Do not call skb_pull() in msft_add_monitor_sync() as
msft_le_monitor_advertisement_cb() expects 'status' to be
part of the skb.
Same applies for msft_remove_monitor_sync().
Signed-off-by: Manish Mandlik <[email protected]>
---
net/bluetooth/msft.c | 2 --
1 file changed, 2 deletions(-)
diff --git a/net/bluetooth/msft.c b/net/bluetooth/msft.c
index f43994523b1f..9990924719aa 100644
--- a/net/bluetooth/msft.c
+++ b/net/bluetooth/msft.c
@@ -387,7 +387,6 @@ static int msft_remove_monitor_sync(struct hci_dev *hdev,
return PTR_ERR(skb);
status = skb->data[0];
- skb_pull(skb, 1);
msft_le_cancel_monitor_advertisement_cb(hdev, status, hdev->msft_opcode,
skb);
@@ -506,7 +505,6 @@ static int msft_add_monitor_sync(struct hci_dev *hdev,
return PTR_ERR(skb);
status = skb->data[0];
- skb_pull(skb, 1);
msft_le_monitor_advertisement_cb(hdev, status, hdev->msft_opcode, skb);
--
2.36.0.512.ge40c2bad7a-goog
This is automated email and please do not reply to this email!
Dear submitter,
Thank you for submitting the patches to the linux bluetooth mailing list.
This is a CI test results with your patch series:
PW Link:https://patchwork.kernel.org/project/bluetooth/list/?series=639914
---Test result---
Test Summary:
CheckPatch PASS 1.54 seconds
GitLint PASS 0.96 seconds
SubjectPrefix PASS 0.87 seconds
BuildKernel PASS 31.03 seconds
BuildKernel32 PASS 27.59 seconds
Incremental Build with patchesPASS 37.79 seconds
TestRunner: Setup PASS 469.43 seconds
TestRunner: l2cap-tester PASS 17.48 seconds
TestRunner: bnep-tester PASS 6.09 seconds
TestRunner: mgmt-tester PASS 100.65 seconds
TestRunner: rfcomm-tester PASS 9.66 seconds
TestRunner: sco-tester PASS 9.45 seconds
TestRunner: smp-tester PASS 9.36 seconds
TestRunner: userchan-tester PASS 6.37 seconds
---
Regards,
Linux Bluetooth
Dear Manish,
Thank you for your patch.
Am 09.05.22 um 23:05 schrieb Manish Mandlik:
> Do not call skb_pull() in msft_add_monitor_sync() as
> msft_le_monitor_advertisement_cb() expects 'status' to be
> part of the skb.
Please reflow for 75 characters per line.
> Same applies for msft_remove_monitor_sync().
Was this found by code review, or were there noticeable problems? If the
later, please add a note, how to reproduce it.
Also, maybe also add a Fixes tag, referencing the commit introducing the
problem.
Kind regards,
Paul
> Signed-off-by: Manish Mandlik <[email protected]>
> ---
>
> net/bluetooth/msft.c | 2 --
> 1 file changed, 2 deletions(-)
>
> diff --git a/net/bluetooth/msft.c b/net/bluetooth/msft.c
> index f43994523b1f..9990924719aa 100644
> --- a/net/bluetooth/msft.c
> +++ b/net/bluetooth/msft.c
> @@ -387,7 +387,6 @@ static int msft_remove_monitor_sync(struct hci_dev *hdev,
> return PTR_ERR(skb);
>
> status = skb->data[0];
> - skb_pull(skb, 1);
>
> msft_le_cancel_monitor_advertisement_cb(hdev, status, hdev->msft_opcode,
> skb);
> @@ -506,7 +505,6 @@ static int msft_add_monitor_sync(struct hci_dev *hdev,
> return PTR_ERR(skb);
>
> status = skb->data[0];
> - skb_pull(skb, 1);
>
> msft_le_monitor_advertisement_cb(hdev, status, hdev->msft_opcode, skb);
>
Hi Manish,
On Wed, May 11, 2022 at 5:56 PM Manish Mandlik <[email protected]> wrote:
>
> Hi Luiz,
>
> On Wed, May 11, 2022 at 2:23 PM Luiz Augusto von Dentz <[email protected]> wrote:
>>
>> Hi Manish,
>>
>> On Mon, May 9, 2022 at 2:05 PM Manish Mandlik <[email protected]> wrote:
>> >
>> > Do not call skb_pull() in msft_add_monitor_sync() as
>> > msft_le_monitor_advertisement_cb() expects 'status' to be
>> > part of the skb.
>> >
>> > Same applies for msft_remove_monitor_sync().
>> >
>> > Signed-off-by: Manish Mandlik <[email protected]>
>> > ---
>> >
>> > net/bluetooth/msft.c | 2 --
>> > 1 file changed, 2 deletions(-)
>> >
>> > diff --git a/net/bluetooth/msft.c b/net/bluetooth/msft.c
>> > index f43994523b1f..9990924719aa 100644
>> > --- a/net/bluetooth/msft.c
>> > +++ b/net/bluetooth/msft.c
>> > @@ -387,7 +387,6 @@ static int msft_remove_monitor_sync(struct hci_dev *hdev,
>> > return PTR_ERR(skb);
>> >
>> > status = skb->data[0];
>> > - skb_pull(skb, 1);
>> >
>> > msft_le_cancel_monitor_advertisement_cb(hdev, status, hdev->msft_opcode,
>> > skb);
>> > @@ -506,7 +505,6 @@ static int msft_add_monitor_sync(struct hci_dev *hdev,
>> > return PTR_ERR(skb);
>> >
>> > status = skb->data[0];
>> > - skb_pull(skb, 1);
>>
>> Well if it expects it to be part of the skb then there is no reason to
>> pass it as argument in addition to the skb itself.
>
> The problem is msft_le_monitor_advertisement_cb() is invoked directly via msft_add_monitor_sync() and also from __msft_add_monitor_pattern() as a callback from hci_req_run_skb(). So, when it is invoked from hci_req_run_skb() it sends status separately as an argument along with the skb and that's why that argument is required.
>
> Looks like some parts of msft.c still use the old way i.e. hci_req_run_skb() instead of __hci_cmd_sync() after hci_sync related refactoring. I am wondering if it was left like this intentionally? If not, then we probably need to refactor msft.c to use __hci_cmd_sync() for all hci requests. In that case, I can work on refactoring and we can discard this patch altogether. Please let me know.
Yes, if you have time please convert it to use hci_sync.c since we
would like to completely deprecate/remove hci_request.c eventually, if
you think that will take some time we can perhaps merge this changes
first though.
>>
>> > msft_le_monitor_advertisement_cb(hdev, status, hdev->msft_opcode, skb);
>> >
>> > --
>> > 2.36.0.512.ge40c2bad7a-goog
>> >
>>
>>
>> --
>> Luiz Augusto von Dentz
>
> Regards,
> Manish.
--
Luiz Augusto von Dentz