有谁读过ntfs的write的代码的（要求3遍以上，我不相信3000多行的代码读1-2遍就能完全理解的）？请加我msn

qianjunhua2000@hotmail.com

见我的
《六一献小礼：完整可编译NT4's NTFS源代码小规模修改工程》吧。

太高深了，不知所云

看上去像招聘广告。。。

引用第22楼killvxk于2008-02-03 16:29发表的 Re:有谁读过ntfs的write的代码的（要求3遍以上，我不相信3000多行的代码读1-2遍就能完全理解的）？请加 .. :
文件系统要搞读写磁盘的位置，大小等；
好的文件系统设计时就要考虑速度问题，分布问题，磁盘碎片等等~~

恩是个文件系统都要考虑这些问题，那么请问ntfs是怎么处理这些问题呢？它有什么特点恩？还有文件系统作为一个唯一的系统和磁盘打交道的路径，文件系统处理向磁盘read和write数据以外 和系统中别的组件 打交道 不会是不值得一提吧？

文件系统要搞读写磁盘的位置，大小等；
好的文件系统设计时就要考虑速度问题，分布问题，磁盘碎片等等~~

引用第20楼GoodOnline于2008-02-02 18:22发表的 Re:.....fsasdfsadf :


NTFS 在哪里啊?没看到源代码

NTFS文件系统应该和磁盘没有关系, 只关心clusters,什么东西是cluster无所谓.

killvxk 这么说可能有他的道理吧，不知killvxk 能否说明 文件驱动和磁盘到底有什么关系呢？

引用第7楼killvxk于2008-01-30 22:40发表的 .....fsasdfsadf :
我读了EXT3的Read和Write，还有udfs的Read和Create,觉得就是文件系统的设计和磁盘是分不开的，有时候看看计算机系统组成和结构会有意向不到的收获~

NTFS 在哪里啊?没看到源代码

NTFS文件系统应该和磁盘没有关系, 只关心clusters,什么东西是cluster无所谓.

不是牛，只是我的思路不同...

我也是每天在学习新的东西~

引用第16楼killvxk于2008-02-01 18:34发表的 Re:有谁读过ntfs的write的代码的 :
我知道，且已经投入使用~
哎~
VMWare的文档很全，intel的VT文档也很全，VBox和各种虚拟机的代码都有~~
有些东西说白很简单，都是思路问题。
~想法不同哦~
.......

此人确实真的很牛 从上面他写的东西来看。佩服佩服。真的没有想到驱动网上还有如此级别的牛人。
以后我能向你请教 开发文件过滤驱动的问题吗？

引用第16楼killvxk于2008-02-01 18:34发表的 Re:有谁读过ntfs的write的代码的 :
我知道，且已经投入使用~
哎~
VMWare的文档很全，intel的VT文档也很全，VBox和各种虚拟机的代码都有~~
有些东西说白很简单，都是思路问题。
~想法不同哦~
.......

跟不上时代了，郁闷。。。。。。

我知道，且已经投入使用~
哎~
VMWare的文档很全，intel的VT文档也很全，VBox和各种虚拟机的代码都有~~
有些东西说白很简单，都是思路问题。
~想法不同哦~

把基础搞好，做什么都得心应手~

难道你所说的这些技术 你都知道怎么去实现吗？知道还是不知道？还是听别人说的？

.........
沙箱技术，虚拟机都可以做到，从文件到网络，彻底的隔离，甚至可以做到一个文件多种明文的技术~~

拿起ida干活去了~

有些产品是没有这个能力，至于怎么实现，拿起IDA动手自己干吧~

现在都虚拟机技术和sanbox技术了，透明过程完全可以是另一种思路了

不要说密文就是给两个进程完全不一样的明文内容都可以~~~

哈哈哈哈哈哈哈哈哈哈

甚至两个进程的网络数据都可以...不多说了~

还是讨论实质的问题吧。
如果有2个程序一个授权的一个是没有授权的，透明加密系统怎么才能给授权的程序以明文，给
没有授权的程序以密文呢？在加密系统能实现吗？怎么实现？ 还有就是在产品在应用中有没有必要实现这个功能？欢迎大家讨论。

同步，操作系统设计里有更多的同步，如果不能了解到本质，看再多的代码，也是停留在别人的水平之下的~

他们是设计整个ntfs的框架，等于从基础开始做。我们现在是在这个框架里面做，你说要不要先通过源代码的阅读来了解这个框架呢？然后再在这个框架里面寻找灵感呢？ 不因为别的，就因为我们现在还在人家的框架里面！

俺听说先杰们搞NTFS 都是抛开电脑，带上纸和笔，划着小船去航海。可见灵感并不一定在于读这3000多行代码嘿嘿。关键是境界把？俺对于这等豪情只有膜拜的份。

Miller gathered some pens and pads, two weeks' worth of provisions and prepared for a lengthy trip on his twenty-eight-foot sailboat. Miller felt that spec writing benefited from solitude, and the ocean offered plenty of it. But he had worked on enough specs "to know that if you do something entirely in isolation you won't get the best crack at it. You need someone to at least bounce ideas off of." Kimura, the obvious trip mate, couldn't come. He had a date with the other file systems. Rather than sail alone, Miller arranged with Perazzoli, who officially took care of the file team, to fly in a programmer Miller knew well. He lived in Switzerland.
In August, Miller and his sidekick set sail for two weeks. The routine was easy: Work in the morning, talking and scratching out notes on a pad, then sail somewhere, then talk and scratch out more notes, then anchor by evening and relax. "You can't do this stuff too intensely," Miller said.
On the water, Miller pondered dozens of technical questions relating to the file system, but he thought most deeply about recovering and restoring files in the event the computer lost power or totally failed. This was the feature that would make or break NT's file system. Certain versions of Unix, the high-end operating program that Cutler sought to outdo with NT, were able to restore files following these situations. But when a PC lost data from a file, it usually lost it for good. This was because unless the PC was specifically instructed to store new entries in the permanent storage of a hard disk, it would save time by keeping many minutes' worth of entries in speedier, temporary DRAM memory. In event of a failure, this memory was always wiped clean.
Miller's answer was to create a duplicate record of all entries. The first record was the ordinary record, which resided first in temporary memory and then in permanent storage. The duplicate entry would live in a special log, which would be sent separately to the hard disk. This log would not contain entries made at the precise moment of the failure, but everything until a few seconds before the disaster. Miller called this state the "last known good." Once the computer came on again, the files in this state would appear.

At the end of his sail, Miller had the raw stuff for his spec. He then spent a week at home typing his notes and polishing the language. He finished with a fifty-page description of the guts of a new file system, which met both his and Cutler's aims. But back at Microsoft, he ran into a wall. Despite Cutler's support, Miller found that he, Kimura and two colleagues were swamped making the DOS and OS/2 file systems run faster and better on NT. Work on the new file system had to wait.

想请问一下为啥说文件系统的设计和磁盘分不开呢？还有计算机系统组成和结构对于理解ntfs有什么帮助呢？   我到现在的感觉是 ntfs里面的同步，它的所有的处理都是要考虑同步。也许这个是废话，那个复杂点的程序没有同步呢？ 但是它的同步有一个特点就是它要同步的情况是比较多，这样对于理解起来就比较困难。