On Thu, Oct 12, 2023 at 10:15:45AM -0400, Sean Anderson wrote: > On 10/12/23 02:39, Heinrich Schuchardt wrote: > > On 10/12/23 03:56, Sean Anderson wrote: > > > fs.c thinks that the sandbox filesystem is available if SANDBOX is > > > enabled, > > > but it is not in SPL. Compile it in SPL to avoid linker errors. > > > > > > Signed-off-by: Sean Anderson <[email protected]> > > > --- > > > > > > fs/Makefile | 1 + > > > 1 file changed, 1 insertion(+) > > > > > > diff --git a/fs/Makefile b/fs/Makefile > > > index 4bed2ff2d99..592c7542bde 100644 > > > --- a/fs/Makefile > > > +++ b/fs/Makefile > > > @@ -9,6 +9,7 @@ obj-$(CONFIG_FS_LOADER) += fs.o > > > obj-$(CONFIG_SPL_FS_FAT) += fat/ > > > obj-$(CONFIG_SPL_FS_EXT4) += ext4/ > > > obj-$(CONFIG_SPL_FS_CBFS) += cbfs/ > > > +obj-$(CONFIG_SANDBOX) += sandbox/ > > > > Why wouldn't you use CONFIG_SANDBOX_SPL here? > > Because that's what the check is in fs.c. Maybe it should be > CONFIG_IS_ENABLED(SANDBOX) in there.
"CONFIG_SANDBOX_SPL" is not "CONFIG_SPL_SANDBOX", so CONFIG_IS_ENABLED doesn't work. I would suggest that fs/Makefile needs a bit of cleaning to make use of obj-$(CONFIG_$(SPL_)_... logic instead of ifdef CONFIG_SPL_BUILD, but I think Simon dislikes that form. We could just move the existing line outside of the SPL/else conditional. -- Tom
signature.asc
Description: PGP signature
