Hi Moritz, On Sat, 5 Feb 2022 at 13:17, Moritz Fischer <mori...@google.com> wrote: > > Commit d953137526cc ("x86: Move SSDT table to a writer function") > introduced a bug where the actual MCFG entries are no longer generated. > > Cc: Simon Glass <s...@chromium.org> > Fixes: d953137526cc ("x86: Move SSDT table to a writer function") > Signed-off-by: Moritz Fischer <mori...@google.com> > --- > > Hi Simon, Andriy, > > it looks like this got dropped when moving stuff around, I don't have HW > to test this, but it seemed off. > > - Moritz > > --- > > arch/x86/lib/acpi_table.c | 2 ++ > 1 file changed, 2 insertions(+)
Reviewed-by: Simon Glass <s...@chromium.org> Tested on: coral Tested-by: Simon Glass <s...@chromium.org> > > diff --git a/arch/x86/lib/acpi_table.c b/arch/x86/lib/acpi_table.c > index c0534343f1..753bf39619 100644 > --- a/arch/x86/lib/acpi_table.c > +++ b/arch/x86/lib/acpi_table.c > @@ -499,6 +499,8 @@ int acpi_write_mcfg(struct acpi_ctx *ctx, const struct > acpi_writer *entry) > header->length = sizeof(struct acpi_mcfg); > header->revision = 1; > > + current = acpi_fill_mcfg(current); > + > /* (Re)calculate length and checksum */ > header->length = current - (u32)mcfg; > header->checksum = table_compute_checksum(mcfg, header->length); > -- > 2.35.0.263.gb82422642f-goog >