Dev Summit (2)
活动很有意思,今天
下午的会议主要是新近的一些专题。pjd@介绍了jail service,包括ZFS与jail之间的一些联动等。通过ZFS提供的一些能力,我们能够让jail对更多的资源进行限制(例如磁盘空间,过去只能通过在vnode上面建FS这种土办法),并且更像一台真实的机器。之后是gnn老大长达近两小时的KSE代码导读;比较遗憾的是,davidxu和很多其他重量级的threading hacker都没能参加这次会议。
共9篇文章。
活动很有意思,今天
下午的会议主要是新近的一些专题。pjd@介绍了jail service,包括ZFS与jail之间的一些联动等。通过ZFS提供的一些能力,我们能够让jail对更多的资源进行限制(例如磁盘空间,过去只能通过在vnode上面建FS这种土办法),并且更像一台真实的机器。之后是gnn老大长达近两小时的KSE代码导读;比较遗憾的是,davidxu和很多其他重量级的threading hacker都没能参加这次会议。
It seems that I have caught a race condition which can be triggered by renice(1), and we need to take PRS_NEW into account when we need to
access several attributes of a process object.
Robert has pointed out that we may want to just initialize the attribute
early to avoid the race condition. After reading some old 4.4BSD code it
seems that the process object is linked into allproc list before
everything is initialized.
今天,John Birrell同学commit了一个不大不小的patchset——-CURRENT上,KSE不再是一个必选项了。
KSE是FreeBSD 5.0-CURRENT时的一个非常重要的技术探索,它是对由华盛顿大学Thomas E. Anderson等人提出的调度器激活概念的一个实现。
对于M:N线程来说,调度器激活是一项十分先进的概念。在这个模型中,内核并不需要了解用户态线程的细节,而用户态线程也并不是由内核直接调度,相反,这些线程被绑定到一系列「调度实体」上,内核能够看到这些调度实体,并在上下文切换时激活用户态的线程调度器。由于减少了内核-用户态的上下文切换操作,因此理论上这一模型能够获得更好的性能。
I have noticed that today’s cvsup brings this to my local repo: src. From the version it seems to be from kern_thread.c. Waiting commitmail to explain this.
Now we have RELENG_4_10 branch :-) There are a lot of -CURRENT improvements has been MFC’ed, however, the most important part of -CURRENT, say, KSE and SMPng, will never be MFC’ed because of the architecture change introduced by this.
Scott Long today announced the release. This is a point release over 5.2-RELEASE which addresses many critical bugs.
The -CURRENT have already differentiated from the RELENG_5_2 branch, say, 5.2.1-RELEASE by this time. The most significant differences are:
Daniel Eischen today committed a new patch and this will make KSE default threading library on FreeBSD!
FreeBSDChina社区最近采访了徐逸锋(David Xu),一位来自中国大陆的FreeBSD src committer。徐逸锋先生在FreeBSD KSE项目中做出了相当多的贡献,在这次访谈中,他谈到了关于自己参加FreeBSD开发的一些经历、KSE、SMPng以及关于FreeBSD的一些其他事情。我们非常感谢徐先生接受我们的采访,也希望这次访谈能够帮助大家进一步了解FreeBSD。
Today folks at FreeBSDChina suggested me to have an interview with FreeBSD’s David XU (Yifeng XU), who is an active committer, and possibly the only src committer living in the mainland China.