Messages in this thread | | | Date | Sat, 20 Sep 2008 13:03:25 -0700 (PDT) | Subject | Re: [PATCH 2/2]: softirq: Add support for triggering softirq work on softirqs. | From | David Miller <> |
| |
From: Daniel Walker <dwalker@mvista.com> Date: Sat, 20 Sep 2008 08:43:33 -0700
> On Fri, 2008-09-19 at 23:48 -0700, David Miller wrote: > > > @@ -6,6 +6,8 @@ > > * Distribute under GPLv2. > > * > > * Rewritten. Old one was good in 2.2, but in 2.3 it was immoral. --ANK (990903) > > + * > > + * Remote softirq infrastructure is by Jens Axboe. > > */ > > This goes in the GIT log so I hear, so you shouldn't need to add it to > the top.. It sounds like your saying Jens is the author, but I'm sure > you are..
No, Jens wrote this code, I just merely made it generic. He also deserves a mention at the top of the file. If we had left it in the block layer, he would have received such a mention.
Why the heck does this even bother you?
> This list your adding is rather confusing .. You add to it, but never > remove anything.. You've got it in the header file, so you must use it > someplace else .. Then I don't see what else it could be used for other > than triggering the softirq..
The layers that use this stuff dequeue from their softirq handler, directly from the list, that's why it's exported.
> This whole patch really needs ifdefs. There's no value here on UP, since > what other cpu are you going to send softirqs to?
On UP we get a single queue, which the layer needs anyways. It just always queues to the one queue.
Unlike Andrew's, your review comments have been completely and utterly useless, as well as a total waste of my time.
| |