Re: Bluetooth broken for some people with 6.8.2 [Was: [PATCH 6.8 308/715] Bluetooth: hci_core: Cancel request on command timeout]

[Greg KH]

On Sat, Mar 30, 2024 at 03:59:22PM +0100, Linux regression tracking (Thorsten Leemhuis) wrote:
> On 24.03.24 23:28, Sasha Levin wrote:
> > From: Luiz Augusto von Dentz <luiz.von.dentz@xxxxxxxxx>
> > 
> > [ Upstream commit 63298d6e752fc0ec7f5093860af8bc9f047b30c8 ]
> > 
> > If command has timed out call __hci_cmd_sync_cancel to notify the
> > hci_req since it will inevitably cause a timeout.
> > 
> > This also rework the code around __hci_cmd_sync_cancel since it was
> > wrongly assuming it needs to cancel timer as well, but sometimes the
> > timers have not been started or in fact they already had timed out in
> > which case they don't need to be cancel yet again.
> > 
> > Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@xxxxxxxxx>
> > Stable-dep-of: 2615fd9a7c25 ("Bluetooth: hci_sync: Fix overwriting request callback")
> > Signed-off-by: Sasha Levin <sashal@xxxxxxxxxx>
> 
> Hey stable team, I wonder if it might be wise to pick up 1c3366abdbe884
> ("Bluetooth: hci_sync: Fix not checking error on
> hci_cmd_sync_cancel_sync") from next
> (https://git.kernel.org/pub/scm/linux/kernel/git/next/linux-next.git/commit/?id=1c3366abdbe884)
> for the next releases of all series that a few days ago received
> 63298d6e752fc0 ("Bluetooth: hci_core: Cancel request on command timeout").
> 
> The latter patch sadly on quite a few systems causes a Oops due to a
> NULL pointer dereference and breaks Bluetooth. This was reported for
> mainline here (yes, coincidentally it was reported by yours truly):
> https://lore.kernel.org/all/08275279-7462-4f4a-a0ee-8aa015f829bc@xxxxxxxxxxxxx/
> 
> Now that the patch landed in 6.8.2 it seems to happen there as well
> (guess in 6.7 and others, too), as can be seen from this bug report
> where multiple people already joined:
> https://bugzilla.kernel.org/show_bug.cgi?id=218651
> 
> The fix mentioned above is on the way to Linus, but due to unlucky
> timing missed this weeks network pull, hence will likely only reach
> mainline next Thursday. But the fix afaics has a stable commit id, so
> might be worth picking up soon for the stable releases to fix the
> regression quickly.

Now queued up, thanks for letting us know.

greg k-h