Control: fixed 825214 2016.05+dfsg1-1 On 2016-06-01, Karsten Merker wrote: >> > On 2016-05-24, Benedikt Wildenhain wrote: >> > > The last version of u-boot-sunxi which enabled the system to boot >> > > properly was 2015.10+dfsg1-3. Switching to Linux 4.6.0 from experimental >> > > did not help. > [...] >> IIRC during the v2016.03 development cycle there was some issue with >> enabling/disabling an unused voltage regulator or setting/not setting >> the regulator to a specific voltage that resulted in kernel hangs on >> certain boards while trying to load a certain kernel module (I2C driver?). >> That could be the problem we are seeing here, although I would need >> to look through the mailing list archives for more specific >> information.
> cherry-picking affa020559bca31d6531e19cb1f009c22705a73d (which > addresses the aforementioned regulator issues) from v2016.05 on > top of v2015.03 solved the issue for me, Thanks for tracking that down! > but I am unsure whether > uploading another v2015.03 package with that patch applied would > make sense now. V2016.05 is already packaged, includes this fix > and in addition to that also has a number of phy handling fixes > that should solve the network stability issues with gigabit > ethernet links on the Lime2, so it probably makes more sense to > aim at getting v2016.05 into unstable in the near future than at > patching v2016.03 further. I've hesitated uploading v2016.05 to unstable as I've been experiencing regressions with other boards (am57xx_evm and firefly-rk3288). I just uploaded 2016.07-rc1 to experimental, after testing on several boards, and it fixes am57xx_evm at least. That will need some wider regression testing, too... live well, vagrant
signature.asc
Description: PGP signature
