From mboxrd@z Thu Jan 1 00:00:00 1970 From: Tom Rini Date: Wed, 22 Aug 2012 16:14:41 -0700 Subject: [U-Boot] [PATCH v2 15/18] ARM: SPL: Convert davinci to CONFIG_SPL_FRAMEWORK In-Reply-To: <20120822191903.GD29180@bill-the-cat> References: <1345481154-7050-1-git-send-email-trini@ti.com> <1345481154-7050-16-git-send-email-trini@ti.com> <20120822174619.GC29180@bill-the-cat> <20120822191903.GD29180@bill-the-cat> Message-ID: <20120822231441.GA3476@bill-the-cat> List-Id: MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit To: u-boot@lists.denx.de On Wed, Aug 22, 2012 at 12:19:03PM -0700, Tom Rini wrote: > On Wed, Aug 22, 2012 at 10:46:19AM -0700, Tom Rini wrote: > > On Wed, Aug 22, 2012 at 12:04:27PM +0200, Christian Riesch wrote: > > > Hi Tom, > > > > > > On Mon, Aug 20, 2012 at 6:45 PM, Tom Rini wrote: > > > > - Convert the non-relocation part of board_init_f to spl_board_init, > > > > turn on CONFIG_SPL_BOARD_INIT in the configs. > > > > - Remove duplicated code. > > > > - Add spl_boot_device() that returns the statically chosen boot device. > > > > > > I tested the patchset on the LogicPD AM1808 experimenter's kit: > > > > > > make da850evm > > > and flashing to SPI flash with > > > mono sfh_OMAP-L138.exe -targetType AM1808 -p /dev/ttyUSB0 -flash_noubl > > > u-boot.ais > > > > Confirmed, and I'll update the README to say that is how you can write > > the u-boot.ais in recovery mode. > > > > > The board boots up fine, but after pressing the reset button or > > > issuing the u-boot "reset" command, it is dead. Only power-cycling > > > helps. Before enabling the SPL framework in this patch reset works > > > fine. Any ideas? > > > > OK, I don't like the SPL malloc fix from before and I think it just > > masked, partially, a real problem. The SPI code did not change in the > > OK, I get what that fixes now, oops. But still, something is going on > here, debugging more. OK, got it, and a v3 shortly. In short, using CONFIG_SPL_BOARD_INIT as the hook for doing our low-level init isn't completely sufficient, so I've reworked things a little more and tossed in another hook point. Now I'm also seeing the initial console output I had been missing. -- Tom