diff options
| author |   Peter Zijlstra <peterz@infradead.org> | 2011-08-24 15:32:31 +0200 |
|---|---|---|
| committer |   Michal Simek <monstr@monstr.eu> | 2011-10-14 12:28:42 +0200 |
| commit | 84ac218f0257a0cfd7689d9a44f53118dbf307b3 (patch) | |
| tree | 015d3fc9489fbb00da12dbf4eee72752f3307b9e /kernel/sysctl.c | |
| parent | microblaze: Use delay slot in __strnlen_user, __strncpy_user (diff) | |
| download | wireguard-linux-84ac218f0257a0cfd7689d9a44f53118dbf307b3.tar.xz wireguard-linux-84ac218f0257a0cfd7689d9a44f53118dbf307b3.zip | |
microblaze: Remove __ARCH_WANT_INTERRUPTS_ON_CTXSW usage
As far as I can tell the only reason microblaze has
__ARCH_WANT_INTERRUPTS_ON_CTXSW is because it initializes new task state
with interrupts enabled so that on switch_to() interrupts get enabled.
So change copy_thread() to clear MSR_IE instead of set it, this will
ensure switch_to() will always keep IRQs disabled.
The scheduler will disable IRQs when taking rq->lock in schedule() and
enable IRQs in finish_lock_switch() after its done its magic.
This leaves ARM the only __ARCH_WANT_INTERRUPTS_ON_CTXSW user.
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Signed-off-by: Michal Simek <monstr@monstr.eu>
Diffstat (limited to 'kernel/sysctl.c')
0 files changed, 0 insertions, 0 deletions