Re: [PATCH v2] HID: cp2112: support large i2c transfers in hid-cp2112

[Ellen Wang <ellen-qUQiAmfTcIp+XZJcv9eMoEEOCMrvLtNR@public.gmane.org>]

On 07/08/2015 05:07 AM, Vaibhav Hiremath wrote:
>
>
> On Wednesday 08 July 2015 03:55 PM, Ellen Wang wrote:
>> cp2112_i2c_xfer() only reads up to 61 bytes, returning EIO
>> on longers reads.  The fix is to wrap a loop around
>> cp2112_read() to pick up all the returned data.
>>
>> Signed-off-by: Ellen Wang <ellen-qUQiAmfTcIp+XZJcv9eMoEEOCMrvLtNR@public.gmane.org>
>> ---
>> This is the updated patch with a check for 0 return from
>> cp2112_read().  I tested it with a suitable delay in the loop
>> to trigger the cp2112_raw_event() overrun bug, which must
>> be fixed before this patch is applied.
>> ---
>>   drivers/hid/hid-cp2112.c | 33 ++++++++++++++++++++++++++-------
>>   1 file changed, 26 insertions(+), 7 deletions(-)
>>
>> diff --git a/drivers/hid/hid-cp2112.c b/drivers/hid/hid-cp2112.c
>> index 3318de6..e2ffac0 100644
>> --- a/drivers/hid/hid-cp2112.c
>> +++ b/drivers/hid/hid-cp2112.c
>> @@ -509,13 +509,32 @@ static int cp2112_i2c_xfer(struct i2c_adapter
>> *adap, struct i2c_msg *msgs,
>>       if (!(msgs->flags & I2C_M_RD))
>>           goto finish;
>>
>> -    ret = cp2112_read(dev, msgs->buf, msgs->len);
>> -    if (ret < 0)
>> -        goto power_normal;
>> -    if (ret != msgs->len) {
>> -        hid_warn(hdev, "short read: %d < %d\n", ret, msgs->len);
>> -        ret = -EIO;
>> -        goto power_normal;
>> +    for (count = 0; count < msgs->len;) {
>> +        ret = cp2112_read(dev, msgs->buf + count, msgs->len - count);
>> +        hid_warn(hdev, "read returned %d for %zd\n",
>> +             ret, msgs->len - count);
>
> Do you always want to throw warning here, unconditionally ?

Yeah.  Sorry.  I had debugging code in my workspace then ran git diff 
with the wrong options.  I'll resend.

>> +        if (ret < 0)
>> +            goto power_normal;
>> +        if (ret == 0) {
>> +            hid_err(hdev, "read returned 0\n");
>> +            ret = -EIO;
>> +            goto power_normal;
>> +        }
>
> bit simplified, I guess :)
>
> if (ret < 0 || ret == 0) {
>      hid_err(hdev, "read returned %d", ret);
>      ret = ret == 0 ? -EIO : ret;
>      goto power_normal;
> }
>
>
>> +        count += ret;
>> +        if (count > msgs->len) {
>> +            /*
>> +             * The hardware returned too much data.
>> +             * This is mostly harmless because cp2112_read()
>> +             * has a limit check so didn't overrun our
>> +             * buffer.  Nevertheless, we return an error
>> +             * because something is seriously wrong and
>> +             * it shouldn't go unnoticed.
>> +             */
>> +            hid_err(hdev, "long read: %d > %zd\n",
>> +                ret, msgs->len - count + ret);
>
> You may want to take another look here.
> 'ret' will be either,
>
>      - ret = msgs->len
>          Not applicable
>      - ret > msgs->len
>          (count > msgs->len) will happen in one single
>          iteration, and will
>      - ret < msgs->len
>          (count > msgs->len) will happen in multiple iterations
>          where count keeps incrementing based on ret
>
> In the 2 scenarios above, I believe you would want to show,
>
>      actual read bytes > requested read bytes
>
>
> Am I missing something here?

(count > msgs->len) should never happen, so there's really no predicting 
it.  Or do you mean something else?


> Thanks,
> Vaibhav

Thank you.