> fortunately only the console code uses this function, which is mostly
> protected by the kernel lock. So i dont think this is an 'active' bug, but
If it would be protected by the big kernel lock it would have no raced in
first place. I think the reason the bh_mask setting it's _not_ racing at
all is a luck timing issue.
The disable/enable_bh() SMP race in the console code if not atomic
operation are used because one CPU is running the console_bh handler (from
the keyboard irq handler) and the other CPU is in the write(2) path from
a proggy.
Andrea Arcangeli
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/