Am 12. November 2025 16:27:50 MEZ schrieb Tom Rini <[email protected]>: >On Wed, Nov 12, 2025 at 04:24:34PM +0100, Marek Vasut wrote: >> On 11/12/25 2:35 PM, Tom Rini wrote: >> > On Wed, Nov 12, 2025 at 03:02:21AM +0100, Marek Vasut wrote: >> > > The 'extension' test would set 'extension_overlay_addr' variable to >> > > decimal 4096 due to conversion in python. The 'extension_overlay_addr' >> > > is however sampled using env_get_hex("extension_overlay_addr", 0); >> > > which converts the 4096 to 0x4096 and uses that as DT overlay address, >> > > which is unaligned. Fix this by setting extension_overlay_addr to 0x1000 >> > > as intended, which is aligned.
Please, do not use fixed address 0x1000 which may not exist on real hardware. $kernel_addr_r would be more versatile. Best regards Heinrich >> > > >> > > Signed-off-by: Marek Vasut <[email protected]> >> > > --- >> > > Cc: Heinrich Schuchardt <[email protected]> >> > > Cc: Mattijs Korpershoek <[email protected]> >> > > Cc: Simon Glass <[email protected]> >> > > Cc: Tom Rini <[email protected]> >> > > Cc: [email protected] >> > > --- >> > > test/py/tests/test_extension.py | 2 +- >> > > 1 file changed, 1 insertion(+), 1 deletion(-) >> > > >> > > diff --git a/test/py/tests/test_extension.py >> > > b/test/py/tests/test_extension.py >> > > index 61223496054..ddc62e7d5ca 100644 >> > > --- a/test/py/tests/test_extension.py >> > > +++ b/test/py/tests/test_extension.py >> > > @@ -38,7 +38,7 @@ def test_extension(ubman): >> > > assert('overlay1.dtbo' in output) >> > > ubman.run_command_list([ >> > > - 'setenv extension_overlay_addr %s' % (overlay_addr), >> > > + 'setenv extension_overlay_addr 0x%x' % (overlay_addr), >> > > 'setenv extension_overlay_cmd \'host load hostfs - >> > > ${extension_overlay_addr} %s${extension_overlay_name}\'' % >> > > (os.path.join(ubman.config.build_dir, OVERLAY_DIR))]) >> > > output = ubman.run_command('extension apply 0') >> > >> > As we often get patches to remove the "0x" portion, can you just correct >> > the format character? Thanks! >> Do you mean s@0x%x@%x@ in this patch ? > >Yes, thanks. >
