Hi Jonas,
On Thu, 26 Sept 2024 at 18:13, Jonas Karlman jonas@kwiboo.se wrote:
Hi Simon,
On 2024-09-27 00:35, Simon Glass wrote:
Now that SPL means SPL (only) and is not defined for TPL, simplify some checks for SPL.
Signed-off-by: Simon Glass sjg@chromium.org
arch/arm/include/asm/arch-rockchip/boot0.h | 3 +-- arch/arm/mach-rockchip/rk3399/rk3399.c | 2 +- drivers/clk/rockchip/clk_rk3368.c | 8 ++++---- drivers/ram/rockchip/sdram_rk3288.c | 10 +++++----- drivers/ram/rockchip/sdram_rk3399.c | 2 +- drivers/ram/rockchip/sdram_rv1126.c | 9 +++------ 6 files changed, 15 insertions(+), 19 deletions(-)
diff --git a/arch/arm/include/asm/arch-rockchip/boot0.h b/arch/arm/include/asm/arch-rockchip/boot0.h index edb2a31c348..918f0495fa0 100644 --- a/arch/arm/include/asm/arch-rockchip/boot0.h +++ b/arch/arm/include/asm/arch-rockchip/boot0.h @@ -54,7 +54,6 @@ _start: ARM_VECTORS #endif
-#if !defined(CONFIG_TPL_BUILD) && defined(CONFIG_XPL_BUILD) && \
(CONFIG_ROCKCHIP_SPL_RESERVE_IRAM > 0)+#if defined(CONFIG_SPL_BUILD) && (CONFIG_ROCKCHIP_SPL_RESERVE_IRAM > 0)
This was changed from SPL to XPL just to be changed back to SPL in this patch?, such change probably just clutter git blame, was the change to use XPL really necessary in the first place?
Because SPL_BUILD was used together with !TPL_BUILD it is pretty obvious that SPL_BUILD did not need to change to XPL_BUILD in the first place.
Yes this is a bit of a corner case.
For the first patch, CONFIG_SPL_BUILD is defined for any 'SPL' build. For the second it is defined only for SPL. So in this case, yes, it doesn't matter. But across the whole of U-Boot, it does.
I got pretty deep down the rabbit hole on this one, and perhaps should not have gone ahead with the 'clean-up'. So I think we can just drop this patch. I'll look at what others might be dropped too.
Regards, Simon