From mboxrd@z Thu Jan 1 00:00:00 1970 From: Vikas MANOCHA Date: Wed, 12 Apr 2017 19:47:09 +0000 Subject: [U-Boot] [PATCH] spl: make image arg or fdt blob address reconfigurable In-Reply-To: <20170412133322.GY19897@bill-the-cat> References: <1491604694-20521-1-git-send-email-vikas.manocha@st.com> <20170410122919.GA19897@bill-the-cat> <20170412133322.GY19897@bill-the-cat> Message-ID: List-Id: MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit To: u-boot@lists.denx.de Hi Tom, > -----Original Message----- > From: Tom Rini [mailto:trini at konsulko.com] > Sent: Wednesday, April 12, 2017 6:33 AM > To: Vikas MANOCHA > Cc: u-boot at lists.denx.de; Marek Vasut ; Stefan Agner ; Jeremy Hunt > > Subject: Re: [U-Boot] [PATCH] spl: make image arg or fdt blob address reconfigurable > > On Tue, Apr 11, 2017 at 11:44:00PM +0000, Vikas MANOCHA wrote: > > Hi Tom, > > > > > -----Original Message----- > > > From: Tom Rini [mailto:trini at konsulko.com] > > > Sent: Monday, April 10, 2017 5:29 AM > > > To: Vikas MANOCHA > > > Cc: u-boot at lists.denx.de; Marek Vasut ; Stefan Agner > > > ; Jeremy Hunt > > > > > > Subject: Re: [U-Boot] [PATCH] spl: make image arg or fdt blob > > > address reconfigurable > > > > > > On Fri, Apr 07, 2017 at 03:38:13PM -0700, Vikas Manocha wrote: > > > > > > > At present fdt blob or argument address being passed to kernel is > > > > fixed at compile time using macro CONFIG_SYS_SPL_ARGS_ADDR. FDT > > > > blob from different media like nand, nor flash are copied to the > > > > address pointed by the macro. > > > > The problem is, it makes args/fdt blob compulsory to copy which is > > > > not required in cases like for NOR Flash. This patch removes this limitation. > > > > > > > > Signed-off-by: Vikas Manocha > > > > --- > > > > arch/arm/lib/spl.c | 7 +++---- > > > > arch/microblaze/cpu/spl.c | 6 +++--- > > > > arch/powerpc/lib/spl.c | 8 ++++---- > > > > common/spl/spl.c | 6 ++++-- > > > > common/spl/spl_nor.c | 8 +------- > > > > include/spl.h | 5 ++--- > > > > > > I assume you've tested the spl_nor case afterwards, yes? Did this > > > result in some measurable boot time decrease? Thanks! > > > > Yes, I tested it's working on board. Not sure how to measure the impact on boot time. > > There's always good old grabserial. But if you didn't measure bootspeed, did this decrease the code size? Or fix some other issue? Thanks Tom for the suggestion. The benefit of this patch is : It removes copying FDT blob from NOR flash to ram in case of booting from nor flash, text size reduction is only 4 Bytes, no change in data/bss size. It might reduce the boot time but boot time impact would depend on the FDT blob copying time & ram vs flash read speed. Also it provides a way to change fdt blob address at run-time which might be useful e.g. choosing between two FDT blobs or manipulating FDT address. Cheers, Vikas > Thanks! > > -- > Tom