Re: Should block-number-in-file vars be using blkcnt_t?

Andreas Dilger wrote:
> On Aug 21, 2006  10:31 -0500, Eric Sandeen wrote:
>> The blkcnt_t type recently went into to 2.6 to handle "large single 
>> files" as a configuration option; i.e. more than 32 bits of blocks in a 
>> file offset.  Right now this seems to be used only in the inode's 
>> i_blocks field.  Looking over Takashi Sato's patches to increase the 
>> maximum size of a file in ext3, I wonder if it might be a good idea to 
>> start using blkcnt_t in ext3, wherever we are counting blocks in a file?
> 
> I think Takashi also submitted a follow-on patch to introduce ext3_blkcnt
> for this purpose.  It makes the code more clear as to what type of block
> number is being used, and facilitates increasing this number.

Ok, I can't find that patch... but I'm glad that you agree that a type 
makes sense here.

Digging back through the archives I see that I'm re-inventing a few 
wheels here, trying to get things up to a full 32 bits.

Can anyone summarize the status of the various patchsets which have been 
submitted for 2^32-1 -block filesystems?  What can I do to help get this 
finished up and committed?  I'm currently comparing my *cough* 
independent work to what was already done; is Takashi's patchset from 
5/25/06 [1] the most current & viable so far?

I'd like to see the patches split out large filesystems from large files 
to a greater degree, just to keep things saner IMHO... but if they're on 
the verge of going in then no point to that I suppose.

Thanks,

-Eric

[1] http://marc.theaimsgroup.com/?l=ext2-devel&m=114856080926308&w=2

-------------------------------------------------------------------------
Using Tomcat but need to do more? Need to support web services, security?
Get stuff done quickly with pre-integrated technology to make your job easier
Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo
http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642