[PATCH 0/4] usb: renesas_usbhs: fixes this driver in gadget mode

[PATCH 0/4] usb: renesas_usbhs: fixes this driver in gadget mode

[Yoshihiro Shimoda]

If we tested this driver as gadget and g_ncm driver using iperf,
some issues below happened:
 - Oops happened.
 - A transaction didn't finish in PIO mode.
 - A transaction didn't finish in DMA mode if the "zero" flag is set.
 - The driver didn't cancel a transaction if dequeue() is called.

This patch series fixes them. If possible, they would like to be
merged to v3.17.
This patch series is based on Greg's usb.git / usb-linus branch.
(commit id: 6835a3a02b1ae222cbfb167a7e2d5e8df5e9854e)

Yoshihiro Shimoda (4):
  usb: renesas_usbhs: fix the behavior of some usbhs_pkt_handle
  usb: renesas_usbhs: protect mod->irq_{bemp,brdy}sts by spin lock
  usb: renesas_usbhs: fix the condition of is_done in
    usbhsf_dma_push_done
  usb: renesas_usbhs: fix the usb_pkt_pop()

 drivers/usb/renesas_usbhs/fifo.c |   72 ++++++++++++++++++++++++++++++++++----
 drivers/usb/renesas_usbhs/mod.c  |    5 +++
 drivers/usb/renesas_usbhs/pipe.c |   13 +++++++
 drivers/usb/renesas_usbhs/pipe.h |    4 +++
 4 files changed, 88 insertions(+), 6 deletions(-)

-- 
1.7.9.5

[PATCH 0/4] usb: renesas_usbhs: fixes this driver in gadget mode (part2)

[Yoshihiro Shimoda]

If we tested this driver as gadget, some issues below happened:
 - Oops happened if g_audio.
 - A usb enumeration may fail if we do insmod during connected the usb cable.
 - A transaction will not finish if g_zero.
 - A usb enumeration may fail after we re-connected the usb cable.

This patch serias fixes them.
This patch serias is based on Felipe's usb.git, testing/fixes branch.
(commit id: 0b93a4c838fa10370d72f86fe712426ac63804de)

Yoshihiro Shimoda (4):
  usb: renesas_usbhs: gadget: fix NULL pointer dereference in
    ep_disable()
  usb: renesas_usbhs: gadget: fix the behavior of pullup
  usb: renesas_usbhs: fix the timing of dcp_control_transfer_done
  usb: renesas_usbhs: fix usbhs_pipe_clear() for DCP PIPE

 drivers/usb/renesas_usbhs/common.c     |    6 ++++--
 drivers/usb/renesas_usbhs/fifo.c       |   34 ++++++++++++++++---------------
 drivers/usb/renesas_usbhs/fifo.h       |    1 +
 drivers/usb/renesas_usbhs/mod_gadget.c |   35 ++++++++++++++++++++++++++++++--
 drivers/usb/renesas_usbhs/mod_host.c   |    2 +-
 drivers/usb/renesas_usbhs/pipe.c       |    8 ++++++--
 6 files changed, 63 insertions(+), 23 deletions(-)

-- 
1.7.9.5

Re: [PATCH 0/4] usb: renesas_usbhs: fixes this driver in gadget mode (part2)

[Felipe Balbi]

[-- Attachment #1: Type: text/plain, Size: 1065 bytes --]

On Wed, Sep 10, 2014 at 07:33:27PM +0900, Yoshihiro Shimoda wrote:
> If we tested this driver as gadget, some issues below happened:
>  - Oops happened if g_audio.
>  - A usb enumeration may fail if we do insmod during connected the usb cable.
>  - A transaction will not finish if g_zero.
>  - A usb enumeration may fail after we re-connected the usb cable.
> 
> This patch serias fixes them.
> This patch serias is based on Felipe's usb.git, testing/fixes branch.
> (commit id: 0b93a4c838fa10370d72f86fe712426ac63804de)

Can you start sending your patches as a thread ? It makes it easier to
find them. If you're already formatting patches to a directory, then all
you have to git send-email the difectory, instead of all files in that.

e.g.:

	$ git format-patch -o patches --cover-letter -4
	$ vim patches/0000-cover-letter*
	$ git send-email --to balbi@ti.com --cc linux-usb@vger.kernel.org \
		patches

With that said, do we need these patches in v3.17 final ? or only v3.18?
It's getting pretty late for v3.17-final...

-- 
balbi

[-- Attachment #2: Digital signature --]
[-- Type: application/pgp-signature, Size: 819 bytes --]

Re: [PATCH 0/4] usb: renesas_usbhs: fixes this driver in gadget mode (part2)

[Yoshihiro Shimoda]

Hi Felipe,

(2014/09/10 22:49), Felipe Balbi wrote:
> On Wed, Sep 10, 2014 at 07:33:27PM +0900, Yoshihiro Shimoda wrote:
>> If we tested this driver as gadget, some issues below happened:
>>  - Oops happened if g_audio.
>>  - A usb enumeration may fail if we do insmod during connected the usb cable.
>>  - A transaction will not finish if g_zero.
>>  - A usb enumeration may fail after we re-connected the usb cable.
>>
>> This patch serias fixes them.
>> This patch serias is based on Felipe's usb.git, testing/fixes branch.
>> (commit id: 0b93a4c838fa10370d72f86fe712426ac63804de)
> 
> Can you start sending your patches as a thread ? It makes it easier to
> find them. If you're already formatting patches to a directory, then all
> you have to git send-email the difectory, instead of all files in that.
> 
> e.g.:
> 
> 	$ git format-patch -o patches --cover-letter -4
> 	$ vim patches/0000-cover-letter*
> 	$ git send-email --to balbi@ti.com --cc linux-usb@vger.kernel.org \
> 		patches

Thank you for the comment. Yes, I will try this command from next time.

> With that said, do we need these patches in v3.17 final ? or only v3.18?
> It's getting pretty late for v3.17-final...

I understand it's getting pretty late for v3.17-final.
So, I will send these patches for v3.18.

Best regards,
Yoshihiro Shimoda