[PATCH v1 0/3] Bluetooth: fix recv_buf() return value

[PATCH v1 0/3] Bluetooth: fix recv_buf() return value

[Francesco Dolcini]

From: Francesco Dolcini <francesco.dolcini@toradex.com>

Serdev recv_buf() callback is supposed to return the amount of bytes consumed, therefore an int in between 0 and count.

Do not return negative number in case of issue, just print an error and return count. This fixes a WARN in ttyport_receive_buf().

In addition to that a small cleanup patch is added on btnxpuart to remove a useless assignment.

Francesco Dolcini (3):
  Bluetooth: btnxpuart: fix recv_buf() return value
  Bluetooth: btmtkuart: fix recv_buf() return value
  Bluetooth: btnxpuart: remove useless assignment

 drivers/bluetooth/btmtkuart.c | 11 +++--------
 drivers/bluetooth/btnxpuart.c |  8 +++-----
 2 files changed, 6 insertions(+), 13 deletions(-)

-- 
2.25.1

[PATCH v1 2/3] Bluetooth: btmtkuart: fix recv_buf() return value

[Francesco Dolcini]

From: Francesco Dolcini <francesco.dolcini@toradex.com>

Serdev recv_buf() callback is supposed to return the amount of bytes
consumed, therefore an int in between 0 and count.

Do not return negative number in case of issue, just print an error and
return count. This fixes a WARN in ttyport_receive_buf().

Link: https://lore.kernel.org/all/087be419-ec6b-47ad-851a-5e1e3ea5cfcc@kernel.org/
Fixes: 7237c4c9ec92 ("Bluetooth: mediatek: Add protocol support for MediaTek serial devices")
Signed-off-by: Francesco Dolcini <francesco.dolcini@toradex.com>
---
 drivers/bluetooth/btmtkuart.c | 11 +++--------
 1 file changed, 3 insertions(+), 8 deletions(-)

diff --git a/drivers/bluetooth/btmtkuart.c b/drivers/bluetooth/btmtkuart.c
index 935feab815d9..203a000a84e3 100644
--- a/drivers/bluetooth/btmtkuart.c
+++ b/drivers/bluetooth/btmtkuart.c
@@ -336,7 +336,7 @@ mtk_stp_split(struct btmtkuart_dev *bdev, const unsigned char *data, int count,
 	return data;
 }
 
-static int btmtkuart_recv(struct hci_dev *hdev, const u8 *data, size_t count)
+static void btmtkuart_recv(struct hci_dev *hdev, const u8 *data, size_t count)
 {
 	struct btmtkuart_dev *bdev = hci_get_drvdata(hdev);
 	const unsigned char *p_left = data, *p_h4;
@@ -375,25 +375,20 @@ static int btmtkuart_recv(struct hci_dev *hdev, const u8 *data, size_t count)
 			bt_dev_err(bdev->hdev,
 				   "Frame reassembly failed (%d)", err);
 			bdev->rx_skb = NULL;
-			return err;
+			return;
 		}
 
 		sz_left -= sz_h4;
 		p_left += sz_h4;
 	}
-
-	return 0;
 }
 
 static int btmtkuart_receive_buf(struct serdev_device *serdev, const u8 *data,
 				 size_t count)
 {
 	struct btmtkuart_dev *bdev = serdev_device_get_drvdata(serdev);
-	int err;
 
-	err = btmtkuart_recv(bdev->hdev, data, count);
-	if (err < 0)
-		return err;
+	btmtkuart_recv(bdev->hdev, data, count);
 
 	bdev->hdev->stat.byte_rx += count;
 
-- 
2.25.1

Re: [PATCH v1 0/3] Bluetooth: fix recv_buf() return value

[Francesco Dolcini]

Hello Jiri,

On Mon, Nov 27, 2023 at 08:14:05PM +0100, Francesco Dolcini wrote:
> From: Francesco Dolcini <francesco.dolcini@toradex.com>
> 
> Serdev recv_buf() callback is supposed to return the amount of bytes
> consumed, therefore an int in between 0 and count.

I have also a patch ready to convert the return value of serdev
recv_buf() from int to size_t.

I would be inclined to wait for this series to go though first, given
that these are fixes, while the change from int to size_t is just a
cleanup to prevent future mistakes. Do you agree of would you do it
differently?

Francesco

Re: [PATCH v1 0/3] Bluetooth: fix recv_buf() return value

[Jiri Slaby]

Hi,

On 27. 11. 23, 20:23, Francesco Dolcini wrote:
> On Mon, Nov 27, 2023 at 08:14:05PM +0100, Francesco Dolcini wrote:
>> From: Francesco Dolcini <francesco.dolcini@toradex.com>
>>
>> Serdev recv_buf() callback is supposed to return the amount of bytes
>> consumed, therefore an int in between 0 and count.
> 
> I have also a patch ready to convert the return value of serdev
> recv_buf() from int to size_t.
> 
> I would be inclined to wait for this series to go though first, given
> that these are fixes, while the change from int to size_t is just a
> cleanup to prevent future mistakes. Do you agree of would you do it
> differently?

Fine by me either way. You can include it in this series at the end. 
Fixes can be picked up by stable too, the rest would go to mainline only.

thanks,
-- 
js
suse labs