on 11/22/00 11:07 AM, Larry Jones at [EMAIL PROTECTED] wrote:
>> Why in the world does CVS need over 33 Megs of memory to
>> check in a 1.7 Meg binary file? That seems a little outrageous!
>
> How big is the repository file? CVS needs to read it into memory in
> order to update it.
6,170,889 bytes (after the last check in).
on 11/22/00 11:01 AM, Derek R. Price at [EMAIL PROTECTED] wrote:
> http://cvshome.org/docs/manual/cvs_2.html#SEC27
I had already read that (thanks). The question this brings to mind is that
if CVS can't store diffs of binary files (correct me if I'm wrong, but the
docs lead us to believe it just stores a whole new copy when new binary
versions are committed), why does it need to run a fancy diff that consumes
huge amounts of memory?
Or does it actually diff binaries correctly and only store the changes as
revisions are added (and we read the docs incorrectly)?
Thanks,
Dave
_______________________________________________
Info-cvs mailing list
[EMAIL PROTECTED]
http://mail.gnu.org/mailman/listinfo/info-cvs
