From mboxrd@z Thu Jan  1 00:00:00 1970
From: Marek Vasut <marex@denx.de>
Date: Fri, 19 Apr 2013 19:04:28 +0200
Subject: [U-Boot] [PATCH 3/6] arm: imx: Fix u-boot-with-nand-spl.imx
	target
In-Reply-To: <51713096.5090302@ti.com>
References: <1366344655-8535-1-git-send-email-marex@denx.de>
	<201304191351.04346.marex@denx.de> <51713096.5090302@ti.com>
Message-ID: <201304191904.28927.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 Tom Rini,

> On 04/19/2013 07:51 AM, Marek Vasut wrote:
> > Dear Beno?t Th?baudeau,
> > 
> >> Dear Marek Vasut,
> >> 
> >> On Friday, April 19, 2013 1:16:31 PM, Marek Vasut wrote:
> >>> Dear Beno?t Th?baudeau,
> >>> 
> >>>> Dear Marek Vasut,
> >>>> 
> >>>> On Friday, April 19, 2013 6:10:52 AM, Marek Vasut wrote:
> >>>>> This target is currently concatenating u-boot SPL in
> >>>>> imximage format with u-boot.bin. The NAND SPL can load a
> >>>>> raw binary, but the preffered format with much less
> >>>>> limitations is uImage format. Fix the target so u-boot.bin
> >>>>> is first converted into uImage format and only after that
> >>>>> is concatenated.
> >>>> 
> >>>> Please reword: This is not a fix, but an enhancement.
> >>>> 
> >>>>> Signed-off-by: Marek Vasut <marex@denx.de> Cc: Albert
> >>>>> ARIBAUD <albert.u.boot@aribaud.net> Cc: Beno?t Th?baudeau
> >>>>> <benoit.thebaudeau@advansee.com> Cc: Fabio Estevam
> >>>>> <fabio.estevam@freescale.com> Cc: Scott Wood
> >>>>> <scottwood@freescale.com> Cc: Stefano Babic
> >>>>> <sbabic@denx.de> Cc: Tom Rini <trini@ti.com> ---
> >>>>> 
> >>>>> arch/arm/imx-common/Makefile | 7 +++++-- 1 file changed, 5
> >>>>> insertions(+), 2 deletions(-)
> >>>>> 
> >>>>> diff --git a/arch/arm/imx-common/Makefile
> >>>>> b/arch/arm/imx-common/Makefile index 44b6822..ba31d3e
> >>>>> 100644 --- a/arch/arm/imx-common/Makefile +++
> >>>>> b/arch/arm/imx-common/Makefile @@ -69,8 +69,11 @@
> >>>>> $(OBJTREE)/u-boot-with-nand-spl.imx: $(OBJTREE)/SPL
> >>>>> $(OBJTREE)/u-boot.bin
> >>>>> 
> >>>>> -I binary -O binary $(OBJTREE)/spl/u-boot-nand-spl.imx \
> >>>>> $(OBJTREE)/spl/u-boot-nand-spl-pad.imx  rm
> >>>>> $(OBJTREE)/spl/u-boot-nand-spl.imx
> >>>>> 
> >>>>> -	cat $(OBJTREE)/spl/u-boot-nand-spl-pad.imx
> >>>>> $(OBJTREE)/u-boot.bin
> >>>>> 
> >>>>> $@ -	rm $(OBJTREE)/spl/u-boot-nand-spl-pad.imx +
> >>>>> $(OBJTREE)/tools/mkimage -A arm -O U-Boot -a
> >>>>> $(CONFIG_SYS_TEXT_BASE) \ +		-e $(CONFIG_SYS_TEXT_BASE) -C
> > 
> > none -d
> > 
> >>>>> $(OBJTREE)/u-boot.bin \ +		$(OBJTREE)/u-boot.uim +	cat
> >>>>> $(OBJTREE)/spl/u-boot-nand-spl-pad.imx
> >>>>> $(OBJTREE)/u-boot.uim
> >>>>> 
> >>>>> $@ +	rm $(OBJTREE)/spl/u-boot-nand-spl-pad.imx
> >>>>> $(OBJTREE)/u-boot.uim
> >>>>> 
> >>>>> ####################################################################
> 
> #### #
> 
> >>>>> -- 1.7.11.7
> >>>> 
> >>>> This change is correct.
> >>>> 
> >>>> I don't know the uImage format very well. Can you give
> >>>> examples of what this patch can bring compared to the raw
> >>>> binary format?
> >>>> 
> >>>> If this change is useful, it should probably be made for
> >>>> u-boot-with-spl.imx too.
> >>> 
> >>> Please see common/spl/spl_nand.c which calls
> >>> spl_parse_image_header() on the payload loaded from NAND. This
> >>> is implemented in common/spl/spl.c and if proper header is not
> >>> found, the payload is assumed to be of 200kB of size, which is
> >>> far less than fully configured U-Boot.
> >> 
> >> Or you could #define CONFIG_SYS_MONITOR_LEN to what you need.
> > 
> > YUCK!
> > 
> >>> So using the uimage format is much more flexible.
> >> 
> >> I agree.
> >> 
> >> So I'd see the following changes for this patch: - Reword "fix"
> >> to show enhancement, and detail the flexibility rationale in the
> >> description. - Extend to u-boot-with-spl.imx.
> > 
> > Ok.
> > 
> > Tom, shall we merge this one into current release after it's
> > fixed?
> 
> This is an enhancement not a bug fix, right?  If so, next release.

Since the release is nigh, let's wrap it into the next one. But the usability of 
the spl is lower :(