All of lore.kernel.org
 help / color / mirror / Atom feed
* [PATCH v2 0/1] Bluetooth: Suspend improvements
@ 2021-03-02 18:49 Abhishek Pandit-Subedi
  2021-03-02 18:49 ` [PATCH v2 1/1] Bluetooth: Remove unneeded commands for suspend Abhishek Pandit-Subedi
  0 siblings, 1 reply; 5+ messages in thread
From: Abhishek Pandit-Subedi @ 2021-03-02 18:49 UTC (permalink / raw)
  To: marcel
  Cc: chromeos-bluetooth-upstreaming, Hans de Goede, linux-bluetooth,
	Abhishek Pandit-Subedi, David S. Miller, Johan Hedberg, netdev,
	linux-kernel, Jakub Kicinski, Luiz Augusto von Dentz


Hi Marcel (and linux bluetooth),

Here are a few suspend improvements based on user reports we saw on
ChromeOS and feedback from Hans de Goede on the mailing list.

I have tested this using our ChromeOS suspend/resume automated tests
(full SRHealth test coverage and some suspend resume stress tests).

Thanks
Abhishek


Changes in v2:
* Removed hci_dev_lock from hci_cc_set_event_filter since flags are
  set/cleared atomically

Abhishek Pandit-Subedi (1):
  Bluetooth: Remove unneeded commands for suspend

 include/net/bluetooth/hci.h |  1 +
 net/bluetooth/hci_event.c   | 24 ++++++++++++++++++++
 net/bluetooth/hci_request.c | 44 +++++++++++++++++++++++--------------
 3 files changed, 52 insertions(+), 17 deletions(-)

-- 
2.30.1.766.gb4fecdf3b7-goog


^ permalink raw reply	[flat|nested] 5+ messages in thread
* [PATCH 1/2] Bluetooth: Notify suspend on le conn failed
@ 2021-03-01 20:06 Abhishek Pandit-Subedi
  2021-03-01 21:13 ` Bluetooth: Suspend improvements bluez.test.bot
  0 siblings, 1 reply; 5+ messages in thread
From: Abhishek Pandit-Subedi @ 2021-03-01 20:06 UTC (permalink / raw)
  To: marcel
  Cc: chromeos-bluetooth-upstreaming, Hans de Goede, linux-bluetooth,
	Abhishek Pandit-Subedi, Archie Pusaka, David S. Miller,
	Johan Hedberg, netdev, linux-kernel, Jakub Kicinski,
	Luiz Augusto von Dentz

When suspending, Bluetooth disconnects all connected peers devices. If
an LE connection is started but isn't completed, we will see an LE
Create Connection Cancel instead of an HCI disconnect. This just adds
a check to see if an LE cancel was the last disconnected device and wake
the suspend thread when that is the case.

Signed-off-by: Abhishek Pandit-Subedi <abhishekpandit@chromium.org>
Reviewed-by: Archie Pusaka <apusaka@chromium.org>
---
Here is an HCI trace when the issue occurred.

< HCI Command: LE Create Connection (0x08|0x000d) plen 25                                           #187777 [hci0] 2021-02-03 21:42:35.130208
        Scan interval: 60.000 msec (0x0060)
        Scan window: 60.000 msec (0x0060)
        Filter policy: White list is not used (0x00)
        Peer address type: Random (0x01)
        Peer address: D9:DC:6B:61:EB:3A (Static)
        Own address type: Public (0x00)
        Min connection interval: 15.00 msec (0x000c)
        Max connection interval: 30.00 msec (0x0018)
        Connection latency: 20 (0x0014)
        Supervision timeout: 3000 msec (0x012c)
        Min connection length: 0.000 msec (0x0000)
        Max connection length: 0.000 msec (0x0000)
> HCI Event: Command Status (0x0f) plen 4                                                           #187778 [hci0] 2021-02-03 21:42:35.131184
      LE Create Connection (0x08|0x000d) ncmd 1
        Status: Success (0x00)
< HCI Command: LE Create Connection Cancel (0x08|0x000e) plen 0                                     #187805 [hci0] 2021-02-03 21:42:37.183336
> HCI Event: Command Complete (0x0e) plen 4                                                         #187806 [hci0] 2021-02-03 21:42:37.192394
      LE Create Connection Cancel (0x08|0x000e) ncmd 1
        Status: Success (0x00)
> HCI Event: LE Meta Event (0x3e) plen 19                                                           #187807 [hci0] 2021-02-03 21:42:37.193400
      LE Connection Complete (0x01)
        Status: Unknown Connection Identifier (0x02)
        Handle: 0
        Role: Master (0x00)
        Peer address type: Random (0x01)
        Peer address: D9:DC:6B:61:EB:3A (Static)
        Connection interval: 0.00 msec (0x0000)
        Connection latency: 0 (0x0000)
        Supervision timeout: 0 msec (0x0000)
        Master clock accuracy: 0x00
... <skip a few unrelated events>
@ MGMT Event: Controller Suspended (0x002d) plen 1                                                 {0x0002} [hci0] 2021-02-03 21:42:39.178780
        Suspend state: Controller running (failed to suspend) (0)
@ MGMT Event: Controller Suspended (0x002d) plen 1                                                 {0x0001} [hci0] 2021-02-03 21:42:39.178780
        Suspend state: Controller running (failed to suspend) (0)
... <actual suspended time>
< HCI Command: Set Event Filter (0x03|0x0005) plen 1                                                #187808 [hci0] 2021-02-04 09:23:07.313591
        Type: Clear All Filters (0x00)

 net/bluetooth/hci_conn.c | 10 ++++++++++
 1 file changed, 10 insertions(+)

diff --git a/net/bluetooth/hci_conn.c b/net/bluetooth/hci_conn.c
index 6ffa89e3ba0a85..468d31f3303d7a 100644
--- a/net/bluetooth/hci_conn.c
+++ b/net/bluetooth/hci_conn.c
@@ -772,6 +772,16 @@ void hci_le_conn_failed(struct hci_conn *conn, u8 status)
 
 	hci_conn_del(conn);
 
+	/* The suspend notifier is waiting for all devices to disconnect and an
+	 * LE connect cancel will result in an hci_le_conn_failed. Once the last
+	 * connection is deleted, we should also wake the suspend queue to
+	 * complete suspend operations.
+	 */
+	if (list_empty(&hdev->conn_hash.list) &&
+	    test_and_clear_bit(SUSPEND_DISCONNECTING, hdev->suspend_tasks)) {
+		wake_up(&hdev->suspend_wait_q);
+	}
+
 	/* Since we may have temporarily stopped the background scanning in
 	 * favor of connection establishment, we should restart it.
 	 */
-- 
2.30.1.766.gb4fecdf3b7-goog


^ permalink raw reply related	[flat|nested] 5+ messages in thread

end of thread, other threads:[~2021-03-03 19:07 UTC | newest]

Thread overview: 5+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2021-03-02 18:49 [PATCH v2 0/1] Bluetooth: Suspend improvements Abhishek Pandit-Subedi
2021-03-02 18:49 ` [PATCH v2 1/1] Bluetooth: Remove unneeded commands for suspend Abhishek Pandit-Subedi
2021-03-03  0:11   ` Bluetooth: Suspend improvements bluez.test.bot
2021-03-03 15:47   ` [PATCH v2 1/1] Bluetooth: Remove unneeded commands for suspend Marcel Holtmann
  -- strict thread matches above, loose matches on Subject: below --
2021-03-01 20:06 [PATCH 1/2] Bluetooth: Notify suspend on le conn failed Abhishek Pandit-Subedi
2021-03-01 21:13 ` Bluetooth: Suspend improvements bluez.test.bot

This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.