jesperrosenkilde:reducingsyscalloverhead

Differences

This shows you the differences between two versions of the page.

Link to this comparison view

Both sides previous revision Previous revision
Next revision		 Previous revision
jesperrosenkilde:reducingsyscalloverhead [2007/07/20 16:30] – JesperRosenkildejesperrosenkilde:reducingsyscalloverhead [2022/10/06 01:00] (current) – Add Category: GsoC, move footnote to intro KubilayKocak
Line 1: Line 1:
 ===== Avoiding syscall Overhead ===== ===== Avoiding syscall Overhead =====
 +
 +This project is for the 2007 Google Summer of Code. [[/JesperRosenkilde|JesperRosenkilde]] will be working on it with [[/JeffRoberson|JeffRoberson]] as mentor.
  
 ==== Synopsis ==== ==== Synopsis ====
 +
 At the moment the setproctitle call is implemented with a sysctl, this has the unfortunate side effect that this simple call locks the Giant-lock. As this call is a simple matter of setting a value, it could be better implemented with shared memory between the kernel and  user-space. At the moment the setproctitle call is implemented with a sysctl, this has the unfortunate side effect that this simple call locks the Giant-lock. As this call is a simple matter of setting a value, it could be better implemented with shared memory between the kernel and  user-space.
  
 ==== Project ==== ==== Project ====
 +
 This project purposes a scheme to securely share process specific data, between the kernel and a user-space process. This is done by having each process allocate a special memory page, on which the kernel and user-space process can share data. This will give the security needed, as the VM-system will make sure that no outside processes can fiddle with a process' data. As everything is going on it user-space, there is no concern about a rogue process could write inside the kernel memory. There is still a locking concern, which will be addressed either by locking the entire page, or micro-locking each data field on the page. To test the scheme, I will implement the setproctitle call with share memory as oppose to the current syscall version. This project purposes a scheme to securely share process specific data, between the kernel and a user-space process. This is done by having each process allocate a special memory page, on which the kernel and user-space process can share data. This will give the security needed, as the VM-system will make sure that no outside processes can fiddle with a process' data. As everything is going on it user-space, there is no concern about a rogue process could write inside the kernel memory. There is still a locking concern, which will be addressed either by locking the entire page, or micro-locking each data field on the page. To test the scheme, I will implement the setproctitle call with share memory as oppose to the current syscall version.
  
-//Update: Howard Su has sugested a a multi page scheme, where a read/write page is used for things like get/setproctitle and a read-only page for things like getpid. And maybe a system wide read-only page for things like getdomain, gethostname etc. More on this to follow.//+//Update: Howard Su has suggested a a multi page scheme, where a read/write page is used for things like get/setproctitle and a read-only page for things like getpid. And maybe a system wide read-only page for things like getdomain, gethostname etc. More on this to follow.//
  
 ==== Milestone ==== ==== Milestone ====
 +
 This is a list of milestone, and the order in which I will take them. This is a list of milestone, and the order in which I will take them.
  
Line 15: Line 20:
   - Deallocate the page. ✔   - Deallocate the page. ✔
   - Write something to the page. ✔   - Write something to the page. ✔
-  - Read and write to the page from user space. +  - Read and write to the page from user space. ✔ 
-  - Figure out a suitable data structure.+  - Figure out a suitable data structure. ✔
   - Figure out a locking scheme.   - Figure out a locking scheme.
   - Add the new scheme to setproctitle, running along side the old.   - Add the new scheme to setproctitle, running along side the old.
Line 30: Line 35:
  
 All weeks included. All weeks included.
 +
 +==== Patch/Source ====
 +
 +This should add a working getpid() through the page-scheme. A page is allocated on fork() and exec() and its address is accessible through the kern.usrsysshm sysctl. There is a struct on the page which is hold the pid, the pid is set from the kernel, and is accessible directly from the user process' vmspace.
 +
 +http://www.cs.aau.dk/~jbr/patch.diff
 +http://www.cs.aau.dk/~jbr/tests.tar.bz2
  
 ---- ----
  
-//This project is for the 2007 Google Summer of Code. [:[[/JesperRosenkilde|JesperRosenkilde]]:Jesper B. Rosenkilde] will be working on it with [:[[/ScottLong|ScottLong]]:Scott Long] as mentor//+{{tag>Gsoc}}
  
  • jesperrosenkilde/reducingsyscalloverhead.1184949044.txt.gz
  • Last modified: 2007/07/20 16:30
  • by JesperRosenkilde