討論串HEADS UP - major structure size changes in HEAD
共 9 篇文章
內容預覽:
On 10/06/2010 05:16 쾀弮, Matthew Dillon wrote:. [...]. > For the most part these tokens are not going to be held for more than> 1uS or so, and often qu
(還有533個字)
內容預覽:
Well, my point here is that if the VM token is held for only a short. period of time on the other cpu, then while the first cpu is running. through th
(還有1824個字)
內容預覽:
On 09/06/2010 08:57 弮弮, Matthew Dillon wrote:. > :I fully agree that tokens are a more convenient API than, say, lockmgr. > :locks, but AFAIK nobody h
(還有5091個字)
內容預覽:
Tokens and the big-giant-lock behave the same way when under contention.. Both basically call lwkt_switch(). The LWKT scheduler then essentially. spin
(還有4016個字)
內容預覽:
The cache management traffic is definitely an issue. That's where. something like MONITOR/MWAIT could help a lot, or even just a PAUSE. instruction..
(還有93個字)