2014-02-17 21:24:51

by Malcolm Priestley

Subject: [PATCH 4/5] staging: vt6656: Remove variable fKillEventPollingThread

We already kill the urb and since patch
s_nsInterruptUsbIoCompleteRead add urb status returns.
have error handling

There is no need for this variable.

Signed-off-by: Malcolm Priestley <[email protected]>

Conflicts:
drivers/staging/vt6656/usbpipe.c
---
drivers/staging/vt6656/device.h | 1 -
drivers/staging/vt6656/int.c | 5 +++--
drivers/staging/vt6656/main_usb.c | 2 --
drivers/staging/vt6656/usbpipe.c | 4 +---
4 files changed, 4 insertions(+), 8 deletions(-)

diff --git a/drivers/staging/vt6656/device.h b/drivers/staging/vt6656/device.h
index ec20b14..e5f0be3 100644
--- a/drivers/staging/vt6656/device.h
+++ b/drivers/staging/vt6656/device.h
@@ -421,7 +421,6 @@ struct vnt_private {

/* Variables to track resources for the Interrupt In Pipe */
INT_BUFFER intBuf;
- int fKillEventPollingThread;
int bEventAvailable;

/* default config from file by user setting */
diff --git a/drivers/staging/vt6656/int.c b/drivers/staging/vt6656/int.c
index ec135b4..34c4528 100644
--- a/drivers/staging/vt6656/int.c
+++ b/drivers/staging/vt6656/int.c
@@ -70,8 +70,9 @@ void INTvWorkItem(struct vnt_private *pDevice)
DBG_PRT(MSG_LEVEL_DEBUG, KERN_INFO"---->Interrupt Polling Thread\n");

spin_lock_irq(&pDevice->lock);
- if (pDevice->fKillEventPollingThread != true)
- ntStatus = PIPEnsInterruptRead(pDevice);
+
+ ntStatus = PIPEnsInterruptRead(pDevice);
+
spin_unlock_irq(&pDevice->lock);
}

diff --git a/drivers/staging/vt6656/main_usb.c b/drivers/staging/vt6656/main_usb.c
index 4f913d3..bea1ad9 100644
--- a/drivers/staging/vt6656/main_usb.c
+++ b/drivers/staging/vt6656/main_usb.c
@@ -1000,7 +1000,6 @@ static int device_open(struct net_device *dev)
pDevice->eEncryptionStatus = Ndis802_11EncryptionDisabled;

pDevice->bIsRxWorkItemQueued = true;
- pDevice->fKillEventPollingThread = false;
pDevice->bEventAvailable = false;

pDevice->bWPADEVUp = false;
@@ -1084,7 +1083,6 @@ static int device_close(struct net_device *dev)
MP_SET_FLAG(pDevice, fMP_DISCONNECTED);
MP_CLEAR_FLAG(pDevice, fMP_POST_WRITES);
MP_CLEAR_FLAG(pDevice, fMP_POST_READS);
- pDevice->fKillEventPollingThread = true;

cancel_delayed_work_sync(&pDevice->run_command_work);
cancel_delayed_work_sync(&pDevice->second_callback_work);
diff --git a/drivers/staging/vt6656/usbpipe.c b/drivers/staging/vt6656/usbpipe.c
index 1ab1bbc..3721937 100644
--- a/drivers/staging/vt6656/usbpipe.c
+++ b/drivers/staging/vt6656/usbpipe.c
@@ -393,7 +393,6 @@ static void s_nsInterruptUsbIoCompleteRead(struct urb *urb)

// if (ntStatus == STATUS_NOT_CONNECTED )
// {
- pDevice->fKillEventPollingThread = true;
// }
DBG_PRT(MSG_LEVEL_DEBUG, KERN_INFO"IntUSBIoCompleteControl STATUS = %d\n", ntStatus );
} else {
@@ -403,7 +402,6 @@ static void s_nsInterruptUsbIoCompleteRead(struct urb *urb)
INTnsProcessData(pDevice);
}

- if (pDevice->fKillEventPollingThread != true) {
ntStatus = usb_submit_urb(pDevice->pInterruptURB, GFP_ATOMIC);
if (ntStatus) {
DBG_PRT(MSG_LEVEL_DEBUG, KERN_INFO
@@ -411,7 +409,7 @@ static void s_nsInterruptUsbIoCompleteRead(struct urb *urb)
} else {
pDevice->intBuf.bInUse = true;
}
- }
+
//
// We return STATUS_MORE_PROCESSING_REQUIRED so that the completion
// routine (IofCompleteRequest) will stop working on the irp.
--
1.9.rc1

2014-02-18 20:22:44

by Malcolm Priestley

Subject: Re: [PATCH 4/5] staging: vt6656: Remove variable fKillEventPollingThread

On 18/02/14 19:00, Greg KH wrote:
> On Mon, Feb 17, 2014 at 09:24:33PM +0000, Malcolm Priestley wrote:
>> We already kill the urb and since patch
>> s_nsInterruptUsbIoCompleteRead add urb status returns.
>> have error handling
>>
>> There is no need for this variable.
>>
>> Signed-off-by: Malcolm Priestley <[email protected]>
>>
>> Conflicts:
>> drivers/staging/vt6656/usbpipe.c
>
> Why is this in your Changelog entry?
>
Oh sorry, I should have deleted it.

It was cherry picked from another git branch where patch 1
didn't exist and was fixed.

2014-02-18 18:59:21

by Greg Kroah-Hartman

Subject: Re: [PATCH 4/5] staging: vt6656: Remove variable fKillEventPollingThread

On Mon, Feb 17, 2014 at 09:24:33PM +0000, Malcolm Priestley wrote:
> We already kill the urb and since patch
> s_nsInterruptUsbIoCompleteRead add urb status returns.
> have error handling
>
> There is no need for this variable.
>
> Signed-off-by: Malcolm Priestley <[email protected]>
>
> Conflicts:
> drivers/staging/vt6656/usbpipe.c

Why is this in your Changelog entry?