On 01/12/2017 09:59 PM, Joe Perches wrote:
On Thu, 2017-01-12 at 21:24 -0800, David VomLehn wrote:
On 01/12/2017 09:06 PM, Joe Perches wrote:
On Thu, 2017-01-12 at 21:02 -0800, Alexander Loktionov wrote:
From: David VomLehn <voml...@texas.net>

Patches to create the make and configuration files.
This patch should _really_ be the last in the series
not the first.

Could you explain the basis for this? By convention, we put tables of
content at the beginning of books and only indices at the back.
Analogously, make and config files can be used to established the
context for what follows, making it easier to understand. Once
committed, of course, the order no longer matters except as bisection is
concerned.
As I wrote the first time:

On Tue, 2016-12-27 at 08:15 -0800, Joe Perches wrote:
On Tue, 2016-12-27 at 05:17 -0800, David VomLehn wrote:
Patches to create the make and configuration files.
[]
Patch 1 will not build if CONFIG_AQTION is enabled.
Patch 1/12 should be reordered to be patch 12/12 and
all the other patches moved up appropriately.
You don't create the files until later patches.

If you applied just this first patch and tried to
add CONFIG_AQTION=y to the .config, make fails.

That's bad for git bisect.
Every patch in this series should build properly.

If you delay the adding of the Makefile and Kconfig
until all the files are added, then it'd bisect fine.
Please go back and re-read the latest patches; I think you will find your concern about CONFIG_AQTION addressed in the v5 patchset. The 01/13 patch no longer has the changes to Makefile and Kconfig in drivers/net/ethernet that will pull in the Makefile and Kconfig from drivers/net/ethernet/aquantia. Those changes are now in the 13/13 patch, which should make it bisectable. If I am missing something, please let me know.

--
David VL

Reply via email to