diff options
| author | Simon Horman <horms+renesas@verge.net.au> | 2013-03-05 15:40:42 +0900 |
|---|---|---|
| committer | Simon Horman <horms+renesas@verge.net.au> | 2013-03-13 02:24:37 +0900 |
| commit | 342896a5c565e38dfb87954f362735f03ae1efb0 (patch) | |
| tree | 03aa26a8cc20cf89529a77070a3922c204c3f22f /drivers/clocksource/sh_mtu2.c | |
| parent | 09acc3a1e0dee537fcfcd5a6c17a1e9b26586066 (diff) | |
| download | linux-342896a5c565e38dfb87954f362735f03ae1efb0.tar.gz linux-342896a5c565e38dfb87954f362735f03ae1efb0.tar.bz2 linux-342896a5c565e38dfb87954f362735f03ae1efb0.zip | |
clocksource: sh_mtu2: Set initcall level to subsys
The reason for this is to ensure that MTU2 is probed earlier
than with its previous initcall level, module init.
This came up as a problem with using CMT as a clock source kzm9g-reference
which does not make use of early timers or devices. In that scenario
initialisation of SDHI and MMCIF both stall on msleep() calls due to the
absence of a initialised clock source.
The purpose of this change is to keep the MTU2 code in sync with the CMT code
which has been modified in a similar manner..
Compile tested only using se7206_defconfig.
I do not believe I have any boards that support the MTU2.
Signed-off-by: Simon Horman <horms+renesas@verge.net.au>
Diffstat (limited to 'drivers/clocksource/sh_mtu2.c')
| -rw-r--r-- | drivers/clocksource/sh_mtu2.c | 2 |
1 files changed, 1 insertions, 1 deletions
diff --git a/drivers/clocksource/sh_mtu2.c b/drivers/clocksource/sh_mtu2.c index 83943e27cfac..4aac9ee0d0c0 100644 --- a/drivers/clocksource/sh_mtu2.c +++ b/drivers/clocksource/sh_mtu2.c @@ -386,7 +386,7 @@ static void __exit sh_mtu2_exit(void) } early_platform_init("earlytimer", &sh_mtu2_device_driver); -module_init(sh_mtu2_init); +subsys_initcall(sh_mtu2_init); module_exit(sh_mtu2_exit); MODULE_AUTHOR("Magnus Damm"); |