Mistral AM37x (with AM3715) devel board networking broken

Mistral AM37x (with AM3715) devel board networking broken

[Rick Bronson]

Hi,

  Should I be able to use the latest omap-linux-2.6 on this board?  I
tried:

git clone git://git.kernel.org/pub/scm/linux/kernel/git/tmlind/linux-omap-2.6.git
cd linux-omap-2.6; make omap2plus_defconfig ARCH=arm; make

  I get some warnings when it boots but it fails to setup the
networking (smsc911x) correctly.  The boot log is here
http://www.efn.org/~rick/pub/boot1.log

  Just wondered if anyone has any ideas?

  Thanks much for any help.

  Rick
                                                                   _
                                                                  | |
                                                                  / /__
.----------------------------------------------------------._____/ (___)
| Rick Bronson    rick at efn dot org    Tel 541-485-7264  |       (___)
| Amazonia Computing http://www.efn.org/~rick      __ o    |_____  (___)
| 5050 Donald Street       Embedded Linux         _`\<,    |     \_(___)
| Eugene, OR  97405     Software Engineering   __( )/( )__ |
`----------------------------------------------------------'

RE: Mistral AM37x (with AM3715) devel board networking broken

[Premi, Sanjeev]

> From: linux-omap-owner@vger.kernel.org [linux-omap-owner@vger.kernel.org] On Behalf Of Rick Bronson [rick@efn.org]
> Sent: Saturday, November 06, 2010 8:50 PM
> To: linux-omap@vger.kernel.org
> Cc: srinath@mistralsolutions.com
> Subject: Mistral AM37x (with AM3715) devel board networking broken
> 
> Hi,
> 
>   Should I be able to use the latest omap-linux-2.6 on this board?  I
> tried:
> 
> git clone git://git.kernel.org/pub/scm/linux/kernel/git/tmlind/linux-omap-2.6.git
> cd linux-omap-2.6; make omap2plus_defconfig ARCH=arm; make
> 
>   I get some warnings when it boots but it fails to setup the
> networking (smsc911x) correctly.  The boot log is here
> http://www.efn.org/~rick/pub/boot1.log
> 
>   Just wondered if anyone has any ideas?
> 
[sp] On first looks I would suspect the u-boot. 2009.11 seems to be quite old.
     Can you try using version 2010.09?

~sanjeev
> 
>   Thanks much for any help.
> 
>   Rick
>                                                                    _
>                                                                   | |
>                                                                   / /__
> .----------------------------------------------------------._____/ (___)
> | Rick Bronson    rick at efn dot org    Tel 541-485-7264  |       (___)
> | Amazonia Computing http://www.efn.org/~rick      __ o    |_____  (___)
> | 5050 Donald Street       Embedded Linux         _`\<,    |     \_(___)
> | Eugene, OR  97405     Software Engineering   __( )/( )__ |
> `----------------------------------------------------------'
> --
> To unsubscribe from this list: send the line "unsubscribe linux-omap" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html

RE: Mistral AM37x (with AM3715) devel board networking broken

[Rick Bronson]

sanjeev,

  Thanks much for the help.

> [sp] On first looks I would suspect the u-boot. 2009.11 seems to be quite old.
>      Can you try using version 2010.09?

  I had tried 2010.09 on this board originally but found that it
wasn't stable, it was almost like the SDRAM wasn't set up correctly.
I'd get random panic's on bad pointers in Linux.  But I tried it
anyway on this new kernel and it does set up the networking about 1 in
5 boots.  Here is a boot where it set up the network
http://www.efn.org/~rick/pub/boot4.log, it still has the warnings though.

  Please check my setup of u-boot, here is what I did:

wget ftp://ftp.denx.de/pub/u-boot/u-boot-2010.09.tar.bz2
tar jxf u-boot-2010.09.tar.bz2
cd u-boot-2010.09; make omap3_evm_config; make

  Is this the correct config (omap3_evm_config) to use for this board?
Maybe you could tell me what exact u-boot and kernel you'd recommend
and the steps to get them for this board.

  Thanks,

  Rick

RE: Mistral AM37x (with AM3715) devel board networking broken

[Premi, Sanjeev]

> -----Original Message-----
> From: linux-omap-owner@vger.kernel.org 
> [mailto:linux-omap-owner@vger.kernel.org] On Behalf Of Rick Bronson
> Sent: Sunday, November 07, 2010 9:15 PM
> To: srinath@mistralsolutions.com; linux-omap@vger.kernel.org
> Subject: RE: Mistral AM37x (with AM3715) devel board networking broken
> 
> sanjeev,
> 
>   Thanks much for the help.
> 
> > [sp] On first looks I would suspect the u-boot. 2009.11 
> seems to be quite old.
> >      Can you try using version 2010.09?
> 
>   I had tried 2010.09 on this board originally but found that it
> wasn't stable, it was almost like the SDRAM wasn't set up correctly.
> I'd get random panic's on bad pointers in Linux.  But I tried it
> anyway on this new kernel and it does set up the networking about 1 in
> 5 boots.  Here is a boot where it set up the network
> http://www.efn.org/~rick/pub/boot4.log, it still has the 
> warnings though.

[sp] If you noticed, earlier u-boot version was identifying your board as:

U-Boot 2009.11 (Oct 30 2010 - 13:14:45)                                         
                                                                                
OMAP34xx/35xx-GP ES1.0, CPU-OPP2 L3-165MHz                                      
OMAP3 EVM board + LPDDR/NAND                                                    

Now,

U-Boot 2010.09 (Nov 07 2010 - 07:08:44)                                         
                                                                                
OMAP3630/3730-GP ES1.0, CPU-OPP2, L3-165MHz, Max CPU Clock 1 Ghz                
OMAP3 EVM board + LPDDR/NAND     

That's ai improvement.

However, in your log, I see folloing:

[    0.000000] Total of 96 interrupts on 1 active controller
[    0.000000] Could not get gpios_ick
[    0.000000] Could not get gpios_fck

Looks like there are some changes - either in your Kconfig or
elsewhere in the code that "get" gpio clocks is failing.

The SMSC chip is "virtually" non-existent for the Linux kernel
here onwards.

My kernel version is "Linux-2.6.36-rc8-09072-g661df74-"

I have not yet looked at, but there is possible "break" on on mainline for EVM due to gpio related changes.
You could be experiencing this break...

~sanjeev

> 
>   Please check my setup of u-boot, here is what I did:
> 
> wget ftp://ftp.denx.de/pub/u-boot/u-boot-2010.09.tar.bz2
> tar jxf u-boot-2010.09.tar.bz2
> cd u-boot-2010.09; make omap3_evm_config; make
> 
>   Is this the correct config (omap3_evm_config) to use for this board?
> Maybe you could tell me what exact u-boot and kernel you'd recommend
> and the steps to get them for this board.
> 
>   Thanks,
> 
>   Rick
> --
> To unsubscribe from this list: send the line "unsubscribe 
> linux-omap" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html
> 

RE: Mistral AM37x (with AM3715) devel board networking broken

[Premi, Sanjeev]

> -----Original Message-----
> From: linux-omap-owner@vger.kernel.org 
> [mailto:linux-omap-owner@vger.kernel.org] On Behalf Of Premi, Sanjeev
> Sent: Monday, November 08, 2010 5:37 PM
> To: Rick Bronson; srinath@mistralsolutions.com; 
> linux-omap@vger.kernel.org
> Subject: RE: Mistral AM37x (with AM3715) devel board networking broken
> 
> > -----Original Message-----
> > From: linux-omap-owner@vger.kernel.org 
> > [mailto:linux-omap-owner@vger.kernel.org] On Behalf Of Rick Bronson
> > Sent: Sunday, November 07, 2010 9:15 PM
> > To: srinath@mistralsolutions.com; linux-omap@vger.kernel.org
> > Subject: RE: Mistral AM37x (with AM3715) devel board 
> networking broken
> > 
> > sanjeev,
> > 
> >   Thanks much for the help.
> > 
> > > [sp] On first looks I would suspect the u-boot. 2009.11 
> > seems to be quite old.
> > >      Can you try using version 2010.09?
> > 
> >   I had tried 2010.09 on this board originally but found that it
> > wasn't stable, it was almost like the SDRAM wasn't set up correctly.
> > I'd get random panic's on bad pointers in Linux.  But I tried it
> > anyway on this new kernel and it does set up the networking 
> about 1 in
> > 5 boots.  Here is a boot where it set up the network
> > http://www.efn.org/~rick/pub/boot4.log, it still has the 
> > warnings though.
> 
> [sp] If you noticed, earlier u-boot version was identifying 
> your board as:
> 
> U-Boot 2009.11 (Oct 30 2010 - 13:14:45)                       
>                   
>                                                               
>                   
> OMAP34xx/35xx-GP ES1.0, CPU-OPP2 L3-165MHz                    
>                   
> OMAP3 EVM board + LPDDR/NAND                                  
>                   
> 
> Now,
> 
> U-Boot 2010.09 (Nov 07 2010 - 07:08:44)                       
>                   
>                                                               
>                   
> OMAP3630/3730-GP ES1.0, CPU-OPP2, L3-165MHz, Max CPU Clock 1 
> Ghz                
> OMAP3 EVM board + LPDDR/NAND     
> 
> That's ai improvement.
> 
> However, in your log, I see folloing:
> 
> [    0.000000] Total of 96 interrupts on 1 active controller
> [    0.000000] Could not get gpios_ick
> [    0.000000] Could not get gpios_fck
> 
> Looks like there are some changes - either in your Kconfig or
> elsewhere in the code that "get" gpio clocks is failing.
> 
> The SMSC chip is "virtually" non-existent for the Linux kernel
> here onwards.
> 
> My kernel version is "Linux-2.6.36-rc8-09072-g661df74-"
> 
> I have not yet looked at, but there is possible "break" on on 
> mainline for EVM due to gpio related changes.
> You could be experiencing this break...
> 
> ~sanjeev

[snip]

I tried building the latest on l-o master and was able to boot consistently
in 4 times I tried. Didn't notice the gpio clock related error I noticed in
your boot log.


U-Boot 2010.09 (Nov 08 2010 - 17:11:45)

OMAP3630/3730-GP ES2.0, CPU-OPP2, L3-165MHz, Max CPU Clock 1 Ghz
OMAP3 EVM board + LPDDR/NAND
...
[    0.000000] Clocking rate (Crystal/Core/MPU): 26.0/200/600 MHz
[    0.000000] Reprogramming SDRC clock to 200000000 Hz
[    0.000000] dpll3_m2_clk rate change failed: -22
[    0.000000] GPMC revision 5.0
[    0.000000] IRQ: Found an INTC at 0xfa200000 (revision 4.0) with 96 interrupts
[    0.000000] Total of 96 interrupts on 1 active controller
[    0.000000] OMAP GPIO hardware version 2.5
[    0.000000] OMAP clockevent source: GPTIMER1 at 32768 Hz
[    0.000000] Console: colour dummy device 80x30
[    0.000000] Lock dependency validator: Copyright (c) 2006 Red Hat, Inc., Ingo Molnar
...
[    2.061950] net eth0: SMSC911x/921x identified at 0xc8880000, IRQ: 336
[    3.075561] Sending DHCP requests .., OK
[    5.169769] IP-Config: Got DHCP answer from 0.0.0.0, my address is xx.yy.zz.00(edited)
...

# uname -a
Linux OMAP3EVM 2.6.37-rc1-08879-g9988a6b-dirty #1 Mon Nov 8 19:50:18 IST 2010 armv7l unknown
#

RE: Mistral AM37x (with AM3715) devel board networking broken

[Rick Bronson]

Hi Sanjeev,

  Thanks much for the help.

> I tried building the latest on l-o master and was able to boot consistently
> in 4 times I tried. Didn't notice the gpio clock related error I noticed in
> your boot log.
> 
> 
> U-Boot 2010.09 (Nov 08 2010 - 17:11:45)
> 
> OMAP3630/3730-GP ES2.0, CPU-OPP2, L3-165MHz, Max CPU Clock 1 Ghz
> OMAP3 EVM board + LPDDR/NAND

 Note that mine is ES1.0 (yours is ES2.0):

OMAP3630/3730-GP ES1.0, CPU-OPP2, L3-165MHz, Max CPU Clock 1 Ghz                

  Not sure if that's any clue.

  Tonight I got the latest stuff:

git clone git://git.kernel.org/pub/scm/linux/kernel/git/tmlind/linux-omap-2.6.git

  and built it up with:

make omap2plus_defconfig ARCH=arm

  And got this

http://www.efn.org/~rick/pub/boot6.log

  It's basically the same, I get the 3 warnings and sometimes eth0 works
and sometimes it doesn't.  Note that my board works flawlessly with
the kernel on the SD Card.

  What compiler are you using?  I tried CodeSourcery
2009q1 and 2010q1 with the same results.

  How did you configure your u-boot?

  Could you send me your boot up text?  Could you send me your
u-boot.bin (and or your uImage) so I can try it?  I might help me
figure out what's going on.

 Thanks again.

 Rick

Re: Mistral AM37x (with AM3715) devel board networking broken

[Gadiyar, Anand]

On Tue, Nov 9, 2010 at 10:23 AM, Rick Bronson <rick@efn.org> wrote:
> Hi Sanjeev,
>
>  Thanks much for the help.
>
>> I tried building the latest on l-o master and was able to boot consistently
>> in 4 times I tried. Didn't notice the gpio clock related error I noticed in
>> your boot log.
>>
>>
>> U-Boot 2010.09 (Nov 08 2010 - 17:11:45)
>>
>> OMAP3630/3730-GP ES2.0, CPU-OPP2, L3-165MHz, Max CPU Clock 1 Ghz
>> OMAP3 EVM board + LPDDR/NAND
>
>  Note that mine is ES1.0 (yours is ES2.0):
>
> OMAP3630/3730-GP ES1.0, CPU-OPP2, L3-165MHz, Max CPU Clock 1 Ghz
>
>  Not sure if that's any clue.
>

That's interesting - there shouldn't b e an ES2.0 of the 3630/3730. The highest
revision is an ES1.2.

>  Tonight I got the latest stuff:
>
> git clone git://git.kernel.org/pub/scm/linux/kernel/git/tmlind/linux-omap-2.6.git
>
>  and built it up with:
>
> make omap2plus_defconfig ARCH=arm
>
>  And got this
>
> http://www.efn.org/~rick/pub/boot6.log
>
>  It's basically the same, I get the 3 warnings and sometimes eth0 works
> and sometimes it doesn't.  Note that my board works flawlessly with
> the kernel on the SD Card.
>
>  What compiler are you using?  I tried CodeSourcery
> 2009q1 and 2010q1 with the same results.
>
>  How did you configure your u-boot?
>
>  Could you send me your boot up text?  Could you send me your
> u-boot.bin (and or your uImage) so I can try it?  I might help me
> figure out what's going on.
>
>  Thanks again.
>
>  Rick
> --
> To unsubscribe from this list: send the line "unsubscribe linux-omap" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html
>
--
To unsubscribe from this list: send the line "unsubscribe linux-omap" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

RE: Mistral AM37x (with AM3715) devel board networking broken

[Premi, Sanjeev]

> -----Original Message-----
> From: linux-omap-owner@vger.kernel.org 
> [mailto:linux-omap-owner@vger.kernel.org] On Behalf Of Gadiyar, Anand
> Sent: Tuesday, November 09, 2010 10:32 AM
> To: Rick Bronson
> Cc: linux-omap@vger.kernel.org; Premi, Sanjeev
> Subject: Re: Mistral AM37x (with AM3715) devel board networking broken
> 
> On Tue, Nov 9, 2010 at 10:23 AM, Rick Bronson <rick@efn.org> wrote:
> > Hi Sanjeev,
> >
> >  Thanks much for the help.
> >
> >> I tried building the latest on l-o master and was able to 
> boot consistently
> >> in 4 times I tried. Didn't notice the gpio clock related 
> error I noticed in
> >> your boot log.
> >>
> >>
> >> U-Boot 2010.09 (Nov 08 2010 - 17:11:45)
> >>
> >> OMAP3630/3730-GP ES2.0, CPU-OPP2, L3-165MHz, Max CPU Clock 1 Ghz
> >> OMAP3 EVM board + LPDDR/NAND
> >
> >  Note that mine is ES1.0 (yours is ES2.0):
> >
> > OMAP3630/3730-GP ES1.0, CPU-OPP2, L3-165MHz, Max CPU Clock 1 Ghz
> >
> >  Not sure if that's any clue.
> >
> 
> That's interesting - there shouldn't b e an ES2.0 of the 
> 3630/3730. The highest
> revision is an ES1.2.
> 

Now, I notice it too. Checking the u-boot sources now.

[snip]...[snip]

--
To unsubscribe from this list: send the line "unsubscribe linux-omap" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

RE: Mistral AM37x (with AM3715) devel board networking broken

[Premi, Sanjeev]

> -----Original Message-----
> From: Rick Bronson [mailto:rick@efn.org] 
> Sent: Tuesday, November 09, 2010 10:24 AM
> To: linux-omap@vger.kernel.org
> Cc: Premi, Sanjeev
> Subject: RE: Mistral AM37x (with AM3715) devel board networking broken
> 
> Hi Sanjeev,
> 
>   Thanks much for the help.
> 
> > I tried building the latest on l-o master and was able to 
> boot consistently
> > in 4 times I tried. Didn't notice the gpio clock related 
> error I noticed in
> > your boot log.
> > 
> > 
> > U-Boot 2010.09 (Nov 08 2010 - 17:11:45)
> > 
> > OMAP3630/3730-GP ES2.0, CPU-OPP2, L3-165MHz, Max CPU Clock 1 Ghz
> > OMAP3 EVM board + LPDDR/NAND
> 
>  Note that mine is ES1.0 (yours is ES2.0):
> 
> OMAP3630/3730-GP ES1.0, CPU-OPP2, L3-165MHz, Max CPU Clock 1 
> Ghz                
> 
>   Not sure if that's any clue.
> 
>   Tonight I got the latest stuff:
> 
> git clone 
> git://git.kernel.org/pub/scm/linux/kernel/git/tmlind/linux-oma
> p-2.6.git
> 
>   and built it up with:
> 
> make omap2plus_defconfig ARCH=arm
> 
>   And got this
> 
> http://www.efn.org/~rick/pub/boot6.log
> 
>   It's basically the same, I get the 3 warnings and sometimes 
> eth0 works
> and sometimes it doesn't.  Note that my board works flawlessly with
> the kernel on the SD Card.
> 
>   What compiler are you using?  I tried CodeSourcery
> 2009q1 and 2010q1 with the same results.

[sp] I am mostly using 2009q1; but have use 2010q1 earlier in JUL;
     didn't see any issues back then.

> 
>   How did you configure your u-boot?

[sp] It is default configuration on the u-boot master. 

> 
>   Could you send me your boot up text?  Could you send me your
> u-boot.bin (and or your uImage) so I can try it?  I might help me
> figure out what's going on.

[sp] Will send offline

~sanjeev

> 
>  Thanks again.
> 
>  Rick
> 

RE: Mistral AM37x (with AM3715) devel board networking broken

[Rick Bronson]

Hi,

  I've found the problem and below is the fix.  It appears your
SMSC9200 is faster than mine ;-)  Once in a while mine gets through the
smsc911x_soft_reset() in 100 us but most of the time it takes a full
250 ms.

  Rick

--- linux/drivers/net/smsc911x.c.~1~	2010-11-08 19:53:03.000000000 -0800
+++ linux/drivers/net/smsc911x.c	2010-11-11 12:00:18.000000000 -0800
@@ -791,8 +791,8 @@
 		return -ENODEV;
 	}
 
-	SMSC_TRACE(PROBE, "PHY %d: addr %d, phy_id 0x%08X",
-			phy_addr, phydev->addr, phydev->phy_id);
+	SMSC_TRACE(PROBE, "PHY %s: addr %d, phy_id 0x%08X",
+			dev_name(&phydev->dev), phydev->addr, phydev->phy_id);
 
 	ret = phy_connect_direct(dev, phydev,
 			&smsc911x_phy_adjust_link, 0,
@@ -1124,11 +1124,11 @@
 
 	/* Reset the LAN911x */
 	smsc911x_reg_write(pdata, HW_CFG, HW_CFG_SRST_);
-	timeout = 10;
-	do {
-		udelay(10);
-		temp = smsc911x_reg_read(pdata, HW_CFG);
-	} while ((--timeout) && (temp & HW_CFG_SRST_));
+	timeout = 10000;  /* this can take up to 1/4 sec */
+	while ((smsc911x_reg_read(pdata, HW_CFG) & HW_CFG_SRST_) && timeout) {
+		udelay(100);
+		timeout--;
+	}
 
 	if (unlikely(temp & HW_CFG_SRST_)) {
 		SMSC_WARNING(DRV, "Failed to complete reset");