2009-07-27 14:47:50

by Tim Gardner

[permalink] [raw]
Subject: [PATCH] dell-laptop: Fix rfkill state setting

Matthew,

I think the rfkill state change logic is inverted. I've tried the original
code on 3 different Dell models. Once 'rfkill block all' is run, then you
can never unblock 'dell-wifi: Wireless LAN'. With this change you can get
it unblocked, but you need to run 'rfkill unblock all' twice (which is
likely an issue with rfkill).

rtg

>From 778aec563a251418e455d63f711aab1c936bff73 Mon Sep 17 00:00:00 2001
From: Tim Gardner <[email protected]>
Date: Mon, 27 Jul 2009 08:30:54 -0600
Subject: [PATCH] UBUNTU: [Upstream] dell-laptop: Fix rfkill state setting.

rfkill enable/disable transitions are predicated on the state of the
external hardware switch, i.e., if the external switch is in the on position,
then no rfkill state transitions are allowed.

Signed-off-by: Tim Gardner <[email protected]>
---
drivers/platform/x86/dell-laptop.c | 7 +++++--
1 files changed, 5 insertions(+), 2 deletions(-)

diff --git a/drivers/platform/x86/dell-laptop.c b/drivers/platform/x86/dell-laptop.c
index 74909c4..cf40c4e 100644
--- a/drivers/platform/x86/dell-laptop.c
+++ b/drivers/platform/x86/dell-laptop.c
@@ -197,8 +197,11 @@ static void dell_rfkill_query(struct rfkill *rfkill, void *data)
dell_send_request(&buffer, 17, 11);
status = buffer.output[1];

- if (status & BIT(bit))
- rfkill_set_hw_state(rfkill, !!(status & BIT(16)));
+ /*
+ * Don't change state unless the read-only HW rfkill switch is disabled.
+ */
+ if (status & BIT(16))
+ rfkill_set_hw_state(rfkill, !!(status & BIT(bit)));
}

static const struct rfkill_ops dell_rfkill_ops = {
--
1.6.2.4



2009-07-27 15:23:58

by Matthew Garrett

[permalink] [raw]
Subject: Re: [PATCH] dell-laptop: Fix rfkill state setting

On Mon, Jul 27, 2009 at 05:10:18PM +0200, Johannes Berg wrote:

> Hmm. The previous code was
>
> - if (status & (1<<16))
> - new_state = RFKILL_STATE_SOFT_BLOCKED;
> -
> - if (status & (1<<bit))
> - *state = new_state;
> - else
> - *state = RFKILL_STATE_UNBLOCKED;
> -
> - return 0;
>
> where new_state was initialised to RFKILL_STATE_HARD_BLOCKED.
>
> So doesn't that mean that 1<<bit is the hard-block bit? Or was the
> previous code just bogus, but happened to work since rfkill didn't
> separate the hard/soft kill concepts?

The previous code may well have been bogus. The code is basically the
only documentation I have here - bit 16 indicates that the switch is on
(and thus everything else is hard blocked), while bits 17, 18 and 19
indicate blocked wifi, bluetooth and wwan respectively. I'd assumed that
those indicated that they were soft blocked independently of the
hardware block state, but it may be that the hardware doesn't expose the
soft state if the hard switch is blocking the devices.

--
Matthew Garrett | [email protected]

2009-07-27 15:10:58

by Johannes Berg

[permalink] [raw]
Subject: Re: [PATCH] dell-laptop: Fix rfkill state setting

On Mon, 2009-07-27 at 08:47 -0600, Tim Gardner wrote:
> Matthew,
>
> I think the rfkill state change logic is inverted. I've tried the original
> code on 3 different Dell models. Once 'rfkill block all' is run, then you
> can never unblock 'dell-wifi: Wireless LAN'. With this change you can get
> it unblocked, but you need to run 'rfkill unblock all' twice (which is
> likely an issue with rfkill).
>
> rtg
>
> From 778aec563a251418e455d63f711aab1c936bff73 Mon Sep 17 00:00:00 2001
> From: Tim Gardner <[email protected]>
> Date: Mon, 27 Jul 2009 08:30:54 -0600
> Subject: [PATCH] UBUNTU: [Upstream] dell-laptop: Fix rfkill state setting.
>
> rfkill enable/disable transitions are predicated on the state of the
> external hardware switch, i.e., if the external switch is in the on position,
> then no rfkill state transitions are allowed.
>
> Signed-off-by: Tim Gardner <[email protected]>
> ---
> drivers/platform/x86/dell-laptop.c | 7 +++++--
> 1 files changed, 5 insertions(+), 2 deletions(-)
>
> diff --git a/drivers/platform/x86/dell-laptop.c b/drivers/platform/x86/dell-laptop.c
> index 74909c4..cf40c4e 100644
> --- a/drivers/platform/x86/dell-laptop.c
> +++ b/drivers/platform/x86/dell-laptop.c
> @@ -197,8 +197,11 @@ static void dell_rfkill_query(struct rfkill *rfkill, void *data)
> dell_send_request(&buffer, 17, 11);
> status = buffer.output[1];
>
> - if (status & BIT(bit))
> - rfkill_set_hw_state(rfkill, !!(status & BIT(16)));
> + /*
> + * Don't change state unless the read-only HW rfkill switch is disabled.
> + */
> + if (status & BIT(16))
> + rfkill_set_hw_state(rfkill, !!(status & BIT(bit)));

Hmm. The previous code was

- if (status & (1<<16))
- new_state = RFKILL_STATE_SOFT_BLOCKED;
-
- if (status & (1<<bit))
- *state = new_state;
- else
- *state = RFKILL_STATE_UNBLOCKED;
-
- return 0;

where new_state was initialised to RFKILL_STATE_HARD_BLOCKED.

So doesn't that mean that 1<<bit is the hard-block bit? Or was the
previous code just bogus, but happened to work since rfkill didn't
separate the hard/soft kill concepts?

johannes


Attachments:
signature.asc (801.00 B)
This is a digitally signed message part

2009-07-27 15:20:08

by Tim Gardner

[permalink] [raw]
Subject: Re: [PATCH] dell-laptop: Fix rfkill state setting

Johannes Berg wrote:
> On Mon, 2009-07-27 at 08:47 -0600, Tim Gardner wrote:
>> Matthew,
>>
>> I think the rfkill state change logic is inverted. I've tried the original
>> code on 3 different Dell models. Once 'rfkill block all' is run, then you
>> can never unblock 'dell-wifi: Wireless LAN'. With this change you can get
>> it unblocked, but you need to run 'rfkill unblock all' twice (which is
>> likely an issue with rfkill).
>>
>> rtg
>>
>> From 778aec563a251418e455d63f711aab1c936bff73 Mon Sep 17 00:00:00 2001
>> From: Tim Gardner <[email protected]>
>> Date: Mon, 27 Jul 2009 08:30:54 -0600
>> Subject: [PATCH] UBUNTU: [Upstream] dell-laptop: Fix rfkill state setting.
>>
>> rfkill enable/disable transitions are predicated on the state of the
>> external hardware switch, i.e., if the external switch is in the on position,
>> then no rfkill state transitions are allowed.
>>
>> Signed-off-by: Tim Gardner <[email protected]>
>> ---
>> drivers/platform/x86/dell-laptop.c | 7 +++++--
>> 1 files changed, 5 insertions(+), 2 deletions(-)
>>
>> diff --git a/drivers/platform/x86/dell-laptop.c b/drivers/platform/x86/dell-laptop.c
>> index 74909c4..cf40c4e 100644
>> --- a/drivers/platform/x86/dell-laptop.c
>> +++ b/drivers/platform/x86/dell-laptop.c
>> @@ -197,8 +197,11 @@ static void dell_rfkill_query(struct rfkill *rfkill, void *data)
>> dell_send_request(&buffer, 17, 11);
>> status = buffer.output[1];
>>
>> - if (status & BIT(bit))
>> - rfkill_set_hw_state(rfkill, !!(status & BIT(16)));
>> + /*
>> + * Don't change state unless the read-only HW rfkill switch is disabled.
>> + */
>> + if (status & BIT(16))
>> + rfkill_set_hw_state(rfkill, !!(status & BIT(bit)));
>
> Hmm. The previous code was
>
> - if (status & (1<<16))
> - new_state = RFKILL_STATE_SOFT_BLOCKED;
> -
> - if (status & (1<<bit))
> - *state = new_state;
> - else
> - *state = RFKILL_STATE_UNBLOCKED;
> -
> - return 0;
>
> where new_state was initialised to RFKILL_STATE_HARD_BLOCKED.
>
> So doesn't that mean that 1<<bit is the hard-block bit? Or was the
> previous code just bogus, but happened to work since rfkill didn't
> separate the hard/soft kill concepts?
>
> johannes

I'm not as familiar with the semantics of soft/hard block in the
previous incarnation of rfkill, but the above code kind of looks
correct. At least the transmitter states are forced to blocked if the
rfkill switch is enabled.

rtg
--
Tim Gardner [email protected] http://www.tpi.com
OR 503-601-0234 x102 MT 406-443-5357

2009-07-27 14:53:51

by Matthew Garrett

[permalink] [raw]
Subject: Re: [PATCH] dell-laptop: Fix rfkill state setting

On Mon, Jul 27, 2009 at 08:47:47AM -0600, Tim Gardner wrote:
> Matthew,
>
> I think the rfkill state change logic is inverted. I've tried the original
> code on 3 different Dell models. Once 'rfkill block all' is run, then you
> can never unblock 'dell-wifi: Wireless LAN'. With this change you can get
> it unblocked, but you need to run 'rfkill unblock all' twice (which is
> likely an issue with rfkill).

Yes, there's clearly an issue of some sort here. I'll dig out a Dell and
test this, but it looks good.

--
Matthew Garrett | [email protected]