Re: [RFC 1/2] Eliminate over- and under-counting of io_ticks

Hello! On 6/9/20 6:41 PM, Hou Tao wrote: > Hi, > > For the following case, the under-counting is still possible if io2 wins > cmpxchg(): > > t 0123456 > io1|-| > io2 |--| > stamp 0 6 > io_ticks 0 3 I hadn't noticed that bug. It looks like

Re: [RFC 1/2] Eliminate over- and under-counting of io_ticks

Hi, On 2020/6/9 12:07, Josh Snyder wrote: > Previously, io_ticks could be under-counted. Consider these I/Os along > the time axis (in jiffies): > > t 012345678 > io1|| > io2|---| > > Under the old approach, io_ticks would count up to 6, like so: > > t

Re: [RFC 1/2] Eliminate over- and under-counting of io_ticks

On Mon, Jun 08, 2020 at 09:07:23PM -0700, Josh Snyder wrote: > Previously, io_ticks could be under-counted. Consider these I/Os along > the time axis (in jiffies): > > t 012345678 > io1|| > io2|---| In current way, when io2 is done, io tickes should be 5, si

[RFC 1/2] Eliminate over- and under-counting of io_ticks

Previously, io_ticks could be under-counted. Consider these I/Os along the time axis (in jiffies): t 012345678 io1|| io2|---| Under the old approach, io_ticks would count up to 6, like so: t 012345678 io1|| io2|---|