>
>On Tue, 7 Dec 1999, Andrea Arcangeli wrote:
>> >This allows me to see the irqdis_nobh in my ktrace, however, I never
>> >do see it. [..]
>>
>> That's fine because you must not see it. If you see it it means it should
>> BUG() or you won't be able to achieve the best latency.
>>
>What is BUG()?
BUG() in 2.3.x is a nicer way to say *(int *)0 = 0 (and BUG() probably
will work also if the virtual address zero is mapped to a valid physical
page but that's not an issue for most code).
>Sorry, not anymore. I have new test results and I will provide a plain
>(mostly) ktrace dump. I did not enable CPU info since I am not SMP.
Fine thanks. I'll look into it ASAP.
>Why are the hex addresses interesting?
Because with hex I understand without possible mistakes where the CPU was
running inside do_IRQ. (I don't need the System.map if the trace is long
enough)
>Will enabling CPU info be useful in my case?
No, the trace you sent should be just fine. I didn't known you was UP (and
the CPU info will never harm so I suggested you the safe way ;).
Andrea
-
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/