Hi Tom, On Mon, 3 Feb 2025 at 19:12, Tom Rini <tr...@konsulko.com> wrote: > > On Mon, Feb 03, 2025 at 05:38:44PM -0700, Simon Glass wrote: > > Hi Tom, > > > > On Mon, 3 Feb 2025 at 13:09, Tom Rini <tr...@konsulko.com> wrote: > > > > > > On Mon, Feb 03, 2025 at 10:42:08AM -0700, Simon Glass wrote: > > > > > > > The app does not have this symbol. Also the memory where the app is > > > > loaded is not under U-Boot's control. Disable this reservation for > > > > the EFI app. > > > > > > > > Signed-off-by: Simon Glass <s...@chromium.org> > > > > --- > > > > > > > > lib/lmb.c | 3 ++- > > > > 1 file changed, 2 insertions(+), 1 deletion(-) > > > > > > > > diff --git a/lib/lmb.c b/lib/lmb.c > > > > index 588787d2a90..420eae4d0b8 100644 > > > > --- a/lib/lmb.c > > > > +++ b/lib/lmb.c > > > > @@ -560,7 +560,8 @@ static void lmb_reserve_uboot_region(void) > > > > lmb_reserve_flags(rsv_start, bank_end - rsv_start + 1, > > > > LMB_NOOVERWRITE); > > > > > > > > - if (gd->flags & GD_FLG_SKIP_RELOC) > > > > + if (!IS_ENABLED(CONFIG_EFI_APP) && > > > > + (gd->flags & GD_FLG_SKIP_RELOC)) > > > > lmb_reserve_flags((phys_addr_t)(uintptr_t)_start, > > > > gd->mon_len, LMB_NOOVERWRITE); > > > > > > What prevents U-Boot from overwriting itself, with U-Boot commands? Also > > > how is this linking today? I see lmb_reserve_uboot_region in > > > efi-x86_app64 for example. > > > > The only memory U-Boot can use when running as an EFI app is that > > which it has allocated using EFI calls. The actual U-Boot code is > > certainly in memory somewhere, but does not lie between gd->ram_start > > and (gd->ram_start + gd->ram_size) > > That's not what the commit message means to me. But the next question > is, is EFI_APP setting SKIP_RELOC?
Yes, there is no sense in relocating from one alloced region to another. U-Boot is not able to relocate to the top of memory, nor decide what other parts of memory are used for. This is in the nature of EFI, after all. We have an API, but not policy. Regards, Simon
