Machine checks support waking up the mcelog daemon quickly.
The original wake up code for this was pretty ugly, relying on
a idle notifier and a special process flag. The reason it did
it this way is that the machine check handler is not subject
to normal interrupt locking rules so it's not safe
to call wake_up(). Instead it set a process flag
and then either did the wakeup in the syscall return
or in the idle notifier.
This patch adds a new "bootstraping" method as replacement.
The idea is that the handler checks if it's in a state where
it is unsafe to call wake_up(). If it's safe it calls it directly.
When it's not safe -- that is it interrupted in a critical
section with interrupts disables -- it uses a new "self IPI" to trigger
an IPI to its own CPU. This can be done safely because IPI
triggers are atomic with some care. The IPI is raised
once the interrupts are reenabled and can then safely call
wake_up().
When APICs are disabled the event is just queued and will be picked up
eventually by the next polling timer. I think that's a reasonable
compromise, since it should only happen quite rarely.
Contains fixes from Ying Huang.
[ solve conflict on irqinit, make it work on 32bit (entry_arch.h) - HS ]
Signed-off-by: Andi Kleen <ak@linux.intel.com>
Signed-off-by: Hidetoshi Seto <seto.hidetoshi@jp.fujitsu.com>
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
Merge reason: arch/x86/kernel/irqinit_{32,64}.c unified in irq/numa
and modified in x86/mce3; this merge resolves the conflict.
Conflicts:
arch/x86/kernel/irqinit.c
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
Impact: fix crash with many devices
I found this crash:
[ 552.616646] general protection fault: 0403 [#1] SMP
[ 552.620013] last sysfs file:
/sys/devices/pci0000:00/0000:00:02.0/usb1/1-1/1-1:1.0/host13/target13:0:0/13:0:0:0/block/sr0/size
[ 552.620013] CPU 0
[ 552.620013] Modules linked in:
[ 552.620013] Pid: 0, comm: swapper Not tainted 2.6.30-rc1-tip-01931-g8fcafd8-dirty #28 Sun Fire X4440
[ 552.620013] RIP: 0010:[<ffffffff8023bada>] [<ffffffff8023bada>] default_idle+0x7d/0xda
[ 552.620013] RSP: 0018:ffffffff81345e68 EFLAGS: 00010246
[ 552.620013] RAX: 0000000000000000 RBX: ffffffff8133d870 RCX: ffffc20000000000
[ 552.620013] RDX: 00000000001d0620 RSI: ffffffff8023bad8 RDI: ffffffff802a3169
[ 552.620013] RBP: ffffffff81345e98 R08: 0000000000000000 R09: ffffffff812244a0
[ 552.620013] R10: ffffffff81345dc8 R11: 7ebe1b6fa0bcac50 R12: 4ec4ec4ec4ec4ec5
[ 552.620013] R13: ffffffff813a54d0 R14: ffffffff813a7a40 R15: 0000000000000000
[ 552.620013] FS: 00000000006d1880(0000) GS:ffffc20000000000(0000) knlGS:0000000000000000
[ 552.620013] CS: 0010 DS: 0018 ES: 0018 CR0: 000000008005003b
[ 552.620013] CR2: 00007fec9d936a50 CR3: 000000007d1a9000 CR4: 00000000000006e0
[ 552.620013] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
[ 552.620013] DR3: 0000000000000000 DR6: 00000000ffff0ff0 DR7: 0000000000000400
[ 552.620013] Process swapper (pid: 0, threadinfo ffffffff81344000,task ffffffff812244a0)
[ 552.620013] Stack:
[ 552.620013] 0000000000000000 ffffc20000000000 00000000001d0620 7ebe1b6fa0bcac50
[ 552.620013] ffffffff8133d870 4ec4ec4ec4ec4ec5 ffffffff81345ec8 ffffffff8023bd84
[ 552.620013] 4ec4ec4ec4ec4ec5 ffffffff813a54d0 7ebe1b6fa0bcac50 ffffffff8133d870
[ 552.620013] Call Trace:
[ 552.620013] [<ffffffff8023bd84>] c1e_idle+0x109/0x124
[ 552.620013] [<ffffffff8023314b>] cpu_idle+0xb8/0x101
[ 552.620013] [<ffffffff80c16d6a>] rest_init+0x7e/0x94
[ 552.620013] [<ffffffff81357efc>] start_kernel+0x3dc/0x3fd
[ 552.620013] [<ffffffff813572a9>] x86_64_start_reservations+0xb9/0xd4
[ 552.620013] [<ffffffff813573b2>] x86_64_start_kernel+0xee/0x109
[ 552.620013] Code: 48 8b 04 25 f8 b4 00 00 83 a0 3c e0 ff ff fb 0f ae f0 65 48 8b 04 25 f8 b4 00 00 f6 80 38 e0 ff ff 08 75 09 e8 71 76 06 00 fb f4 <eb> 06 e8 68 76 06 00 fb 65 48 8b 04 25 f8 b4 00 00 83 88 3c e0
[ 552.620013] RIP [<ffffffff8023bada>] default_idle+0x7d/0xda
[ 552.620013] RSP <ffffffff81345e68>
[ 552.828646] ---[ end trace 4cbfc5c01382af7f ]---
Joerg Roedel said
"The 0403 error code means that there was an external interrupt with vector
0x80. Yinghai, my theory is that the kernel on this machine has no 32bit
emulation compiled in, right? In this case the selector points to a zero entry
which may cause the #gpf right after the hlt.
But I have no idea where the external int 0x80 comes from"
it turns out that we could use 0x80 for external device on 64-bit
when 32-bit emulation is disabled.
But we forgot to set the gate for it.
try to set gate for it by checking used_vectors.
Also move apic_intr_init() early to avoid setting
that gate two times.
Signed-off-by: Yinghai Lu <yinghai@kernel.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Joerg Roedel <joerg.roedel@amd.com>
LKML-Reference: <49E62DFD.6010904@kernel.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
To make the topic merge life easier for tip:perfcounters/core,
include two (inactive in this topic) IRQ vector initializations
here.
Also fix build bug - missing kprobes.h inclusion.
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Impact: cleanup
We can remove some #ifdefs if we define IA32_SYSCALL_VECTOR on 32-bit.
Reviewed-by Cyrill Gorcunov <gorcunov@openvz.org>
Signed-off-by: Pekka Enberg <penberg@cs.helsinki.fi>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Andi Kleen