SEO
vBulletin Search Engine Optimization
| |||||||
| Register | FAQ | Members List | Calendar | Search | Today's Posts | Mark Forums Read |
02-18-2008, 07:34 AM
| ||||
| ||||
 Re: Simple NFS v3 write speedup > > It might look safe, but the size of the diff does not say anything about > > it being safe or not. This really requires very wide testing, and we have > > no time left for that. > > My assertion that the first patch looked safe was not based on the size > of the diff, rather for these reasons: > > 1. It doesn't add any code, but merely changes the logic of one > conditional so that it evaluates differently and the execution path > follows an already existing branch. > > 2. It has already been applied to NetBSD. > > But if it's too late then it's too late. To explain a bit further, your logic still does not constitute it safe. By having much faster NFS writes, it is not unlikely to expose old VFS bugs and race conditions that we have never hit before...      |
Linear Mode
