Hello Michael,

I guess we are back on business?
Can you please address these point?

Thanks!
Alon Bar-Lev.

On Tuesday 16 October 2007, Alon Bar-Lev wrote:
> Hello Michael,
> 
> Here are some of my thoughts regarding the key module interface. I
> wrote some in the past but I got no response regarding this.
> 
> 1. Introduce statefull mode.
> 
> Current implementation is stateless, every call to encrypt or decrypt
> results in deserialization, create the crypto context, do the crypto
> operation and release the context.
> 
> The above is done *TWICE*, as once it is called to get size, and the
> second to do the actual operation.
> 
> I suggest adding two callbacks, one for deserialize and one for free,
> the deserialize will allocate key specific context that will be used
> instead of the blob for crypto operations. The free will be called
> when the key is not used anymore.
> 
> This will be much more efficient than current implementation, and will
> also enable storing more data on the context so that next calls may
> proceed much quicker.
> 
> 2. Allow the key module to report that a key is unusable.
> 
> For example, if the user specify wrong passphrase for smartcard key
> (statically), and this passphrase is wrong, if more than X retries are
> done, the card will lock itself. So we need a way to report that this
> key is unusable, so it be deleted from store.
> 
> 3. Key module context & parameters
> 
> Allow key module specific parameter that are not related to a specific
> key, this will allow to specify some module wide options. It should be
> read from some configuration file and forwarded to the key module
> during initialization.This also suggests adding key module wide
> context that may be returned from module initialization callback, and
> free during module termination, this context should be passed to
> deserialization callback so it may be stored in the key context.
> 
> 4. [Not directly related] Pluggable Random Generator
> 
> eCryptfs gathers its random for the encryption keys using default
> system random, but I guess it should be pluggable. I believe the
> simplest way is to add sysfs entry and set a random device... Another
> solution is to use the netlink socket in order to get the random. As
> maybe users wish to choose a better random source for eCryptfs than
> the system.
> 
> Best Regards,
> Alon Bar-Lev.
> 



-------------------------------------------------------------------------
This SF.net email is sponsored by: Splunk Inc.
Still grepping through log files to find problems?  Stop.
Now Search log events and configuration files using AJAX and a browser.
Download your FREE copy of Splunk now >> http://get.splunk.com/
_______________________________________________
eCryptfs-devel mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/ecryptfs-devel

Reply via email to