2015-12-21 17:40:47

by Larry Finger

[permalink] [raw]
Subject: Regression: Kernel unbootable since commit 4d6b4e6 - found by bisection

Since early in the 4.3-rcx series, my Dell Latitude D600 with a 32-bit kernel
fails to boot. Unfortunately, I did not discover this until late in the 4.4-rcX
cycle. The symptom is that the kernel echos the "Loading initial ramdisk ..."
message and then hangs.

The problem was bisected to commit 4d6b4es" x86/PCI/ACPI: Use common interface
to support PCI host bridge" from Jiang Liu <[email protected]>. The
bisection was tested by reverting the commit in question. I have not yet tried
to analyze the patch to see where it might be going wrong.

Please suggest any trial prints to debug this issue.

Thanks,

Larry


2015-12-21 21:24:32

by Keith Busch

[permalink] [raw]
Subject: RE: Regression: Kernel unbootable since commit 4d6b4e6 - found by bisection

> Since early in the 4.3-rcx series, my Dell Latitude D600 with a 32-bit kernel
> fails to boot. Unfortunately, I did not discover this until late in the 4.4-rcX
> cycle. The symptom is that the kernel echos the "Loading initial ramdisk ..."
> message and then hangs.
>
> The problem was bisected to commit 4d6b4es" x86/PCI/ACPI: Use common interface
> to support PCI host bridge" from Jiang Liu <[email protected]>. The
> bisection was tested by reverting the commit in question. I have not yet tried
> to analyze the patch to see where it might be going wrong.
>
> Please suggest any trial prints to debug this issue.

Does this fix help you? It is included with the most recent 4.4 rc.

https://git.kernel.org/cgit/linux/kernel/git/torvalds/linux.git/commit/?id=727ae8be30b428082d3519817f4fb98b712d457d

2015-12-21 22:45:47

by Larry Finger

[permalink] [raw]
Subject: Re: Regression: Kernel unbootable since commit 4d6b4e6 - found by bisection

On 12/21/2015 03:24 PM, Busch, Keith wrote:
>> Since early in the 4.3-rcx series, my Dell Latitude D600 with a 32-bit kernel
>> fails to boot. Unfortunately, I did not discover this until late in the 4.4-rcX
>> cycle. The symptom is that the kernel echos the "Loading initial ramdisk ..."
>> message and then hangs.
>>
>> The problem was bisected to commit 4d6b4es" x86/PCI/ACPI: Use common interface
>> to support PCI host bridge" from Jiang Liu <[email protected]>. The
>> bisection was tested by reverting the commit in question. I have not yet tried
>> to analyze the patch to see where it might be going wrong.
>>
>> Please suggest any trial prints to debug this issue.
>
> Does this fix help you? It is included with the most recent 4.4 rc.
>
> https://git.kernel.org/cgit/linux/kernel/git/torvalds/linux.git/commit/?id=727ae8be30b428082d3519817f4fb98b712d457d

Yes, that patch does fix my problem even though the commit message says it is
needed for legacy AMD platforms. My laptop has a Celeron CPU, but the same
considerations must apply.

Thanks,

Larry