board_get_usable_ram_top() was designed for getting the top most location
for U-Boot allocation that's why function itself supports via total_size
parameter to find out where the right location for U-Boot is.
But function itself is also reused by different (EFI) which is passing
total_size as 0 to find out where the usable ram top is. For this case
doesn't make sense (a waste time) to call any lmb functions.
That's why simply return gd->ram_top.
And gd->ram_top is filled already based on previous call for U-Boot iself.
The same solution is also used by stm32mp by commit
92b611e8b003 ("stm32mp:
correctly handle board_get_usable_ram_top(0)") and commit
c8510e397fad
("stm32mp: Fix board_get_usable_ram_top()").
Signed-off-by: Michal Simek <michal.simek@amd.com>
Link: https://lore.kernel.org/r/44470619e74f3e480b70deac24578e3e0d5c907e.1651225945.git.michal.simek@amd.com
phys_addr_t reg;
struct lmb lmb;
+ if (!total_size)
+ return gd->ram_top;
+
/* found enough not-reserved memory to relocated U-Boot */
lmb_init(&lmb);
lmb_add(&lmb, gd->ram_base, gd->ram_size);
phys_addr_t reg;
struct lmb lmb;
+ if (!total_size)
+ return gd->ram_top;
+
if (!IS_ALIGNED((ulong)gd->fdt_blob, 0x8))
panic("Not 64bit aligned DT location: %p\n", gd->fdt_blob);