On 06/05/2013 12:08 PM, Maciej W. Rozycki wrote:
On Wed, 5 Jun 2013, Steven J. Hill wrote:diff --git a/arch/mips/mti-malta/malta-init.c b/arch/mips/mti-malta/malta-init.c index ff8caff..76e0205 100644 --- a/arch/mips/mti-malta/malta-init.c +++ b/arch/mips/mti-malta/malta-init.c @@ -106,6 +106,8 @@ extern struct plat_smp_ops msmtc_smp_ops; void __init prom_init(void) { + set_micromips_exception_mode(); + mips_display_message("LINUX"); /* diff --git a/arch/mips/mti-sead3/sead3-init.c b/arch/mips/mti-sead3/sead3-init.c index bfbd17b..9e314cb 100644 --- a/arch/mips/mti-sead3/sead3-init.c +++ b/arch/mips/mti-sead3/sead3-init.c @@ -130,6 +130,8 @@ static void __init mips_ejtag_setup(void) void __init prom_init(void) { + set_micromips_exception_mode(); + board_nmi_handler_setup = mips_nmi_setup; board_ejtag_handler_setup = mips_ejtag_setup;Shouldn't this be in a generic place such as trap_init instead?
I think it is fine here. If it spreads to more systems, then factoring them out into trap_init might make sense. For now it doesn't seem like we should clutter up trap_init when there aren't many microMIPS systems in existence.
David Daney
Maciej
