CentOS - Nov 2016 - linux-3.10.0-514.el7.tar.xz - process of creation

I try to create own patch for kernel, and I look in process of
generation binary rpm file.
As I can see, for creation binary package, rpmbuild doesn't uses any set
of patches and doesn't uses file of original (vanilla) version of
kernel. Instead it, I see some xz archive of compressed tree of kernel
probably with all changes (linux-3.10.0-514.el7.tar.xz). How and where I
can find process of creation of this archive and to see separately all
set of patches? I want to remove some changes and add own.

Many thanks to all, who have such info and can give link on process or
link on documentations of process of creation of this archive.

best regards, Andrei

upwork.link

On Fri, Nov 25, 2016 at 6:03 AM, Staazul <info at upwork.link>
wrote:
> I try to create own patch for kernel, and I look in process of
> generation binary rpm file.
> As I can see, for creation binary package, rpmbuild doesn't uses any
set
> of patches and doesn't uses file of original (vanilla) version of
> kernel. Instead it, I see some xz archive of compressed tree of kernel
> probably with all changes (linux-3.10.0-514.el7.tar.xz). How and where I
> can find process of creation of this archive and to see separately all
> set of patches? I want to remove some changes and add own.
>
> Many thanks to all, who have such info and can give link on process or
> link on documentations of process of creation of this archive.
The individual kernel patches are available to RHEL subscribers. No
public access though.

Do you really need to see them to create your patches? Can't you just
look at the code and figure out the changes you'd like to make?

Akemi

Am 25.11.2016 um 15:03 schrieb Staazul:
> I try to create own patch for kernel, and I look in process of
> generation binary rpm file.
> As I can see, for creation binary package, rpmbuild doesn't uses any
set
> of patches and doesn't uses file of original (vanilla) version of
> kernel. Instead it, I see some xz archive of compressed tree of kernel
> probably with all changes (linux-3.10.0-514.el7.tar.xz). How and where I
> can find process of creation of this archive and to see separately all
> set of patches? I want to remove some changes and add own.
>
> Many thanks to all, who have such info and can give link on process or
> link on documentations of process of creation of this archive.
>
> best regards, Andrei
You cannot easily bisect the kernel tarball provided. Some time ago Red 
Hat had dediced to use this way of prividing the sources of their kernel 
because well knowns had started an aggressive business to hijack RHEL 
customers.

See i.e. 
https://linux.slashdot.org/story/11/03/04/1550231/red-hat-stops-shipping-kernel-changes-as-patches

Alexander

On Fri, Nov 25, 2016 at 9:13 AM, Alexander Dalloz <ad+lists at uni-x.org>
wrote:
> Am 25.11.2016 um 15:03 schrieb Staazul:
>>
>> I try to create own patch for kernel, and I look in process of
>> generation binary rpm file.
>> As I can see, for creation binary package, rpmbuild doesn't uses
any set
>> of patches and doesn't uses file of original (vanilla) version of
>> kernel. Instead it, I see some xz archive of compressed tree of kernel
>> probably with all changes (linux-3.10.0-514.el7.tar.xz). How and where
I
>> can find process of creation of this archive and to see separately all
>> set of patches? I want to remove some changes and add own.
>>
>> Many thanks to all, who have such info and can give link on process or
>> link on documentations of process of creation of this archive.
>>
>> best regards, Andrei
>
>
> You cannot easily bisect the kernel tarball provided. Some time ago Red Hat
> had dediced to use this way of prividing the sources of their kernel
because
> well knowns had started an aggressive business to hijack RHEL customers.
>
> See i.e.
>
https://linux.slashdot.org/story/11/03/04/1550231/red-hat-stops-shipping-kernel-changes-as-patches
>
> Alexander
That article is not quite accurate. As I noted in my earlier post, the
patches are available. As a RHEL subscriber, I can see and search
them:

https://access.redhat.com/articles/1130403

I would be surprised if no one at Oracle can afford to pay for an
entitlement. ;)

Akemi