date:20140227

Re: [PATCH 1/2] dma: cppi41: start tear down only if channel is busy

2014-02-27 Thread Shevchenko, Andriy

On Thu, 2014-02-27 at 10:44 +0530, George Cherian wrote:
 Start the channel tear down only if the channel is busy, else just
 bail out. In some cases its seen that by the time the tear down is
 initiated the cppi completes the DMA, especially in ISOCH transfers.
 
 Signed-off-by: George Cherian george.cher...@ti.com
 ---
  drivers/dma/cppi41.c | 7 +--
  1 file changed, 5 insertions(+), 2 deletions(-)
 
 diff --git a/drivers/dma/cppi41.c b/drivers/dma/cppi41.c
 index c18aebf..d028f36 100644
 --- a/drivers/dma/cppi41.c
 +++ b/drivers/dma/cppi41.c
 @@ -620,12 +620,15 @@ static int cppi41_stop_chan(struct dma_chan *chan)
   u32 desc_phys;
   int ret;
  
 + desc_phys = lower_32_bits(c-desc_phys);
 + desc_num = (desc_phys - cdd-descs_phys) / sizeof(struct cppi41_desc);
 + if (!cdd-chan_busy[desc_num])
 + return 0;
 +
   ret = cppi41_tear_down_chan(c);
   if (ret)
   return ret;
  
 - desc_phys = lower_32_bits(c-desc_phys);
 - desc_num = (desc_phys - cdd-descs_phys) / sizeof(struct cppi41_desc);
   WARN_ON(!cdd-chan_busy[desc_num]);

Do you still need this WARN_ON?

   cdd-chan_busy[desc_num] = NULL;
  


-- 
Andy Shevchenko andriy.shevche...@intel.com
Intel Finland Oy
-
Intel Finland Oy
Registered Address: PL 281, 00181 Helsinki 
Business Identity Code: 0357606 - 4 
Domiciled in Helsinki 

This e-mail and any attachments may contain confidential material for
the sole use of the intended recipient(s). Any review or distribution
by others is strictly prohibited. If you are not the intended
recipient, please contact the sender and delete all copies.

RE: [PATCH 2/6] usb: gadget: mv_udc: disable HW zlt for ep0

2014-02-27 Thread Neil Zhang

-Original Message-
From: Felipe Balbi [mailto:ba...@ti.com]
Sent: 2014年2月26日 23:19
To: Neil Zhang
Cc: ba...@ti.com; Peter Chen; gre...@linuxfoundation.org;
linux-usb@vger.kernel.org; linux-ker...@vger.kernel.org; Alexander Shishkin
Subject: Re: [PATCH 2/6] usb: gadget: mv_udc: disable HW zlt for ep0

Hi,

On Tue, Feb 25, 2014 at 05:48:17PM -0800, Neil Zhang wrote:
Hardware zlt will try to send the zero length packet
automatically when the data transferd is multiple times
of max packet, this will cause issues on Windows.
So let's disable HW zlt by default.

Would you have description that what kinds of issue on
Windows if zlt is is selected?

Enumeration will fail.

What causes enumeration fail, why it does not occur before?

A unexpected zero packet cause enumeration fail.
It's not easy that the descriptor is actually 1024 bytes, so not
easy to be found.

Chipidea bug too? Does it follow ch 8.5.3.2 Variable-length Data
Stage, USB
2.0 spec?

wait, this is a chipidea core ? Why aren't you guys using the
chipidea driver yet ? You need to switch over to that driver dude,
we can't have duplicated code in the tree.

I'm sorry, but I won't be taking this series, please use chipidea
driver, it should be very simple to add a glue layer for your core to the
chipidea driver.

Yes, it use chipidea IP.
But the driver is earlier than the chipidea one and we use it for our
products.
So it may be not that easy to switch to chipidea driver due to the
stability.

that's nonsense, the average chipidea glue layer is ~80 LOCs. You can write
that in less than 2 hours and give it a try. We cannot have duplicate drivers
in
the tree and development effort *must* be shared.

If you guys use the same IP, why wouldn't you use the same chipidea driver ?

sorry, you didn't convince me.

It's too sad!
Anyway we will estimate the chipidea driver to see whether it meets our
requirement.

--
Balbi

Best Regards,
Neil Zhang
N�Р骒r��yb�X�肚�v�^�)藓{.n�+�伐�{焙柒��^n�r■�z���h�ㄨ��Ⅷ�G���h�(�茛j���m赇z罐��帼f＂�h���~�m�

76 matches

Mail list logo