On Sat, May 30, 2026 at 04:22:56PM +0200, Marek Vasut wrote:
> On 5/30/26 3:44 PM, Peter Robinson wrote:
>
> > > The revert is wrong as that breaks other things , like use of DRAM above
> > > 4 GiB boundary, which is exactly why this change was implemented. If RPi
> > > has limitations, then those limitations should be imposed on RPi, not
> > > globally, so add the LMB reservations on RPi.
> >
> > I have no idea how LMB works so please send a patch that does that.
>
> Try this:
>
> // Enable CONFIG_LMB_ARCH_MEM_MAP=y
>
> #include <lmb.h>
>
> void lmb_arch_add_memory(void)
> {
> int i;
> phys_addr_t bank_end;
> phys_size_t size;
> u64 ram_top = gd->ram_top;
> struct bd_info *bd = gd->bd;
>
> /* Assume a 4GB ram_top if not defined */
> if (!ram_top)
> ram_top = 0x100000000ULL;
>
> for (i = 0; i < CONFIG_NR_DRAM_BANKS; i++) {
> size = bd->bi_dram[i].size;
> bank_end = bd->bi_dram[i].start + size;
>
> if (size) {
> /*
> * Reserve memory above ram_top as
> * no-overwrite so that it cannot be
> * allocated
> */
> if (bd->bi_dram[i].start >= ram_top)
> lmb_reserve(bd->bi_dram[i].start, size,
> LMB_NOOVERWRITE);
> else if (bank_end > ram_top)
> lmb_reserve(ram_top, bank_end - ram_top,
> LMB_NOOVERWRITE);
> }
> }
> }I see our last emails crossed, thanks for posting this to help figure out what's going, Marek. -- Tom
signature.asc
Description: PGP signature
