[PATCH] libata: pata_bf54x: handle portmuxing of pins through GPIO PORTs

[PATCH] libata: pata_bf54x: handle portmuxing of pins through GPIO PORTs

[Mike Frysinger]

From: Sonic Zhang <sonic.zhang@analog.com>

By default, the PATA pins are routed to the async address lines in which
case, no peripheral muxing needs to be done.  However, if the pins get
routed through the GPIO PORTs pins, we need to make sure to request them
so that the muxing is properly set up.

Signed-off-by: Sonic Zhang <sonic.zhang@analog.com>
Signed-off-by: Bryan Wu <cooloney@kernel.org>
Signed-off-by: Mike Frysinger <vapier@gentoo.org>
---
Andrew: could you pick this up ?  it was posted over a month ago to the
Linux/IDE list w/no feedback.

 drivers/ata/pata_bf54x.c |   19 +++++++++++++++++++
 1 files changed, 19 insertions(+), 0 deletions(-)

diff --git a/drivers/ata/pata_bf54x.c b/drivers/ata/pata_bf54x.c
index c4b47a3..02c81f1 100644
--- a/drivers/ata/pata_bf54x.c
+++ b/drivers/ata/pata_bf54x.c
@@ -1557,6 +1557,25 @@ static unsigned short atapi_io_port[] = {
 	P_ATAPI_DMARQ,
 	P_ATAPI_INTRQ,
 	P_ATAPI_IORDY,
+	P_ATAPI_D0A,
+	P_ATAPI_D1A,
+	P_ATAPI_D2A,
+	P_ATAPI_D3A,
+	P_ATAPI_D4A,
+	P_ATAPI_D5A,
+	P_ATAPI_D6A,
+	P_ATAPI_D7A,
+	P_ATAPI_D8A,
+	P_ATAPI_D9A,
+	P_ATAPI_D10A,
+	P_ATAPI_D11A,
+	P_ATAPI_D12A,
+	P_ATAPI_D13A,
+	P_ATAPI_D14A,
+	P_ATAPI_D15A,
+	P_ATAPI_A0A,
+	P_ATAPI_A1A,
+	P_ATAPI_A2A,
 	0
 };
 
-- 
1.6.3.3

Re: [PATCH] libata: pata_bf54x: handle portmuxing of pins through GPIO PORTs

[Jeff Garzik]

Mike Frysinger wrote:
> From: Sonic Zhang <sonic.zhang@analog.com>
> 
> By default, the PATA pins are routed to the async address lines in which
> case, no peripheral muxing needs to be done.  However, if the pins get
> routed through the GPIO PORTs pins, we need to make sure to request them
> so that the muxing is properly set up.
> 
> Signed-off-by: Sonic Zhang <sonic.zhang@analog.com>
> Signed-off-by: Bryan Wu <cooloney@kernel.org>
> Signed-off-by: Mike Frysinger <vapier@gentoo.org>
> ---
> Andrew: could you pick this up ?  it was posted over a month ago to the
> Linux/IDE list w/no feedback.

Although it is disappointing that your patch was missed -- our apologies 
-- I don't think the best response was to elide linux-ide and the libata 
maintainer, in favor of a higher noise list.  :)

I have the patch queued, now.

	Jeff

Re: [PATCH] libata: pata_bf54x: handle portmuxing of pins through GPIO PORTs

[Mike Frysinger]

On Tue, Jul 14, 2009 at 15:07, Jeff Garzik wrote:
> Mike Frysinger wrote:
>> From: Sonic Zhang <sonic.zhang@analog.com>
>>
>> By default, the PATA pins are routed to the async address lines in which
>> case, no peripheral muxing needs to be done.  However, if the pins get
>> routed through the GPIO PORTs pins, we need to make sure to request them
>> so that the muxing is properly set up.
>>
>> Signed-off-by: Sonic Zhang <sonic.zhang@analog.com>
>> Signed-off-by: Bryan Wu <cooloney@kernel.org>
>> Signed-off-by: Mike Frysinger <vapier@gentoo.org>
>> ---
>> Andrew: could you pick this up ?  it was posted over a month ago to the
>> Linux/IDE list w/no feedback.
>
> Although it is disappointing that your patch was missed -- our apologies --
> I don't think the best response was to elide linux-ide and the libata
> maintainer, in favor of a higher noise list.  :)

the assumption was that linux-ide was dead rather than missed as is
the cases with some sections and Andrew can pick up simple things.
i'll make sure to send to you in the future as well.

> I have the patch queued, now.

thanks
-miek

Re: [PATCH] libata: pata_bf54x: handle portmuxing of pins through GPIO PORTs

[Mike Frysinger]

On Tue, Jul 14, 2009 at 14:07, Jeff Garzik wrote:
> Mike Frysinger wrote:
>> By default, the PATA pins are routed to the async address lines in which
>> case, no peripheral muxing needs to be done.  However, if the pins get
>> routed through the GPIO PORTs pins, we need to make sure to request them
>> so that the muxing is properly set up.
>> ---
>> Andrew: could you pick this up ?  it was posted over a month ago to the
>> Linux/IDE list w/no feedback.
>
> Although it is disappointing that your patch was missed -- our apologies --
> I don't think the best response was to elide linux-ide and the libata
> maintainer, in favor of a higher noise list.  :)
>
> I have the patch queued, now.

so it's been 8 months since it was posted, 6 months since it was
queued, and multiple libata merges have been made, but this patch has
yet to make it into mainline ...
-mike

Re: [PATCH] libata: pata_bf54x: handle portmuxing of pins through GPIO PORTs

[Alan Cox]

> so it's been 8 months since it was posted, 6 months since it was
> queued, and multiple libata merges have been made, but this patch has
> yet to make it into mainline ...

In those situations Cc Linus with an included rant (points for the best
rant) and the relevant subject lines to show when it was posted. It
happens - confusion about which tree is being merged, generally only
being human etc...

For an added bonus make sure the rant includes something highly quotable
and funny as a put down, then you'll make lwn.net as well ;)

Alan

Re: [PATCH] libata: pata_bf54x: handle portmuxing of pins through GPIO PORTs

[Jeff Garzik]

On 12/21/2009 09:21 AM, Mike Frysinger wrote:
> On Tue, Jul 14, 2009 at 14:07, Jeff Garzik wrote:
>> Mike Frysinger wrote:
>>> By default, the PATA pins are routed to the async address lines in which
>>> case, no peripheral muxing needs to be done.  However, if the pins get
>>> routed through the GPIO PORTs pins, we need to make sure to request them
>>> so that the muxing is properly set up.
>>> ---
>>> Andrew: could you pick this up ?  it was posted over a month ago to the
>>> Linux/IDE list w/no feedback.
>>
>> Although it is disappointing that your patch was missed -- our apologies --
>> I don't think the best response was to elide linux-ide and the libata
>> maintainer, in favor of a higher noise list.  :)
>>
>> I have the patch queued, now.
>
> so it's been 8 months since it was posted, 6 months since it was
> queued, and multiple libata merges have been made, but this patch has
> yet to make it into mainline ...

No idea how it got lost :(   Just sent a pull request to Linus with this 
in it.

	Jeff

Re: [PATCH] libata: pata_bf54x: handle portmuxing of pins through GPIO PORTs

[Jeff Garzik]

On 07/14/2009 01:39 PM, Mike Frysinger wrote:
> From: Sonic Zhang<sonic.zhang@analog.com>
>
> By default, the PATA pins are routed to the async address lines in which
> case, no peripheral muxing needs to be done.  However, if the pins get
> routed through the GPIO PORTs pins, we need to make sure to request them
> so that the muxing is properly set up.
>
> Signed-off-by: Sonic Zhang<sonic.zhang@analog.com>
> Signed-off-by: Bryan Wu<cooloney@kernel.org>
> Signed-off-by: Mike Frysinger<vapier@gentoo.org>
> ---
> Andrew: could you pick this up ?  it was posted over a month ago to the
> Linux/IDE list w/no feedback.
>
>   drivers/ata/pata_bf54x.c |   19 +++++++++++++++++++
>   1 files changed, 19 insertions(+), 0 deletions(-)


Added to the tip of libata-dev...

Linus, if you pull

master.kernel.org:/pub/scm/linux/kernel/git/jgarzik/libata-dev.git 
upstream-linus

then you will receive the previous pull (two pata_cmd64x fixes), plus 
the above fix to pata_bf54x.