[PATCH USB 1/2] usb: raw-gadget: fix return value of ep read ioctls

[PATCH USB 1/2] usb: raw-gadget: fix return value of ep read ioctls

[Andrey Konovalov]

They must return the number of bytes transferred during the data stage.

Signed-off-by: Andrey Konovalov <andreyknvl@google.com>
---
 drivers/usb/gadget/legacy/raw_gadget.c | 8 ++++++--
 1 file changed, 6 insertions(+), 2 deletions(-)

diff --git a/drivers/usb/gadget/legacy/raw_gadget.c b/drivers/usb/gadget/legacy/raw_gadget.c
index ca7d95bf7397..7b241992ad5a 100644
--- a/drivers/usb/gadget/legacy/raw_gadget.c
+++ b/drivers/usb/gadget/legacy/raw_gadget.c
@@ -669,12 +669,14 @@ static int raw_ioctl_ep0_read(struct raw_dev *dev, unsigned long value)
 	if (IS_ERR(data))
 		return PTR_ERR(data);
 	ret = raw_process_ep0_io(dev, &io, data, false);
-	if (ret)
+	if (ret < 0)
 		goto free;
 
 	length = min(io.length, (unsigned int)ret);
 	if (copy_to_user((void __user *)(value + sizeof(io)), data, length))
 		ret = -EFAULT;
+	else
+		ret = length;
 free:
 	kfree(data);
 	return ret;
@@ -964,12 +966,14 @@ static int raw_ioctl_ep_read(struct raw_dev *dev, unsigned long value)
 	if (IS_ERR(data))
 		return PTR_ERR(data);
 	ret = raw_process_ep_io(dev, &io, data, false);
-	if (ret)
+	if (ret < 0)
 		goto free;
 
 	length = min(io.length, (unsigned int)ret);
 	if (copy_to_user((void __user *)(value + sizeof(io)), data, length))
 		ret = -EFAULT;
+	else
+		ret = length;
 free:
 	kfree(data);
 	return ret;
-- 
2.26.2.303.gf8c07b1a785-goog

[PATCH USB 2/2] usb: raw-gadget: fix typo in uapi headers

[Andrey Konovalov]

Signed-off-by: Andrey Konovalov <andreyknvl@google.com>
---
 include/uapi/linux/usb/raw_gadget.h | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/include/uapi/linux/usb/raw_gadget.h b/include/uapi/linux/usb/raw_gadget.h
index ea375082b3ac..8544802b25bd 100644
--- a/include/uapi/linux/usb/raw_gadget.h
+++ b/include/uapi/linux/usb/raw_gadget.h
@@ -119,7 +119,7 @@ struct usb_raw_ep_io {
  * received on endpoint 0, provided that was an IN (OUT for READ) request and
  * waits until the urb is completed. Copies received data to user for READ.
  * Accepts a pointer to the usb_raw_ep_io struct as an argument.
- * Returns length of trasferred data on success or negative error code on
+ * Returns length of transferred data on success or negative error code on
  * failure.
  */
 #define USB_RAW_IOCTL_EP0_WRITE		_IOW('U', 3, struct usb_raw_ep_io)
@@ -145,7 +145,7 @@ struct usb_raw_ep_io {
  * request and waits until the urb is completed. Copies received data to user
  * for READ.
  * Accepts a pointer to the usb_raw_ep_io struct as an argument.
- * Returns length of trasferred data on success or negative error code on
+ * Returns length of transferred data on success or negative error code on
  * failure.
  */
 #define USB_RAW_IOCTL_EP_WRITE		_IOW('U', 7, struct usb_raw_ep_io)
-- 
2.26.2.303.gf8c07b1a785-goog

Re: [PATCH USB 2/2] usb: raw-gadget: fix typo in uapi headers

[Greg Kroah-Hartman]

On Fri, Apr 24, 2020 at 03:09:59AM +0200, Andrey Konovalov wrote:
> Signed-off-by: Andrey Konovalov <andreyknvl@google.com>
> ---

I can't take patches without any changelog text, sorry.

greg k-h

Re: [PATCH USB 2/2] usb: raw-gadget: fix typo in uapi headers

[Felipe Balbi]

[-- Attachment #1: Type: text/plain, Size: 149 bytes --]

Andrey Konovalov <andreyknvl@google.com> writes:

> Signed-off-by: Andrey Konovalov <andreyknvl@google.com>

return -ENOCHANGELOG

-- 
balbi

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 832 bytes --]

Re: [PATCH USB 2/2] usb: raw-gadget: fix typo in uapi headers

[Andrey Konovalov]

On Tue, May 5, 2020 at 9:50 AM Felipe Balbi <balbi@kernel.org> wrote:
>
> Andrey Konovalov <andreyknvl@google.com> writes:
>
> > Signed-off-by: Andrey Konovalov <andreyknvl@google.com>
>
> return -ENOCHANGELOG

I've sent v2 with changelog shortly after Greg's response. Would you
like me to resend all raw-gadget fixes formed as a series?

BTW, to whom should I send them, to you or to Greg? I've noticed that
some of my gadget patches from before got into the mainline twice.

Thanks!

Re: [PATCH USB 2/2] usb: raw-gadget: fix typo in uapi headers

[Greg Kroah-Hartman]

On Tue, May 05, 2020 at 02:11:44PM +0200, Andrey Konovalov wrote:
> On Tue, May 5, 2020 at 9:50 AM Felipe Balbi <balbi@kernel.org> wrote:
> >
> > Andrey Konovalov <andreyknvl@google.com> writes:
> >
> > > Signed-off-by: Andrey Konovalov <andreyknvl@google.com>
> >
> > return -ENOCHANGELOG
> 
> I've sent v2 with changelog shortly after Greg's response. Would you
> like me to resend all raw-gadget fixes formed as a series?
> 
> BTW, to whom should I send them, to you or to Greg? I've noticed that
> some of my gadget patches from before got into the mainline twice.

They really should all go through Felipe, I was just trying to help out
a bit in the past.  I'll let him deal with all of these :)

thanks,

greg k-h

Re: [PATCH USB 1/2] usb: raw-gadget: fix return value of ep read ioctls

[Greg Kroah-Hartman]

On Fri, Apr 24, 2020 at 03:09:58AM +0200, Andrey Konovalov wrote:
> They must return the number of bytes transferred during the data stage.
> 
> Signed-off-by: Andrey Konovalov <andreyknvl@google.com>
> ---

Is this a bugfix?  If so, does it need to go to older kernels?  A
"Fixes:" tag would be nice...

thanks,

greg k-h

Re: [PATCH USB 1/2] usb: raw-gadget: fix return value of ep read ioctls

[Dan Carpenter]

On Fri, Apr 24, 2020 at 03:09:58AM +0200, Andrey Konovalov wrote:
> They must return the number of bytes transferred during the data stage.
> 
> Signed-off-by: Andrey Konovalov <andreyknvl@google.com>

This was my mistake.  Please add a Fixes tag.

Fixes: 068fbff4f860 ("usb: raw-gadget: Fix copy_to/from_user() checks")

I should have seen that bug...  I thought I was being careful and I
even singled out that part of the commit and mentioned it in the
commit message but I messed up.  Sorry.

regards,
dan carpenter

Re: [PATCH USB 1/2] usb: raw-gadget: fix return value of ep read ioctls

[Andrey Konovalov]

On Fri, Apr 24, 2020 at 10:43 AM Dan Carpenter <dan.carpenter@oracle.com> wrote:
>
> On Fri, Apr 24, 2020 at 03:09:58AM +0200, Andrey Konovalov wrote:
> > They must return the number of bytes transferred during the data stage.
> >
> > Signed-off-by: Andrey Konovalov <andreyknvl@google.com>
>
> This was my mistake.  Please add a Fixes tag.
>
> Fixes: 068fbff4f860 ("usb: raw-gadget: Fix copy_to/from_user() checks")
>
> I should have seen that bug...  I thought I was being careful and I
> even singled out that part of the commit and mentioned it in the
> commit message but I messed up.  Sorry.

No worries, the bug was actually present before your change, but in a
slightly different form. So FWIW we can also add:

Fixes: f2c2e717642c ("usb: gadget: add raw-gadget interface")

However raw-gadget is not present in any released kernel yet, so need
of backporting AFAIU.

Re: [PATCH USB 1/2] usb: raw-gadget: fix return value of ep read ioctls

[Dan Carpenter]

On Fri, Apr 24, 2020 at 03:16:35PM +0200, Andrey Konovalov wrote:
> On Fri, Apr 24, 2020 at 10:43 AM Dan Carpenter <dan.carpenter@oracle.com> wrote:
> >
> > On Fri, Apr 24, 2020 at 03:09:58AM +0200, Andrey Konovalov wrote:
> > > They must return the number of bytes transferred during the data stage.
> > >
> > > Signed-off-by: Andrey Konovalov <andreyknvl@google.com>
> >
> > This was my mistake.  Please add a Fixes tag.
> >
> > Fixes: 068fbff4f860 ("usb: raw-gadget: Fix copy_to/from_user() checks")
> >
> > I should have seen that bug...  I thought I was being careful and I
> > even singled out that part of the commit and mentioned it in the
> > commit message but I messed up.  Sorry.
> 
> No worries, the bug was actually present before your change, but in a
> slightly different form. So FWIW we can also add:
> 
> Fixes: f2c2e717642c ("usb: gadget: add raw-gadget interface")
> 
> However raw-gadget is not present in any released kernel yet, so need
> of backporting AFAIU.

The Fixes: tag lets scripts know that it's not required to back port it.

regards,
dan carpenter