Re: [PATCH v4] HID: ft260: improve error handling of ft260_hid_feature_report_get()

[Tom Rix]

On 5/11/21 7:34 AM, Michael Zaidman wrote:
> On Tue, May 11, 2021 at 06:10:36AM -0700, Tom Rix wrote:
> > Generally change is fine.
> >
> > a nit below.
> >
> > On 5/11/21 3:12 AM, Michael Zaidman wrote:
> > > Fixes: 6a82582d9fa4 ("HID: ft260: add usb hid to i2c host bridge driver")
> > >
> > > The ft260_hid_feature_report_get() checks if the return size matches
> > > the requested size. But the function can also fail with at least -ENOMEM.
> > > Add the < 0 checks.
> > >
> > > In ft260_hid_feature_report_get(), do not do the memcpy to the caller's
> > > buffer if there is an error.
> > >
> > > ---
> > > v4   Fixed commit message
> > > ---
> > > v3   Simplify and optimize the changes
> > > ---
> > > v2:  add unlikely()'s for error conditions
> > > ---
> > >
> > > Signed-off-by: Tom Rix <trix@xxxxxxxxxx>
> > > Signed-off-by: Michael Zaidman <michael.zaidman@xxxxxxxxx>
> > > ---
> > >    drivers/hid/hid-ft260.c | 24 ++++++++++++------------
> > >    1 file changed, 12 insertions(+), 12 deletions(-)
> > >
> > > diff --git a/drivers/hid/hid-ft260.c b/drivers/hid/hid-ft260.c
> > > index 047aa85a7c83..7f4cb823129e 100644
> > > --- a/drivers/hid/hid-ft260.c
> > > +++ b/drivers/hid/hid-ft260.c
> > > @@ -249,7 +249,10 @@ static int ft260_hid_feature_report_get(struct hid_device *hdev,
> > >    	ret = hid_hw_raw_request(hdev, report_id, buf, len, HID_FEATURE_REPORT,
> > >    				 HID_REQ_GET_REPORT);
> > > -	memcpy(data, buf, len);
> > > +	if (likely(ret == len))
> > > +		memcpy(data, buf, len);
> > > +	else if (ret >= 0)
> > > +		ret = -EIO;
> > >    	kfree(buf);
> > >    	return ret;
> > >    }
> > > @@ -298,7 +301,7 @@ static int ft260_xfer_status(struct ft260_device *dev)
> > >    	ret = ft260_hid_feature_report_get(hdev, FT260_I2C_STATUS,
> > >    					   (u8 *)&report, sizeof(report));
> > > -	if (ret < 0) {
> > > +	if (unlikely(ret < 0)) {
> > >    		hid_err(hdev, "failed to retrieve status: %d\n", ret);
> > >    		return ret;
> > >    	}
> > > @@ -720,10 +723,9 @@ static int ft260_get_system_config(struct hid_device *hdev,
> > >    	ret = ft260_hid_feature_report_get(hdev, FT260_SYSTEM_SETTINGS,
> > >    					   (u8 *)cfg, len);
> > > -	if (ret != len) {
> > > +	if (ret < 0) {
> > nit: should be consistent and use unlikely(ret < 0) for this and other
> > similar checks.
> >
> > Tom
> I preserved the likely/unlikely hints in the critical path where the
> performance matters. And for the sake of consistency, I removed them from
> the rest of the places that are called rarely and are not performance-critical
> to be aligned to the other "if" statements in the code.

This is fine.

Thanks,

Tom

> Michael