From: Marcel Holtmann <marcel@holtmann.org>
To: Andre Guedes <andre.guedes@openbossa.org>
Cc: linux-bluetooth@vger.kernel.org
Subject: Re: [PATCH v4 07/14] Bluetooth: Handle race condition in Discovery
Date: Tue, 20 Sep 2011 14:37:13 +0200 [thread overview]
Message-ID: <1316522234.1937.81.camel@aeonflux> (raw)
In-Reply-To: <1316468136-12472-8-git-send-email-andre.guedes@openbossa.org>
Hi Andre,
> If MGMT_OP_STOP_DISCOVERY command is issued before the kernel
> receives the HCI Inquiry Command Status Event from the controller
> then that command will fail and the discovery procedure won't be
> stopped. This situation may occur if a MGMT_OP_STOP_DISCOVERY
> command is issued just after a MGMT_OP_START_DISCOVERY.
>
> This race condition can be handled by checking for pending
> MGMT_OP_STOP_DISCOVERY command in inquiry command status event
> handler. If we have a pending MGMT_OP_STOP_DISCOVERY command we
> cancel the ongoing discovery.
>
> Signed-off-by: Andre Guedes <andre.guedes@openbossa.org>
> ---
> include/net/bluetooth/hci_core.h | 1 +
> net/bluetooth/hci_event.c | 3 +++
> net/bluetooth/mgmt.c | 14 +++++++++++++-
> 3 files changed, 17 insertions(+), 1 deletions(-)
>
> diff --git a/include/net/bluetooth/hci_core.h b/include/net/bluetooth/hci_core.h
> index 4a79a50..36e15cc 100644
> --- a/include/net/bluetooth/hci_core.h
> +++ b/include/net/bluetooth/hci_core.h
> @@ -876,6 +876,7 @@ int mgmt_discovering(u16 index, u8 discovering);
> int mgmt_device_blocked(u16 index, bdaddr_t *bdaddr);
> int mgmt_device_unblocked(u16 index, bdaddr_t *bdaddr);
> int mgmt_discovery_complete(u16 index, u8 status);
> +int mgmt_has_pending_stop_discov(u16 index);
>
> /* HCI info for socket */
> #define hci_pi(sk) ((struct hci_pinfo *) sk)
> diff --git a/net/bluetooth/hci_event.c b/net/bluetooth/hci_event.c
> index b44f362..c9d641b 100644
> --- a/net/bluetooth/hci_event.c
> +++ b/net/bluetooth/hci_event.c
> @@ -962,6 +962,9 @@ static inline void hci_cs_inquiry(struct hci_dev *hdev, __u8 status)
> set_bit(HCI_INQUIRY, &hdev->flags);
>
> mgmt_discovering(hdev->id, 1);
> +
> + if (mgmt_has_pending_stop_discov(hdev->id))
> + hci_cancel_inquiry(hdev);
> }
>
> static inline void hci_cs_create_conn(struct hci_dev *hdev, __u8 status)
> diff --git a/net/bluetooth/mgmt.c b/net/bluetooth/mgmt.c
> index d84f242..58cf33a 100644
> --- a/net/bluetooth/mgmt.c
> +++ b/net/bluetooth/mgmt.c
> @@ -1674,7 +1674,11 @@ static int stop_discovery(struct sock *sk, u16 index)
> goto failed;
> }
>
> - err = hci_cancel_inquiry(hdev);
> + if (test_bit(HCI_INQUIRY, &hdev->flags))
> + err = hci_cancel_inquiry(hdev);
> + else
> + err = 0;
> +
do we really just wanna always return success here? Even if
stop_discovery is called for a none existing discovery.
And btw. since you just changed the hci_cancel_inquiry() to return
-EPERM if the HCI_INQUIRY flag is not set you could do this simpler by
just checking the return value directly. No double check of the
HCI_INQUIRY flag.
Regards
Marcel
next prev parent reply other threads:[~2011-09-20 12:37 UTC|newest]
Thread overview: 38+ messages / expand[flat|nested] mbox.gz Atom feed top
2011-09-19 21:35 [PATCH v4 00/14] Discovery support Andre Guedes
2011-09-19 21:35 ` [PATCH v4 01/14] Bluetooth: Periodic Inquiry and mgmt discovering event Andre Guedes
2011-09-20 12:23 ` Marcel Holtmann
2011-09-23 19:12 ` Andre Guedes
2012-03-19 12:40 ` Andre Guedes
2011-09-19 21:35 ` [PATCH v4 02/14] Bluetooth: Add mgmt_discovery_complete() Andre Guedes
2011-09-20 12:29 ` Marcel Holtmann
2011-09-23 19:13 ` Andre Guedes
2011-09-19 21:35 ` [PATCH v4 03/14] Bluetooth: Check pending command in start_discovery() Andre Guedes
2011-09-20 12:26 ` Marcel Holtmann
2011-09-23 19:13 ` Andre Guedes
2011-09-19 21:35 ` [PATCH v4 04/14] Bluetooth: Check pending commands in stop_discovery() Andre Guedes
2011-09-19 21:35 ` [PATCH v4 05/14] Bluetooth: Create hci_do_inquiry() Andre Guedes
2011-09-20 12:31 ` Marcel Holtmann
2011-09-23 19:13 ` Andre Guedes
2011-09-19 21:35 ` [PATCH v4 06/14] Bluetooth: Create hci_cancel_inquiry() Andre Guedes
2011-09-20 12:33 ` Marcel Holtmann
2011-09-19 21:35 ` [PATCH v4 07/14] Bluetooth: Handle race condition in Discovery Andre Guedes
2011-09-20 12:37 ` Marcel Holtmann [this message]
2011-09-23 19:13 ` Andre Guedes
2011-09-19 21:35 ` [PATCH v4 08/14] Bluetooth: Prepare for full support discovery procedures Andre Guedes
2011-09-20 12:43 ` Marcel Holtmann
2011-09-23 19:14 ` Andre Guedes
2011-09-19 21:35 ` [PATCH v4 09/14] Bluetooth: Send mgmt_discovering events Andre Guedes
2011-09-20 12:45 ` Marcel Holtmann
2011-09-23 19:15 ` Andre Guedes
2011-09-19 21:35 ` [PATCH v4 10/14] Bluetooth: Add 'eir_len' param to mgmt_device_found() Andre Guedes
2011-09-20 12:47 ` Marcel Holtmann
2011-09-19 21:35 ` [PATCH v4 11/14] Bluetooth: Report LE devices Andre Guedes
2011-09-20 12:49 ` Marcel Holtmann
2011-09-23 19:15 ` Andre Guedes
2011-09-19 21:35 ` [PATCH v4 12/14] Bluetooth: LE scan infra-structure Andre Guedes
2011-09-20 12:53 ` Marcel Holtmann
2011-09-23 19:15 ` Andre Guedes
2011-09-19 21:35 ` [PATCH v4 13/14] Bluetooth: Support LE-Only discovery procedure Andre Guedes
2011-09-20 13:00 ` Marcel Holtmann
2011-09-23 19:16 ` Andre Guedes
2011-09-19 21:35 ` [PATCH v4 14/14] Bluetooth: Support BR/EDR/LE " Andre Guedes
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=1316522234.1937.81.camel@aeonflux \
--to=marcel@holtmann.org \
--cc=andre.guedes@openbossa.org \
--cc=linux-bluetooth@vger.kernel.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox