Since many kernel versions I regularly faced reproducibly kernel hangs when
compiling some bigger files. The machine suddenly just seemed to hang.
With spinlock debugging turned on I found this:
BUG: spinlock recursion on CPU#0, tool/7263
lock: 10644000, .magic: dead4ead, .owner: tool/7263, .owner_cpu: 0
Backtrace:
[<10113a94>] show_stack+0x18/0x28
BUG: spinlock lockup on CPU#0, tool/7263, 10644000
Backtrace:
[<10113a94>] show_stack+0x18/0x28
BUG: soft lockup - CPU#0 stuck for 61s! [tool:7263]
IASQ: 00000000 00000000 IAOQ: 102d55dc 102d557c
IIR: 03c008b3 ISR: 00000000 IOR: 00000000
CPU: 0 CR30: 7d1a4000 CR31: 11111111
ORIG_R28: 00000000
IAOQ[0]: _raw_spin_lock+0x15c/0x1c0
IAOQ[1]: _raw_spin_lock+0xfc/0x1c0
RP(r2): _raw_spin_lock+0x18c/0x1c0
Backtrace:
[<102d560c>] _raw_spin_lock+0x18c/0x1c0
Kernel panic - not syncing: softlockup: hung tasks
Backtrace:
[<10113a94>] show_stack+0x18/0x28
The lock at "10644000" references pa_tlb_lock, which is used to lock the PxTLB
broadcasts (see comment in cache.c).
So, I think we need to use the irq-blocking functions for purge_tlb_start() and
purge_tlb_end(). I did tested the patch below, and since then I didn't faced one
single hang any longer.
The patch below is just a hack. It will need a cleanup and is just to open
discussions here on the list.
Helge
diff --git a/arch/parisc/include/asm/tlbflush.h b/arch/parisc/include/asm/tlbflush.h
index 1f6fd4f..9b92fbe 100644
--- a/arch/parisc/include/asm/tlbflush.h
+++ b/arch/parisc/include/asm/tlbflush.h
@@ -18,8 +18,8 @@
*/
extern spinlock_t pa_tlb_lock;
-#define purge_tlb_start(x) spin_lock(&pa_tlb_lock)
-#define purge_tlb_end(x) spin_unlock(&pa_tlb_lock)
+#define purge_tlb_start(x) do { unsigned long flags; spin_lock_irqsave(&pa_tlb_lock, flags)
+#define purge_tlb_end(x) spin_unlock_irqrestore(&pa_tlb_lock, flags); } while (0)
extern void flush_tlb_all(void);
extern void flush_tlb_all_local(void *);
--
To unsubscribe from this list: send the line "unsubscribe linux-parisc" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
