lkml.org 
[lkml]   [2010]   [Jul]   [26]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
Date
From
SubjectRe: [PATCH UPDATED 1/3] vhost: replace vhost_workqueue with per-vhost kthread
On 07/26/2010 06:23 PM, Michael S. Tsirkin wrote:
>> * Can you please keep the outer goto repeat loop? I just don't like
>> outermost for (;;).
>
> Okay ... can we put the code in a {} scope to make it clear
> where does the loop starts and ends?

If we're gonna do that, it would be better to put it inside a loop
construct. The reason why I don't like it is that loops like that
don't really help read/writeability much while indenting the whole
logic unnecessarily and look more like a result of obsession against
goto rather than any practical reason. It's just a cosmetic
preference and I might as well be the weirdo here, so if you feel
strong about it, please feel free to put everything in a loop.

>> * Placing try_to_freeze() could be a bit annoying. It shouldn't be
>> executed when there's a work to flush.
>
> It currently seems to be executed when there is work to flush.
> Is this wrong?

Oh, does it? As I wrote in the other mail, things like that wouldn't
necessarily break correctness but I think it would be better to avoid
surprises in the generic code if not too difficult.

>> * I think A - B <= 0 test would be more familiar. At least
>> time_before/after() are implemented that way.
>
> I am concerned that this overflows a signed integer -
> which I seem to remeber that C99 disallows.

Really? Overflows of pointer isn't expected and that's why we have
weird RELOC_HIDE() macro for such calculations but integers not
expected to overflow is a news to me. Are you sure? That basically
means time_before/after() aren't safe either.

> timer macros are on data path so might be worth the risk there,
> but flush is slow path so better be safe?

I don't think performance matters much here. I just think the sign
test is clearer / more familiar for the logic.

Thanks.

--
tejun


\
 
 \ /
  Last update: 2010-07-26 21:07    [W:0.283 / U:0.060 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site