[U-Boot] [ANN] U-Boot v2017.03-rc2 released
Hey all,
It's release day and v2017.03-rc2 is out. I think my patchwork queue is looking good currently but I'm open to being told there's important stuff outstanding. I'm probably going to grab some Kconfig things and perhaps other cleanups I can mechanically confirm are correct. Things are on track for -rc3 on the 27th.
Thanks all!
Hello Tom,
Am 14.02.2017 um 00:01 schrieb Tom Rini:
Hey all,
It's release day and v2017.03-rc2 is out. I think my patchwork queue is looking good currently but I'm open to being told there's important stuff outstanding. I'm probably going to grab some Kconfig things and perhaps other cleanups I can mechanically confirm are correct. Things are on track for -rc3 on the 27th.
Hmm.. my weekly test of current mainline u-boot shows a problem with the at91 based smartweb and corvus boards:
smartweb log: U-Boot 2017.03-rc2-00002-g6484233 (Feb 14 2017 - 03:01:05 +0100)
CPU: AT91SAM9260 Crystal frequency: 18.432 MHz CPU clock : 198.656 MHz Master clock : 99.328 MHz Watchdog enabled DRAM: 64 MiB
NAND: 256 MiB
In: serial@fffff200
Out: serial@fffff200
Err: serial@fffff200
Net: No ethernet found.
U-Boot#
corvus: U-Boot SPL 2017.03-rc2-00002-g0d9386a (Feb 14 2017 - 03:41:27) Trying to boot from NAND
U-Boot 2017.03-rc2-00002-g0d9386a (Feb 14 2017 - 03:41:27 +0100)
CPU: AT91SAM9G45 Crystal frequency: 12 MHz CPU clock : 400 MHz Master clock : 133.333 MHz DRAM: 128 MiB NAND: 256 MiB In: serial@ffffee00 Out: serial@ffffee00 Err: serial@ffffee00 Net: No ethernet found. Hit any key to stop autoboot U-Boot>
No ethernet found with current mainline,
Last week it ethernet worked: xeidos.ddns.net/tests/test_db_auslesen.php#237 xeidos.ddns.net/tests/test_db_auslesen.php#236
I try to find time to look into it ... added Andreas to it, may he know something ...
bye, Heiko
Hello Tom,
Am 14.02.2017 um 06:38 schrieb Heiko Schocher:
Hello Tom,
Am 14.02.2017 um 00:01 schrieb Tom Rini:
Hey all,
It's release day and v2017.03-rc2 is out. I think my patchwork queue is looking good currently but I'm open to being told there's important stuff outstanding. I'm probably going to grab some Kconfig things and perhaps other cleanups I can mechanically confirm are correct. Things are on track for -rc3 on the 27th.
Hmm.. my weekly test of current mainline u-boot shows a problem with the at91 based smartweb and corvus boards:
smartweb log: U-Boot 2017.03-rc2-00002-g6484233 (Feb 14 2017 - 03:01:05 +0100)
CPU: AT91SAM9260 Crystal frequency: 18.432 MHz CPU clock : 198.656 MHz Master clock : 99.328 MHz Watchdog enabled DRAM: 64 MiB NAND: 256 MiB In: serial@fffff200 Out: serial@fffff200 Err: serial@fffff200 Net: No ethernet found. U-Boot#
corvus: U-Boot SPL 2017.03-rc2-00002-g0d9386a (Feb 14 2017 - 03:41:27) Trying to boot from NAND
U-Boot 2017.03-rc2-00002-g0d9386a (Feb 14 2017 - 03:41:27 +0100)
CPU: AT91SAM9G45 Crystal frequency: 12 MHz CPU clock : 400 MHz Master clock : 133.333 MHz DRAM: 128 MiB NAND: 256 MiB In: serial@ffffee00 Out: serial@ffffee00 Err: serial@ffffee00 Net: No ethernet found. Hit any key to stop autoboot U-Boot>
No ethernet found with current mainline,
Last week it ethernet worked: xeidos.ddns.net/tests/test_db_auslesen.php#237 xeidos.ddns.net/tests/test_db_auslesen.php#236
I try to find time to look into it ... added Andreas to it, may he know something ...
Could bisect it (not yet automated with tbot, as without ethernet, I had to load the new u-boot bin with "loadb" ...)
pollux:u-boot-smartweb hs [(kein Branch, binäre Suche begonnen bei master)] $ git bisect visualize commit 577aa3b3587193ee078ce83d918604da6ac84dd5 Author: Wenyou Yang wenyou.yang@atmel.com Date: Wed Nov 2 10:06:56 2016 +0800
net: macb: Add the clock support
Due to introducing the at91 clock driver, add the clock support.
Signed-off-by: Wenyou Yang wenyou.yang@atmel.com Acked-by: Joe Hershberger joe.hershberger@ni.com pollux:u-boot-smartweb hs [(kein Branch, binäre Suche begonnen bei master)] $ git bisect log git bisect start # bad: [6484233c463a1b79bae651a7fa8504a838425eb7] dfu: introduce dfu_mtd support git bisect bad 6484233c463a1b79bae651a7fa8504a838425eb7 # good: [446d4e048ee3b00f7907e15cd02aa404cc714c77] x86: make LOAD_FROM_32_BIT visible for platforms git bisect good 446d4e048ee3b00f7907e15cd02aa404cc714c77 # good: [6f57b19857b514f6e94e423c9bbe2d4b8973a0db] Merge branch 'master' of git://git.denx.de/u-boot-mmc git bisect good 6f57b19857b514f6e94e423c9bbe2d4b8973a0db # bad: [2a48b3a2c4ec0f1775deeac5cfe7e61072b6b894] omap_hsmmc.c: Fix build warning on non-omap3 git bisect bad 2a48b3a2c4ec0f1775deeac5cfe7e61072b6b894 # bad: [83cfbeb0df9f5962a16e8737e08cf59ed84e0cff] net: phy: Fix mask so that we can identify Marvell 88E1518 git bisect bad 83cfbeb0df9f5962a16e8737e08cf59ed84e0cff # bad: [9f455bcb34fcab31b0eb8eec5702105c8d4bde18] net: cosmetic: Make the MAC address string less magical git bisect bad 9f455bcb34fcab31b0eb8eec5702105c8d4bde18 # bad: [577aa3b3587193ee078ce83d918604da6ac84dd5] net: macb: Add the clock support git bisect bad 577aa3b3587193ee078ce83d918604da6ac84dd5 # good: [ebcb40a5a02adc153db949d59e9df9bf0d89b6ec] net: Kconfig: Add CONFIG_MACB option git bisect good ebcb40a5a02adc153db949d59e9df9bf0d89b6ec # first bad commit: [577aa3b3587193ee078ce83d918604da6ac84dd5] net: macb: Add the clock support pollux:u-boot-smartweb hs [(kein Branch, binäre Suche begonnen bei master)] $
A fast revert of this patch, and u-boot works with ethernet again on the smartweb board ... any ideas?
bye, Heiko
Hi Heiko,
-----Original Message----- From: Heiko Schocher [mailto:hs@denx.de] Sent: 2017年2月14日 14:26 To: Tom Rini trini@konsulko.com; Tony Yang - A41535 Wenyou.Yang@microchip.com Cc: u-boot@lists.denx.de; Andreas Bießmann andreas.devel@googlemail.com; Joe Hershberger joe.hershberger@ni.com Subject: Re: [U-Boot] [ANN] U-Boot v2017.03-rc2 released
Hello Tom,
Am 14.02.2017 um 06:38 schrieb Heiko Schocher:
Hello Tom,
Am 14.02.2017 um 00:01 schrieb Tom Rini:
Hey all,
It's release day and v2017.03-rc2 is out. I think my patchwork queue is looking good currently but I'm open to being told there's important stuff outstanding. I'm probably going to grab some Kconfig things and perhaps other cleanups I can mechanically confirm are correct. Things are on track for -rc3 on the 27th.
Hmm.. my weekly test of current mainline u-boot shows a problem with the at91 based smartweb and corvus boards:
smartweb log: U-Boot 2017.03-rc2-00002-g6484233 (Feb 14 2017 - 03:01:05 +0100)
CPU: AT91SAM9260 Crystal frequency: 18.432 MHz CPU clock : 198.656 MHz Master clock : 99.328 MHz Watchdog enabled DRAM: 64 MiB NAND: 256 MiB In: serial@fffff200 Out: serial@fffff200 Err: serial@fffff200 Net: No ethernet found. U-Boot#
corvus: U-Boot SPL 2017.03-rc2-00002-g0d9386a (Feb 14 2017 - 03:41:27) Trying to boot from NAND
U-Boot 2017.03-rc2-00002-g0d9386a (Feb 14 2017 - 03:41:27 +0100)
CPU: AT91SAM9G45 Crystal frequency: 12 MHz CPU clock : 400 MHz Master clock : 133.333 MHz DRAM: 128 MiB NAND: 256 MiB In: serial@ffffee00 Out: serial@ffffee00 Err: serial@ffffee00 Net: No ethernet found. Hit any key to stop autoboot U-Boot>
No ethernet found with current mainline,
Last week it ethernet worked: xeidos.ddns.net/tests/test_db_auslesen.php#237 xeidos.ddns.net/tests/test_db_auslesen.php#236
I try to find time to look into it ... added Andreas to it, may he know something ...
Could bisect it (not yet automated with tbot, as without ethernet, I had to load the new u-boot bin with "loadb" ...)
pollux:u-boot-smartweb hs [(kein Branch, binäre Suche begonnen bei master)] $ git bisect visualize commit 577aa3b3587193ee078ce83d918604da6ac84dd5 Author: Wenyou Yang wenyou.yang@atmel.com Date: Wed Nov 2 10:06:56 2016 +0800
net: macb: Add the clock support Due to introducing the at91 clock driver, add the clock support. Signed-off-by: Wenyou Yang <wenyou.yang@atmel.com> Acked-by: Joe Hershberger <joe.hershberger@ni.com> pollux:u-boot-smartweb hs [(kein Branch, binäre Suche begonnen bei master)] $ git bisect log git bisect start # bad: [6484233c463a1b79bae651a7fa8504a838425eb7] dfu: introduce dfu_mtd support git bisect bad 6484233c463a1b79bae651a7fa8504a838425eb7 # good: [446d4e048ee3b00f7907e15cd02aa404cc714c77] x86: make LOAD_FROM_32_BIT visible for platforms git bisect good 446d4e048ee3b00f7907e15cd02aa404cc714c77 # good: [6f57b19857b514f6e94e423c9bbe2d4b8973a0db] Merge branch 'master' of git://git.denx.de/u-boot-mmc git bisect good 6f57b19857b514f6e94e423c9bbe2d4b8973a0db # bad: [2a48b3a2c4ec0f1775deeac5cfe7e61072b6b894] omap_hsmmc.c: Fix build warning on non-omap3 git bisect bad 2a48b3a2c4ec0f1775deeac5cfe7e61072b6b894 # bad: [83cfbeb0df9f5962a16e8737e08cf59ed84e0cff] net: phy: Fix mask so that we can identify Marvell 88E1518 git bisect bad 83cfbeb0df9f5962a16e8737e08cf59ed84e0cff # bad: [9f455bcb34fcab31b0eb8eec5702105c8d4bde18] net: cosmetic: Make the MAC address string less magical git bisect bad 9f455bcb34fcab31b0eb8eec5702105c8d4bde18 # bad: [577aa3b3587193ee078ce83d918604da6ac84dd5] net: macb: Add the clock support git bisect bad 577aa3b3587193ee078ce83d918604da6ac84dd5 # good: [ebcb40a5a02adc153db949d59e9df9bf0d89b6ec] net: Kconfig: Add CONFIG_MACB option git bisect good ebcb40a5a02adc153db949d59e9df9bf0d89b6ec # first bad commit: [577aa3b3587193ee078ce83d918604da6ac84dd5] net: macb: Add the clock support pollux:u-boot-smartweb hs [(kein Branch, binäre Suche begonnen bei master)] $
A fast revert of this patch, and u-boot works with ethernet again on the smartweb board ... any ideas?
The clock code should be enclosed by the conditional macro, such as #ifdef CONFIG_CLK ... #endif.
I missed it. Sorry.
Will send a patch to fix.
bye, Heiko -- DENX Software Engineering GmbH, Managing Director: Wolfgang Denk HRB 165235 Munich, Office: Kirchenstr.5, D-82194 Groebenzell, Germany
Best Regards, Wenyou Yangls
On Tue, Feb 14, 2017 at 07:25:33AM +0100, Heiko Schocher wrote:
Hello Tom,
Am 14.02.2017 um 06:38 schrieb Heiko Schocher:
Hello Tom,
Am 14.02.2017 um 00:01 schrieb Tom Rini:
Hey all,
It's release day and v2017.03-rc2 is out. I think my patchwork queue is looking good currently but I'm open to being told there's important stuff outstanding. I'm probably going to grab some Kconfig things and perhaps other cleanups I can mechanically confirm are correct. Things are on track for -rc3 on the 27th.
Hmm.. my weekly test of current mainline u-boot shows a problem with the at91 based smartweb and corvus boards:
smartweb log: U-Boot 2017.03-rc2-00002-g6484233 (Feb 14 2017 - 03:01:05 +0100)
CPU: AT91SAM9260 Crystal frequency: 18.432 MHz CPU clock : 198.656 MHz Master clock : 99.328 MHz Watchdog enabled DRAM: 64 MiB NAND: 256 MiB In: serial@fffff200 Out: serial@fffff200 Err: serial@fffff200 Net: No ethernet found. U-Boot#
corvus: U-Boot SPL 2017.03-rc2-00002-g0d9386a (Feb 14 2017 - 03:41:27) Trying to boot from NAND
U-Boot 2017.03-rc2-00002-g0d9386a (Feb 14 2017 - 03:41:27 +0100)
CPU: AT91SAM9G45 Crystal frequency: 12 MHz CPU clock : 400 MHz Master clock : 133.333 MHz DRAM: 128 MiB NAND: 256 MiB In: serial@ffffee00 Out: serial@ffffee00 Err: serial@ffffee00 Net: No ethernet found. Hit any key to stop autoboot U-Boot>
No ethernet found with current mainline,
Last week it ethernet worked: xeidos.ddns.net/tests/test_db_auslesen.php#237 xeidos.ddns.net/tests/test_db_auslesen.php#236
I try to find time to look into it ... added Andreas to it, may he know something ...
Could bisect it (not yet automated with tbot, as without ethernet, I had to load the new u-boot bin with "loadb" ...)
Good catch, thanks!
On Tue, Feb 14, 2017 at 1:01 AM, Tom Rini trini@konsulko.com wrote:
Hey all,
It's release day and v2017.03-rc2 is out. I think my patchwork queue is looking good currently but I'm open to being told there's important stuff outstanding. I'm probably going to grab some Kconfig things and perhaps other cleanups I can mechanically confirm are correct. Things are on track for -rc3 on the 27th.
+Cc: Simon
This build weirdly breaks on Intel Edison. -rc1 works fine, while just rebase on -rc2 becomes unsable
On the first boot *original Edison* Linux kernel (v3.10.98 based) doesn't print anything (looks like internal crash in U-Boot), on the sequential we have corrupted media, so, board is basically bricked.
I dunno how to debug this since Tangier/Edison support is not (yet) in U-Boot upstream.
Any hints?
Hi Andy,
On 14 February 2017 at 10:21, Andy Shevchenko andy.shevchenko@gmail.com wrote:
On Tue, Feb 14, 2017 at 1:01 AM, Tom Rini trini@konsulko.com wrote:
Hey all,
It's release day and v2017.03-rc2 is out. I think my patchwork queue is looking good currently but I'm open to being told there's important stuff outstanding. I'm probably going to grab some Kconfig things and perhaps other cleanups I can mechanically confirm are correct. Things are on track for -rc3 on the 27th.
+Cc: Simon
This build weirdly breaks on Intel Edison. -rc1 works fine, while just rebase on -rc2 becomes unsable
On the first boot *original Edison* Linux kernel (v3.10.98 based) doesn't print anything (looks like internal crash in U-Boot), on the sequential we have corrupted media, so, board is basically bricked.
We have the beginnings of 64-bit support in there now, so I suspect something there is breaking Edison. Can you bisect it?
I dunno how to debug this since Tangier/Edison support is not (yet) in U-Boot upstream.
Any hints?
Hmmm it would be good to get this support into mainline so we can avoid these sorts of problems. I did get the files at some point but have not made time to do it. Do you know if anyone is planning to do this? From memory the patches looked pretty clean.
-- With Best Regards, Andy Shevchenko
Regards, Simon
+Bin
On 14 February 2017 at 10:33, Simon Glass sjg@chromium.org wrote:
Hi Andy,
On 14 February 2017 at 10:21, Andy Shevchenko andy.shevchenko@gmail.com wrote:
On Tue, Feb 14, 2017 at 1:01 AM, Tom Rini trini@konsulko.com wrote:
Hey all,
It's release day and v2017.03-rc2 is out. I think my patchwork queue is looking good currently but I'm open to being told there's important stuff outstanding. I'm probably going to grab some Kconfig things and perhaps other cleanups I can mechanically confirm are correct. Things are on track for -rc3 on the 27th.
+Cc: Simon
This build weirdly breaks on Intel Edison. -rc1 works fine, while just rebase on -rc2 becomes unsable
On the first boot *original Edison* Linux kernel (v3.10.98 based) doesn't print anything (looks like internal crash in U-Boot), on the sequential we have corrupted media, so, board is basically bricked.
We have the beginnings of 64-bit support in there now, so I suspect something there is breaking Edison. Can you bisect it?
I dunno how to debug this since Tangier/Edison support is not (yet) in U-Boot upstream.
Any hints?
Hmmm it would be good to get this support into mainline so we can avoid these sorts of problems. I did get the files at some point but have not made time to do it. Do you know if anyone is planning to do this? From memory the patches looked pretty clean.
-- With Best Regards, Andy Shevchenko
Regards, Simon
On Tue, Feb 14, 2017 at 7:33 PM, Simon Glass sjg@chromium.org wrote:
Hi Andy,
On 14 February 2017 at 10:21, Andy Shevchenko andy.shevchenko@gmail.com wrote:
On Tue, Feb 14, 2017 at 1:01 AM, Tom Rini trini@konsulko.com wrote:
Hey all,
It's release day and v2017.03-rc2 is out. I think my patchwork queue is looking good currently but I'm open to being told there's important stuff outstanding. I'm probably going to grab some Kconfig things and perhaps other cleanups I can mechanically confirm are correct. Things are on track for -rc3 on the 27th.
+Cc: Simon
This build weirdly breaks on Intel Edison. -rc1 works fine, while just rebase on -rc2 becomes unsable
On the first boot *original Edison* Linux kernel (v3.10.98 based) doesn't print anything (looks like internal crash in U-Boot), on the sequential we have corrupted media, so, board is basically bricked.
We have the beginnings of 64-bit support in there now, so I suspect something there is breaking Edison. Can you bisect it?
I don't think so, because each iteration would require to port patches on it. It might take days/weeks since it's not a main work for me.
I dunno how to debug this since Tangier/Edison support is not (yet) in U-Boot upstream.
Any hints?
Hmmm it would be good to get this support into mainline so we can avoid these sorts of problems. I did get the files at some point but have not made time to do it. Do you know if anyone is planning to do this?
Seems we are doing this.
From memory the patches looked pretty clean.
Can you point to the patches repository or any mail archive which we can look at? Currently our work based on latest stuff from github:01org/edison-u-boot repository. There patches for our opinion are quite far from upstream quality.
On Tue, Feb 14, 2017 at 7:40 PM, Andy Shevchenko andy.shevchenko@gmail.com wrote:
On Tue, Feb 14, 2017 at 7:33 PM, Simon Glass sjg@chromium.org wrote:
Hi Andy,
On 14 February 2017 at 10:21, Andy Shevchenko andy.shevchenko@gmail.com wrote:
On Tue, Feb 14, 2017 at 1:01 AM, Tom Rini trini@konsulko.com wrote:
Hey all,
It's release day and v2017.03-rc2 is out. I think my patchwork queue is looking good currently but I'm open to being told there's important stuff outstanding. I'm probably going to grab some Kconfig things and perhaps other cleanups I can mechanically confirm are correct. Things are on track for -rc3 on the 27th.
+Cc: Simon
This build weirdly breaks on Intel Edison. -rc1 works fine, while just rebase on -rc2 becomes unsable
On the first boot *original Edison* Linux kernel (v3.10.98 based) doesn't print anything (looks like internal crash in U-Boot), on the sequential we have corrupted media, so, board is basically bricked.
We have the beginnings of 64-bit support in there now, so I suspect something there is breaking Edison. Can you bisect it?
I don't think so, because each iteration would require to port patches on it. It might take days/weeks since it's not a main work for me.
I dunno how to debug this since Tangier/Edison support is not (yet) in U-Boot upstream.
Any hints?
Hmmm it would be good to get this support into mainline so we can avoid these sorts of problems. I did get the files at some point but have not made time to do it. Do you know if anyone is planning to do this?
Seems we are doing this.
From memory the patches looked pretty clean.
Can you point to the patches repository or any mail archive which we can look at? Currently our work based on latest stuff from github:01org/edison-u-boot repository. There patches for our opinion are quite far from upstream quality.
In addition. Just confirmed that rc1 works fine.
I have pushed to my repo [1] devel branches, so
edison -- rc1-based edison-devel-rc2 -- rc2-based not working edison-devel -- quite modified version of edison-devel-rc2
[1] https://github.com/andy-shev/u-boot/tree/edison
On Tue, Feb 14, 2017 at 7:46 PM, Andy Shevchenko andy.shevchenko@gmail.com wrote:
On Tue, Feb 14, 2017 at 7:40 PM, Andy Shevchenko andy.shevchenko@gmail.com wrote:
On Tue, Feb 14, 2017 at 7:33 PM, Simon Glass sjg@chromium.org wrote:
On 14 February 2017 at 10:21, Andy Shevchenko andy.shevchenko@gmail.com wrote:
On Tue, Feb 14, 2017 at 1:01 AM, Tom Rini trini@konsulko.com wrote:
+Cc: Simon
This build weirdly breaks on Intel Edison. -rc1 works fine, while just rebase on -rc2 becomes unsable
In addition. Just confirmed that rc1 works fine.
I have pushed to my repo [1] devel branches, so
edison -- rc1-based edison-devel-rc2 -- rc2-based not working edison-devel -- quite modified version of edison-devel-rc2
Oops, sorry guys for the noise, it seems I found a culprit and it has nothing to do with -rc2.
Hi Andy,
On 14 February 2017 at 10:40, Andy Shevchenko andy.shevchenko@gmail.com wrote:
On Tue, Feb 14, 2017 at 7:33 PM, Simon Glass sjg@chromium.org wrote:
Hi Andy,
On 14 February 2017 at 10:21, Andy Shevchenko andy.shevchenko@gmail.com wrote:
On Tue, Feb 14, 2017 at 1:01 AM, Tom Rini trini@konsulko.com wrote:
Hey all,
It's release day and v2017.03-rc2 is out. I think my patchwork queue is looking good currently but I'm open to being told there's important stuff outstanding. I'm probably going to grab some Kconfig things and perhaps other cleanups I can mechanically confirm are correct. Things are on track for -rc3 on the 27th.
+Cc: Simon
This build weirdly breaks on Intel Edison. -rc1 works fine, while just rebase on -rc2 becomes unsable
On the first boot *original Edison* Linux kernel (v3.10.98 based) doesn't print anything (looks like internal crash in U-Boot), on the sequential we have corrupted media, so, board is basically bricked.
We have the beginnings of 64-bit support in there now, so I suspect something there is breaking Edison. Can you bisect it?
I don't think so, because each iteration would require to port patches on it. It might take days/weeks since it's not a main work for me.
I dunno how to debug this since Tangier/Edison support is not (yet) in U-Boot upstream.
Any hints?
Hmmm it would be good to get this support into mainline so we can avoid these sorts of problems. I did get the files at some point but have not made time to do it. Do you know if anyone is planning to do this?
Seems we are doing this.
From memory the patches looked pretty clean.
Can you point to the patches repository or any mail archive which we can look at?
Sure - can you please send me an email from your Intel address?
Currently our work based on latest stuff from github:01org/edison-u-boot repository. There patches for our opinion are quite far from upstream quality.
Regards, Simon
participants (5)
-
Andy Shevchenko -
Heiko Schocher -
Simon Glass -
Tom Rini -
Wenyou.Yang＠microchip.com