LinuxLists.cc - [PATCH] ath9k-htc: fix probe failure

2010-11-01 14:32:39

Subject: [PATCH] ath9k-htc: fix probe failure

From: Ming Lei <[email protected]>

Obviously, we should set the transfer type consitent with
endpoint descriptor type to urb, this patch fixes the probe
failure below:

[ 8.018768] PM: Removing info for No Bus:1-2.2
[ 8.019470] usb 1-2.2: BOGUS urb xfer, pipe 3 != type 1
[ 8.031707] usb 1-2.2: ath9k_htc: Unable to allocate URBs
[ 8.037506] ath9k_hif_usb: probe of 1-2.2:1.0 failed with error -22
[ 8.044250] usbcore: registered new interface driver ath9k_hif_usb

Signed-off-by: Ming Lei <[email protected]>
---
drivers/net/wireless/ath/ath9k/hif_usb.c | 4 ++--
1 files changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/net/wireless/ath/ath9k/hif_usb.c b/drivers/net/wireless/ath/ath9k/hif_usb.c
index 728d904..04242bc 100644
--- a/drivers/net/wireless/ath/ath9k/hif_usb.c
+++ b/drivers/net/wireless/ath/ath9k/hif_usb.c
@@ -541,7 +541,7 @@ static void ath9k_hif_usb_reg_in_cb(struct urb *urb)
}

usb_fill_int_urb(urb, hif_dev->udev,
- usb_rcvbulkpipe(hif_dev->udev,
+ usb_rcvintpipe(hif_dev->udev,
USB_REG_IN_PIPE),
nskb->data, MAX_REG_IN_BUF_SIZE,
ath9k_hif_usb_reg_in_cb, nskb, 1);
@@ -721,7 +721,7 @@ static int ath9k_hif_usb_alloc_reg_in_urb(struct hif_device_usb *hif_dev)
goto err;

usb_fill_int_urb(hif_dev->reg_in_urb, hif_dev->udev,
- usb_rcvbulkpipe(hif_dev->udev,
+ usb_rcvintpipe(hif_dev->udev,
USB_REG_IN_PIPE),
skb->data, MAX_REG_IN_BUF_SIZE,
ath9k_hif_usb_reg_in_cb, skb, 1);
--
1.7.3

2010-11-01 16:39:10

by Rajkumar Manoharan

[permalink] [raw]

Subject: RE: [PATCH] ath9k-htc: fix probe failure

2010-11-03 14:13:20

by Ming Lei

[permalink] [raw]

Subject: Re: [PATCH] ath9k-htc: fix probe failure

2010/11/2 Rajkumar Manoharan <[email protected]>:
> From: Ming Lei <[email protected]>
>
> Obviously, we should set the transfer type consitent with
> endpoint descriptor type to urb, this patch fixes the probe
> failure below:
>
> [ ? ?8.018768] PM: Removing info for No Bus:1-2.2
> [ ? ?8.019470] usb 1-2.2: BOGUS urb xfer, pipe 3 != type 1
> [ ? ?8.031707] usb 1-2.2: ath9k_htc: Unable to allocate URBs
> [ ? ?8.037506] ath9k_hif_usb: probe of 1-2.2:1.0 failed with error -22
> [ ? ?8.044250] usbcore: registered new interface driver ath9k_hif_usb
>
> NACK. To optimize CPU consumption and scan duration, ?HTC device firmwares
> were patched to change descriptor types of EP3 and EP4 from Interrupt to Bulk.
> After firmware download in initialization, endpoint descriptors got changed.
> Please use the latest firmwares from linux-firmware git tree.
>
> http://git.kernel.org/?p=linux/kernel/git/dwmw2/linux-firmware.git;a=summary

My card is UB91, so you means I should update to this one(ar9271.fw) blew:

http://git.kernel.org/?p=linux/kernel/git/dwmw2/linux-firmware.git;a=blob;f=ar9271.fw;h=0ec953b404c1912dc64fa3b0171cffae82b0f375;hb=8f919160792e4702c6b7a67a243cea4f757407e4

This one still can't prevent the probe failure...

Any suggestions?

thanks,
--
Lei Ming

2010-11-03 16:41:06

by Rajkumar Manoharan

[permalink] [raw]

Subject: RE: [PATCH] ath9k-htc: fix probe failure

>2010/11/2 Rajkumar Manoharan <[email protected]>:
>> From: Ming Lei <[email protected]>
>>
>> Obviously, we should set the transfer type consitent with
>> endpoint descriptor type to urb, this patch fixes the probe
>> failure below:
>>
>>[ 8.018768] PM: Removing info for No Bus:1-2.2
>> [ 8.019470] usb 1-2.2: BOGUS urb xfer, pipe 3 != type 1
>> [ 8.031707] usb 1-2.2: ath9k_htc: Unable to allocate URBs
>> [ 8.037506] ath9k_hif_usb: probe of 1-2.2:1.0 failed with error -22
>> [ 8.044250] usbcore: registered new interface driver ath9k_hif_usb
>>
>> NACK. To optimize CPU consumption and scan duration, HTC device firmwares
>> were patched to change descriptor types of EP3 and EP4 from Interrupt to Bulk.
>> After firmware download in initialization, endpoint descriptors got changed.
>> Please use the latest firmwares from linux-firmware git tree.
>>
>> http://git.kernel.org/?p=linux/kernel/git/dwmw2/linux-firmware.git;a=summary
>
>My card is UB91, so you means I should update to this one(ar9271.fw) blew:
>
>http://git.kernel.org/?p=linux/kernel/git/dwmw2/linux->firmware.git;a=blob;f=ar9271.fw;h=0ec953b404c1912dc64fa3b0171cffae82b0f375;hb=8f919160792e4702c6b7a67a243cea4f757407e4
>
>This one still can't prevent the probe failure...

>Any suggestions?

Could you please try it without CONFIG_USB_DEBUG option.

Thanks,
Rajkumar

2010-11-04 02:51:15

by Rajkumar Manoharan

[permalink] [raw]

Subject: RE: [PATCH] ath9k-htc: fix probe failure

>2010/11/4 Rajkumar Manoharan <[email protected]>:
>>>2010/11/2 Rajkumar Manoharan <[email protected]>:
>>>> From: Ming Lei <[email protected]>
>>>>
>>>> Obviously, we should set the transfer type consitent with
>>>> endpoint descriptor type to urb, this patch fixes the probe
>>>> failure below:
>>>>
>>>>[ 8.018768] PM: Removing info for No Bus:1-2.2
>>>> [ 8.019470] usb 1-2.2: BOGUS urb xfer, pipe 3 != type 1
>>>> [ 8.031707] usb 1-2.2: ath9k_htc: Unable to allocate URBs
>>>> [ 8.037506] ath9k_hif_usb: probe of 1-2.2:1.0 failed with error -22
>>>> [ 8.044250] usbcore: registered new interface driver ath9k_hif_usb
>>>>
>>>> NACK. To optimize CPU consumption and scan duration, HTC device firmwares
>>>> were patched to change descriptor types of EP3 and EP4 from Interrupt to Bulk.
>>>> After firmware download in initialization, endpoint descriptors got changed.
>>>> Please use the latest firmwares from linux-firmware git tree.
>>>>
>>>> http://git.kernel.org/?p=linux/kernel/git/dwmw2/linux-firmware.git;a=summary
>>>
>>>My card is UB91, so you means I should update to this one(ar9271.fw) blew:
>>>
>>>http://git.kernel.org/?p=linux/kernel/git/dwmw2/linux->>firmware.git;a=blob;f=ar9271.fw;h=0ec953b404c1912dc64fa3b0171cffae82b0f375;hb=8f919160792e4702c6b7a67a243cea4f757407e4
>>>
>>>This one still can't prevent the probe failure...
>>
>>>Any suggestions?
>>
>> Could you please try it without CONFIG_USB_DEBUG option.
>
>I don't think disabling CONFIG_USB_DEBUG can prevent the .probe failure,
>which only don't print some messages.
>
>So this patch does make sense, doesn't it?

The recent commit "check the endpoint type against the pipe type" in usb under DEBUG is causing probe failure,
It's conditional on CONFIG_USB_DEBUG, so it won't affect normal users. But this probe failure under DEBUG
needs to be addressed without causing any IOT issue.

Thanks,
Rajkumar

2010-11-04 02:02:37

by Ming Lei

[permalink] [raw]

Subject: Re: [PATCH] ath9k-htc: fix probe failure

2010/11/4 Rajkumar Manoharan <[email protected]>:
>>2010/11/2 Rajkumar Manoharan <[email protected]>:
>>> From: Ming Lei <[email protected]>
>>>
>>> Obviously, we should set the transfer type consitent with
>>> endpoint descriptor type to urb, this patch fixes the probe
>>> failure below:
>>>
>>>[ ? ?8.018768] PM: Removing info for No Bus:1-2.2
>>> [ ? ?8.019470] usb 1-2.2: BOGUS urb xfer, pipe 3 != type 1
>>> [ ? ?8.031707] usb 1-2.2: ath9k_htc: Unable to allocate URBs
>>> [ ? ?8.037506] ath9k_hif_usb: probe of 1-2.2:1.0 failed with error -22
>>> [ ? ?8.044250] usbcore: registered new interface driver ath9k_hif_usb
>>>
>>> NACK. To optimize CPU consumption and scan duration, ?HTC device firmwares
>>> were patched to change descriptor types of EP3 and EP4 from Interrupt to Bulk.
>>> After firmware download in initialization, endpoint descriptors got changed.
>>> Please use the latest firmwares from linux-firmware git tree.
>>>
>>> http://git.kernel.org/?p=linux/kernel/git/dwmw2/linux-firmware.git;a=summary
>>
>>My card is UB91, so you means I should update to this one(ar9271.fw) blew:
>>
>>http://git.kernel.org/?p=linux/kernel/git/dwmw2/linux->firmware.git;a=blob;f=ar9271.fw;h=0ec953b404c1912dc64fa3b0171cffae82b0f375;hb=8f919160792e4702c6b7a67a243cea4f757407e4
>>
>>This one still can't prevent the probe failure...
>
>>Any suggestions?
>
> Could you please try it without CONFIG_USB_DEBUG option.

I don't think disabling CONFIG_USB_DEBUG can prevent the .probe failure,
which only don't print some messages.

So this patch does make sense, doesn't it?

thanks,
--
Lei Ming