[U-Boot] openrd_ultimate GigEth ports not working
Hi list,
I've just bought an OpenRD Ultimate board and I've managed to get u-boot running on it. I am building it like this, using the latest u-boot GIT repo:
make mrproper CROSS_COMPILE=arm-none-linux-gnueabi- make openrd_ultimate_config CROSS_COMPILE=arm-none-linux-gnueabi- make u-boot.kwb CROSS_COMPILE=arm-none-linux-gnueabi-
and then I burn the .kwb file into the flash using openocd.
Unfortunately I cannot get the Gig Eth ports to work in u-boot at the moment. When I run the following at the command prompt the console just hangs and I have to reboot:
Marvell>> setenv ipaddr 192.168.1.1 Marvell>> ping 192.168.1.30 ... board hangs ...
Has anybody else seen this issue?
Thanks in advance for any help you can give.
Alex
Alex Zeffertt writes:
and then I burn the .kwb file into the flash using openocd.
Unfortunately I cannot get the Gig Eth ports to work in u-boot at the moment. When I run the following at the command prompt the console just hangs and I have to reboot:
Marvell>> setenv ipaddr 192.168.1.1 Marvell>> ping 192.168.1.30 ... board hangs ...
Has anybody else seen this issue?
I see something similar with the Seagate Dockstar (also Kirkwood) when using dhcp. It also hangs. I did not test ping. The 2011.06 snapshot works fine.
-Ralph
On 25 July 2011 18:14, Ralph Metzler rjkm@metzlerbros.de wrote:
Alex Zeffertt writes:
and then I burn the .kwb file into the flash using openocd.
Unfortunately I cannot get the Gig Eth ports to work in u-boot at the moment. When I run the following at the command prompt the console just hangs and I have to reboot:
Marvell>> setenv ipaddr 192.168.1.1 Marvell>> ping 192.168.1.30 ... board hangs ...
Has anybody else seen this issue?
I see something similar with the Seagate Dockstar (also Kirkwood) when using dhcp. It also hangs. I did not test ping. The 2011.06 snapshot works fine.
-Ralph
Thanks Ralph,
I can confirm that the issue is the same with OpenRD Ultimate. I.e. ethernet is not working in HEAD but it is fine on tag v2011.06.
Alex
Dear Alex Zeffertt,
In message CANNiwJKvYkei0qmRC8V5LvTar+q8CzYeTYCpqDjzV34HtCL6kQ@mail.gmail.com you wrote:
I can confirm that the issue is the same with OpenRD Ultimate. I.e. ethernet is not working in HEAD but it is fine on tag v2011.06.
And what says "git bisect" ?
Best regards,
Wolfgang Denk
git bisect says:
c2dd0d45540397704de9b13287417d21049d34c6 is the first bad commit commit c2dd0d45540397704de9b13287417d21049d34c6 Author: Aneesh V aneesh@ti.com Date: Thu Jun 16 23:30:49 2011 +0000
armv7: integrate cache maintenance support
- Enable I-cache on bootup - Enable MMU and D-cache immediately after relocation - Do necessary initialization before enabling d-cache and MMU - Changes to cleanup_before_linux() - Make changes according to the new framework
Signed-off-by: Aneesh V aneesh@ti.com
:040000 040000 2e80477b0d05ad168feb89ef0e00d9bbd79b0cd0 5505c5f890b2af6e9b54bba915fdf3368abdfb16 M arch
Thanks,
Alex
On 26 July 2011 12:09, Wolfgang Denk wd@denx.de wrote:
Dear Alex Zeffertt,
In message < CANNiwJKvYkei0qmRC8V5LvTar+q8CzYeTYCpqDjzV34HtCL6kQ@mail.gmail.com> you wrote:
I can confirm that the issue is the same with OpenRD Ultimate. I.e. ethernet is not working in HEAD but it is fine on tag v2011.06.
And what says "git bisect" ?
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@denx.de Superior ability breeds superior ambition. -- Spock, "Space Seed", stardate 3141.9
[Cc-ing Prafulla the author of the mvgbe driver]
Hi All, I contacted the author of that commit and this was his response:
Yes, most likely it's the caching. Guess your driver is doing DMA andis not handling the cache flushing/invalidation well to avoid coherency issues. First of all to confirm that this is indeed the issue you may try re-compile your u-boot with CONFIG_SYS_DCACHE_OFF in your config header. That should solve the problem, but the right solution will be to fix your driver.
br, Aneesh
Prafulla, does this make any sense to you?
Regards,
Alex
On 26 July 2011 13:29, Alex Zeffertt azeffertt@cambridgesys.com wrote:
git bisect says:
c2dd0d45540397704de9b13287417d21049d34c6 is the first bad commit commit c2dd0d45540397704de9b13287417d21049d34c6 Author: Aneesh V aneesh@ti.com Date: Thu Jun 16 23:30:49 2011 +0000 armv7: integrate cache maintenance support
- Enable I-cache on bootup - Enable MMU and D-cache immediately after relocation - Do necessary initialization before enabling d-cache and MMU - Changes to cleanup_before_linux() - Make changes according to the new framework
Signed-off-by: Aneesh V aneesh@ti.com :040000 040000 2e80477b0d05ad168feb89ef0e00d9bbd79b0cd0 5505c5f890b2af6e9b54bba915fdf3368abdfb16 M arch
Thanks, Alex
On 26 July 2011 12:09, Wolfgang Denk wd@denx.de wrote:
Dear Alex Zeffertt,
In message CANNiwJKvYkei0qmRC8V5LvTar+q8CzYeTYCpqDjzV34HtCL6kQ@mail.gmail.com you wrote:
I can confirm that the issue is the same with OpenRD Ultimate. I.e. ethernet is not working in HEAD but it is fine on tag v2011.06.
And what says "git bisect" ?
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@denx.de Superior ability breeds superior ambition. -- Spock, "Space Seed", stardate 3141.9
-----Original Message----- From: Alex Zeffertt [mailto:azeffertt@cambridgesys.com] Sent: Monday, August 01, 2011 1:52 PM To: Wolfgang Denk Cc: Ralph Metzler; u-boot@lists.denx.de; Prafulla Wadaskar Subject: Re: [U-Boot] openrd_ultimate GigEth ports not working
[Cc-ing Prafulla the author of the mvgbe driver]
Hi All, I contacted the author of that commit and this was his response:
Yes, most likely it's the caching. Guess your driver is doing DMA
andis not handling the cache flushing/invalidation well to avoid coherency
issues. First of all to confirm that this is indeed the issue you may try re-compile your u-boot with CONFIG_SYS_DCACHE_OFF in your config header. That should solve the problem, but the right solution will be to fix your driver.
br, Aneesh
Prafulla, does this make any sense to you?
Yes, it is. In the mainlined driver, D-cache is considered to be off. In the subsequent patches is it enabled leading the problem.
We need to address some code to handle this or disable D-cache
Regards.. Prafulla . . .
Regards,
Alex
On 26 July 2011 13:29, Alex Zeffertt azeffertt@cambridgesys.com wrote:
git bisect says:
c2dd0d45540397704de9b13287417d21049d34c6 is the first bad commit commit c2dd0d45540397704de9b13287417d21049d34c6 Author: Aneesh V aneesh@ti.com Date: Thu Jun 16 23:30:49 2011 +0000 armv7: integrate cache maintenance support
- Enable I-cache on bootup - Enable MMU and D-cache immediately after relocation - Do necessary initialization before enabling d-cache and MMU - Changes to cleanup_before_linux() - Make changes according to the new framework
Signed-off-by: Aneesh V aneesh@ti.com :040000 040000 2e80477b0d05ad168feb89ef0e00d9bbd79b0cd0
5505c5f890b2af6e9b54bba915fdf3368abdfb16 M arch
Thanks, Alex
On 26 July 2011 12:09, Wolfgang Denk wd@denx.de wrote:
Dear Alex Zeffertt,
In message
CANNiwJKvYkei0qmRC8V5LvTar+q8CzYeTYCpqDjzV34HtCL6kQ@mail.gmail.com you wrote:
I can confirm that the issue is the same with OpenRD Ultimate.
I.e.
ethernet is not working in HEAD but it is fine on tag v2011.06.
And what says "git bisect" ?
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@denx.de Superior ability breeds superior ambition. -- Spock, "Space Seed", stardate 3141.9
participants (4)
-
Alex Zeffertt -
Prafulla Wadaskar -
Ralph Metzler -
Wolfgang Denk