Hi,
It took me quite some time to get v4.4.24 to successfully boot (this
was a 32 bits build on Fedora 24). Especially annoying was that it
failed silently on real hardware. Only after messing around with Qemu
it became clear what might be going on.
It turned out Fedora 24 ships binutils-2.26, and using that version of
binutils apparently requires commit 6d92bc9d483a ("x86/build: Build
compressed x86 kernels as PIE"). Does is makes sense to add that commit
to the v4.4.y stable tree?
Thanks,
Paul Bolle
On Fri, Oct 14, 2016 at 03:40:22PM +0200, Paul Bolle wrote:
> Hi,
>
> It took me quite some time to get v4.4.24 to successfully boot (this
> was a 32 bits build on Fedora 24). Especially annoying was that it
> failed silently on real hardware. Only after messing around with Qemu
> it became clear what might be going on.?
>
> It turned out Fedora 24 ships binutils-2.26, and using that version of
> binutils apparently requires commit?6d92bc9d483a ("x86/build: Build
> compressed x86 kernels as PIE"). Does is makes sense to add that commit
> to the v4.4.y stable tree?
Did you apply that patch and it worked properly for you? If so, yes,
I'll be glad to queue it up.
Also, stable questions/issues should be cc:ed to [email protected].
thanks,
greg k-h
On Fri, 2016-10-14 at 16:08 +0200, Greg KH wrote:
> Did you apply that patch and it worked properly for you? If so, yes,
> I'll be glad to queue it up.
Yes I did and it did.
But I do find this commit as magical as they can get so I'd hoped that
hjl would speak up. But if my word is enough for you, then feel free to
queue it up.
> Also, stable questions/issues should be cc:ed to [email protected].
Sorry about that. I've received one too many of your formletters, so
now I try to avoid [email protected] as much as I can.
Thanks,
Paul Bolle
On Fri, Oct 14, 2016 at 9:21 AM, Paul Bolle <[email protected]> wrote:
> On Fri, 2016-10-14 at 16:08 +0200, Greg KH wrote:
>> Did you apply that patch and it worked properly for you? If so, yes,
>> I'll be glad to queue it up.
>
> Yes I did and it did.
>
> But I do find this commit as magical as they can get so I'd hoped that
> hjl would speak up. But if my word is enough for you, then feel free to
> queue it up.
>
Yes, commit 6d92bc9d483a is needed to support x86 relocatable
compressed kernel with binutils 2.26.
--
H.J.