summaryrefslogtreecommitdiffstats
path: root/arch/microblaze
diff options
context:
space:
mode:
authorRabin Vincent <rabin@rab.in>2015-02-19 18:20:09 +0100
committerJesper Nilsson <jespern@axis.com>2015-03-25 10:54:06 +0100
commit06aca92424c2a53a9e26cb81c1d6a829af9491b8 (patch)
treee85258187b8d566464427228df1fb7703284b826 /arch/microblaze
parent9a7449d3e975fe5c5ca12b7fea4f4bd69188a5f9 (diff)
downloadlinux-stable-06aca92424c2a53a9e26cb81c1d6a829af9491b8.tar.gz
linux-stable-06aca92424c2a53a9e26cb81c1d6a829af9491b8.tar.bz2
linux-stable-06aca92424c2a53a9e26cb81c1d6a829af9491b8.zip
CRISv32: don't enable irqs in INIT_THREAD
INIT_THREAD enables interrupts in the thread_struct's saved flags. This means that interrupts get enabled in the middle of context_switch() while switching to new tasks that get forked off the init task during boot. Don't do this. Fixes the following splat on boot with spinlock debugging on: BUG: spinlock cpu recursion on CPU#0, swapper/2 lock: runqueues+0x0/0x47c, .magic: dead4ead, .owner: swapper/0, .owner_cpu: 0 CPU: 0 PID: 2 Comm: swapper Not tainted 3.19.0-08796-ga747b55 #285 Call Trace: [<c0032b80>] spin_bug+0x2a/0x36 [<c0032c98>] do_raw_spin_lock+0xa2/0x126 [<c01964b0>] _raw_spin_lock+0x20/0x2a [<c00286c8>] scheduler_tick+0x22/0x76 [<c003db2c>] update_process_times+0x5e/0x72 [<c0007a94>] timer_interrupt+0x4e/0x6a [<c00378d6>] handle_irq_event_percpu+0x54/0xf2 [<c00379c4>] handle_irq_event+0x50/0x74 [<c003988e>] handle_simple_irq+0x6c/0xbe [<c0037270>] generic_handle_irq+0x2a/0x36 [<c0004c40>] do_IRQ+0x38/0x84 [<c000662e>] crisv32_do_IRQ+0x54/0x60 [<c0006204>] IRQ0x4b_interrupt+0x34/0x3c [<c0192baa>] __schedule+0x24a/0x532 [<c00056b4>] ret_from_kernel_thread+0x0/0x14 Signed-off-by: Rabin Vincent <rabin@rab.in> Signed-off-by: Jesper Nilsson <jespern@axis.com>
Diffstat (limited to 'arch/microblaze')
0 files changed, 0 insertions, 0 deletions