Hi Stephen, On Mon, Jan 9, 2012 at 8:46 AM, Stephen Warren <swar...@nvidia.com> wrote: > On 01/06/2012 05:25 PM, Simon Glass wrote: >> Since Tegra executes early code on an ARMv4T (but the rest on ARMv7) >> we must mark the new board init files for compilation with ARMv4t >> architecture. >> >> Signed-off-by: Simon Glass <s...@chromium.org> > > This and patch 28 seem fine to me (but I didn't look at patches 1..26 to > see what the implications of patch 28 really is!) One question below:
A wise move. The intent is basically to create a generic board init file so that most of the code between start.S and main.c is the same for all architectures. > >> diff --git a/arch/arm/cpu/armv7/tegra2/config.mk >> b/arch/arm/cpu/armv7/tegra2/config.mk > ... >> ifdef CONFIG_TEGRA2 >> CFLAGS_arch/arm/lib/board.o += -march=armv4t >> +CFLAGS_common/board_f.o += -march=armv4t >> +CFLAGS_lib/initcall.o += -march=armv4t >> endif > > A pre-existing issue, but given this is a Tegra2-specific file, how come > those CFLAG assignments are ifdef'd on CONFIG_TEGRA2? I thought we had fixed this at some point but obviously not. Worth a patch at some point because it is not needed, as you say. Regards, Simon > > -- > nvpublic _______________________________________________ U-Boot mailing list U-Boot@lists.denx.de http://lists.denx.de/mailman/listinfo/u-boot