> I've been playing with large files (ext2fs) on the Alpha,
> kernel 2.1.116.
>
> It appears I am able to create a 16 1/16 GB file - but not 16 1/8 GB
> file. Somewhere in between lies a limit or problem...
You hit the triple-indirect limit for 1 kB block filesystems.
> Now, it seems to me, it should be possible to create files large
> than this, even several hundred GD, am I correct?
Yes, but you need to create a new filesystem with larger blocks.
Here is a log-log graph showing how max file size varies with the
filesystem block size.
http://www.cs.uml.edu/~acahalan/linux/ext2.gif
Summary:
1 kB -> 16 GB
2 kB -> 256 GB
4 kB -> 4 TB
8 kB -> 32 TB (limit is block counter)
That assumes you won't hit a 32-bit sector number limit in some
other part of the kernel, cutting you off at 1 TB or 2 TB.
> P.S. On the Intel, we are restricted to 2GB files because of mm
> considerations. But, since we have a 3/1 split, why isn't the
> maximum file size 3GB?
It seems lseek() would get ugly. I suppose absolute positions could
always treat the offset as unsigned. That would mean you couldn't
cover the whole file in one relative seek, but could reach it all
with absolute seeks. There may be other problems.
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.altern.org/andrebalsa/doc/lkml-faq.html
