[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

JFFS2 performance (fwd)




Please reply to John G Dorsey <john+@xxxxxxx.edu> since I don't know if he's
subscribed to the list.

---------- Forwarded message ----------
Date: Sun, 18 Mar 2001 14:10:42 -0500
From: John G Dorsey <john+@xxxxxxx.edu>
To: nico@xxxxxxx.ORG
Subject: JFFS2 performance

Nicolas,

Are there any documented discussions of the performance issues
associated with some of the JFFS2 compression algorithms? I'm looking at
the "DYNRUBIN" method which was chosen by mkfs.jffs2 to compress most of
the inodes of my zImage, and it seems to be much more computationally
expensive than, say, zlib compression.

I'm feeling kind of discouraged by performance in general surrounding
the JFFS2 support in bootldr. It takes 11 seconds to replay the log for
an 8MB filesystem, and that's with flash caching enabled. Add to _that_
the cost for decompressing each inode, and the boot process starts to
become lengthy. What's most annoying, I suppose, is the fact that any
compression is being applied to the zImage inodes at all, since nearly
all of them have already been gzipped.

Thanks,
-jd


To unsubscribe from this list: send the line "unsubscribe jffs-dev" in
the body of a message to majordomo@xxxxxxx.com