netdev
[Top] [All Lists]

Re: MSEC_TO_JIFFIES is messed up...

To: Andrew Morton <akpm@xxxxxxxx>
Subject: Re: MSEC_TO_JIFFIES is messed up...
From: Paul Wagland <paul@xxxxxxxxxxxxxxxx>
Date: Thu, 13 May 2004 21:50:41 +0200
Cc: greg@xxxxxxxxx, linux-kernel@xxxxxxxxxxxxxxx, mingo@xxxxxxx, jgarzik@xxxxxxxxx, netdev@xxxxxxxxxxx, wli@xxxxxxxxxxxxxx, davidel@xxxxxxxxxxxxxxx, Valdis.Kletnieks@xxxxxx
In-reply-to: <20040513121141.37f32035.akpm@xxxxxxxx>
References: <40A26FFA.4030701@xxxxxxxxx> <20040512193349.GA14936@xxxxxxx> <200405121947.i4CJlJm5029666@xxxxxxxxxxxxxxxxxxxxxxx> <Pine.LNX.4.58.0405121255170.11950@xxxxxxxxxxxxxxxxxxxxxxx> <200405122007.i4CK7GPQ020444@xxxxxxxxxxxxxxxxxxxxxxx> <20040512202807.GA16849@xxxxxxx> <20040512203500.GA17999@xxxxxxx> <20040512205028.GA18806@xxxxxxx> <20040512140729.476ace9e.akpm@xxxxxxxx> <20040512211748.GB20800@xxxxxxx> <20040512221823.GK1397@xxxxxxxxxxxxxx> <61D92BA6-A504-11D8-BD91-000A95CD704C@xxxxxxxxxxx> <20040513121141.37f32035.akpm@xxxxxxxx>
Sender: netdev-bounce@xxxxxxxxxxx

On May 13, 2004, at 21:11, Andrew Morton wrote:

Paul Wagland <paul@xxxxxxxxxxx> wrote:
 This changes behaviour when HZ==(z)000

 JIFFIES_TO_MSECS  goes from
 ((x) * 1000) / (z)000  to (((x) + (z) - 1)/(z))

i.e. for x=1, z=2 this goes from ((1)*1000)/2000)=0 to (((1)+(2)-1)/2)=1

hm, so you're saying that we now round 0.5 up to 1 rather than down to zero?

More precisely, we round .x up, where before it was rounded down, but yeah, _and_ only when HZ is a multiple of 1000, and greater than 1000. This is also only the case for the patch as proposed by wli, currently I don't know of any architectures that have a HZ of 2000 or more... but I just note that it _is_ a behaviour change in those cases, whether or not it is important is for other people to decide :-)

Cheers,
Paul

Attachment: PGP.sig
Description: This is a digitally signed message part

<Prev in Thread] Current Thread [Next in Thread>