signed-off-by: Dan Carpenter
Signed-off-by: Felipe Balbi
From: Felipe Balbi
commit bc5ba2e0b829c9397f96df1191c7d2319ebc36d9 upstream.
When going into bus suspend/resume we _must_
call gadget driver's ->suspend/->resume callbacks
accordingly. This patch implements that very feature
which has been missing forever.
Cc: # 3.14
Signed-off-by: Felipe Balb
From: Dan Carpenter
commit 73a30bfc0d526db899033165db6f95c427e70505 upstream.
These were cut and paste from the ->disconnect function.
Fixes commit 30d577b9bcc4 ('usb: dwc3: gadget: call gadget driver's
->suspend/->resume')
Cc: # 3.14
Signed-off-by: Dan Carpenter
Signed-off-by: Felipe Balbi
Hi,
The following patches were merged on kernel >= v3.15 but the issue they fix
exists on v3.14 too.
We are currently seeing BC1.2 compliance test failure on some Intel devices
supported by v3.14 LTS kernel. The issue is fixed by this series.
So, I propose to add these to stable.
Br, David
---
D
Hi Greg,
On Fri, Apr 17, 2015 at 09:42:57PM +0200, Greg KH wrote:
> On Fri, Apr 17, 2015 at 11:41:56AM -0700, David Cohen wrote:
> > From: Felipe Balbi
> >
> > When going into bus suspend/resume we _must_
> > call gadget driver's ->suspend/->resume callbacks
> > accordingly. This patch implement
Hi Felipe,
On Fri, Apr 17, 2015 at 02:43:27PM -0500, Felipe Balbi wrote:
> On Fri, Apr 17, 2015 at 11:41:56AM -0700, David Cohen wrote:
> > From: Felipe Balbi
>
> missing the required:
>
> [ Upstream commit bc5ba2e0b829c9397f96df1191c7d2319ebc36d9 ]
>
> >
> > When going into bus suspend/resum
On Thu, Apr 23, 2015 at 09:50:14PM +0100, Chris Bainbridge wrote:
> Hardware: IvyBridge Macbook (Panther Point PCH), 2 external USB3 ports
> Verified in kernels: 4.0.0+ (1fc1499), Debian 3.16.4
>
> All USB3 devices appear as USB2 high-speed (480mbps) rather than USB3
> super-speed (5000mbps).
Are
Hardware: IvyBridge Macbook (Panther Point PCH), 2 external USB3 ports
Verified in kernels: 4.0.0+ (1fc1499), Debian 3.16.4
All USB3 devices appear as USB2 high-speed (480mbps) rather than USB3
super-speed (5000mbps).
usb1 and usb2 are the XHCI USB3 buses. usb3 and usb4 are the EHCI USB2 buses
(t
On 04/23/2015 08:06 AM, Johan Hovold wrote:
Make sure only to copy any actual data rather than the whole buffer,
when releasing the temporary buffer used for unaligned non-isochronous
transfers.
Compile-tested only.
Tested-by: Stephen Warren
(Tested a USB network device attached to Jetson TK
If you are looking for a good C websockets library, gpl etc try:
https://libwebsockets.org/trac/libwebsockets
I used it in a project and was very pleased with how it worked and the
support where it is hosted.
Regards, Steve
On Wed, Apr 22, 2015 at 11:52 PM, fx IWATA NOBUO
wrote:
> Hi Bjørn
>
I know. Honestly I never expect it to be fixed. I just want to draw
some attention so that people can be aware of this problem (some may
be suffering from this without understand what's going on). Though I
did filed a bug report on bugzilla some time ago to request making
this function a no-op (so
On Thu, 23 Apr 2015, Johan Hovold wrote:
> Make sure only to copy any actual data rather than the whole buffer,
> when releasing the temporary buffer used for unaligned non-isochronous
> transfers.
>
> Compile-tested only.
>
> Signed-off-by: Johan Hovold
> ---
> drivers/usb/host/ehci-tegra.c |
On Thu, Apr 23, 2015 at 04:31:51PM +0200, Frans Klaver wrote:
> On Thu, Apr 23, 2015 at 4:06 PM, Johan Hovold wrote:
> > static void free_dma_aligned_buffer(struct urb *urb)
> > {
> > struct dma_aligned_buffer *temp;
> > + size_t length;
> >
> > if (!(urb->transfer_flags &
On Thu, 23 Apr 2015, Tom Yan wrote:
> I'm not saying that the kernel shouldn't initialize the attributes or
> have a default. But it should only set the default when the attribute
> is initialized (It doesn't even matter to me whether it's enabled or
> disabled).
>
> It's just there should not be
On Thu, 23 Apr 2015, Tal Shorer wrote:
> Why is drivers/usb/Makefile even read when CONFIG_USB is not set?
CONFIG_USB means that the kernel will support host-side USB. But
drivers/usb and the directories underneath it also contain drivers for
peripheral-side USB. Therefore the Makefiles in th
On Mon, Apr 13, 2015 at 10:15 AM, Mathias Nyman
wrote:
> Hi
>
> On 08.04.2015 20:45, Rodrigo Severo wrote:
>>
>> At that time I even tested enabling XHCI_TRUST_TX_LENGTH quirk for the
>> ASM1042A USB 3.0 Host Controller which eliminated the warnings on my
>> logs but didn't solve the hang issues.
On Thu, Apr 23, 2015 at 4:06 PM, Johan Hovold wrote:
> Make sure only to copy any actual data rather than the whole buffer,
> when releasing the temporary buffer used for unaligned non-isochronous
> transfers.
>
> Compile-tested only.
>
> Signed-off-by: Johan Hovold
> ---
> drivers/usb/host/ehci
On Wed, 22 Apr 2015, Arun Ramamurthy wrote:
> This patch set adds a new API to get phy by index when multiple
> phys are present. This patch is based on discussion with Arnd Bergmann
> about dt bindings for multiple phys.
>
> History:
> v1:
> - Removed null pointers on Dmitry's suggestion
>
On Mon, Apr 13, 2015 at 10:15 AM, Mathias Nyman
wrote:
> Hi
>
> On 08.04.2015 20:45, Rodrigo Severo wrote:
>>
>> When connecting cx231xx video grab devices (which are USB 2.0
>> themselves) on ASMedia ASM1042A USB 3.0 Host Controllers the video
>> capture process hangs after a few minutes. Besides
Make sure to copy the whole transfer buffer when releasing the temporary
buffer used for unaligned isochronous transfers as the data is not
necessarily contiguous in that case.
Signed-off-by: Johan Hovold
---
drivers/staging/octeon-usb/octeon-hcd.c | 12 +---
1 file changed, 9 insertions
Make sure only to copy any actual data rather than the whole buffer,
when releasing the temporary buffer used for unaligned non-isochronous
transfers.
Signed-off-by: Johan Hovold
---
drivers/usb/musb/musb_host.c | 9 +++--
1 file changed, 7 insertions(+), 2 deletions(-)
diff --git a/drivers
Make sure only to copy any actual data rather than the whole buffer,
when releasing the temporary buffer used for unaligned non-isochronous
transfers.
Compile-tested only.
Signed-off-by: Johan Hovold
---
drivers/usb/host/ehci-tegra.c | 12 +---
1 file changed, 9 insertions(+), 3 deletio
These patches (for 4.1) make sure that only the received data is copied
from the temporary buffers used for unaligned transfers.
I discovered this when debugging an issue where the Beaglebone Black
would lock up on disconnect.
Turns out it was related to the transfer_buffers not being properly
al
On Thu, Apr 23, 2015 at 09:10:04AM -0400, Kynn Jones wrote:
> NB: This is an updated version of
> https://bugzilla.kernel.org/show_bug.cgi?id=97131. I'm reposting here by the
> request of that bug's assignee.
>
> In this update I've added the output of some more diagnostic commands (at the
> end
NB: This is an updated version of
https://bugzilla.kernel.org/show_bug.cgi?id=97131. I'm reposting here by the
request of that bug's assignee.
In this update I've added the output of some more diagnostic commands (at the
end of the post). The list of diagnostic commands run is now this:
$ c
Hi,
On 23-04-15 01:04, Arun Ramamurthy wrote:
This patch set adds a new API to get phy by index when multiple
phys are present. This patch is based on discussion with Arnd Bergmann
about dt bindings for multiple phys.
History:
v1:
- Removed null pointers on Dmitry's suggestion
- Impro
On Thu, Apr 23, 2015 at 10:07:35AM +0530, Niranjan H Y wrote:
> Hello Everyone,
>
> I'm newbie to driver development & have a question on CDC NCM gadget driver.
>
> When I load the module g_ncm.ko on the target board
> and execute "lsusb" on host, there is some delay in response.
> (about 5-10 se
> > +#include "../../../../include/uapi/linux/usbip_ux.h"
> So you can just refer this as
I will fix in v3.
Thank you,
n.iwata
//
> I see various parameters, such as --busid, followed by an "=" sign.
> This looks like a thing thoughout the manual.
> Although this syntax works as well, I think we should be consistent
> and not include "=".
> I, personally, prefer having "--busid "--busid=
Hi Valentina,
> * I noticed in some places you've used labels such as err0 and err1.
> Please replace them with something more relevant to their
> functionality,
> such as what you're cleaning up after. There are examples of this
> both in
> USB/IP code and throughout the kernel.
I will fix all
37 matches
Mail list logo
