Percona Server with XtraDB

Remove all references to autogenerated files in all our patches

Reported by Alexey Kopytov on 2011-01-10
6
This bug affects 1 person
Affects Status Importance Assigned to Milestone
Percona Server
High
Laurynas Biveinis
5.1
High
Oleg Tsarev
5.5
High
Laurynas Biveinis
Rnt-5.1
Undecided
Oleg Tsarev

Bug Description

We should remove all changes in autogenerated files (such as 'configure', 'Makefile.in' and so on) from our patches. They cause a lot of headache when doing some mass editing of patches (for example, when cleaning up warnings/errors).

Stewart Smith (stewart) wrote :

I'm very surprised this hasn't horribly broken things.

Changed in percona-server:
status: New → Confirmed
importance: Undecided → High
Stewart Smith (stewart) on 2011-05-24
Changed in percona-server:
status: Confirmed → Triaged
Stewart Smith (stewart) wrote :

Is this currently actually 'In Progress'? if not, should just be Triaged

Affected patches in 5.1:

innodb_swap_builtin_plugin.patch (configure)
show_patches.patch (sql/Makefile.in)
userstat.patch (configure)
innodb_show_enhancements.patch (Makefile.in)
bugfix48929.patch (configure)
response-time-distribution.patch (sql/Makefile.in)

If the "configure.in" was patched, does the execute of "./configure" detect always the obsoleted the files and regenerate them automatically?
(I remember, at least in ancient, I met some dependency problem about regeneration the files (It might be MySQL's bug, which was already fixed))

Or we should document the procedure needed before "./configure"

I agree if no problem for "all environments", now.
(I mean, for example, older distribution might not have enough version of automake)

On Mon, 18 Jul 2011 08:57:57 -0000, Yasufumi Kinoshita <email address hidden> wrote:
> If the "configure.in" was patched, does the execute of "./configure" detect always the obsoleted the files and regenerate them automatically?
> (I remember, at least in ancient, I met some dependency problem about regeneration the files (It might be MySQL's bug, which was already fixed))
>
> Or we should document the procedure needed before "./configure"

With Jenkins we should be able to find out if it gets it wrong anywhere,
but I've found it usually works fine... otherwise we can add in
regeneration steps.
--
Stewart Smith

Some patches already do not include the generated files and thus correct re-generation is already required for correct builds.

5.5 is already clean (side effect of CMake).

In 5.1 config.h.in is also a generated file in bugfix48929.patch and userstat.patch.

To post a comment you must log in.
This report contains Public information  Edit
Everyone can see this information.

Other bug subscribers