Hi,
2015-05-23 07:46, Or Gerlitz:
> On Fri, May 15, 2015 at 6:56 PM, Bruce Richardson
> <bruce.richardson at intel.com> wrote:
> > move mlx4 PMD to drivers/net directory
> >
> > Signed-off-by: Bruce Richardson <bruce.richardson at intel.com>
> > ---
> > drivers/net/Makefile | 2 +-
> > drivers/net/mlx4/Makefile | 121 +
> > drivers/net/mlx4/mlx4.c | 4686
> > ++++++++++++++++++++++++++
> > drivers/net/mlx4/mlx4.h | 157 +
> > drivers/net/mlx4/rte_pmd_mlx4_version.map | 4 +
> > lib/Makefile | 1 -
> > lib/librte_pmd_mlx4/Makefile | 121 -
> > lib/librte_pmd_mlx4/mlx4.c | 4686
> > --------------------------
> > lib/librte_pmd_mlx4/mlx4.h | 157 -
> > lib/librte_pmd_mlx4/rte_pmd_mlx4_version.map | 4 -
> > 10 files changed, 4969 insertions(+), 4970 deletions(-)
> > create mode 100644 drivers/net/mlx4/Makefile
> > create mode 100644 drivers/net/mlx4/mlx4.c
> > create mode 100644 drivers/net/mlx4/mlx4.h
> > create mode 100644 drivers/net/mlx4/rte_pmd_mlx4_version.map
> > delete mode 100644 lib/librte_pmd_mlx4/Makefile
> > delete mode 100644 lib/librte_pmd_mlx4/mlx4.c
> > delete mode 100644 lib/librte_pmd_mlx4/mlx4.h
> > delete mode 100644 lib/librte_pmd_mlx4/rte_pmd_mlx4_version.map
>
> There's a way to do source movements with git such that history
> remains, see for example in the upstream kernel the way the ethernet
> drivers were moved from drivers/net to drivers/net/ethernet couple of
> years ago.
I don't see anything special here:
https://git.kernel.org/cgit/linux/kernel/git/torvalds/linux.git/commit/?id=5a2cc190eb3fe
It doesn't seem different of
http://dpdk.org/browse/dpdk/commit/?id=98a1f3776fc9a3de442
By the way, what do you mean by "history remains"?
Are you thinking about git log --follow which is known to be a hack?
http://permalink.gmane.org/gmane.comp.version-control.git/147089
My understanding is that file history doesn't fit well with git:
http://article.gmane.org/gmane.comp.version-control.git/217
> Please make sure to use that practice here too.
The change was applied before you sent this message:
http://dpdk.org/ml/archives/dev/2015-May/018064.html
But I'm still interested about best practices.
