[uClinux-dev] CRAMFS and mtime/ctime

John Williams jwilliams at itee.uq.edu.au
Thu Aug 25 20:11:56 EDT 2005


Are there known issues with cramfs and mtime/ctime?

If I loop mount the cramfs image on my workstation (recent 2.4 kernel), 
the mtimes and ctimes are different from the timestamps on the actual 
source files.

At first I thougt the mtime was set to the time at which the cramfs 
image was mounted, but lookiung more closely the times are all over the 
place - and certainly don't match the mtimes on the original files, from 
which the cramfs image was created.

When this image is mounted as the root fs on my uClinux system (latest 
kernel), all the mtimes are set to Jan 1 1970 (although so is my system 
clock, so that might explain something).

Does anybody have the 25 words or less explanation on this?



PS The reason I care is that I'm working with precompiled python 
modules, and the interpreter is complaining that the ctimes on the .py 
vs .pyc files are inconsistent.

More information about the uClinux-dev mailing list