Thanks Markus, On 9/24/2026 12:40 AM, Markus Schneider-Pargmann wrote: > On HS-SE platforms TIFS takes longer to initialize. TIFS only removes > the firewall on the registers at the end of initialization. On HS-SE the > u-boot ctrl_mmr_unlock() can happen before TIFS removed the firewall > leading to a race condition. This leads to a broken watchdog since the > signal configuration doesn't succeed because the write doesn't take > effect. > > Add a get_ti_sci_handle() call before ctrl_mmr_unlock(). > get_ti_sci_handle() leads to fetching the TIFS version which only comes > back after TIFS initialization is done. If a response is not received > get_ti_sci_handle() returns after a timeout and continues as it would > have done without this patch. get_ti_sci_handle() has to happen after > spl_early_init() as it requires DM to be setup. > > Link: > https://lore.kernel.org/u-boot/[email protected]/ > Link: https://lore.kernel.org/u-boot/[email protected]/
Here tom advice to include all platforms https://lore.kernel.org/u-boot/20260226143756.GC1593142@bill-the-cat/ > Signed-off-by: Markus Schneider-Pargmann <[email protected]> > --- > arch/arm/mach-k3/am62x/am625_init.c | 13 +++++++++++-- > 1 file changed, 11 insertions(+), 2 deletions(-) > > [..]
