From mboxrd@z Thu Jan  1 00:00:00 1970
From: Marek Vasut <marex@denx.de>
Date: Fri, 21 Sep 2012 14:51:33 +0200
Subject: [U-Boot] [PATCH 03/11] DM: add block controller core
In-Reply-To: <11832942.v7278ihCq2@bloomfield>
References: <1348169867-2917-1-git-send-email-morpheus.ibis@gmail.com>
	<201209202205.36972.marex@denx.de> <11832942.v7278ihCq2@bloomfield>
Message-ID: <201209211451.33805.marex@denx.de>
List-Id: <u-boot.lists.denx.de>
MIME-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Content-Transfer-Encoding: 7bit
To: u-boot@lists.denx.de

Dear Pavel Herrmann,

> On Thursday 20 of September 2012 22:05:36 Marek Vasut wrote:
> > Dear Pavel Herrmann,
> > 
> > > This core provides unified access to different block controllers (SATA,
> > > SCSI).
> > 
> > Description of the patch missing or is sub-par. You should work on this
> > skill.
> > 
> > > Signed-off-by: Pavel Herrmann <morpheus.ibis@gmail.com>
> > > ---
> > > 
> > >  Makefile                   |   1 +
> > >  drivers/blockctrl/Makefile |  42 ++++++
> > >  drivers/blockctrl/core.c   | 349
> > > 
> > > +++++++++++++++++++++++++++++++++++++++++++++ include/dm/blockctrl.h   
> > >  |
> > > 
> > >  75 ++++++++++
> > >  4 files changed, 467 insertions(+)
> > >  create mode 100644 drivers/blockctrl/Makefile
> > >  create mode 100644 drivers/blockctrl/core.c
> > >  create mode 100644 include/dm/blockctrl.h
> > > 
> > > diff --git a/Makefile b/Makefile
> > > index e43fd9d..4420484 100644
> > > --- a/Makefile
> > > +++ b/Makefile
> > > @@ -304,6 +304,7 @@ LIBS-y += test/libtest.o
> > > 
> > >  LIBS-$(CONFIG_DM) += common/dm/libdm.o
> > >  LIBS-$(CONFIG_DM) += drivers/demo/libdemo.o
> > >  LIBS-${CONFIG_DM_BLOCK} += drivers/blockdev/libblockdev.o
> > > 
> > > +LIBS-${CONFIG_DM_BLOCK} += drivers/blockctrl/libblockctrl.o
> > 
> > ${} ? What is this ?

Why not just reuse drivers/block and in drivers/block compile in the libblock.o 
so you don't polute the top-level makefile ? Easy as that.

> > [..]
> > 
> > This handles SCSI? Sata ? what ?
> > 
> > Should this not be called scsi_core ? sata_core ? What did the previous
> > core do? sata?  scsi? block? I'm lost.
> 
> the previous core handled disks (and cards and stuff) and partitions (think
> /dev/sdxy), and was largely a replacement of /disk
> this core handles any interface those disks are connected to (SATA, PATA,
> SCSI), and should replace /drivers/block

Why is this not in the commit message then ? I have a proposal, before you 
submit a patchset, prepare it, work on something else for a bit, then read again 
the commit message only and see if you still understand what it means.

Am I correct that this will look as such:
user -> [ 01/11 ] -> [ 03/11 or something else ] -> [ if 03/11, then disc ]

> > I stop here, I don't know what this is all about, sorry.
> > 
> > Best regards,
> > Marek Vasut

Best regards,
Marek Vasut