[U-Boot] [PATCH 1/1] SMC911x driver fixed for NFS boot

[U-Boot] [PATCH 1/1] SMC911x driver fixed for NFS boot

[Josh Karabin]

Manikandan Pillai wrote:
> eth_halt() function in the smc911x drivers used to call the
> smc911x_reset() function. eth_halt() used to be called after
> tftp transfers. This used to put the ethernet chip in reset
> while the linux boots up resulting in the ethernet driver
> not coming up. NFS boot used to fail as a result.
> 
> This patch calls smc911x_shutdown() instead of smc911x_reset().
> Some comments received has also been fixed.

This patch fixes NFS boot in the case where the kernel is first loaded
with tftpboot, as described.

If the kernel is loaded using some other mechanism, like flash, the smc
device is never initialized.  At least in the case where there is no
valid EEPROM connected to the smc device, NFS boot will fail.  I'm not
sure whether or not a loaded MAC address has anything to do with it, or
if the kernel is expecting the smc device to be in a different state.

Relying on tftpboot leaving the ethernet device in the right state
doesn't seem right.  Is there an existing clean way to deal with this,
short of adding an initramfs to fix it up?  I'll work on a patch if
someone can suggest an approach that makes sense.


Regards,

- Josh

[U-Boot] [PATCH 1/1] SMC911x driver fixed for NFS boot

[Wolfgang Denk]

Dear Josh Karabin,

In message <4A12CD18.2040608@vocollect.com> you wrote:
> 
> If the kernel is loaded using some other mechanism, like flash, the smc
> device is never initialized.  At least in the case where there is no
> valid EEPROM connected to the smc device, NFS boot will fail.  I'm not
> sure whether or not a loaded MAC address has anything to do with it, or
> if the kernel is expecting the smc device to be in a different state.
> 
> Relying on tftpboot leaving the ethernet device in the right state
> doesn't seem right.  Is there an existing clean way to deal with this,
> short of adding an initramfs to fix it up?  I'll work on a patch if
> someone can suggest an approach that makes sense.

Please check the FAQ, especially here:
http://www.denx.de/wiki/view/DULG/EthernetDoesNotWorkInLinux

Best regards,

Wolfgang Denk

-- 
DENX Software Engineering GmbH,     MD: Wolfgang Denk & Detlev Zundel
HRB 165235 Munich, Office: Kirchenstr.5, D-82194 Groebenzell, Germany
Phone: (+49)-8142-66989-10 Fax: (+49)-8142-66989-80 Email: wd at denx.de
Another war ... must it always be so?  How many comrades have we lost
in this way? ...  Obedience.  Duty.  Death, and more death ...
	-- Romulan Commander, "Balance of Terror", stardate 1709.2

[U-Boot] [PATCH 1/1] SMC911x driver fixed for NFS boot

[Manikandan Pillai]

eth_halt() function in the smc911x drivers used to call the
smc911x_reset() function. eth_halt() used to be called after
tftp transfers. This used to put the ethernet chip in reset
while the linux boots up resulting in the ethernet driver
not coming up. NFS boot used to fail as a result.

This patch calls smc911x_shutdown() instead of smc911x_reset().
Some comments received has also been fixed.

Signed-off-by: Manikandan Pillai <mani.pillai@ti.com>
---
 drivers/net/smc911x.c |   23 ++++++++++++++++++++++-
 1 files changed, 22 insertions(+), 1 deletions(-)

diff --git a/drivers/net/smc911x.c b/drivers/net/smc911x.c
index 1ded8f0..5bc3914 100644
--- a/drivers/net/smc911x.c
+++ b/drivers/net/smc911x.c
@@ -116,6 +116,27 @@ static int smc911x_phy_reset(void)
 	return 0;
 }
 
+static void smc911x_shutdown(void)
+{
+	unsigned int cr;
+
+	/* Turn of Rx and TX */
+	cr = smc911x_get_mac_csr(MAC_CR);
+	cr &= ~(MAC_CR_TXEN | MAC_CR_RXEN | MAC_CR_HBDIS);
+	smc911x_set_mac_csr(MAC_CR, cr);
+
+	/* Stop Transmission */
+	cr = smc911x_get_mac_csr(TX_CFG);
+	cr &= ~(TX_CFG_STOP_TX);
+	smc911x_set_mac_csr(TX_CFG, cr);
+	/* Stop receiving packets */
+	cr = smc911x_get_mac_csr(RX_CFG);
+	cr &= ~(RX_CFG_RXDOFF);
+	smc911x_set_mac_csr(RX_CFG, cr);
+
+}
+
+
 static void smc911x_phy_configure(void)
 {
 	int timeout;
@@ -224,7 +245,7 @@ int eth_send(volatile void *packet, int length)
 
 void eth_halt(void)
 {
-	smc911x_reset();
+	smc911x_shutdown();
 }
 
 int eth_rx(void)
-- 
1.6.0.3

[U-Boot] [PATCH 1/1] SMC911x driver fixed for NFS boot

[Wolfgang Denk]

Dear Ben,

In message <1239162275-13087-1-git-send-email-mani.pillai@ti.com> Manikandan Pillai wrote:
> eth_halt() function in the smc911x drivers used to call the
> smc911x_reset() function. eth_halt() used to be called after
> tftp transfers. This used to put the ethernet chip in reset
> while the linux boots up resulting in the ethernet driver
> not coming up. NFS boot used to fail as a result.
> 
> This patch calls smc911x_shutdown() instead of smc911x_reset().
> Some comments received has also been fixed.
> 
> Signed-off-by: Manikandan Pillai <mani.pillai@ti.com>
> ---
>  drivers/net/smc911x.c |   23 ++++++++++++++++++++++-
>  1 files changed, 22 insertions(+), 1 deletions(-)

Any comments on this patch?

Best regards,

Wolfgang Denk

-- 
DENX Software Engineering GmbH,     MD: Wolfgang Denk & Detlev Zundel
HRB 165235 Munich, Office: Kirchenstr.5, D-82194 Groebenzell, Germany
Phone: (+49)-8142-66989-10 Fax: (+49)-8142-66989-80 Email: wd at denx.de
"I haven't lost my mind - it's backed up on tape somewhere."

[U-Boot] [PATCH 1/1] SMC911x driver fixed for NFS boot

[Ben Warren]

Wolfgang Denk wrote:
> Dear Ben,
>
> In message <1239162275-13087-1-git-send-email-mani.pillai@ti.com> Manikandan Pillai wrote:
>   
>> eth_halt() function in the smc911x drivers used to call the
>> smc911x_reset() function. eth_halt() used to be called after
>> tftp transfers. This used to put the ethernet chip in reset
>> while the linux boots up resulting in the ethernet driver
>> not coming up. NFS boot used to fail as a result.
>>
>> This patch calls smc911x_shutdown() instead of smc911x_reset().
>> Some comments received has also been fixed.
>>
>> Signed-off-by: Manikandan Pillai <mani.pillai@ti.com>
>> ---
>>  drivers/net/smc911x.c |   23 ++++++++++++++++++++++-
>>  1 files changed, 22 insertions(+), 1 deletions(-)
>>     
>
> Any comments on this patch?
>
>   
It's on my list of patches to pull into next.

regards,
Ben

[U-Boot] [PATCH 1/1] SMC911x driver fixed for NFS boot

[Wolfgang Denk]

Dear Manikandan Pillai,

In message <1239162275-13087-1-git-send-email-mani.pillai@ti.com> you wrote:
> eth_halt() function in the smc911x drivers used to call the
> smc911x_reset() function. eth_halt() used to be called after
> tftp transfers. This used to put the ethernet chip in reset
> while the linux boots up resulting in the ethernet driver
> not coming up. NFS boot used to fail as a result.
> 
> This patch calls smc911x_shutdown() instead of smc911x_reset().
> Some comments received has also been fixed.

This patch (commit ID ca9c8a1e in the master branch) causes compile
warnings on all systems that use this driver, for example:

+ ./MAKEALL imx31_litekit
Configuring for imx31_litekit board...
smc911x.c: In function 'smc911x_shutdown':
smc911x.c:130: warning: large integer implicitly truncated to unsigned type
smc911x.c:132: warning: large integer implicitly truncated to unsigned type
smc911x.c:134: warning: large integer implicitly truncated to unsigned type
smc911x.c:136: warning: large integer implicitly truncated to unsigned type
   text    data     bss     dec     hex filename
 114556    5228   24584  144368   233f0 /work/wd/tmp-arm/u-boot


The problem seems to come from this part of the code:

> +static void smc911x_shutdown(void)
> +{
> +	unsigned int cr;
> +
> +	/* Turn of Rx and TX */
> +	cr = smc911x_get_mac_csr(MAC_CR);
> +	cr &= ~(MAC_CR_TXEN | MAC_CR_RXEN | MAC_CR_HBDIS);
> +	smc911x_set_mac_csr(MAC_CR, cr);
> +
> +	/* Stop Transmission */
> +	cr = smc911x_get_mac_csr(TX_CFG);
> +	cr &= ~(TX_CFG_STOP_TX);
> +	smc911x_set_mac_csr(TX_CFG, cr);
> +	/* Stop receiving packets */
> +	cr = smc911x_get_mac_csr(RX_CFG);
> +	cr &= ~(RX_CFG_RXDOFF);
> +	smc911x_set_mac_csr(RX_CFG, cr);

wher eyou ar epssing long constants (TX_CFG, RX_CFG) into
smc911x_set_mac_csr(), but smc911x_set_mac_csr() is declared
in "drivers/net/smc911x.h" to take an "unsigned character" argument
only:

	static u32 smc911x_get_mac_csr(u8 reg)

Please check what's wrong.

Please either provide a fix, or I will have to consider to revert this
patch for the upcoming release.


Thanks.

Best regards,

Wolfgang Denk

-- 
DENX Software Engineering GmbH,     MD: Wolfgang Denk & Detlev Zundel
HRB 165235 Munich, Office: Kirchenstr.5, D-82194 Groebenzell, Germany
Phone: (+49)-8142-66989-10 Fax: (+49)-8142-66989-80 Email: wd at denx.de
"He was so narrow minded he could see through  a  keyhole  with  both
eyes ..."

[U-Boot] [PATCH 1/1] SMC911x driver fixed for NFS boot

[Wolfgang Denk]

Dear Manikandan Pillai,

In message <20090609220854.A1377832E416@gemini.denx.de> I wrote:
> 
...
> This patch (commit ID ca9c8a1e in the master branch) causes compile
> warnings on all systems that use this driver, for example:
...
> Please either provide a fix, or I will have to consider to revert this
> patch for the upcoming release.

Please respond urgently. Otherwise I will have to revert this patch.

Best regards,

Wolfgang Denk

-- 
DENX Software Engineering GmbH,     MD: Wolfgang Denk & Detlev Zundel
HRB 165235 Munich, Office: Kirchenstr.5, D-82194 Groebenzell, Germany
Phone: (+49)-8142-66989-10 Fax: (+49)-8142-66989-80 Email: wd at denx.de
Anything that is worth doing at all is worth doing well.
                                       -- Philip Earl of Chesterfield

[U-Boot] [PATCH 1/1] SMC911x driver fixed for NFS boot

[Jean-Christophe PLAGNIOL-VILLARD]

On 23:04 Fri 12 Jun     , Wolfgang Denk wrote:
> Dear Manikandan Pillai,
> 
> In message <20090609220854.A1377832E416@gemini.denx.de> I wrote:
> > 
> ...
> > This patch (commit ID ca9c8a1e in the master branch) causes compile
> > warnings on all systems that use this driver, for example:
> ...
> > Please either provide a fix, or I will have to consider to revert this
> > patch for the upcoming release.
> 
> Please respond urgently. Otherwise I will have to revert this patch.
the following fix this and a function implementation location

as I've no board with me currently with a smc911x I can not test it

but it's normaly work

Best Regards,
J.