diff options
author | Carl Fredrik Hammar <hammy.lite@gmail.com> | 2009-06-25 18:30:27 +0200 |
---|---|---|
committer | Carl Fredrik Hammar <hammy.lite@gmail.com> | 2009-06-25 18:30:27 +0200 |
commit | 8146471c38766cac0aff5074823aa958dda101b0 (patch) | |
tree | 4b627d32a3d1516a065706f7c677726716c9ceca | |
parent | c838f31395bb2243fa0bf336069f9eef0c0ea912 (diff) |
hurd/translator/short-circuiting: Polish and elaborate
-rw-r--r-- | hurd/translator/short-circuiting.mdwn | 48 |
1 files changed, 30 insertions, 18 deletions
diff --git a/hurd/translator/short-circuiting.mdwn b/hurd/translator/short-circuiting.mdwn index 064e4abb..9de9f7b8 100644 --- a/hurd/translator/short-circuiting.mdwn +++ b/hurd/translator/short-circuiting.mdwn @@ -34,34 +34,40 @@ special files: * `/hurd/ifsock` So all special files are implemented through special-purpose translators, -right? Well, actually there aren't even implementations of either `chrdev` or -`blkdev` in the Hurd. Instead these are implemented purely by *translator -short-circuiting*, as well as the others from the above list are, if possible. - -Translator short-circuiting is when a file system server, instead of starting a -passive translator -- say, for resolving a symbolic link -- implements the -functionality itself. For instance, by continuing a file name look-up by -simply appending the path of a symlink's target without returning to the caller -a `FS_RETRY_*` reply, which is what the [[`symlink`|symlink]] translator would -do. (And that's all the `symlink` translator ever does, by the way.) - -In fact this list's translators that actually are implemented (`symlink`, +right? Not quite, instead the translators of this list are often +implemented in their underlying filesystem through *translator +short-circuiting*. In fact, `chrdev` and `blkdev` aren't even implemented +as translators at all. + +Translator short-circuiting is when a file system server implements the +functionality of a passive translator itself, instead of actually starting +it. For instance, all the [[`symlink`|symlink]] translator does is return +a `FS_RETRY_*` reply to the caller. So instead of starting it, the file +system server can simply continue the file name look-up internally by +appending the target of the symbolic link to the path being looked-up. + +This way, we can skip starting the `symlink` translator, skip retrying +the look-up on the newly started translator, and we might also skip a +retry to the same file system server again, if the target of the symbolic +link is in it. + +In fact, the list's translators that actually are implemented (`symlink`, `fifo`, `ifsock`) are only used as a default implementation if the underlying file system's translator does not implement the functionality itself, i.e., if it doesn't short-circuit it. -To make sure that you use one of these translators, that you bypass the +To make sure that you use one of these translators, there by bypassing the short-circuiting mechanism, you can either start it as -an active translator, or you can use a different path from the one in +an active translator, or use a different path from the one in `hurd/path.h`, e.g. `settrans bar /hurd/./symlink foo`. The best example of how short-circuiting is implemented can be found in [[`libdiskfs`|libdiskfs]]. Notice how it detects if a translator to store -is a special file in `diskfs_S_file_set_translator` and (if possible) instead -of storing a real passive translator setting on the disk, simply has (for -example) [[`ext2fs`|ext2fs]]' `diskfs_create_symlink_hook` handle this case. +is a special file in `diskfs_S_file_set_translator` and instead +of storing a real passive translator setting on the disk, stores it as a +symlink node (using `diskfs_create_symlink_hook` or a generic implementation). -When reading (resolving), it checks the node's `stat` structure in +In later look-ups to the node, it checks the node's `stat` structure in `diskfs_S_file_get_translator`, or `diskfs_S_dir_lookup` and handles special file types appropriately. @@ -71,6 +77,12 @@ functionality, but it also has advantages: using functionality that the file system's data structures nevertheless already provide -- storing symbolic links in `ext2fs`' inodes instead of storing passive translator settings -- and thus staying compatible with other operating systems mounting that file system. + Also, this short-circuiting does preserve system resources, as it's no longer required to start a `symlink` translator for resolving each symbolic link, as well as it does reduce the [[RPC]] overhead. + +It can also confuse users who expect the passive translator to start. +For instance, if a user notices that [[`symlink`|symlink]]'s code is +lacking some functionality, but that it unexpectedly works when the user +tries to run it. |