[lkml]   [2006]   [Jan]   [3]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
Messages in this thread
SubjectRe: [PATCH 8/11] Time: i386 Conversion - part 4: ACPI PM variable renaming and config change.
john stultz <> wrote:
> Andrew, All,
> The conversion of i386 to use the generic timeofday subsystem
> has been split into 6 parts. This patch, the fourth of six, renames
> some ACPI PM variables and removes the CONFIG_X86_PM_TIMER option.
> It applies on top of my timeofday-arch-i386-part3 patch. This patch is
> part the timeofday-arch-i386 patchset, so without the following parts
> it is not expected to compile.
> Andrew, please consider for inclusion into your tree.


- The changelogs aren't terribly good.

When preparing changelogs, please think of how they will look in the
final git record. This means that metainfo concerning preceding patch
dependencies, the number of patches in the series, etc is irrelevant.
The sequence numbering in the Subject: (which you have done correctly) is

Text such as "please apply", while nice, just has to be removed by
myself. I'm not offended if it doesn't get sent ;)

Alert readers will note that I also strip out text such as "this patch
does X" from the changelogs. Because such text is redundant once the
patch is merged up. We _know_ it's a patch - otherwise it wouldn't be in
the git tree.

And once we strip out the above irrelevencies, your changelogs are
awfully spartan for such a substantial piece of work. Isn't there more
to be said?

- The fact that the kernel won't compile without the succeeding patches
in the series is unfortunate. If someone is trying to locate an
unrelated regression via `git bisect' and the bisection process happens
to land them at one of your it-wont-compile points then they have a big
problem. It can pretty much screw up the whole bisection process.

So please try to ensure that the kernel will compile and probably run
at all points of a patch series. If that's too hard then just fold the
separate patches into one larger patch (with appropriately expanded
changelog) to achieve the desired result.

Anyway, I'll tenatively merge these patches into next -mm so they can get a
bit of testing. Which causes a problem, because you don't then have a tree
against which to raise a new patch series.

So perhaps it would be best if you were to

a) Tell me which patches to fold into which other patches to generate a
series which compiles at every stage and

b) Send me a new set of changelogs for the resulting patch series.

To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to
More majordomo info at
Please read the FAQ at

 \ /
  Last update: 2006-01-04 01:39    [W:0.102 / U:2.768 seconds]
©2003-2018 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site