[email protected] wrote:
action/file.py:
- line 113ff: we do this a couple times now; is it worth breaking out
into a utility function? I might also bump the buffer size up to 128k.
If we're going to try to match the buffer size to the underlying
blocksize of the filesystem, it would be a good idea to get this
information from statvfs(2). If Python doesn't have support for this
call natively, it would be a simple C module to write.
I'd be happy to write the module, when I get some more free time.
Provided we'll actually use it to make sensible decisions about buffer
size.
This would be great to have. In the bit of the search was I'm working
on, I think changing the buffer size from the default up to 128k gave
about a 20-30% speed boost in overall install time. I would imagine
anyplace we're reading or writing large files, this would be an
improvement. What I don't know is whether we're doing that often enough
for this change to make a significant difference.
brock
-j
_______________________________________________
pkg-discuss mailing list
[email protected]
http://mail.opensolaris.org/mailman/listinfo/pkg-discuss
_______________________________________________
pkg-discuss mailing list
[email protected]
http://mail.opensolaris.org/mailman/listinfo/pkg-discuss
