From mboxrd@z Thu Jan 1 00:00:00 1970 From: Heiko Stuebner Date: Tue, 11 Dec 2018 07:25:07 +0100 Subject: [U-Boot] ARM: rockchip: rv1108: Fix booting with initramfs In-Reply-To: References: <20181204134926.28794-1-otavio@ossystems.com.br> <20181210005657.GG8702@bill-the-cat> Message-ID: <3396602.VQmpVtay8p@phil> List-Id: MIME-Version: 1.0 Content-Type: text/plain; charset="utf-8" Content-Transfer-Encoding: 8bit To: u-boot@lists.denx.de Hi Philipp, Am Dienstag, 11. Dezember 2018, 00:12:40 CET schrieb Philipp Tomsich: > + Heiko > > > On 10.12.2018, at 01:56, Tom Rini wrote: > > > > On Mon, Dec 10, 2018 at 01:38:36AM +0100, Philipp Tomsich wrote: > >> Tom, > >> > >> On 10.12.2018, at 01:28, Tom Rini wrote: > >>> > >>> On Mon, Dec 10, 2018 at 01:01:52AM +0100, Philipp Tomsich wrote: > >>>>> We move the ramdisk_addr_r to 0x68000000 and disable the initrd and > >>>>> fdt relocation, so the initramfs works out of box. > >>>>> > >>>>> Signed-off-by: Otavio Salvador > >>>>> Reviewed-by: Philipp Tomsich > >>>>> --- > >>>>> > >>>>> include/configs/rv1108_common.h | 4 +++- > >>>>> 1 file changed, 3 insertions(+), 1 deletion(-) > >>>>> > >>>> > >>>> Applied to u-boot-rockchip, thanks! > >>> > >>> Ugh, sorry for not spotting this sooner. Please don't disable > >>> fdt/initrd relocation and instead use bootm_size. > >> > >> Thanks for spotting this (just in time). > >> I’ll drop it and rerun Travis for tomorrow’s PR. > > > > Thanks. And you might want to audit the rest of rockchip (and no, my > > own house isn't 100% in order) as it looks like in general at least > > you're using fdt_high to (good) set an upper bound but I think > > bootm_size is more robust as we can set that and it covers fdt and > > initrd if present (which not all rk3xxx_common.h are setting). > > We may have a problem on rk3188 and rk3288 configurations. While I can > convert these to bootm_size, I would have to do so blindly, as I don’t have > any boards. > > @Heiko: you have added the fdt_high and initrd_high in rk3188_common.h, > do you have more info and would you have able to test this changed to > bootm_size instead? I think I more or less only copied that from rk3288 (adjusted for actual ram-location) and I'm of course able to test changes you want to do on rk3188 [Though only starting from thursday]. Heiko