On 06/03/2014 10:34 AM, Ralf Baechle wrote: > On Tue, Jun 03, 2014 at 09:46:17AM +0100, Markos Chandras wrote: > >> Warning: the 32-bit microMIPS architecture does not support the `smartmips' >> extension >> arch/mips/kernel/entry.S:90: Error: unrecognized opcode `mtlhx $24' >> [...] >> arch/mips/kernel/entry.S:109: Error: unrecognized opcode `mtlhx $24' >> >> Link: https://dmz-portal.mips.com/bugz/show_bug.cgi?id=1021 >> Reviewed-by: Steven J. Hill <Steven.Hill@xxxxxxxxxx> >> Signed-off-by: Markos Chandras <markos.chandras@xxxxxxxxxx> >> --- >> arch/mips/Kconfig | 2 +- >> 1 file changed, 1 insertion(+), 1 deletion(-) >> >> diff --git a/arch/mips/Kconfig b/arch/mips/Kconfig >> index 2fe8e60..ffde3d6 100644 >> --- a/arch/mips/Kconfig >> +++ b/arch/mips/Kconfig >> @@ -2063,7 +2063,7 @@ config ARCH_PHYS_ADDR_T_64BIT >> def_bool 64BIT_PHYS_ADDR >> >> config CPU_HAS_SMARTMIPS >> - depends on SYS_SUPPORTS_SMARTMIPS >> + depends on SYS_SUPPORTS_SMARTMIPS && !CPU_MICROMIPS >> bool "Support for the SmartMIPS ASE" >> help >> SmartMIPS is a extension of the MIPS32 architecture aimed at > > From a user's perspective that's a bit quirky; a user has to first > disable CPU_MICROMIPS before he can enable CPU_HAS_SMARTMIPS. So I > think this should become a choice statement. > > Ralf > James (now on CC) suggested the same thing and I told him that choice statements make sense for symbols that are somehow related. For example SMVP or SMTC or no-MT. In this case, these symbols totally different so in my opinion a choice symbol will be rather confusing. But I will not object. Whatever works best. -- markos
