Bug 5959

Summary:	bluetooth CF card is broken by suspend/resume
Product:	Networking	Reporter:	Pavel Machek (pavel)
Component:	Other	Assignee:	Rafael J. Wysocki (rjwysocki)
Status:	CLOSED CODE_FIX
Severity:	normal
Priority:	P2
Hardware:	i386
OS:	Linux
Kernel Version:	2.6.9-2.6.16-rc1	Subsystem:
Regression:	---	Bisected commit-id:

Description Pavel Machek 2006-01-25 11:52:07 UTC

Distribution: Debian
Hardware Environment: i386

Steps to reproduce: Suspend to RAM (and I think suspend-to-disk too, will
verify) breaks bluetooth running over CF card.

Problem probably is because CF card is initialized from userland (by
hciattach?), and userland has no idea that we have just suspended. After resume,
I have endless stream of errors in syslog.

Jan 25 17:52:26 amd kernel: bcsp_pkt_cull: Peer acked invalid packet
Jan 25 17:52:26 amd kernel: bcsp_handle_le_pkt: Found a LE sync pkt, card has reset
Jan 25 17:52:26 amd kernel: bcsp_pkt_cull: Peer acked invalid packet
Jan 25 17:52:26 amd kernel: bcsp_handle_le_pkt: Found a LE sync pkt, card has reset
Jan 25 17:52:26 amd kernel: bcsp_pkt_cull: Peer acked invalid packet
Jan 25 17:52:26 amd kernel: bcsp_handle_le_pkt: Found a LE sync pkt, card has reset
Jan 25 17:52:26 amd kernel: bcsp_pkt_cull: Peer acked invalid packet
Jan 25 17:52:26 amd kernel: bcsp_handle_le_pkt: Found a LE sync pkt, card has reset

...perhaps error handling in bluetooth-over-serial needs some improvements...

Comment 1 Rafael J. Wysocki 2006-09-29 07:14:55 UTC

Has it been fixed after 2.6.16-rc1?

Comment 2 Rafael J. Wysocki 2006-10-25 10:10:01 UTC

Due to the lack of response, I assume that the problem has been fixed.  Please
reopen if that's not the case.