Messages in this thread | | | Date | Mon, 12 Oct 1998 09:36:03 +0900 (JST) | From | NIIBE Yutaka <> | Subject | Re: [PATCH] plip.c in 2.1.125 |
| |
Ely Wilson <plexus@ionet.net> wrote: > It removes a few redundant calles to synchronize_irq(), which is called by > disable_irq() *always* before return.
Point taken. Thanks.
> Removed a check from the irq handler for a plip devices,
I'll follow the opinion by Andrea Arcangeli <andrea@e-mind.com> about checking irq.
> Oh yes, also, I changed NET_DEBUG to PLIP_DEBUG and removed the runtime > checks making them preprocesor checks. (surprise! surprise!) so that there > isn't any extra overhead (mainly in the irq handler, so i just changed it > everywhere else) in checking if PLIP_DEBUG was set to a value.
I don't like it. Firstly, the NET_DEBUG scheme is not invented by PLIP driver itself, other network drivers use the runtime checking scheme, and PLIP driver followed it. Secondly, I don't think the overhead is so big. Thirdly, changing the value at runtime is easy, but it requires source code and compiler if it's the preprocessor checks.
Anyway, thank you for your constructive feedback.
Best Regards, -- NIIBE Yutaka
- 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/
| |